V O LU M E
F I F T Y
ADVANCES
T H R E E
IN
MARINE BIOLOGY
Advances in MARINE BIOLOGY
Series Editor
DAVID W. SIMS
Marine Biological Association of the United Kingdom,
The Laboratory Citadel Hill, Plymouth, United Kingdom
Editors Emeritus
LEE A. FUIMAN
University of Texas at Austin
ALAN J. SOUTHWARD
Marine Biological Association of the United Kingdom,
The Laboratory Citadel Hill, Plymouth, United Kingdom
CRAIG M. YOUNG
Oregon Institute of Marine Biology
Advisory Editorial Board
ANDREW J. GOODAY
Southampton Oceanography Centre
GRAEME C. HAYS
University of Wales Swansea
SANDRA E. SHUMWAY
University of Connecticut
ROBERT B. WHITLATCH
University of Connecticut
V O LU M E
F I F T Y
ADVANCES
T H R E E
IN
MARINE BIOLOGY
Edited by
DAVID W. SIMS
Marine Biological Association of the United Kingdom
The Laboratory, Citadel Hill
Plymouth, United Kingdom
AMSTERDAM • BOSTON • HEIDELBERG • LONDON
NEW YORK • OXFORD • PARIS • SAN DIEGO
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Academic Press is an imprint of Elsevier
Academic Press is an imprint of Elsevier
84 Theobald’s Road, London WC1X 8RR, UK
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
Linacre House, Jordan Hill, Oxford OX2 8DP, UK
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
First edition 2008
Copyright # 2008 Elsevier Ltd. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form
or by any means electronic, mechanical, photocopying, recording or otherwise without the prior
written permission of the publisher.
Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford,
UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: permissions@elsevier.com.
Alternatively you can submit your request online by visiting the Elsevier web site at
http://www.elsevier.com/locate/permissions, and selecting
Obtaining permission to use Elsevier material
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or property as
a matter of products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions or ideas contained in the material herein. Because of rapid advances in
the medical sciences, in particular, independent verification of diagnoses and drug dosages should
be made.
ISBN: 978-0-12-374119-6
ISSN: 0065-2881
For information on all Academic Press publications
visit our Web site at www.books.elsevier.com
Printed and bound in The Netherlands
08 09 10 11 12
10 9 8 7 6 5 4 3 2 1
CONTRIBUTORS TO VOLUME 53
Shannon Gowans
Texas A&M University, Galveston, Texas 77551, USA
Eckerd College, Petersburg, Florida 33711, USA
Ray Hilborn
School of Aquatic and Fishery Sciences, University of Washington, Washington
98195, USA
Daniel Huppert
School of Marine Affairs, University of Washington, Seattle, Washington 98195,
USA
Leszek Karczmarski
Texas A&M University, Galveston, Texas 77551, USA
University of Pretoria, Mammal Research Institute, South Africa
Michael J. Keough
Department of Zoology, University of Melbourne, Victoria 3010, Australia
Phillip S. Levin
Northwest Fisheries Science Centre, NOAA Fisheries, Seattle, Washington 98122,
USA
Dustin J. Marshall
School of Integrative Biology/Centre for Marine Studies, University of Queensland,
Queensland 4072, Australia
Kerry A. Naish
School of Aquatic and Fishery Sciences, University of Washington, Washington
98195, USA
Thomas P. Quinn
School of Aquatic and Fishery Sciences, University of Washington, Washington
98195, USA
Joseph E. Taylor, III
Departments of History and Geography, Simon Fraser University, British Columbia,
Canada, USA
Bernd Würsig
Texas A&M University, Galveston Texas 77551, USA
James R. Winton
Western Fisheries Research Center, US Geological Survey, Seattle, Washington
98115, USA
v
This page intentionally left blank
CONTENTS
Contributors to Volume 53
Series Contents for Last Ten Years
v
ix
1. The Evolutionary Ecology of Offspring Size in Marine Invertebrates
1
Dustin J. Marshall and Michael J. Keough
1. Introduction
2. How Variable is Offspring Size Within Species?
3. Effects of Offspring Size
4. Sources of Variation in Offspring Size
5. Offspring-Size Models
6. Summary
Appendix
Acknowledgements
References
2. An Evaluation of the Effects of Conservation and Fishery
Enhancement Hatcheries on Wild Populations of Salmon
3
6
10
28
32
39
46
50
50
61
Kerry A. Naish, Joseph E. Taylor III, Phillip S. Levin, Thomas P. Quinn,
James R. Winton, Daniel Huppert, and Ray Hilborn
1. Introduction
2. Historical Overview of Hatchery Activities
3. Political Dynamics of Hatchery Programmes
4. Geographical Extent of Activities
5. Potential Consequences of Enhancement Activities
6. Economic Perspectives on Hatchery Programmes
7. Discussion
Acknowledgements
References
63
71
78
84
100
150
160
170
170
3. The Social Structure and Strategies of Delphinids:
Predictions Based on an Ecological Framework
195
Shannon Gowans, Bernd Würsig, and Leszek Karczmarski
1. Biological Pressures on Social Strategies
2. Dolphin Ecology
197
205
vii
viii
3. Resident Communities
4. Wide-Ranging Communities
5. Intermediate-Ranging Patterns
6. Demographic, Social and Cultural Influences
7. Comparisons with Other Cetaceans
8. Conservation Implications
9. Concluding Comments
Acknowledgements
References
Taxonomic Index
Subject Index
Contents
223
239
253
267
268
274
276
278
278
295
299
SERIES CONTENTS
FOR
LAST TEN YEARS*
Volume 30, 1994.
Vincx, M., Bett, B. J., Dinet, A., Ferrero, T., Gooday, A. J., Lambshead, P. J. D., Pfannküche, O., Soltweddel, T. and Vanreusel, A.
Meiobenthos of the deep Northeast Atlantic. pp. 1–88.
Brown, A. C. and Odendaal, F. J. The biology of oniscid isopoda of
the genus Tylos. pp. 89–153.
Ritz, D. A. Social aggregation in pelagic invertebrates. pp. 155–216.
Ferron, A. and Legget, W. C. An appraisal of condition measures for
marine fish larvae. pp. 217–303.
Rogers, A. D. The biology of seamounts. pp. 305–350.
Volume 31, 1997.
Gardner, J. P. A. Hybridization in the sea. pp. 1–78.
Egloff, D. A., Fofonoff, P. W. and Onbé, T. Reproductive behaviour
of marine cladocerans. pp. 79–167.
Dower, J. F., Miller, T. J. and Leggett, W. C. The role of microscale
turbulence in the feeding ecology of larval fish. pp. 169–220.
Brown, B. E. Adaptations of reef corals to physical environmental
stress. pp. 221–299.
Richardson, K. Harmful or exceptional phytoplankton blooms in the
marine ecosystem. pp. 301–385.
Volume 32, 1997.
Vinogradov, M. E. Some problems of vertical distribution of mesoand macroplankton in the ocean. pp. 1–92.
Gebruk, A. K., Galkin, S. V., Vereshchaka, A. J., Moskalev, L. I. and
Southward, A. J. Ecology and biogeography of the hydrothermal
vent fauna of the Mid-Atlantic Ridge. pp. 93–144.
Parin, N. V., Mironov, A. N. and Nesis, K. N. Biology of the Nazca
and Sala y Gomez submarine ridges, an outpost of the Indo-West
Pacific fauna in the eastern Pacific Ocean: composition and distribution of the fauna, its communities and history. pp. 145–242.
Nesis, K. N. Goniatid squids in the subarctic North Pacific: ecology,
biogeography, niche diversity, and role in the ecosystem. pp. 243–324.
Vinogradova, N. G. Zoogeography of the abyssal and hadal zones.
pp. 325–387.
Zezina, O. N. Biogeography of the bathyal zone. pp. 389–426.
*The full list of contents for volumes 1–37 can be found in volume 38.
ix
x
Series Contents for Last Ten Years
Sokolova, M. N. Trophic structure of abyssal macrobenthos.
pp. 427–525.
Semina, H. J. An outline of the geographical distribution of oceanic
phytoplankton. pp. 527–563.
Volume 33, 1998.
Mauchline, J. The biology of calanoid copepods. pp. 1–660.
Volume 34, 1998.
Davies, M. S. and Hawkins, S. J. Mucus from marine molluscs.
pp. 1–71.
Joyeux, J. C. and Ward, A. B. Constraints on coastal lagoon fisheries.
pp. 73–199.
Jennings, S. and Kaiser, M. J. The effects of fishing on marine
ecosystems. pp. 201–352.
Tunnicliffe, V., McArthur, A. G. and McHugh, D. A biogeographical
perspective of the deep-sea hydrothermal vent fauna. pp. 353–442.
Volume 35, 1999.
Creasey, S. S. and Rogers, A. D. Population genetics of bathyal and
abyssal organisms. pp. 1–151.
Brey, T. Growth performance and mortality in aquatic macrobenthic
invertebrates. pp. 153–223.
Volume 36, 1999.
Shulman, G. E. and Love, R. M. The biochemical ecology of marine
fishes. pp. 1–325.
Volume 37, 1999.
His, E., Beiras, R. and Seaman, M. N. L. The assessment of marine
pollution—bioassays with bivalve embryos and larvae. pp. 1–178.
Bailey, K. M., Quinn, T. J., Bentzen, P. and Grant, W. S. Population
structure and dynamics of walleye pollock, Theragra chalcogramma.
pp. 179–255.
Volume 38, 2000.
Blaxter, J. H. S. The enhancement of marine fish stocks. pp. 1–54.
Bergström, B. I. The biology of Pandalus. pp. 55–245.
Volume 39, 2001.
Peterson, C. H. The ‘‘Exxon Valdez’’ oil spill in Alaska: acute indirect
and chronic effects on the ecosystem. pp. 1–103.
Johnson, W. S., Stevens, M. and Watling, L. Reproduction and
development of marine peracaridans. pp. 105–260.
Series Contents for Last Ten Years
xi
Rodhouse, P. G., Elvidge, C. D. and Trathan, P. N. Remote sensing
of the global light-fishing fleet: an analysis of interactions with
oceanography, other fisheries and predators. pp. 261–303.
Volume 40, 2001.
Hemmingsen, W. and MacKenzie, K. The parasite fauna of the
Atlantic cod, Gadus morhua L. pp. 1–80.
Kathiresan, K. and Bingham, B. L. Biology of mangroves and mangrove ecosystems. pp. 81–251.
Zaccone, G., Kapoor, B. G., Fasulo, S. and Ainis, L. Structural,
histochemical and functional aspects of the epidermis of fishes.
pp. 253–348.
Volume 41, 2001.
Whitfield, M. Interactions between phytoplankton and trace metals
in the ocean. pp. 1–128.
Hamel, J.-F., Conand, C., Pawson, D. L. and Mercier, A. The sea
cucumber Holothuria scabra (Holothuroidea: Echinodermata): its
biology and exploitation as beche-de-Mer. pp. 129–223.
Volume 42, 2002.
Zardus, J. D. Protobranch bivalves. pp. 1–65.
Mikkelsen, P. M. Shelled opisthobranchs. pp. 67–136.
Reynolds, P. D. The Scaphopoda. pp. 137–236.
Harasewych, M. G. Pleurotomarioidean gastropods. pp. 237–294.
Volume 43, 2002.
Rohde, K. Ecology and biogeography of marine parasites. pp. 1–86.
Ramirez Llodra, E. Fecundity and life-history strategies in marine
invertebrates. pp. 87–170.
Brierley, A. S. and Thomas, D. N. Ecology of southern ocean pack
ice. pp. 171–276.
Hedley, J. D. and Mumby, P. J. Biological and remote sensing
perspectives of pigmentation in coral reef organisms. pp. 277–317.
Volume 44, 2003.
Hirst, A. G., Roff, J. C. and Lampitt, R. S. A synthesis of growth
rates in epipelagic invertebrate zooplankton. pp. 3–142.
Boletzky, S. von. Biology of early life stages in cephalopod molluscs.
pp. 143–203.
Pittman, S. J. and McAlpine, C. A. Movements of marine fish and
decapod crustaceans: process, theory and application. pp. 205–294.
Cutts, C. J. Culture of harpacticoid copepods: potential as live feed
for rearing marine fish. pp. 295–315.
xii
Series Contents for Last Ten Years
Volume 45, 2003.
Cumulative Taxonomic and Subject Index.
Volume 46, 2003.
Gooday, A. J. Benthic foraminifera (Protista) as tools in deep-water
palaeoceanography: environmental influences on faunal characteristics. pp. 1–90.
Subramoniam, T. and Gunamalai, V. Breeding biology of the intertidal
sand crab, Emerita (Decapoda: Anomura). pp. 91–182
Coles, S. L. and Brown, B. E. Coral bleaching—capacity for acclimatization and adaptation. pp. 183–223.
Dalsgaard J., St. John M., Kattner G., Müller-Navarra D. and Hagen
W. Fatty acid trophic markers in the pelagic marine environment.
pp. 225–340.
Volume 47, 2004.
Southward, A. J., Langmead, O., Hardman-Mountford, N. J., Aiken, J.,
Boalch, G. T., Dando, P. R., Genner, M. J., Joint, I., Kendall, M. A.,
Halliday, N. C., Harris, R. P., Leaper, R., Mieszkowska, N., Pingree,
R. D., Richardson, A. J., Sims, D.W., Smith, T., Walne, A. W. and
Hawkins, S. J. Long-term oceanographic and ecological research in the
western English Channel. pp. 1–105.
Queiroga, H. and Blanton, J. Interactions between behaviour and
physical forcing in the control of horizontal transport of decapod
crustacean larvae. pp. 107–214.
Braithwaite, R. A. and McEvoy, L. A. Marine biofouling on fish
farms and its remediation. pp. 215–252.
Frangoulis, C., Christou, E. D. and Hecq, J. H. Comparison of
marine copepod outfluxes: nature, rate, fate and role in the carbon
and nitrogen cycles. pp. 253–309.
Volume 48, 2005.
Canfield, D. E., Kristensen, E. and Thamdrup, B. Aquatic Geomicrobiology. pp. 1–599.
Volume 49, 2005.
Bell, J. D., Rothlisberg, P. C., Munro, J. L., Loneragan, N. R., Nash,
W. J., Ward, R. D. and Andrew, N. L. Restocking and stock
enhancement of marine invertebrate fisheries. pp. 1–358.
Volume 50, 2006.
Lewis, J. B. Biology and ecology of the hydrocoral Millepora on coral
reefs. pp. 1–55.
Series Contents for Last Ten Years
xiii
Harborne, A. R., Mumby, P. J., Micheli, F., Perry, C. T., Dahlgren,
C. P., Holmes, K. E., and Brumbaugh, D. R. The functional value of
Caribbean coral reef, seagrass and mangrove habitats to ecosystem
processes. pp. 57–189.
Collins, M. A. and Rodhouse, P. G. K. Southern ocean cephalopods.
pp. 191–265.
Tarasov, V. G. EVects of shallow-water hydrothermal venting on
biological communities of coastal marine ecosystems of the western
Pacific. pp. 267–410.
Volume 51, 2006.
Elena Guijarro Garcia. The fishery for Iceland scallop (Chlamys
islandica) in the Northeast Atlantic. pp. 1–55.
JeVrey, M. Leis. Are larvae of demersal fishes plankton or nekton?
pp. 57–141.
John C. Montgomery, Andrew Jeffs, Stephen D. Simpson, Mark
Meekan and Chris Tindle. Sound as an orientation cue for the
pelagic larvae of reef fishes and decapod crustaceans. pp. 143–196.
Carolin E. Arndt and Kerrie M. Swadling. Crustacea in Arctic
and Antarctic sea ice: Distribution, diet and life history strategies.
pp. 197–315.
Volume 52, 2007.
Leys, S. P., Mackie, G. O. and Reiswig, H. M. The Biology of Glass
Sponges. pp. 1–145.
Garcia E. G. The Northern Shrimp (Pandalus borealis) Offshore
Fishery in the Northeast Atlantic. pp. 147–266.
Fraser K. P. P. and Rogers A. D. Protein Metabolism in Marine
Animals: The underlying Mechanism of Growth. pp. 267–362.
The Evolutionary Ecology
of Offspring Size in
Marine Invertebrates
Dustin J. Marshall* and Michael J. Keough†
Contents
3
6
6
10
10
13
21
28
28
31
32
33
36
38
39
39
40
42
42
43
43
45
46
50
50
1. Introduction
2. How Variable is Offspring Size Within Species?
2.1. Meta-analysis of the literature
3. Effects of Offspring Size
3.1. Fertilization
3.2. Development
3.3. Post-metamorphosis
4. Sources of Variation in Offspring Size
4.1. Within populations
4.2. Among populations
5. Offspring-Size Models
5.1. Offspring size-number trade-off
5.2. Offspring size-fitness function
5.3. Reconciling within-clutch variation
5.4. Summary of offspring-size models
6. Summary
6.1. Planktotrophs
6.2. Non-feeding
6.3. Direct developers
6.4. Ecological implications
6.5. Evolutionary implications
6.6. Future research directions
Appendix
Acknowledgements
References
*
{
School of Integrative Biology/Centre for Marine Studies, University of Queensland,
Queensland 4072, Australia
Department of Zoology, University of Melbourne, Victoria 3010, Australia
Advances in Marine Biology, Volume 53
ISSN 0065-2881, DOI: 10.1016/S0065-2881(07)53001-4
#
2008 Published by Elsevier Ltd.
1
2
Dustin J. Marshall and Michael J. Keough
Abstract
Intraspecific variation in offspring size is of fundamental ecological and evolutionary importance. The level of provisioning an organism receives from its
mother can have far reaching consequences for subsequent survival and performance. In marine systems, the traditional focus was on the remarkable variation
in offspring size among species but there is increasing focus on variation in
offspring size within species. Here we review the incidence and consequences
of intraspecific offspring-size variation for marine invertebrates.
Offspring size is remarkably variable within and among marine invertebrate
populations. We examined patterns of variation in offspring size within populations using a meta-analysis of the available data for 102 species across 7 phyla.
The average coefficient of variation in offspring size within populations is 9%,
while some groups (e.g., direct developers) showed much more variation (15%),
reflecting a fourfold difference between the largest and smallest offspring in
any population.
Offspring-size variation can have for reaching consequences. Offspring size
affects every stage of a marine invertebrate’s life history, even in species in
which maternal provisioning accounts for only a small proportion of larval
nutrition (i.e., planktotrophs). In species with external fertilization, larger
eggs are larger targets for sperm and as such, the sperm environment may
select for different egg sizes although debate continues over the evolutionary
importance of such effects. Offspring size affects the planktonic period in many
species with planktotrophic and lecithotrophic development, but we found that
this effect is not universal. Indeed, much of the evidence for the effects of
offspring size on the planktonic period is limited to the echinoids and in this
group and other taxa there is variable evidence, suggesting further work is
necessary. Post-metamorphic effects of offspring size were strong in species
with non-feeding larvae and direct development: bigger offspring generally
have higher post-metamorphic survival, higher growth rates and sometimes
greater fecundity. Although there is limited evidence for the mechanisms
underlying these effects, the size of post-metamorphic feeding structures and
resistance to low-food availability appear to be good candidates. There was
limited evidence to assess the effects of offspring size on post-metamorphic
performance in planktotrophs but surprisingly, initial indications suggest that
such effects do exist and in the same direction as for species with other
developmental modes. Overall, we suggest that for direct developers and
species with non-feeding larvae, the post-metamorphic effects of offspring
size will be greatest source of selection.
Offspring-size variation can arise through a variety of sources, both within
and among populations. Stress, maternal size and nutrition, and habitat quality
all appear to be major factors affecting the size of offspring, but more work on
sources of variation is necessary. While theoretical considerations of offspring
size can now account for variation in offspring size among mothers, they
struggle to account for within-brood variation. We suggest alternative
approaches such as game theoretic models that may be useful for reconciling
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
3
within-clutch variation. While some of the first theoretical considerations of
offspring size were based on marine invertebrates, many of the assumptions of
these models have not been tested, and we highlight some of the important
gaps in understanding offspring-size effects. We also discuss the advantages of
using offspring size as a proxy for maternal investment and review the evidence
used to justify this step.
Overall, offspring size is likely to be an important source of variation in the
recruitment of marine invertebrates. The quality of offspring entering a population could be as important as the quantity and further work on the ecological
role of offspring size is necessary. From an evolutionary standpoint, theoretical
models that consider every life-history stage, together with the collection of
more data on the relationship between offspring size and performance at each
stage, should bring us closer to understanding the evolution of such a wide
array of offspring sizes and developmental modes among species.
1. Introduction
Offspring size is a trait of fundamental interest to evolutionary biologists and ecologists. Offspring size, for our purposes, will be defined as
the volume of a propagule once it has become independent of maternal
nutritional investment. According to this definition, the size of freely
spawned egg is the appropriate measure of offspring size but the size of a
direct developing snail egg before the embryo has ingested nurse eggs is not.
The enormous range of offspring sizes observed among species, even closely
related ones (Fig. 1.1), has long fascinated biologists as to the selection
pressures that led to such divergent sizes (Lack, 1947). Offspring size is of
particular interest because while it has fitness consequences for both the
offspring and mother, selection acts to maximize maternal fitness only
(Bernardo, 1996). Thus, mother and individual offspring may be in conflict
with regards to the strategy that maximizes fitness (Einum and Fleming,
2000). Similarly ecologists have long been interested in the role of maternal
investment in population dynamics (Bagenal, 1969) and for most organisms,
offspring size is the sole source of maternal investment. If we hope to
understand the evolutionary pressures acting on offspring size, then we
must first determine the ecological consequences of this variation within
species. In his excellent review of propagule size effects, Bernardo (1996,
pp. 219–220) points out ‘The ecological and evolutionary implications of
variation in propagule size are evaluated by selection and modelled by
theoreticians as a within-population variance component. . . . It is at the
within- to among-population (intraspecific) level that we should seek evidence of the ecological role and evolutionary dynamics of propagule size
and their relationship to parental phenotypes . . .’. The goals of evolutionary
4
Dustin J. Marshall and Michael J. Keough
Figure 1.1 Micrographs of the offspring from four species of closely related Australian
sea stars within the Asterinidae, left to right is the brooded Parvulustra parvivipara, the
benthic lecithotroph Parvulustra exigua, the lecithotrophic broadcast spawner Meridiastra
calcar and the planktotrophic broadcast spawner Patiriella regularis. Scale bar is 100 mm
and note that for P. parvivipara, the scale bar is half the size of the others. Micrographs
courtesy of Maria Byrne.
biology and ecology can be achieved by the same means then—the examination of the effects of offspring size within individual species.
Marine invertebrates have one of the most diverse and striking range of
offspring sizes exhibited among species. For example, latitudinal patterns
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
5
in offspring size were noted more than 50 years ago by Thorson (1950).
Consequently, one of the first attempts at modelling offspring size was
done with marine invertebrates in mind (Vance, 1973a). In their excellent
review of Conus life-history strategies, Kohn and Perron (1994) concluded
that, ‘. . . egg size is the single most important attribute of understanding
(1) reproductive energetics; (2) the temporal patterns of embryonic development and larval biology; (3) dispersal potential, which is tightly linked to
(1) and (2) but is an evolutionary ‘‘byproduct’’. . .’. However, despite a
long fascination with offspring-size evolution, intraspecific variation in
offspring size and its effects have only recently been examined in detail.
One of the first studies on intraspecific offspring-size variation in marine
invertebrates was done in the late 1970s (Turner and Lawrence, 1977),
but for the most part earlier work tended to focus on the effects of
interspecific variation (e.g., Berrill, 1935; Dickie et al., 1989; Emlet
et al., 1987; Hoegh-Guldberg and Pearse, 1995; Staver and Strathmann,
2002; Steele, 1977 but see Kohn and Perron, 1994). Indeed, it has been
our personal experience that people are surprised when we present
remarkably high levels of variation in offspring size within species.
However, in both the terrestrial and marine literature, offspring-size
studies have increasingly focused on within-species variation. We believe
that this is appropriate: as pointed out by Bernardo (1996), it is inappropriate to use interspecific studies to infer ecological effects or evolutionary
transitions without controlling for species relationships (for detailed discussion, see Harvey and Pagel, 1991). More importantly, intraspecific
variation in offspring size has the potential to dramatically change our
view of the dynamics of marine invertebrate populations. Settling larvae
are traditionally viewed (and modelled) as being homogenous in their
chances of recruiting and their post-settlement performance (Eckman,
1996; Vance, 1973). However, it has become clear that settling larvae
vary greatly in their potential to survive and grow to reproduction.
Exposure to pollutants, increased swimming durations or larval activity
levels and larval nutrition can strongly affect post-metamorphic performance in a range of taxa (Highsmith and Emlet, 1986; Marshall et al.,
2003b; Ng and Keough, 2003; Pechenik et al., 1998, 2001; Phillips and
Gaines, 2002; Wendt, 1998). Many of these ‘carry-over’ effects are
thought to be mediated by variation in larval energetic reserves (Bennett
and Marshall, 2005; Wendt, 2000) such that if larger offspring have more
energetic reserves than smaller offspring, then similar effects would be
expected. Thus, offspring size could be an important source of variation in
larval quality and, consequently, variation in recruitment. Traditionally,
we have viewed marine invertebrate populations as being strongly affected
by the quantity of larvae entering a population; intraspecific variation in
offspring could also mean that the quality of larvae could have equally
important effects.
6
Dustin J. Marshall and Michael J. Keough
In light of the evolutionary and ecological importance of intraspecific
variation in offspring size, we have several aims for this review:
1. Document and quantify the amount of variation in offspring size within
marine invertebrate species.
2. Review the known effects of offspring size across the various life-history
stages of marine invertebrates.
3. Identify the common sources of intraspecific variation in offspring size
within and among marine invertebrate populations.
4. Summarize the findings of the theoretical literature on offspring-size
effects in marine invertebrates.
5. Identify the key knowledge gaps that currently limit our understanding
of the ecological and evolutionary consequences of offspring-size
variation.
Our first aim represents an attempt to highlight the fact that offspring sizes
are extremely variable within and among marine invertebrate populations.
Our second aim is a first attempt at integrating the various findings for
different life-history stages, and we hope to demonstrate that selection is
likely to act on offspring size across multiple, if not all, life-history stages.
We will demonstrate that offspring-size variation can have pervasive and
important effects on performance at each life-history stage, so the next
obvious step is to identify some of the sources of this variation. We will
then examine whether current theoretical considerations of the issue match
our empirical findings and the problems associated with various approaches.
We will then attempt to identify the appropriate next steps in understanding
the evolutionary and ecological consequences of offspring-size variation
within species.
2. How Variable is Offspring Size
Within Species?
2.1. Meta-analysis of the literature
In this section, we summarize the degree of variation in offspring sizes from
species, from a range of taxa and from 7 phyla (including 35 orders,
58 families and 102 species) and compare the relative variation from each
of the three major developmental modes (planktotrophic, lecithotrophic
and direct development). Here, lecithotrophic is used as a general term for
non-feeding larvae; however, it is recognized that not all non-feeding larvae
are necessarily ‘yolk feeding’. Facultative planktotrophs (species that can
feed but do not necessarily need to in order to become competent to
metamorphose McEdward, 1997) are considered to be lecithotrophic for
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
7
the purposes of this chapter. For the purposes of our analysis and our review
more generally, we define ‘direct development’ here as any development
whereby the offspring are fully formed juveniles independent of maternal
nutrition sources (although not necessarily maternal nutrition: these juveniles may still be utilizing maternal yolk reserves). The two groups with
planktonic development were further partitioned into internal and external
fertilizers. We compiled data on variation in offspring size from the available
literature and from our own unpublished data (see Appendix). The most
commonly reported measure of offspring size was length of embryos/newly
hatched larvae or egg diameter. For a number of species (especially the
gastropods), the sizes of a range of different developmental stages were
available. Because we were interested in variation in total maternal investment per offspring, the measure that best reflected this investment was
utilized. For example, for gastropods that fed on nurse eggs prior to
hatching, we utilized measures of size and variation in size for newly
hatched juveniles rather than those parameters for newly laid eggs
(cf. Kohn and Perron, 1994). We used data for species only where the
eggs of two or more individuals were measured. Often the source of
the variation (among broods or within broods) was not reported and so for
the majority of cases we cannot determine the principal source of the variation.
Data on offspring size were compiled from studies that collected females
from the same population and in most cases the same time, although in some
cases data were compiled from a single reproductive season. Many other
studies were excluded because no details were provided of the numbers of
individuals on which the summary of offspring size was based. Egg volume
was also a commonly reported measure, although we did not include data
using this parameter in most of the analyses because variance [and more
importantly, coefficients of variation (CVs)] in diameter and volume are not
equivalent and, more importantly, do not scale linearly. Thus, we would
discourage the approach used by Einum and Fleming (2002) whereby CVs
in egg volume and diameter are pooled because this could introduce biases
to the analysis.
To overcome the problems associated with traditional comparative
analyses (i.e., treating individual species as replicates), we used the method
of higher node contrasts (Harvey and Pagel, 1991). We tested the effects of
developmental type and reproductive mode at the species, family and order
level. The effects of developmental type (direct, lecithotrophic and planktotrophic as different levels) on CVs of offspring size within populations
were tested with ANOVA. For the planktotrophs and lecithotrophs, we
also compared the CV of offspring size within developmental types for two
reproductive modes: internal and external fertilization.
The level of within-population variability in offspring size differed considerably between species, with CVs ranging from 0.7% to 51% (Fig. 1.2).
The average CV across the entire set of species was approximately 9% and is
8
Dustin J. Marshall and Michael J. Keough
Coefficient of variation (%)
30
20
10
hic
rop
tot
nk
Pla
Le
cit
ho
tro
Di
ph
rec
ic
t
0
Figure 1.2 Mean (SE) CV for offspring size within populations of marine invertebrates. Data are compiled from the literature for three developmental types: direct
developers, lecithotrophic and planktotrophic.
similar to previous averages reported for Conus species (Kohn and Perron,
1994). This level of variation in diameter means that within any group
of eggs, about a third of all the offspring will be 25% larger or smaller in
volume than the average size and about 5% will be 50% larger or smaller in
volume than the average size. Note that our calculations assume a normal
distribution of egg sizes, and the data appear to reflect this distribution.
Intra-population variation in offspring size varied strongly with developmental mode, and this pattern was consistent at all taxonomic levels that were
tested (analysis using species, F2,99 ¼ 27.13; families, F2,64 ¼ 24.6; orders,
F2,40 ¼ 22.51; all P < 0.001; and all pairwise comparisons P < 0.001).
Variation was greatest in the direct developers, least in planktotrophs and
intermediate in lecithotrophs (Fig. 1.2).
For the direct developers only, a CV of 14% means that about a third of all
the offspring will be 48% larger or smaller in volume than the average size and
5% of offspring will be more than twice the average size. Put in another way,
this means a fourfold difference between the smallest 5% and largest 5%
of individuals within any one population. These figs. show that there is
an impressive range of offspring sizes being produced within any one population of marine invertebrate. Given that the size of direct developers was, on
average, greater than indirect developers, we were concerned that the effects
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
9
of developmental type were confounded with offspring size. Generally,
however, CV was not correlated with offspring size within all three reproductive modes (direct developers: R2 ¼ 0.104, n ¼ 20, P ¼ 0.165; planktotrophs:
R2 ¼ 0.03, n ¼ 39, P ¼ 0.3; lecithotrophs: R2 ¼ 0.02, n ¼ 43, P ¼ 0.8).
Within the lecithotrophs, internal fertilizers had higher levels of withinpopulation variation in offspring size than external fertilizers (F1,37 ¼ 10.85,
P ¼ 0.002, Fig. 1.3). For planktotrophs, there was no effect of fertilization
mode on the level of within-population variation in offspring size (F1,40 < 0.01,
P ¼ 0.99, Fig. 1.3) despite the fact that the power to detect an effect similar
to that seen for non-feeding larvae was high at 0.9.
Given the strong effects of relatively small differences in offspring size
discussed later in this chapter, it seems that among all the developmental
modes, the quality and performance of offspring will be highly variable
within any single population. Our results suggest that the relative importance of larval quality and quantity for subsequent population dynamics will
depend on developmental type. Overall, variation in offspring size in direct
developers was very high—thus, larval quality may be particularly important
for explaining variation in recruitment in this group because different
individuals can vary markedly in quality.
It is difficult to speculate as to the cause of the pronounced differences in
offspring-size variation between developmental and fertilization modes as a
30
Internal
Coefficient of variation (%)
External
20
10
0
Lecithotrophic
Planktotrophic
Figure 1.3 Mean (SE) CV for offspring size within populations for marine invertebrates with planktotrophic or lecithotrophic development. The shaded bars represent
species with external fertilization and open bars represent species with internal
fertilization.
10
Dustin J. Marshall and Michael J. Keough
number of factors could be driving this effect. However, it is worth noting
that there appears to be a strong, negative relationship between the dispersal
potential of offspring and variation in offspring size, the most variation in
the non-dispersing direct developers through to internally fertilized lecithotrophs and the least variation in the highly dispersive externally fertilized
planktotrophs. Further work distinguishing between the two sources of
intrapopulation variation (within brood and among individuals; e.g.,
Kohn and Perron, 1994) may shed light on the causes behind the systematic
differences between the various developmental groups.
3. Effects of Offspring Size
From our analysis of within-population offspring-size variation, it is
clear that offspring sizes can be highly variable within species and populations. In this section, we review the effects of offspring-size variation on
each of the major life-history stages across the various developmental
modes. We do not consider the differential benefits to mothers of brooding
smaller versus larger offspring but note that in some other groups this can be
a major factor (Sakai and Harada, 2001). Unfortunately, there are far too
few data on this important issue in marine invertebrates and so we focus on
each of the life-history stages following the release of gametes/offspring.
3.1. Fertilization
In Thorson’s consideration of free-spawning invertebrates, he concluded
that ‘. . . . failure of insemination, cannot explain the enormous waste
[of eggs] found in most marine invertebrates during development. The
heavy waste takes place after fertilisation, during the free swimming pelagic
life’ (Thorson, 1950). Since then it has become clear that fertilization is not
assured in free-spawners, and the production of zygotes can be a potentially
limiting factor in the population dynamics of some species (Levitan, 1991,
1995; Levitan and Petersen, 1995; Levitan et al., 1992; Pennington, 1985;
Yund, 2000). Here we define ‘free-spawning’ as the release of both sperm
and eggs into the water column. While free-spawning has alternatively been
termed ‘broadcast spawning’ (Byrne et al., 2003; Oliver and Babcock,
1992), we prefer the term free-spawning, partly because in many species,
eggs are not ‘broadcasted’ into the water column but remain in a viscous
matrix near the spawning female (Marshall, 2002; Thomas, 1994; Williams
et al., 1997). However, we should note that we separate free-spawning/
broadcast spawning from species where only sperm are shed into the water
column while eggs are retained (sometimes termed ‘spermcast spawning’;
Pemberton et al., 2003). The principal factor determining the fertilization
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
11
rate of spawned eggs is the collision rate between eggs and sperm (Vogel
et al., 1982). The collision rate between eggs and sperm is affected by a range
of factors, but most important is the concentration of sperm present
(Marshall et al., 2000; Styan, 1998). Thus, any factors that change the
amount of sperm present in the water column will affect female fertilization
success and, accordingly, the density of spawning males and local hydrodynamic conditions will strongly affect fertilization rates in the natural environment (Denny and Shibata, 1989; Denny et al., 1992; Franke et al., 2002;
Lasker et al., 1996; Levitan, 1991; Levitan et al., 1992; Marshall, 2002;
Marshall et al., 2004b; Mead and Denny, 1995; Yund, 1990). Given that
sperm can quickly dilute to ineffective concentrations in the field (Babcock
et al., 1994; Denny and Shibata, 1989), a number of adaptations exist that
enhance fertilization success in free-spawners and most relevant to the
discussion here is the effect of egg size.
Larger eggs present a larger ‘target’ for sperm and are therefore more
likely to be contacted within a given period of time than smaller sperm
(Levitan, 1996a; Marshall et al., 2002; Styan, 1998; Vogel et al., 1982).
Both in the laboratory and the field, when sperm are scarce, larger eggs are
more likely to be fertilized than smaller eggs (Levitan, 1996a,b; Marshall and
Keough, 2003; Marshall et al., 2002). However, when sperm are abundant,
larger eggs are more likely to suffer polyspermy than smaller eggs either
because they are more likely to be contacted by multiple sperm before they
have formed a block to polyspermy or because such blocks are slower
(Marshall and Keough, 2003; Marshall et al., 2002; Millar and Anderson,
2003; Styan, 1998). Therefore, under sperm-limiting conditions, larger
eggs are more likely to be successfully fertilized, while under polyspermy
conditions, smaller eggs are more likely to be fertilized. Debate continues
about the prevalence of sperm limitation and polyspermy under natural
conditions, but it is clear that both can occur simultaneously in the
same spawning population (Brawley, 1992; Franke et al., 2002; Marshall,
2002).
The effects of egg size on fertilization rate have led to speculation
about the evolution of egg sizes of free-spawning marine invertebrates and
the evolution of anisogamy (Levitan, 1993, 1996a; Podolsky, 2001;
Podolsky and Strathmann, 1996). It has been suggested that in habitats
that are conducive to sperm-limiting conditions, larger eggs have evolved
relative to species in habitats where sperm limitation is unlikely (Levitan,
1998, 2002). More broadly, Levitan argues that the evolution of egg size
in marine invertebrates will be strongly influenced by the pre-zygotic
selection associated with fertilization. In contrast, Podolsky and
Strathmann (1996) argue that the benefits of increased egg size for fertilization will be outweighed by the reduction in fecundity associated
with this increase. Furthermore, they argue that post-zygotic selection
(i.e., the effects of offspring size on developmental and post-metamorphic
12
Dustin J. Marshall and Michael J. Keough
performance) will also shape offspring size and overall pre-zygotic selection (i.e., effects on fertilization alone) will be less important (Podolsky
and Strathmann, 1996). This debate has been broadened by the discussion
of the effects of egg accessory structures. Many invertebrates produce eggs
that have large accessory structures such as jelly coats or follicle cells
surrounding the eggs (reviewed in Podolsky, 2001). It has been argued
that, as accessory structures are energetically inexpensive relative to egg
material (Bolton and Thomas, 2000), any selection to increase egg target
size will result in an increase in these structures rather than the egg cell
itself (Podolsky, 2001, 2002; Podolsky and Strathmann, 1996). Farley and
Levitan (2001) propose that despite the effects of accessory structures on
fertilization, there will still be substantial selection on egg size due to
fertilization effects. Podolsky (2001) raises the interesting possibility that
egg accessory structures may mediate sperm–egg interactions in ways other
than just size (e.g., increase the variance in sperm arrival times, thereby
reducing polyspermy); it will be interesting to determine if this is indeed
the case. Overall, this debate remains unresolved. Nevertheless it seems
unlikely that pre-zygotic selection on egg size will be the sole force in the
evolution of egg size in free-spawners, especially given the post-zygotic
effects of egg size outlined later in this chapter.
An important consequence of the size-dependent fertilization of eggs is
that under different sperm concentrations, the same brood of unfertilized
eggs will produce zygotes of different sizes. At low sperm concentrations
larger zygotes will be produced, but if that same brood is exposed to a
high sperm concentration then (because of polyspermy effects) smaller,
viable zygotes will be produced (Levitan, 1996a,b; Marshall and Keough,
2003; Marshall et al., 2002). Therefore, free-spawning marine invertebrates appear to be unusual in that the size distribution of zygotes is a
product not only of maternal investment but also of the local ‘sperm
environment’. This raises the interesting possibility that free-spawning
males and females may be in conflict at fertilization whereby male fitness
is maximized by a strategy that may reduce female fitness (Franke et al.,
2002). Finally, size-dependent fertilization means that producing offspring
of optimal size is further complicated in free-spawning species because
an additional ‘layer’ of size-dependent selection occurs at fertilization.
For example, because free-spawning mothers may be competing for limiting sperm (Marshall and Evans, 2005), selection at this stage could act
to increase optimal egg size (and thus competitive ability) but selection at
another stage could act to decrease optimal egg size (i.e., size-independent
mortality at settlement). Clearly, the effects of egg size at fertilization in
free-spawners have the potential to influence egg size evolution; further
research on the effects of offspring size across the entire life history of an
organism will determine the relative strength of selection on egg size at
each stage.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
13
3.2. Development
Our view of the effects of offspring size on developmental time (time to a
functional stage such as feeding or metamorphic competence) has been
shaped largely by the early models of Vance (1973a,b). In both of Vance’s
original papers, it is assumed that the length of the pre-feeding period (or l)
increased with offspring size, and that the length of feeding period (r)
decreased with offspring size. For non-feeding larvae with planktonic
development, this means that the total planktonic period should be positively
correlated with egg size because r is zero (although Vance restricted his
considerations for S 1). For feeding larvae, Vance (1973a, p. 342) made
another assumption that the influence of egg size on l was much smaller
than its influence on r (in terms of the model P 1) and so concluded that
for planktotrophic species, egg size should be negatively correlated with
planktonic period. Vance’s models predicted that only very small or very
large offspring should be selected and since then various modifications of the
model have been made that better approximate the distribution of egg sizes
observed in nature (Levitan, 1993, 2000; Podolsky and Strathmann, 1996;
Styan, 1998). Since Vance’s work, the assumptions that increases in egg size
result in decreases in the planktonic period for feeding larvae and increases
in the planktonic period of non-feeding larvae have become accepted in
theoretical considerations of the topic and some conceptual works (Levitan,
2000; Ramirez-Llodra, 2002 but see Strathmann, 1977). For direct developers, there has been little speculation on the effects of offspring size on
subsequent development but some general offspring-size models do assume
a relationship between offspring size and developmental time (Sargent et al.,
1987). In this section, we review the available evidence for each of these
assumptions.
3.2.1. Planktotrophs
3.2.1.1. Pre-feeding period There are very few studies examining the
effects of egg size on the pre-feeding period alone in marine invertebrates.
In one of the few studies to examine the effects of intraspecific variation in
egg size on pre-feeding developmental time, McLaren (1965, 1966) showed
that larger eggs took longer time to hatch than smaller eggs for the copepod
Pseudocalanus minutus. Similarly in another copepod, Lonsdale and Levinton
(1985) found across four populations that female Scottolana cnadensis from
the population that produced the largest eggs also produced eggs that took
the longest time to hatch into nauplii. In contrast, Jones et al. (1996) found
no correlation between egg size and embryonic developmental period
among populations in the nudibranch Adalaria proxima. Among species,
the effect of egg size on the length of pre-feeding period is also unclear.
Dickie et al. (1989) compared the length of pre-feeding periods among
strongylocentrotid sea urchin species and found no influence of egg size,
14
Dustin J. Marshall and Michael J. Keough
although only one clutch from each species was used in this study. In a study
across 20 different species with feeding larvae from a number of phyla,
Staver and Strathmann (2002) found that the time until first swimming was
positively correlated with egg size, but within individual groups (Urochordates, Echinoderms and Spiralia), the effects of egg size were less clear and
were limited by small sample sizes. Some of the best evidence comes from
Kohn and Perron’s study showing a positive relationship between the prefeeding (pre-hatching) period and egg size among species of Conus (Kohn
and Perron, 1994).
3.2.1.2. Entire planktonic period: Interspecific comparisons Emlet (1995)
considered the relationship between overall length of the developmental
period (from the fertilization of eggs to metamorphic competence) of
28 echinoids with feeding larvae and 5 echinoids with non-feeding larvae.
He found that the larval period decreased with increased egg size across
both developmental modes (Emlet, 1995). Extending Emlet’s work,
Levitan (2000) examined the relationship between egg size and planktonic
duration for 37 echinoid species, all with feeding larvae. Again, it was found
that those species with larger eggs had a shorter planktonic period than those
with smaller eggs. It should be noted that for both studies, developmental
periods were first adjusted with Q10 values so that comparisons could be made
across species that occurred at different water temperatures. Before adjusting
for temperature, Emlet (1995) found no relationship between egg size and
larval developmental period. Importantly, Levitan (2000) found that the
relationship between egg size and planktonic period was not linear
(as assumed by Vance, 1973a,b), rather, it was curvilinear with initial increases
in egg size resulting in a large decrease in the planktonic period; however,
with increased egg size the concomitant reduction in planktonic period
was less.
Analysing the available data for Asteroidea (Echinodermata), across the
planktotrophs and lecithotrophs, there is the expected relationship between
egg size and planktonic period (Emlet et al., 1987). However, the effects of
egg size on developmental rate within feeding larvae alone are less clear.
Using Q10 values of 2, there is a significant negative correlation between egg
size and planktonic period but using values of 3–3.6 (typical values for
echinoderm larvae, see Emlet, 1995), there is no significant correlation
between egg size and planktonic period. Similar results were reported by
Hoegh-Guldberg and Pearse (1995).
There have been far fewer studies examining the effects of egg size on
the planktonic period of feeding larvae that are not focused on echinoderms.
Kohn and Perron (1994) showed a strong negative relationship between
the minimum planktonic period and egg diameter in Conus (Fig. 1.5).
Havenhand (1993) compared the developmental time (again standardizing
for differences in temperature) of opisthobranch molluscs with a range of
15
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
different egg sizes. He found that across developmental modes (planktotrophic and lecithotrophic) there was a strong negative relationship between
egg size and developmental time, with lecithotrophs having a shorter
developmental period (Havenhand, 1993). However, when lecithotrophs
and planktotrophs are considered separately, there is no relationship
between egg size and developmental time for either group (planktotrophs:
R2 ¼ 0.01, n ¼ 18, P > 0.5; lecithotrophs: R2 ¼ 0.269, n ¼ 13, P ¼ 0.069).
This suggests that apart from the broad differences between planktotrophs
and lecithotrophs, there is no effect of egg size on developmental time in
this group despite the comparisons being made across a broad range of egg
sizes within each group (e.g., for planktotrophs between 65 and 149 mm;
Fig. 1.4). Further studies on other, non-echinoid groups are necessary to
determine if the relationship between egg size and the planktonic period for
feeding larvae is applicable to other groups of organisms.
We compiled the data from the review by Kupriyanova et al. (2001) on
serpulimorph polychaetes, to examine the relationship between egg size and
developmental time and to be conservative, we compared our results across
two values of Q10: 2 and 3.6 (the results were qualitatively independent of
the Q10 that was used). Once again, the polychaete data do not resemble the
data on echinoids. Across feeding and non-feeding larvae, developmental
Developmental time (days)
140
100
60
60
80
100
140
Egg diameter (mm)
Figure 1.4 Relationship between egg size and developmental time (adjusted to
development at 10 C using Q10 values) for opisthobranch molluscs with planktotrophic larvae, each point represents a single species. Data taken from Havenhand (1993),
note the log scale. Developmental time is defined here as the time from egg release/
fertilization through to metamorphic competence.
16
Dustin J. Marshall and Michael J. Keough
Developmental time (days)
40
30
20
10
60
80
100
120
140
Egg diameter (mm)
Figure 1.5 Relationship between egg size and developmental time (adjusted to
development at 20 C using Q10 values) for serpulimorph polychaetes with planktotrophic (as indicated by circles and the dashed line) or lecithotrophic (as indicated by
the crosses and solid line). Data taken from Kupriyanova et al. (2001), and developmental
time was defined as time taken to reach metamorphic competence.
time does decrease with increasing egg size (Q10 ¼ 2, R2 ¼ 0.199, n ¼ 20,
P ¼ 0.048; Q10 ¼ 3.6, R2 ¼ 0.212, n ¼ 20, P ¼ 0.041). However, there is
no statistically significant relationship between egg size and developmental
time (Q10 ¼ 2, R2 ¼ 0.299, n ¼ 13, P ¼ 0.053; Q10 ¼ 3.6, R2 ¼ 0.154, n ¼ 13,
P ¼ 0.184) when planktotrophs are examined on their own (Fig. 1.5). In fact,
there is almost a positive relationship between egg size and developmental
time in planktotrophic polychaetes (Fig. 1.6). Our discussion of interspecific
patterns that have been adjusted using Q10 values should be tempered with
the fact that such measures that assume Q10 values remain constant across
temperature ranges, a condition that is unlikely in some instances (HoeghGuldberg and Pearse, 1995). This further emphasizes the advantages of
examining offspring-size effects within rather than among species.
For holoplanktonic species, the pattern appears to be similar. Guisande
and Harris (1995) found that hatching success and naupliar survival under
conditions of starvation were positively correlated with egg size in the
copepod Calanus helgolandicus. Similarly, Lonsdale and Levinton (1985)
found, among populations, that egg size was positively correlated with
naupliar survival under low-food conditions but not high-food conditions.
In summary, it appears that for planktotrophic larvae, the echinoids and
Conus gastropods are the only groups where egg size consistently affects the
17
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
Developmental time (days)
35
30
25
20
15
10
40
50
60
70
Egg diameter (mm)
80
90
Figure 1.6 Relationship between egg size and developmental time (adjusted to development at 20 C using Q10 values) for serpulimorph polychaetes with planktotrophic
larvae. Data taken from Kupriyanova et al. (2001), note the log scale. Developmental
time is defined here as the time from egg release/fertilization through to metamorphic
competence.
planktonic period, whereas for other groups there appears to be little
support for the idea that larvae from smaller eggs require longer time to
develop. This may be because the influence of egg size on the pre-feeding
period is actually much larger than has previously been assumed, potentially
balancing out (or perhaps even overcoming in the case of polychaetes) the
effects of egg size on the length of the feeding period.
3.2.1.3. Entire planktonic period: Intraspecific comparisons The number
of studies on feeding larvae that examine the effect of egg size on developmental time within species is severely limited. Bertram and Strathmann
(1998) found a small effect of maternal source habitat (and thus egg size)
on the developmental rate of larvae of the urchin Strongylocentrotus droebachiensis fed in the laboratory. However, Meidel et al. (1999) found that
increased egg size in S. droebachiensis resulted in an increase in the rate of
metamorphosis for larvae fed high but not low food rations. George et al.
(1990) found that larvae from larger eggs were competent to settle sooner
than larvae from smaller eggs in the echinoids Arbacia lixula and Paracentrotus
lividus, although these comparisons are based on mothers from different sites.
While examinations of the effect of natural variation in egg size are
rare, several studies have examined the effects of manipulating egg size.
18
Dustin J. Marshall and Michael J. Keough
Hart (1995) halved the size of developing eggs in the echinoid
S. droebachiensis with no effect on time to metamorphose and only a small
effect on juvenile size. Surprisingly, he found no effect of egg size on the
time taken to progress through the larval stage despite strong effects on larval
feeding rates, with larvae coming from halved eggs having lower clearance
rates than those from unmanipulated eggs. Sinervo and McEdward (1988)
manipulated the size of S. droebachiensis eggs and found that egg size strongly
affected developmental rate with larvae coming from smaller eggs taking
longer time to reach metamorphosis than larvae from larger eggs.
3.2.2. Non-feeding larvae
3.2.2.1. Interspecific comparisons Berrill’s (1935) examination of the
effect of egg size on developmental time (measured as time to swimming)
in ascidians is perhaps the best-known example of the effects of offspring size
on developmental rate in non-feeding larvae. Species with larger eggs took
longer to develop than species with smaller eggs at the same temperature.
Interestingly, there are very few other examinations of the effects of offspring size on developmental time in this group. Staver and Strathmann
(2002) found that egg size and time to swimming appeared to positively
(although statistically non-significantly) correlated among three species of
Urochordate. In contrast, Emlet et al. (1987) found a strong, positive
correlation between egg size and planktonic period across 36 species of
non-feeding asteroid larvae.
3.2.2.2. Intraspecific comparisons There are very few examinations of
the intraspecific effects of offspring size on developmental time or planktonic period in species with non-feeding larvae. Isomura and Nishimura
(2001) found that larger larvae had longer lifetimes (measured as the time
from the free-swimming stage to when the larvae died and therefore
probably exceeds the time taken to become metamorphically competent)
than smaller larvae within three species of pocilloporid corals. More
recently, Marshall and Bolton (2007) showed that egg size strongly
affected time until hatching in two species of ascidian (Ciona intestinalis
and Phallusia obesa) and the sea urchin Heliocidaris erythrogramma. The magnitude of the effects of egg size on developmental time in this study was
surprising, with small increases in egg size dramatically increasing developmental time (e.g., for Ciona an approximately 4% increase in offspring size
resulted in a 15% increase in developmental time). This has some interesting
implications for the way in which egg size evolution is modelled.
As discussed above, Vance’s (1973a) model assumed that the effects of egg
size variation on the feeding period were much larger than its effects on
the pre-feeding period, so any increase in egg size would result in a
large decrease in the feeding period and only a small increase in the prefeeding period. Later models even removed the effects of egg size on the
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
19
pre-feeding period, regarding it as constant (e.g., Levitan, 2000, p. 178).
The effect of egg size on the pre-feeding period should not be assumed to be
substantially less than its effect on the feeding period. Given the lack of an
effect of egg size on total developmental time discussed above, we suggest
that the effects of egg size on the pre-feeding period may partially balance
the effects of egg size on the feeding period, thereby obscuring any
relationship between egg size and time to metamorphic competence. Alternatively, differences in the size at metamorphosis among species may also
be obscuring any egg size–feeding period relationship in planktotrophs
(Strathmann, 1977).
Offspring size can also affect larval behaviour. Our own work on brooding species has shown that larger larvae actually spend longer in the plankton
than smaller larvae, at least for species with brooded larvae (Marshall and
Keough, 2003). However, this is not due to differences in the developmental
time of larvae, indeed the larvae of each of the species in our study were
capable of settlement immediately (Marshall and Keough, 2003). The differences appear to be due to the fact that larger larvae are more ‘choosy’ with
respect to settlement surfaces. Hence for any given larva, the larval period is
a product of the availability of suitable substrates and the size of the larva.
The effect of larval size on the developmental time/settlement behaviour of non-feeding larvae has some interesting implications for variation
in the dispersal potential of this group. All else being equal, larger larvae will
have greater dispersal than smaller larvae. Thus, if a mother produces
offspring of variable size, then she will also produce offspring that are likely
to disperse to varying amounts. It is unclear whether this production of
larvae with variable dispersal potentials represents a ‘bet-hedging’ strategy
(Laaksonen, 2004; Raimondi and Keough, 1990) or is merely a physiological artefact (Eckelbarger, 1986). Nonetheless, it appears marine invertebrates that produce non-feeding larvae have the previously unanticipated
potential to control the dispersal of their offspring (Marshall and Keough,
2003c). Interestingly, McGinley et al. (1987) state that for a strategy of
producing offspring of variable size to be adaptive, mothers must be able
to control the dispersal of their offspring into different habitats and this
criterion may apply for this group of organisms. Further work is necessary to
determine the prevalence of the larval size–swimming time relationship, but
we believe a relationship is likely given the demonstrated post-metamorphic
cost of extended larval swimming in marine invertebrates with non-feeding
larvae (Maldonado and Young, 1999; Marshall et al., 2003b; Wendt, 1998,
2000). If larger larvae have greater nutritional resources, they may be better
able to ‘afford’ to delay their metamorphosis until a suitable settlement
surface is found. Alternatively, for the ciliated bryozoan larvae at least, larger
larvae will have a lower ciliated surface area to volume ratio and thus may
expend proportionately less energy than smaller larvae while in the plankton
(Wendt, 2000).
20
Dustin J. Marshall and Michael J. Keough
3.2.3. Direct developers
As far as we are aware, there is very little information on how offspring size
affects developmental time in direct developers. Kohn and Perron (1994)
provide an excellent compilation of the literature on Conus that show that
egg size is strongly related to the length of the pre-hatching period, pooling
across indirect and direct developers and this strongly suggests that among
these species, larger offspring take longer time to develop. In another
example for direct developers, Steer et al. (2004) found that manipulating
maternal nutrition affected offspring size in the squid Euprymna tasmanica
and that hatching success was positively correlated with offspring size.
3.2.4. Summary
Table 1.1 summarizes our findings regarding offspring-size effects on the
developmental/planktonic period. In general, the apparent lack of a clear
effect of egg size on pre-feeding developmental time or the overall planktonic period of planktotrophs was surprising given the number of modelling
studies including this as an explicit assumption. Based on the above information, it would appear premature to conclude that increasing egg size
results in a decrease in the planktonic period of species with feeding larvae,
and further work is needed to establish the relative effects of egg size on the
Table 1.1 Summary of the effects of offspring size on developmental time in
planktotrophs and non-feeding larvae
Group
Planktotrophs: pre-feeding period
Gastropods (Conus)
Copepods
Echinoids
Opisthobranchs
Planktotrophs: planktonic period
Gastropods (Conus)
Echinoids
Asteroids
Opisthobranchs
Polychaetes
Lecithotrophs
Opisthobranchs
Ascidians
Anthozoa
Bryozoa
Interspecific
Intraspecific
P (þve)
P (þve)
O
No data
O
P ( ve)
P ( ve)
O
O
O
Variable
No data
No data
No data
No data
P (þve)a
No data
No data
O
P (þve)
P (þve)
P (þve)
a
Developmental time only.
Ticks indicate an effect of offspring size and crosses indicate that no effect was detected.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
21
pre-feeding and feeding periods. In particular, more information on groups
other than the echinoids and more intraspecific comparisons are necessary if
we hope to understand the crucial relationship between egg size and
planktonic period in planktotrophs.
The effects of offspring size on developmental/planktonic period in
species with non-feeding larvae appear to be more consistent, but in this
group there were also some surprises regarding the settlement behaviour of
larvae. Offspring size may not only affect developmental time as previously
thought but it also appears to affect larval settlement behaviour (and thus
overall planktonic period) in at least two phyla. Together, these data suggest
that marine invertebrate mothers may have more ‘control’ over the dispersal
potential of their offspring than previously thought.
3.3. Post-metamorphosis
3.3.1. Planktotrophs
There have been few studies examining the post-metamorphic effects of egg
size in planktotrophs. Given that most planktotrophic larvae greatly increase
in size from fertilization through to metamorphosis and take in a large
amount of exogenous resources through feeding, it is perhaps unsurprising
that there has been little interest in post-metamorphic egg size effects in this
group. Intuitively, one would expect there to be a weaker link between
maternal provisioning and post-metamorphic performance in this group
than in other groups where the newly metamorphosed individuals rely
entirely on maternal provisioning. However, the lack of studies reduces
our confidence in any predictions about the strength of post-metamorphic
effects of size in this group. Because no study has examined the effects of
offspring size on post-metamorphic performance directly for planktotrophs,
here we examine whether egg size affects post-metamorphic/juvenile size.
Juvenile size in lecithotrophs and direct developers has strong effects on postmetamorphic performance (see Section 3.3.3) and settler size variation due
to larval feeding can have strong effects on the post-metamorphic performance of planktotrophs (Phillips, 2002). Thus, we use post-metamorphic
size as a proxy for performance to examine if egg size can affect this parameter
in planktotrophs.
Interspecific comparisons of egg size and juvenile size in planktotrophs
show variable results. Levitan (2000) found no relationship between egg size
and size at metamorphosis among 25 species of echinoids after adjusting for
phylogenetic effects. He concluded that it was reasonable to assume that size
at metamorphosis was independent of egg size in this group. Similarly,
Hadfield and Miller (1987) found no relationship between egg size and
settler size among species of opisthobranchs, and Kohn and Perron (1994)
found no relationship among species of Conus. In contrast, for calyptraeid
gastropods there is a positive relationship between egg size and settler size.
22
Dustin J. Marshall and Michael J. Keough
Collin (2003) found no relationship overall between egg size and settler
size, but we found a significant positive relationship (R2 ¼ 0.413, n ¼ 10,
P ¼ 0.045) in our analysis of Collin’s data for just the planktotrophs.
Intraspecific comparisons of egg size effects on post-metamorphic size
come exclusively from the echinoids and have also yielded variable results.
For the echinoid A. lixula, George et al. (1990) found strong effects of initial
egg size and population source on the size of metamorphosed post-larvae.
Post-larvae from large eggs from adults from a favourable site were
(mean SE) 346 mm (26.3) in diameter, whereas post-larvae from small
egg from adults from a less favourable site were 290 mm (28.9). However,
studies that utilize natural variation in egg size in planktotrophs without the
confounding effect of coming from different sites appear to be rare. Most of
the remaining comparisons are based on the fairly extreme reduction in
offspring size by halving (or quartering) egg size experimentally during
development. We have reservations regarding the relevance of this technique for examining the ecological role of natural variation in egg size in
planktotrophs, but it does serve to parameterize the potential effects of egg
size. Hart (1995) found that experimentally halving the eggs of S. droebachiensis resulted in juveniles that were significantly smaller than juveniles
from unmanipulated eggs. Although he concluded that the effects were
small, the differences in test diameters that he observed resulted in an
approximately 15% difference in volume between the two groups and
explained the bulk of the variation in juvenile size (Hart, 1995). In contrast,
similar experiments on the same species conducted by Sinervo and
McEdward (1988) found no effect of egg size manipulation on subsequent
juvenile size. Interestingly, Sinervo and McEdward (1988) found an effect
of initial egg size on developmental time (halved eggs took longer to
develop) whereas Hart (1995) did not. It may be that initial egg size can
affect either length of planktonic period or juvenile size even in the same
species. Allen et al. (2006) demonstrated that manipulating egg size had
strong effects on post-settlement size in Clypeaster rosaceus and generally,
halving egg size had a larger effect than manipulating larval food. It is
possible that egg size has more of an effect in planktotrophs with large
eggs (C. rosaceus is facultatively planktotrophic) than species with small eggs.
Alternatively, the different larval food levels could also explain variation in
the results of different studies. Regardless, it appears that in some species at
least, significant differences in initial egg size can persist through to metamorphosis despite a period of larval feeding and therefore, egg size has the
potential to affect post-metamorphic performance in planktotrophs.
3.3.2. Non-feeding larvae
Most of the evidence of post-metamorphic effects of offspring size on
post-metamorphic performance comes from species with a non-feeding
larval stage. This is probably due to the fact that a number of these studies
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
23
focus on colonial invertebrates where post-metamorphic survival and
growth are easily quantified. The majority of studies examining offspringsize effects for this group involve field studies where larvae are measured in
the laboratory, settled onto artificial substrata and then transplanted into the
field. However, the effects of larval size on post-metamorphic growth are
not restricted to colonial organisms. Emlet and Hoegh-Guldberg’s working
on the sea urchin H. erythrogramma was one of the first studies to show that,
in lecithotrophic species, the majority of maternal investment (in their case
lipid reserves) was unnecessary for larval development and was probably
for post-metamorphic performance (Emlet and Hoegh-Guldberg, 1997).
Accordingly, they found that larval lipid reductions did not affect larval
performance but had a strong effect on post-metamorphic survival and
growth in this species. Ito (1997) also showed that time until starvation in
newly metamorphosed ‘benthic larvae’ of the opisthobranch Haloa japonica
was positively correlated with initial egg size.
For the bryozoan Bugula neritina, larval size affects early post-settlement
mortality, early growth, reproduction and the quality of offspring produced
in the subsequent generation (Marshall et al., 2003a). The effects of larval
size differed between populations for this species: an effect of larval size on
survival was detected throughout the life of B. neritina colonies in southern
Australia, but these effects were more transient in Florida, United States.
We suggested that the differential effects of larval size on survival were due
to different intensities and sources of mortality between the two sites.
At Florida, the principal source of mortality appeared to be detachment during
severe storms. While survival was size-dependent initially in Florida, the sizeindependent mortality associated with storms removed any relationship
between larval size and adult survival. In contrast, the only mortality that
Australian B. neritina experienced was early mortality after settlement, and thus
overall mortality was strongly dependent on larval size. This early, sizedependent mortality of sessile invertebrates appears to be relatively common
and one of the few generalizations that can be made about the postmetamorphic effects of offspring size. Settlers coming from larger larvae
have higher initial survivorship than those from smaller larvae in at least four
different taxa (Table 1.2). For non-feeding larvae, the first time an individual
is able to feed is once metamorphosis is complete and feeding structures are
fully functioning. This is clearly an energetically expensive process and values
of 10–60% of total energetic reserves being expended during metamorphosis have been reported (reviewed in Bennett and Marshall, 2005). Given
the high energetic cost of metamorphosis, it may be that settlers from
smaller larvae are closer to their energetic minimum and if conditions are
not ideal (i.e., abundant food available immediately after settlement), then
these larvae are more likely to starve to death. Alternatively, settlers from
smaller larvae may be unable to exploit food resources as efficiently because
they can have smaller feeding structures (Marshall and Keough, 2003a, 2005).
Table 1.2
Summary of studies of post-metamorphic effects of offspring size in marine invertebrates
Species
Size range
Survival
Growth
Reproduction
Marshall et al., 2006
Marshall et al., 2003
Field
Field
3-fold
2-fold
P (0–100%)
P (0–90%)
P
P
N/A
P
Marshall et al., 2003
Field
Botrylloides violaceus
Bugula neritina
(Australia)
Bugula neritina
(Florida)
P
N/A
Marshall and Keough,
2003
Marshall and Keough,
2004a
Field
Ciona intestinalis
1.2-fold
O (early
effects
only)
P (22–65%)
N/A
N/A
Field
2.5-fold
P (47–98%)
O
N/A
O
P
N/A
O (early
effects
only)
P
N/A
1.5-fold
O (early
effects
only)
P (43–62%)a
P
P
P
P
N/A
N/A
Marshall and Keough,
2004
Field
Marshall and Keough,
2005
Field
Watersipora
subtorquata
(settlement plates)
Watersipora
subtorquata
(pier pilings)
Diplosoma listerianum
Moran and Emlet,
2001
Rivest, 1983
Emlet and HoeghGuldberg, 1997
Ito, 1997
Field
Nucella ostrina
Lab
Lab
Searlesia dira
Heliocidaris
erythrogramma
Haloa japonica
Lab
a
Calculated from Figure 4, p. 1604.
Ticks indicate a positive effect of offspring size and crosses indicate that no effect was detected.
P (time until
starvation)
N/A
Dustin J. Marshall and Michael J. Keough
Location
24
Study
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
25
Regardless, the effect of larval size on post-metamorphic survival for species
with non-feeding larvae (at least for filter feeders) appears to be mediated by
nutrition. This is supported by the fact that the effects of larval size on
survival in the ascidian C. intestinalis are exacerbated at higher settler densities when competition for food is more likely to occur (Marshall and
Keough, 2003a). Thus, we predict that when food is more abundant, the
effects of larval size on post-metamorphic survival are likely to be reduced.
Initial studies suggest that disturbance as a source of post-settlement
mortality acts in a size-independent manner. As described above, colony
survival was unrelated to initial larval size in Florida due to storms causing
high mortality of B. neritina colonies throughout the adult stage. In a study
on the encrusting bryozoan Watersipora subtorquata, we found that colony
survival was dependent on initial larval size for colonies on settlement plates
but was independent of larval size for colonies on pier pilings among natural
communities (Marshall and Keough, 2004a). Mortality was much higher for
colonies on pier pilings, and we suggested that increased mortality through
trampling or predation for settlers in this habitat was likely to act in a sizeindependent way. Overall, we propose that larval size will mediate the
nutritional aspects of post-settlement survival but will have little effect on
survival if there are high levels of (both biotic and abiotic) disturbance
(Brockelman, 1975). Generally, more work is necessary to examine explicitly how biotic factors like predation and competition affect the relationship
between larval size and post-settlement survival.
The effects of larval size appear to persist throughout so as to affect
subsequent growth in some marine invertebrates with non-feeding larvae.
For B. neritina, colonies from larger larvae have higher growth rates than
colonies from smaller larvae, and these effects can be detected for up to six
weeks post-settlement (Marshall et al., 2003). Other sources of variation in
larval quality affect B. neritina in a similar manner (Wendt, 1998), and it may be
that larval quality effects are particularly persistent in this species. For two
other species of colonial invertebrate, Diplosoma listerianum and W. subtorquata,
we found initial effects of larval size on colony growth, although these
effects were generally less persistent in both species (Marshall and Keough,
2004a, 2005). We speculate that in ‘weedy’ species, such as D. listerianum
and W. subtorquata, larval-size effects on growth can be quickly obscured
due to factors such as the availability of free space and colony fragmentation
affecting colony growth. This suggestion is partially supported by the
finding that larval-size effects on colony size are far more persistent in the
superior competitor Botrylloides violaceus (Marshall et al., 2006). The mechanism underlying the effect of larval size on post-metamorphic growth
appears to vary between species. For example, in the colonial ascidians
D. listerianum and B. violaceus, larval size positively affects the size of feeding
structures (Marshall and Keough, 2005; Marshall et al., 2006). However,
larval size can also affect the budding rate of new settlers with settlers from
26
Dustin J. Marshall and Michael J. Keough
larger larvae budding at higher rates than settlers from smaller larvae
(Marshall and Keough, 2004a; Marshall et al., 2006). Given the strong
effects of offspring size on growth in colonial invertebrates, it is perhaps
unsurprising that offspring size also strongly affects intraspecific competitive
interactions. For the colonial ascidian B. violaceus, not only were larger
larvae more likely to survive and grow faster as colonies than smaller larvae
but also they were better competitors (Marshall et al., 2006). When established colonies were present, new recruits from larger larvae were more
likely to survive than recruits from smaller larvae, and at higher settler
densities the advantages of increased offspring size were exacerbated.
Finally, when settlers were placed within proximity to one another, settlers
from smaller larvae were more likely to lose territory (without being
overgrown) to settlers derived from larger larvae.
The effect of larval size on subsequent reproduction is the most important
life-history variable to measure as this parameter gives the most relevant
measure of offspring fitness (Stearns, 1992). However, we are aware of only
one species for which the effect of larval size on subsequent reproduction has
been determined: we found that colonies from larger larvae generally have
greater reproduction than colonies from smaller larvae in B. neritina (Marshall,
2005; Marshall et al., 2003a). However, these effects vary among different
populations, apparently according to local selection pressures. For example,
for B. neritina colonies that come from populations that are highly seasonal
with mortality at the end of the summer, larval size affects the time until
reproduction with colonies from larger larvae reproducing before colonies
from smaller larvae (Marshall, 2005). In contrast, for colonies where there are
high rates of predation but colonies as a whole persist year round, larval size
affects growth rates and fecundity much more strongly (Marshall, 2005).
Although direct evidence is limited, the strong effects of larval size on postmetamorphic growth suggest that larval size will also affect reproduction in a
range of taxa (especially colonial organisms) but more tests are needed.
3.3.3. Direct developers
The direct developers as a group have received less attention with regard to
the effects of offspring size on post-metamorphic performance, perhaps
because they are mobile as juveniles and therefore harder to track than sessile
invertebrates. Despite the challenges involved, Moran and Emlet (2001)
examined the effects of offspring size on post-metamorphic performance in
the field for an intertidal, direct developing gastropod. After one month
in the field, they found that larger hatchlings were more likely to be
recovered from the field (the authors inferred greater survivorship from
recovery rates) than smaller hatchlings during the winter but during the
summer, when survival was lower overall, they found no effect of offspring
size on recovery rates. They suggested that desiccation or thermal stress was
the main source of mortality (this was supported by the fact that more snails
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
27
died on the sun-exposed experimental patch than the shaded patch), and that
this source of mortality was likely to be independent of offspring size (Moran
and Emlet, 2001). Moran and Emlet (2001) also showed that larger hatchlings had higher post-metamorphic growth in the field in all experimental
runs. This may be due to the fact that larger hatchlings are more likely to be
successful when attacking larger prey items (barnacles; Palmer, 1990). Rivest
(1983) found that hatchling size affected post-metamorphic growth rates in
the snail Searlesia dira with larger hatchlings growing faster. Interestingly, he
also found that larger hatchlings were less likely to be successfully preyed
upon by the crab Pagurus hirsutiuculus, which showed strong preferences for
smaller hatchlings. This appears to be the only study to have addressed the
effect of offspring size on predation rates in marine invertebrates.
3.3.4. Summary
Offspring size has the potential to affect post-metamorphic survival, growth
and even reproduction although evidence is limited by the paucity of studies.
Table 1.2 compares the average survivorship of individuals across the size
ranges observed for each field study. The effects of size are variable among
species with some small differences in size resulting in large differences in
survivorship (e.g., Nucella) and vice versa (e.g., Watersipora). Thus, even small
differences in maternal provisioning can result in differential survivorship,
and this raises an interesting possibility concerning planktotrophs. Given the
range of offspring sizes over which survival differences are observed in
lecithotrophs and direct developers, we believe that the degree of variation
in settler sizes was induced by egg size in planktotrophs, means that
post-metamorphic effects of egg size in the group should not be ruled out.
We propose a number of predictions with regard to offspring-size effects
on post-metamorphic performance, all of which should be relatively
straightforward to test. First, for species with non-feeding larvae that
produce offspring that are at the smaller end of the size spectrum
(e.g., most solitary ascidians), we predict that offspring size will affect
post-metamorphic survival much more strongly than for those species
with extremely large/yolky eggs or larvae that are brooded (e.g., colonial
ascidian larvae). In other words, we predict that the within-species effects of
offspring size will vary among species with different mean offspring sizes.
Species with smaller eggs are probably closer to their energetic minimum
requirements, and less well-provisioned offspring within this group probably require high-food conditions immediately after settlement to survive.
In contrast, in species with much larger offspring, larvae are probably further
from their energetic minimum requirements and therefore variation in
offspring size is likely to affect subsequent growth rates rather than survival.
At the very least, we expect that the relationship between offspring size
and survival to be more sensitive to variation in food availability in species
with relatively smaller eggs as opposed to those with larger eggs.
28
Dustin J. Marshall and Michael J. Keough
Second, we predict that different sources of mortality will be more or less
likely to act in an offspring size-dependent manner. For example, physical
disturbance and stress (e.g., salinity or heat stress) are likely to cause mortality
irrespective of offspring size whereas competition and food availability are
likely to be size-dependent. The effects of predation on the relationship
between offspring size and performance seem less clear: in some instances
offspring size will strongly affect predation rates (Palmer, 1990), but other
predators, or biological sources of mortality (e.g., bulldosing by limpets of
newly settled barnacles), seem less likely to act in a offspring size-dependent
manner. Regardless, we strongly believe that a simple division between
‘good’ and ‘bad’ environments (Einum and Fleming, 1999; McGinley
et al., 1987; Sargent et al., 1987; Stearns, 1992) is uninformative, and generalizations such as larger offspring being better in a bad environments (while
useful in some taxa, e.g., Fox, 2000; Fox and Mousseau, 1996) are unlikely to
apply to marine invertebrates. Rather than focusing on merely the intensity
of the mortality, the source of mortality should also be considered.
Finally, we predict that offspring size will have very different effects on
different post-metamorphic traits among different populations (such as
those in Marshall, 2005). Traditionally, the effects of offspring size are
viewed as being constant among different environments but a genetic
component of offspring-size effects clearly exists (Reznick, 1981). Therefore, it is reasonable to expect that offspring size will affect different traits
among different populations.
4. Sources of Variation in Offspring Size
Offspring size varies within broods (e.g., Marshall et al., 2000), among
broods from the same mother (Chester, 1996; Jones et al., 1996), among
mothers and among populations (e.g., George, 1994a; George et al., 1990).
Here, we review the sources of variation in offspring size at the within- and
among-population levels. A huge literature is devoted to various parameters
that cause variation in offspring size and to cover them all here would
be tedious and uninformative, so we have attempted to highlight some
common and major sources of variation.
4.1. Within populations
4.1.1. Stress
A variety of stresses can affect the size of offspring. One of the earliest studies
by Bayne et al. (1978) showed that salinity, temperature and food availability
can all strongly affect the size (mass) of eggs produced by Mytilus edulis.
More generally, maternal nutrition can have a strong effect on offspring size
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
29
in a variety of species (Chester, 1996; George, 1995; Meidel et al., 1999;
Qian, 1994; Qian and Chia, 1991; Steer et al., 2004), but not in others
(Lewis and Choat, 1993). Our own studies have shown that when the
bryozoan B. neritina suffers a simulated predation event, colonies reduce
the size of their offspring (and therefore offspring fitness) dramatically
(Marshall and Keough, 2004b). This may be to increase the colony’s ability
to recover (Marshall and Keough, 2004b). Generally, offspring size is often
positively correlated with maternal resource state, but the effects are variable
and more work is needed.
Other stresses can also affect offspring size but have received less attention. Gimenez and Anger (2001) found that salinity stress resulted in an
increase in egg size for the crab Chasmagnathus granulate. Cox and Ward
(2002) found strong effects of pollution on the size of larvae produced by
Montipora capitata with a remarkable 17% decrease in larval volume (assuming spherical larvae) due to exposure to increased ammonium. Studies in
terrestrial organisms suggest that pollution could act as a strong selection
pressure on offspring size (Hendrickx et al., 2003a,b). It will be interesting to
determine the effect that pollution will have on offspring size in other
species as this may represent a previously unrecognized mechanism by
which pollution can negatively affect marine populations.
4.1.2. Maternal size
Within species across a wide variety of taxa, offspring size is correlated with
maternal size (Sakai and Harada, 2001; Stearns, 1992). This correlation can
be negative, for example in Conus marmoreus, egg size is negatively correlated with maternal size (Kohn and Perron, 1994) and in B. neritina, larval
size can be positively or negatively correlated with colony size (Marshall,
2005). However, generally, if there is a correlation between offspring and
maternal size, it is positive (Table 1.3).Table 1.3 is probably not an accurate
representation of reality given that in many cases the absence of a relationship between maternal size and offspring size is unlikely to be reported.
Therefore, the percentage of species where no relationship is present is
probably dramatically underestimated. Nevertheless, it is clear that within a
range of species, larger mothers produce larger offspring. Interestingly,
comparisons among species show the opposite pattern, smaller species
tend to produce larger offspring than larger species (Emlet et al., 1987).
In non-marine species, maternal–offspring size relationships are common
and a variety of adaptive explanations have been proposed for the observed
correlations. For example, Sakai and Harada (2001) propose that if larger
mothers can provision their offspring more efficiently than smaller mothers,
then this will result in a correlation between maternal and offspring size.
Alternatively, Parker and Begon (1986) predict that if competition between
siblings is likely, then larger, more fecund mothers should produce larger
offspring to compensate for increased levels of competition. This may apply
30
Dustin J. Marshall and Michael J. Keough
Table 1.3 Summary of studies reporting a relationship between maternal size and
offspring size in marine invertebrates
Study
Species
Development
Relationship
Dugan et al.,
1991
Damiani, 2003
Emerita analoga
P
þve in 8/22 sites
Pagurus
longicarpus
Chasmagnathus
granulata
Ceratoserolis
trilobitoides
Serolis polita
Ligia oceanica
Homarus
americanus
Palaemon gravieri
P
–
P
þve
B
þve
B
?
P
–
þve
þve
P
–
Gammarus
duebeni
D
þve
Crepidula dilatata
D
þve
Buccinum
undatum
Haloa japonica
Conus spp.
(13 sp.)
D
–
L
P
D
þve
No relationship
in 11/12
species but
ve in
C. armoreus
–
P
–
L
þve
L
þve in 2/3
sites, ve
in 1/3
L
þve
Gimenez and
Anger, 2001
Clarke, 1992
Clarke, 1992
Willows, 1987
Oullet and
Plante, 2004
Kim and Hong,
2004
Dunn and
McCabe,
1995
Chaparro et al.,
1999
Valentinsson,
2002
Ito, 1997
Kohn and
Perron, 1994
Steer et al., 2004
McCarthy et al.,
2003
Bridges and
Heppell, 1996
Marshall,
2005;
Marshall et al.,
2003
Marshall et al.,
2000
Euprymna
tasmanica
Phragmatpoma
lapidosa
Streblospio
benedicti
Bugula stolonifera
Pyura stolonifera
x
31
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
Table 1.3 (continued)
Study
Species
Development
Relationship
Marshall and
Keough,
2003
Marshall and
Keough, 2003
Ciona intestinalis
L
þve
Uniophora
granifera
L
þve
P, B, L and D indicate planktotrophic, brooding, lecithotrophic and direct development, respectively.
‘þve’ indicates a positive correlation between maternal size and offspring size and ‘–’ indicates no
relationship.
to direct developers, but competition among sibling larvae in planktotrophic species seems unlikely. Hendry et al. (2001) suggest that if the
maternal phenotype influences the quality of natal environment (e.g., larger
mothers having access to the best spawning sites), then a correlation
between offspring size and maternal size should be expected. These are all
intriguing possibilities, but we currently have insufficient data to determine
if any of these models apply to marine invertebrates. Alternatively, the
maternal size–offspring size relationship may be non-adaptive and simply
be a product of anatomical scaling constraints (Fox and Czesak, 2000).
Regardless, it appears in some species, larger mothers produce larger
offspring and so not only are larger mothers contributing more offspring to
the next generation but also they are contributing offspring of the highest
quality. This has interesting implications for fisheries and population demographic models because, generally, populations with larger individuals are
more likely to supply recruits in populations that contain mostly small
individuals (Birkeland and Dayton, 2005).
4.2. Among populations
4.2.1. Habitat quality
Given the effects of maternal nutrition and stress on offspring size, it is
perhaps unsurprising that habitat quality also has a strong effect on the size
of offspring. However, as with maternal nutrition, the direction of the
effects of habitat quality on offspring size is highly variable (George, 1994,
1995; George et al., 1990). Habitat quality can vary due to an almost
endless variety of factors but some commonly reported factors include
tidal height, water depth (Bertram and Strathmann, 1998) and wave
exposure (Etter, 1989). Generally, in poorer quality/more stressful habitats, offspring size is smaller (references above in Section 4.1.1); however,
this is not always the case. It is unclear whether these responses are
32
Dustin J. Marshall and Michael J. Keough
adaptive: the lack of information emphasizes the fact that very few studies
have examined the consequences of offspring-size variation in multiple
habitats.
4.2.2. Latitudinal clines
Thorson (1935) first suggested that egg size increases from lower to higher
latitudes and initial evidence in Balanus balanoides supported this idea
(Barnes and Barnes, 1965), but it was noted that the trend was inconsistent
and appeared to be more related to winter temperatures than latitude per se.
Interest in the effects of latitude on offspring size (in crustaceans in particular)
has resumed more recently, and for some species, there is a strong latitudinal
cline in offspring size (Hadfield, 1989; Hagstrom and Lonning, 1967;
Kokita, 2003; Lardies and Castilla, 2001; Lardies and Wehrtmann, 2001;
Wagele, 1987; Wehrtmann and Kattner, 1998). Clearly, selection pressures
that act on offspring size are likely to predictably change along latitudinal clines, but we are unaware of any study that specifically addresses
the relationship between offspring size and performance and how this
changes along latitudes. We believe that such studies are warranted
given the surprising results from similar studies on seed size (Moles
et al., 2004).
5. Offspring-Size Models
In this section, we review the various models examining offspring size
in marine invertebrates, some of the fundamental knowledge gaps that are
currently slowing progress and some problems with the traditional approach
to modelling offspring size.
Given the astonishing range of offspring sizes exhibited in marine
invertebrates, it comes as no surprise that one of the first attempts using
models to understand the selection pressures operating on mothers was done
with reference to benthic marine invertebrates (Vance, 1973). Vance’s
classic work used arbitrary units to determine what size offspring maximized
maternal fitness (the number of settling larvae) with respect to a number of
parameters. While Vance’s work was the earliest work on the issue, Smith
and Fretwell’s (1974) work was probably more influential among the
broader community of ecologists. The two models share two basic features:
a trade-off between the size and number of offspring and offspring sizefitness function, with these parameters forming the basis of most of the
subsequent life-history models of offspring size. We will now examine these
basic components of offspring-size models before examining some of the
variations on the theme.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
33
5.1. Offspring size-number trade-off
This is the simplest and most conserved feature of offspring-size optimality
models, which typically assume that the number of offspring a mother can
produce is inversely proportional to the size of their offspring:
N¼
C
s
where N is the number of offspring produced, C is the total resources
available for reproduction and s the size of offspring.
Typically, this trade-off is presented as an energetic argument: mothers
have a limited amount of energy available for reproduction, so any increase
in offspring size will result in a concomitant decrease in fecundity. Note that
this argument pertains specifically to the energetic costs to the mother rather
than the energetic content of the offspring, a subtle but important difference.
Since the early models of Vance and Smith and Fretwell, there has been a
significant effort devoted to determining the relative energetic content of
large and small offspring among and within species. Across species, there
now appears to be reasonable evidence for a relationship between egg size
and energetic content in annelids and echinoderms at least ( Jaeckle, 1995;
McEdward and Miner, 2001; Pernet and Jaeckle, 2004; Wendt, 2000).
Within species, the relationship between egg size and energy content is
viewed as being more variable.Table 1.4 summarizes those studies that have
examined the relationship between egg size and energetic content within
marine invertebrate species. McEdward and colleagues have repeatedly
suggested that egg size is not a reliable indicator of maternal investment
because some regression equations required large differences in egg size to
predict a difference in energetic content (McEdward and Carson, 1987;
McEdward and Chia, 1991; McEdward and Coulter, 1987; McEdward and
Miner, 2001). We would argue that offspring size is probably a reasonable
reflection of offspring energetic content for a number of reasons. First,
the most common method by which the energetic content of eggs was
estimated in many of the studies that found no relationship between
offspring size and energy content was the dichromate oxidation technique
as modified by McEdward and Coulter (1987). This technique is now viewed
as producing unreliable results for a number of reasons (Gosselin and Qian,
1999; Pernet and Jaeckle, 2004), and so the lack of a relationship may be due
to methodological problems. Second, the lack of a significant relationship
within some species is almost certainly due to a Type II error as a result of a
lack of statistical power. The latter seems likely in some studies where only
a small number of eggs were examined per species [e.g., McEdward and Chia
(1991) used three values for two of the species in their study]. Thus, we
believe it is likely that larger offspring have a higher energetic content than
34
Table 1.4
Summary of studies examining the relationship between offspring size and energetic content
Study
Species
Method
Correlation
n of eggs/
female
McEdward and
Chia, 1991
McEdward and
Chia, 1991
McEdward and
Chia, 1991
McEdward and
Chia, 1991
McEdward and
Chia, 1991
McEdward and
Coulter, 1987
Clarke, 1992
Henricia sp.
PD
No, R ¼ 0.37
18/2
Solaster
endeca
Solaster
dawsoni
Mediaster
aequalis
Pteraster
tesselatus
Pteraster
tesselatus
Chorismus
antarticus
Notocrangon
antarticus
Eulus
gaimadrii
PD
Yes, R ¼ 0.84
20/2
PD
Yes, R ¼ 0.84
20/2
PD
10/1
Calculated from graph
PD
Yes, R ¼ 0.616, P value from 1
tailed test significant
Yes, R ¼ 0.9
20/2
Calculated from graph
PD
No, R ¼ 0.25
44/1
Elemental
carbon
Elemental
carbon
Elemental
carbon
Yes, R ¼ 0.34
?/111
Yes, R ¼ 0.51
?/29
Yes, R ¼ 0.506
?/99
Clarke, 1992
PD ¼ Potassium dichromate micro-oxidation technique.
One point highly
influential
Calculated from graph
Square root from %
variance explained
Square root from %
variance explained
Square root from %
variance explained
Dustin J. Marshall and Michael J. Keough
Clarke, 1992
Notes
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
35
smaller offspring. However, this does not necessarily mean that offspring size
perfectly represents energetic content or, more importantly, that larger
offspring cost more to produce.
Although we believe that offspring size may be a reasonable indicator
of energetic content in marine invertebrates, we suggest that this line of
research is somewhat irrelevant to the central issue of whether mothers face
a trade-off between the size and number of their offspring. The crucial
component of the offspring size-number trade-off is that larger offspring
‘cost’ more to produce than small offspring. Remarkably, there have been
no tests, to our knowledge, that have examined this or that find a trade-off
between the size and number of offspring that are produced. As a first step,
comparative analyses across species could be useful (e.g., Elgar, 1990) and
Kohn and Perron’s (1994) comparisons among Conus species certainly
suggest a trade-off. However, Stearns (1992) highlights the dangers associated with inferring trade-offs from interspecific studies, and empirical
measures of the size and number of offspring produced among different
individuals of the same species may be more informative. The lack of tests is
surprising given that there are some indications that the relative cost of
producing large and small offspring could be non-intuitive. For example,
the costs of embryo packaging are likely to be non-trivial (for direct
developers especially) and may not scale linearly with offspring size (volume). For example, Conus species with larger eggs also produce more
expensive protective capsules than species with smaller eggs (Kohn and
Perron, 1994). Thus, the relative costs of packaging larger offspring will
be different than packaging smaller offspring (assuming that one offspring is
included per package), resulting in differences in provisioning efficiency
between large and small offspring. Finally, Sakai and Harada (2001) have
suggested that if offspring are metabolizing resources as they are provisioned
(e.g., in brooding species) and the rate at which mothers can provision their
offspring is limited, then larger offspring will tend to take longer time to be
provisioned and therefore are less efficient to produce.
Overall the energetic costs of producing offspring may not scale with size
and caution should be exercised regarding this assumption. It could be
argued that the use of size-number trade-offs may still be appropriate
because although offspring size and number may not trade off because of
energetic constraints, they almost certainly will trade off because of simple
space constraints. The brood capacity of mothers to hold eggs or developing
offspring is finite if a mother produces more offspring; she may have to
produce offspring of smaller size so that they still fit within her reproductive
structures. Furthermore, while offspring size may not perfectly represent
energetic investment, it does capture some effects of offspring size that
would not be represented by energetic content alone. As shown earlier,
some offspring-size effects are essentially energetic effects; larger offspring
perform better because they have more resources (e.g., resistance to starvation).
36
Dustin J. Marshall and Michael J. Keough
However, other effects of offspring size are simply a product of the physical
effect of increased size (e.g., increased target size for fertilization, developmental time), and still others are probably composites of the two (larger
ciliated larvae may swim for longer because they have more reserves and a
lower surface area to volume ratio). Therefore, offspring size captures two
aspects of subsequent performance that energetic content alone may not,
and given the relative ease of measuring offspring size we suggest that it
remains a useful proxy. Nevertheless, more work on the energetic costs
of producing large and small offspring is clearly necessary to further resolve
this issue.
5.2. Offspring size-fitness function
This part of the offspring-size optimality models has received the most
attention while it has been revised repeatedly, a common thread remains
at its fundamentals (Table 5). Initial attempts at modelling offspring size
focused solely on planktonic survival with decreases in offspring size resulting in a longer planktonic period and thus higher overall mortality
(Christiansen and Fenchel, 1979; Vance, 1973a,b). However, later attempts
also incorporated the effects of offspring size on fertilization rates (Levitan,
1993; Podolsky and Strathmann, 1996), facultative feeding (McEdward,
1997), generation time (Havenhand, 1993) and post-metamorphic effects
of offspring size (Marshall et al., 2006). As shown in Table 1.5, most of
the models contain a planktonic mortality function and a function
linking offspring size to time spent in the plankton with the majority of
modifications of Vance’s original model occurring at one or both of these
functions.
There are some fundamental knowledge gaps that could drastically
change the predictions of each of the models and some reproductive
modes that are common in marine invertebrates that are completely
ignored. For example, although Christiansen and Fenchel (1979) included
the effects of size-dependent planktonic mortality in their model, there
have been no empirical studies examining whether larger or smaller
larvae (of the same species and age) have a greater probability of survival
[note that Levitan (2000) includes a similar component in his model].
We have shown that larger larvae are more likely to reject poor settlement
surfaces than are smaller larvae, and so are more likely to settle in higher
quality habitats (Marshall and Keough, 2003c). Thus, the benefits of
increasing offspring size could be much higher than current models
would predict. One group that has been mostly ignored in theoretical
models is direct developers. Most marine invertebrate models do not
explicitly include any phases that are relevant to this major group. More
general models have considered how offspring size might be optimally
balanced in a group with high levels of maternal protection (packaging).
37
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
Table 1.5 Summary of offspring-size optimality models that are most relevant
to marine invertebrates
Postmetamorphic
period
Study
Provisioning
Fertilization
Planktonic
period
Vance, 1973a
Christiansen
and
Fenchel,
1979
Sargent et al.,
1987
McEdward,
1997
Levitan,
1993
Levitan,
2000
Marshall
et al., 2006
O
O
O
O
C¼e
C¼e
O
O
O
O
O
C¼e
MT*
O
O
C¼1–e
aS
C¼e
MT
O
O
C¼1–e
aS
C¼e
MT**
O
O
O
O
MT
M*T
O
O
C ¼ EL(S)
C ¼ L(S)
Note that the models have been greatly simplified so that their essential structures can be compared. In
all of the models, M denotes planktonic mortality rate, T denotes time spent in the plankton and S
denotes offspring size. A cross indicates that the model does not include this life-history stage.
C represents offspring fitness. E represents pre-hatching survival: e mS, where m ¼ mortality rate while
in the egg; L(S) represents post-metamorphic performance and is a function of S; T ¼ l þ r where
l ¼ Sb and r ¼ a(1 S), where b and a are constants; M* ¼ P þ FS where P is mortality common to all
sizes and F is a larval size-specific mortality rate; T* ¼ DS þ D(1 S)/F where D ¼ time taken to
develop at the maximum rate and F ¼ food availability; T** ¼ Sfp/(S 1) þ Tfp where Sfp ¼ minimum
size where planktonic feeding is unnecessary and Tfp ¼ a minimum time in the plankton when feeding
does not occur. Note: this assumes that there is no effect of egg size on the pre-feeding period.
The model of Sargent et al. (1987) is relevant to encapsulated direct
developers: this model assumes that maternal care (packaging) increases
offspring survival, larger juveniles have greater performance and larger
offspring have longer periods until hatching. Overall, this model predicts
that for species with greater levels of maternal care, larger offspring sizes
should be favoured (Sargent et al., 1987). For marine direct developers,
there is good evidence for each of these assumptions (Kohn and Perron,
1994; Moran and Emlet, 2001; Strathmann, 1995; Strathmann and
Chaffee, 1984; Strathmann et al., 2002). We suggest that this model
could be applied to direct developers successfully as a way of modelling
offspring-size evolution in this group. However, this assumes that the costs
of egg packaging can reliably be estimated.
38
Dustin J. Marshall and Michael J. Keough
5.3. Reconciling within-clutch variation
All offspring-size optimality models predict at least two stable optima, a ‘large
offspring-size optimum’ where fitness benefits exceed fecundity costs and a
‘small offspring-size optimum’ where producing infinitely small offspring
yields an infinite fecundity (Vance, 1973a,b). The latter optimum is clearly
non-sensical and is simply a product of the functions that are used rather than
an accurate reflection of biology. Thus, if we ignore optima based on unfeasibly small offspring, in a constant environment, we should expect a single
optimum size. Models incorporating the effects of maternal phenotype on the
natal environment have become more common (Hendry et al., 2001; Parker
and Begon, 1986; Sakai and Harada, 2001). For example, Sakai and Harada
(2001) predict that if larger mothers can provision their offspring more
efficiently than smaller mothers, then maternal size and offspring size should
be correlated. Further, in species of fish where the maternal phenotype has
the potential to affect the offspring size–fitness relationship, offspring-size
variation within populations is higher (Einum and Fleming, 2004a). These
models predict the observed variation in offspring sizes among different
mothers and initial empirical evidence is supportive (see Section 4.1.2).
While a substantial theory base can now account for variation in offspring
size among mothers, explaining the variation in offspring sizes from the same
mother has been more problematic. While there are numerous verbal arguments for producing a brood of offspring that vary in size (e.g., Capinera,
1979; Crump, 1981; Dziminski and Roberts, 2005; Lips, 2001), the few
theoretical considerations of intra-brood offspring-size variation struggle to
find an adaptive basis for this variation (e.g., Einum and Fleming, 2004b;
McGinley et al., 1987). In most instances, producing offspring of identical size
has the greatest advantage or if producing variable offspring is advantageous,
it is only under restrictive and unlikely assumptions. For example, McGinley
et al. (1987) found that producing offspring of variable size was advantageous
only when mothers could strictly control the dispersal of their offspring into
the appropriate habitat. Rather than having an adaptive basis, intra-brood
variation is increasingly viewed as a product of physiological or genetic
constraints that prevent mothers from producing offspring of identical size
(Einum and Fleming, 2004b; Fox and Czesak, 2000). In their review of
offspring-size effects on insects, Fox and Czesak (2000; p. 358) concluded
that ‘. . . some authors have suggested that at least some of the variation within
families is an adaptive response to living in a variable environment. At this
time however, there are few experimental studies and too little theoretical
work to generalize’. Therefore, despite the intuitive appeal of intra-clutch
variation in offspring size as a mechanism for coping with environmental
heterogeneity, theoretical evidence for the concept remains elusive.
We suggest that the ubiquitous variation in offspring size seen within
clutches does not solely occur due to constraints on producing offspring of
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
39
uniform size and may yet have an adaptive explanation. We believe that the
lack of theoretical evidence for adaptive within-clutch variation reflects the
modelling approaches that have traditionally been used; optimality models,
by definition, predict a single offspring size to maximize fitness (ignoring the
nonsensical, minimum optimum). Alternative approaches such as game
theoretic models (Geritz, 1995; Geritz et al., 1999) may provide better tools
for exploring adaptive variation in offspring size within clutches whereas
more traditional, optimality models may still be useful for exploring variation
among mothers, populations and species.
5.4. Summary of offspring-size models
One point this chapter has hopefully made clear is that variation in offspring
size can have effects on performance in every life-history stage. Accordingly,
theoretical models need to reflect the pervasive nature of offspring-size
effects across life-history stages. Ideally, a realistic optimality model should
contain the influence of maternal nutritional state, and the relationships
between offspring size and fertilization success, planktonic survival, settlement choice and post-metamorphic performance. Once this model has
been constructed, then perturbations of the environment in each of these
conditions and the relative importance of each life-history stage can be
assessed. We suggest that this is where the real value of optimality modelling
lies, that is, as tools for examining the relative contribution of each lifehistory stage to the selective pressures acting on offspring size and for
making predictions about how the variation in environmental conditions
will influence optimal offspring size. In contrast, these models have been
used inappropriately to explain interspecific patterns in offspring-size distributions (e.g., Sewell and Young, 1997; Vance, 1973a,b), and a great deal
of effort has been expended to try and match optimality model predictions
with the observed distribution of offspring sizes among species (McEdward
and Miner, 2006). Such comparisons are inappropriate given that, depending on the relationship between offspring and performance (which of course
will vary greatly among species), a very different optimal offspring size will be
predicted for each individual species.
6. Summary
Figure 1.7A–C summarize our view of the effects of maternal nutrition on each of the life-history stages for the three broad developmental
modes, highlighting the major unknowns for each group and stage. Size of
arrow represents our view of the relative strength of offspring-size effects
between each stage. The relative strengths also vary with developmental
mode.
40
Dustin J. Marshall and Michael J. Keough
6.1. Planktotrophs
Despite maternal nutrition constituting a lesser proportion of total larval
nutrition in planktotrophs than in other developmental modes, there
are strong effects of offspring size on multiple life-history stages of
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
41
Figure 1.7 Summaries of the effects of offspring size on the various life-history stages
of (A) a‘typical’planktotrophic species with external fertilization, (B) a‘typical’ lecithotrophic species with internal fertilization and (C) a‘typical’species with direct development. Arrows indicate that we believe there is strong effect of offspring size on that
particular life-history stage and question marks indicate that the effects at this stage
have not been examined.
planktotrophs. Egg size has strong effects on fertilization in free-spawning
species; larger eggs are more likely to be contacted by sperm than smaller
eggs (although this may be mitigated by egg accessory features). In many
taxa, larvae from larger eggs also spend less time in the plankton than smaller
eggs because they appear to require fewer resources before reaching metamorphic competence, although this finding is not universal and more tests
are necessary, particularly at the intraspecific level and on ‘non-echinoids’.
There may be synergistic effects between larvae from larger eggs having
greater feeding capacities and more initial resources. The most likely mechanism is increased feeding capacity in larger larvae, leading to higher growth
rates. However, there are currently so few data on intraspecific effects of egg
size that it is difficult to generalize. Similarly, there are too few data to
make definitive conclusions about post-metamorphic effects in planktotrophs but we suggest that post-metamorphic effects cannot be ruled out,
because several inter- and intraspecific studies indicate that larger eggs do
42
Dustin J. Marshall and Michael J. Keough
become larger settlers. Thus, despite its intuitive appeal, the notion that
exogenous, larval nutrition overrides any effect of maternal provisioning
may be incorrect for planktotrophs.
6.2. Non-feeding
Perhaps unsurprisingly, this group shows the strongest effects of offspring
size. Egg size affects fertilization success, developmental time before metamorphic competency in some groups, larval settlement behaviour, maximum larval life span and all elements of post-metamorphic performance,
including reproduction and offspring provisioning in the subsequent generation. However, most of the examples come from our own work on
colonial invertebrates, and more work on unitary species is necessary to
determine the generality of the patterns of offspring-size effects in this
group. Non-feeding taxa as a whole are of particular interest because,
presumably, larval size affects post-metamorphic performance because of
differing levels of larval energetic reserves at metamorphosis. Because larvae
do not feed in this group and extended swimming has post-metamorphic
consequences (Marshall et al., 2003b; Pechenik et al., 1998), an interesting
trade-off exists with regard to whether larger larvae should ‘use’ their extra
resources for swimming and maximize their chances of encountering an
optimal habitat, or for settling immediately and utilizing these extra
resources for enhanced juvenile/adult performance. It would be interesting
to determine the relative impact of larval swimming and larval size on postmetamorphic performance and how these two larval nutrition factors
interact.
6.3. Direct developers
We suspect that offspring size–performance relationships in the field are
likely to be strongest and most consistent in direct developers because there
is no larval stage that can affect (through either extended swimming or larval
feeding) the relationship between juvenile energy reserves and original
maternal provisioning. This is also the group where the highest level of
within-population variation in offspring size was observed across species.
Interestingly, numerous different pathways for increasing per offspring
maternal investment have evolved in species with this developmental
mode, including intragonadal sibling cannibalism (Byrne, 2006), nurse
eggs (Spight, 1976) and even partial predation on maternal body parts
(Emlet, personal communication). Some reproductive strategies will result
in mothers having more ‘control’ over the provisioning of offspring than
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
43
others, for example nurse eggs versus cannibalism. We would expect
mothers with more control of the provisioning of their individual offspring
to be more likely to be able to adaptively adjust the size of their offspring
according to local conditions. Given that direct developers tend to disperse
less than other groups and the post-metamorphic effects of offspring size can
be so strong in this group, we suggest that local population dynamics are
particularly susceptible to variation in offspring quality in this group but this
requires testing.
6.4. Ecological implications
What are the consequences of offspring-size variation for recruitment and
subsequent population dynamics? Consider a broadcast spawning marine
invertebrate with short-lived, lecithotrophic larvae. Clearly, offspring size
will affect the number of eggs that are produced, the percentage of eggs
that are fertilized, the time that the larvae spend in the plankton, the
microhabitats in which the larvae settle and the number of those settled
larvae that will survive to reproduction. Hence, offspring-size variation
can be viewed to act as a ‘filter’ on the number of individuals that pass
through each stage. If offspring sizes are relatively large, then (all else
being equal) more of the eggs will become successful recruits than if
offspring sizes are relatively smaller. In other words, the strength of the
link between adult reproductive output and subsequent recruitment will
be strongly mediated by offspring size. Furthermore, the strength of links
among populations will be strongly affected by offspring-size variation
through two mechanisms: first, by affecting the dispersal potential of
offspring and second, by affecting the chances of those offspring surviving
in the new habitat. Offspring-size effects also change the way in which we
view source and sink populations; a population that produces very few,
large (high quality) offspring may actually contribute more to the recruitment pool than a population that produces many, low quality offspring.
It is particularly interesting that maternal phenotype or habitat quality can
strongly affect the size of offspring that are produced, suggesting recruitment is coupled not only to offspring quality and quantity but also to
broodstock quality.
6.5. Evolutionary implications
In this chapter, we have found strong evidence for offspring size affecting
multiple life-history stages. For species with planktonic larvae, this comprises at least three life-history stages (gamete, larvae and juvenile) and
44
Dustin J. Marshall and Michael J. Keough
three different habitats (maternal habitat, plankton and juvenile habitat)
that can be widely separated in space and time. Consequently, it presents a
remarkable challenge to marine invertebrate mothers to optimally provision their offspring, especially since there could be conflicting selection
pressures on offspring size at each life-history stage. For example, an adult
C. intestinalis could be at very high densities and so during spawning, its
eggs could be more likely to suffer polyspermy, thereby selecting for
smaller eggs. However, smaller eggs are less likely to disperse from this
high-competition environment, and less likely to perform well under a
higher competitive regime than larger eggs (Marshall and Keough, 2003a,b).
The way in which marine invertebrate mothers balance these selection
pressures and which life-history stage gets ‘priority’ as a selection pressure
remains an intriguing unknown. In other organisms, offspring size is
regarded as a highly plastic, adaptive maternal effect (Fox and Czesak,
2000; Fox et al., 1997). Given the complex ‘web of selection’ acting on
marine invertebrate mothers, it will be interesting to determine whether
mothers can adaptively adjust the size of their offspring according to local
environmental conditions. Our review indicates that direct developers as
the most appropriate group for the first examination of this issue as the
offspring are less dispersive, and there are few life-history stages in which
offspring size can affect performance.
The strong offspring-size effects on fertilization and subsequent performance also have some interesting implications for sexual selection in broadcast spawners. First, it seems that in this group, males will determine the
ultimate size of offspring while females determine the range of sizes that can
be fertilized. This is because males can control the sperm environment in
which eggs are fertilized and ultimately zygote size due to size-dependent
fertilization effects (Marshall et al., 2002). Thus, there is the potential for
sexual conflict, with females getting the greatest fitness benefits from all of
their eggs being fertilized whereas males may get greater benefits if they only
release small amounts of sperm, ‘prudently’ fertilizing only the largest eggs
that will have the greatest performance (for discussion of ‘prudent’ males
with regards to sperm release see Wedell et al., 2002). Egg size-dependent
fertilization and performance also has some consequences for males: if a
male accesses eggs after they have been exposed to another male, then he
can only fertilize the ‘left over’ eggs, which may be eggs that were smaller
and less likely to be fertilized by the first male. Therefore, by accessing eggs
second, a male suffers two fitness costs: fertilization of fewer eggs (because
the remaining eggs are smaller and harder to ‘hit’) and siring offspring that
will have lower performance (Marshall et al., 2004a). Consequently, males
will be under strong sexual selection pressure to access a female’s brood
before any other males.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
45
6.6. Future research directions
The study of post-metamorphic effects of offspring size in marine invertebrates has only just begun and so more work generally is needed. However,
we feel there are some priority areas that require addressing. The most
commonly cited mechanism for the advantage of increased offspring size is
nutritional: larger larvae are more resistant to starvation as settlers than
settlers from smaller larvae. There is currently little direct evidence
to support this hypothesis. Simple experiments examining the relative
importance of offspring size under high- and low-food regimes could
address this important gap in our knowledge. Apart from preliminary
laboratory studies, there are few studies examining how offspring size affects
predation rates. Plant ecologists have been addressing the analogous problem in plants for many years and have used caging studies to examine the
interaction between herbivory and optimal seed size (Stanton, 1984, 1985).
These experiments could be easily transferred to sessile invertebrates.
The final priority area that we feel requires urgent attention is examination of offspring-size effects in more diverse habitats. Thus far, postmetamorphic offspring-size effects in marine invertebrates have focused
entirely upon temperate, hard substrate environments (benthic subtidal
and rocky intertidal), whereas soft sediment environments (intertidal and
subtidal) and coral reefs have received no attention whatsoever. Clearly, this
important gap in our understanding of offspring-size effects requires addressing and especially given that offspring sizes show similarly high levels of
variation for organisms in these environments when compared to those that
have been better studied.
Throughout the discussion of post-metamorphic offspring-size effects,
we have emphasized that among different environments or conditions, the
effects of offspring size are highly variable. This is a crucial point because
ultimately, for any particular species the goal is to determine the relative
importance of offspring size. Focusing on any single set of conditions will
have limited applicability. Given that the effects of offspring size are so
sensitive to local conditions, we believe any study examining offspring-size
effects should (1) examine offspring-size effects under as realistic conditions
as possible and (2) examine how these effects vary across a range of ecologically relevant stresses. While laboratory studies are essential for the initial
establishment of any offspring-size effects (particularly for mobile species),
the importance of these effects cannot be estimated until they have been
examined in a field context. Numerous studies have shown that laboratory
examinations of offspring size will produce misleading conclusions about
the strengths of these effects (e.g., Einum and Fleming, 1999; Fox, 2000).
We also suggest that further studies on offspring size be conducted at
46
Dustin J. Marshall and Michael J. Keough
multiple sites with offspring from several populations. By doing so, two
issues can be addressed simultaneously. First, the generality of the offspringsize effects can be determined and second, by using several populations, one
avoids the potentially erroneous assumption that there is no populationlevel variation in the relationship between offspring size and performance
(Marshall, 2005). Finally, the effects of offspring size should be examined
over as much of an organism’s life history as possible so as to gather more
accurate measures of fitness. Obviously the above suggestions greatly
increase the workload of anyone wishing to examine offspring-size effects
and, as an example, thus far our own studies have neglected to include all of
the above components. However, we believe strongly that the sensitivity of
offspring-size effects and their apparent variability among populations lead a
risk of highly over- or underestimating the effects of offspring size.
To date, offspring-size effect studies have largely been limited to one or
two populations but, given the effects of offspring size on postmetamorphic survival, and the levels of among-population variation in
offspring size, it would be interesting to determine whether offspring-size
effects ‘scale up’ to the level of populations. For example, do ‘source’
populations produce offspring of increased size relative to ‘sink’ populations. Of particular interest is the examination of how maternal stresses at
the level of populations (such as anthropogenic factors like pollution and
nutrient enrichment) affect maternal provisioning. In terrestrial systems, it is
becoming clear that like many other traits, offspring size is under novel
selection pressure from human influences (Hendrickx et al., 2003a,b). Given
the likely consequences of offspring-size variation for recruitment in marine
invertebrates and the initial strong effects of pollutants on maternal provisioning (Cox and Ward, 2002), there may be strong underlying impacts of
pollutants that are going undetected.
With regard to the next steps in theoretical studies, we suggest the need
for more integrative models that take the effects of offspring size on every
life-history stage into account, from the production of gametes through to
reproductive maturity of the offspring. Models that examine the influence
of recruit quality and its consequences for population connectivity would
also greatly enhance our ability to determine the relative importance of
larval quality and quantity.
Appendix
Variation in offspring size in marine invertebrates with direct (D),
lecithotrophic (L) and planktotrophic (P) development and internal (I) or
external (E) fertilization. Offspring sizes are given as diameters and CVs are
calculated as total variation.
Phylum,
Class
Cnidaria,
Anthozoa
Cnidaria,
Hydrozoa
Platyhelminthes,
Turbellaria
Annellida, Polycheata
Mollusca, Bivalvia
Mollusca, Gastropoda
Order
Family
Species
Study
Development
Fertilization
Eggs/
mother
Number of
mothers
Offspring
size
CV
Scleractinia
Acroporidae
Acropora spathulata
Baird et al., 2001
L
E
32
6–8
557
9.33
Scleractinia
Scleractinia
Scleractinia
Scleractinia
Scleractinia
Scleractinia
Scleractinia
Scleractinia
Helioporacea
Hydroida
Acroporidae
Acroporidae
Acroporidae
Faviidae
Faviidae
Agariciidae
Acroporidae
Pocillporidae
Helioporidae
Tubulariidae
Baird et al., 2001
Baird et al., 2001
Baird et al., 2001
Baird et al., 2001
Baird et al., 2001
Baird et al., 2001
Baird et al., 2001
Harii et al., 2002
Harii et al., 2002
Yamashita et al., 2003
L
L
L
L
L
L
L
L
L
L
E
E
E
E
E
E
E
I
I
I
18
32
49
20
38
30
39
20
25
30
6–8
6–8
6–8
6–8
6–8
6–8
6–8
5
10
10
553
541
538
401
371
368
337
1000
3700
305
4.70
5.91
3.38
8.72
5.66
5.16
12.16
20
10.81
10.88
Polycladida
Stylochidae
Acropora hyacinthus
Acropora millepora
Astreopora myriophthalma
Favites halicora
Goniastrea retiformis
Pachyseris speciosa
Montipora digitata
Pocillopora damicornis
Heliopora coerulea
Tubularia
mesenbryanthemum
Stylochus ellipticus
Chintala and Kennedy, 1993
P
I
50
?
68.5
4.52
Sabellida
Sabellida
Sabellida
Sabellida
Sabellida
Sabellida
Terebellida
Pteroida
Pteroida
Mytiloida
Veneroida
Ostreina
Veneroida
Archaeogastropoda
Neogastropoda
Neogastropoda
Neogastropoda
Mesogastropoda
Mesogastropoda
Mesogastropoda
Mesogastropoda
Neogastropoda
Mesogastropoda
Mesogastropoda
Serpulidae
Spirorbidae
Spirorbidae
Spirorbidae
Spirorbidae
Spirorbidae
Sabellariidae
Pectinidae
Pectinidae
Mytilidae
Tellinidae
Ostreidae
Tridacnidae
Trochidae
Buccinidae
Columbellidae
Columbellidae
Cypraeidae
Strombidae
Strombidae
Strombidae
Muricidae
Vermetidae
Vermetidae
Hydroides dianthus
Bushiella abnormis
Circeis armoricana
Paradexiospira vitrea
Pileolaria berkelyana
Protolaeospira exima
Phragmatopoma lapidosa
Chlamys bifrons
Chlamys asperrima
Brachidontes virgiliae
Macoma mitchelli
Ostrea edulis
Tridacna squamosa
Cantharidus callichroa
Engoniophos unicinctus
Strombina francesae
Strombina pumilio
Cypraea caputdraconis
Strombus gigas
Strombus costatus
Strombus raninus
Drupella cornus
Vermetus sp.
Dendropoma corrodens
Toonen and Pawlik, 2001
Hess, 1993
Hess, 1993
Hess, 1993
Hess, 1993
Hess, 1993
McCarthy et al., 2003
Styan and Butler, 2000
Styan and Butler, 2000
Bernard et al., 1988
Kennedy and Lutz, 1989
Jonsson et al., 1999
Fitt and Trench, 1981
Ho Sun and Hong, 1994
Miloslavich and Penchazadeh, 1994
Cipriani and Penchazadeh, 1993
Cipriani and Penchazadeh, 1993
Osorio et al., 1992
Davis et al., 1993
Davis et al., 1993
Davis et al., 1993
Turner, 1992
Miloslavich and Penchazadeh, 1992
Miloslavich and Penchazadeh, 1992
P
L
L
L
L
L
P
P
P
L
P
P
P
L
D
D
D
P
P
P
P
P
L
D
E
I
I
I
I
I
E
E
E
I
E
I
E
I
?
?
?
?
?
?
?
10
10
300
25
40
10
30
?
20
11
125
20
30
20
200
33
134
15
9
11
11
5
7
20
8
6
?
?
120
?
?
49
1
?
11
3
3
9
7
49
27
60.7
185
167
196
169
199
90.4
116.5
71.2
383
59
202
158
446
1007.5
571
947
112
225
262
140
170
240
512
8.23
21.62
10.17
9.18
7.69
10.55
4.09
2.66
5.67
13.31
3.89
5.94
4.43
10.76
25.60
6.12
10.24
5.1
7.56
2.29
2.85
1.47
5.83
11.52
I
I
I
I
I
I
(continued)
Appendix (continued)
Phylum,
Class
Mollusca,
Opisthobranchia
Mollusca, Celphalopoda
Crustacea, Malacostraca
Crustacea, Copepoda
Crustacea, Maxillopoda
Eggs/
mother
Number of
mothers
Offspring
size
CV
28
74
?
2
16
35
149
1450
425.7
10.06
5.51
5.40
E
I
I
I
?
?
?
?
40
5
?
40
??
309
69
204
?
?
100
15
15
15
?
?
?
?
?
?
?
5
23?
14
18
33?
?
?
9
1
1
30
1490
1131
1330
671
756
235.4
2200
1270
74
325.8
720
1520
218
234
300
151
216
90
18.12
8.71
13.23
8.64
10.73
2.33
25.71
14.40
2.17
6.59
17.12
19.078
3.66
7.86
3.12
5.03
3.425
3.33
P
P
D
P
P
I
I
?
I
I
12
?
?
?
?
18
15–30
5
12
17
168
84.5
4800
567
731
4.57
4.52
13.12
2.64
3.83
P
P
P
P
P
P
I
I
I
I
I
I
?
290–497
?
?
?
?
49
20
?
?
16
10
410
62.4
283
565
190.9
211.7
6.09
0.70
4.94
5.3
4.6
3.77
Order
Family
Species
Study
Development
Fertilization
Mesogastropoda
Mesogastropoda
Neogastropoda
Cassidae
Vermetidae
Buccinidae
Cypraecassis testiculus
Petaloconchus montereyensis
Babylonia areolata
P
D
P
I
Neogastropoda
Neogastropoda
Neogastropoda
Neogastropoda
Mesogastropoda
Mesogastropoda
Mesogastropoda
Neogastropoda
Stylommatophora
Mesogastropoda
Mesogastropoda
Neogastropoda
Mesogastropoda
Mesogastropoda
Archaeogastropoda
Mesogastropoda
Mesogastropoda
Cephalaspidea
Buccinidae
Muricidae
Muricidae
Muricidae
Vermetidae
Naticidae
Calyptraeidae
Muricidae
Odostomiidae
Calyptraeidae
Calyptraeidae
Buccinidae
Calyptraeidae
Calyptraeidae
Trochidae
Cymatiidae
Cymatiidae
Atyidae
Searlesia dira
Nucella crassilabrum
Thais emarginata
Acanthina spirata
Dendropoma petraeum
Polinices lewisii
Crepidula adunca
Nucella lapillus
Odostomia columbiana
Crucibulum quirqinae
Crucibulum quirqinae
Buccinum cyaneum
Crepidula dilatata
Crepidula dilatata
Calliostoma zizyphinum
Cymatium cutaceum
Cymatium corrugatum
Haminoea vesicula
Hughes and Hughes, 1987
Hadfield, 1989
Chaitanawisuti and Kritsanapuntu,
1997
Rivest, 1983
Gallardo, 1979
Spight, 1976
Spight, 1976
Calvo et al., 1998
Pedersen and Page, 2000
Collin, 2000
Etter, 1989
Collin and Wise, 1997
Veliz et al., 2001
Veliz et al., 2001
Miloslavich and Dufresne, 1994
Gallardo, 1977
Gallardo, 1977
Holmes, 1997
Ramon, 1991
Ramon, 1991
Gibson and Chia, 1989
D
D
D
D
D
P
D
D
P
P
D
D
P
D
L
P
P
P
Nudibranchia
Cephalaspidea
Teuthida
Decapoda
Decapoda
Doridae
Bullidae
Loliginidae
Geryonidae
Geryonidae
Jones et al., 1996
Farfan and Ramirez, 1988
Steer et al., 2003
Hines, 1988
Hines, 1988
Decapoda
Harpacticoida
Thoracia
Thoracia
Thoracia
Thoracia
Paguridae
Harpacticidae
Balanidae
Verrucidae
Chamalidae
Chthamalidae
Aldaria proxima
Bulla gouldiana
Sepioteuthis australis
Geryon (Chaceon) fenneri
Geryon (Chaceon)
quinquedens
Pagurus longicarpus
Euterpina acutifrons
Balanus balanoides
Verruca stroemia
Chthamalus dentatus
Octomeris angulosa
Damiani, 2003
Guisande et al., 1996
Barnes and Barnes, 1965
Barnes, 1953
Achituv and Wortzlavski, 1983
Achituv and Wortzlavski, 1983
I
I
I
I
I
Bryozoa, Gymnolaemata
Cheilostomata
Cheilostomata
Cheilostomata
Cheilostomata
Cheilostomata
Echinodermata, Echinoida Clypeasteroida
Arbacioida
Cidaroida
Clypeasteroida
Clypeasteroida
Echinodermata, Asteroida Platyasterida
Platyasterida
Spinulosida
Echinodermata,
Ophiuroida
Echinordermata,
Holothuroida
Chordata, Ascidia
Bugulidae
Bugulidae
Bugulidae
Bugulidae
Watersiporidae
Dendrasteridae
Arbaciidae
Cidaridae
Clypeasteridae
Clypeasteridae
Luidiidae
Luidiidae
Echinoasteridae
Bugula stolonifera
Bugula simplex
Bugula turrita
Bugula neritina
Watersipora subtorquata
Dendraster excentricus
Arbacia lixula
Phyllacanthus imperialis
Clypeaster rosaceus
Clypeaster subdepressus
Luidia maculata
Luidia foliolata
Echinaster morph 1
Wendt, 2000
Wendt, 2000
Wendt, 2000
Marshall et al., 2003
Marshall and Keough, 2004
Podolsky, 2002
George et al., 1990
Olson et al., 1993
Emlet, 1986
Emlet, 1986
Komatsu et al., 1994
George, 1994
Scheibling and Lawrence, 1982
L
L
L
L
L
P
P
L
FP
P
P
P
L
I
I
I
I
I
E
E
E
E
E
E
E
E
?
?
?
?
?
10
100
10
25
25
?
20
9
E
E
E
60
17
?
30
30
35
?
105
10
37
400
?
?
?
?
?
6
10
20
?
?
?
1
10 12
3
10 12
3
?
?
?
5
4
7
75
1
10
?
16
Spinulosida
Echinoasteridae
Echinaster morph 2
Scheibling and Lawrence, 1982
D
E
12
Platyasterida
Paxillosida
Forcipulata
Forcipulata
Spinulosida
Spinulosida
Spinulosida
Forcipulata
Spinulosida
Spinulosida
Phrynophiurida
Luidiidae
Astropectentidae
Asteriidae
Asteriidae
Poraniidae
Poraniidae
Pterasteridae
Asteriidae
Asterinidae
Asterinidae
Asteroschematidae
Luidia quinaria
Astropecten gisselbrechti
Pisaster brevispinus
Pisaster ochraceus
Porania antarctica
Porania sp.
Pteraster militaris
Diplasterias brucei
Patiriella regularis
Asterina minor
Astrobrachion constrictum
Komatsu et al., 1982
Komatsu and Nojima, 1985
Fraser et al., 1981
Fraser et al., 1981
Bosch, 1989
Bosch, 1989
McClary and Mladenov, 1990
Bosch and Pearse, 1990
Byrne, 1991
Komatsu et al., 1979
Stewart and Mladenov, 1994
P
L
P
P
P
L
D
D
P
L
L
E
E
E
E
E
E
Dendrochirotida
Psolidae
Psolus chitonoides
McEuen and Chia, 1991
Dendrochirotida
Aspidochirotida
Apodida
Stolidobranchia
Aplousobranchia
Phlebobranchia
Stolidobranchia
Stolidobranchia
Psolidae
Holothuriidae
Synaptidae
Pyuridae
Didemnidae
Cionidae
Pyuridae
Styelidae
Psolidium bullatum
Holothuria scabra
Leptosynapta clarki
Pyura stolonifera
Diplosoma listerianum
Ciona intestinalis
Pyura fissa
Styela plicata
McEuen and Chia, 1991
Ramfafia et al., 2000
Sewell, 1994
L
E
10
L
P
D
L
L
L
L
L
E
E
10
?
100
100
10
100
50
50
E
I
E
E
E
160
207
202
271
323.18
129
76.6
507
280.3
152.6
173
144.3
840
7.90
19.43
6.64
6.90
11
3.5
4.30
6.29
2.74
2.29
3.46
4.78
4.76
960
5.2
124
353
165
163
548
554
2171
3000
197
437
415
4.91
5.09
3.33
3.68
9.23
17.08
25.10
20
1.92
6.86
13.49
5
627
5.58
2
?
6
34
12
20
10
10
330
157
2000
269
976
145
175.7832
163
5.15
2.27
52
9.18
9.32
5.17
5.21
7.9
50
Dustin J. Marshall and Michael J. Keough
ACKNOWLEDGEMENTS
We thank Craig Young for inviting this chapter. Maria Byrne kindly provided micrographs
for Fig. 1.1. The final fig. was generously provided by Adrian McMahon. This chapter
benefited greatly from discussions on offspring size and larval biology with Toby Bolton and
Richard Emlet and thorough reviewing from Richard Strathmann, David Sims and an
anonymous reviewer. Australian Research Council grants supported D.J.M. and M.J.K.
during the preparation of this chapter.
REFERENCES
Achituv, Y. and Wortzlavski, A. (1983). Studies in the biochemistry of cirripede
eggs. VII. Changes in the general biochemical composition during development of
Chthamalus dentatus Krauss and Octomeris angulosa Sowerby. J. Exp. Mar. Biol. Ecol. 69,
137–144.
Allen, J. D., Zakas, C. and Podolsky, R. D. (2006). Effects of egg size reduction and larval
feeding on juvenile quality for a species with facultative-feeding development. J. Exp.
Mar. Biol. Ecol. 331, 186–197.
Babcock, R. C., Mundy, C. N. and Whitehead, D. (1994). Sperm diffusion model and in
situ confirmation of long-distance fertilisation in the free-spawning asteroid Acanthaster
planci. Biol. Bull. 186, 17–28.
Bagenal, T. B. (1969). Relationship between egg size and fry survival in brown trout Salmo
trutta L. J. Fish Biol. 1, 349–353.
Baird, A. H., Pratchett, M. S., Gibson, D. J., Koziumi, N. and Marquis, C. P. (2001).
Variable palatability of coral eggs to a planktivorous fish. Mar. Freshwater Res. 52,
865–868.
Barnes, H. (1953). Size variations in the cyprids of some common barnacles. J. Mar. Biol.
Assoc. UK 32, 297–304.
Barnes, H. and Barnes, M. (1965). Egg size, nauplius, and their variations with
local, geographical and specific factors in some common cirripedes. J. Anim. Ecol. 34,
391–402.
Bayne, B. L., Holland, D. L., Moore, M. N., Lowe, D. M. and Widdows, J. (1978). Further
studies on the effects of stress in the adult on the eggs of Mytilus edulis. J. Mar. Biol. Assoc.
UK 58, 825–841.
Bennett, C. E. and Marshall, D. J. (2005). The relative energetic costs of the larval period,
larval swimming and metamorphosis for the ascidian Diplosoma listerianum. Mar. Freshwater Behav. Physiol. 38, 21–29.
Bernard, R. T. F., Davies, B. R. and Hodgson, A. T. (1988). Reproduction in a brackishwater Mytilid: Gametogenesis and embryonic development. Veliger 30, 278–290.
Bernardo, J. (1996). The particular maternal effect of propagule size, especially egg size:
Patterns models, quality of evidence and interpretations. Am. Zool. 36, 216–236.
Berrill, N. J. (1935). Studies in tunicate development III. Differential retardation and
acceleration. Philos. Trans. R. Soc. London Ser. B 225, 255–326.
Bertram, D. F. and Strathmann, R. R. (1998). Effect of maternal and larval nutrition on
growth and form of planktotrophic larvae. Ecology 79, 315–327.
Birkeland, C. and Dayton, P. K. (2005). The importance in fishery management of leaving
the big ones. Trends Ecol. Evol. 20, 356–358.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
51
Bolton, T. F. and Thomas, F. I. M. (2000). Maternal energy investment in eggs and jelly
coats surrounding eggs of the echinoid Arbacia punctulata. Biol. Bull. 199, 1–8.
Bosch, I. (1989). Contrasting modes of reproduction in two antartic asteroids of the genus
Porania, with a description of unusual feeding and non-feeding larval types. Biol. Bull.
177, 77–82.
Bosch, I. and Pearse, J. S. (1990). Developmental types of shallow-water asteroids of
McMurdo Sound, Antartica. Mar. Biol. 104, 41–46.
Brawley, S. H. (1992). Fertilisation in natural populations of the dioecious brown alga Fucus
ceranoides and the importance of the polyspermy block. Mar. Biol. 113, 145–157.
Bridges, T. S. and Heppell, S. (1996). Fitness consequences of maternal effects in Streblospio
benedicti (Annelida: Polychaeta). Am. Zool. 36, 132–146.
Brockelman, W. Y. (1975). Competition, fitness of offspring, and optimal clutch size. Am.
Nat. 109, 677–699.
Byrne, M. (1991). Embryogeneisis and larval development of the asteroid Patiriella regularis
viewed by light and scanning electron microscopy. Biol. Bull. 180, 332–345.
Byrne, M. (2006). Life-history diversity and evolution in the Asterinidae. Integr. Comp. Biol.
46, 243–254.
Byrne, M., Hart, M. W., Cerra, A. and Cisternas, P. (2003). Reproduction and larval
morphology of broadcasting and viviparous species in the Cryptasterina species complex.
Biol. Bull. 205, 285–294.
Calvo, M., Templado, J. and Penchaszadeh, P. E. (1998). Reproductive biology of the
gregarious Mediterranean Vermetid gastropod Dendropoma petraeum. J. Mar. Biol. Assoc.
UK 78, 525–549.
Capinera, J. L. (1979). Qualitative variation in plants and insects—effect of propagule size on
ecological plasticity. Am. Nat. 114, 350–361.
Chaitanawisuti, N. and Kritsanapuntu, A. (1997). Laboratory spawning and juvenile rearing
of the marine gastropod: Spotted babylon, Babylonia areolata Link 1807 (Neogastropoda:
Buccinidae), in Thailand. J. Shellfish Res. 16, 31–37.
Chaparro, O. R., Oyarzun, R. F., Vergara, A. M. and Thompson, R. J. (1999). Energy
investment in nurse eggs and egg capsules in Crepidula dilatata Lamark (Gastropoda,
Calyptraeidae) and its influence on hatching size of juveniles. J. Exp. Mar. Biol. Ecol.
232, 261–274.
Chester, C. M. (1996). The effect of adult nutrition on reproduction and development of the
estuarine nudibranch, Tenellia adspersa (Nordmann, 1845). J. Exp. Mar. Biol. Ecol. 198,
113–130.
Chintala, M. M. and Kennedy, V. S. (1993). Reproduction of Stylochus ellipticus (Platyhelminthes: Polycladida) in response to temperature, food, and presence or absence of a
partner. Biol. Bull. 185, 373–387.
Christiansen, F. B. and Fenchel, T. M. (1979). Evolution of marine invertebrate reproduction patterns. Theor. Popul. Biol. 16, 267–282.
Cipriani, R. and Penchazadeh, P. E. (1993). How does Strombina reproduce? Evidence from
two Venezuelan species (Prosobranchia: Columbellidae). Veliger 36, 178–184.
Clarke, A. (1992). Reproduction in the cold: Thorson revisited. Invertebr. Reprod. Dev. 22,
175–184.
Collin, R. (2000). Sex change, reproduction, and development of Crepidula adunca and
Crepidula lingulata (Gastropoda: Calyptraeidae). Veliger 43, 24–33.
Collin, R. (2003). Worldwide patterns in mode of development in calyptraeid gastropods.
Mar. Ecol. Prog. Ser. 247, 103–122.
Collin, R. and Wise, J. B. (1997). Morphology and development of Odostomia columbiana
Dall and Bartsch (Pyramidellidae): Implications for the evolution of gastropod development. Biol. Bull. 192, 242–252.
52
Dustin J. Marshall and Michael J. Keough
Cox, E. F. and Ward, S. (2002). Impact of elevated ammonium on reproduction in two
Hawaiian scleractinian corals with different life history patterns. Mar. Pollut. Bull. 44,
1230–1235.
Crump, M. L. (1981). Variation in propagule size as a function of environmental uncertainty
for tree frogs. Am. Nat. 117, 724–737.
Damiani, C. C. (2003). Reproductive costs of the symbiotic hydroid Hydractinia symbiolongicarpus (Buss and Yund) to its host hermit crab Pagurus longicarpus (Say). J. Exp. Mar. Biol.
Ecol. 288, 203–222.
Davis, M., Bolton, C. A. and Stoner, A. W. (1993). A comparison of larval development,
growth, and shell morphology in three Caribbean Strombus species. Veliger 36, 236–244.
Denny, M. W. and Shibata, M. F. (1989). Consequences of surf-zone turbulence for
settlement and external fertilisation. Am. Nat. 134, 859–889.
Denny, M. W., Dairiki, J. and Distefano, S. (1992). Biological consequences of topography
on wave-swept rocky shores: I. Enhancement of external fertilization. Biol. Bull. 183,
220–232.
Dickie, L., Hart, M. W. and Helling, R. (1989). Prefeeding larval development time is not
correlated with egg size in regular echinoids (Strongylocentrotus) species. Invertebr. Reprod.
Dev. 15, 229–232.
Dugan, J. E., Wenner, A. M. and Hubbard, D. M. (1991). Geographic variation in the
reproductive biology of the sand crab Emerita analoga (Stimpson) on the California coast.
J. Exp. Mar. Biol. Ecol. 150, 63–81.
Dunn, A. M. and McCabe, J. (1995). Resource allocation to young: Seasonal patterns
within and between Gammarus duebeni populations. Oikos 73, 199–202.
Dziminski, M. A. and Roberts, J. D. (2005). Fitness consequences of variable maternal
provisioning in quacking frogs (Crinia georgiana). J. Evol. Biol. 19, 144–155.
Eckelbarger, K. J. (1986). Vitellogenic mechanisms and the allocation of energy to offspring
in polychaetes. Bull. Mar. Sci. 39, 426–443.
Eckman, J. E. (1996). Closing the larval loop: Linking larval ecology to the population
dynamics of marine benthic invertebrates. J. Exp. Mar. Biol. Ecol. 200, 207–237.
Einum, S. and Fleming, I. A. (1999). Maternal effects of egg size in brown trout (Salmo trutta):
Norms of reaction to environmental quality. Proc. R. Soc. London Ser. B 266, 2095–2100.
Einum, S. and Fleming, I. A. (2000). Highly fecund mothers sacrifice offspring survival to
maximize fitness. Nature 405, 565–567.
Einum, S. and Fleming, I. A. (2002). Does within-population variation in fish egg size reflect
maternal influences on optimal values. Am. Nat. 160, 756–765.
Einum, S. and Fleming, I. A. (2004a). Does within-population variation in egg size reduce
intraspecific competition in Atlantic Salmon, Salmo salar? Funct. Ecol. 18, 110–115.
Einum, S. and Fleming, I. A. (2004b). Environmental unpredictability and offspring size:
Conservative versus diversified bet-hedging. Evol. Ecol. Res. 6, 443–455.
Elgar, M. A. (1990). Evolutionary compromise between a few large and many small eggs:
Comparative evidence in teleost fish. Oikos 59, 283–287.
Emlet, R. B. (1986). Facultative planktotrophy in the tropical echinoid Clypaster rosaceus
(Linnaeus) and a comparison with obligate planktotrophy in Clypaster subdepressus (Gray)
(Clypasteroida: Echinoidea). J. Exp. Mar. Biol. Ecol. 95, 183–202.
Emlet, R. B. (1995). Developmental mode and species geographic range in regular seaurchins (Echinodermata: Echinoidea). Evolution 49, 476–489.
Emlet, R. B. and Hoegh-Guldberg, O. (1997). Effect of egg size on post-larval performance:
Experimental evidence from a sea urchin. Evolution 51, 141–152.
Emlet, R. B., McEdward, L. and Strathmann, R. (1987). Echinoderm larval ecology viewed
from the egg. In ‘‘Echinoderm Studies’’ (M. Langoux and J. Lawrence, eds), Vol. 2,
pp. 55–136. A. A. Balkema, Rotterdam.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
53
Etter, R. J. (1989). Life history variation in the intertidal snail Nucella lapillus across a waveexposure gradient. Ecology 70, 1857–1876.
Farfan, B. C. and Ramirez, L. F. B. (1988). Spawning and ontogeny of Bulla gouldiana
(Gastropoda: Opisthobranchia: Cephalispidea). Veliger 31, 114–119.
Farley, G. S. and Levitan, D. R. (2001). The role of jelly coats in sperm-egg encounters,
fertilization success, and selection on egg size in broadcast spawners. Am. Nat. 157,
626–636.
Fitt, W. K. and Trench, R. K. (1981). Spawning, development, and acquisition of zooxanthellae by Tridacna squamosa (Mollusca, bivalvia). Biol. Bull. 161, 213–235.
Fox, C. W. (2000). Natural selection on seed-beetle egg size in nature and the laboratory:
Variation among environments. Ecology 81, 3029–3035.
Fox, C. W. and Czesak, M. E. (2000). Evolutionary ecology of progeny size in arthropods.
Annu. Rev. Entomol. 45, 341–369.
Fox, C. W. and Mousseau, T. A. (1996). Larval host plant affects fitness consequences of egg
size variation in the seed beetle Stator limbatus. Oecologia 107, 541–548.
Fox, C. W., Thakar, M. S. and Mosseau, T. A. (1997). Egg size plasticity in a seed beetle: An
adaptive maternal effect. Am. Nat. 149, 149–163.
Franke, E. S., Babcock, R. C. and Styan, C. A. (2002). Sexual conflict and polyspermy under
sperm-limited conditions: In situ evidence from field simulations with the free-spawning
marine echinoid Evechinus chloroticus. Am. Nat. 160, 485–496.
Fraser, A., Gomez, J., Hartwick, B. and Smith, M. J. (1981). Observations on the reproduction and development of Pisaster ochraceus (Brandt). Can. J. Zool. 59, 1700–1707.
Gallardo, C. S. (1977). Two modes of development in the morphospecies Crepdula dilatata
(Gastropoda: Calyptraeidae) from Southern Chile. Mar. Biol. 39, 241–251.
Gallardo, C. S. (1979). Developmental pattern and adaptations for reproduction in Nucella
crassilabrum and other muricacean gastropods. Biol. Bull. 157, 453–463.
George, S. B. (1994a). Population differences in maternal size and offspring quality for Leptasterias epichlora (Brandt) (Echinodermata: Asteroidea). J. Exp. Mar. Biol. Ecol. 175, 121–131.
George, S. B. (1994b). ‘‘Echinoderms Through Time’’ (B. David, A. Guille, J. P. Feral and
M. Roux, eds), pp. 297–307. A. A. Balkema, Rotterdam.
George, S. B. (1995). Echinoderm egg and larval quality as a function of adult nutritional
state. Oceanol. Acta 19, 297–308.
George, S. B., Cellario, C. and Fenaux, L. (1990). Population differences in egg quality of
Arbacia lixula (Echinodermata: Echinoidea): Proximate composition of eggs and larval
development. J. Exp. Mar. Biol. Ecol. 141, 107–118.
Geritz, S. A. H. (1995). Evolutionarily stable seed polymorphism and small-scale spatial
variation in seedling density. Am. Nat. 146, 685–707.
Geritz, S. A. H., van der Meijden, E. and Metz, J. A. J. (1999). Evolutionary dynamics of
seed size and seedling competitive ability. Theor. Popul. Biol. 55, 324–343.
Gibson, G. D. and Chia, F. S. (1989). Embryology and larval development of Haminoea
vesicula Gould (Opisthobranchia: Cephalispidea). Veliger 32, 409–412.
Gimenez, L. and Anger, K. (2001). Relationships among salinity, egg size, embryonic
development and larval biomass in the estuarine crab Chasmagnathus granulata, Dana
1851. J. Exp. Mar. Biol. Ecol. 260, 241–257.
Gosselin, L. A. and Qian, P. Y. (1999). Analysing energy content: A new micro-assay and an
assessment of the applicability of acid dichromate assays. Hydrobiologia 390, 141–151.
Guisande, C. and Harris, R. (1995). Effect of total organic content of eggs on hatching
success and naupliar survival in the copepod calanus helgolandicus. Limnol. Oceanogr. 40,
476–482.
Guisande, C., Sanchez, J., Maneiro, I. and Miranda, A. (1996). Trade-off between offspring
number and offspring size in the marine copepod Euterpina acutifrons at different food
concentrations. Mar. Ecol. Prog. Ser. 143, 37–44.
54
Dustin J. Marshall and Michael J. Keough
Hadfield, M. G. (1989). Latitudinal effects on juvenile size and fecundity in Petaloconchus
(Gastropoda). Bull. Mar. Sci. 45, 369–376.
Hadfield, M. G. and Miller, S. E. (1987). On developmental patterns of Opisthobranchs.
Am. Malacol. Bull. 5, 197–214.
Hagstrom, B. E. and Lonning, S. (1967). Experimental studies of Strongylocentrotus droebachiensis and S. pallidus. Sarsia 29, 165–176.
Harii, S., Kayanne, H., Takigawa, H., Hayashibara, T. and Yamamoto, M. (2002). Larval
survivorship, competency periods, and settlement of two brooding corals, Heliopora
coerulea and Pocillporoa damicornis. Mar. Biol. 141, 39–46.
Hart, M. W. (1995). What are the costs of small egg size for marine invertebrate with feeding
planktonic larvae? Am. Nat. 146, 415–426.
Harvey, P. H. and Pagel, M. D. (1991). ‘‘The Comparative Method in Evolutionary
Biology’’ Oxford University Press, Oxford.
Havenhand, J. N. (1993). Egg to juvenile period, generation time, and the evolution of larval
type in marine invertebrates. Mar. Ecol. Prog. Ser. 97, 247–260.
Hendrickx, F., Maelfait, J. P. and Lens, L. (2003a). Relationship between fluctuating
asymmetry and fitness within and between stressed and unstressed populations of the
wolf spider Pirata piraticus. J. Evol. Biol. 16, 1270–1279.
Hendrickx, F., Maelfait, J. P., Speelmans, M. and Van Straalen, N. M. (2003b). Adaptive
reproductive variation along a pollution gradient in a wolf spider. Oecologia 134,
189–194.
Hendry, A. P., Day, T. and Cooper, A. B. (2001). Optimal size and number of propagules:
Allowance for discrete stages and effects of maternal size on reproductive output and
offspring fitness. Am. Nat. 157, 387–407.
Hess, H. C. (1993). The evolution of parental care in brooding spirobid polychaetes: The
effect of scaling constraints. Am. Nat. 141, 577–596.
Highsmith, R. C. and Emlet, R. B. (1986). Delayed metamorphosis—effect on growth and
survival of juvenile sand dollars (Echinoidea, Clypeasteroida). Bull. Mar. Sci. 39, 347–361.
Hines, A. H. (1988). Fecundity and reproductive output in two species of deep-sea crabs,
Geryon fenneri and G. quinquedens (Decapoda: Brachyura). J. Crustacean Biol. 8, 557–562.
Ho Sun, M. and Hong, S. Y. (1994). Spawning, larval development, and larval shell
morphology of Cantharidus callichroa callichroa (Phillipi, 1850) (Gastropoda: Trochidae)
in Korean waters. Veliger 37, 136–140.
Hoegh-Guldberg, O. and Pearse, J. S. (1995). Temperature, food availability and the
development of marine invertebrate larvae. Am. Zool. 35, 415–425.
Holmes, S. (1997). Notes on gamete release and fertilisation in Calliostoma zizyphinum (L.)
(Gastropoda; Trochidae). J. Mollus. Stud. 63, 471–473.
Hughes, R. N. and Hughes, H. P. I. (1987). Spawning and hatching of Cypraecassis testiculis
Linnaeus, 1758 (Tonnacea: Cassidae). Veliger 29, 256–259.
Isomura, N. and Nishimura, M. (2001). Size variation of planulae and its effect on the
lifetime of planulae in three pocilloporid corals. Coral Reefs 20, 309–315.
Ito, K. (1997). Egg-size and -number variations related to maternal size and age, and the
relationship between egg size and larval characteristics in and annual marine gastropod,
Haloa japonica (Opisthobranchia; Cephalaspidea). Mar. Ecol. Prog. Ser. 152, 187–195.
Jaeckle, W. B. (1995). Variation in the size, energy content and biochemical composition of
invertebrate eggs: Correlates to the mode of larval development. In ‘‘Ecology of Marine
Invertebrate Larvae’’ (L. McEdward, ed.), Vol. 1, pp. 49–78. CRC Press, Inc., Boca
Raton.
Jones, H. L., Todd, C. D. and Lambert, W. J. (1996). Intraspecific variation in embryonic
and larval traits of the dorid nudibranch mollusc Adalaria proxima (Alder and Hancock)
around the northern coasts of British Isles. J. Exp. Mar. Biol. Ecol. 202, 29–47.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
55
Jonsson, P. R., Berntsson, K. M., Andre, C. and Wangberg, S. A. (1999). Larval growth and
settlement of the European oyseter (Ostrea edulis) as a function of food quality measured as
fatty acid composition. Mar. Biol. 134, 559–570.
Kennedy, V. S. and Lutz, R. A. (1989). Larval and early postlarval development of Macoma
mitchelli (Bivalvia: Tellinidae). Veliger 32, 29–38.
Kim, S. H. and Hong, S. Y. (2004). Reproductive biology of Palaemon gravieri (Decapoda:
Caridea: Palaemonidae). J. Crustacean Biol. 24, 121–130.
Kohn, A. J. and Perron, F. E. (1994). ‘‘Life History and Biogeography Patterns in Conus’’
Oxford Biogeography Series No. 8. p. 106. Clarendon Press, Oxford.
Kokita, T. (2003). Potential latitudinal variation in egg size and number of a geographically
widespread reef fish, revealed by common-environment experiments. Mar. Biol. 143,
593–601.
Komatsu, M. and Nojima, S. (1985). Development of the seastar, Astropecten gisselbrechti
Doderlein. Pac. Sci. 69, 272–281.
Komatsu, M., Kono, Y. T., Yoshizawa, H. and Akabane, S. (1979). Reproduction and
development of the hermaphroditic sea-star Asterina minor (Hayashi). Biol. Bull. 157,
258–274.
Komatsu, M., Oguro, C. and Kano, Y. (1982). ‘‘Internation Echinoderm Conference’’
(J. M. Lawrence, ed.). A. A. Balkema, Rotterdam, Tampa Bay.
Komatsu, M., Kawai, M., Nojima, S. and Oguro, C. (1994). ‘‘Echinoderms Through Time’’
(B. David, A. Gulle, J. P. Feral and M. Roux, eds), pp. 327–333. A. A. Balkema, Rotterdam.
Kupriyanova, E. K., Nishi, E., Ten Hove, H. A. and Rzhavsky, A. V. (2001). Life-history
patterns in serpulimorph polychaetes: Ecological and evolutionary perspectives. Oceanogr.
Mar. Biol. 39, 1–101.
Laaksonen, T. (2004). Hatching asynchrony as a bet-hedging strategy—an offspring diversity
hypothesis. Oikos 104, 616–620.
Lack, D. (1947). The significance of clutch size. Ibis 89, 302–352.
Lardies, M. A. and Castilla, J. C. (2001). Latitudinal variation in the reproductive biology of
the commensal crab Pinnaxodes chilensis (Decapoda: Pinnotheridae) along the Chilean
coast. Mar. Biol. 139, 1125–1133.
Lardies, M. A. and Wehrtmann, I. S. (2001). Latitudinal variation in the reproductive
biology of Betaeus truncatus (Decapoda: Alpheidae) along the Chilean coast. Ophelia 55,
55–67.
Lasker, H. R., Brazeau, D. A., Clalderon, J., Coffroth, M. A., Coma, R. and Kim, K.
(1996). In situ rates of fertilization among broadcast spawning gorgonian corals. Biol. Bull.
190, 45–55.
Levitan, D. R. (1991). Influence of body size and population density on fertilisation success
and reproductive output in a free-spawning invertebrate. Biol. Bull. 191, 261–268.
Levitan, D. R. (1993). The importance of sperm limitation to the evolution of egg size in
marine invertebrates. Am. Nat. 141, 517–536.
Levitan, D. R. (1995). The ecology of fertilization in free-spawning invertebrates. In
‘‘Ecology of Marine Larvae’’ (L. R. McEdward, ed.), pp. 124–152. CRC Press, Inc.,
Boca Raton.
Levitan, D. R. (1996a). Effects of gamete traits on fertilisation in the sea and the evolution of
sexual dimorphism. Nature 382, 153–155.
Levitan, D. R. (1996b). Predicting optimal and unique egg size in free spawning marine
invertebrates. Am. Nat. 148, 174–188.
Levitan, D. R. (1998). Does Bateman’s principle apply to broadcast spawning organisms?
Egg traits influence in situ fertilisation rates among congeneric sea urchins. Evolution 52,
1043–1056.
56
Dustin J. Marshall and Michael J. Keough
Levitan, D. R. (2000). Optimal egg size in marine invertebrates: Theory and phylogenetic
analysis of the critical relationship between egg size and development time in echinoids.
Am. Nat. 156, 175–192.
Levitan, D. R. (2002). Density-dependent selection on gamete traits in three congeneric sea
urchins. Ecology 83, 464–479.
Levitan, D. R. and Petersen, C. (1995). Sperm limitation in the sea. Trends Ecol. Evol. 10,
228–231.
Levitan, D. R., Sewell, M. A. and Chia, F.-S. (1992). How distribution and abundance
influences fertilization success in the sea urchin Strongylocentrotus franciscanus. Ecology 73,
248–254.
Lewis, A. R. and Choat, J. H. (1993). Spawning mode and reproductive output of the
tropical Cephalopod Idiosepius pygmaeus. Can. J. Fish. Aquat. Sci. 50, 20–28.
Lips, K. R. (2001). Reproductive trade-offs and bet-hedging in Hyla calypsa, a Neotropical
treefrog. Oecologia 128, 509–518.
Lonsdale, D. J. and Levinton, J. S. (1985). Latitudinal differentiation in embryonic duration,
egg size, and newborn survival in a harpacticoid copepod. Biol. Bull. 168, 419–431.
Maldonado, M. and Young, C. M. (1999). Effects of the duration of larval life on post-larval
stages of the demosponge Sigmadocia caerulea. J. Exp. Mar. Biol. Ecol. 232, 9–21.
Marshall, D. J. (2002). In situ measures of spawning synchrony and fertilization success in an
intertidal, free-spawning invertebrate. Mar. Ecol. Prog. Ser. 236, 113–119.
Marshall, D. J. (2005). Geographical variation in offspring size effects across generations.
Oikos 108, 602–608.
Marshall, D. J. and Evans, J. P. (2005). Does egg competition occur in marine broadcast
spawners? J. Evol. Biol. 18, 1244–1252.
Marshall, D. J. and Keough, M. J. (2003a). Effects of settler size and density on early
post-settlement survival of Ciona intestinalis in the field. Mar. Ecol. Prog. Ser. 259,
139–144.
Marshall, D. J. and Keough, M. J. (2003b). Sources of variation in larval quality for freespawning marine invertebrates: Egg size and the local sperm environment. Invertebr.
Reprod. Dev. 44, 63–70.
Marshall, D. J. and Keough, M. J. (2003c). Variation in the dispersal potential of non-feeding
invertebrate larvae: The desperate larva hypothesis and larval size. Mar. Ecol. Prog. Ser.
255, 145–153.
Marshall, D. J. and Keough, M. J. (2004a). Variable effects of larval size on postmetamorphic performance in the field. Mar. Ecol. Prog. Ser. 279, 73–80.
Marshall, D. J. and Keough, M. J. (2004b). When the going gets rough: Effect of maternal
size manipulation on offspring quality. Mar. Ecol. Prog. Ser. 272, 301–305.
Marshall, D. J. and Keough, M. J. (2005). Offspring size effects in the marine environment:
A field test for a colonial invertebrate. Austral Ecology 30, 275–280.
Marshall, D. J., Styan, C. A. and Keough, M. J. (2000). Intraspecific co-variation between
egg and body size affects fertilisation kinetics in free-spawning marine invertebrates. Mar.
Ecol. Prog. Ser. 195, 305–309.
Marshall, D. J., Styan, C. A. and Keough, M. J. (2002). Sperm environment affects offspring
quality in broadcast spawning marine invertebrates. Ecol. Lett. 5, 173–176.
Marshall, D. J., Bolton, T. F. and Keough, M. J. (2003a). Offspring size affects the
post-metamorphic performance of a colonial marine invertebrate. Ecology 84,
3131–3137.
Marshall, D. J., Pechenik, J. A. and Keough, M. J. (2003b). Larval activity levels and delayed
metamorphosis affect post-larval performance in the colonial, ascidian Diplosoma listerianum. Mar. Ecol. Prog. Ser. 246, 153–162.
Marshall, D. J., Evans, J. P. and Steinberg, P. D. (2004a). The early sperm gets the good egg:
Mating order effects in free-spawners. Proc. R. Soc. London Ser. B 271, 1585–1589.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
57
Marshall, D. J., Semmens, D. and Cook, C. (2004b). Consequences of spawning at low tide:
Limited gamete dispersal for a rockpool anemone. Mar. Ecol. Prog. Ser. 266, 135–142.
Marshall, D. J., Cook, C. N. and Emlet, R. B. (2006). Offspring size effects mediate
competitive interactions in a colonial marine invertebrate. Ecology 87, 214–225.
McCarthy, D. A., Young, C. M. and Emson, R. H. (2003). Influence of wave-induced
disturbance on seasonal spawning patterns in the sabellariid Polychaete Phragmatopoma
lapidosa. Mar. Ecol. Prog. Ser. 256, 123–133.
McClary, D. J. and Mladenov, P. V. (1990). Brooding biology of the sea star Pteraster militaris
(O. F. Muller): Energetic and histological evidence for nutrient translocation to brooded
juveniles. J. Exp. Mar. Biol. Ecol. 142, 183–199.
McEdward, L. R. (1997). Reproductive strategies of marine benthic invertebrates revisited:
Facultative feeding by planktotrophic larvae. Am. Nat. 150, 48–72.
McEdward, L. R. and Carson, S. F. (1987). Variation in egg organic content and its relationship with egg size in the starfish Solaster stimpsoni. Mar. Ecol. Prog. Ser. 37, 159–169.
McEdward, L. R. and Chia, F. S. (1991). Size and energy content of eggs from echinoderms
with pelagic lecithotrophic development. J. Exp. Mar. Biol. Ecol. 147, 95–102.
McEdward, L. R. and Coulter, L. K. (1987). Egg volume and energetic content are not
correlated among sibling offspring of starfish: Implications for life-history theory. Evolution 41, 914–917.
McEdward, L. R. and Miner, B. G. (2001). Larval and life-cycle patterns in echinoderms.
Can. J. Zool. 79, 1125–1170.
McEdward, L. R. and Miner, B. G. (2006). Estimation and interpretation of egg provisioning in marine invertebrates. Integr. Comp. Biol. 46, 224–232.
McEuen, F. S. and Chia, F. S. (1991). Development and metamorphosis of two psolid seacucumbers, Psolus chitonoides and Psolidium bullatum, with a review of reproductive
patterns in the family Psolidae (Holothurodea: Echinodermata). Mar. Biol. 109, 267–279.
McGinley, M. A., Temme, D. H. and Geber, M. A. (1987). Parental investment in offspring in
variable environments: Theoretical and empirical considerations. Am. Nat. 130, 370–398.
McLaren, I. A. (1965). Some relationships between temperature and egg size, body size,
development rate, and fecundity, of the copepod Pseudocalanus. Limnol. Oceanogr. 10,
528–538.
McLaren, I. A. (1966). Predicting development rate of copepod eggs. Biol. Bull. 131,
457–469.
Mead, K. S. and Denny, M. W. (1995). The effects of hydrodynamic shear stress on
fertilisation and early development of the purple sea urchin Strongylocentrotus purpuratus. Biol. Bull. 188, 46–56.
Meidel, K., Scheibling, R. E. and Metaxas, A. (1999). Relative importance of parental and
larval nutrition on the larval development and metamorphosis of the sea urchin Strongylocentrotus droebachiensis. J. Exp. Mar. Biol. Ecol. 240, 161–178.
Millar, R. B. and Anderson, M. J. (2003). The kinetics of monospermic and polyspermic
fertilization in free-spawning marine invertebrates. J. Theor. Biol. 224, 79–85.
Miloslavich, P. and Dufresne, L. (1994). Development and effect of female size on egg and
juvenile production in the neogastropod Buccinum cyoneum from the Saguenay ford. Can.
J. Fish. Aquat. Sci. 51, 2766–2872.
Miloslavich, P. and Penchazadeh, P. E. (1992). Reproductive biology of Vermetus sp. and
Dendropoma corrodens (Orbigny, 1842): Two vermetid gastropods from the southern
Caribbean. Veliger 35, 78–88.
Miloslavich, P. and Penchazadeh, P. E. (1994). Spawn and development of Engoniophos
unicinctus (Say, 1825) (Gastropoda: Prosobranchia) from the southern Caribbean sea.
Veliger 37, 425–429.
Moles, A. T., Warton, D. I., Stevens, R. D. and Westoby, M. (2004). Does a latitudinal
gradient in seedling survival favour larger seeds in the tropics? Ecol. Lett. 7, 911–914.
58
Dustin J. Marshall and Michael J. Keough
Moran, A. L. and Emlet, R. B. (2001). Offspring size and performance in variable environments: Field studies on a marine snail. Ecology 82, 1597–1612.
Ng, T. Y. T. and Keough, M. J. (2003). Delayed effects of larval exposure to Cu in the
bryozoan Watersipora subtorquata. Mar. Ecol. Prog. Ser. 257, 77–85.
Oliver, J. and Babcock, R. C. (1992). Aspects of the fertilization ecology of broadcast
spawning corals: Sperm dilution effects and in situ measures of fertilization. Biol. Bull.
183, 409–417.
Olson, R. R., Cameron, J. L. and Young, C. M. (1993). Larval development (with
observations on spawning) of the pencil urchin Pyllacanthus imperialis: A new intermediate
form? Biol. Bull. 185, 77–85.
Osorio, C., Gallardo, C. and Atan, H. (1992). Egg mass and intracapsular development of
Cypraea caputdraconis Melvill, 1888, from Easter Islan (Gastropoda: Cypraeidae). Veliger
35, 316–322.
Oullet, P. and Plante, F. (2004). An investigation of the sources of variability in American
lobster (Homarus americanus) eggs and larvae: Female size and reproductive status, and
interannual and interpopulation comparisons. J. Crustacean Biol. 24, 481–495.
Palmer, A. R. (1990). Predator size, prey size and the scaling of vulnerability: Hatchling
gastropods vs. barnacles. Ecology 71, 759–775.
Parker, G. A. and Begon, M. (1986). Optimal egg size and clutch size—effects of environment and maternal phenotype. Am. Nat. 128, 573–592.
Pechenik, J. A., Wendt, D. E. and Jarrett, J. N. (1998). Metamorphosis is not a new
beginning. Bioscience 48, 901–910.
Pechenik, J. A., Gleason, T., Daniels, D. and Champlin, D. (2001). Influence of larval
exposure to salinity and cadmium stress on juvenile performance of two marine invertebrates (Capitella sp I and Crepidula fornicata). J. Exp. Mar. Biol. Ecol. 264, 101–114.
Pedersen, R. V. K. and Page, L. R. (2000). Development and metamorphosis of the
planktotrophic larvae of the moon snail, Policines lewisii (Gould, 1847) (Caengastropoda:
Naticoidea). Veliger 43, 58–63.
Pemberton, A. J., Hughes, R. N., Manriquez, P. H. and Bishop, J. D. D. (2003). Efficient
utilization of very dilute aquatic sperm: Sperm competition may be more likely than sperm
limitation when eggs are retained. Proc. Royal Soc. London Ser. B: Biol. Sci. 270, S223–S226.
Pennington, J. T. (1985). The ecology of fertilization of echinoid eggs: The consequences of
sperm dilution, adult aggregation and synchronous spawning. Biol. Bull. 169, 417–430.
Pernet, B. and Jaeckle, W. B. (2004). Size and organic content of eggs of marine annelids,
and the underestimation of egg energy content by dichromate oxidation. Biol. Bull. 207,
67–71.
Phillips, N. E. (2002). Effects of nutrition-mediated larval condition on juvenile performance in a marine mussel. Ecology 83, 2562–2574.
Phillips, N. E. and Gaines, S. D. (2002). Spatial and temporal variability in size at settlement
of intertidal mytilid mussels from around Pt. Conception, California. Invertebr. Reprod.
Dev. 41, 171–177.
Podolsky, R. D. (2001). Evolution of egg target size: An analysis of selection on correlated
characters. Evolution 55, 2470–2478.
Podolsky, R. D. (2002). Fertilization ecology of egg coats: Physical versus chemical contributions to fertilization success of free-spawned eggs. J. Exp. Biol. 205, 1657–1668.
Podolsky, R. D. and Strathmann, R. R. (1996). Evolution of egg size in free spawners:
Consequences of the fertilisation-fecundity trade-off. Am. Nat. 148, 160–173.
Qian, P.-Y. (1994). Effect of food quantity on growth and reproductive characteristics of
Capitella sp. (Annelida: Polychaeta). Invertebr. Reprod. Dev. 26, 175–185.
Qian, P. Y. and Chia, F. S. (1991). Fecundity and egg size are mediated by food quality in
the polychaete worm Capitella sp. J. Exp. Mar. Biol. Ecol. 148, 11–25.
The Evolutionary Ecology of Offspring Size in Marine Invertebrates
59
Raimondi, P. T. and Keough, M. J. (1990). Behavioural variability in marine larvae. Aust.
J. Ecol. 15, 427–437.
Ramfafia, C., Battaglene, S. and Byrne, M. (eds) (2000). ‘‘Larval Development in the
Tropical Sea Cucumber Holothuria Scabra’’. Balkema, Lisse.
Ramirez-Llodra, E. R. (2002). Fecundity and life-history strategies in marine invertebrates.
In ‘‘Advances in Marine Biology’’ (C. M. Young, ed.), Vol. 43, pp. 88–172. Academic
Press, London.
Ramon, M. (1991). Spawning and development characteristics of Cymatium cutaceum and
C. corrugatum (Gastropoda: Prosobranchia) in the laboratory. Ophelia 33, 31–43.
Reznick, D. N. (1981). ‘‘Grandfather effects’’: The genetics of interpopulation differences in
offspring size in the mosquito fish Gambusia affinis. Evolution 35, 941–953.
Rivest, B. R. (1983). Development and the influence of nurse egg allotment on hatching size
in Searlesia dira (Reeve, 1846) (Prosobranchia: Buccinidae). J. Exp. Mar. Biol. Ecol. 69,
217–241.
Sakai, S. and Harada, Y. (2001). Why do large mothers produce large offspring? Theory and
a test. Am. Nat. 157, 348–359.
Sargent, R. C., Taylor, P. D. and Gross, M. R. (1987). Parental care and the evolution of egg
size in fishes. Am. Nat. 129, 32–46.
Scheibling, R. E. and Lawrence, J. M. (1982). Differences in reproductive strategies of
morph of the genus Echinaster (Echinodermata: asteriodea) from the easter Gulf of
Mexico. Mar. Biol. 70, 51–62.
Sewell, M. A. (1994). Birth, recruitment and juvenile growth in the intraovarian brooding
sea cucumber Leptosynapta clarki. Mar. Ecol. Prog. Ser. 114, 149–156.
Sewell, M. A. and Young, C. M. (1997). Are echinoderm egg size distributions bi-modal?
Biol. Bull. 193, 297–305.
Sinervo, B. and McEdward, L. R. (1988). Developmental consequences of an evolutionary
change in egg size: An experimental test. Evolution 42, 885–899.
Smith, C. C. and Fretwell, S. D. (1974). The optimal balance between size and number of
offspring. Am. Nat. 108, 499–506.
Spight, T. M. (1976). Hatching size and the distribution of nurse eggs among prosobranch
embryos. Biol. Bull. 150, 491–499.
Stanton, M. L. (1984). Seed variation in wild radish: Effect of seed size on components of
seedling and adult fitness. Ecology 65, 1105–1112.
Stanton, M. L. (1985). Seed size and emergence time within a stand of wild radish (Raphanus
raphanistrum L)—the establishment of a fitness hierarchy. Oecologia 67, 524–531.
Staver, J. M. and Strathmann, R. R. (2002). Evolution of fast development of planktonic
embryos to early swimming. Biol. Bull. 203, 58–69.
Stearns, S. C. (1992). ‘‘The Evolution of Life Histories’’. Oxford University Press, Oxford.
Steele, D. H. (1977). Correlation between egg size and developmental period. Am. Nat. 111,
371–372.
Steer, M. A., Moltschaniwskyj, N. A. and Jordan, A. R. (2003). Embryonic development of
southern calamary (Sepioteuthis australis) within the constraint of an aggregated egg mass.
Mar. Freshwater Res. 54, 217–226.
Steer, M. A., Moltschaniwskyj, M. A., Nichols, D. S. and Miller, M. (2004). The role of
temperature and maternal ration in embryo survival: Using the dumpling squid Euprymna
tasmanica as a model. J. Exp. Mar. Biol. Ecol. 307, 73–89.
Stewart, B. and Mladenov, P. V. (1994). ‘‘Echinoderms Through Time’’ (B. David,
A. Gulle, J. P. Feral and M. Roux, eds), pp. 491–497. A. A. Balkema, Rotterdam.
Strathmann, R. R. (1977). Egg size, larval development, and juvenile size in benthic marineinvertebrates. Am. Nat. 111, 373–376.
Strathmann, R. R. (1995). Peculiar constraints on life histories imposed by protective or
nutritive devices for embryos. Am. Zool. 35, 426–433.
60
Dustin J. Marshall and Michael J. Keough
Strathmann, R. R. and Chaffee, C. (1984). Constraints on egg masses. II. Effect of spacing,
size and number of eggs on ventilation of masses of embryos in jelly, adherent groups or
thin walled capsules. J. Exp. Mar. Biol. Ecol. 84, 85–93.
Strathmann, R. R., Staver, J. M. and Hoffman, J. R. (2002). Risk and the evolution of cellcycle durations of embryos. Evolution 56, 708–720.
Styan, C. A. (1998). Polyspermy, egg size and the fertilisation kinetics of free-spawning
marine invertebrates. Am. Nat. 152, 290–297.
Styan, C. A. and Butler, A. J. (2000). Fitting fertilisation kinetics models for free-spawning
marine invertebrates. Mar. Biol. 137, 943–951.
Thomas, F. I. M. (1994). Transport and mixing of gametes in three free-spawning polychaete annelids, Phragmatopoma californica (Fewkes), Sabellaria cementarium (Moore),
and Shizobranchia insignis (Bush). J. Exp. Mar. Biol. Ecol. 179, 11–27.
Thorson, G. (1950). Reproductive and larval ecology of marine bottom invertebrates. Biol.
Rev. 25, 1–45.
Toonen, R. J. and Pawlik, J. R. (2001). Settlement ofthe gregarious tube worm Hydroides
dianthus (Polychaeta: Serpulidae). II. Testing the desperate larva hypothesis. Mar. Ecol.
Prog. Ser. 224, 115–131.
Turner, R. and Lawrence, J. M. (1977). Volume and composition of echinoderm eggs:
Implications for the use of egg size in life-history models. In ‘‘Reproductive Ecology of
Marine Invertebrates’’ (S. E. Stancyk, ed.), pp. 25–40. University of South Carolina
Press, Columbia.
Turner, S. J. (1992). The egg capsules and early life history of the corallivorous gastropod
Drupella cornus (Roding, 1978). Veliger 35, 16–25.
Valentinsson, D. (2002). Reproductive cycle and maternal effects on offspring size and
number in the neogastropod Buccinum undatum (L.). Mar. Biol. 140, 1139–1147.
Vance, R. R. (1973a). More on reproductive strategies in marine benthic invertebrates. Am.
Nat. 107, 353–361.
Vance, R. R. (1973b). On reproductive strategies in marine benthic invertebrates. Am. Nat.
107, 339–352.
Veliz, D., Guisado, C. and Winkler, F. M. (2001). Morphological, reproductive, and genetic
variability among three populations of Crucibulum quirqinae (Gastropoda: Calyptraeidae)
in northern Chile. Mar. Biol. 139, 527–534.
Vogel, H., Czihak, G., Chang, P. and Wolf, W. (1982). Fertilization kinetics of sea-urchin
eggs. Math. Biosci. 58, 189–216.
Wagele, J.-W. (1987). On the reproductive biology of Ceratoserolis trilobitoides (crustacea:
Isopoda): Latitudinal variation of fecundity and embryonic development. Polar Biol. 7,
11–24.
Wedell, N., Gage, M. J. C. and Parker, G. A. (2002). Sperm competition, male prudence
and sperm-limited females. Trends Ecol. Evol. 17, 313–319.
Wehrtmann, I. S. and Kattner, G. (1998). Changes in volume, biomass, and fatty acids of
developing eggs in Nauticaris magellanica (Decapoda: Caridea): A latitudinal comparison.
J. Crustacean Biol. 18, 413–422.
Wendt, D. E. (1998). Effect of larval swimming duration on growth and reproduction of
Bugula neritina (Bryozoa) under field conditions. Biol. Bull. 195, 126–135.
Wendt, D. E. (2000). Energetics of larval swimming and metamorphosis in four species of
Bugula (Bryozoa). Biol. Bull. 198, 346–356.
Williams, M. E., Bentley, M. G. and Hardege, J. D. (1997). Assessment of field fertilisation
success in the infaunal polychaete Arenicola marina (L.). Invertebr. Reprod. Dev. 31, 1–3.
Willows, R. I. (1987). Intrapopulation variation in the reproductive characteristics of
2 populations of Ligia oceanica (Crustacea, Oniscidea). J. Anim. Ecol. 56, 331–340.
An Evaluation of the Effects of
Conservation and Fishery
Enhancement Hatcheries on
Wild Populations of Salmon1
Kerry A. Naish,*,2 Joseph E. Taylor, III† Phillip S. Levin,‡ Thomas
P. Quinn,* James R. Winton,§ Daniel Huppert,} and Ray Hilborn*
Contents
1. Introduction
1.1. Scope of the review
1.2. Motivations and objectives of hatcheries
1.3. Content overview
2. Historical Overview of Hatchery Activities
3. Political Dynamics of Hatchery Programmes
4. Geographical Extent of Activities
4.1. Enhancement of indigenous salmonids: Conservation,
production and mitigation hatcheries
4.2. Enhancement of non-indigenous salmon and trout:
Introductions
5. Potential Consequences of Enhancement Activities
5.1. Genetic risks associated with salmon hatchery programmes
5.2. Behavioural and ecological interactions between wild and
hatchery-produced salmon
5.3. The effects of harvest on wild salmon populations
5.4. Disease effects of salmonid enhancement
*
{
{
}
}
1
2
63
65
68
70
71
78
84
84
99
100
100
127
133
141
School of Aquatic and Fishery Sciences, University of Washington, Washington 98195, USA
Departments of History and Geography, Simon Fraser University, British Columbia, Canada, USA
Northwest Fisheries Science Centre, NOAA Fisheries, Seattle, Washington 98122, USA
Western Fisheries Research Center, US Geological Survey, Seattle, Washington 98115, USA
School of Marine Affairs, University of Washington, Seattle, Washington 98195, USA
Lead authors of specific sections: K.A.N., genetics; J.E.T., historical and political perspectives; P.S.L., current
status; T.P.Q., competition; R.H., harvest; J.R.W., disease; D.H., economic analyses, USA
Corresponding author: Email Knaish@u.washington.edu
Advances in Marine Biology, Volume 53
ISSN 0065-2881, DOI: 10.1016/S0065-2881(07)53002-6
#
2008 Elsevier Ltd.
All rights reserved.
61
62
Kerry A. Naish et al.
6. Economic Perspectives on Hatchery Programmes
6.1. Measuring costs, effectiveness and benefits
6.2. Cost-effectiveness of hatchery programmes
6.3. BCA of hatchery programmes
6.4. Complicating factors
6.5. Conclusions
7. Discussion
7.1. Release objectives and release sizes
7.2. Interactions between hatchery and wild fish
7.3. Economic issues
7.4. Moving forward: Scientific and social dimensions
7.5. Conclusions
Acknowledgements
References
150
151
153
156
159
160
160
161
162
167
168
169
170
170
Abstract
The historical, political and scientific aspects of salmon hatchery programmes
designed to enhance fishery production, or to recover endangered populations,
are reviewed. We start by pointing out that the establishment of hatcheries has
been a political response to societal demands for harvest and conservation; given
this social context, we then critically examined the levels of activity, the biological
risks, and the economic analysis associated with salmon hatchery programmes.
A rigorous analysis of the impacts of hatchery programmes was hindered by the
lack of standardized data on release sizes and survival rates at all ecological
scales, and since hatchery programme objectives are rarely defined, it was also
difficult to measure their effectiveness at meeting release objectives. Debates on
the genetic effects of hatchery programmes on wild fish have been dominated by
whether correct management practices can reduce negative outcomes, but we
noted that there has been an absence of programmatic research approaches
addressing this important issue. Competitive interactions between hatchery and
wild fish were observed to be complex, but studies researching approaches to
reduce these interactions at all ecological scales during the entire salmon life
history have been rare, and thus are not typically considered in hatchery management. Harvesting of salmon released from fishery enhancement hatcheries
likely impacts vulnerable wild populations; managers have responded to this
problem by mass marking hatchery fish, so that fishing effort can be directed
towards hatchery populations. However, we noted that the effectiveness of this
approach is dependant on accurate marking and production of hatchery fish with
high survival rates, and it is not yet clear whether selective fishing will prevent
overharvest of wild populations. Finally, research demonstrating disease transmission from hatchery fish to wild populations was observed to be equivocal;
evidence in this area has been constrained by the lack of effective approaches to
studying the fate of pathogens in the wild. We then reviewed several approaches
to studying the economic consequences of hatchery activities intended to inform
the social decisions surrounding programmes, but recognized that placing
Evaluation of the Effect of Hatcheries on Wild Salmon
63
monetary value on conservation efforts or on hatcheries that mitigate cultural
groups’ loss of historical harvest opportunities may complicate these analyses.
We noted that economic issues have rarely been included in decision making on
hatchery programmes. We end by identifying existing major knowledge gaps,
which, if filled, could contribute towards a fuller understanding of the role that
hatchery programmes could play in meeting divergent goals. However, we also
recognized that many management recommendations arising from such research
may involve trade-offs between different risks, and that decisions about these
trade-offs must occur within a social context. Hatcheries have played an important role in sustaining some highly endangered populations, and it is possible
that reform of practices will lead to an increase in the number of successful
programmes. However, a serious appraisal of the role of hatcheries in meeting
broader needs is urgently warranted and should take place at the scientific, but
more effectively, at the societal level.
1. Introduction
Enhancement is increasingly seen as an important fishery management
tool (Leber et al., 2005a), especially in light of the worldwide decline in
wild fish populations. Broadly defined as the deliberate release of cultured
organisms to increase population abundance for conservation or harvest
objectives, enhancement of fish and invertebrate populations has been
implemented extensively since the turn of the century. However, there has
been considerable debate about the efficacy of releasing cultured organisms,
the impact of these organisms on conspecific wild populations, and the
relevance of this approach for meeting societal needs [reviewed in Taylor
(1999a) and in Section 2]. Thus, enhancement has fallen out of favour as a
management technique at various periods since it was first implemented.
However, improvements in seed production, rearing technology, disease
control, tagging and genetic and ecological approaches to management have
invigorated renewed research in the field (Blaxter, 2000). These technological improvements have coincided with a changing philosophy; namely, that
enhancement should be conducted in a scientifically based and sustainable
manner (Leber et al., 2005a), rather than providing a means of supplying
unlimited fishery resources, or replacing extirpated natural populations
without addressing the reasons for the decline.
Given this shift in philosophy and the renewed interest in the field, it is
not surprising that several reviews and edited volumes on the topic have been
published recently. In a comprehensive evaluation of marine fish enhancement, Blaxter (2000) has shown that success depends on the life history stage,
the season at release, and the size of the enhanced region. However, doubt
remains over whether enhancement can be used to recover declining fish
populations occurring in the high seas (Blaxter, 2000). In their introduction
64
Kerry A. Naish et al.
to an edited symposium on the topic, Leber et al. (2005a) identified several
key emerging issues associated with recent advances in the field, but pointed
out that there are few examples of the successful implementation of these
advances (Leber et al., 2005b). This theme is echoed in a review
of enhancement of marine invertebrates (Bell et al., 2005); although programmes aim to either rebuild depleted populations or increase their
productivity, objectives are infrequently identified and success is rarely
measured, so any advances cannot be effectively evaluated.
A review of salmon enhancement activities adds an interesting dimension to discussions within the field. Salmon populations exhibit an extensive
range of life history strategies (Allendorf and Waples, 1996; Quinn, 2005);
they can be locally adapted to their rearing and spawning habitats (Taylor,
1991) and are genetically differentiated from each other on a fine scale (e.g.,
Waples et al., 2001). Therefore, many enhancement efforts are aimed at
discrete stocks or populations of salmon, and often take the form of hatchery
programmes sited near spawning grounds. Additionally, programmes vary
in their objectives and range from fishery enhancement to conservation
hatcheries (Utter and Epifanio, 2002). There is considerable interest in the
interactions between hatchery-produced fish and conspecific wild populations, many of which are endangered or declining (National Research
Council, 1996; Parrish et al., 1998). Above all, salmon are culturally and
economically significant, and their management is usually driven by competing societal demands (Taylor, 1999a). Taken together, these issues provide a broad basis for evaluating a range of enhancement activities in a
variety of species, each represented by a large number of independent
stocks, and each aimed at fulfilling an assortment of societal needs.
Salmonid hatchery programmes have aroused considerable debate in
the last few decades. Many critics have noted that hatcheries have failed to
stem the decline of salmon stocks and, in some cases, have exacerbated
this decline (Hilborn, 1992a; Larkin, 1974; Myers et al., 2004; National
Research Council, 1996). Biological problems that may arise following
hatchery releases include changes in the genetic diversity of wild populations (Utter and Epifanio, 2002; Waples, 1991), risk of transmission of
disease pathogens to wild stocks (Elliott et al., 1997), exceeding the carrying
capacity of streams and oceans (Beamish et al., 1997; Levin et al., 2001) and
over-harvest of wild stocks due to mixed-stock fishing (Beamish et al., 1997;
Hilborn, 1985a; Unwin and Glova, 1997). On the other hand, a defence of
hatchery programmes has been mounted on the basis that evidence of these
problems is either lacking or the product of poor scientific rigour (Brannon
et al., 2004b; Heard, 2001), or that critics have specific social agendas
(Brannon et al., 2004b; Buchal, 1998; Robbins, 2004). The debate over
hatchery programmes reached a peak in the mid-1990s, which led to
advocates on both sides agreeing to rein in the rhetoric, if not substance,
Evaluation of the Effect of Hatcheries on Wild Salmon
65
of their views (Hilborn, 1999; Schramm, 1996). However, in the United
States, hatchery-related disputes moved to the courts after 2000, and legal
challenges have included the interpretation of the language within the
Endangered Species Act and a state’s right to direct recovery efforts (Alsea
Valley Alliance, 2001; Maine v Norton, 2003; California State Grange, 2005).
Admittedly, these debates are rarely as polarized outside the United States,
yet the character of these contests helps to illustrate the ecological and social
implications of salmonid hatcheries, and how other societies fall along the
spectrum of these views and responses.
The interaction between societal demands and science in the context of
the hatchery debate is a complicated one. Social advocates on both sides often
selectively employ scientific papers that further their view, while science is
often confined to researching systems that have been established by public
demand based on material needs (such as placing dams across rivers for
hydroelectric power). In light of this complicated relationship, it is important
to state from the outset that it is not our aim to enter the social debate on
whether hatcheries should, or should not, exist. Many commentators have
pointed out that hatcheries provide one of many tools that can be used in
salmonid management (e.g., Mobrand et al., 2005; Waples, 1999) and, in
many cases, viable alternatives have rarely been offered. Thus, we acknowledge that enhancement activities are likely to persist in the foreseeable future,
given their societal framework. Rather, we confine our review to the major
social and scientific issues associated with the use of hatchery-raised salmon
for conservation purposes and for fishery enhancement.
This chapter focuses largely on areas in which salmonid hatcheries could
impact wild stocks. It should be pointed out that it is not the intent of this
chapter to suggest that hatcheries should not have a role in salmonid enhancement activities, especially where their use represents an important means to
recover critically endangered stocks. For example, it is likely that certain
populations might well have gone extinct by this date without captive propagation programmes that have been largely successful (e.g., the Snake River
sockeye salmon in the northeastern United States; Utter and Epifanio, 2002).
We also attempt to identify major knowledge gaps associated with these issues.
The topic is a large one, and cannot include all aspects of the debate, and thus
we initiate our treatment by first describing the focus of this chapter.
1.1. Scope of the review
The term ‘enhancement’ takes in a wide variety of activities that humans
have engaged in on behalf of salmon species and the fisheries that capture
them. Thus, it is necessary to identify the kinds of activities that we will
review, and those that we will not consider.
66
Kerry A. Naish et al.
At one end of the continuum, there are ‘habitat enhancement’ projects
that add woody debris to streams. These ‘stream enhancement’ efforts may
or may not succeed in increasing fish densities, depending on whether the
wood that was added remained in the channel, and whether it was needed
in the first place (e.g., Cederholm et al., 1997; Roni and Quinn, 2001).
There have also been efforts at enhancing the productivity of rearing
environments. Application of inorganic nutrients or the introduction of
plant material- or marine-derived nutrients to freshwater may sometimes
accelerate juvenile salmon growth (e.g., Hyatt and Stockner, 1985; Mason,
1976; Mundie et al., 1983; Stockner and MacIsaac, 1996). Whether or not
the fast growth in freshwater is translated into more adults (the real objective) is a more complex issue (Koenings et al., 1993), but we will not review
these studies.
In addition to activities directed at juvenile habitat or growth, there
have been three main types of projects pertaining to the enhancement of
reproduction: use of in-stream egg incubation boxes, spawning channels
and hatcheries. The egg incubation box is used simply to protect developing
embryos during their vulnerable stage by forcing the upwelling of water
through gravel, where the eggs are placed. Spawning channels are artificial
channels, supplied with water diverted from natural rivers or fed by
groundwater and designed to optimize the conditions for spawning and
incubation of embryos. In most wild populations, the survival from egg
deposition to emergence is about 10–30%, depending on density and
physical factors (Quinn, 2005), but survival rates in spawning channels can
be about 50–80% (Essington et al., 2000; Hilborn, 1992b). In species or
populations where spawning habitat is the limiting factor rather than rearing
space or food, the channels can be successful. Consequently, they are most
widely used for pink, Oncorhynchus gorbuscha, and chum, O. keta, salmon
(species that migrate to sea after emerging from the gravel) and sockeye
salmon, O. nerka (that migrate to lakes) rather than for the species that rear in
streams (e.g., coho, O. kisutch; and Chinook salmon, O. tshawytscha; steelhead trout, O. mykiss and Atlantic salmon, Salmo salar) because the production of these latter species is generally limited by rearing capacity rather
than spawning capacity. We have elected to avoid reviewing the literature
on spawning channels and outplanting of egg incubation boxes, and so will
only consider enhancement projects that actually remove gametes from adult
salmon for incubation (i.e., hatcheries). This is a very important distinction
because some (though not all) of the issues related to hatcheries stem inexorably from the circumvention of natural processes of selection on the wild
fishes such as spawn timing, nest site selection, preparation and defence by
females and mate choice and competition by both males and females.
Our chapter does not include operations based on deliberately domesticated salmon that are maintained throughout their life cycle in aquaculture
67
Evaluation of the Effect of Hatcheries on Wild Salmon
facilities for the purposes of food production. These types of operations have
been the subject of a recent review (Naylor et al., 2005). Although many of
the issues associated with such activities are related to those examined here,
fish from these facilities are not intended for deliberate release and it is the
consequences of this management action that are the focus of this chapter.
Activities on anadromous salmonids in the genus Salmo (the Atlantic
salmonids) and Oncorhynchus (the Pacific salmonids) will be examined, with
a few examples from freshwater salmon within both genera and from
Salvelinus (the charrs) (Table 2.1). However, this chapter places an emphasis
on anadromous Pacific salmon for a number of reasons. First, the authors of
this chapter are most familiar with this species. Many of the issues that will
be addressed here are relevant to all species, and a comprehensive review of
the issues in Pacific salmon, with supporting evidence from Atlantic salmon,
is intended as illustrative. Second, Pacific salmon hatchery management has
largely been under governmental control since the building of the first
facility in California in 1871. Thus, the debate about enhancement has
always been a very public one, and affected by legislation and court decisions. The use of Pacific salmon hatcheries involves public lands, and a large
component of these operations is driven by the continued importance of
commercial and recreational salmon fisheries. Third, most enhancement
efforts in the eastern Atlantic are focused primarily on providing salmon for
recreational fishing in the face of the decline of native populations and, to a
Table 2.1 Scientific names and common names of salmon species (family
Salmonidae, subfamily Salmoninae) used frequently throughout this chapter
Genus
Salmo (the Atlantic
salmonids)
Oncorhynchus (the
Pacific salmonids)
Salvelinus (the charrs)
Common name (anadromous/
freshwater)
Atlantic salmon
Sea trout/brown trout
Chinook salmon
Chum salmon
Coho salmon
Cutthroat trout
(predominantly freshwater)
Masou (cherry) salmon
Pink salmon
Sockeye salmon/kokanee
Steelhead/rainbow trout
Arctic charr
Infrequent examples are named in the text.
Scientific name
S. salar
S. trutta
O. tshawytscha
O. keta
O. kisutch
O. clarki
O. masou
O. gorbuscha
O. nerka
O. mykiss
S. alpinus
68
Kerry A. Naish et al.
smaller extent, on harvest and conservation (Section 3). The more significant issues in Atlantic salmon are around the interactions between commercially farmed fish and wild stocks and, as we outlined above, outside the
scope of our review.
Finally, we recognize that hatchery releases have substantial ecological
impacts on the systems in which they operate, but do not review this aspect
in detail. Instead, the chapter focuses primarily on the interactions between
hatchery fish and their wild counterparts, since most hatchery operations are
justified on the basis of supporting the very stocks with which they interact,
either by directing harvest pressure away from wild stocks, or by supportive
breeding for the recovery of weakened stocks. It is these justifications that
will be largely examined here.
1.2. Motivations and objectives of hatcheries
The term ‘hatchery’ encompasses a broad spectrum of operations, each with
different objectives and practises. Many critiques of hatchery practises fail to
discriminate between these goals and hence the range of impacts that
various activities will have on wild populations (Allendorf and Ryman,
1997). In order to provide a full evaluation of the state of knowledge of
hatchery activities, it is therefore necessary to describe the different categories into which hatcheries fall. This attempt at definition should be
qualified. It is recognized that hatcheries have rarely been categorized
(Section 2) and many modern enhancement activities continue to lack
clear defining objectives. Further, the purposes of hatcheries may change
and yet may retain their founding broodstock. For example, a number of
hatcheries in the northeast Pacific are defined as having conservation goals,
but the majority of these were founded on the principle of providing
opportunities for harvest. Thus, current hatchery practises are most likely
to fall along a continuum of the definitions given here. Finally, many of the
terms used below have been applied loosely. For example, ‘supplementation’ has been used to describe activities varying from conservation to
fishery enhancement. ‘Stocking’ has been used in a generic sense to describe
the release of cultured fish into the wild, but has also specific definitions in
the context of enhancement, mitigation and conservation activities (Cowx,
1998). Here, we attempt to more clearly define many of these categories
below.
Hatcheries are classified broadly by having either conservation or fishery
objectives. The former are intended to restore extinct, endangered or
threatened populations or to reduce the risk of extinction. The latter are
used to increase population sizes for fishery opportunities. The aquaculture
classification of Utter and Epifanio (2002) is largely followed here, with an
emphasis on salmon hatcheries.
Evaluation of the Effect of Hatcheries on Wild Salmon
69
Captive broodstock hatcheries are conservation oriented, with the sole
purpose of maintaining populations that cannot be supported in their wild
habitat for even part of their life cycle (Utter and Epifanio, 2002). Typically,
the broodstock is maintained in captivity until the population threats
have been removed, at which point the captively reared fish will be
restored. For example, populations of Atlantic salmon indigenous to the
Iijoki and Oulujoki rivers in Finland have been maintained in captivity since
dam construction removed suitable spawning habitat (Saisa et al., 2003) and
may be reintroduced as part of the International Baltic Sea Fishery Commission’s ‘Salmon Action Plan’ to increase wild population returns. Sockeye
salmon returning to Redfish Lake, part of the Columbia River drainage
system on the west coast of North America, declined to very small numbers
of returning adults from 1991 to 1996; all were taken into captivity, and a
portion of the population has since been utilized in a continuing programme
of reintroduction (Utter and Epifanio, 2002).
Supplementation hatcheries also share a conservation ethic. Waples et al.
(2007) defined supplementation as ‘the intentional demographic integration
of hatchery and natural production, with the goal of improving the status of
an existing natural population’. While the intention is to incorporate
the broodstock into wild stocks, the degree of integration can vary, with
different outcomes (Section 5.1). Supplementation activities have been
implemented extensively on the west coast of North America in an attempt
to mitigate losses due to anthropogenic activities such as dam construction,
forestry, agriculture or urbanization (Section 3.1).
Production hatcheries, or fishery enhancement hatcheries (Utter and Epifanio,
2002), are hatcheries that seek to augment the abundance of salmon in order
to increase fishing opportunities. ‘Ocean ranching’ has been defined as the
release of ‘juvenile specimens of species of fishery importance raised or
reared in hatcheries and nurseries into the sea for subsequent harvest at
the adult stage or manipulating fishery habitat to improve growth of the
wild stocks’ (Mustafa, 2003), which can often include domesticated stocks,
and thus falls under this category. One potential outcome of such activities
is that the resulting demographic increases may redirect harvest pressures
away from natural production. In many cases, the wild populations are
viable. Production hatcheries are used extensively throughout the world.
For example, many European countries release anadromous Atlantic
salmon and brown trout (S. trutta) populations; many hatchery strains are
derived from exogenous stocks. The pink salmon fishing industry in Alaska
is supported by releases from production hatcheries in Prince William
Sound (PWS; Section 4).
Mitigation hatcheries are production hatcheries that have typically been
founded to compensate for lost harvest opportunities following substantial
reduction or extirpation of an indigenous stock due to losses of habitat or
other anthropogenic activities (Utter and Epifanio, 2002). Such hatcheries
70
Kerry A. Naish et al.
have been established with the understanding that the habitat that is essential
for part of the salmon life cycle will not be replaced within the predictable
future, and thus continuation of the population is dependant on artificial
propagation. The most extensive programme in this category is the chum
salmon programme operated in Japan (Section 4). In many cases, mitigation
hatcheries are maintained in order to meet a mandate imposed by prior
rights of a group to the fishery. For example, access to the Chinook salmon
fishery was lost by the native peoples of the Columbia River Basin following the construction of the Grand Coulee Dam, and mitigation hatcheries
were constructed in response to tribal treaties (Utter and Epifanio, 2002).
Given increasing concern that endangered or threatened stocks may be
caught in mixed-stock fisheries (Section 5.4), the release of hatchery fish
at a remote acclimation site has been explored. In the Columbia River, for
example, an ongoing programme has placed hatchery juveniles into a net
pen at a location that has not been frequented by migrating endangered
salmon stocks (ISRP/IEAB, 2005). The project’s intent is that the fish
acclimate and return to the remote site where they can be harvested, thus
reducing risk to the endangered wild stocks.
Hatcheries providing fishing opportunities for non-indigenous fisheries (Introduced
fish) are production hatcheries (including ‘put and take’ aquaculture) operated to provide harvest or recreational fishing opportunities on species that
are exotic to the region in which they are released (Utter and Epifanio,
2002). For example, rainbow trout has been extensively introduced to
countries in the Southern Hemisphere, and Chinook salmon has been
introduced to the North American Great Lakes and New Zealand. Many
of these operations involve a single introduction, while others are maintained by hatchery programmes (e.g., Chinook salmon in the Great Lakes)
with the notion that the introduced species may go extinct once such
activities cease.
1.3. Content overview
In the following sections, we provide a social context for an evaluation
of hatchery operations through an overview of the history of hatchery
activities and the political dynamics associated with hatchery programmes.
We then provide a survey of the geographical extent of anadromous
and freshwater hatchery programmes throughout the world. In an attempt
to understand the impacts of such programmes, we evaluate in detail
the types of biological risks that hatchery programmes may pose to wild
stocks of salmon species and return to the social aspect of such programmes
by examining the economic issues associated with hatchery programmes.
The chapter ends with a discussion on the risks associated with conservation
and fishery enhancement hatcheries, and on the social drivers and costs of
hatchery activities.
Evaluation of the Effect of Hatcheries on Wild Salmon
71
2. Historical Overview of Hatchery Activities
Western salmonid fish culture dates to eighteenth-century Westphalia.
In 1747, army officer and naturalist Ludwig Jacobi used ancient Asian
techniques to fertilize and rear trout in an artificial environment, but not
until he published his memoirs in 1770 did his achievements gain notice.
In the next century, a number of Europeans emulated his efforts, including
Karl Lund, Karl Vogt, John Shaw, Joseph Rémy and Antoine Géhin. Each
refined and expanded upon Jacobi’s work by innovating new methods for
raising an array of freshwater and anadromous species. These were limited
efforts by individual enthusiasts and scientists, people primarily interested in
studying and reproducing small stocks of fish for fulfilment or profit rather
than for professional or industrial interests (Marsh, 1857; Prince, 1900).
The development of a hatchery programme, in the modern sense of the
systematic management of fish and fisheries, required a more institutionalized approach. France was the first to adopt this tack. In 1850, inspired by
the work of Rémy and Géhin at Bresse and the writings of naturalist
Armand de Quatrefages, the French Minister of Agriculture built a fishbreeding station at Huningen to repopulate the Rhine and Rhône rivers.
Soon Switzerland, Germany, England and Scotland had established similar
efforts to restore their fisheries. In most cases, these hatcheries were
designed to serve both angling and commercial interests, and while none
produced immediate, demonstrable successes in rebuilding stocks, they
represented a new movement in fisheries policy that drew the attention of
North Americans. Interest in fish culture began relatively late in Canada and
the United States. The first documented case of reproducing trout was by
Ohioan Theodatus Garlick in 1853. This work quickly gained notice.
Fishmongers and anglers saw a hope for reversing decades of decline in
the fisheries. As a result, they lobbied legislators throughout the eastern
states to establish fish hatcheries, and from 1855 to 1857, Massachusetts,
Connecticut and Vermont commissioned studies on approaches to restore
local fisheries. Each report became a treatise on the technical and sociopolitical implications of breeding fish, and each concluded that while fish
culture had not yet restored a troubled fishery, the technology held tremendous
promise for ameliorating the material consequences of progress.
Government support for hatchery programmes did not begin in the
United States until the end of the Civil War. In 1865, New Hampshire
built the first state-run hatchery, and California, Connecticut, Maine, New
Jersey, New York, Pennsylvania and Rhode Island followed by 1870
(Bowen, 1970). These operations were primarily dedicated to game fisheries. The major expansion of fish culture in the next decade was more tied
to commercial interests however. In 1867, at the request of several New
England fish commissions, a for-profit fish culturist named Seth Green bred
72
Kerry A. Naish et al.
shad eggs on the Connecticut River; the next year he extended this work
to the Hudson, Potomac and Susquehanna rivers (Goode, 1881; Norris,
1868). In 1871, Charles Atkins and Livingston Stone persuaded the state of
Maine to build a hatchery at Bucksport specifically for hatching Atlantic
salmon, and Frank Clark hatched whitefish on the Great Lakes in 1872
(Milner, 1874; Stone, 1897).
Canadian fish culture loosely paralleled American activities. In the late
1850s, Richard Nettle first hatched brook trout and Atlantic salmon in
Quebec city. By the mid-1860s, enthusiasts were hatching these species and
lake herring on the St. Lawrence and Great Lakes. The most prominent
culturist was Samuel Wilmot, a zealous self-promoter who billed himself
as the leading fish culturist in North America. In 1866, he persuaded the
province of Ontario to hire him as a fishery officer and fund his hatchery
at Newcastle on Lake Ontario. Two years later, the Dominion of Canada
took over the hatchery and made Wilmot an officer of the Department of
Marine and Fisheries. In the early 1870s, he built additional salmon hatcheries on the Restigouche (1872), Miramichi (1873), Gaspe (1874) and
Tadoussac (1875) rivers, and in 1876 he was promoted to Superintendent
of Fish Breeding for Canada. The Dominion built its first hatchery
for Pacific salmon on the Fraser River in 1884. Wilmot overstated his
achievements on occasions—he had not been the first to hatch salmon or
whitefish—but he was, without doubt, the driving force behind federalization of artificial propagation in North America (Prince, 1900;
Lasenby et al., 2001).
Influenced by Canada’s hatchery programme, the US Congress soon
followed suit. It created the US Fish Commission (USFC) in 1871, and the
following year the first commissioner, Spencer Fullerton Baird, assigned
Livingston Stone to transplant salmon eggs from the Sacramento River to
eastern streams. In 1873, Baird hired Seth Green to plant shad in Midwestern
streams and the Sacramento River. Other American fish culturists refined
methods for inseminating and incubating eggs, and Baird turned his USFC
employees into an army of researchers, surveying habitat and species abundance, investigating egg development and experimenting with fish feeds.
As in Canada, federal support for fish culture grew. By the 1880s, Congress
was funding hatcheries from the Bay of Fundy to San Francisco Bay, from the
Columbia River to the Savannah River and from the Gulf of Mexico to the
Great Lakes. By the end of the century, Canada and the United States had
built extensive hatchery programmes. Almost every major fishing stream was
affected by at least one federal, state, provincial or private hatchery, and fish
culture had become an intrinsic tool of managing game and commercial
fisheries across North America (Allard, 1978).
Both countries also conducted extensive fish transportation programmes. The motivations for transplant programmes were complex.
Some of it was driven by emigrants’ desires to recreate the ecologies of
Evaluation of the Effect of Hatcheries on Wild Salmon
73
home or, as Wisconsin’s Commissioners of Fisheries boasted in 1888, in
hopes of creating one vast ‘Summer Paradise’ (quoted in Bougue, 2000; see
also Lampman, 1946). Economic opportunity was another major influence
in shipping salmon around the world (Taylor, 1999a), and, as Spencer
Fullerton Baird himself admitted in 1877, political considerations also
drove transplant projects: ‘The object is to introduce [fish eggs] into as
many states as possible and have credit with Congress accordingly. If they
are there, they are there, and we can so swear, and that is the end of it’
(quoted in Allard, 1978). Using railways and steamships, hatchery programmes sent species to every corner of the continent and beyond. In the
last quarter of the nineteenth century, the USFC and Canada Department of
Marine and Fisheries transplanted a menagerie of species. Pacific salmon
were transplanted to the Great Lakes, South Dakota’s Belle Fourche River,
and the Great Salt Lake, not to mention Europe, Asia, South America and
Australasia (Colpitts, 2002). By 1900, global hatchery ecology was emerging
in which salmonids played a key, but hardly singular, role. Brown trout
were shipped from England to California, California Chinook salmon were
sent to New Zealand, Japanese koi (Cyprinus carpio carpio) were cultivated in
Massachusetts, Rhode Island shad were released in Oregon streams, Oregon
steelhead were exchanged with Germany, German carp (C. carpio carpio)
were placed in the Great Basin and black bass were released just about
everywhere (Allard, 1978; Bogue, 2000; Bowen, 1970). Salmonids were far
from the only species introduced to new environments.
The paradox of most hatchery programmes was that institutional successes went hand-in-hand with ecological disaster. Despite growing fiscal
support—Congress increased the Division of Fish Culture’s budget from
$25,000 in 1873 to $331,000 by 1900—optimism was deserting fish culturists (Cart, 1968). North American and European salmon runs had been
declining for centuries (Netboy, 1980). Whitefish populations in the Great
Lakes had collapsed in recent decades, as had shad and alewife stocks along
the Atlantic (Bogue, 2000; McPhee, 2002; Steinberg, 1991). Shad (Alosa
sapidissima) and striped bass (Morone saxatilis) were colonizing west coast
streams, but few markets existed for these species and some observers
worried that exotics would disrupt native species. German carp were
denuding rearing habitat for western North American trout (Langston,
2003), and brown trout and rainbow trout were displacing indigenous
species in Australia and New Zealand (Crowl et al., 1992). The more things
fell apart, the more politically potent hatchery programmes seemed. By the
early twentieth century crises had become fish culture’s raison d’etre. Declining stocks and degraded habitat made artificial propagation the default
solution for many governments. And if the results often fell short, the
achievements of these programmes were no less significant. North American
fishery agencies had become the gold standard for fishery management.
The USFC had developed into a model agency for supporting fisheries
74
Kerry A. Naish et al.
through economic surveying, scientific research and artificial propagation,
and professional scientists were turning the American Fisheries Society and,
after 1912, the Biological Board of Canada, into premier organizations for
fisheries science ( Johnstone, 1977; Smith, 1994).
The principal exceptions to the mounting problems of hatchery work
were a select few game fish species. Brown trout (S. trutta), rainbow trout
(O. mykiss), bass (Micropterus dolomieu and M. salmoides), catfish (Ictalurus
furcatus, I. punctatus, Ameiurus catus), crappie (Pomoxis nigromaculatus), perch
(Perca flavescens, Morone americana) and pike (Esox lucius) adapted well to
pond culture, but, equally important, each also had enthusiastic angling
constituencies. In Europe, angling had been a primary motive for planting
brown trout and Atlantic salmon since the 1700s. Since the majority of
streams were privately owned, most releases were performed by individuals
and fishing associations. Conditions in Japan and North America were
different. Most waters were publicly owned, so both individuals and the
state released fish into the waterways. In Japan, for example, the Agricultural
Bureau initiated hatchery work in 1878 on streams in Niigata-ken, Naganoken, Ibaraki-ken and Hokkaido. Budgets and plantings grew significantly
by 1900 (Imperial Fisheries Bureau, 1904). Canada followed a slightly
different path. At first small, privately run hatcheries operated in the Maritimes, Quebec and Ontario, and a few provinces gained some authority
over fisheries by 1900. The federal government retained primary authority
for oceans and ultimate authority for all other fisheries, however
(Thompson, 1975). Thus, similar to Japan, the Dominion of Canada operated in support of both angling and commercial interests well into the
twentieth century.
Hatchery development in the United States was more complicated due
to a constitutional division of authority. The federal government held
jurisdiction of seas, navigable rivers, and territories, but states ruled all
other waters including the oceans within three miles of the coast.
In practise, this meant that the USFC and US Bureau of Fisheries (USBF)
propagated commercial species that frequented the seas or interstate waters
such as the Great Lakes and Columbia River (Allard, 1978; Taylor, 1999a).
Federal hatcheries planted game fish in national parks and federal forests, as
well as across Alaska, but by 1900 states were taking over the primary
responsibility for much of the hatchery work (Pritchard, 1999). Fishery
agencies were evolving into a huge apparatus for breeding and distributing
fish. Large and growing bureaucracies existed in nearly every state, and an
immense amount of fish were bred and planted each year by fish culturists
(Lampman, 1946; Reiger, 2001; Taylor, 1999a).
The next half-century was a period of elaboration rather than change.
The size and scope of fishery agencies continued to grow as stock depletion
and habit decline accelerated after 1900. The policy of compensating for
problems with fish culture, begun in New England in the 1860s, extended
Evaluation of the Effect of Hatcheries on Wild Salmon
75
to the Pacific Northwest in the 1910s when Washington State adopted
an ‘in lieu’ policy that would eschew fish-ways if the dam owner funded
a hatchery (Steinberg, 1991; Taylor, 1999a). Although few governments made this formal policy, all were increasingly inclined to mitigate
losses rather than restrain development. When alewife (A. pseudoharengus),
eel (Anguilla rostrata), salmon, shad and sturgeon (Acipenser transmontanus)
populations dropped because of habitat loss, agencies in Canada, France,
Great Britain, Ireland, Japan, Russia and the United States responded by
advocating fish-ways and hatcheries. Few wild fish populations recovered,
but the bureaucracies overseeing them thrived (Netboy, 1980; Pritchard,
2001; Taylor, 1999a).
Hatchery programmes also experienced an institutional mitosis. The
fisheries had been torn by rivalries for centuries. Towards the end of the
nineteenth century, state legislatures in the United States began to formalize
the divisions between sport and commercial interests in separate fish and
game departments. Fish culture was influenced by these events as it was
embedded in such agencies. Individual hatcheries began to specialize,
serving the desires of constituencies interested only in market or game
fish, or specifically in one species. In the Pacific Northwest, for example,
Oregon’s Fish Commission began to favour Chinook salmon in coastal
streams where coho had been the dominant native fish, while the Game
Commission planted bass and walleye (Sander vitreus) in inland streams
where salmon and trout had predominated (Taylor, 1999a). Transplanting
coho to Lake Michigan in the 1960s precipitated similar upheavals in the
Great Lakes (Chiarappa and Szylvian, 2003). Institutional specialization and
ecological reorganization occurred across many Northern Hemispheric
fisheries during this period, including Hokkaido Island, Vancouver Island,
Yellowstone National Park and the Barents Sea and White Sea (Harris,
2001; Imperial Fisheries Bureau, 1904; Pritchard, 2001).
Scientific research also became an increasingly important institutional
activity. The USFC had been created as a research agency in 1871, and
Commissioner Baird insisted that research remain a high priority even as
fish culture dominated budgets (Allard, 1978). Annual reports included
essays by top scientists, and the Bulletin of the Bureau of Fisheries was a science
publication from its inception in 1881. The other major research publication in the United States was, of course, the Proceedings of the American FishCultural Association and its successor, the Transactions of the American Fisheries
Society (Smith, 1994). The Biological Board (later the Fisheries Research
Board) of Canada began its own research programme in 1912 ( Johnstone,
1977), and Europeans had developed a tradition of scientific cooperation
long before the International Committee on the Exploration of the Seas
(ICES) formed in 1902 (Rozwadowski, 2002). In addition, various state and
provincial fishery agencies in the United States and Canada began their own
research programmes (Taylor, 1999a).
76
Kerry A. Naish et al.
The impact of this science on hatchery operations was uneven. Some
research proved functional, such as designing efficient fish-ways and fish
screens, understanding predators and parasites and refining effective and
economical feeds (Bowen, 1970; Eicher, 1970; McHugh, 1970). Fish
tagging studies confirmed the home-stream theory for salmon and
influenced fishing treaties (Taylor, 1999a). Administrators of hatchery programmes in Canada and the United States embraced this work enthusiastically, but they rejected critical research. Harley White’s study of Prince
Edward Island trout (White, 1924), Willis Rich’s statistical analysis of
Columbia River hatcheries (Rich, 1922) and Russell Foerster’s work of
Cultus Lake sockeye (Foerster, 1936) all cast doubts on claims that fish
culture had made significant differences in the size of salmonid populations.
Although economic considerations were also important, this research did
influence decisions to halt hatchery work in British Columbia and Alaska
during the 1930s. Fish culturists responded by attacking White and ignoring
the broader implications of his research ( Johnstone, 1977; Taylor, 1998b).
By 1950, salmonid hatchery programmes in North America were on a path
best described as scientific yet without scrutiny (Hilborn and Winton, 1993;
Lichatowich, 1999; Taylor, 1999a). Similarly aggressive hatchery programmes have been initiated around the Baltic Sea in recent decades, but
here, too, familiar problems with disease, interbreeding, mixed-stock fisheries and declining wild populations have emerged (Khristoforov and
Murza, 2003; Paaver et al., 2003).
Research during the 1940s and 1950s created the technical and intellectual foundation of the last half-century. Studies of parasites and diseases,
and advances in medical and food science led to new prophylactic treatments. Fish culturists devised ways to address epizootics, and extruding
machines produced pelleted feeds that avoided age-old problems with
nutritional deficiencies and contamination (Stickney, 1996). These innovations allowed fish culturists to raise more fish, more economically for far
longer. As hatcheries began to raise greater numbers of much larger fish,
fishery bureaucracies expanded again. Great Lakes managers started replacing failing whitefish stocks with trout, and west coast hatchery workers
used new feeds to retain Pacific salmon to smoltification (Chiarappa and
Szylvian, 2003; Taylor, 1999a). Meanwhile, the Scandinavians succeeded in
raising rainbow trout and then Atlantic salmon from eggs to harvest. By
1960, a far more technically based era of salmonid culture had emerged
(Sedgwick, 1982).
Developments in North America since 1960 have exposed lingering
problems with hatchery programmes. Practises learned from the commercial farming of Atlantic salmon, rainbow trout and several Pacific salmons,
and the transporting of juvenile Pacific salmon around dams with trucks and
barges were regarded as significant advances in the 1960s and early 1970s.
The salmon hatchery programme was even revived in British Columbia,
Evaluation of the Effect of Hatcheries on Wild Salmon
77
where Canada’s Department of Fisheries and Oceans (DFO) constructed a
number of production hatcheries, and the International Pacific Salmon
Fisheries Commission (IPSFC) began to experiment with artificial spawning channels (Mead and Woodall, 1968; Roos, 1991). The appearance
of success faded in the 1970s as worries surfaced about the deleterious effects
of mixed-stock fisheries, genetic interactions with wild stocks, threats of
disease transmission, mass hatchery releases out-competing wild stocks and
continuing declines in many fisheries (Lichatowich, 1999; Orr et al., 2002;
Taylor, 1999a). Many western inland trout and Pacific salmon stocks have
declined due to habitat loss, competition from and hybridization with
exotic species (Cone and Ridlington, 1996; Leary et al., 1995). Hatcheries
on the Great Lakes produced so many lake trout and Pacific salmon that
they annihilated the last significant whitefish stocks (Chiarappa and
Szylvian, 2003). In addition, the collapse of Labrador’s Atlantic salmon
fishery in the 1970s revealed the inability of hatcheries in the eastern United
States and the Maritime Provinces to compensate fully for the effects of
intense harvests and declining habitat (Netboy, 1980).
During this period, hatchery programmes in Europe and Scandinavia
showed mixed results. In the 1970s and 1980s commercial salmon farms in
Norway, Scotland and Ireland gained a foothold in the marketplace due to
declining salmon fisheries in the western Atlantic and northeastern Pacific.
The Norwegian salmon farming programme at the time also aided small,
outlying communities; however, a market collapse in the 1990s, caused in
part by competition from a rapidly expanding Chilean salmon aquaculture
industry, led to industry consolidation by a few corporations and growing
concerns about the ecological impacts of industry practises (Milstein, 2003).
More recently, efforts to rebuild extinct runs in Belgium’s Meuse River
basin revealed, once again, that the success of artificial propagation depends
on healthy habitat and competent fish passage technologies (Prignon et al.,
1999), while work on the Asón and Nansa rivers in Spain demonstrated that
transplanted salmon stocks fare more poorly than native wild stocks
(Verspoor and de Leaniz, 1997). Species hybridization between Atlantic
salmon and trout populations in Sweden has been attributed to the release of
too many fish by hatcheries ( Jansson and Oest, 1997), and hatcheries
throughout the Gulf of Bothnia have contributed to genetic homogenization in wild Atlantic salmon populations (Khristoforov and Murza, 2003;
Paaver et al., 2003; Vasemägi et al., 2005).
The persistence of these problems, most of which have plagued salmonid
hatcheries for a century or more, has inspired ever more urgent calls for
reform and even termination. Critics have demanded that salmonid hatcheries be independently evaluated, that hatchery managers define goals which
are rigorous and testable and that administrators develop policies based on
the best available science (Brown, 1982; Lichatowich, 1999; Mobrand et al.,
2005; National Research Council, 1996; Taylor, 1999a). At the same time,
78
Kerry A. Naish et al.
however, it is clear that some wild salmonid stocks are in such great
peril that the only hope for recovery seems to be the sort of intensive
hatchery programmes that exacerbate problems within the region
(Schiewe et al., 1997). Thus, the last 15 years has been a period of reassessment. History casts a troubling light on the practises and goals of many
salmonid hatchery programmes, but no consensus has emerged yet about
the future of these programmes.
3. Political Dynamics of Hatchery Programmes
Although usually considered a scientifically based activity, salmonid
hatcheries must also be understood as political technologies fundamentally
shaped by economic and cultural concerns. Scientists have been key players
in creating and shaping hatchery programmes, and fish culturists have
conducted important research, yet the history of these programmes reveals
that science has often played only a secondary, legitimizing role in policymaking, or has been implemented to maintain the status quo. Social and
economic politics has been the primary influence on governmentsponsored hatchery programmes in the last two centuries. Thus, examining
the historical context of the political aspects of fish culture is essential for
understanding the development and consequences of salmonid hatcheries
during this period.
This historical perspective is particularly crucial in the current, highly
politicized climate that surrounds salmonid hatcheries. In recent years some
programmes have earned withering criticism. Scholars have cited a number
of problems, including insufficient scientific rigour, economic worth and
ecological viability in many enhancement programmes (Anonymous, 2004;
Hilborn and Eggers, 2000; Myers et al., 2004; National Research Council,
1996). With less care, critics have also tried to explain the technology’s
origins by variously blaming zealous founders, short-term thinking and even
Francis Bacon for replacing a holistic appreciation of nature with instrumentalist thinking (Cone and Ridlington, 1996). The disparity between
careful studies of technical issues and vague assertions about historical roots
has hindered our understanding of why science has not been a stronger
guide in hatchery policies (McEvoy, 1986).
Closer attention to the past illustrates how thoroughly blurred politics
and science were in early hatchery programmes. From the beginning,
proponents of hatcheries noted that manual fertilization of fish eggs produced far greater hatch rates than natural reproduction. Many were led by
an arithmetical logic to predict astonishing increases in fish populations by
even meagre efforts. According to US Fish Commissioner Spencer Fullerton Baird, fish culture would allow the government ‘not only to maintain
Evaluation of the Effect of Hatcheries on Wild Salmon
79
the present supply of fish, but to increase it if desirable’ (Baird, 1874).
Framing these insights in agrarian terms, proponents argued that an acre
of water was more productive than one, five, or ten acres of land. One even
insisted that ‘one acre of the waters of any salmon stream in Oregon . . . is
worth more as a medium for the product of a food supply than forty acres of
the best land in the State’ (Hume, 1893). Such boasts drew notice in many
countries, but what made fish culture most appealing for government
officials were its political implications. Few could resist a technology that
French scientist Jules Haime claimed (Marsh, 1857) was ‘destined to
solve one of the important terms of the great problem of cheap living’
(McPhee, 2002).
The cultural context in which fish culture emerged was another factor in
its popularity. Although the rhetoric surrounding fish breeding emphasized
bountiful harvests, contemporary politicians were less motivated by Malthusian fears of hunger than by the growing conflicts between aquatic and
terrestrial interests. George Perkins Marsh, who was squarely on one side of
this contest, explained to Vermont’s legislature in 1857 that ‘We cannot
destroy our dams, or provide artificial water-ways for the migration of
fish. . .; we cannot wholly prevent the discharge of deleterious substances
from our industrial establishments’, nor was it probable ‘that any mere
protective legislation, however faithfully obeyed, would restore the ancient
abundance of our public fisheries’. For Marsh and others the ‘final extinction of the larger wild quadrupeds and birds, as well as the diminution of
fish, and other aquatic animals, is everywhere a condition of advanced
civilization and the increase and spread of a rural and industrial population’
(Marsh, 1857). Destruction was thus the sad but inevitable cost of progress.
Most people exploring the feasibility of fish culture programmes were
not worried about an imminent implosion of food sources. Rather, it was
growing contests over dwindling fish stocks that forced politicians in many
countries to address the issue. In other words, it was the political implications
of finite resources, not hunger, which first inspired modern hatchery programmes. Conflicts among fishers led France to build a government hatchery at Huningen in 1850 to repopulate the Rhine and Rhône rivers.
Lobbying by sport and commercial interests led New England states and
Canada to fund trout, shad and salmon work during the 1860s (Prince,
1900; Reiger, 2001), and tourism agendas influenced Japan to build a
hatchery on Lake Chuzenji in the 1890s (Imperial Fisheries Bureau, 1904).
Economic possibilities helped leverage funding, but as Spencer Baird
noted in an 1875 letter to the US Senate, the tangled problems of regulation
were also salient. ‘In the United States’, he observed, ‘it has always
been found very difficult to enforce laws in regard to the fisheries. When
passed by the States they involve an extensive police for their execution’,
and, crucially, no state had built such a force. Thus he ‘unhesitatingly’
recommended that instead of the passage of protective laws which cannot
80
Kerry A. Naish et al.
be enforced except at very great expense and with much ill feeling, ‘measures
be taken . . . for the immediate erection of a hatching establishment on the
Columbia river, and the initiation during the present year [1875] of the
method of artificial hatching of these fish’ (Baird, 1875).
Here were the political underpinnings of the modern hatchery.
Government officials in many countries, most of whom knew or cared
little about fish or about fish culture, nevertheless supported large government budgets for these programmes not simply because the technology
promised to sustain fisheries but because Haime and Armand de Quatrefages
in France, Frank Buckland in Britain, Samuel Wilmot in Canada and Marsh
and Baird in the United States assured officials that hatcheries would avert
the social conflicts between declining fish stocks and modern development
(Bogue, 2000; Gardner-Thorpe, 2001; MacCrimmon, 1965; Prince, 1900;
Taylor, 1999a). For governments and the public alike, much of fish culture’s
appeal emanated from its panacean qualities (Taylor, 1999a). Hatcheries
seemed to alleviate the need to make hard choices about limiting access to
fish or habitat.
Yet governments did make choices and did intervene, and one of fish
culture’s first impacts was on property rights. By 1850 many state legislatures
in the United States had already forced dam owners to maintain fish-ways to
protect migrating species (Steinberg, 1991). Such limitations became precedents for further restrictions to facilitate artificial propagation. Between
1861 and 1865 the British Parliament passed acts for England, Ireland,
Scotland and Wales that exempted fish culturists from regulations and
limited private claims on some Scottish and Irish streams (Great Britain,
1861, 1862, 1863, 1865). Americans were more hesitant. Marsh advised
Vermont legislators to rely ‘upon the enterprise and ingenuity of private
citizens’ and to create economic incentives by according property rights to
fish produced by entrepreneurial hatcheries (Marsh, 1857). Vermonters did
not agree, believing that fish and game should remain free until capture
(McEvoy, 1986). The same held true across the United States, Canada and
Japan, but experiences varied in European nations. In countries with customs of privatized fish and game, parliaments had to finely tune rules to
specific bodies of water before initiating public or private hatcheries during
the last two centuries (Prat, 1998).
Even then, however, fish culture did not prevent further restraints on
people’s interests in fish, water and land. One unavoidable conflict created
by hatcheries was the need to harvest fish. The very effort to enhance fish
stocks put fish culturists in conflict with other resource users. One hatchery
on Oregon’s Clackamas River inspired repeated conflicts because the
hatchery weir blocked both migrating salmon from upstream settlers and
logs floating to downstream lumber mills (Taylor, 1999a). Far more common were the ways hatcheries abetted the dispossession of resources. In
Canada (the Maritimes, St. Lawrence and British Columbia), in the United
Evaluation of the Effect of Hatcheries on Wild Salmon
81
States (New England, the Great Lakes and Pacific Northwest) and in Japan
(Hokkaido Island), hatcheries divided indigenous peoples from their
salmon, trout and whitefish fisheries (Harris, 2001; Newell, 1993;
Parenteau, 1998; Taylor, 1999a; Walker, 2001). In 1877, for example,
Livingston Stone evicted Clackamas Indians from their fishery to prevent
competition with his hatchery (Taylor, 1999a). Over time, hatcheries built
to mitigate dams also reorganized the spaces of reproduction in ways that
deprived Native peoples of their historic fisheries (Allen, 2003; Evenden,
2004; Taylor, 1999a). In a few instances, such as the introduction of Pacific
salmon to the Great Lakes, exotic transplants for the benefit of recreational
anglers also undermined commercial and aboriginal fisheries (Chiarappa and
Szylvian, 2003). In recent years, however, the lines of these conflicts have
shifted. In Europe, notably in Britain, and in North America, growing
concerns about the genetic implications of declining stocks of wild salmonids have increasingly pitted conservation groups against the only remaining
significant group of harvesters: anglers.
Another consequence was that salmonid hatcheries became contested
prizes. Relentlessly shrinking fish stocks exacerbated existing tensions
among fishers, and the rarer a population or species became, the more
hatcheries became an explicit prize in political battles. In Canada and the
United States, commercial and sport fishers engaged in what was essentially
an ecological tug of war, battling over the control of hatcheries and the
release of game and commercial species. In Oregon and Washington,
industrial fishers also fought over which hatcheries and rivers would receive
financial support (Parenteau, 1998; Taylor, 1999b). In the Great Lakes,
Pacific Northwest and Japan, commercial and sport fishers also tried to
deny aboriginal fishers access to hatchery fish (Blumm, 2002; Chiarappa
and Szylvian, 2003; Shigeru, 1994). The advent of salmon aquaculture has
had similar implications in Norway, where commercial fishers were
excluded from harvesting Atlantic salmon to protect sport and farming
interests (Otterstad, 1998). On the other hand, Alaskans chose to bar
Atlantic salmon aquaculture from their state and restrict activities to ‘private,
non-profit’ (PNP) hatcheries, fearing that the ecological and economic
destabilizations that have accompanied farming operations elsewhere
would negatively affect their Pacific salmon fisheries (Herbst, 2003).
Such consequences illustrate why technology must be understood
within its historical context. Fish culture was not inherently racist or classist,
even if some fish culturists were bigots (Chiarappa and Szylvian, 2003), but
when hatcheries were used to serve the interests of some at the expense of
others, then technology was politicized. This held true not only in those
internecine battles that plagued sport and industrial fisheries during the last
two centuries ( Jacoby, 2001; Thompson, 1975), but, more insidiously, it
also applied to industrialization. Fish hatcheries meshed seamlessly with an
ideology of production that defined value narrowly in terms of economic
82
Kerry A. Naish et al.
wealth and regarded development as a natural good. Once industrial growth
emerged as the ultimate goal, fish culture thrived politically because it
promised to enable such an agenda. This partnership between fish culture
and development began very early. The first hatcheries were built in Europe
and eastern North America in the 1850s and 1860s to stem declining fish
stocks due to habitat loss and industrialization (Kinsey 2006; Netboy, 1968;
Pisani, 1984; Smith, 1994). In western North America, fears of overfishing
were the fulcrum for establishing hatchery programmes (Harris, 2001;
Taylor, 1998b). In all cases, fish culture was popular because neither the
public nor legislators wanted to regulate economic activity. This pattern of
relying on technological solutions was a deliberate, politically influenced
choice that continued throughout the twentieth century (Meffe, 1992).
After 1910, state, provincial and federal governments allowed entrepreneurs
to mitigate damaging activities by funding state-run fish hatcheries (Blumm,
2002; Evenden, 2004; Harris, 2001).
This ‘in-lieu’ policy saw its most aggressive application with dambuilding programmes in Washington State on the Cowlitz, Lewis, White
Salmon and Columbia rivers, but it was a popular solution from Japan to
Western Europe. In 1923, Herbert Hoover declared ‘We have only to
preserve and increase the supplies of our fish by moderate restraint and
scientific propagation’ (Taylor, 2004). In the 1950s, France dammed the
Rhône for navigation and hydroelectricity and hoped hatcheries would
mitigate fisheries losses (Pritchard, 2001). In the 1960s, the Hokkaido Development Agency confiscated Ainu lands to build the Nibutani Dam for the
sake of national progress (Sonohara, 1997). The culture that inspired each
event has not changed in many cases. In 2003, the US President George W.
Bush insisted that dams posed no problems for salmon runs (Reichmann,
2003). In all cases governments essentially institutionalized George Perkins
Marsh’s assumption that damage to wild stocks was an unavoidable consequence of progress. In a few places such as the United Kingdom, fishery
institutions have begun to embrace a more risk-adverse approach to managing fragile stocks of salmonids, but in these cases the role and operation of
hatcheries has also undergone significant change. Hatchery programmes have
continued to thrive because they remain the most politically appealing, least
controversial way to address the material consequences of development.
The political appeal of fish hatcheries was underscored by the technology’s resistance to scientific criticism. Researchers have noted many basic
flaws with fish culture. In the 1890s Knut Dahl, Johan Hjort, Ernest Holt
and John Moore cast doubt on the efficacy of fish culture by citing poor
statistics, insufficient controls and inconsistent results. These were leading
European scientists, yet government officials ignored them in favour of
voices that said what politicians wanted to hear (Smith, 1994). Criticism
nevertheless increased with time. In the 1910s Americans Barton Warren
Evermann and Willis Rich lodged similar complaints, as did Canadians
Evaluation of the Effect of Hatcheries on Wild Salmon
83
Russell Foerster, William Ricker and Harley White in the 1920s and 1930s
(Taylor, 1998a, 1999a). After World War II, scientists expanded the litany of
criticisms to include problems related to genetics, mixed-stock fisheries and
habitat loss. Peter Larkin and James Lichatowich have been only two of many
scientists who continued to point out the fundamental problems with
salmonid enhancement in recent decades (Larkin, 1974; Lichatowich, 1999).
More than any other aspect of fish culture, it has been the selective
acceptance of this criticism that underscores the importance of a broad
historical perspective. Calls for ‘better science’ to guide hatchery programmes sound sensible (Koenings, 2000; Parent, 2003), yet the more we
consider the tangled history of science and politics, the less science seems
able to resolve basic problems with these programmes. Calls for adaptive
management—for framing policies around assumptions of a complex and
unpredictable nature informed by incomplete and evolving knowledge—
suggest growing acceptance of this messy state of affairs. Yet even this
approach leaves many issues unresolved (Lackey et al., 2006; Langston,
2003). An underlying assumption of adaptive management has been that,
if not now then eventually, science will lead (Lee, 1993). The problem is not
simply that science has never been a primary guide, but that our yearning for
objectivity ignores the import of history. Not only has politics intrinsically
shaped the agendas and practises of modern salmonid hatcheries, but the
consequences of these actions have also narrowed managers’ options.
For example, the use of spawning channels and acclimation ponds were
not simply new scientifically based technological approaches to hatchery
problems but also politically based decisions to rely on environments that
mimic natural conditions rather than on wild environments themselves.
Longing for objective science obscures the historically produced circumstances that continue to constrain both our policy options and the ecological
and social consequences of our choices.
This is not a call to abandon dispassionate science for subjective politics,
but a request that readers learn to recognize the intrinsic social and ecological implications of salmonid hatcheries. Science is a necessary tool for
developing effective practises, but it cannot resolve the social politics that
have framed the structure and intent of fish culture policies since the midnineteenth century. Thus, expectations that more science will necessarily
lead to better policies tends only to mask the social implications of various
policy choices, a tactic that many interests have used in the Pacific salmon
crisis (Taylor, 1999a). Such approaches only perpetuate conflict because, as
we have seen in fishery after fishery in North America and Europe, any
policy not reached through the messy, compromise-laden process of consensus building quickly migrates to the courts. Many of the problems
attending salmonid hatcheries can only be resolved through political negotiation, and a prerequisite to a stable outcome will be an understanding of
the historical development of those hatchery programmes. We must first
84
Kerry A. Naish et al.
step back and examine the historical lessons of these technologies before we
can move intelligibly into the future.
4. Geographical Extent of Activities
Here, we report on the numbers of salmon released from conservation
and fishery enhancement hatcheries in the northern Atlantic and Pacific
oceans. This section is intended to provide an understanding of the relative
numbers of fish released in different regions, and to identify the main
purposes for their releases.
4.1. Enhancement of indigenous salmonids: Conservation,
production and mitigation hatcheries
4.1.1. Atlantic salmon (S. salar)
Atlantic salmon have been artificially propagated and released into the wild
on a large scale for more than a century. While the goals of Atlantic salmon
hatcheries are as varied as Pacific salmon hatcheries (e.g., supplementation
of at-risk populations, compensation for human-mediated loss of production, reestablishment of extinct populations, and increased catch), Atlantic
salmon hatcheries differ in several ways from those in the Pacific. Most
obviously, hatchery operations in the Atlantic are based on a few species
compared to the seven species propagated in Pacific hatcheries. The scale of
hatchery operations also differs dramatically. For instance, the number of
Atlantic salmon smolts released is about two orders of magnitude less than
releases of coho and Chinook smolts in the Pacific (Isaksson, 1988). Atlantic
salmon hatcheries frequently employ a ‘delayed release’ technique in which
out-migrating juveniles are released directly into marine waters rather than
rivers (Salminen et al., 1995). While this practise appears to improve survival,
it may increase straying (Gunnerd et al., 1988).
4.1.1.1. Western Atlantic The United States releases Atlantic salmon in an
effort to recover populations that have been extirpated or severely depleted
for decades. Over the last 10 years, annual releases from hatcheries in the
United States have averaged greater than 10 million (M), with a maximum
15.3 M in 2000 (Fig. 2.1). The vast majority of fish releases are fry (Fig. 2.2).
An analysis has shown that of the nearly 193 M salmon released in New
England since 1969 (Fig. 2.2), 79% were fry, 12.4% were smolts and 8.5%
were parr (US Atlantic Salmon Assessment Committee, 2003). Additionally, adult salmon that were spent, or were excess to hatchery broodstock
needs, have been released into US rivers, although these numbers are low
85
Evaluation of the Effect of Hatcheries on Wild Salmon
Millions of atlantic salmon
released
18
16
14
12
10
8
6
4
2
0
1960
1965
1970
1975 1980 1985
Release year
1990
1995
2000
sm
ol
t
+
sm
ol
2
t
+
sm
ol
3
t
+
s
2– mo
8 lt
w
8– ee
14 k
14 we
–2 ek
0
20 we
–2 ek
6
26 we
–5 ek
2
w
U eek
nf
ed
1 fry
+
pa
rr
2
+
pa
r
3
+ r
ad
ul
t
20
18
16
14
12
10
8
6
4
2
0
1
0
+
Percentage of releases
Figure 2.1 Number, in millions, of Atlantic salmon (S. salar) released from hatcheries
into New England waters from 1960 to 2000 (all life stages). Data are from the US
Atlantic Salmon Assessment Committee (2003).
Age class
Figure 2.2 Percentage of Atlantic salmon (S. salar) released into New England waters
by age class. Data are from the US Atlantic Salmon Assessment Committee (2003).
(e.g., 3576 in 2002). Twenty river systems have routinely received hatchery
Atlantic salmon, but three of these rivers (Merrimack, Connecticut and
Penobscot) account for more than 80% of total releases.
Despite the many millions of hatchery fry released over the last several
decades, 88% of Atlantic salmon returning to US waters originated as
hatchery smolts. The remaining 12% of returning fish originated from either
natural spawning fish or hatchery fry. In general, however, return rates
of hatchery fish have been very low. In 2000, for instance, the return rate of
hatchery smolts released in the Penobscot River was 0.10% (US Atlantic
Salmon Assessment Committee, 2003).
86
Kerry A. Naish et al.
Canadian Atlantic salmon hatcheries have been built primarily to compensate for the loss or degradation of freshwater habitat as a consequence of
hydropower development or other human activities. As with US releases,
returns from Canadian smolt releases have been low, ostensibly as a result of
harvesting in the Greenlandic fishery (Isaksson, 1988). From 1976 to 2002
nearly 80 M Atlantic salmon were released from hatcheries in the Canadian
Maritimes. Of these, most fish released were unfed or young fry, while
nearly 17% were 1þ smolts (Fig. 2.3).
6000
5000
4000
3000
2000
1000
0
19
76
19
78
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
00
20
02
Thousands of fish released
4.1.1.2. Eastern Atlantic While Scotland, Ireland and Norway are the
major worldwide producers of Atlantic salmon in commercial programmes,
production of salmon reared in conservation or fishery enhancement
hatcheries is small.
Legislation in Scotland governing the use of freshwater habitats limits the
development of hatcheries in these environments (Thorpe, 1980). Nevertheless, about 14 M fish are released annually throughout the region, most
of which comprise eyed ova, unfed and fed fry (B. Davidson, Association of
Salmon Fishery Boards/Institute of Fisheries Management, personal communication). Approximately 15% are released as parr. The majority of fish
released in these waters provide harvest opportunities, although some
are released to compensate for habitat lost to hydroelectric power schemes.
The region has seen better recovery following habitat restoration, and is
turning to this approach as the primary conservation measure.
Production of juvenile salmon from Irish hatcheries is also modest
relative to other countries. Unfed fry comprise the vast bulk of releases from
hatcheries. Approximately 2.9 M fish were released in 2002. Releases of
parr also take place, with 598,000 released in 2002 compared to 349,000
in 1999. About half a million smolts have been released each year since 1995
Release year
Figure 2.3 Number, in thousands, of Atlantic salmon (S. salar) (all life stages) released
from hatcheries into waters of the Canadian Maritime provinces (Novo Scotia, New
Brunswick, and Prince Edward Island) from 1976 to 2002.
87
Evaluation of the Effect of Hatcheries on Wild Salmon
(WGNAS, 2003). The major goal of these hatcheries is the preservation of
fisheries which are affected by hydroelectric development (Isaksson, 1988).
In England and Wales, releases of Atlantic salmon are modest (Fig. 2.4).
Currently (2000–2004), 1.4 M salmon are released annually, of which 42%
are parr and smolts (Fig. 2.5). This release size contrasts with historical
releases of 4.4 M in 1965–1969 and 3.0 M in 1987–1991 (N. Milner,
Environment Agency, UK, personal communication). Hatchery releases
in this region are intended to recover salmon stocks that declined as a result
of poor estuarine water quality and loss of spawning and rearing habitat
(Milner et al., 2004).
In Norway, several hatcheries release Atlantic salmon in an effort to
compensate for loss of spawning and juvenile rearing areas due to hydropower development. As with other East Atlantic countries, there was a rapid
expansion of hatchery production in the 1980s, with recent annual releases
of 8–9 M fry ( Jonsson et al., 1993). Even so, production of adult salmon
from hatchery releases in Norway is small relative to other countries in the
region.
Iceland is in a somewhat unique position relative to other countries in
the Atlantic because salmon harvest is limited to terminal fisheries in streams
(Isaksson, 1988). Iceland’s hatchery programme began with experimental
smolt releases in 1964 (Isaksson et al., 1997; Fig. 2.6). After achieving return
rates of 4–9% over the following 15 years (Isaksson, 1988), commercial
releases began with the goal of supplying a privately owned terminal fishery.
After a period of low return rates in the 1990s, commercial operations ended;
Thousands of fish released
4500
Smolts
Parr
Fry
Ova
4000
3500
3000
2500
2000
1500
1000
500
0
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Release year
Figure 2.4 Number, in thousands, of Atlantic salmon S. salar ova, fry, parr and smolts
released from hatcheries into waters of England and Wales from 1991 to 2003. Data are
from Milner et al. (2004).
88
3000
1965–1969
2500
1987–1991
2001–2004
2000
1500
1000
500
1+
/2
+
0+
/1
+
sm
ol
ts
pa
rr
fry
Fe
d
U
nf
ed
fry
0
Eg
gs
Thousands of Atlantic salmon released
Kerry A. Naish et al.
Age class
Figure 2.5 Number, in thousands, of Atlantic salmon (S. salar) released into waters of
England and Wales by age class. Data are from Milner et al. (2004).
Millions of fish released
6
5
4
3
2
1
0
1964
1969
1974
1979
1984
1989
1994
Release year
Figure 2.6 Number, in millions, of Atlantic salmon (S. salar) smolts released into Icelandic waters. Data redrawn from Isaksson et al. (1997).
however, there is still ongoing enhancement to increase abundance for
angling (Isaksson and Oskarsson, 2002).
4.1.1.3. Baltic Sea Since the early 1950s, several extensive hatchery programmes have been in place throughout the Baltic Sea region with the aim
of preserving and maintaining fisheries for Atlantic salmon stocks that have
been drastically reduced by hydropower development and other forms of
habitat degradation (Eriksson and Eriksson, 1993; Salminen and Erkamo,
1998). Specifically, countries surrounding the Baltic Sea have developed a
89
Evaluation of the Effect of Hatcheries on Wild Salmon
‘Salmon Action Plan’ (ICES, 2005), which aims to restore healthy runs of
Atlantic salmon and sea trout. A number of rivers in Finland, Sweden,
Estonia, Latvia, Lithuania and Russia have been identified for population
restoration efforts (including hatchery intervention), with the eventual goal
of creating self-sustaining populations of both species.
By the mid-1980s, natural production of Atlantic salmon had been largely
replaced by artificial propagation throughout the Baltic [Eskelinen and Eriksson
(1987) cited in Isaksson (1988)]. For instance, in 2001 total smolt production
for the Baltic region was 7.9 M Atlantic salmon, of which 6.6 M (83.5%) were
of hatchery origin (ICES, 2002). However, the proportion of hatchery-reared
fish varies substantially among regions, with the proportion of hatcheryreared smolts as high as 98.4% in the Gulf of Finland (ICES, 2002).
Sweden began releasing hatchery fish during the 1950s and over the next
decade the number of released smolts increased to about 1.5 M (Fig. 2.7).
By the middle of the 1980s Swedish hatchery production increased to about
2 M (mainly 2-year-old) smolts. In the early 1970s, Finland began a smolt
release programme that grew rapidly in the 1980s. In recent years, Finland
has released nearly 2.5 M smolts into the Baltic (Fig. 2.7; ICES, 2005).
A number of other Baltic countries have hatchery programmes, and their
contribution to regional hatchery production increased in the late 1980s.
In 2001, in addition to Sweden and Finland, Poland, Latvia, Estonia and
Russia released salmon into the Baltic (Fig. 2.8). Denmark and members of
Finland
Other countries
7
6
5
4
3
2
1
2000
1995
1990
1985
1980
1975
1970
1965
1960
1955
0
1950
Millions of Atlantic salmon released
Sweden
Release year
Figure 2.7 Number, in millions, of hatchery-raised Atlantic salmon (S. salar) smolts
released into the Baltic Sea. Data extracted from Eriksson and Eriksson (1993) and ICES
(2002).
90
3000
2500
2000
1500
1000
Fi
nl
an
d
Sw
ed
en
R
us
s
ia
n
fe
de
ra
tio
n
to
ni
a
Es
La
tv
ia
500
0
Po
la
nd
Thousands of Atlantic
salmon released in 2001
Kerry A. Naish et al.
Countries
Figure 2.8 Number of hatchery-raised Atlantic salmon (S. salar) smolts released by
country into the Baltic Sea. Data from ICES (2002).
the European Union also released small numbers of salmon in the 1990s
(ICES, 2002). Mixed-stock analysis using genetic approaches have shown
that hatchery fish comprise 20–75% of the total Finnish catch, depending on
the region harvested (ICES, 2005).
4.1.2. Sea trout (S. trutta)
Sea trout the anadromous form of S. trutta are subject to hatchery production, but this is small in most countries. There are two exceptions.
The Baltic Sea countries release sea trout to compensate for spawning
habitat lost through anthropogenic activities such as pollution, harvest,
damming and dredging. The majority of hatchery fish support fisheries,
although a few are used to rehabilitate threatened or extirpated populations
(ICES, 2005). All early life history stages (to 2-year-old smolts) are released.
Hatchery production has been fairly constant since 1988 (Fig. 2.9); Sweden,
Poland and Finland are the greatest contributors of hatchery fish to the
Baltic Sea. Finland and Estonia release about half of their smolt production
directly to the coastline, while the remaining fish are released in dammed
rivers. The majority of countries collect broodstock from naturally returning fish, but Poland’s production relies almost entirely on hatchery fish; the
wild populations are believed to be very small in this country (ICES, 2005).
In Ireland sea trout fisheries were, before the late 1980s, very important
sport fisheries, but they collapsed dramatically due to the impacts of sea lice
infection associated with marine salmon farming (Whelan and Poole, 1996).
Similar decline occurred in Scottish west coast sea trout fisheries, also linked
to salmon farming.
91
4000
3500
3000
Sweden
Poland
Latvia
Finland
Estonia
Denmark
2500
2000
1500
1000
500
0
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
Thousands of sea trout smolt releases
Evaluation of the Effect of Hatcheries on Wild Salmon
Release year
Figure 2.9 Number of hatchery-raised anadromous sea trout (S. trutta) smolts released
into the Baltic Sea by country from 1988 to 2004. Russia and Lithuania released small
numbers in the early years of this time series but are not included here. Data from ICES
(2005).
4.1.3. Pacific salmon (Oncorhynchus spp.)
Salmon hatcheries in the Pacific date from the 1870s, when the US Fish
Commision established a hatchery on the McCloud River in California
(Lichatowich, 1999). Early hatcheries were directed towards the enhancement
of depleted stocks or mitigation for habitat loss, but these hatcheries achieved
limited success because nearly all operations relied on releases of young fry that
had low survival rates (Mahnken et al., 1998). Important innovations in feeding
technology, disease control and rearing early life history stages occurred in the
1950s and 1960s leading to releases of larger fish with higher survival rates
(Lichatowich and McIntyre, 1987), and since this time, hatcheries have played
a major role in the management of Pacific Rim salmonids.
4.1.3.1. British Columbia Hatcheries have existed in British Columbia
since the first sockeye (O. nerka) hatchery began production in the 1890s.
These early hatcheries were substantial, with total output in 1910 around
500 M fish (Peterson et al., 2002), more than the current production in
British Columbia (Fig. 2.10). Hatchery production ceased after a couple of
decades because it was difficult to demonstrate any increase in production as
a result of artificial propagation (Section 2; Wood et al., 2002), but began
again in the 1960s (Section 2). In 1977, the Salmon Enhancement
Programme (SEP) of Canada was started with the aim of doubling the
catch of Pacific salmonids by protecting, rehabilitating and enhancing fish
stocks throughout British Columbia. Artificial propagation has played a
major role in the hatcheries formed under the SEP; spawning and rearing
92
Kerry A. Naish et al.
800
Millions of fish released
700
600
Sockeye
Chum
Pink
Chinook
Coho
500
400
300
200
100
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Release year
Figure 2.10 Number of hatchery fish by species (all life stages) released from British
Columbia hatcheries and spawning channels. Included are sockeye (O. nerka), chum
(O. keta), pink (O. gorbuscha), Chinook (O. tshawytscha) and coho (O. kisutch) salmon.
Data from the North Pacific Anadromous Fish Commission (http://www.npafc.org/).
channels and in-stream incubation boxes are intended to speed the recovery
of depleted stocks. British Columbia currently has 38 federally operated
hatcheries and an additional 150 public involvement projects (e.g., classroom
hatcheries or classroom incubators; Wood et al., 2002).
Production from British Columbia hatcheries and spawning channels
peaked in 1992 when nearly 700 M fish were released (Fig. 2.10). Since
then production has declined, with recent releases totalling around 330 M
fish. Throughout the time series examined here, sockeye and chum
(O. keta) have dominated hatchery and spawning channel production,
with sockeye comprising 41% (mostly spawning channel production) and
chum 34% of production in recent years. Additionally, an average of 20 M
coho, 53 M Chinook (O. tshawytscha) and 41 M pink salmon (O. gorbuscha)
have been released in recent years [Fig. 2.10; North Pacific Anadromous
Fish Commission (http://www.npafc.org/)].
4.1.3.2. Japan Japan operates the most extensive hatchery operation in
the world, with the goal of supporting its salmon fisheries. Japanese hatchery
programmes date from 1888, when the Chitose River Central Hatchery
was built in Hokkaido ( Johnson et al., 1997). Within 12 years, 45 chum
salmon hatcheries were constructed. However, as in other regions of the
Pacific Rim, fish culture practises were not well developed and the majority
of releases comprised unfed fry. Beginning in the 1960s, improvements in
feeding led to an increase in the size of juveniles, and as the percentage of fed
93
Evaluation of the Effect of Hatcheries on Wild Salmon
fry increased, return rates to hatcheries also increased substantially, reaching
2.3% after 1966 (Isaksson, 1988; Mahnken et al., 1998).
Currently more than 300 chum salmon hatcheries are located on the
islands of Honshu and Hokkaido, and at least 262 rivers are managed almost
entirely for artificial propagation. Over 2 billion salmon are released annually from these hatcheries. Chum salmon have made up about 93% of recent
releases (Fig. 2.11). Additionally, an average of about 132 M pink and 15 M
masu salmon (O. masou) have been released annually into Japanese waters
over the last decade (North Pacific Anadromous Fish Commission,
http://www.npafc.org/).
4.1.3.3. Russia The first salmon hatcheries in Russia began production in
the 1920s in tributaries to the Amur and Kamchatka rivers. At this time, the
Japanese also built hatcheries on Sakhalin Island and the Kuril Islands, which
came under Russian control after World War II (Johnson et al., 1997).
By the 1960s, 25 hatcheries operated in Russia and recently 44 hatcheries
have produced 500–550 M salmon annually in an effort to enhance fisheries
(Environment and Natural Resources Institute, 2001). Production is almost
entirely based on chum and pink salmon and is approximately evenly
distributed between the two species (Fig. 2.12). In addition, close to 3 M
coho, 5 M sockeye, 400,000 Chinook and 200,000 masu have been released
annually (Environment and Natural Resources Institute, 2001).
Millions of fish released
2500
2000
Masu
Chum
Pink
1500
1000
500
0
1950
1955
1960
1965
1970 1975 1980
Release year
1985
1990
1995
2000
Figure 2.11 Number, in millions, of hatchery fish by species (all life stages) released
from Japanese hatcheries from 1950 to 2000. Included are masu (O. masou), chum
(O. keta) and pink (O. gorbuscha) salmon. Data from the North Pacific Anadromous Fish
Commission (http://www.npafc.org/).
94
1600
1400
1200
1000
Chum
Pink
800
600
400
200
No data
0
19
73
19
75
19
77
19
79
19
81
19
83
19
85
19
87
19
89
19
91
19
93
19
95
19
97
19
99
20
01
Millions of fish released
Kerry A. Naish et al.
Release year
Figure 2.12 Number of hatchery chum (O. keta) and pink (O. gorbuscha) salmon (all life
stages) released from Russian hatcheries from 1973 to 2001. Sockeye, masu, chinook and
coho are not included; see text for numbers released during this period.
Russian hatcheries differ from Japanese programmes in that they were
not constructed to manage rivers exclusively for hatchery fish. Consequently, Russian hatcheries generally have used local fish for broodstock
and there is no attempt to prevent natural spawning. Even so, natural
production was not afforded a high priority as, historically, eggs were widely
exchanged among hatcheries and excess hatchery fish were allowed to
spawn with wild fish. In recent years, however, hatchery managers have
recognized the problems associated with egg transfers and such exchanges
have been reduced ( Johnson et al., 1997).
4.1.3.4. Alaska Construction of hatcheries in Alaska began in the early
1900s, but they were often badly sited and had poor water quality. As a
consequence, these hatcheries achieved little success and by 1936 Alaska’s
hatcheries ceased production (Roppel, 1982). However, after a protracted
decline in salmon catches in the early 1970s, the Alaska Department of Fish
and Game developed a coordinated SEP and the state of Alaska passed
legislation that encouraged ‘PNP’ hatcheries. Over the next several years,
there was an explosion of hatchery construction (Fig. 2.13; Farrington,
2003) and corresponding hatchery releases (Fig. 2.14). A unique feature
of Alaska’s hatchery system is that most hatcheries are operated by private
associations of fishers, environmentalists and local civic interests (Heard
et al., 2003). These associations can not only build and operate hatcheries,
but they also assist the Alaska Department of Fish and Game in the development of regional salmon plans, authorize taxes on salmon catches to
support hatcheries and sell returning hatchery fish to offset operational
expenses (Heard et al., 2003). Currently, there are eight regional aquaculture
associations in Alaska.
95
Evaluation of the Effect of Hatcheries on Wild Salmon
60
Number of hatcheries
50
Federal
Private non-profit
State
40
30
20
10
19
3
19 5
4
19 0
4
19 5
5
19 0
5
19 5
6
19 0
6
19 5
7
19 0
7
19 5
8
19 0
8
19 5
9
19 0
9
20 5
00
0
Year
Figure 2.13 Number of federal, state and private Pacific salmon (Oncorhynchus spp.)
hatcheries operating in Alaska (Farrington, 2003).
1800
Millions of fish released
1600
1400
1200
1000
Sockeye
Chum
Pink
Chinook
Coho
800
600
400
200
0
1970
1975
1980
1985
Release year
1990
1995
2000
Figure 2.14 Number, in millions, of hatchery salmon by species (all life stages)
released from Alaska hatcheries. Data from the North Pacific Anadromous Fish Commission, 2003 (http://www.npafc.org/). Included are sockeye (O. nerka), chum (O. keta),
pink (O. gorbuscha), Chinook (O. tshawytscha) and coho (O. kisutch) salmon.
Pink and chum salmon have made up the bulk of the salmon produced
in Alaska hatcheries (Fig. 2.14). In recent years, more than 1.4 billion
salmon have been released annually; 61% are pink and 32% are chum,
respectively. In addition, recent annual sockeye releases have averaged
more than 70 M (5%), coho releases have averaged about 18 M (1.3%)
96
Kerry A. Naish et al.
and Chinook releases have averaged nearly 9 M (0.6%). The PWS and
Southeast Alaska regions are the largest producers of hatchery salmon.
The Prince William Sound Aquaculture Corporation releases more than
400 M pink salmon each year and operates the largest hatchery operation in
North America (Environment and Natural Resources Institute, 2001).
Hatchery-produced fish appear to contribute significantly to harvest
levels in Alaska. In 2000, hatchery fish comprised 42% of Alaska’s pink,
64% of chum, 19% of Chinook, 24% of coho and 4% of sockeye catches
(Heard et al., 2003). However, the proportion of hatchery fish in the catch
varied greatly among regions. For instance, 82% of the pink and 88% of the
chum harvest in PWS was of hatchery origin. In contrast, hatchery fish
comprised only 10% of the total salmon harvest (2% pink, 0% chum) in
Cook Inlet (Heard et al., 2003).
4.1.3.5. United States, Contiguous Pacific States Salmon hatcheries in
the US Pacific Northwest have played an increasingly prominent role in
salmon management. Most public hatcheries were originally constructed to
rebuild depleted stocks and to mitigate for loss of natural spawning habitat,
and their goal was simply focused on enhancing the harvest of adults in the
commercial fisheries (Flagg et al., 2000). The number of hatcheries increased
gradually throughout the first half of the twentieth century; facilities were
constructed at a rate of about 1.5 per year from 1900 until 1950. However,
the pace of construction increased rapidly in the latter part of the century, at a
rate of nearly 6 per year from 1951 to 2000.
Total hatchery production peaked in the early 1980s with the release of
nearly 600 M salmon (Fig. 2.15). More recently, total annual hatchery
releases have averaged about 400 M. Chinook salmon dominate the releases
in the Pacific Northwest with average annual releases of 256 M fish from
1990 to 2000. The centre of Chinook production is the Columbia River
Basin, which accounts for about 27% of the world’s total Chinook release
(Mahnken et al., 1998). Coho and chum are also produced in large numbers,
with annual average releases from 1990 to 2000 of 77 and 66 M fish, respectively. Additionally, hatcheries in the region annually release steelhead
(O. mykiss; 28 M per year), sockeye (11.6 M per year) and pink salmon
(1.8 M per year). Interestingly, hatchery releases do not correspond directly
to the number of hatcheries constructed (Fig. 2.16). For Chinook, for
example, there appears to be a step function, with average annual production
increasing abruptly in the 1950s and subsequently varying around a greater
mean than in previous years (Fig. 2.16).
While salmon hatcheries operate in California, Idaho, Oregon and
Washington, the majority of hatchery fish are produced in Washington. In
1998, more than 70% of Pacific salmon released in the mainland United States
originated from Washington hatcheries, with 16% from California, 10.4%
97
Evaluation of the Effect of Hatcheries on Wild Salmon
Millions of fish released
700
600
500
400
Sockeye
Steelhead
Chum
Pink
Chinook
Coho
300
200
100
0
1900
1910
1920
1930
1940 1950 1960
Release year
1970
1980
1990
2000
450
1.6
400
1.4
350
1.2
300
1.0
250
0.8
200
150
100
50
0.6
0.4
0.2
Millions of Chinook released per
hatchery
Total number of Chinook hatcheries
Figure 2.15 Number, in millions, of hatchery Pacific salmon (Oncorhynchus spp.) (all
life stages) released from hatcheries in the continental United States. Data from the
North Pacific Anadromous Fish Commission (http://www.npafc.org/) and Mahnken
et al. (1998). Included are sockeye (O. nerka), chum (O. keta), pink (O. gorbuscha), Chinook
(O. tshawytscha) and coho (O. kisutch) salmon.
0
0
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
Figure 2.16 Cumulative number of Chinook salmon hatcheries (solid line) and the
average number of hatchery Chinook salmon O. tshawytscha (dashed line) released into
continental US water per hatchery from 1900 to 2000. Data extracted from Myers et al.
(1998) and the North Pacific Anadromous Fish Commission (http://www.npafc.org/).
from Oregon and 2.7% from Idaho (North Pacific Anadromous Fisheries
Commission Statistical Yearbook, available at http://www.npafc.org/).
As part of the management process in the Pacific Northwest, hatcheries
are required to state the purpose of their operations (Drake et al., 2003). We
grouped these operational goals into three of the categories outlined earlier
98
Kerry A. Naish et al.
(Section 1.2): mitigation, supplementation or production hatcheries.
The purpose of hatcheries varied greatly among species (Fig. 2.17). The stated
purpose of most coho and steelhead hatcheries, for example, was production,
while conservation was the primary purpose of most chum hatcheries.
Drake et al. (2003) also reviewed available published literature and
unpublished studies and subjectively classified hatchery stocks based on
the amount of genetic divergence between the hatchery and wild stocks
and the source of the hatchery stock relative to wild stock. The hatchery fish
were classified as local or non-local, and if the latter, as within or outside
evolutionary significant units (ESUs; Waples, 1995).
The source of most Chinook stocks was found to be local or non-local but
within the ESU (Fig. 2.18). In general, Drake et al. (2003) determined that
while there were moderate to few wild fish in most Chinook hatchery broodstock, there was no more than moderate divergence of the hatchery stocks
from the wild fish. Additionally, the ratio of hatchery Chinook to the natural
population was high for a number of stocks. This result meant that there were
substantial numbers of natural origin fish in the Chinook broodstock and there
was minimal divergence between hatchery and wild fish. A minority of
hatchery stocks comprised broodstock whose source was outside the ESU,
and in these stocks, there was extreme divergence between hatchery and wild
fish. A similar pattern emerged for steelhead; many hatchery stocks exhibited
little divergence from natural populations, but a significant number of stocks
showed substantial or extreme divergence (Fig. 2.18). This latter result may be
partly explained by the use of a deliberately domesticated hatchery stock with a
different return timing from wild populations. Coho hatchery stocks generally
had no more than moderate divergence from wild stocks, and chum hatchery
stocks showed little divergence from fish of natural origin.
Percentage of hatcheries
100
Mitigation
Supplementation
80
Harvest augmentation
60
40
20
0
Chum
Chinook Steelhead
Species
Coho
Figure 2.17 Declared purpose of hatcheries for each of four species of Pacific salmon
in the Pacific Northwest. Data are from Drake et al. (2003).
99
Evaluation of the Effect of Hatcheries on Wild Salmon
Coho
Chinook
30
25
20
Number of
15
hatcheries
10
5
0
1
2
3
4
4
c
b Source of
a hatchery stock
Relationship to
natural population
30
25
20
Number of
15
hatcheries
10
5
0
1
2
3
4
4
c
b Source of
a hatchery stock
Relationship to
natural population
Chum
1
30
25
20
Number of
15
hatcheries
10
5
0
2
Relationship to
natural population
3
4
4
c
b Source of
a hatchery stock
Steelhead
30
25
20
Number of
15
hatcheries
10
5
0
1
2
3
4
4
c
b Source of
a hatchery stock
Relationship to
natural population
Figure 2.18 Number of steelhead (O. mykiss), Chinook (O. tshawytscha), coho
(O. kisutch) and chum (O. keta) salmon hatcheries organized by the amount of genetic
divergence between the hatchery and wild stocks (relationship to natural population)
on a scale of 1^4, and the source of the hatchery stock relative to wild stock (source of
hatchery stock) on a scale of a^c for stocks within an ESU, and 4 for those outside an
ESU. Data are from Drake et al. (2003).
4.2. Enhancement of non-indigenous salmon and trout:
Introductions
The literature on worldwide salmon and trout introductions is extensive,
and will be reviewed elsewhere. Several general conclusions relevant to our
discussion can be drawn from these findings. First, data on release sizes for
anadromous salmonids in conservation, production and mitigation hatcheries were not readily obtained (Section 4.1), and similar data on introductions were not available in most cases. In addition, it appears difficult to
discriminate between those introductions that were successful versus those
that were not. Second, there were inconsistencies in the scales over which
the data were reported, and these scales varied historically. Third, it appears
that the extent of introductions worldwide has declined, possibly because
most potential sites for salmon fisheries have been explored. However, there
is a paucity of data on ongoing ‘put and take’ hatcheries based on nonindigenous salmon, and it is difficult to gauge the level, extent and impact of
these activities. Finally, the ongoing tension between economic and ecological incentives is identified as being a primary determining force as to
100
Kerry A. Naish et al.
whether exotic introductions will continue. We do not explicitly address
the consequences of non-indigenous salmon introductions in most of the
risks that we review. However, the introduction of exotic salmon may
result in extreme outcomes of these risks; for example, competition with
local species, the introduction of a non-endemic pathogen or hybridization
with vulnerable indigenous species.
5. Potential Consequences of
Enhancement Activities
5.1. Genetic risks associated with salmon
hatchery programmes
The genetic effects of hatchery programmes have been the subject of
substantial review in the literature. Most authors agree that releases can
have detrimental effects on wild populations (e.g., Allendorf and Ryman,
1997; Aprahamian et al., 2003; Busack and Currens, 1995; Cross, 2000;
Hindar et al., 1991; Utter, 1998; Waples, 1991) and others have suggested
management steps that may be taken to reduce such effects (Hard et al.,
1992; Mobrand et al., 2005; Waples and Drake, 2005). The debate in this
area is not around whether hatchery programmes pose a threat to wild
populations, but whether the risks are sufficiently large to compromise
wild populations and if true, whether they may be reduced or avoided
through correct management actions (Brannon et al., 2004a; Campton,
1995; Waples, 1999).
The aim of this section of the chapter is not to simply repeat the
information provided in many of the papers written on this topic, but to
update and evaluate our understanding of the genetic consequences of
hatchery programmes. First, we will review the current knowledge of the
genetic risks involved. Second, most would argue that the varying objectives of hatchery programmes will pose different types and magnitudes of
effects on wild populations, and so we will examine the impacts of hatchery
programmes in the context of their release objectives (conservation or
fishery enhancement hatcheries, following the classification outlined in
Section 1). Finally, we will identify research directions that will assist
designing management steps that may be taken to reduce the genetic risks
associated with different salmon hatchery programmes.
5.1.1. Genetic risks associated with hatchery fish
The potential genetic outcomes of rearing and releasing hatchery fish into
the wild fall broadly into three categories: the effect of hatcheries on
hatchery fish, the direct effect of hatchery fish on wild fish and the indirect
effect of hatchery fish on wild fish (Campton, 1995; Hindar et al., 1991;
Evaluation of the Effect of Hatcheries on Wild Salmon
101
Waples, 1991). The latter occurs through processes such as competition,
disease transfer and increased fishing mortality on the wild component of a
stock, all of which result in demographic changes or selection in the wild
populations. The first two factors will be reviewed in detail below, primarily
because much is known about these factors, while the third has not been
empirically studied to any significant extent (Campton, 1995; Waples,
1991), and so will not be discussed further in this section.
5.1.1.1. The genetic effects of hatcheries on hatchery fish Procedures
implemented in the collection, mating, rearing and release of hatchery
salmon may lead to a change and perhaps a reduction in genetic diversity
of the source population in two ways. First, as the population size decreases,
the random loss of genetic variation in a population can be expected to
increase. A loss of genetic variation may lead to inbreeding and an associated
decrease in fitness, termed inbreeding depression. Second, the hatchery
environment can be a poor imitator of wild conditions, and hatchery fish
may become adapted to their environment through a process known as
domestication selection. The result of both effects may ultimately lead to the
release of a genetically altered population that may interact negatively with
any wild stocks present, by decreasing the overall fitness of the combined
populations.
5.1.1.1.1. Loss of genetic diversity and inbreeding The genetic diversity
relevant to the long-term survival of a species is quantitative in nature,
that is, several gene loci interact with each other and with the environment
to create a range of phenotypes. Quantitative genetic diversity is difficult to
measure at present (Hard, 1995), although substantial advances are being
made in characterizing this functional diversity (reviewed in Danzmann and
Gharbi, 2001; Vasemägi and Primmer, 2005). Thus, effective population
size is often used as a proxy metric for quantitative genetic diversity because
the theoretical link between this measure and loss of genetic variation is well
known (Frankham et al., 2002).
The effective population size, Ne, is the size of an ideal population
whose genetic composition is influenced by random processes in the same
way as a real population. It is important to realize that Ne can be very
different from the census size, Nc, of a population, and the ratio of Ne to Nc
can be affected by factors such as sex ratio, family size, fluctuations in
population size, overlapping generations and variance in reproductive success. The measurement of Ne in salmon populations is usually confounded
by overlapping generations, seen in most species; hence Nb, the number of
breeders corrected for generation length, is often the preferred measure
(Waples, 1990, 2004). The ratio of Nb (or Ne) to Nc in wild salmon
populations is often low (e.g., Shrimpton and Heath, 2003), and has been
102
Kerry A. Naish et al.
noted to vary between 0.01 and 0.71 (Bartley et al., 1992; Hedrick et al.,
1995; Waples et al., 1993).
A number of practises associated with the management of hatchery fish
may lead to a decrease in Ne (Busack and Currens, 1995). For example,
hatcheries may cause accelerated growth, which in turn may cause early
maturation in male salmon and thus skew male to female ratios (Larsen et al.,
2004). Spawning protocols that deviate from 1:1 sex ratios are known to
reduce Ne (e.g., Allendorf, 1993). Genetic variation may also be lost at the
founding of the hatchery strain, during a bottleneck or during prolonged
periods of reduced population size. Most studies have not examined the
underlying causes of a decline in genetic diversity, but such decreases of
diversity in hatchery strains have been reported at neutral genetic markers
(Jones et al., 1997; Koljonen et al., 1999; Nielsen et al., 1997; Primmer et al.,
1999; Tessier et al., 1997; Was and Wenne, 2002). Where a decrease in Ne
has been directly measured, it is often possible to attribute the decrease to
broodstock practises that may be avoidable (Koljonen et al., 2002).
On a positive note, it is widely recognized that if management steps
are taken to avoid a loss of variation, then the Ne to Nc ratio within a
captive population can be higher than that in the wild. For example, if
founder population sizes are adequate (Allendorf and Ryman, 1997;
Frankham et al., 2002), sex ratios are equalized at mating (Campton,
2004), family sizes are also equalized (Allendorf, 1993) and long-term fluctuations in population sizes are avoided, then Ne to Nc ratios can exceed 1.0.
Thus, captive populations may be used to enhance the genetic diversity in a
depleted wild population (Hedrick et al., 2000).
However, a loss of genetic variation may be unavoidable in some cases.
The inevitable consequence is that a self-perpetuated broodstock will eventually comprise individuals with higher average relatedness, and mating
between these individuals will result in inbreeding (identity by descent).
Inbreeding by itself does not result in a change in gene frequencies; it does,
however, result in an increase in homozygotes. Inbreeding is often associated with a decline in fitness-related phenotypes (Keller and Waller,
2002), termed inbreeding depression, which in turn may lead to a reduction
in population size and a change in gene frequencies through genetic drift
(Bijlsma et al., 2000; Saccheri et al., 1998). Inbreeding depression is usually
more prevalent in life history traits than in morphological traits (DeRose
and Roff, 1999) and tends to be more severe in wild than in captive
populations (Crnokrak and Roff, 1999; Kalinowski and Hedrick, 1999).
The manifestation of inbreeding depression is usually attributed to one
of two mechanisms, either the loss of dominance (‘masking’ of deleterious
alleles) or of over-dominance (heterozygote advantage) at genetic loci
encoding fitness traits (Frankham et al., 2002). Deleterious alleles may increasingly occur in the homozygous state following inbreeding, thus reducing
the dominance interactions between advantageous and deleterious alleles
Evaluation of the Effect of Hatcheries on Wild Salmon
103
in heterozygotes. It has been argued that a fitness decline is controlled by the
rate of inbreeding because deleterious alleles occur more frequently as
homozygotes in small populations, and are no longer masked by positive
or neutral dominant alleles. Selection can then act by ‘purging’ these alleles
from the population. However, many researchers have demonstrated that
selection against deleterious alleles cannot be relied on to decrease the rate
of extinction because these alleles can also become fixed in a population
through genetic drift (Reed et al., 2003). An alternative explanation for a
decline in fitness is attributed to the advantage conferred by overdominance at heterozygous loci; that is, the sum of two alleles at a locus
may outweigh either of the two homozygotes. If heterozygotes are lost,
then overall fitness will decline in a population. If over-dominance is the
underlying mechanism for inbreeding depression, then purging cannot be
implemented as a management tool and in fact, the over-dominance
hypothesis may partly explain why this approach is unpredictable in
many cases.
Wang et al. (2002) provided several examples of inbreeding depression in
their review of inbreeding in salmonids. Perhaps their most significant
finding was that experimental designs have varied, and general inferences
about the incidence and manifestation of inbreeding depression in the
salmonids cannot easily be drawn. There are three major approaches
to testing inbreeding depression in a population (Keller and Waller,
2002): (1) the experimental comparison of inbred with outbred lines,
(2) the outcrossing of small inbred populations to examine whether an
increase in heterozygosity results in an increase in fitness, and (3) the
comparison of the phenotypic values of related versus unrelated individuals
within a population. Inbreeding studies on salmonids have favoured the first
method and have differed in the rates of inbreeding reported (Wang et al.,
2002). In a typical experiment using sexually reproducing organisms, inbred
individuals are produced only after at least two generations of mating. Such
experiments can be especially protracted in salmon, most of which reach
maturity after several years. Hence, most studies have generated individuals
with high inbreeding values as early as possible. Wang et al. (2002) proposed
that the contradictory results seen by different researchers at the same traits
and in the same species can be explained by the rates of inbreeding, most of
which are not characteristic of wild populations. For example, Gjerde et al.
(1983) reported inbreeding depression in adult body weight of rainbow
trout O. mykiss under fast inbreeding, whereas Pante et al. (2001) do not
observe this outcome under slower inbreeding. This contrast supports the
notion that purging may play a role in avoiding fitness declines in salmonids,
a view which needs to be verified by further research.
Other equally important factors may also affect the manifestation
of inbreeding depression in the reported studies. For example, the initial
inbreeding coefficient, F, of the baseline population may differ between
104
Kerry A. Naish et al.
experimental populations (Wang et al., 2002), as may the prevailing environmental or social conditions (Gallardo and Neira, 2005). Inbreeding
depression also varies by the trait measured (DeRose and Roff, 1999) and
this trend certainly appears to be supported within salmonids. For example,
early growth-related traits and survival appear to be more prone to inbreeding
than adult weight and size (Wang et al., 2002), and inbred Chinook salmon
have been found to be more susceptible to the pathogen Myxobolus cerebralis,
but not to Listonella anguillarum or infectious hematopoietic necrosis virus
(IHNV; Arkush et al., 2002).
Perhaps the most conspicuous point here is that the majority of such
studies have been carried out on cultured fish. If the incidence of inbreeding
depression in salmon increases in the wild, as in other taxa (Crnokrak and
Roff, 1999; Kalinowski and Hedrick, 1999), then more emphasis on wild
populations is needed. The design of such experiments should be systematic,
include both resident and anadromous forms and populations with
inbreeding history and those without. Such broadened studies likely will
be more amenable to generalizations about the effects of inbreeding in
systems incorporating hatchery releases.
5.1.1.1.2. Domestication selection Domestication selection that arises in a
supplementation programme is often unintentional, resulting from ‘natural’
adaptation of the species to the hatchery. Simply, life history theory predicts
that selection imposed by a novel environment will rapidly alter the genetic
architecture of life history traits of a population, and lead to divergence
between the founding and the new population. The strength of this change
will be dependent on the selection regimes between the hatchery and the
wild environment, the numbers of generations that the broodstock are held
in captivity and the magnitude of genetic variation underlying the fitness
trait under selection. Thus, the opportunity for domestication selection to
produce divergence between wild and captively reared individuals is largest
when the latter are cultured throughout their life histories for many generations (Hard, 1995). However, simulations have shown that domestication
selection in hatchery fish can have rapid and substantial negative genetic
effects on targeted wild populations, even when wild captive breeders are
always used (Lynch and O’Hely, 2001; Ford, 2002).
Examples of differences between hatchery and wild fish are widely
published. However, it should be noted that in many of these cases it is
difficult to implicate domestication selection alone. For example, a series of
experiments have demonstrated behavioural and morphological differences
between wild populations and hatchery coho salmon (Fleming and Gross,
1992, 1993) and Atlantic salmon originating from aquaculture facilities
(Fleming and Einum, 1997; Fleming et al., 1994, 1996). While many of
these experiments serve to demonstrate the rapid phenotypic divergences
that may be obtained following deliberate domestication, the cultured
Evaluation of the Effect of Hatcheries on Wild Salmon
105
strains used in some of these studies were not derived from the same
watershed as the wild populations. In fact, Reisenbichler and Rubin
(1999) pointed out that the prevalence of this experimental design in most
studies, and the fact that many researchers examine only one or two fitnessrelated traits, has resulted in the charge that domestication selection has yet
to be demonstrated in hatchery salmonids. On the west coast of North
America, hatchery coho salmon have been compared to their wild source
counterparts. Juvenile cultured coho salmon are less aggressive at emergence
and adult hatchery fish are not as successful at mating as wild hatchery fish
(Berejikian et al., 1997, 1999, 2001a). The two groups also differ morphologically (Hard et al., 2000). However, in an informative series of experiments, dissimilarities between hatchery and wild steelhead (O. mykiss) have
been substantially reduced by rearing hatchery fish in enriched environments (Berejikian et al., 2000, 2001b). Thus, an explanation for many of the
variations observed between hatchery and wild fish is that the different
rearing environments have acted to change the phenotypes without substantially changing the underlying genotype, thus confounding a genetic
interpretation of the results of these studies. However, there are cases where
the evidence indicates that a hatchery population diverged from the wild
population from which it was derived (e.g., anti-predator and aggressive
behaviour of juvenile steelhead trout Berejikian, 1995; Berejikian et al.,
1996). In addition, the adults selected for spawning in a hatchery are often
the early arrivals, with the result that the distribution of spawning may
change, often quite dramatically (Flagg et al., 1995; Ford et al., 2006; Quinn
et al., 2002). Such differences in spawning date are likely to have large fitness
consequences, as this trait is closely linked to the prevailing regimes of
temperature, flow and productivity of the ecosystem.
The genetic outcomes of domestication selection and their potential
solutions have rarely been tested empirically in salmonids due, in part, to
the fact that such experiments require a breeding design and these species
are long lived. Even in those populations in which domestication selection
has been reported in controlled experiments, few studies have been
designed to detect this phenomenon directly. An experiment conducted
over several generations (Hershberger et al., 1990) implicated domestication
selection for increased weight in coho salmon cultured over four generations in marine net pens. However, the underlying genetic model in this
study has been criticized (Hard, 1995) because the experiment did not
maintain controls that may have discriminated environmental versus genetic
changes during culture.
In a seminal study, wild steelhead embryos released in small streams
generally had a higher survival to 1 year than those of either hatchery or
hybrid offspring (Reisenbichler and McIntyre, 1977). In this case, hatchery
fish were derived from the wild population and separated for only two
generations. Results differed between streams, suggesting a genotype by
106
Kerry A. Naish et al.
environment interaction that was magnified in some environments but not
others (Hard, 1995). Hence, the severity of domestication selection can be
expected to vary in wild environments. As a comparison, in a recent
experiment mimicking the experimental design of Reisenbichler and
McIntyre (1977), differences in growth and survival were not seen between
the offspring of wild brown trout, a sea-ranched strain and their hybrids
when they were reared in the wild (Dahl et al., 2006; Dannewitz et al.,
2003). The sea-ranched strain had been separated from its source population
by seven generations. These contrasting examples highlight the importance
of experimental design—the sea trout experiments were conducted in a
common environment, were replicated and took into account genetic
effects that may explain variation between individual families comprising
each cross type.
It is quite clear that the risks posed by domestication selection have not
been quantified in a systematic fashion. In articles examining genetic changes
in hatchery salmon populations, many authors have recognized domestication selection as a potentially significant problem (Busack and Currens,
1995; Waples, 1999), but have concluded that scant evidence exists to
evaluate its significance to management approaches. Little is known about
the relationship between selection on specific fitness traits and population
size, the number of generations in captivity that may lead to genetic
differences with the wild population and whether such selection is reversible or avoidable through different management approaches. Such knowledge is essential for conservation planning, and there is an urgent need for
research on the extent and consequences of domestication selection in
salmonids, as well as steps that may be taken to reduce its effects.
5.1.1.2. Hatchery regime effects on wild fish If the hatchery regime
results in a change in the genetic composition of the captive population,
then such changes can have negative consequences on the wild populations
with which they interact. These changes can be demographic in nature; the
release of a genetically under-represented hatchery population into the wild
can reduce the overall effective population size Ne of the two components
together or decrease the existing population structure. The changes can also
affect the fitness traits of a wild population through hybridization with less
fit hatchery fish. Taken together, the results of these processes can lead
ultimately to the decline and extinction of an endangered wild population.
5.1.1.2.1. Changes in effective population size A simple simulation based
on the Ryman–Laikre model (Ryman and Laikre, 1991) can be used to
illustrate the effects of hatchery release size on a population’s effective size,
Ne (Hedgecock and Coykendall, 2007). The model examines the effect of
hatchery recruitment to a wild population over a single generation, and relies
107
Evaluation of the Effect of Hatcheries on Wild Salmon
on known values of effective size for hatchery (Neh) and wild (New) fish.
Outcomes vary with the relative proportion of hatchery fish in the total
census population (Fig. 2.19). Intuitively, supplementation with a hatchery
population with a large Ne is beneficial when the wild population has a small
Ne (Fig. 2.19A), but detrimental with the opposite scenario (Fig. 2.19C).
The most interesting lesson to be learned from this model, however, is that
the effects of hatchery supplementation can be very detrimental to wild
populations of moderate effective size (Fig. 2.19B).
One needs to keep in mind that Ne is less than N in most cases. For
example, the Ryman–Laikre method assumes that all fish are spawning
adults, but many hatcheries release smolts with different return rates than
the wild fish. Similarly, it is assumed that hatchery and wild fish have an
A
B
Neh = 50
0.9
Initial effective size, New
C
x, Proportion of
hatchery fish
300
0.9
260
0.7
220
0.7
180
0.5
140
0.5
100
0.3
60
0.1
0.3
20
0.1
300
260
220
180
140
100
60
20
Neh = 10
Initial effective size, New
Neh = 100
0.1
0.3
x, Proportion of
hatchery fish
0.5
0.7
300
260
220
180
140
100
60
20
0.9
Initial effective size, New
Figure 2.19 Proportional change in the effective population size of a supplemented
population (Nes), following a single release of a hatchery population of varying effective
sizes (Neh). Shaded areas represent decreases in the effective size of the supplemented
population. Each contour line represents a proportional increase or decrease by a factor
of 2. Based on approaches developed by Ryman and Laikre (1991) and Hedgecock and
Coykendall (2007) where Nes ¼ (NehNew)/(x2New þ y2Neh), and New is the effective size
of the wild population prior to supplementation, x is the relative proportion of
hatchery-origin fish, y of wild fish, and x þ y ¼ 1.0.
108
Kerry A. Naish et al.
equal chance of reproducing successfully, but many studies have shown that
this is not the case. The approach also assumes that the numbers of progeny
from captive or wild fish are distributed binomially, and that there are no
overlapping generations. This model has been further developed, taking
many of these caveats into account, and the newer models will be discussed
in a later section. The central point to be made from this model, however, is
that attempts to increase the growth rate of a population may be detrimental
to the genetic diversity of a population if few individuals are used to recover
that population (Ryman and Laikre, 1991).
5.1.1.2.2. Hybridization and outbreeding depression The impacts of hatchery releases may be limited to a change in effective population size if the
hatchery stock is genetically identical to the wild population with which it
interacts. However, if genetic variation in hatchery stocks changes following inbreeding or domestication selection, or if hatchery fish are exogenous
to the wild system, then hybridization between the cultivated fish and the
wild fish may have unintended consequences. Hybridization between unrelated populations can lead to a reduction in fitness, known as outbreeding
depression. Outbreeding depression has been attributed to two mechanisms,
each of which can have different long-term consequences.
First, outbreeding depression can result from a loss of local adaptation
(known as ecological outbreeding depression). Populations can become
adapted to a specific environment following selection. Such ‘local adaptation’ is extensive in the salmonids (Allendorf and Waples, 1996; Taylor,
1991; Waples et al., 2001), and there is substantial concern that hybridization will result in its loss (Allendorf and Waples, 1996; Busack and Currens,
1995). If individuals from unrelated populations mate, hybrids have only
half the ‘adapted’ alleles in either parental environment and are not as fit as
the parental populations. This reduction in fitness is often seen in the first
(F1) hybrid generation. Second, outbreeding depression may follow a
disruption of interactions between co-adapted genetic loci underlying fitness traits (known as physiological outbreeding depression). These epistatic
interactions can arise either through selection, which can act concurrently
on genes that are inherited together, or through chance associations (Lynch
and Walsh, 1998; Templeton, 1986). Significantly, co-adapted gene complexes can differ between populations occupying similar environments, and
recombination between divergent genomes can disrupt such complexes.
Typically, hybrid vigour (heterosis) observed in the F1 hybrids is a poor
indicator of the performance of subsequent generations (Lynch and Walsh,
1998) because recombination between the parental chromosomes occurs
for the first time in this generation.
It has been hypothesized that the underlying mechanisms for outbreeding
depression will have different outcomes for hybrid populations. Simulations
have shown that a disturbance of local adaptation results in a greater initial
Evaluation of the Effect of Hatcheries on Wild Salmon
109
fitness decline than a disruption of co-adapted gene complexes (Edmands and
Timmerman, 2003). However, the simulation results differ over the long
term. Much of the genetic variation explaining local adaptation is additive and
provides a better opportunity for population recovery through selection than
does the epistatic variation that results from co-adapted gene complexes.
Epistatic variation arises through genetic drift and indirect selection (Lynch
and Walsh, 1998).
The severity of outbreeding depression is expected to change with a
number of parameters and there is a considerable body of literature investigating whether such parameters can be used to predict the outcomes of
hybridization. For example, the incidence of outbreeding depression is
expected to increase with greater genetic distance between hybridizing
taxa (Emlen, 1991; Lynch, 1991). This prediction is true across a wide
range of species, but not in others (Edmands, 2002). Declines in fitness may
depend on prevailing environmental conditions (Lynch, 1991). These
environments may fluctuate temporally and results may vary accordingly
(Gharrett et al., 1999; Gilk et al., 2004). Most studies investigating the
consequences of hybridization are on first generation hybrids only, but
outbreeding depression can be expected to vary across generations
(Edmands and Timmerman, 2003). Thus, fitness increases in first-generation
hybrids are not necessarily repeated in the second generation (Fenster and
Galloway, 2000), and population recovery can vary (Edmands and
Timmerman, 2003; Templeton, 1986). Finally, the expression of outbreeding
depression differs across fitness traits (Andersen et al., 2002).
A meta-analysis on several studies in fishes has shown that the outcomes
of hybridization are difficult to predict (McClelland and Naish, 2007).
Response varies across traits, taxon and generation but, significantly, genetic
distance based on neutral genetic markers cannot be used reliably as an
indicator of the incidence of outbreeding depression. Such an outcome can
be explained by the unpredictable nature of the different mechanisms
underlying outbreeding depression (Lynch and Walsh, 1998), but also by
the fact that a measure of genetic distance at fitness traits may be more
appropriate for this task (McClelland and Naish, 2007). In a recent review,
Utter (2001) proposed that the complexity of life history might be a better
predictor of outbreeding depression because introgression is more likely in
freshwater than in anadromous salmonids, and hybrids of the latter may be
less likely to survive.
In salmonid fishes, both increases and decreases in fitness have been
observed in the F1 generation. Decreases have been observed in pink
salmon homing ability (Bams, 1976), disease resistance in coho salmon
(Hemmingsen et al., 1986), salinity tolerance in kokanee hybrids (Foote
et al., 1992), growth rate in coho salmon (McClelland et al., 2005;
Tymchuck et al., 2006) and rainbow trout (Tymchuck and Devlin, 2005)
and development rate in coho (Granath et al., 2004). However, such
110
Kerry A. Naish et al.
decreases were not seen in the F1 of pink salmon (Gharrett and Smoker,
1991; Gharrett et al., 1999).
A few studies have followed outbred populations through to the F2 or
subsequent generations (McClelland and Naish, 2007), and individual case
studies have shown a continuation in reduced fitness in these generations
(McClelland et al., 2005; McGinnity et al., 2003; Tymchuck and Devlin,
2005; Tymchuck et al., 2006). In all of these examples, fitness loss was
attributed to genetic mechanisms underlying ecological outbreeding
depression. However, it is the systematic treatment of hybridization over
varying evolutionary distances that will provide researchers with the ability
to predict the genetic outcomes of mixing populations. In a series of
experiments on pink salmon populations, researchers in Alaska have performed crosses between populations of varying genetic distance with the
intent of detecting the point at which outbreeding depression will no longer
be demonstrable. At one extreme, reproductively isolated pink salmon
returning strictly in either the odd or the even years were crossed and
released. The survival of their F1 hybrids in the wild was comparable to
the controls (Gharrett and Smoker, 1991; Gharrett et al., 1999), but the F2
hybrid survival rate was severely depressed (Gharrett et al., 1999). In this
case, outbreeding depression was attributed to a breakdown in co-adapted
gene complexes, which is perhaps not surprising. The odd and even year
pink salmon return to similar habitats and any genetic differences that
accumulate between them must be due, in large part, to genetic drift.
In contrast, a second experiment has demonstrated that the second mechanism underlying outbreeding depression is loss of local adaptation. Pink
salmon from the same year class that were spatially separated by about
1000 km were hybridized and both the F1 and F2 generations exhibited
outbreeding depression (Gilk et al., 2004). In contrast to these findings,
coho salmon spawning populations separated over spatial distances of
130–340 km and displaying different development rates showed no change
in fitness over two generations following hybridization (Smoker et al.,
2004), although the authors point out that the power of the data may not
have been sufficient to detect outbreeding depression.
These experiments serve to illustrate the potential outcomes of hybridizing unrelated populations, but are most useful when they are systematic
in nature. The continuation of experiments such as those conducted on
pink salmon over different distances will provide a very interesting insight
into whether evolutionary distance can be used as a predictor of the
potential for outbreeding depression within a single anadromous salmonid
species, a point relevant to effective management.
5.1.1.2.3. Population structure Increasingly, attention is being paid on
the effects of hatchery releases on metapopulation structure of a wild
population (Utter, 2004; Ward, 2006). Hatchery activity may affect
Evaluation of the Effect of Hatcheries on Wild Salmon
111
population structure through two means: by transfers between different
locations and by continued release of hatchery fish. For example,
Vasemägi et al. (2005) demonstrated that ongoing releases of nonindigenous Atlantic salmon resulted in homogenization of population structure of wild fish over time. The impact of the number of fish transferred and
released on population structure has not been frequently reported, and yet
an understanding of this relationship is important for risk assessments. In
coho salmon, it has been shown that the number of fish transferred might
reduce population structuring, even between closely related populations
(Eldridge and Naish, 2007). More importantly, the numbers of fish released
from hatchery programmes that collect broodstock locally resulted in a
reduction in fine scale population structure in this species (Eldridge and
Naish, 2007), a finding that has clear implications for ongoing hatchery
programmes. Population structure reflects evolutionary processes, some of
which lead to local adaptation, and levels of migration between neighbouring
populations are related to the long-term genetic viability of a species as a
whole (Waples, 2002). The greater the spatial diversity of a species, the more
likely that species will exhibit resilience to extinction risk (McElhany et al.,
2000). Ongoing hatchery programmes may need to control the size of their
release and numbers of fish transferred between programmes in order to
reduce their impact on this resiliency.
5.1.2. Evidence for the genetic impacts of different types of
hatchery programmes
In an interesting evaluation of the genetic risks associated with hatchery
practises, Campton (1995) raised two key points that are often ignored
when evaluating whether hatchery fish can be effectively used as a management tool for conservation or harvest. First, most studies fail to discriminate
between the underlying biological or management-based causes of any
detrimental effects. In the years between Campton’s and this chapter, this
distinction has rarely been elucidated. Second, there is a paucity of data
providing a clear understanding of the biological causes of such effects.
In Section 5.1.2, we examine the evidence for the genetic outcomes of
hatchery practises in the context of hatchery type and programme goals, the
risks associated with such goals and the evidence, if any, for impacts that
may be attributed to biological effects rather than to management effects.
5.1.2.1. Captive broodstock The greatest genetic risks associated with the
maintenance of an entirely captive broodstock in culture over long periods
of time are the loss of within-population genetic diversity and domestication selection. Losses due to genetic drift may be avoided by maintaining
high Ne/Nc ratios and inbreeding can be reduced for as long as possible by
maintaining pedigrees and minimizing kinship during mating (RodriguezClark, 1999). In very small populations, selection theoretically becomes
112
Kerry A. Naish et al.
almost negligible (unless the selection differential is very high), and some
authors have suggested that captive endangered populations be managed as
fragmented subpopulations in different rearing facilities in order to avoid
domestication selection (Margan et al., 1998). This strategy is risky because
significant reductions can be incurred if subpopulations are lost and to date,
the approach has been supported empirically in experimental populations of
fruit flies only (Woodworth et al., 2002). Perhaps some of the best management steps that can be taken to avoid domestication selection in a captive
broodstock are the reduction of the number of generations held in captivity
(initiating reintroduction as soon as possible), and decreasing selection
differentials between hatchery and wild environments as far as possible.
The scientific community has started to accumulate evidence on the
consequences of genetic drift, inbreeding and domestication selection in
captive salmon populations, but there are few studies that examine the
effectiveness of management steps in mitigating these risks in fishes,
let alone salmonids. For example, studies comparing modern to archival
samples, such as that conducted on captive Atlantic salmon in the Baltic
region (Saisa et al., 2003), demonstrate that long-term programmes have
resulted in reduced genetic diversity and effective sizes. However, the
extent to which such losses could have been avoided through careful
management has not been determined, especially since our awareness of
the risks has post-dated the initiation of such programmes. Realistically, the
mating of relatives (and hence inbreeding) is inevitable in a closed system
despite the best measures (Myers et al., 2001). An inbreeding rate of around
1% is generally deemed acceptable in benign agricultural environments
(Franklin, 1980), but this tenet has yet to be tested in salmonids that will
eventually be released to the wild.
5.1.2.2. Supplementation through supportive breeding The goal of
many conservation-oriented hatcheries is to support declining populations
and, thus, most seek to enhance numbers without compromising the
genetic diversity of the wild populations. This goal may be difficult to attain
because a change associated with broodstock collection and release is
probably inevitable (Waples, 1999). Supplementation hatcheries face similar
challenges as those described for captive broodstock, but have an advantage
in some cases. Many genetic changes such as inbreeding and domestication
selection can be theoretically reduced by replenishment from the wild populations, and many recommendations focus on this practise (e.g., Mobrand
et al., 2005). However, these programmes may also have a major disadvantage: through their practises, they could alter the genetic composition of the
wild stocks with which they interact. This alteration may occur through a
change in effective population size, homogenization of locally adapted stocks
or through outbreeding depression, and can affect the ability of the vulnerable
populations to adapt to a changing environment.
Evaluation of the Effect of Hatcheries on Wild Salmon
113
Systematic treatment of the genetic effects of supportive breeding has
only occurred through theoretical modelling. Earlier, we described how a
single generation of supplementation could result in a decrease in the
effective population size of a wild population, even if supplementation
leads to an overall increase in the census size (the Ryman–Laikre effect).
This first model was important in alerting managers to an important risk
associated with supportive breeding, but examined the outcomes of supplementation over a single generation only. Supportive breeding programmes
are usually carried out over several generations and are typically considered
successful if the programme results in a viable, self-sustaining population.
In a series of modelling exercises, a number of authors have examined
the genetic impacts of supplementation under different management and
demographic scenarios. The approaches used can be divided into two
groups: one focused on effective size or inbreeding and the related effects
on drift (Duchesne and Bernatchez, 2002; Wang and Ryman, 2001; Waples
and Do, 1994), and the other on measures related to fitness differences
between the two components of the population (Ford, 2002; Lynch and
O’Hely, 2001; Theodorou and Couvet, 2004). All models were based on
several necessary assumptions and attempted to identify the conditions
under which supplementation programmes are detrimental or beneficial
to vulnerable wild populations. Both groups have implicit links to the
other, but an integrated model that addresses the effects of both drift and
domestication (or relaxation of selection) has yet to be developed.
All of the studies demonstrated conditions under which supplementation would be negative. For example, Waples and Do (1994) showed that if
a small number of breeders were used in a hatchery, an ‘inbreeding crash’
would result in the wild population after the cessation of an unsuccessful
programme. The relaxation of selection in a hatchery may lead to the
accumulation of deleterious mutations through drift (Lynch and O’Hely,
2001), which may in turn compromise any numerical gains in the population. A wild population’s mean phenotype can rapidly change towards that
of a captive population (Ford, 2002) even when migration between the two
is small.
Several management steps such as increasing the Ne of the hatchery
population may reduce genetic risks associated with releases, but some
theory has shown that this benefit is realized only if the census size of the
entire population increases (Wang and Ryman, 2001) or if the contribution
of captive populations is moderated (Theodorou and Couvet, 2004).
The rate of inbreeding could be reduced if Ne/Nc ratios were high in the
hatchery population (Waples and Do, 1994). Steps such as increasing the
migration rate between the hatchery and wild stocks through broodstock
replenishment from the wild have the advantage of reducing negative
genetic changes (Duchesne and Bernatchez, 2002; Ford, 2002; Wang and
Ryman 2001), but the outcomes of using exclusively wild fish for
114
Kerry A. Naish et al.
broodstock are complex and depend on the scenarios modelled (Duchesne
and Bernatchez, 2002; Wang and Ryman, 2001).
Many models point towards an optimal programme duration. At the
initiation of a program, supplementation could be detrimental to the Ne of
the wild population because the initial demographic effect of sampling wild
individuals to create a broodstock may be negative, and must be compensated for by several generations of support (Duchesne and Bernatchez, 2002;
Wang and Ryman, 2001). On the other hand, all the studies caution that
supportive breeding programmes may not be genetically beneficial in the
long term in many situations. For example, in the selection model of Ford
(2002), it was demonstrated that a shift in a wild population’s phenotype can
still occur within 50 generations in some scenarios modelled, even if
hatchery broodstock comprise natural spawners, and that outcomes may
depend in part on population growth rate and carrying capacity in the native
environment.
Perhaps the strongest message derived from a reading of these six studies
is that the outcomes of supplementation are difficult to predict and may
be programme-specific [an examination of the scenarios modelled by
Duchesne and Bernatchez (2002) supports this view]. Although some
broad conclusions could be drawn, each study has caveats even if the
model assumptions are correct. Thus, strong emphasis must be placed on
monitoring changes in genetic diversity very closely and in developing
meaningful performance measures for hatchery programmes.
Little empirical proof supports theoretical predictions of the outcome of
management practises, partly because most of this theory is very recent, and
many supplementation hatcheries have existed for longer than our awareness of the genetic risks involved. However, several case studies support
theoretical predictions; namely, that genetic diversity can be maintained or
reduced by hatchery founder numbers (Primmer et al., 1999), sex ratios at
mating (Tessier et al., 1997), hatchery population size (Hansen et al., 2000)
and effective population size of released hatchery fish (Eldridge and
Killebrew, 2007; Hedrick et al., 2000; Tessier et al., 1997). Heggenes et al.
(2006) reported that measures of neutral genetic variation and population
structure did not significantly change after 20 years of supportive breeding,
an outcome attributed to the use of overlapping year classes and frequent
integration of wild fish into the broodstock. On the other hand, some
studies do not fit predictions. In Sweden, a hatchery population of sea
trout received no new broodstock from its source wild populations for
approximately seven generations (Palm et al., 2003), but was used to supplement the wild populations, thereby creating unidirectional gene flow.
Effective population size was high in the captive population and, while
small genetic differences were seen between both captive and wild fish on
a yearly basis, these differences were outweighed by temporal variation
(Palm et al., 2003). The hatchery stock used in this study was the same
Evaluation of the Effect of Hatcheries on Wild Salmon
115
population that showed little sign of domestication selection described
earlier (Dannewitz et al., 2003).
Studies on stray rates from supplementation hatcheries are rare and the
extents to which fish from such programmes interbreed with wild fish are
unknown. There are several examples of close genetic relationships between
locally derived hatchery fish and their wild counterparts (e.g., Hansen et al.,
2000; Primmer et al., 1999) and one case in which a wild Chinook population in the Columbia River appears to have maintained its integrity in the
face of supplementation in nearby rivers (Marshall et al., 2000). In contrast,
Williamson and May (2005) suggested that releases of supplementation
hatchery fish in areas that do not correspond to their natal spawning grounds
have led to reduced imprinting and widespread straying and homogenization
of Chinook populations in California.
A key question relevant to evaluating the potential risks and benefits of
supportive breeding is whether hatchery fish are as fit as their wild counterparts and whether they may effectively contribute to conservation efforts
(Berejikian and Ford, 2004; ISAB, 2002). Two studies based on measuring
the reproductive success of locally derived hatchery fish provide some
information on their relative lifetime fitness over the short term. Locally
derived coho salmon (Ford et al., 2006) and steelhead (Araki et al., 2007)
reared in a hatchery to the smolt stage and released were as successful
reproducing in the wild as naturally produced wild fish. These results
provide a clear contrast with the reduced reproductive success of exogenous, domesticated Atlantic salmon (aquaculture escapees; McGinnity et al.,
1997, 2003) and steelhead (Araki et al., 2007) that have been propagated
over several generations. While both studies on the locally derived broodstock provide a somewhat optimistic outlook for conservation programmes,
there are caveats attached to both. Ford et al. (2006) pointed out that their
study was performed on a system which had experienced hatchery releases
for over 60 years, and the naturally produced fish were themselves likely
propagated in a hatchery in the previous one or two generations. Araki et al.
(2007) reported that hatchery fish reproducing with each other in the wild
produced fewer offspring than expected, which has implications for cumulative fitness losses over several generations of propagation. Both studies
emphasize that the long-term effects of supportive breeding programmes are
still unknown.
Returning to our stated aim of evaluating whether negative biological
effects can be avoided by correct management practises, we conclude that
there is insufficient empirical data available, although recent studies on
relative fitness of locally derived hatchery fish provide some insight on
their possible contribution to conservation efforts, and should be replicated
and continued over several generations. The theoretical information has
demonstrated that there are scenarios under which correctly managed
hatcheries may benefit declining populations, and empirical studies should
116
Kerry A. Naish et al.
be carried out within the framework of this theory. However, it is quite
clear there are few general statements that are applicable to supportive
breeding programmes. Theory has shown that the ‘window of opportunity’
for rehabilitating populations may be limited to short time periods, and
practically, programmes must be accurately monitored to avoid negative
effects. The permanent use of conservation-based hatcheries may be risky,
since there appear to be substantial genetic risks associated with the failure to
sustain large and genetically diverse populations. The maintenance of such
hatcheries therefore depends on risk trade-offs that necessarily relies on clear
dialogue between science and policy (Waples and Drake, 2005).
5.1.2.3. Production hatcheries in the presence of wild stocks The typical
production hatchery practise of releasing a closed, and at least partially
domesticated, population for harvest opportunity can result in introgression.
Introgression may, in turn, lead to a change in the mean phenotype towards
that of the hatchery fish, to outbreeding depression and to complete
displacement of the wild population.
Although the effects of production hatcheries on wild fish have not been
explicitly modelled, many of the theoretical treatments examined above
provide insight on the outcomes of such programmes. If a closed, captive
population makes a large contribution to the breeding pool, genetic load
may increase substantially over the long term (Lynch and O’Hely, 2001;
Theodorou and Couvet, 2004), but even low levels of gene flow from the
hatchery to the wild populations can move the optimal wild phenotype
towards that of the hatchery fish (Ford, 2002). None of the theoretical
studies explicitly model the use of exogenous and domesticated stocks that
have been typically used in production hatcheries. Some recent empirical
evidence for the outcomes of releasing hatchery fish are summarized in
Table 2.2, but few general trends can be inferred from the systems studied
because release numbers, duration of releases and broodstock management
have been poorly documented. Hatchery releases pre-date any concerted
scientific studies and in many cases in Europe stocking has been practised for
150 years or more (Berrebi et al., 2000a; Laikre, 1999).
There is extensive evidence that hatchery-wild hybrids from production
hatcheries are less fit than wild fish (Table 2.2) and rates of introgression vary
with life history strategies. For example, studies in brown trout indicate that
introgression rates are higher in resident than anadromous forms. The more
complex life history of the anadromous forms probably precludes completion of the life cycle (Ruzzante et al., 2004). In a review paper, Utter (2001)
noted that anadromous fish from different evolutionary lineages are less
likely to hybridize with each other than those within lineages. In support of
this view, Ford et al. (2004) reported that introduced coho populations from
Washington State did not appear to persist in Oregon, whereas transfers
between closely related populations appear to have affected population
Summary of genetic effects of hatchery fish on wild fish, categorized by source of hatchery fish and species
Species and life history
Region
Exogenous source of hatchery fish
S. trutta, anadromous
Denmark
Norway
S. trutta,
resident
Norway
Spain
Outcome
Study
Reduction in fine-scale population structure,
level of introgression correlated with
intensity of release activity
Low incidence of released fish. Hatchery fish
provided little harvest opportunity for the
anadromous population, but introgressed
with the resident forms
Decline in incidence of domesticated fish
following cessation of releases
Fritzner et al., 2001;
Ruzzante et al.,
2001
Hansen, 2002;
Hansen et al., 2002
Mortality of hatchery fish at early life history
stages was higher than wild fish, may
reduce incidence of introgression
Hatchery fish hybridized with wild fish, but
survival of hybrids was lower than wild
Little impact of hatchery on wild fish
population structure, despite 40 years of
stocking
Extensive introgression in all populations
studied, reduction of population structure
Fritzner et al., 2001;
Ruzzante et al.,
2004
Borgstrom et al.,
2002
Evaluation of the Effect of Hatcheries on Wild Salmon
Table 2.2
Skaala et al., 1996
Heggenes et al., 2002
Cagigas et al., 1999;
Garcia-Marin
et al., 1998;
Machordom et al.,
1999, 2000
117
(continued)
118
Table 2.2
(continued)
Species and life history
Region
France
Outcome
Study
Release of hatchery fish did not improve
population size
Straying of hatchery fish from areas that
received releases versus ones that did not.
Hatchery genotypes persisted after
cessation of releases
Lower incidence of introgression in
harvested regions than in protected areas
Decline in alleles of domesticated origin 7
years after cessation of releases
Some selection against hatchery fish, but
cessation of releases for 6 years did not lead
to recovery of wild genotypes
Levels of introgression varied with intensity
of release activity
Machordom et al.,
1999
Cagigas et al., 1999
Aurelle et al., 1999;
Poteaux et al.,
2000
Berrebi et al., 2000a
Kerry A. Naish et al.
Incidence of introgression was small
annually, but accumulation of hybrid
genotypes increased over time.
Incorporation of locally derived
broodstock may have maintained
population variation
Garcia-Marin et al.,
1998, 1999
Almodovar et al.,
2001
Poteaux et al.,
1998a,b
Italy
S. salar, anadromous
North America, east
coast
Denmark
Estonia
O. mykiss, anadromous
United States, west
coast
Introduction of domesticated strain reduced
reproductive barriers between two
indigenous forms
Degree of admixture was site specific and
may be linked to management actions—
intensity of release activity and fishing
Hybrids between hatchery and wild fish were
less fit than wild fish, mortality of hybrids
greater between age 1þ and 2þ. Hybrids
were more vulnerable to fishing
Reduction in genetic diversity and structure,
replacement of wild subspecies with
hatchery fish
Evidence that past release activities reduced
stock structure in some drainage systems
Translocations or hatchery releases have had
little effect on long-term population
structure
Release of hatchery fish compromised
recolonization of restored habitat by
indigenous populations
Increased straying by non-native hatchery
fish
Largiader and Scholl,
1995
Largiader and Scholl,
1996; Mezzera and
Largiader, 2001a;
Mezzera and
Largiader, 2001b
Mezzera and
Largiader,
2001a,b
Marzano et al., 2003
King et al., 2001
Evaluation of the Effect of Hatcheries on Wild Salmon
Switzerland
Nielsen et al., 1999
Vasemägi et al., 2001
Schroeder et al.,
2001
(continued)
119
120
Table 2.2
(continued)
Species and life history
Region
United States, west
coast
United States, west
coast
O. tshawytscha, anadromous
United States, west
coast
European Alps
Salvelinus umbla, resident
Salvelinus nyamacush, resident
United States,
southeast
Study
Hatchery females had lower reproductive
success than wild fish
Introduced summer run hatchery fish
hybridized with indigenous winter run
populations
Hatchery fish were more abundant on
spawning grounds and produced more
offspring, despite lower reproductive
success
Introgression led to a reduction in disease
resistance in indigenous population
Rainbow trout/cutthroat trout hybrids were
difficult to detect in generations later than
the F1 in the wild, provided a mechanism
for undetected introgression
Population structure persisted, despite
extensive hatchery releases
Extent of introgression was small and varied
across environments
Higher introgression of hatchery fish in a lake
disturbed by pollution than one that
remained undisturbed
Extent of introgression with non-native
northern form varied from replacement to
Kostow et al., 2003
Mackey et al., 2001;
Kostow et al., 2003
Kostow et al., 2003
Currens et al., 1997
Campbell et al., 2002
Utter et al., 1995
Brunner et al., 1998
Englbrecht et al.,
2002
Kerry A. Naish et al.
O. mykiss,
resident
O. clarkii,
resident
Outcome
S. salar, anadromous
Ireland
O. tshawytscha, anadromous
New Zealand
O. gorbuschi, anadromous
Alaska
Galbreath et al.,
2001; Hayes et al.,
1996;
Reduction in genetic diversity within
hatchery stocks
No differences in survivorship observed
between hatchery and wild fry at
18 months
Locally derived hatchery fish exhibited
similar return numbers but different life
history. Translocated population from
neighbouring drainage had lower return
rates
Significant changes observed in male life
history characters and in return timing.
Population was introduced from California
Hatchery derived from local broodstock
appeared to have no affect on wild
population structure in the same locality
Was and Wenne,
2003
Crozier and Moffett,
1995
McGinnity et al.,
2004
Unwin and Glova,
1997
Evaluation of the Effect of Hatcheries on Wild Salmon
Endogenous source of hatchery fish
S. trutta, anaromous
Poland
no detectable hybridization. Erosion of
population structure, level of introgression
not related to intensity of release activity
Seeb et al., 1999
121
122
Kerry A. Naish et al.
structure in the same species (Eldridge and Naish, 2007). The incidence of
hybridization within freshwater salmonids appears to vary greatly
(Table 2.2). Hybridization between hatchery and wild fish can be extensive
and detrimental (Cagigas et al., 1999; Garcia-Marin et al., 1998), with little
benefit to the population size (Machordom et al., 1999). On the other hand,
descendants of hatchery fish may be undetectable even after a long period of
hatchery releases (Heggenes et al., 2002). Many studies report greater
introgression with higher release numbers (Aurelle et al., 1999; Mezzera
and Largiader, 2001a), but others did not (Hayes et al., 1996). Some authors
observed a decline in hybrid genotypes after releases are stopped
(Almodovar et al., 2001), but these genotypes persisted in other areas
(Poteaux et al., 1998a). There is also some evidence that the incidence of
hybridization may be environmentally or ecologically dependent. Arctic
charr (Salvelinus umbla) released into an undisturbed lake in the Alps were
less successful at becoming established than those released into a historically
polluted one (Englbrecht et al., 2002).
Management strategies for production hatcheries have been proposed in
order to avoid the negative impacts of hatchery fish on wild populations
(ISAB, 2002; Mobrand et al., 2005; Utter, 2004). Utter (2004) has suggested
that the domestication of hatchery broodstock can be expected to lead to
substantially reduced fitness of hatchery fish in the wild. The release of such
stocks may be beneficial under management scenarios that are aimed at
deliberately segregating hatchery and wild fish, because reduced fitness of
hatchery fish would minimize concerns about the impacts of colonization
and hybridization. There is not yet sufficient data to determine whether this
is a viable strategy; while many studies report hybridization between lessadapted hatchery and wild fish (Table 2.2), there have been few concerted
efforts at deliberately domesticating ‘maladapted’ hatchery fish for segregated
programmes and tracing their reproductive success in the wild. There are
some existing approaches that may support this goal, however. For example,
certain steelhead production hatcheries in the United States have introduced non-native populations that differ in their run timing and spawn
timing from indigenous populations. While the effects of these programmes
have not yet been fully characterized, it appears that the life histories of
the hatchery fish may change in response to the new environment and in
some examples, return timing and spatial distribution have been seen
to overlap with that of the wild fish (Mackey et al., 2001), making introgression with wild individuals possible, but in one case, limited (Kostow,
2004). Cutthroat trout (O. clarki) are deliberately hybridized with rainbow
trout (O. mykiss) because the hybrids can be identified from the parental
species phenotypically and can be targeted by anglers. However, hybrid
individuals of generations later than the F1 cannot be reliably identified
and hence escape capture, leading to ongoing inter-specific introgression
(Campbell et al., 2002). Segregation can be controlled, to some extent,
Evaluation of the Effect of Hatcheries on Wild Salmon
123
by complete harvest of the hatchery population. Brown trout hatchery fish
in one study appear to be more susceptible to angling (Poteaux et al., 1998a),
and a review of programmes in Spain showed that there are lower levels of
introgression in harvested than protected regions (Garcia-Marin et al.,
1998). However, such an approach relies on efficiency of capture.
Given these caveats, Mobrand et al. (2005) have recommended that an
alternative strategy could be considered, the integration of hatchery fish and
wild fish. Fish surplus to the maintenance of the wild population may be
harvested. The use of endogenous sources of broodstock for production
hatcheries has both positive and negative aspects. Using native stocks may
reduce losses associated with the production of less fit hybrid individuals
typical of exogenous fish releases. For example, a pink salmon hatchery
using locally derived broodstock appeared to have had little effect on the
population structure of indigenous wild populations (Seeb et al., 1999).
A second hatchery collects broodstock migrating into PWS destined for
different regions, and the hatchery fish released comprised a mixture of
stocks with greater potential to affect local populations. On the other hand,
it may also be argued that exogenous fish are more likely to be purged from
a wild population, especially if a programme is terminated. There is also
some evidence for change in life histories, even when endogenous fish have
been used and fish have been reared for part of their life cycle in the
hatchery (McGinnity et al., 2004) sometimes causing a subsequent shift in
such traits in the wild population (Unwin and Glova, 1997). Finally,
theoretical approaches have shown that long-term integration between
hatchery and wild stocks is not always a sustainable strategy, which is further
exacerbated if that stock is subject to harvest (Goodman, 2005). These
studies emphasize the need for further research on the impact of broodstock
management and release.
In summary, published studies show that production hatcheries have
been detrimental to local wild populations where the two populations
interact, although there are many examples where distantly related populations do not appear to have persisted. It should be noted that many genetic
studies have focused primarily on reporting levels of introgression only, and
results are rarely correlated with the size of release. It is also possible that
less fit hybrid individuals may have reduced the overall effective population
size and structure of wild fish, thus causing changes in the life history of
wild populations, and this aspect should also be studied in greater depth.
It is still difficult to ascribe outcomes of production hatcheries to management or biological causes. Management strategies for production
hatcheries advanced thus far would be to maintain the hatchery and wild
fish as separate populations (Mobrand et al., 2005; Utter, 2004), or to
integrate hatchery and wild populations (Mobrand et al., 2005), but the
efficacy of these approaches over the long term has yet to be demonstrated
empirically.
124
Kerry A. Naish et al.
5.1.2.4. Introduced species The greatest genetic risk associated with the
introduction of a new salmon species to a habitat is hybridization with
native indigenous populations. Hybridization in this case has three primary
genetic outcomes: introgression, repeated introgression leading to hybrid
swarms in which neither of the parental genotypes persist, or sterility and
hence lost reproductive potential (Allendorf et al., 2001; Leary et al., 1995).
Hybrids of brown trout and Atlantic salmon tend to be unidirectional
and may compromise one species through introgression, but not the other
(Garcia-Vazquez et al., 2004; McGowan and Davidson, 1992). Hatchery
releases have led to hybridization between these two species (Jansson and
Oest, 1997), and although outside the realm of this chapter, has increased
following escapes from salmon farms in Europe (Matthews et al., 2000).
The widespread introduction of rainbow trout for angling has threatened
the genetic integrity of many western freshwater species in North America
(Leary et al., 1995; Scribner et al., 2000). For example, rainbow trout form
fertile hybrids with cutthroat trout (Allendorf and Leary, 1988) and introgression can be greater in regions of hatchery introduction than in areas
where the two species co-occur naturally (Docker et al., 2003). Many
populations have been replaced by hybrid swarms (Allendorf and Leary,
1988; Williams et al., 1996), which are of no evolutionary or conservation
value (Allendorf et al., 2001). Similar results have been recorded in Europe,
for example, in areas where brown trout have been introduced to marble
trout (S. marmoratus) habitats (Berrebi et al., 2000b). Interactions between
native bull trout (Salvelinus confluentus) and introduced brook trout
(Salvelinus fontinalis) result in unidirectional hybridization (Kanda et al., 2002),
and reproductive effort is substantially compromised in the former species
because bull trout tend to be the maternal contributor. The majority of fish
are F1 hybrids, with very few backcrosses detected (Kanda et al., 2002).
Therefore, bull trout populations are demographically compromised by the
reduced reproductive output following the introduction of brook trout for
fishing purposes. Finally, the introduction of a new species for harvest opportunity may also have indirect genetic effects on native populations through
competition and restriction of the ranges of native populations.
5.1.3. Can management practises negate genetic impacts?
To summarize this section of the chapter, we evaluate whether the current
state of knowledge provides guidance on management steps that may be
taken to reduce the genetic risks associated with different salmon hatchery
programmes. In the decade since Campton (1995) noted that distinction
between management and biological risks were rarely elucidated, the
majority of research has shown that hatcheries can affect genetic diversity
within hatchery populations, and that interactions between hatchery and
wild populations can be detrimental. However, a growing number of
studies have shown that specific steps in broodstock management have led
Evaluation of the Effect of Hatcheries on Wild Salmon
125
to both negative and positive outcomes. In recent years, a shift in the
research is beginning to move from reporting problems associated with
individual case studies towards researching possible solutions, driven in
part, by attempts to reform hatchery practises. A key need in this area,
therefore, is the development of a strong understanding of the degree to
which specific activities pose a risk and whether proposed management
approaches are effective at reducing these risks.
If we consider the diverse biological outcomes of hatchery rearing, there
are still a significant number of unanswered questions. Accumulating evidence has shown that inbreeding in salmonids leads to fitness declines, and
these declines vary by the trait measured. However, the relationship
between inbreeding and the point at which inbreeding depression becomes
manifest is still unclear. In other words, at what population sizes and
generation can we expect a decline in fitness due to inbreeding? The answer
is unlikely to be simple and can be extended to exploring how the incidence
of inbreeding depression is related to historical inbreeding levels, to different
wild environments, to life history strategy and to rate of inbreeding.
The related management questions would therefore be concerned with
developing strategies to both avoid and recover a population from suffering
inbreeding depression.
Domestication selection remains a controversial topic, and research thus
far has been directed at describing individual case studies that provide
evidence of this phenomenon. Measuring the magnitude and direction of
domestication selection under different selection regimes typical of hatcheries, and testing whether the genetic outcomes are reversible if selection is
relaxed, is necessary. Research relevant to management should be directed
towards evaluating strategies to reduce the magnitude of domestication
selection by integrating wild individuals into the broodstock, as well as
understanding the relationship between selection and population size and
generation number in order to gain an understanding of the duration over
which hatchery programmes should be maintained.
Theoretical treatments of the demographic and fitness effects of hatchery
releases on the genetic variation and effective population size of wild
populations have proved very informative and have illustrated the potential
of various management approaches. It is important to provide empirical
support of steps aimed at maximizing effective size and reducing demographic effects of releases on the wild populations. Systematic approaches
are ambitious and long-term, but can be accommodated by close monitoring
of a large range of existing hatchery programmes.
Fitness declines associated with outbreeding have been clearly demonstrated in a large number of studies. However, if a threatened population
requires rehabilitation by the introduction of new broodstock, it is still
unclear how closely related donor and recipient populations should be in
order to avoid outbreeding depression. It appears that genetic distance may
126
Kerry A. Naish et al.
prove an unreliable measure, but few studies have examined the relationship
between measures of a range of evolutionary distances and the probability of
outbreeding depression within a given salmonid species. Managementrelated research should be directed towards determining acceptable levels
of introgression and understanding how evolutionary distances relate to
these levels.
Many of these questions are difficult to answer, particularly because they
require prolonged experimental periods and substantial support. We
emphasize that hatchery-directed research has thus far focused on reporting
the genetic outcomes of specific programmes, with only recent exploration
of the relationship between these results and management actions. In our
chapter [and that of Campton (1995)], we have found little evidence of this
delineation and, unfortunately, little insight into whether research programmes are now being directed towards exploring this relationship.
This weakness has been exploited by opposing viewpoints on whether
hatchery fish should be permitted to spawn in the wild (Brannon et al.,
2004a; ISAB, 2002).
If we accept the likely outcome that hatchery programmes will persist,
then two general research directions should be implemented in order to
provide practical management advice. The first should focus on developing
a clear understanding of the relationship between genetic risks involved in
hatchery releases, and steps to reduce these risks, even if these experiments
are expected to take place over several generations. Most hatcheries have
been established without research programmes, and a strong emphasis
should be placed on devoting at least a portion of the rearing space to
experimental releases. It is only by directly comparing a network of experiments in hatcheries with similar goals that many of the risks and management approaches may be quantified. The second direction should focus on
developing a risk averse approach to management, as advocated by Waples
(1991, 1999), which implements strict monitoring protocols. These protocols should track fitness changes in hatchery and wild populations using a
mixture of approaches. Such data could also contribute significantly to a
large meta-analysis that would allow evaluation of the genetic risks posed by
releasing cultured fish into the wild. Hatchery programmes have existed for
many decades, yet surprisingly, a large programmatic approach to answering
many outstanding questions about genetic risks and remedial management
practises has yet to be executed. We strongly advocate such research.
Finally, it is important at this point to raise the social context in which
research in this discipline is conducted. Waples (1999) and Waples and
Drake (2005) pointed out that genetic changes associated with hatcheries
are inevitable. Research will likely determine the genetic consequences of
hatchery programmes, but societal values must be consulted to determine
the degree to which these consequences are acceptable. Efforts to improve
the interaction between the two are strongly encouraged.
Evaluation of the Effect of Hatcheries on Wild Salmon
127
5.2. Behavioural and ecological interactions between
wild and hatchery-produced salmon
There are important implicit (though seldom explicit) assumptions of
hatchery programmes regarding ecological and behavioural processes.
Violation of these assumptions may result in lower than anticipated production either from the hatchery or from the region that includes hatchery and
coexisting wild populations. First, it is assumed that the hatchery increases
the abundance or survival of salmon populations during the life history stage
that limits the wild population size and that subsequent stages are not
limiting. Second, it is assumed that there are no significant interactions
between wild and hatchery fish that would limit the enhancement effort.
This section briefly reviews these issues, with emphasis on competition
between wild and hatchery fish, and a brief consideration of predation.
5.2.1. Competition between wild and hatchery fish
The majority of lifetime mortality in salmonids generally takes place during
the period from fertilization to emergence from the gravel several months
later. Much of this mortality results from poor circulation of water or low
dissolved oxygen concentration (often related to fine sediment), scour from
flooding, desiccation or freezing and disturbance by the digging of other
female salmon (Quinn, 2005). It has been known for well over a century
that salmonid eggs and milt can be taken from adults, mixed and the
embryos incubated with higher survival rates than commonly occur in
nature [reviewed by Bottom (1997) and Lichatowich (1999); see also
Section 2]. Early hatchery programmes were predicated on the assumption
that increased number of fry released into the rivers would produce commensurate increases in adults. However, the extent to which this is true
depends on the species involved. Almost all salmonid species characteristically emerge from stream gravels and rear for months or years in the stream
before migrating to the sea, a lake or a larger river. The generally low
productivity of streams caused these species (e.g., coho and Chinook
salmon, rainbow and cutthroat trout, Atlantic salmon, brown trout, Arctic
charr) to evolve territorial behaviour. Juveniles defend territories from
conspecifics and heterospecifics with stereotyped displays and overt aggression. Decades of research have indicated that food and space limit the
density of juveniles and production of smolts from streams (Bradford et al.,
1997; Chapman, 1966), though habitat quality (e.g., Sharma and Hilborn,
2001) and environmental conditions cause production to vary among sites
and years.
The ability of individuals to acquire and retain high-quality feeding
territories depends on a number of interrelated factors. Not surprisingly,
larger fish dominate smaller ones, and even a small size disparity is sufficient
to determine the outcome of a contest, but territorial possession also
128
Kerry A. Naish et al.
strongly influences competition (e.g., Abbott et al., 1985; Huntingford and
Garcia de Leaniz, 1997; Rhodes and Quinn, 1998). Both of these factors
favour early emerging fry because they will hold territories and will have
grown by the time that later emerging fry try to compete with them (Einum
and Fleming, 2000), though predation may serve as a countervailing pressure
(Brännäs, 1995). There are many other factors affecting dominance relationships, including recent experience in territorial bouts, individual recognition,
sibling recognition and metabolic rate. However, individuals that are unable
to obtain territories may adopt non-territorial ‘floater’ behaviour patterns
(Nielsen, 1992) or be forced to emigrate (Chapman, 1962).
Given the limited food and space in streams, salmonids evolved under
high levels of competition among juveniles. Even under some fishing
pressure, far more juveniles emerge from the gravel than can be supported
in the stream. Thus, for these species it is the fry to smolt period rather than
the egg to fry period that is really the limiting life history stage, assuming an
adequate number of adults return to spawn. Therefore, the release of
hatchery-produced fry or parr into a stream may not increase the number
of seawards migrating smolts due to simple competition. However, the
hatchery fish may differ from the wild fish in phenotypic traits affecting
dominance. For example, if they are fed for some period in the hatchery
prior to release then they may be larger than the wild fish. In addition, the
timing of spawning in many hatchery populations diverges, commonly
becoming earlier than the wild population from which it was derived
(e.g., Flagg et al., 1995; Quinn et al., 2002). This difference would magnify
any disparity in size between wild and hatchery fish. Nickelson et al. (1986)
studied 30 streams, half of which had received presmolt coho salmon from
hatcheries, and half were unaffected by such activity. Hatchery releases
increased the overall density of coho salmon but decreased the abundance
of wild coho. Similar numbers of adult salmon returned to the two types of
streams but the hatchery-origin fish tended to return earlier in the season
and produced fewer offspring, so the hatchery releases failed to increase the
productivity of the recipient streams. The authors of this finding suggested
that competitive displacement may have been a mechanism underlying this
outcome, but this mechanism was not explicitly tested in their study.
There have been many comparisons between the behaviour of wild and
hatchery fish in laboratory experiments and also many field studies of the
effects of adding hatchery fish to a population of wild or naturally rearing
fish, for example, brown trout (Berg and Jorgensen, 1991; Jorgensen and
Berg, 1991) and coho salmon (Rhodes and Quinn, 1999). The latter type of
study is relevant but, as Weber and Fausch (2003) pointed out, very few
have distinguished the effects of competition per se from the effects of
increased density. In most cases, growth or some other performance measure was recorded in a population of wild fish, and compared to that in a
group of wild fish to which hatchery-produced fish were added. In such
Evaluation of the Effect of Hatcheries on Wild Salmon
129
cases a ‘substitutive’ experimental design that controlled for overall fish
density might be more informative about the processes of competition,
though perhaps less representative of the normal management practise.
Competition for food and space in streams may limit many salmonid
species, but this is not the case for pink, chum and sockeye salmon.
These species commonly spawn at much higher densities than the other
species of Pacific salmon and are much more numerous overall. Pink and
chum salmon migrate directly to sea after emergence and make little or no
use of streams for rearing, whereas sockeye salmon typically migrate to
lakes. Conventional wisdom had maintained that salmonids were limited
by freshwater constraints but that the ocean had the capacity to rear more
salmonids than the rivers could produce. Thus, increases in production of
juvenile pink and chum salmon should be accompanied by proportional
increases in adults; sockeye salmon might be limited by either spawning or
lacustrine rearing capacity. However, between the streams and the ocean
lies the estuary, a habitat whose role in salmonid ecology is not fully
understood (Thorpe, 1994). Is the estuary a critical habitat, merely a
highway through which they must migrate or possibly a hazardous place
filled with predators? Generally speaking, the species that enter the estuary
at a large size move through it more rapidly than smaller salmonids. Atlantic
salmon and sea trout are large when they migrate to sea, as are steelhead,
cutthroat, sockeye, coho and yearling Chinook salmon. Chum salmon
smolts are small, as are populations of Chinook salmon that migrate to sea
in their first year of life, and these species make the most extensive use of
estuaries (Healey, 1982a; Simenstad et al., 1982). Pink salmon are something
of a paradox as they are the smallest in size on entry into the ocean but seem
to move through estuaries faster than chum salmon. Size of smolts and
growth in the estuary provide an advantage in survival at sea (Healey,
1982b; Neilson and Geen, 1986; Reimers, 1971). Though growth rates in
estuaries are often rapid, the vast majority of juvenile salmonids leave after a
few days or weeks, and there is evidence for food limitation in estuaries
(Reimers, 1971; Sibert, 1979; Wissmar and Simenstad, 1988). However,
the extent to which estuaries present a bottleneck may vary among species.
In the Columbia River in the northwestern United States, for example,
steelhead, coho, sockeye and yearling Chinook salmon tend to swim in the
pelagic zone and remain for only a short time, whereas the under-yearling
Chinook salmon are primarily in the littoral zone and are present over a
much longer period (Dawley et al., 1986). We know of no systematic,
controlled study of the effects of density on wild salmon, or of interactions
between wild and hatchery salmon, nor on the duration of estuarine
residence and survival of salmon, though such effects might occur.
It is plausible that the estuary is a limiting habitat, given the many
millions of smolts that may enter over a short period of time, but can the
ocean also be a limiting habitat? Mathews (1980) used data on density, growth
130
Kerry A. Naish et al.
and survival of coho salmon in Puget Sound and the Columbia River in the
northeastern Pacific to test the hypothesis that increased numbers of hatcheryproduced fish depressed the growth or survival of the species, but the data
were equivocal. Rogers (1980) noted the strong environmental controls over
growth at sea but he concluded that there was a density-dependent reduction
in growth of sockeye salmon in Bristol Bay, Alaska, an area where this
species is very abundant. Subsequent to that report, a body of evidence
developed indicating that the density of salmon at sea affected their growth
and age at maturity. Within large ‘stock complexes’ such as Bristol Bay
sockeye salmon and Japanese chum salmon, years with high abundance
were usually associated with smaller size (e.g., Kaeriyama, 1998; Rogers
and Ruggerone, 1993), though interactions with physical conditions
(e.g., temperature) were also noted. For example, the increases in abundance
of chum salmon from Hokkaido hatcheries were accompanied by reduced
size at age and increased age at maturity (Kaeriyama, 1998). Rogers and
Ruggerone (1993) found that the growth of Bristol Bay sockeye salmon
was depressed in the final period at sea by their own density but was not
affected by other salmon (they were especially interested in possible growth
reduction related to the abundance of Japanese chum salmon). However,
McKinnell (1995) and Pyper and Peterman (1999) both reported evidence of
competition between stock complexes of sockeye salmon.
There is evidence, at least for some of the very large stock complexes, of
density-dependent growth. Thus increases in hatchery production might be
associated with smaller size and lower survival of those fish, and perhaps for
sympatric salmon of the same and even other species (Levin and Williams,
2002), and authors such as Cooney and Brodeur (1998) have discussed the
possible implications of marine carrying capacity for salmonid enhancement
efforts. However, the extent to which these effects occur in areas with
more dispersed production and lower overall densities is unclear. Perhaps
more fundamentally, does high density depress only growth or survival as
well? Evidence on this crucial point is much less clear, but recently
Ruggerone et al. (2003) reported that not only the growth but also the
survival of Bristol Bay sockeye salmon was depressed by the abundance of
Asian pink salmon. In addition, Levin and Schiewe (2001) concluded that
under conditions of naturally low ocean productivity, high densities of
hatchery Chinook salmon depress survival rates of wild conspecifics.
In general, the 1980s and 1990s have seen high abundance and survival
rates of Pacific salmon from the northern part of their North American
range, and ‘predator swamping’ effects might lead to a positive relationship
between abundance and survival rather than a negative one. Indeed, earlier
analysis indicated a positive relationship between survival of Babine Lake
sockeye salmon in British Columbia and the abundance of juvenile pink
salmon (Peterman, 1982). However, the question certainly needs further
work before this finding can be accepted as a general conclusion.
Evaluation of the Effect of Hatcheries on Wild Salmon
131
In addition to the potential competition for food and space between
wild and hatchery-produced juvenile salmon in streams, and for food in
estuaries and the ocean, there are possible competitive effects and behavioural interactions on the spawning grounds. If all wild fish spawned in the
river where they originated, and all hatchery fish returned and were
spawned in the hatchery where they were produced, these interactions
would not occur. However, this kind of segregation seldom occurs. First,
there is some straying of hatchery-produced fish into other rivers (Candy
and Beacham, 2000; Labelle, 1992; Quinn et al., 1991). Moreover, even if
the salmon return to their river of origin, there are often opportunities for
exchange between wild and hatchery populations. Nicholas and Downey
(1983) reported that the proportion of hatchery-produced Chinook salmon
entering Elk River Hatchery, Oregon, averaged 22.8% over a 9-year period
(range, 5.9–52.2%). Hence, in most years, the majority of fish produced in
the hatchery did not spawn there but rather in the river. In another case of
interaction between wild and hatchery salmon, Nicholas and Van Dyke
(1982) estimated that 2022 (64.7%) of the 3124 wild coho salmon returning
to the Yaquina River watershed in 1981 entered the Oregon Aqua-Foods
hatchery. Such decoying of wild salmon into hatcheries both reduces the
number of wild fish in the stream and contributes to genetic mixing.
On the other hand, hatchery fish commonly spawn with naturally produced fish and can outnumber them in some systems. In an extreme example,
Nicholas and Van Dyke (1982) estimated that 6% of the adult coho salmon
returning after release from the Oregon Aqua-Foods, a private production
hatchery, strayed to spawn in the Yaquina River watershed, Oregon.
However, they were so numerous (and the wild fish so scarce) that hatchery
fish constituted 74% of the naturally spawning coho salmon in 1981 (Nicholas
and Van Dyke, 1982) and 91% in 1985 ( Jacobs, 1988). If a stream’s carrying
capacity has already been met by the wild salmon, then the addition of hatcheryproduced fish will diminish the productivity of the group as a whole, even if
there are no functional differences between wild and hatchery fish.
Just as juvenile wild and hatchery salmon differ in phenotype and
genotype, with complex implications for their behavioural and ecological
interactions, adults differ as well. The nature and extent of the differences
vary greatly, depending on whether the wild and hatchery populations have
been managed as a single unit (e.g., wild salmon used for spawning in the
hatchery and hatchery-produced salmon allowed to spawn naturally, and all
salmon subjected to common fisheries) or managed separately. For example,
in Washington State, salmon hatcheries have tended to employ the former
approach. Until recently, most hatchery-produced salmon were not
marked, so fisheries operated equally on wild and hatchery fish, and there
was considerable exchange between populations. In such cases the wild and
hatchery populations may be essentially the same, and most differences
between them may result from culture practises. For example, hatchery
132
Kerry A. Naish et al.
smolts are often larger than wild smolts, and this tends to reduce the age at
maturity (e.g., Norris et al., 2000), and hence overall size.
On the other hand, steelhead in Washington State have been managed
under a very different set of goals. Hatchery steelhead have been selectively
bred to return and spawn earlier in the winter than wild fish (e.g., Ayerst,
1977). This approach was initially implemented to lengthen the growing
season for juveniles in the hatchery so they could reach a suitable size for
smolt transformation and release after 1 year rather than 2 years as is typical
of wild steelhead. However, the high genetic variability underlying return
date allowed managers to select early returning fish, and hence open early
fisheries that targeted the hatchery-produced fish and close the fisheries later
if the wild populations needed protection. In this situation, when the
hatchery-produced fish spawn in the river, they do so earlier in the year
than the wild fish. This approach may expose the hatchery fish to less
favourable physical conditions (e.g., flow, temperature or loss of redds to
later-arriving wild adults) since presumably the wild fish evolved an optimal
spawning season to maximize embryo survival or fry growth. It is therefore
perhaps not surprising that the hatchery fish do not produce as many
surviving offspring per capita as do wild fish when spawning in the river,
as indicated by genetic analyses (e.g., Leider et al., 1990).
5.2.2. Predation
Although most research on behavioural interactions between wild and
hatchery-produced salmon has emphasized competition, predation is
another important ecological interaction. Salmonids tend to eat invertebrates (e.g., insects in streams and zooplankton in lakes) when they are small
but they become more piscivorous once they reach about 10–20 cm
(Keeley and Grant, 2001). Studies in freshwater (e.g., Hunter, 1959) and
at sea (e.g., Parker, 1971) identified coho salmon as a significant predator on
pink and chum salmon. Coho smolts (usually about 10- to 12-cm long)
can easily consume newly emerged members of the other species (about
3–4 cm). As the pink and chum salmon grow they become progressively less
vulnerable (Hargreaves and LeBrasseur, 1986), but recently Briscoe et al.
(2005) reported that the survival of Auke Creek coho salmon was positively
correlated with the numbers of pink and chum salmon fry released from
nearby hatcheries. Likewise, coho salmon in lakes can be a significant
predator on sockeye salmon (Ruggerone and Rogers, 1992), and Arctic
charr can congregate and eat sockeye salmon smolts (Ruggerone and
Rogers, 1984). These latter studies were conducted on wild populations,
but releases of large numbers of hatchery-produced coho salmon coincident
with the presence or migration of wild salmon could result in significant
mortality. Pearsons and Fritts (1999) reported that juvenile coho could
eat Chinook that were over 40% of their length (e.g., a 140-mm coho ate
a 64-mm Chinook).
Evaluation of the Effect of Hatcheries on Wild Salmon
133
There has been a tendency to focus on predation by hatchery-produced
smolts on wild fry, but other kinds of predator–prey interactions involving
wild and hatchery fish can occur as well. For example, in Lake Washington,
Washington State, there is a large population of adfluvial cutthroat trout that
prey heavily on wild and hatchery-produced sockeye salmon fry (Nowak
et al., 2004). The presence of hatchery fry might buffer predation on the
wild fry (as predation is buffered by the abundance of longfin smelt, an
alternative forage species for trout), assuming that other factors limit the
abundance of trout. However, if the availability of hatchery-produced
salmonids increases the abundance or modifies the distribution of predators,
increases in predation on wild fish might occur.
5.3. The effects of harvest on wild salmon populations
The underlying principle in the theory of sustainable salmon harvesting is
the stock concept. Due to their ability to home to their natal streams,
salmon have adapted to a wide range of freshwater habitats, and consist of
thousands of reproductively isolated stocks (Helle, 1981; McDonald, 1981).
The population dynamics of each stock will be determined by the habitat it
uses, and a convenient metric of the overall productivity of each stock is the
potential recruits per spawner. A stock that spawns in good gravel with
stable flows, little scouring and few fine sediments can be expected to have
higher egg to fry survival than a stock spawning in an unstable stream with
frequent floods and scour, siltation and intense predation. Similarly, through
the rest of the freshwater and marine life history, a stock using better habitat
would be expected to have higher survival rates. Higher survival through
their life history results in more individuals surviving to return to spawn
for every spawner in their parental generation.
On average, a habitat that has less than one recruit per spawner would
not be able to support a stock of salmon without frequent immigration.
Stocks in good habitat can often produce two to ten recruits per spawner from
adults spawning at low density. The sustainable harvest rate for a population
depends on the number of recruits per spawner. A population producing two
recruits per spawner can be harvested at 50%, one spawner produces two
adults, one is harvested and one remains to replace the parental generation
and complete the cycle. A population with three recruits per spawner can be
harvested at 66%, and a population with four recruits per spawner can
be harvested at 75%.
In the absence of harvesting, populations would be expected to increase
until competition for resources (breeding space for adults or food and space
for their offspring) reduces the recruits per spawner to 1.0; that is, populations cannot grow forever. Thus, when we attempt to estimate the productivity of a salmon population, we normally attempt to estimate the potential
recruits per spawner at low densities. Table 2.3 shows the estimated
134
Kerry A. Naish et al.
potential maximum recruits per spawner and the exploitation rate at maximum sustainable yield (MSY) for a range of natural populations of Pacific
salmon. These estimates were derived from data sets accumulated by
R. Myers (Dalhousie University, Canada), and represent only stocks that
have been well studied and have survived many generations of harvesting.
Therefore, the estimates are almost certainly biased towards the higher end
of natural productivity.
Salmon enhancement projects intervene at specific life history stages to
improve the survival rates, and thus ideally increase both the recruits per
spawner and the potential sustainable harvest rate. For example, if a population of salmon produced 1600 eggs per spawner, and egg to fry survival
and fry to adult survival rates were 5% and 2.5%, respectively, the population would produce two recruits per spawner and could be harvested at
50%. The same stock, if placed in a hatchery with 90% egg to fry survival,
would produce 36 recruits per spawner, and could be harvested at 97%.
Table 2.4 shows how the sustainable harvest rate depends on the release to
adult survival for a hatchery population with 1600 eggs per spawner and
90% egg to release survival.
Table 2.3 Maximum recruits per spawner for some Pacific salmon populations
Species
Number of
stocks
Average maximum
recruits per spawner
Exploitation rate at
MSY (%)
Chinook
Chum
Pink
Sockeye
6
7
52
23
4.4
3.0
2.8
3.5
67
55
54
60
Included are Chinook (O. tshawytscha), chum (O. keta), pink (O. gorbuscha) and sockeye (O. nerka)
salmon.
Table 2.4 Sustainable harvest rate for hatchery fish as
function of smolt-to-adult survival
Ocean survival (%)
Sustainable harvest rate
0.08
0.10
0.50
1.00
2.00
3.00
5.00
0.13
0.31
0.86
0.93
0.97
0.98
0.99
135
Evaluation of the Effect of Hatcheries on Wild Salmon
An extensive tagging programme using coded wire tags since the 1970s
has tagged tens of millions of hatchery fish each year in North America,
allowing us to estimate the survival rate of hatchery fish for most hatcheries.
Figure 2.20 shows the distribution of survival rates from hatchery released
Chinook and coho salmon from this database. The average sustainable
harvest rate for these three species is between 86% and 98%, far in excess
of the sustainable harvest rate for wild stocks.
5.3.1. Sustainable harvest in mixed-stock fisheries
When SEPs are technologically successful, the stocks they produce can be
harvested at very high rates, and this creates one of the primary problems in
management of enhancement programmes. When natural stocks with lower
Coho
Frequency
A
0.25
0.20
0.15
0.10
0.05
0.00
0
0.01
0.02
0.03
0.04
0.05
0.04
0.05
0.04
0.05
Ocean survival rate
Fall Chinook
Frequency
B
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0
0.01
0.02
0.03
Ocean survival rate
Spring Chinook
Frequency
C
0.20
0.16
0.12
0.08
0.04
0.00
0
0.01
0.02
0.03
Ocean survival rate
Figure 2.20 The frequency of ocean survival rates for hatchery release groups of coho
(O. kisutch) (A), fall Chinook (B) and spring Chinook (C) (O. tshawytscha) from all
hatchery releases in the Pacific salmon Coded Wire Tag database (Magnusson, 2002).
The arrows show the average value for the salmon species.
136
Kerry A. Naish et al.
sustainable exploitation rates are mixed in the same fisheries, how do you
harvest the hatchery stocks without over-harvesting the naturally producing
stocks? In theory, we would like to harvest each stock individually, be it
wild or hatchery. In such an ideal world, this practise would allow us to
harvest the hatchery stocks at very hard rates and the wild stocks at the
appropriate rate. Unfortunately, two factors prevent this practise from
happening.
First, for historical reasons, most salmon harvesting does not take place
where the different stocks (natural and hatchery) are separated. This separation often takes place within a river system, since most enhancement
facilities are located well above tidewater, and by tradition most commercial
and recreational harvesting takes place in salt water. In the extreme of large
river systems in North America such as the Sacramento, Columbia, Fraser,
Skeena, Copper and Yukon, dozens (or hundreds) of discrete stocks are
found in the watersheds, often diverging from one another at spawning
grounds hundreds of miles upstream. A primary reason most commercial
fisheries take place in salt water is that the quality of the flesh deteriorates as
the fish enter freshwater, reducing their economic value. Thus, for commercial reasons, freshwater harvesting is very undesirable. It is an unfortunate fact of life that most salmon fisheries are to some extent mixed-stock
fisheries, and the majority of enhanced salmon populations will be harvested
with naturally producing fish when fishery enhancement takes place in a
geographic region with natural production.
The problem of harvesting stocks of differential productivity in a common fishery is commonly called the ‘mixed-stock harvesting problem’, and
has long been recognized and analysed (Hilborn, 1976, 1985b; Kope, 1992;
Paulik et al., 1967; Ricker, 1958; Shaklee et al., 1999; Walters, 1988).
Figure 2.21 shows the relationship between harvest rate and sustainable
yield for a weak wild stock with potentially 1.5 recruits per spawner, and a
stronger hatchery stock that produces 6 recruits to the fishery per spawner.
Panel (A) shows the case where the wild stock has a potential return of 1000
spawners and the hatchery stock of 100. Fishing near the optimum
rate for the wild stock (about 20%) maximizes the total harvest from the
mix of stocks. However, if the hatchery is larger (panel B), its potential
return is 600 spawners (still well below the potential return of the
wild stock), and harvesting at about 70% maximizes the yield. This harvest
rate is near the optimum for the hatchery stock but drives the wild stock
extinct.
5.3.2. Salmon harvesting and impacts of hatchery fish on wild fish
Salmon fisheries can be broadly divided into two types: terminal fisheries
near river mouths targeting fish as they return to a particular watershed, and
mixed-stock or ‘interception’ fisheries that harvest a range of stocks that are
intermingled. Most fisheries near the mouths of larger rivers are actually
137
Evaluation of the Effect of Hatcheries on Wild Salmon
Sustainable yield
A
140
120
100
80
60
40
20
0.05
B
Sustainable yield
300
250
Hatchery
Natural
0.25 0.45 0.65
Harvest rate
0.85
Hatchery
Natural
200
150
100
50
0.05
0.25 0.45 0.65
Harvest rate
0.85
Figure 2.21 Total sustainable yield (TSY) for a fishery with a mix of a weak natural
stock and a hatchery stock.TSYwhen (A) the hatchery stock is small (N ¼ 100) relative
to the wild stock (N ¼ 1000) and (B) when the hatchery stock (N ¼ 600) is just over half
as large as the wild stock (N ¼ 1000).
mixed-stock fisheries, since there is normally a range of stock complexes
within any river system. But we generally do draw a contrast between the
mixed-stock fisheries for immature Chinook and coho salmon that are highly
intermingled along the west coast of North America, with the much more
terminal (and less mixed) fisheries that take place in river mouths. Similar
problems are found in the Atlantic, where many of the traditional fisheries
take place on stocks of very mixed origin (Crozier et al., 2004).
The Chinook fisheries are a very good illustration of the mixed-stock
problem. Figure 2.22 shows the distribution of exploitation rates on Chinook
salmon from four specific hatcheries: Robertson Creek, located on the west
coast of Vancouver Island; Big Qualicum, located on the east coast of
Vancouver Island in the Straight of Georgia; the Nisqually hatchery located
in southern Puget sound and the Upriver Brights (URB) from a hatchery
located on the Hanford Reach of the Columbia River. The fig. shows that the
distribution of exploitation rates for different stocks differs spatially, with the
Robertson Creek stock caught primarily in northern British Columbia and
Alaska, the Big Qualicum and Nisqually stocks caught primarily in the
more local sport and commercial fisheries of the interior waters of British
138
Kerry A. Naish et al.
Alaska Troll
Alaska net
Alaska sport
North Troll
Robertson Creek
Big Qualicum
Central Troll
Nisqually
N/CBC net
URB
N/CBC sport
WCVI Troll
GeoSt Tr&Sp
Canada net
Canada sport
US Troll
US net
US sport
0
20
40
60
Percentage of stock in fishery
Figure 2.22 Distribution of Chinook salmon (O. tshawytscha) stocks in mixed-stock
fisheries on the west coast of North America.
Columbia and Washington, while the URB stock is caught over most of the
outer coast. Data suggest that wild stocks from the same geographic locations
have similar distributions of catch, indicating that almost all of the major
Chinook fisheries are heavily mixed. In other words, harvesters cannot put
their lines or nets in the water without catching fish from many locations,
including a mix of wild and hatchery fish.
It is these mixed-stock fisheries that pose the primary problem for wild
stock managers faced with significant hatchery production. More importantly, the rise in hatchery production of Chinook and coho in the
1960–1980s led to high harvest rates in the coastwide fisheries that led, in
turn, to over-harvest of the wild fish.
139
Evaluation of the Effect of Hatcheries on Wild Salmon
Figure 2.23 shows the historical pattern in exploitation rates on Big
Qualicum hatchery Chinook as estimated from coded wire tagging data.
Each year almost all hatchery stocks on the Pacific coast have a significant
portion of their released fish tagged with small wire tags (coded wire tags),
and catches and escapements are sampled to determine survival after release,
harvest rates and stock contribution to mixed-stock fisheries. When the data
became available, it was clear that the harvest rate on this stock was high,
and indeed higher than that sustainable by wild fish (Pacific Salmon
Commission Joint Chinook Technical Committee, 2002). The pattern
observed for Big Qualicum hatchery fish was generally consistent with
patterns of most wild Chinook stocks on the east coast of Vancouver island;
that is, the harvest rates in the 1970s and 1980s were in excess of the
sustainable rates for wild fish, and it was only in the 1990s that the harvest
rates were reduced.
As we have seen (Section 3), most hatchery programmes on the west
coast of North America have produced Chinook and coho salmon, the
major exception being the recently established large programmes for pink
salmon in PWS in Alaska (Hilborn and Eggers, 2000; Pinkerton, 1994).
There are no fisheries for pink salmon outside of PWS, but the nature of the
fisheries within this area is complex and there are significant concerns that
the harvesting of hatchery fish has impacted the wild stocks. The nature of
the geography and the migration of stocks have certainly posed significant
concerns to the area managers. There is a tension between the desire of the
managers to harvest the hatchery stocks as close as possible to the hatchery to
reduce mixing with wild fish, and the economic desire to harvest the fish
away from the hatcheries while the flesh quality is higher. Hilborn and
Eggers (2000, 2001) showed that the advent of the large pink salmon
hatchery programme in PWS coincided with a decline in the abundance
and productivity of the wild fish at the same time that other wild pink
salmon populations in Alaska were increasing. They suggested that the
Exploitation rate
100%
80%
Fishing-induced mortality
Landed catch
60%
40%
20%
0%
1973 1976 1979 1982 1985 1988 1991 1994 1997 2000
Brood year
Figure 2.23 Historical exploitation rates on Chinook salmon (O. tshawytscha) from the
Big Qualicum hatchery, British Columbia, from 1973 to 2000.
140
Kerry A. Naish et al.
primary impact of the hatchery programme in PWS was to replace wild
with hatchery fish rather than to significantly increase total pink salmon
returns.
There is general agreement that fisheries agencies, in their desire to
maximize the harvest of wild fish, systematically overfished many wild
stocks, which led to the development of wild fish policies in Oregon and
Washington in the 1990s. In 1997, the environmental impact statement for
the Washington Department of Fish and Wildlife’s Wild Salmonid Policy
stated bluntly that ‘current fish management plans and practices overfish 89
wild stocks in order to harvest co-mingled hatchery fish at rates that are not
sustainable by wild populations’ (Washington Department of Fish and
Wildlife, 1997). This problem was not at all unique to Washington and
has been found in every salmon jurisdiction that has had significant hatchery
production.
5.3.3. Selective harvesting
Possible solutions to the mixed-stock harvesting problem include
(1) continuing the overexploitation of wild stocks and relying on hatchery
production, (2) closing of hatcheries, (3) reducing mixed-stock fishery
exploitation rates to levels sustainable by wild stocks and (4) attempting to
selectively harvest hatchery fish, in many cases by permitting fishermen
to retain only hatchery fish (Lawson and Sampson, 1996; Zhou, 2002).
Management agencies on the west coast of North America have mostly
chosen to reduce exploitation rates while trying to selectively harvest
hatchery fish at the same time. Selective fishing relies primarily on marking
hatchery fish and encouraging fishermen to release unmarked fish, often by
law. Thus, in some jurisdictions, all hatchery fish released have their adipose
fin clipped and fishers can only retain adipose clipped fish. Selective fishing
requires not only the ability to identify hatchery fish, but also that the
survival rate of released fish is high.
5.3.4. Impacts of harvest: Summary
When hatchery programmes first became successful at producing significant
numbers of fish for harvesting, the harvest of wild fish in mixed-stock
fisheries was a very serious threat to the viability of the wild stocks. In the
1990s, growing recognition of the problem, aided both by better data from
marking programmes and increasing concern about wild fish, led to a
significant change in harvest policies in the Atlantic and the Pacific.
The adoption of formal policies for protection of wild salmon has led to
dramatic reductions in harvest rates in mixed-stock fisheries that should
allow wild stocks to rebound where their freshwater habitat remains suitable
and ocean conditions are favourable. There remains much discussion and
controversy over the ability of selective fishing to continue to harvest
hatchery surpluses without adversely affecting wild stocks. It remains to
Evaluation of the Effect of Hatcheries on Wild Salmon
141
be seen if these efforts will be successful and, should the results prove
negative, whether society will respond accordingly by reducing or redirecting
demand for harvestable fish.
5.4. Disease effects of salmonid enhancement
In a strict sense, disease can be defined as a departure from normal and may
include alterations in histology, physiology, behaviour or function. Diseases
may have either infectious (e.g., tuberculosis, hepatitis) or non-infectious
(e.g., botulism, cystic fibrosis) causes. Although fish provide many interesting examples of disease resulting from non-infectious etiologies
(Leatherland and Woo, 1998), for the purposes of this chapter we will not
consider diseases of non-infectious origin because they typically are
not transmissible between fish. Nevertheless, there is concern that hatchery
practises can affect levels of non-infectious diseases among wild fish by
amplifying diseases that have a genetic etiology (e.g., certain cancers) or
by the release of chemicals or pollutants. As an example, the use of malachite
green for control of fungus infections in hatchery fish has been largely
discontinued in Europe and North America due to its demonstrated carcinogenicity and concern about its release into the environment via the
hatchery effluent (Srivastava et al., 2004).
Disease is a natural process and one of the factors (along with age and
predation) that determines rates of population mortality. It is important to
remember that infectious disease is a normal component of ecosystems and
that all species live in association with a broad suite of pathogens.
Nevertheless, the presence of a pathogen in nature does not inevitably
lead to infection and, should infection occur, it does not inevitably lead to
disease. Thus, infections of fish can be acute, subacute, chronic or unapparent, and the infected fish may die, recover or become long-term carriers.
Several factors control the disease process in both wild and cultured
populations of fish. These factors rest with the host, the pathogen and the
environment (Hedrick, 1998). For the host, factors might include the
species, stock, age, immune status and nutritional state. For the pathogen,
factors include virulence, number and strain. In a normal environment,
most endemic pathogens are in a relatively balanced relationship with their
natural hosts. Both innate and adaptive immune mechanisms help protect
the host against endemic pathogens, although pathogens with a high rate of
mutation (e.g., RNA viruses) can be described as being in an ‘arms race’
with the host immune system. Because fish live in close association with
their environment, changing environmental factors can have important
effects in altering the balance of the host–pathogen relationship. Such
factors include the presence of stressors, adverse water quality and abnormal
water temperatures. The anthropogenic and natural stressors that reduce
resistance or exacerbate disease in wild fish are typically local, for example,
142
Kerry A. Naish et al.
hydroelectric dams, thermal effluents from power plants, contamination from
mining or industrial activities and altered flows or water temperatures
from natural causes or agricultural practises (Arkoosh et al., 1991); however,
global or large-scale effects may also cause changes in disease ecology (Kocan
et al., 2004).
5.4.1. Infectious diseases of salmonids
A wide variety of viruses, bacteria, parasites and fungi can cause disease in
salmonids and for more detailed information the reader is referred to various
fish health texts (e.g., Hoffman, 1999; Noga, 2000; Plumb, 1999; Roberts,
2001; Wolf, 1988; Woo, 1995; Woo and Bruno, 1998). While the initial
exposure of a population to an exotic disease is often devastating
(e.g., whirling disease in rainbow trout), differences in the host specificity,
virulence and the nature of the resulting disease are frequently seen with
different strains of endemic pathogens as well. For example, IHNV is
endemic among a wide range of anadromous salmonids on the west coast
of North America; however, significant genetic differences have been
shown among the different strains of IHNV that infect sockeye salmon,
Chinook salmon, and the rainbow and steelhead trout (Kurath et al., 2003).
This finding underscores the need for caution to avoid the translocation of
both exotic pathogens and non-native strains of endemic pathogens with
the movement of hatchery fish. In addition, there are significant differences
in the innate resistance to a given pathogen among the various salmonid
species (Nichols et al., 2003; Vincent, 2002) and even among stocks of the
same species (Vincent, 2002). Finally, differences in environmental conditions and other factors (e.g., strains of alternate hosts) can affect the distribution and ecology of disease in various geographic areas. An example is the
difference in the severity of whirling disease among populations of naturally
spawning rainbow trout in different regions of the United States (Kerans
et al., 2004).
5.4.2. Infectious diseases in wild and hatchery salmonids
Typically, the sources of pathogens that can infect fish are endemic among
free-living, facultative pathogens in the aquatic environment (e.g., Flavobacterium psychrophilum, the causative agent of bacterial cold water disease) or
from obligate pathogens that are maintained among reservoirs in freeranging aquatic animals (e.g., Renibacterium salmoninarum, the causative
agent of bacterial kidney disease). Except for a few specific instances
where exotic pathogens have been introduced to a new area by the intentional movement of hatchery fish (see below), these natural sources and
endemic reservoirs among wild fish are the origins for the infectious diseases
that affect both wild and hatchery salmonids (Amos and Thomas, 2002;
Anderson et al., 2000; Mitchum and Sherman, 1981; Olivier, 2002).
Evaluation of the Effect of Hatcheries on Wild Salmon
143
Although infectious diseases are common in populations of wild salmonids, their effects are hard to observe (especially in the ocean) and difficult to
study. Many infectious diseases in wild fish occur at chronic or relatively
low levels unless a significant environmental stressor is present or the
population reaches an abnormally large size. Disease outbreaks that have
resulted in large-scale mortality events among wild fish have been documented for several marine fish species (Hedrick et al., 2003; Rahimian and
Thulin, 1996) and some populations of free-ranging salmonids (Williams
and Amend, 1976). In some cases, these outbreaks have resulted in losses
approaching 90% of the wild stock.
Not surprisingly, much of what we know about infectious diseases of
salmonids comes from experience with captively reared fish, where disease
outbreaks are easily observed and there is an incentive for action, and
because, at least in some cases, various disease control options may be
available. As a result, most research on infectious diseases of salmonids has
focused on those infectious agents causing large economic losses at commercial aquaculture facilities or large impacts at salmonid hatcheries
supporting state, tribal and federal fisheries programmes. This has led some
to the incorrect, but common, perception that fish disease is a hatchery
phenomenon.
In addition to being more easily observed, when infectious diseases
occur among fish in hatcheries, they are frequently found to have a higher
prevalence or intensity than among wild stocks, although exceptions have
been noted (Elliott et al., 1997). Hatchery fish may experience greater
impacts from infectious diseases due to higher densities, higher levels of
stress and poorer water quality leading to an increased level of susceptibility
and lowered ability to recover from infection. Other reasons that outbreaks
of disease are more commonly observed in hatcheries might include a lower
level of genetic diversity in some cases, and the fact that hatcheries typically
rear the most susceptible life stages of fish, especially fry and juveniles.
5.4.3. Disease risks associated with salmon hatchery programmes
While an important area of concern, there are but a few well-documented
cases in which hatchery fish have been shown to affect directly the health or
infectious disease status of wild stocks (McVicar, 1997). Nevertheless, this
remains a considerable area of debate and a major source of scientific
uncertainty requiring additional research. However, there are several
potential mechanisms by which hatcheries could affect the disease status of
wild stocks.
5.4.3.1. Introduction of exotic pathogens While principally associated
with the intentional movement of cultured fish harbouring an undetected
infectious agent, this remains the most dangerous and best-documented threat
to the health of wild stocks. Often cited examples include the introduction
144
Kerry A. Naish et al.
and distribution of M. cerebralis, the causative agent of whirling disease, and
Gyrodactylus salaris, the causative agent of gyrodatylosis. Both of these important salmonid diseases have impacted wild or free-ranging stocks, sometimes
with devastating consequences.
Whirling disease was believed to have been initially introduced into
cultured rainbow trout in the United States sometime in the 1950s either by
direct importation of infected fish from Europe where the causative agent is
endemic or by use of imported fish as trout food (Bartholomew and Reno,
2002). Now present in both captive and free-ranging salmonids in at least 23
states, the parasite continues to spread both by natural means and by the
intentional or unintentional movement of infected fish by commercial farms
and fisheries agencies. Among wild-spawning rainbow trout in the western
United States, declines approaching 90% have been observed in certain
populations (Baldwin et al., 1998). Because several species of anadromous
salmonids are highly susceptible, there is significant concern for wild stocks
of Chinook salmon, sockeye salmon and steelhead trout in the western
United States (Hedrick et al., 2001).
G. salaris is a trematode parasite that is cited as having caused significant
damage to wild Atlantic salmon populations in 44 Norwegian rivers (Peeler
and Murray, 2004). Spread from endemic areas by the movement of
infected fish used in commercial aquaculture (Johnsen and Jensen, 1986),
the parasite is now present in many rivers in Norway with little chance of
eradication.
While the greatest risk of introducing exotic pathogens is associated with
the deliberate movement of infected fish between watersheds, other pathways have been postulated. These include birds, anglers, ballast water and
straying fish (Bartholomew et al., 2005; Peeler and Thrush, 2004). The
operational plans of most conservation hatcheries preclude many of these
risks because they rely on local stocks, have good fish health inspections and
restrict the movement of fish to the same, or nearby, watersheds. Because
fish pathogens are detected most readily when they affect stocks in hatcheries, it is common to assume that a newly discovered pathogen is a result of
an introduction, however, this is frequently not true (Mork et al., 2004).
5.4.3.2. Amplification of endemic pathogens in hatchery fish A second
method by which hatcheries are assumed to impact the health status of wild
stocks involves the creation of a point source of infection from disease
outbreaks that occur in hatchery fish. Since hatcheries often contain high
densities of susceptible fish, such outbreaks can result in the release of
significant quantities of infectious agents in the effluent (Watanabe et al.,
1988); although high levels of pathogens can also be released from wild
salmonids in natural systems (Mulcahy et al., 1983). The threat to
wild stocks from pathogens in hatchery effluents is related to the number
and concentration of infectious units that are released, the dilution of the
Evaluation of the Effect of Hatcheries on Wild Salmon
145
effluent by the receiving waters, the stability of the infectious agent in
the environment and the opportunity to contact a susceptible wild fish.
An example of this type of risk is the amplification of sea lice (Lepeophtheirus
salmonis) by commercial Atlantic salmon farms in Europe and North America
(Krkosek et al., 2005; Morton et al., 2004; Peeler and Murray, 2004; Tully
et al., 1999), and the dramatic collapse of sea trout stocks on the west coasts of
Ireland and Scotland has been attributed to infection of post-smolts during
their migration past such farms (Butler and Walker, 2006; Gargan et al., 2006).
However, others suggest that the role of commercial salmon farms in contributing to local infections is less clear and that infection pressure on wild fish
depends on multiple factors (Brooks, 2005; Costelloe et al., 1998). It should
be noted that few, if any, examples are reported in which fishery enhancement or conservation hatcheries rearing Pacific salmon have been shown to
amplify endemic pathogens in a manner that has resulted in an increase in
disease prevalence or intensity among wild stocks in the watershed. However,
because the same, or very closely related, strains of endemic pathogens infect
both hatchery and wild stocks, it is currently difficult or even impossible to
determine the origin of the infectious agent with certainty (Todd et al., 2004).
Nevertheless, a large hatchery operating on a small watershed that contains a
substantial number of susceptible wild fish could present a source of risk to the
wild cohort.
5.4.3.3. Intentional release of infected fish that contact wild stocks In
addition to the release of pathogens in hatchery effluents, conservation
hatcheries will typically release fish into systems at times or in ways that
attempt to mimic the natural production cycles. In some cases, these
captively reared fish may be undergoing a disease outbreak or harbouring
pathogens that can result in a greater than normal risk of infection for the
wild stock. While some fraction of naturally produced fish may also be
infected with the same endemic pathogens, there may be times or circumstances when highly infected hatchery fish will be in close proximity with
wild stocks having lower levels of infection. In such cases, concerns about
disease transmission from hatchery to wild fish have been raised. One
example is the possibility of increased disease transmission during barging
of salmon around dams in the Columbia River, where both wild and
hatchery-reared salmonids are held together in close proximity and in a
relatively stressful environment during collection and transportation (Elliott
et al., 1997).
5.4.3.4. Reservoir for exposure of wild fish at abnormal times Another
way in which a fishery enhancement or conservation hatchery might affect
the health of wild fish is to serve as a long-term reservoir of infection. In this
way, captive stocks that are chronically infected might continually release,
albeit at low levels, pathogens that could initiate infections among wild fish
146
Kerry A. Naish et al.
during life stages in which they are most susceptible or do not normally
encounter the pathogen in nature. For example, in western North America,
IHNV is commonly found in spawning adult sockeye and is spread among
these highly susceptible fish through the water where high levels of virus
have been detected (Mulcahy et al., 1983). Out-migrating fry are also highly
susceptible to infection with IHNV, but by the time of fry emergence, adult
carcasses have largely been removed from the system and the infection
pressure on fry is low. A hatchery that provided a significant source of
IHNV to the watershed at these times could have an adverse effect on this
highly susceptible life stage in a manner not typical in nature.
5.4.3.5. Genetic effects of hatchery releases on disease resistance of wild
stocks There are several methods by which hatchery operations could
affect the innate disease resistance of wild stocks, including the stock or
strain chosen for rearing in the hatchery. While less common at hatcheries
using local stocks and exercising care to prevent inbreeding, stocks of
hatchery fish having lower resistance to endemic pathogens could spread
less favourable alleles at genes involved in resistance following interbreeding
with wild fish (Currens et al., 1997; Lawlor and Hutchings, 2004). This
effect might be more likely for hatchery stocks having relatively modest
differences in susceptibility compared to wild stocks because hatchery
programmes choosing to rear and release stocks with significantly lower
disease resistance than the wild stock have experienced very poor returns
when such hatchery stocks undergo intense negative selection by endemic
pathogens such as Ceratomyxa shasta (Bartholomew, 1998).
There is an increased effort to determine the genetic basis of disease
resistance in fish. As an example, some alleles have been identified that are
associated with increased resistance to IHNV, while others are associated
with increased susceptibility (Miller et al., 2004). Thus, even if local stocks
are used, it is possible that hatcheries with highly effective disease control
methods for endemic pathogens (e.g., a pathogen-free water supply)
may provide a form of relaxed selection, leading to a greater frequency of
alleles associated with susceptibility among the population of hatchery
fish. If large numbers of these fish are released and do not encounter
sufficient levels of infection in the wild, they can be expected to survive
and return. If these hatchery fish are allowed to spawn with the wild stock,
this relaxed selection might, over time, lower the overall resistance of the
population.
While the genetic diversity of populations helps ensure survivors, hatchery
diseases can exert intense selection. Some hatchery stocks that were founded
from a natural population have been shown to have significantly greater
resistance after a few generations of selection by disease (e.g., Chinook salmon
in the Great Lakes). Such strong selection by one pathogen may be accompanied by a loss of resistance to a second pathogen (Hard et al., 1992).
Evaluation of the Effect of Hatcheries on Wild Salmon
147
5.4.3.6. Release of unexposed stocks from hatcheries Hatcheries with
effective disease control programmes and a source of pathogen-free water
are able to rear and to release large numbers of uninfected fish. While this is
generally assumed to be highly desirable, these unexposed fish may remain
susceptible and could become infected with certain endemic pathogens
following release. If large numbers of such fish suffer a significant disease
outbreak while co-habiting with wild stocks (e.g., during barging operations), they could generate sufficient infection pressure to produce an added
risk to the unexposed portion of the wild stock; although the magnitude of
this risk is unknown. Such fish, infected later than their wild cohort, could
also serve as carriers during in-river or ocean migration to infect portions of
the wild stocks in areas where the disease is not endemic or at times at which
it does not normally occur.
5.4.3.7. Introduction of pollutants or stressors that alter disease
ecology A final method by which hatcheries could increase disease risk
to wild stocks is by altering the ecology of a watershed. Naturally, this
would be most likely for large hatcheries on small watersheds (Tervet,
1981). Effects could range from changes in stream temperature by large
inputs of hatchery water, or phosphorous or organic matter that can increase
algal growth or lower dissolved oxygen levels. Such stressors could be
expected to affect the host–pathogen relationship for endemic diseases
among wild fish.
Naturally, different types of hatcheries will show differing levels of these
effects. In fact, each situation is probably unique. Compared with commercial aquaculture, conservation hatcheries can be expected to have significantly fewer of the most serious risks because they typically do not transport
fish from outside the watershed and because they rear species, stocks and life
stages that are usually derived from local, wild stocks. Nevertheless,
additional research to assess the magnitude of these risks is needed.
5.4.4. Approaches to reduce effects of disease in hatcheries
Unlike their wild counterparts, there are many approaches that can be used
to control the risk or reduce the severity of infectious diseases among
hatchery fish. The choice, however, will depend to a significant degree
on the type of facility involved. For example, disease control strategies that
substantially increase overall costs tend to find few applications in commercial aquaculture but may be very appropriate for conservation hatcheries
attempting to help recover threatened wild stocks. Hatcheries involved in
recovery of local stocks are usually not involved with the movement of
fish from distant watersheds and the associated risk of the introduction
of exotic pathogens or new strains of existing pathogens. For these facilities,
good fish health practises include good sanitation, sound nutrition,
regular health examinations and disease monitoring (American Fisheries
148
Kerry A. Naish et al.
Society, 2004; Winton, 2001). Further benefits can be achieved through
reduction of stress by controlling fish density, water flow and temperature
(Wedemeyer, 1998), and, when necessary, by careful application of drugs
and chemicals (Stoskopf, 1993), vaccines (Leong and Fryer, 1993;
Newman, 1993) or immunomodulators (Iwama and Nakanishi, 1996).
Whenever possible, improvements to hatchery facilities should be a high
priority, including using a pathogen-free water supply or disinfecting of
incoming water (e.g., with ozone or ultraviolet light) and effluent treatment
in some cases.
5.4.5. Application of risk assessment tools for disease
management and control
In recent years, a topic of increasing interest is the application of the tools of
risk assessment to the management and control of aquatic animal diseases
(Bartholomew et al., 2005; Office International des Epizooties, 2003).
In addition to assessing the risk of the introduction, establishment and
likelihood of adverse effects from the spread of a disease into a new
geographic area via the movement of fish, the principles and methods of
risk assessment can also be applied to help analyse ways to reduce disease
risks for wild fish. While the reduction of disease in hatchery fish can lead to
lowered risk for wild fish, the idea is not to simply compare the risks of
hatcheries versus natural rearing, but to assess the various types and levels of
risk posed by different strategies and to identify factors under management
control that can affect risk.
5.4.6. Future work and knowledge gaps
As can be seen, there is considerable uncertainty surrounding many aspects
of the disease risk posed by hatchery operations to wild stocks. Additional
research or effort is needed in the following areas:
1. Improved methods for the detection of important salmonid pathogens
(including non-lethal sampling techniques) and increased disease surveys
of wild fish stocks to gain a better understanding of the distribution and
level of these pathogens in nature. Additional work is also needed to
validate these standard methods to ensure uniformity.
2. Studies to determine the host specificity and virulence of various strains
of important viral, bacterial, protozoan and fungal pathogens affecting
both wild and cultured salmonids.
3. Research to better understand the genetic basis of host resistance among
salmonids and to map these traits on the salmonid genome in order to
identify the genes involved in susceptibility and resistance. Genetic
tools are needed to assess the levels of diversity required to maintain
healthy populations, the heritability of resistance to infectious disease
of salmonids and the genetic changes associated with the development of
Evaluation of the Effect of Hatcheries on Wild Salmon
4.
5.
6.
7.
8.
9.
149
resistance, domestication and interbreeding between hatchery and
wild fish.
New information on the nature of the innate and adaptive immune
systems of salmonids, including development of novel tools and assays
to assess the factors that control the susceptibility of various species and
strains of salmonids to various classes of pathogens.
Research to develop new vaccines to protect fish in hatcheries. This
includes new-generation vaccines (e.g., DNA vaccines) and novel delivery methods.
Improved knowledge of the environmental factors that affect the ecology of infectious diseases of wild salmonids.
Improved and standardized legislation to prevent introduction, movement or spread of exotic pathogens and strains of endemic pathogens
between watersheds.
Information on the risk to wild fish from the various types and levels of
pathogens released from hatcheries. This includes studies on the pathogen shedding rate from infected fish, the environmental stability of the
agent, effective dose/infection pressure that occurs in the wild and the
transmission efficiency between fish in the wild.
Develop methods of risk reduction for various modifications in facilities
or operations. These include effluent treatment, vaccination, disinfection, disease management and stress reduction. Apply risk analysis
approaches to the introduction of exotic pathogens. Analyse and compare pathways and risks from aquaculture, ballast water, anglers, birds and
other factors.
5.4.7. Conclusions
Infectious disease is an important component of the environment that affects
both wild and cultured salmonids. Infections of salmonids may occur in
watersheds, estuaries and the open ocean and, where stocks or species of
wild and hatchery-reared salmonids overlap, many of the same pathogens
will be shared by both. While hatchery operations can have impacts on the
level of disease in wild fish that range from devastating (e.g., introduction of
exotic pathogens) to inconsequential, the origin of infectious disease in
hatcheries is nearly always from the aquatic environment itself or from
reservoirs of infection that are maintained among free-ranging wild stocks.
Additionally, the application of sound hatchery management practises and
application of effective disease control strategies can do much to reduce the
disease risk to wild stocks. This is especially true for conservation hatcheries
where threats from introduction of exotic pathogens or different levels of
disease resistance are lessened by the use of local stocks. Nevertheless,
additional research is needed to provide information to better understand
and quantify the risks to wild fish from infectious disease.
150
Kerry A. Naish et al.
6. Economic Perspectives on
Hatchery Programmes
Salmon enhancement efforts programmes absorb large amounts of
economic resources, and they often claim to provide substantial benefits
in terms of commercial fisheries harvest, recreational fishing or conservation
benefits. Because hatcheries constitute but one means of conserving wild
broodstocks, or enhancing fisheries, economic assessments typically focus
on estimating their costs and benefits, or their cost-effectiveness. The result
can help salmon enhancement planners to select projects that achieve
substantial results at reasonable cost. Successful economic assessments
require comprehensive information on programme costs, reliable and quantitative measures of outcomes achieved and a means of measuring the economic value of outcomes. The assembling of reliable and adequate information
covering all programme dimensions is relatively rare. Consequently, this
section cannot provide a comprehensive economic summary of worldwide
salmon hatchery programmes. It will lay out the basic conceptual framework
for an economic assessment, summarize a handful of economic studies and
provide some insight into complicating factors that make conclusive economic
assessments difficult in practise.
Each of the salmon hatchery types that have been described in Section 1
has a characteristic operational pattern, incurs costs associated with operations and, frequently, a blend of objectives. Figure 2.24 provides a useful
scheme for evaluating hatcheries by type of operation, based on broodstock
origin (hatchery origin, wild origin or permanent captive broodstock),
release location, release objectives and location where the adult hatchery
fish return to a fishery or spawning site, and provides a basis for economic
analyses of the operations. Project outcomes can be measured in physical
Broodstock origin
Release location
Release objectives Adult return location
Fishery
Remote
enhancement
acclimation site
1. Hatchery origin
2. Wild origin
River
Fishery
enhancement
River or ocean
harvest
Conservation
River spawning
grounds
Conservation
Hatchery
3. Captive broodstock
Hatchery
pond
Acclimation site
harvest
Figure 2.24 Depiction of various origins, handling routes and destinations for
hatchery-spawned salmon.
Evaluation of the Effect of Hatcheries on Wild Salmon
151
terms (juveniles released, adult returns or size of broodstock preserved) as
well as in economic terms (increased value of fisheries or benefits to the
public). When enhancement projects are motivated by legal requirements
or multi-purpose objectives that are not easily assigned an economic value,
outcomes are expressed only in physical or biological units (e.g., number of
returning spawners or increase in survival at some life stage). In such cases, it
is useful to pursue cost-effectiveness analysis (CEA) of alternative projects or
facility designs (see IEAB, 2002). The CEA reveals which projects provide
the most performance for the cost incurred. When both project costs and
economic benefit estimates are available, enhancement projects can be
appraised via benefit-cost analysis (BCA). The BCA approach is most
applicable for programmes that aim to augment or enhance fisheries or to
establish or protect a salmon run with known value to people. To better
inform policy makers, both BCA and CEA may need to be augmented by
evaluation of other consequences such as regional employment or income
impacts to account for broader socio-economic consequences (Fraser and
Friedlander, 1980).
We could find few examples of economic analyses in Europe, and
therefore we focus this discussion on SEPs in western North America.
These programmes include private and public hatcheries that release juvenile fish for both enhancement of fisheries and conservation of wild stocks.
For example, Wahle et al. (1974) and Wahle and Vreeland (1978) evaluated
the Columbia River enhancement programmes, Boyce et al. (1993) assessed
Alaska’s salmon hatchery programme based on the increased economic
value of the fisheries and Pearse (1994) evaluated costs and benefits of
diverse projects in the salmon stock enhancement programme in British
Columbia. In this chapter, we review some of these studies and comment
on the use of economic evaluation of the hatchery and other enhancement
facilities. Economic assessment methods can be applied to any form of
salmon stock enhancement, including riparian habitat restoration and fish
passage improvements (Paulsen and Wernstedt, 1995; Willis et al., 1998;
Wu et al., 2000). Generally, an economic assessment is contingent on, and
may be severely limited by, the availability of quantitative predictions of key
biological outcomes of enhancement projects. Where the effects of
enhancement projects on salmon populations cannot be quantified, an
economic assessment may be premature.
6.1. Measuring costs, effectiveness and benefits
6.1.1. Costs
Project costs include both capital costs and annual operating costs. The
capital costs comprise all initial and periodic investment expenses associated
with planning, design, construction, equipment installation and replacement and land acquisition for the facilities. Operating costs involve salary
152
Kerry A. Naish et al.
and wages, personnel benefits, transportation, utilities and routine maintenance associated with the activities of trapping and holding adult spawners,
hatching eggs, rearing fry and juveniles, maintaining water supply and
quality, and research and monitoring. The costs are typically revealed in
budget documents of the responsible agencies, but annual budgets often do
not place expenditures in the accounting categories needed for economic
analysis. Agencies rarely maintain accurate capital investment and depreciation accounts, and the lack of this information makes project assessment
difficult. Also, administrative, monitoring and research costs are often
assigned to separate offices rather than to individual hatchery operations.
To provide an accurate synopsis of individual hatchery programmes, the
costs that are shared among a number of hatcheries (often administration
and research) need to be allocated on some basis to individual projects.
To properly account for capital costs, both start-up costs and periodic
maintenance or replacement costs of a facility need to be annualized over
appropriate time spans. This is typically done by treating the capital cost as
the principle on a loan, and calculating the annualized capital cost as
equivalent to the payment required to pay off (or amortize) the loan over
a specified period. For example, the capital cost could be annualized over a
30-year period with an annual interest rate of 5%. Annualizing the capital
cost facilitates comparison of annual costs (operating and capital costs) with
the value of the hatchery’s contribution to the fishery.
The full costs may be assembled into a summary table, displayed in
accounting categories (such as labour, materials, transportation, utilities,
feed, maintenance, capital expense). Where costs for a large number of
similar projects are available, the results may be a statistically derived
functional relationship between total or component project costs and fish
release numbers, fish species, hatchery type, location and other variables that
influence costs (Loomis and Fix, 1999). The accounting display provides a
detailed snapshot of a particular project (or class of projects), while the
functional cost equation provides a means to forecast how costs vary with
hatchery size or design.
6.1.2. Effectiveness
Effectiveness should be measured to reflect the main purpose, or purposes,
of the enhancement project. A fishery enhancement hatchery could be
judged by the magnitude of the run size increase or harvest contribution.
A conservation hatchery might be judged by the magnitude of increase in a
wild salmon population. To be a useful planning device, CEA must incorporate information from a range of alternative enhancement projects. If a
fixed budget for enhancement were available, a cost-effective group of
projects would be those that achieve the most effectiveness for the budget.
On the other hand, if a fixed enhancement objective were firmly established, the CEA would assist in selecting a mix of projects that achieves that
Evaluation of the Effect of Hatcheries on Wild Salmon
153
objective at lowest cost—that is, a programme that meets the objective
cost-effectively.
Measuring the contribution of a hatchery to the size of salmon populations is a difficult research task, involving tag release and recapture data,
smolt-to-adult survival estimates, harvest rate estimates (often for several
geographically separate and mixed-stock fisheries) and hatchery return and
straying estimates. Further, where a hatchery brood interacts with wild
stocks or other hatchery stocks (via competition for food and space, disease
transmission or interbreeding and genetic modification), the contribution of
the hatchery to run size and harvests must be adjusted to account for possible
negative changes in the other stocks. When a hatchery depends on
continued capture of natural spawners for broodstock, the net increase in
run size attributed to the hatchery should reflect the hatchery-origin run
minus any reduction in natural spawning run.
6.1.3. Benefit-cost analysis
BCA tackles the more complex task of estimating economic value of the
project outcomes. For production hatcheries aimed at commercial fishing
(e.g., the Alaska SEP), the benefit is simply the net economic value of
increased fish harvests (i.e., sales value minus increased harvesting cost).
Where recreational fisheries take some or all of the fish, recreation benefits
can be assessed using recreational demand models based on the travel cost
method (see Brown et al., 1983) or one of the more sophisticated recreational choice models (see Berman et al., 1997). Subsistence fishing, especially treaty-obligated fishing by Native peoples, presents a more difficult
conceptual task that has, frankly, not been addressed adequately by salmon
economics research. Further, people who appreciate the existence or preservation of unique salmon runs hold non-use values, which do not depend
on harvesting fish. Non-use values can be assessed using actual or hypothetical payments in response to questions posed in surveys, using the contingent valuation method (see Bell et al., 2003). In some circumstances, salmon
enhancement projects may sometimes produce all four types of benefits—
commercial, recreational, subsistence and non-use value—making the
benefits assessment a challenge. Further, conservation hatcheries typically
provide benefits through an increase in wild stocks, and all the types of
economic benefits would be applicable to these as well.
6.2. Cost-effectiveness of hatchery programmes
Two recent attempts at CEA serve to illustrate the method and the complications associated with the method. The Northwest Power Planning
Council’s Independent Economic Analysis Board (IEAB, 2002) assessed
cost-effectiveness of six hatchery programmes and one acclimation and
release programme in the Columbia River Basin. The IEAB’s objective
154
Kerry A. Naish et al.
was to provide advice regarding expenditures in the Council’s Fish and
Wildlife Programme, which spends roughly $40 M a year on salmon
enhancement projects. The initial phase of that work studied five ongoing
production hatcheries, including a lower river fall Chinook hatchery; an
upper Snake River summer Chinook hatchery in Idaho; a multi-hatchery
complex in the upper Columbia at Leavenworth, Washington; a steelhead
hatchery in the mid-Columbia and a fall Chinook hatchery operated by a
public utility district. The resulting short report and technical appendix
(IEAB, 2002) were reviewed by the agencies and other technical staff.
At about the same time, the Audits Division of Oregon’s Secretary of
State examined Oregon’s hatchery programmes for cost-effectiveness.
This study focused on 12 coho and Chinook hatcheries in western Oregon
(Oregon Secretary of State, 2002). In both of these reports, each project was
characterized by total releases, estimated smolt-to-adult survival, estimated
total catch (all fisheries combined) and annualized costs. The costs are summarized in three forms: cost per fish (or pound) released, cost per adult survival
and cost per adult caught. As with the BCAs discussed below, neither of these
CEAs incorporates the effects of hatchery fish impacts on wild stocks.
An example of this type of economic analysis pertains to a salmon
hatchery in McCall, Idaho, operated for fishery enhancement by the US
Federal government. The hatchery rears summer Chinook in a facility with
2, large outdoor ponds, 14 indoor rearing tanks and incubation facilities.
Initial construction costs in 1981 for the facilities were $5,453,000. Updating this fig. to 2000 by applying the US Gross National Product (GNP)
price deflator yields a capital cost of $10,755,424. Annualizing this cost over
50 years at a 3% interest rate generates an estimated annual capital cost of
$418,015. The costs and the production of smolts, the smolt-to-adult
survival rates (SARs) and harvest rates were obtained for a 13-year period,
brood years 1984–1997. The average cost per smolt released for that period
was $1.09, the average cost per adult fish returning (to the fishery or the
hatchery) was $271.80 and the cost per fish caught in the fishery was
$1051.01. This cost occurred during a period of time when the salmon
were experiencing relatively low ocean survival rates. The cost might be
significantly lower during other periods of time.
The IEAB research results found that the costs per smolt (measured in
2001) varied from $0.08 for fall Chinook (sub-yearling smolts) released at a
mid-Columbia public utility district hatchery to $2.60 for Chinook released
as yearling smolts from the Nez Perce tribal hatchery. Based on both the
data and discussions with hatchery managers, the IEAB found that the cost
of producing sub-yearling smolts (fall Chinook) was substantially lower than
cost per yearling smolts (spring and summer Chinook, steelhead, coho) for
the obvious reason that the yearlings are reared and fed for a longer period.
The cost per adult survivor ranged from a low of $12 for mid-Columbia fall
Chinook to $3707 for spring and fall Chinook from the Nez Perce tribal
Evaluation of the Effect of Hatcheries on Wild Salmon
155
hatchery. Again, the two lowest cost per adult estimates were for fall
Chinook hatcheries. Finally, using estimated contribution rates to fisheries,
the cost per adult fish caught ranged from a low of $23 for Priest Rapid fall
Chinook to $68,031 for spring Chinook from an upper Columbia river
hatchery on the Entiat River. The wide range of cost-effectiveness estimates
suggest that reallocation of funds to better-performing facilities would offer
an opportunity to achieve more harvest enhancement per dollar spent. On
the other hand, the fig. for cost per harvest were the least reliable of the costeffectiveness measures because the sampling of tags from in-river fisheries
was of unknown accuracy. Further, the Nez Perce tribal hatchery has just
begun production and has focused on supplementation (conservation) of
local runs, not harvest per se. The objective of that programme is not simply
to produce some fish for catch somewhere but to produce a particular substock returning to a particular tributary. Hence, comparability across hatcheries is not as transparent as the numbers might suggest.
The Oregon cost-effectiveness audit found cost per pound of fish
released to vary from $4.08 to $9.09 (measured in 2001); cost per adult
survivor (hatchery return plus catch) ranged from $14 (Salmon River fall
Chinook) to $530 (coastal coho at Bandon, Oregon) and cost per adult
caught ranged from $27 (Salmon River fall Chinook) to $1442 (coastal
coho at the Trask River). As with the Columbia Basin hatcheries, yearling
releases are more expensive than sub-yearling releases, and the cost per fish
caught depends strongly on both SAR and harvest rate. Harvest rates in
Oregon salmon fisheries are geared to protect the weak stocks, and they
have been tightly regulated in recent years to protect coastal coho and
Columbia River Chinook stocks that are listed as threatened or endangered
under the Endangered Species Act. Hence, a hatchery with a reasonably low
cost per adult survivor may have a high cost per adult caught simply because
their fish mix with protected wild fish and harvest rates are kept low.
To change the locations or operations of hatcheries to improve the future
harvest rate (and to lower the cost per catch) would require adapting to
future fishing regulations that will respond to perceived depletion of various
salmon runs with shorter fishing seasons and lower catches.
One way to lower the cost per fish caught would be to move the smolts
from the hatchery location to an acclimation site away from protected
stocks, with the intent of getting the fish to return to a site where they
can be fished at a high rate. The Clatsop County Economic Development
Council in Oregon funded a project of this sort starting in 1977 to enhance
the lower Columbia River gillnet fishery. In recent years, this project has
acclimated salmon in net pens in Young’s Bay, west of Astoria (IEAB,
2002). The programme includes fall and spring Chinook and coho from
various sources and fish are released at sites in the Columbia River estuary.
The cost per fish caught from the programme range from $14 for coho to
$233 per spring Chinook.
156
Kerry A. Naish et al.
6.3. BCA of hatchery programmes
An early BCA of salmon hatcheries was conducted by the US National
Marine Fisheries Service on the Columbia River Development Programme
(Wahle and Vreeland, 1978; Wahle et al., 1974). These studies were tied to
very ambitious mark-recapture research efforts that provided estimates of
hatchery contributions from 13 Chinook and 20 coho hatcheries to ocean
and river fisheries. The ocean fisheries ranged from southeast Alaska to
California, and the river fisheries include commercial gillnet, sport fishing
and Native peoples’ fishing. Total estimated contributions to coho harvests
were 1.13 and 1.05 M fish for the 1965 and 1966 brood years, and contributions to Chinook harvests ranged from 11,000 to 602,000 fish for
brood years 1963 through 1966. The capital costs (costs associated with
design and construction of the facilities) were annualized over 30 years at a
3.5% interest rate. Operating costs were compiled for the spawning, hatching and rearing for two brood years of coho salmon (1965 and 1966) and
four brood years for Chinook salmon (1963–1966). The commercial harvests were valued by multiplying the number of fish contributed to harvest,
multiplied by the estimated weight per fish and the current ex-vessel price.
Recreational harvests of coho were valued by dividing the economic value
per angler day of $20 (estimated by Brown et al., 1983) by catch per day and
then multiplying by number of fish caught by sport fisheries. Recreational
harvests of fall Chinook were valued at a straight $18.35 per fish.
Overall, Wahle et al. (1974) estimated economic benefits for the coho
fisheries at $9.07 and $8.51 M for the two brood years. When compared to
the coho hatchery costs of $1.29 and $1.23 M, the benefit-cost ratios were
7.4 and 6.6 for the 1965 and 1966 brood years, respectively. For the fall
Chinook hatchery programme, annual estimated benefits ranged from $1.3
to $5.2 M, while the annual hatchery costs fell in a narrow range of
$659,000–$748,000. Benefit-cost ratios for the fall Chinook hatcheries
ranged from 2.0 for the 1962 brood year to 7.2 for the 1963 brood year,
and had a 4-year average of 4.2. Of equal interest is the estimated variation
in benefit-cost ratios for individual hatcheries that ranged from 11.2 for the
Spring Creek hatchery to 0.3 for the Elokomin hatchery (1961 brood year
only). In principle, reliable estimates of benefit-cost ratios for individual
hatcheries, or even individual batches of fish within a hatchery, could be
used to score and rank the underlying rearing regimes, locations and species.
This information would feed into subsequent decisions regarding design
and allocation of funds within the hatchery programme.
A drawback of these Columbia River hatchery studies is the use of exvessel price for economic value per pound of harvest. The logic for this
procedure, outlined in Wahle et al. (1974), is that because the hatcheries are
augmenting the harvest of an open access and economically inefficient
fishery, the additional catch will add little or nothing to the harvesting cost.
Evaluation of the Effect of Hatcheries on Wild Salmon
157
Normally, economists would subtract additional harvesting cost from the
additional catch, assuming that an increase in gross revenue from the fishery
(especially in an open access fishery like the salmon fisheries of the 1960s)
would attract additional fishing effort, thus increasing costs. Had this been
done for these hatchery projects, the benefit-cost ratios would have been
substantially lower.
The State of Alaska began a major SEP in the early 1970s, encompassing
both state-run projects, under the new Division of Fisheries Rehabilitation
Enhancement and Development (FRED) of Alaska Department of Fish and
Game, and PNP hatcheries owned and operated by regional aquaculture
associations (see also Section 5.3). During 1972–1992, the State appropriated $210.3 M to the new FRED division, and total expenditure by
PNP hatcheries was just over $200 M (Boyce et al., 1993). About 42% of
PNP funds came from the State and the rest came from landings fees and
cost-recovery fisheries. Cost-recovery fisheries, which are organized by the
PNP hatchery associations, involve contract harvest for the association with
sales revenue used to cover the costs of operating the hatcheries. In 1992,
the Alaska State Senate sponsored the research reported by Boyce et al.
(1993) in order to evaluate seven alternative actions being considered,
including eliminating the pink and sockeye salmon hatcheries, and increasing or decreasing the two species’ production levels by 15%.
The BCA analysis was approached using an Alaska accounting stance
(Boyce et al., 1993), that is, only costs and benefits accruing to Alaska
fishermen, processors and agencies were counted. The authors used a
biological model (Collie, 1993) to project catches by species and region
over a 30-year time period. The prices for salmon under each alternative
were computed from an international salmon market model (Herrman,
1993), and the benefits to the fishing industry were defined as the total
revenue from sales of fish minus the costs of harvesting the fish. The net
economic benefits to the State equal the benefits to fishers minus the costs of
the enhancement programme. With these assumptions and estimates they
estimated the following 30-year, statewide totals for the existing system
(Alternative 1): total catch (includes all wild and hatchery fish), 353 million
kilograms; gross revenue, $557 M; benefit to industry, $222 M; hatchery
costs, $23.4 M and net benefits of $199 M to Alaska.
The main results were associated with Alternatives 2 and 3, which
eliminated the pink and sockeye salmon enhancement facilities, respectively. For Alternative 2 (eliminating pink salmon hatcheries), gross revenues dropped by $5.5 M, industry benefits increased by $9.7 M, hatchery
costs dropped by $6.4 M and Alaska net benefits increased by $16.1 M.
The implied negative net benefit from pink salmon hatcheries occurred
outside of PWS, where major pink salmon hatcheries generate benefits for
the local fishery. Alternative 3 (eliminating sockeye hatcheries) reduced gross
revenues by $8.75 M, increased industry benefits by $ 8 M, decreased hatchery
158
Kerry A. Naish et al.
costs by $4.1 M and caused an increase in state benefits of $12 M. Overall,
this analysis suggested that Alaska would be better off economically without
the pink salmon and sockeye salmon enhancement facilities. The report,
however, warned that no recreational or subsistence fishing benefits were
evaluated and that there may be some un-analysed strategic value to maintaining hatchery production to fend off the expanding salmon farming
business in Europe and South America.
Canada’s SEP in British Columbia was launched in 1977 with the
objective of doubling salmon catches on Canada’s Pacific coast through
construction of hatcheries, spawning channels and other works. Over
17 years they spent $526 M, built more than 300 facilities and expanded
the fish harvests by roughly 14,000 metric tons, or 13% of the annual salmon
catch. This was somewhat of a disappointment, given the ambitious goal of
the SEP. A very broad and creative BCA was performed by Peter Pearse for
the DFO’s Internal Audit and Evaluation Branch (Pearse, 1994). The Pearse
report followed a series of previous evaluations by a Royal Commission,
a Ministerial Task Force and three previous audits.
Pearse (1994) estimated the catch contributions, gross and net harvest
values (for commercial, Native and recreational fishing) and lifetime costs
(construction and operating) for the SEP facilities. The catch contribution
due to SEP was estimated at 17,361 metric tons (50% chum, 19% sockeye,
10% pink, 12% coho and 9% Chinook). These contributions were not
adjusted to account for interactions between hatchery and wild stocks
because Pearse (1994) was unsuccessful at getting a consensus expert judgement on the extent of interactions. For the commercial fishery, net benefits
were the sum of three pieces: vessel owner benefits (50% of gross revenue
minus crew share), crew benefits (crew share minus estimated labour cost,
valued at mean wage) and cannery benefits (50% of wholesale value minus
fish costs minus variable costs of canning operation). The Native fishery
benefits were valued at the 1993 ex-vessel price, with no deductions for
harvest costs. The recreational fishery was valued by multiplying increased
coho catch by $14 and increased Chinook catch by $54. Finally, Pearse
(1994) used an 8% interest rate to value past costs and benefits as of 1993 and
to discount future costs and benefits (out to 2017) back to 1993. Overall, the
estimated present value of SEP in 1993 costs ($1.51 billion) exceeded the
estimated net benefits ($919.9 M) by $592 M, leaving the programme with
a benefit-cost ratio of 0.6. The benefit-cost ratios varied widely among
enhancement projects; the spawning channels had a 2.2 benefit-cost and the
lake fertilization projects a 1.3 benefit-cost ratio.
Pearse (1994) also provided a reasonable approach to additional decision
making by dropping the past capital costs (the ‘sunk’ costs) and the benefits
occurring before 1993. The result was an evaluation of the project from
1993 on, which is an important perspective for decision makers at that point
in time. For this short-term decision framework, the benefit-cost ratio for
Evaluation of the Effect of Hatcheries on Wild Salmon
159
the overall SEP programme rose to 1.6, with a net benefit of $165.3 M.
In other words, given that the costs of investing in the facilities are irretrievable, there are net benefits to continuing the programme over the
expected lifespan of the facilities. If we could go back in time to the
programme start-up date, possessing the economic assessment produced in
1993, we might decide not to start the programme at all. Further, because
the report contains specific estimated benefits for each major project, it is
useful information for planning and adapting the content of the SEP.
6.4. Complicating factors
Several conditions add to the complexity and unreliability of SEP evaluations in practise. First and foremost, measuring programme effectiveness is
absolutely reliant on biological and ecological modelling and analysis. As is
evident from other sections of this chapter, the full impact of hatchery
releases on aggregate run size depends on interactions among stocks, and
particularly the effects of hatchery smolts on wild smolts and the effect of
straying hatchery-origin spawners on natural spawning populations. These
effects are often only roughly quantifiable and frequently controversial
among experts. The economic studies reviewed above basically assume
that the hatchery run represents a net increase in the volume of salmon
returns from the ocean, despite evidence that this is not true in some cases
(Hilborn and Eggers, 2001). A second complication is that large volumes of
returns to hatcheries can affect market prices for salmon, at least within the
region impacted and during the harvest season (Herrman, 1993). When the
market price varies with the hatchery output, the economic benefits to
consumers should be measured as the increased consumer surplus (i.e., the
increased area under the estimated demand curve as the price falls). This
measure requires additional research on the market demand for the salmon
products.
A third complication is that enhancement projects can have a range of
complex objectives that defy even concerted attempts at quantification.
Experimental and research hatcheries focused on supplementation of
endangered populations that may contribute to the long-term survival of
listed species. While economists have estimated non-use values for salmon
protection and restoration (Bell et al., 2003; Loomis, 1996), it remains
difficult to attribute specific values to specific projects that protect narrow
sub-populations with known levels of risk. Further, many of the hatcheries
in the Columbia River Basin were authorized in conjunction with multipurpose river development projects (hydropower dams and irrigation projects).
The construction and operation of the hatcheries represent a portion of
the multi-purpose project objective to preserve some specific salmon or
steelhead runs in the affected tributaries. Some observers note that the
associated costs are attributable to the other project objectives, and that
160
Kerry A. Naish et al.
the trade-offs made in designing and authorizing the projects should not be
recalculated later based solely on the performance of the hatcheries. There
is necessarily a political balancing and negotiating aspect to these decisions
that is not transparent in the economic analysis. Hence, the most that can
be claimed for the economic assessment is that it is useful information for
decision making when trade-offs among quantifiable objectives are being
weighed.
6.5. Conclusions
Given the size and costs of public salmon hatchery-release programmes,
careful and extensive benefit-cost and cost-effectiveness analyses would
appear to have a useful role in salmon enhancement project selection and
programme design. US Federal and State laws require that new programmes
be evaluated for both environmental and economic consequences. However, salmon hatchery programmes have generally not been subjected to
standard benefit-cost or cost-effectiveness analyses. Salmon hatcheries in the
Pacific Northwest, British Columbia and Alaska are justified on other
diverse grounds such as: (1) to mitigate for loss of spawning/rearing habitat,
(2) to meet requirements of treaties, (3) to compensate for destruction of a
natural salmon run via dam construction, (4) to augment commercial or
recreational fishing and (5) to support threatened or endangered stocks.
Only the short-lived commercial ‘salmon ranching’ operations in Oregon
had the simple economic objective of producing harvestable fish that could
sell for more than the cost of production. Hence, it is not surprising that
the standard economic project evaluation techniques are rarely aimed at
public salmon hatchery programmes. Nevertheless, it is also clear that SEP
decisions strongly influence the magnitude of economic costs and benefits
and that these decisions need not be made in ignorance of the economic
consequences. The benefit-cost and cost-effectiveness analyses reviewed in
this chapter show that a moderate research effort, using information normally collected for hatchery fish monitoring and budgetary purposes, can
provide a reasonably constructed economic assessment of SEPs.
7. Discussion
This chapter joins a growing number of papers that attempt to collate
information on enhancement activities (National Research Council, 1996)
and to evaluate the available evidence for the biological effects of such
activities. We have presented the historical context and political underpinnings of hatchery programmes, reviewed the current level of releases
from hatchery facilities in the North Pacific and Atlantic, discussed possible
Evaluation of the Effect of Hatcheries on Wild Salmon
161
outcomes of interactions between hatchery and wild fish and have evaluated
economic issues associated with the design and running of hatchery
programmes.
Broadly, our chapter points to three major issues. First, decisions to
initiate or sustain hatchery programmes are typically motivated by political
objectives, which are rooted in historical contexts. These decisions are
infrequently revisited and are rarely driven by biological or ecological
reasoning. A holistic view of the effects of the production of salmonids on
the ecosystem has not, in most cases, been taken into consideration. Rather,
the effects have tended to be viewed in isolation. Second, there remains a
dearth of information on the consequences of interactions between wild and
hatchery fish despite the fact that hatchery programmes have been operating
for since the nineteenth century. Third, the outcomes of hatchery releases
and management steps are not fully understood because robust, systematic
and coordinated scientific assessments are rare.
Such broad statements, of course, are only constructive when key gaps in
the state of knowledge are identified, and placed in context of the objectives
of conservation and fishery-enhancement hatchery programmes. We attempt
this task here, while acknowledging that reviewers examining the same data
sets often reach different conclusions (e.g., Brannon et al., 2004a; Myers et al.,
2004). Nevertheless, recent political events have motivated individual
scientists and advisory groups to formulate guidelines for the management of
these types of hatcheries (Mobrand et al., 2005; Waples and Drake, 2005), and
we examine below some of these guidelines in the context of our assessment of
the major knowledge gaps in the field.
7.1. Release objectives and release sizes
We initiated the review by providing a classification system for enhancement activities in recognition of the fact that differing objectives for hatchery programmes would lead to a range of biological outcomes. We then
pointed out that these objectives have rarely been identified and subsequently enacted upon (Section 4). Without these defining objectives, individual programmes cannot be held accountable if they do not have a clear
set of measurable guidelines. The same issue has been raised by a number of
authors (Waples and Drake, 2005; Waples et al., 2007), and has been
identified as a key guiding principle in formulating recommendations for
hatchery programme reform (Mobrand et al., 2005).
We noted an absence of standardized approaches towards the collection
and archiving of data on hatchery release sizes. This outcome is not
surprising since hatchery activities are defined by political boundaries.
We also noted that data quality varied across the countries we surveyed.
Reporting would be most useful to the scientific community if the release
goals of hatchery programmes were clearly identified and if attempts were
162
Kerry A. Naish et al.
made by the international community to centralize release data in a readily
accessible format. The effects of hatchery programmes likely transcend
watersheds and international boundaries and the development of a clear
understanding of the extent of these effects would be well served by the
collation of large data sets. It is particularly important to identify the relevant
scale at which this data should be collected. Ideally, data sets should be
collated hierarchically; levels of production and survivorship should be
reported at the freshwater, estuarine and oceanic stages. Hierarchical data
sets will be valuable for identifying the relevant scales over which interactions between hatchery and wild fish should be appraised. Finally, we noted
that the effects of hatchery releases on the ecosystem were difficult to
evaluate because of the paucity of data collected at this level.
7.2. Interactions between hatchery and wild fish
The literature on interactions between wild and hatchery fish was examined
by exploring genetic effects, competition, harvest interactions and disease
transmission. We acknowledged that this list was incomplete, but we also
noted that there have been few concerted experimental approaches to
understanding the outcomes of these interactions.
7.2.1. Genetic issues
Most examples of the genetic interactions between hatchery and wild fish
have been retrospective and case specific, and have rarely been defined in
terms of their release goals. While these studies point to a frequent
outcome—that releases are often detrimental or unsuccessful—there is still
a strong need to gain an understanding of the degree of risk posed by
hatchery fish, and whether these risks can be reduced by correct management. Recent experimentation and a change in philosophy towards
solution-based research appear promising. Part of the problem associated
with research in this area is that most experiments require several generations of returning adults, and the resources needed to complete such
experiments have seldom been available.
Relatively new guidelines have been presented by a scientific advisory
group in the Pacific Northwest (Mobrand et al., 2005). One recommendation aimed at reducing genetic impacts is that hatchery broodstock be either
integrated with, or segregated from, wild populations (Mobrand et al.,
2005). This guideline is based on theoretical treatments that examined
changes in fitness traits with varying levels of migration between hatchery
and wild fish (reviewed in Section 5.1; modified from the model proposed
by Ford, 2002), and is aimed at preventing the negative outcomes of
reproduction between wild fish and hatchery fish that have been subject to
domestication selection. The authors also point out that issues such as genetic
drift, inbreeding, changes in effective size (the Ryman–Laikre effect) and
Evaluation of the Effect of Hatcheries on Wild Salmon
163
outbreeding depression might be reduced by correct management of broodstock collection, mating and rearing protocols and individuals released
(Section 5.1; see also Waples and Drake, 2005).
The recommendations of Mobrand et al. (2005) promote active management approaches rather than risk-averse measures, and will likely be
debated in the scientific community over the next few years. For example, it
has been noted that the impact of an integrated release on a fine scale
metapopulation structure is unknown (Utter, 2004), especially if the other
components are demographically unstable. It is also uncertain whether
hatchery fish may be practically segregated from spawning wild fish once
they have been released to the wild, although it has been argued that
domesticated, less fit individuals might pose lower risks (Utter, 2004).
Discussion on new management approaches are likely to be lively in the
coming years, but recommendations such as those proposed by Mobrand
et al. (2005) provide a framework for future experimentation on ‘problem
solving’ approaches, and such research is strongly supported here.
It is quite clear that genetic issues have been placed at the heart of the
‘hatchery-wild’ debate. If one concern has been identified in this chapter, it
is that many hatchery programmes continue to be operated with few
objectives, and with a poor understanding of the magnitude and importance
of the impacts of genetic effects of hatchery releases and the role of this
information in informing remedial actions. The field has been invigorated
by recent hatchery reform initiatives, but management recommendations
that are implemented broadly without an experimental approach and
without identifying long-term goals will continue to perpetuate this
problem, possibly with the negative consequences that have been widely
reported to date.
7.2.2. Competition
We identified two key assumptions that are embedded in the philosophy
underlying hatchery operations. The first assumes that captive rearing is
appropriately directed at the most limiting life history stage. For many
species, this limitation is not at the egg to smolt stage at which most
operations are directed, but during the juvenile rearing period in freshwater
streams or perhaps during their estuarine or early ocean stage. The second
assumption is that competition between hatchery fish and their wild counterparts does not counteract the aims of the hatchery programmes. If, as was
pointed out, the carrying capacity of the environment is limited (and this has
been demonstrated in a broad range of studies in freshwater and a
limited number in estuarine and marine environments), then competitive
interactions between the two components can have negative outcomes.
The nature and level of behavioural interactions between hatchery and
wild fish may vary with the type of hatchery programme. If conservationbased hatcheries are considered first, then the primary aim of recovering a
164
Kerry A. Naish et al.
threatened population would be best served by producing hatchery fish
whose distribution in physical form and life history characters (and, by
implication, in genetic composition) does not differ from that of their
wild counterparts. The simple notion would be that competitive interactions would not favour one component over the other. However, many
programmes have been established to recover weakened populations without considering the underlying causes of the population decline, and specifically, which element of the habitat has proven limiting. Thus, releases of
wild-type hatchery fish under this scenario may not result in the intended
demographic boost if the original limitation is not addressed. Hatchery fish
released for fisheries augmentation may differ from wild fish in a number of
physical and life history traits and may differ in abundance. These differences can have complex outcomes, depending on the extent of spatial and
temporal overlap between hatchery and wild fish.
Practically, authors have suggested that negative behavioural interactions
can be avoided in a number of ways that depend on the nature of the
hatchery programme (Mobrand et al., 2005; Waples and Drake, 2005). If
the aim is to segregate the hatchery fish from the wild, then interactions
during early freshwater stages can be reduced by releasing smolts that
migrate rapidly to sea or by producing larger smolts that utilize different
habitats than the wild fish. Marine carrying capacity should factor into
estimates of release size, although it is acknowledged that calculations
based on this parameter are unlikely to be realistic over the short term and
only relevant to changes in ocean regimes related to large-scale climate
cycles (Mobrand et al., 2005). Competition for spawning habitat may be
eased by siting hatcheries away from spawning grounds and by maximizing
imprinting to the hatchery itself. However, as was pointed out in Section 5.2,
it is unclear to what degree competition and straying by hatchery fish on the
spawning grounds can be alleviated by these measures, and generally, the
locations of hatcheries are largely fixed.
While each of the approaches identified above may provide some solutions for segregated hatcheries, their utility is less clear for integrated
hatcheries geared towards conservation because most of the steps will result
in genetic differentiation between hatchery and wild components.
For example, the release of fish larger than those found in the wild can be
expected to change a suite of life history characters within the run, most
notably age at maturity. The issue here is whether hatcheries can produce
‘wild-like’ fish in numbers that do not exceed the carrying capacity of the
habitat and do not compromise the wild populations.
Our chapter collated a rapidly growing body of literature that points
towards detrimental behavioural interactions between hatchery and wild
fish. More is known about these interactions in freshwater rearing habitats
than in estuarine and marine environments. There is also, however, a
paucity of information on whether risk avoidance measures are effective at
Evaluation of the Effect of Hatcheries on Wild Salmon
165
reducing competition and predation and, as far as we know, little attention
is directed towards carrying capacity when the size of release is considered.
7.2.3. Harvest
The impact of harvest on wild populations becomes important when
fishery-based hatcheries are considered. In theory, increasing the survival
of a specific life history stage will support elevated harvest rates on the
hatchery component of the population. The success of the programme will
rely on the efficient segregation of the hatchery fish from the wild, which in
turn is largely dependent on where and when the fish are harvested.
Although two types of fisheries are recognized—terminal fisheries near
the mouth of a river and interception fisheries in open waters—in practise,
both target mixed stocks (although the former fishery likely comprises fewer
populations than the latter). Commercial demand favours fishing at sea
because flesh quality is higher during this life history phase. Ocean stocks do,
of course, include a mixture of separate spawning populations, and overproduction of hatchery fish can lead to overexploitation of weaker stocks (often
wild fish) within this mixture.
Several solutions to the ‘mixed-stock harvesting problem’ have been
identified. The most controversial would lead to over-exploitation of the
wild stocks and dependence on the hatchery component of the run for the
persistence of the species, or almost as contentious, the termination of all
production hatchery programmes. Most management agencies have instead
relied on reducing exploitation rates to those sustainable by the weaker wild
stocks and on selectively harvesting hatchery fish, which relies on efficient
mass marking.
The success in setting appropriate exploitation rates depends on the
accurate identification of a wild ‘stock’ so that appropriate forecasting and
in-season management can be implemented. In Europe, it is recognized that
the use of genetically isolated units within rivers is impracticable, and
groupings based on populations experiencing similar abundance trends are
being implemented instead in some places (Crozier et al., 2004). Researchers
monitoring mixed populations of Pacific salmon in the high seas often
depend on genetic definitions of stocks (Beacham et al., 2004; Seeb et al.,
2004).
Selective fishing requires that hatchery fish are accurately identified and
that the survival rates of hatchery fish are high prior to harvest. Mass
marking methods have, to a large degree, been successfully implemented
in North America when the marks are clearly visible. However, the use of
approaches such as otolith marking does not permit identification of
hatchery fish until they are dead and, thus, they are of limited utility.
Several data sources point towards mixed success in consistently producing hatchery fish with high survival rates. Shifts in ocean regimes and marine
productivity affect these rates with unintended consequences. If survival
166
Kerry A. Naish et al.
rates are low, then fisheries may inadvertently be redirected towards vulnerable wild stocks. On the other hand, excess fish escaping the fishery in a
productive year can return to the spawning grounds, raising heated debates
about the fate of those individuals (ISAB, 2002). In this case, interest groups
have argued that returning hatchery fish can augment declining wild populations, but this view ignores the negative outcomes that are the subject of
this chapter. Regardless, social dimensions often intervene, and fish from
production hatcheries have been allowed to spawn in the wild in the past
(examples given in ISAB, 2002). Finally, to our knowledge, there are no
studies evaluating whether selective harvesting has been effective in reducing
harvest risks to wild populations, and research on this issue is needed.
7.2.4. Disease
Historical movement of infected fish or contaminated eggs and the practise
of using raw, unpasteurized salmon viscera as fish food have contributed
significantly to the introduction or spread of many fish pathogens. However,
awareness of these issues, implementation of strategies to control infectious
agents in hatcheries and development of standards and guidelines for movement of aquatic animals have done much to reduce the spread of pathogens
and the impact of infectious disease (Winton, 2001). The more controversial
aspect of the ‘hatchery-wild’ debate is around the role of hatchery fish
in amplifying and/or transmitting endemic pathogens to susceptible wild
populations. While this issue is often raised as a criticism against hatchery
operations, very little is actually known about this specific source of risk to
wild fish.
Our lack of understanding in this area is explained partly by the fact that
standard methods have been developed for the detection of fish pathogens
(American Fisheries Society, 2004; Office International des Epizooties,
2003), there are few published studies that have determined levels of
pathogens in populations of wild fish or in environmental samples and
fewer still that have tried to assess the risk that infected hatchery fish or
contaminated hatchery effluents might pose to wild populations. Current
methods for epidemiological strain typing of pathogens typically cannot
distinguish hatchery from wild origin, and thus it has been difficult to
demonstrate the direction of transmission for pathogens affecting both
hatchery and wild fish. Similarly, there is a poor understanding of the
factors that control the ecology of infectious disease among populations
of wild fish, the likelihood that wild fish will develop disease following
exposure to a pathogen under natural environmental conditions or the
effect of disease on the survival of salmonids in either freshwater or marine
environments. What is clear from the few examples given in Section 5.4 is
that the disease interactions between hatchery and wild fish are complex and
may be case-specific.
Evaluation of the Effect of Hatcheries on Wild Salmon
167
Several approaches for reducing disease risks to wild fish include the
following hatchery practises: sound sanitation, routine screening of spawning adults for pathogens, disinfecting fertilized eggs, maintaining families
separately to reduce horizontal transmission and frequent disease monitoring during the rearing period. Additionally, lower rearing densities and
good nutrition can reduce stresses that exacerbate disease. Finally, hatchery
water supplies should be from pathogen-free source and the hatchery
effluent treated, wherever feasible. Many of these practises are in place at a
wide range of hatcheries (Waples and Drake, 2005).
In summary, the role that hatchery fish play in affecting the disease ecology
of wild salmonid populations is highly equivocal. Research focused on
the factors controlling the disease cycle in wild fish is needed to assist
in determining the risk, if any, that hatchery fish pose to their wild counterparts.
7.3. Economic issues
To adequately consider the economic consequences of SEPs, at least two
lines of inquiry need to be pursued further. First, the standard project
evaluation and selection tools—BCA and CEA—are designed to assist in
setting priorities and choosing projects for funding. As noted earlier, these
emphasize efficiency in decision making and proper balancing of government funding when outcomes are quantifiable and economic consequences
can be measured. Our review of past BCA studies shows that public salmon
hatchery programmes generate economic consequences from high to low in
terms of a benefit-cost ratio. By applying BCA to the sub-parts of the British
Columbia salmon fishery enhancement programme, Pearse (1994) found
some elements with high benefit-cost ratios even though the programme as
a whole performed poorly by this standard. This information should assist in
the selection of fishery enhancement projects that yield positive economic
benefits. Existing CEAs show that hatcheries in the Columbia basin and
Oregon have widely varying costs, ranging from $23 to $68,031 per
additional fish caught. Clearly, where projects aim to increase fish harvests,
hatcheries achieving a lower cost per fish represent a better public
investment in fishery enhancement.
Because these objectives for conservation hatcheries and mitigation
hatcheries (e.g., fishing opportunities for Native Americans) are less easily
quantified in economic terms, BCA is less relevant to selecting projects of
this type. Still, CEA is an appropriate decision tool where a range of
alternatives is being considered for species protection or fisheries enhancement. Second, the project selection process inevitably triggers shifts in
locations of government facilities and expenditures, and these fuel local
economic impacts. Hence, impacts on small, rural communities become a
focus for government decisions when salmon enhancement projects are
168
Kerry A. Naish et al.
being considered. This brings the discussion of hatchery openings and
closing directly into the political crossfire as those most likely to be affected
raise their concerns through democratic processes. Overall, since little
economic analysis has been included in the design and project selection
process for SEPs, it is not surprising that these programmes are not demonstrated to be strong contributors to our economic prosperity. Inclusion of
improved economic analysis in project design and selection could improve
the situation.
7.4. Moving forward: Scientific and social dimensions
Most enhancement activities are operated under the principle of ‘adaptive
management’ (Section 3), that is, that hatchery practises should change as
new scientific information becomes available. Practically, attempts to
address many of the knowledge gaps we identify in our review have only
recently been implemented and thus hatcheries have been slow to reform.
We note, too, that there has rarely been a coordinated and programmatic
approach to managing hatcheries within a given region. Throughout this
chapter, it has been difficult to identify whether hatchery risks are due to
inherent biological problems or due to poor management decisions. This
criticism is not new (Hilborn, 1992a; Rich, 1922). Rarely have programmes
been set up to effectively track any question, and, although a difficult goal to
fulfil, they have not generated sufficiently systematic data to prove success or
failure.
In some cases, political developments have led to a growing number of
attempts to reform hatchery practises. We mentioned earlier that an independent scientific panel was mandated to review hatchery programmes in
the Pacific Northwest and provide broad recommendations and guidelines
for reforming existing hatcheries (Mobrand et al., 2005). The process
identified several key guiding principles that the nature and objectives of
hatchery programmes must be clearly identified and programme success be
measured against these goals, that operations and establishment of programmes should be scientifically defensible and that hatcheries should
respond rapidly to new information as it becomes available.
While few people will quibble with such clear recommendations based
on scientific principles, it is important to consider the social and political
contexts in which the recommendations were made (Section 1). The reader
is reminded of the arguments presented on the political aspects underpinning hatcheries (Section 3); namely that it is not science, but economic and
cultural issues that motivate hatchery programmes. The hatchery reform
process inherently acknowledges a priori that hatcheries have a role to play in
recovering threatened populations or in enhancing fisheries, and it is in this
political framework that the science is conducted. The alternative, that all
hatcheries be closed, is unlikely to be seriously considered in the near future.
Evaluation of the Effect of Hatcheries on Wild Salmon
169
Thus, the successful implementation of any scientific approach is dependent
on sustained political support.
Recent attempts to reform hatchery practises are a positive move.
However, if the political process does not include ongoing attempts to
answer the key knowledge gaps (some of which are identified above) then
hatchery management will not have the appropriate tools for long-term
monitoring and will continue to be managed without a sound scientific
foundation. Without these tools, the larger question of whether hatcheries
can, in fact, support conservation and harvest activities while minimizing
risks to wild populations in a socially acceptable framework will remain
largely unanswered. (This is not a trivial question: the counting of hatchery
fish in listing decisions under the Endangered Species Act has been debated
in court, and has resulted in policy reformulations in the United States; Alsea
Valley Alliance v. Evans; NOAA Federal Register June 2005.) It should be
noted, too, that the reform process attempts to change the practises of
existing programmes and should not be interpreted as an excuse for creating
new ones. Yet this is a possible outcome. The social and economic processes
driving hatchery reform will inevitably use different measures of success
than will biological approaches, and the formulation of a set of recommendations may be seen as that success. It should be emphasized that the reform
of hatchery practises inevitably involves trade-offs between different risks
(e.g., reducing competition between hatchery and wild fish in freshwater by
releasing hatchery fish at outmigration may increase genetic changes due to
domestication; Waples and Drake, 2005). The weighting of these risks will
likely occur at the societal level. Finally, it should be noted that the
implementation of hatchery reform is limited to the regions of the world
in which the tenor of the political debate is at its strongest. It is still unclear
whether there is sufficient social impetus to implement such changes
worldwide, and yet it is clear that they are needed.
7.5. Conclusions
We conclude by restating the intent of this chapter. The subject matter has
focused largely on areas in which hatcheries could adversely impact wild
stocks. We do not suggest that hatcheries should not have a role in salmon
enhancement where their use represents an important means to recover
critically endangered stocks. In Section 1, we list several populations (e.g.,
the Snake River Sockeye in the Pacific Northwest) that would be extinct
without a captive propagation programme.
However throughout this chapter, it has been difficult to separate
biological factors from social factors in problems associated with salmon
hatchery programmes. Despite the fact that hatcheries have been operated
over many decades, it is still unclear whether such activities can support
conservation and fishery goals. A greater emphasis should be placed on
170
Kerry A. Naish et al.
experimental approaches to reforming hatchery practises by conducting
coordinated research within the existing and extensive hatchery system
using appropriate controls. This research should be supported by a climate
of active debate about the role of salmon hatcheries in today’s society.
ACKNOWLEDGEMENTS
We are very grateful to Nigel Milner and Brian Davidson for providing unpublished data for
Section 4. Additionally, we are indebted to a number of colleagues who have provided
excellent comments on the chapter, although the views expressed here are entirely our own;
Nigel Milner, Jeff Hard, Fred Utter, Robin Waples, Nils Ryman, Barry Berejikian, Walt
Dickhoff, Michael Dauer and Jordan Watson. We also thank the editor, David Sims, and
previous editors, Alan Southward and Lee Fuiman, for their guidance throughout the
process of writing.
REFERENCES
Abbott, J. C., Dunbrack, R. L. and Orr, C. D. (1985). The interaction of size and experience
in dominance relationships of juvenile steelhead trout (Salmo gairdneri). Behaviour 92,
241–253.
Allard, D. C. (1978). ‘‘Spencer Fullerton Baird and the U.S. Fish Commission.’’ Arno Press,
New York.
Allen, C. (2003). Replacing salmon—Columbia River Indian fishing rights and the geography of fisheries mitigation. Oreg. Hist. Q. 104, 196–227.
Allendorf, F. W. (1993). Delay of adaptation to captive breeding by equalizing family size.
Conserv. Biol. 7, 416–419.
Allendorf, F. W. and Leary, R. F. (1988). Conservation and distribution of genetic variation
in a polytypic species, the cutthroat trout. Conserv. Biol. 2, 170–184.
Allendorf, F. W. and Ryman, N. (1987). Genetic management of hatchery stocks.
In ‘‘Population Genetics and Fishery Management’’ (N. Ryman and F. Utter, eds),
pp. 141–160. Washington Sea Grant Program, University of Washington Press, Seattle.
Allendorf, F. W. and Waples, R. S. (1996). Conservation and genetics of salmonid fishes.
In ‘‘Conservation Genetics: Case Histories from Nature’’ ( J. C. Avise and J. L. Hamrick,
eds), pp. 238–239. Chapman and Hall, New York.
Allendorf, F. W., Leary, R. F., Spruell, P. and Wenburg, J. K. (2001). The problems with
hybrids: Setting conservation guidelines. Trends Ecol. Evol. 16, 613–622.
Almodovar, A., Suarez, J., Nicola, G. G. and Nuevo, M. (2001). Genetic introgression between
wild and stocked brown trout in the Douro River basin, Spain. J. Fish Biol. 59, 68–74.
American Fisheries Society (2004). Fish health section blue book: suggested Procedures for the
Detection and Identification of Certain Finfish and Shellfish Pathogens, 2004 edition
American Fisheries Society Bethesda, MD.
Amos, K. and Thomas, J. (2002). Disease interactions between wild and cultured fish: Observations and lessons learned in the Pacific Northwest. Bull. Eur. Assoc. Fish Pathol. 22, 95–102.
Andersen, D. H., Pertoldi, O., Scali, V. and Loeschcke, V. (2002). Intraspecific
hybridization, developmental stability and fitness in Drosophila mercatorum. Evol. Ecol.
Res. 4, 603–621.
Anderson, E. D., Engelking, H. M., Emmenegger, E. J. and Kurath, G. (2000). Molecular
epidemiology reveals emergence of a virulent infectious hematopoietic necrosis (IHN) virus
strain in wild salmon and its transmission to hatchery fish. J. Aquat. Anim. Health 12, 85–99.
Evaluation of the Effect of Hatcheries on Wild Salmon
171
Anonymous, G. (2004). Why Montana went wild. In ‘‘Montana Outdoors,’’ Vol. 35.
Aprahamian, M. W., Smith, K. M., McGinnity, P., McKelvey, S. and Taylor, J. (2003).
Restocking of salmonids—Opportunities and limitations. Fish. Res. 62, 211–227.
Araki, H., Ardren, W. R., Olsen, E., Cooper, B. and Blouin, M. S. (2007). Reproductive
success of captive-bred steelhead trout in the wild: Evaluation of three hatchery programs
in the Hood river. Conserv. Biol. 21, 181–190.
Arkoosh, M. R., Casillas, E., Clemons, E., McCain, B. and Varanasi, U. (1991). Suppression
of immunological memory in juvenile Chinook salmon (Oncorhynchus tshawytscha) from
an urban estuary. Fish Shellfish Immunol. 1, 261–277.
Arkush, K. D., Giese, A. R., Mendonca, H. L., McBride, A. M., Marty, G. D. and
Hedrick, P. W. (2002). Resistance to three pathogens in the endangered winter-run
Chinook salmon (Oncorhynchus tshawytscha): Effects of inbreeding and major histocompatibility complex genotypes. Can. J. Fish. Aquat. Sci. 59, 966–975.
Aurelle, D., Lek, S., Giraudel, J. L. and Berrebi, P. (1999). Microsatellites and artificial
neural networks: Tools for the discrimination between natural and hatchery brown trout
(Salmo trutta, L.) in Atlantic populations. Ecol. Model. 120, 313–324.
Ayerst, J. D. (1977). The role of hatcheries in rebuilding steelhead runs of the Columbia
River system. In ‘‘Columbia River Salmon and Steelhead’’ (E. Schwiebert, ed.), Vol. 10,
pp. 84–88. American Fisheries Society, Bethesda, MD.
Baird, S. F. (1874).‘‘Letter to the Chairman of the Committee on Commerce,’’ Vol. 79.
Senate Miscellaneous Documents, 43rd Congress, 2nd Session.
Baird, S. F. (1875). Letter to Hon. Z. Chandler, 25 Jan. 1875. ‘‘Records of the United States
Fish and Wildlife Service, Letters Received 1882–1900.’’ National Archives, College
Park, Maryland.
Baldwin, T. J., Peterson, J. E., McGhee, G. C., Staigmiller, K. D., Motteram, E. S.,
Downs, C. C. and Stanek, D. R. (1998). Distribution of Myxobolus cerebralis in salmonid
fishes in Montana. J. Aquat. Anim. Health 10, 361–371.
Bams, R. A. (1976). Survival and propensity for homing as affected by presence or absence of
locally adapted paternal genes in 2 transplanted populations of pink salmon (Oncorhynchus
gorbuscha). J. Fish. Res. Board Canada 33, 2716–2725.
Bartholomew, J. L. (1998). Host resistance to infection by the myxosporean parasite
Ceratomyxa shasta: A review. J. Aquat. Anim. Health 10, 112–120.
Bartholomew, J. L. and Reno, P. W. (2002). The history and dissemination of whirling
disease. In ‘‘Whirling Disease: Reviews and Current Topics. American Fisheries Society
Symposium’’ (J. L. Bartholomew and J. C. Wilson, eds), Vol. 29, pp. 3–24. American
Fisheries Society, Bethesda, MD.
Bartholomew, J. L., Kerans, B. L., Hedrick, R. P., MacDiarmid, S. C. and Winton, J. R.
(2005). A risk assessment based approach for the management of whirling disease. Rev.
Fish. Sci. 13, 205–230.
Bartley, D. M., Bagley, M., Gall, G. and Bentley, B. (1992). Use of linkage disequilibrium data to
estimate effective size of hatchery and natural fish populations. Conserv. Biol. 6, 365–375.
Beacham, T. D., Lapointe, M., Candy, J. R., Miller, K. M. and Withler, R. E. (2004). DNA
in action: Rapid application of DNA variation to sockeye salmon fisheries management.
Conserv. Genet. 5, 411–416.
Beamish, R. J., Mahnken, C. and Neville, C. M. (1997). Hatchery and wild production of
Pacific salmon in relation to large-scale, natural shifts in the productivity of the marine
environment. ICES J. Mar. Sci. 54, 1200–1215.
Bell, K., Huppert, D. and Johnson, R. (2003). Willingness to pay for local coho salmon
enhancement in coastal communities. Mar. Resour. Econ. 18, 15–32.
Bell, J. D., Riothisbreg, P. C., Munro, J. L., Loneragan, N. R., Nash, W. J., Ward, R. D.
and Andrew, N. L. (2005). Restocking and stock enhancement of marine invertebrate
fisheries. Adv. Mar. Biol. 49, 1–374.
172
Kerry A. Naish et al.
Berejikian, B. A. (1995). The effects of hatchery and wild ancestry and experience on the
relative ability of steelhead trout fry (Oncorhynchus mykiss) to avoid a benthic predator.
Can. J. Fish. Aquat. Sci. 52, 2476–2482.
Berejikian, B. and Ford, M. J. (2004). In ‘‘Review of the Relative Fitness of Hatchery and
Natural Salmon.’’ Rep. No. NOAA Technical Memorandum NMFS-NWFSC-61.
Northwest Fisheries Science Center, Seattle. http://www.nwfsc.noaa.gov/publications/.
Berejikian, B. A., Mathews, S. B. and Quinn, T. P. (1996). Effects of hatchery and wild
ancestry and rearing environments on the development of agonistic behavior in steelhead
trout (Oncorhynchus mykiss) fry. Can. J. Fish. Aquat. Sci. 53, 2004–2014.
Berejikian, B. A., Tezak, E. P., Schroder, S. L., Knudsen, C. M. and Hard, J. J. (1997).
Reproductive behavioral interactions between wild and captively reared coho salmon
(Oncorhynchus kisutch). ICES J. Mar. Sci. 54, 1040–1050.
Berejikian, B. A., Tezak, E. P., Schroder, S. L., Flagg, T. A. and Knudsen, C. M. (1999).
Competitive differences between newly emerged offspring of captive-reared and wild
coho salmon. Trans. Am. Fish. Soc. 128, 832–839.
Berejikian, B. A., Tezak, E. P., Flagg, T. A., LaRae, A. L., Kummerow, E. and
Mahnken, C. V. W. (2000). Social dominance, growth, and habitat use of age-0 steelhead
(Oncorhynchus mykiss) grown in enriched and conventional hatchery rearing environments. Can. J. Fish. Aquat. Sci. 57, 628–636.
Berejikian, B. A., Tezak, E. P., Park, L., LaHood, E., Schroder, S. L. and Beall, E. (2001a).
Male competition and breeding success in captively reared and wild coho salmon
(Oncorhynchus kisutch). Can. J. Fish. Aquat. Sci. 58, 804–810.
Berejikian, B. A., Tezak, E. P., Riley, S. C. and LaRae, A. L. (2001b). Competitive ability
and social behaviour of juvenile steelhead reared in enriched and conventional hatchery
tanks and a stream environment. J. Fish Biol. 59, 1600–1613.
Berg, S. and Jorgensen, J. (1991). Stocking experiments with 0þ and 1þ trout parr, Salmo
trutta L., of wild and hatchery origin: 1. Post-stocking mortality and smolt yield. J. Fish
Biol. 39, 151–169.
Berman, M., Haley, S. and Kim, H. (1997). Estimating net benefits of reallocation: Discrete
choice models of sport and commercial fishing. Mar. Resour. Econ. 12, 307–327.
Berrebi, P., Poteaux, C., Fissier, M. and Cattaneo-Berrebi, G. (2000a). Stocking impact and
allozyme diversity in brown trout from Mediterranean southern France. J. Fish Biol. 56,
949–960.
Berrebi, P., Povz, M., Jesensek, D., Cattaneo-Berrebi, G. and Crivelli, A. J. (2000b). The
genetic diversity of native, stocked and hybrid populations of marble trout in the Soca
river, Slovenia. Heredity 85, 277–287.
Bijlsma, R., Bundgaard, J. and Boerema, A. C. (2000). Does inbreeding affect the extinction
risk of small populations? Predictions from Drosophila. J. Evol. Biol. 13, 502–514.
Blaxter, J. H. S. (2000). The enhancement of marine fish stocks. Adv. Mar. Biol. 38, 1–54.
Blumm, M. C. (2002). ‘‘Sacrificing the Salmon: A Legal and Policy History of the Decline of
Columbia Basin Salmon.’’ BookWorld Publications, Den Bosch, The Netherlands.
Bogue, M. B. (2000). ‘‘Fishing the Great Lakes: An Environmental History, 1783–1933.’’
University of Wisconsin Press, Madison.
Borgstrom, R., Skaala, O. and Aastveit, A. H. (2002). High mortality in introduced brown
trout depressed potential gene flow to a wild population. J. Fish Biol. 61, 1085–1097.
Bottom, D. L. (1997). To till the water—A history of ideas in fisheries conservation.
In ‘‘Pacific Salmon and Their Ecosystems’’ (D. J. Stouder, P. A. Bisson and
R. J. Naiman, eds), pp. 569–597. Chapman and Hall, New York.
Bowen, J. T. (1970). A history of fish culture as related to the development of fisheries
programme. In ‘‘A Century of Fisheries in North America’’ (N. G. Benson, ed.),
pp. 71–93. American Fisheries Society, Washington.
Evaluation of the Effect of Hatcheries on Wild Salmon
173
Boyce, J., Herrmann, M., Bischak, D. and Greenberg, J. (1993). The Alaska salmon
enhancement programme: A cost/benefit analysis. Mar. Resour. Econ. 8, 293–312.
Bradford, M. J., Taylor, G. C. and Allan, J. A. (1997). Empirical review of coho salmon
smolt abundance and the prediction of smolt production at the regional level. Trans. Am.
Fish. Soc. 126, 49–64.
Brännäs, E. (1995). First access to territorial space and exposure to strong predation pressure:
A conflict in early emerging Atlantic salmon (Salmo salar L.) fry. Evol. Ecol. 9, 411–420.
Brannon, E. L., Amend, D. F., Cronin, M. A., Lannan, J. E., LaPatra, S., McNeil, W. J.,
Noble, R. E., Smith, C. E., Talbot, A. J., Wedemeyer, G. A. and Westers, H. (2004a).
The controversy about salmon hatcheries. Fisheries 29, 12–31.
Brannon, E. L., Powell, M. S., Quinn, T. P. and Talbot, A. (2004b). Population structure of
Columbia River Basin Chinook salmon and steelhead trout. Rev. Fish. Sci. 12, 99–232.
Briscoe, R. J., Adkinson, M. D., Wertheimer, A. and Taylor, S. G. (2005). Biophysical
factors associated with the marine survival of Auke Creek, Alaska coho salmon. Trans.
Am. Fish. Soc. 134, 817–828.
Brooks, K. M. (2005). The effects of water temperature, salinity, and currents on the survival
and distribution of the infective copepodid stage sea lice (Lepeophtheirus salmonis)
originating on Atlantic salmon farms in the Broughton Archepelago of British Columbia,
Canada. Rev. Fish. Sci. 13, 177–204.
Brown, B. (1982). ‘‘Mountain in the Clouds: A Search for the Wild Salmon.’’ Simon and
Schuster, New York.
Brown, W. B., Sorhus, C., Chou-Yang, B.-L. and Richards, J. A. (1983). Using individual
observations to estimate recreation demand functions: A caution. Am. J. Agric. Econ. 65,
154–157.
Brunner, P. C., Douglas, M. R. and Bernatchez, L. (1998). Microsatellite and mitochondrial
DNA assessment of population structure and stocking effects in Arctic charr Salvelinus
alpinus (Teleostei: Salmonidae) from central Alpine lakes. Mol. Ecol. 7, 209–223.
Buchal, J. L. (1998). ‘‘The Great Salmon Hoax.’’ Iconoclast Publishing, Aurora, OR.
Busack, C. A. and Currens, K. P. (1995). Genetic risks and hazards in hatchery operations;
fundamental concepts and issues. Am. Fish. Sci. Symp. 15, 71–80.
Butler, J. R. A. and Walker, A. F. (2006). Characteristics of the sea trout (Salmo trutta) stock
collapse in the River Ewe (Wester Ross), Scotland, in the Late 1980s. In ‘‘Sea Trout:
Biology, Conservation and Management. Proceedings of First International Sea Trout
Symposium, Cardiff, July 2004’’ (G. S. Harris and N. J. Milner, eds), pp. 45–59. Fishing
News Books, Blackwell Scientific Publications, Oxford.
Cagigas, M. E., Vazquez, E., Blanco, G. and Sanchez, J. A. (1999). Genetic effects of
introduced hatchery stocks on indigenous brown trout (Salmo trutta L.) populations in
Spain. Ecol. Freshwater Fish 8, 141–150.
Campbell, M. R., Dillon, J. and Powell, M. S. (2002). Hybridization and introgression in a
managed, native population of Yellowstone cutthroat trout: Genetic detection and
management implications. Trans. Am. Fish. Soc. 131, 364–375.
Campton, D. E. (1995). Genetic effects of hatchery fish on wild populations of Pacific
salmon and steelhead: What do we really know? Trans. Am. Fish. Soc. 15, 337–353.
Campton, D. E. (2004). Sperm competition in salmon hatcheries: The need to institutionalize genetically benign spawning protocols. Trans. Am. Fish. Soc. 133, 1277–1289.
Candy, J. R. and Beacham, T. D. (2000). Patterns of homing and straying in southern British
Columbia coded-wire tagged Chinook salmon (Oncorhynchus tshawytscha) populations.
Fish. Res. 47, 41–56.
Cart, T. W. (1968). ‘‘Federal Fisheries Service, 1871–1940.’’ University of North Carolina,
MA.
Cederholm, C. J., Bilby, R. E., Bisson, P. A., Bumstead, T. W., Fransen, B. R.,
Scarlett, W. J. and Ward, J. W. (1997). Response of juvenile coho salmon and steelhead
174
Kerry A. Naish et al.
to placement of large woody debris in a coastal Washington stream. North Am. J. Fish.
Manag. 17, 947–963.
Chapman, D. W. (1962). Aggressive behavior in juvenile coho salmon as a cause of
emigration. J. Fish. Res. Board Canada 19, 1047–1080.
Chapman, D. W. (1966). Food and space as regulators of salmonid populations in streams.
Am. Nat. 100, 345–357.
Chiarappa, M. J. and Szylvian, K. M. (2003). ‘‘Fish for all: An Oral History of Multiple
Claims and Divided Sentiment on Lake Michigan.’’ Michigan State University Press,
East Lansing.
Collie, J. (1993). ‘‘Salmon Population and Exploitation Modeling. Prepared for the Senate
Special Committee on Domestic and International Commercial Fisheries.’’ Juneau, Alaska.
Colpitts, G. (2002). ‘‘Game in the Garden: A Human History of Wildlife in Western Canada
to 1940.’’ University of British Columbia Press, Vancouver.
Cone, J. and Ridlington, S. (1996). ‘‘The Northwest Salmon Crises: A Documentary
History.’’ Oregon State University Press, Corvallis.
Cooney, R. T. and Brodeur, R. D. (1998). Carrying capacity and North Pacific salmon
production: Stock-enhancement implications. Bull. Mar. Sci. 62, 443–464.
Costelloe, M., Costelloe, J., O’Donohoe, G., Coghlan, N., Oonk, M. and
Heijden, Y. V. D. (1998). Planktonic distribution of sea lice larvae, Lepeophtheirus
salmonis, in Killary Harbour, West Coast of Ireland. J. Mar. Biol. Assoc. UK 78, 853–874.
Cowx, I. G. (1998). Stocking strategies: Issues and options for future enhancement
programs. In ‘‘Stocking and Introduction of Fish’’ (I. G. Cowx, ed.), pp. 3–13. Fishing
News Books, Oxford.
Crnokrak, P. and Roff, D. A. (1999). Inbreeding depression in the wild. Heredity 83, 260–270.
Cross, T. F. (2000). Genetic implications of translocation and stocking of fish species, with
particular reference to Western Australia. Aquacult. Res. 31, 83–94.
Crowl, T. A., Townsend, C. R. and Mcintosh, A. R. (1992). The impact of introduced brown
and rainbow trout on native fish: The case of Australasia. Rev. Fish Biol. Fish. 2, 217–241.
Crozier, W. W. and Moffett, I. J. J. (1995). Application of low frequency genetic marking at
GPI-3* and MDH-B1,2* loci to assess supplementary stocking of Atlantic salmon,
Salmo salar L., in a Northern Irish stream. Fish. Manag. Ecol. 2, 27–36.
Crozier, W. W., Schon, P. J., Chaput, G., Potter, E. C. E., Maoileidigh, N. O. and
MacLean, J. C. (2004). Managing Atlantic salmon (Salmo salar L.) in the mixed stock
environment: Challenges and considerations. ICES J. Mar. Sci. 61, 1344–1358.
Currens, K. P., Hemmingsen, A. R., French, R. A., Buchanan, D. V., Schreck, C. B. and
Li, H. W. (1997). Introgression and susceptibility to disease in a wild population of
rainbow trout. North Am. J. Fish. Manag. 17, 1065–1078.
Dahl, J., Pettersson, E., Dannewitz, J., Jarvi, T. and Lof, A. C. (2006). No difference in
survival, growth and morphology between offspring of wild-born, hatchery and hybrid
brown trout (Salmo trutta). Ecol. Freshw. Fish 15, 388–397.
Dannewitz, J., Petersson, E., Prestegaard, T. and Jarvi, T. (2003). Effects of sea-ranching and
family background on fitness traits in brown trout Salmo trutta reared under near-natural
conditions. J. Appl. Ecol. 40, 241–250.
Danzmann, R. G. and Gharbi, K. (2001). Gene mapping in fishes: a means to an end.
Genetica 111, 3–23.
Dawley, E. M., Ledgerwood, R. D., Blahm, T. H., Sims, C. W., Durkin, J. T.,
Kirn, R. A., Rankis, A. E., Monan, G. E. and Ossiander, F. J. (1986). ‘‘Migrational
Characteristics, Biological Observations, and Relative Survival of Juvenile Salmonids
Entering the Columbia River Estuary, 1966–1983.’’ Rep. No. 81–102. Bonneville
Power Administration,U.S. Department of Energy.
DeRose, M. A. and Roff, D. A. (1999). A comparison of inbreeding depression in lifehistory and morphological traits in animals. Evolution 53, 1288–1292.
Evaluation of the Effect of Hatcheries on Wild Salmon
175
Docker, M. F., Dale, A. and Heath, D. D. (2003). Erosion of interspecific reproductive
barriers resulting from hatchery supplementation of rainbow trout sympatric with
cutthroat trout. Mol. Ecol. 12, 3515–3521.
Drake, J., Ford, M., Kostow, K., Myers, J., Waknitz, B. and Adams, P. (2003). ‘‘Hatchery
Broodstock Summaries and Assessments for Chum Coho, Chinook Salmon and Steelhead Stocks Within Evolutionarily Significant Units Listed Under the Endangered
Species Act.’’ NOAA Fisheries, Northwest Fisheries Science Center, Seattle,
Washington.
Duchesne, P. and Bernatchez, L. (2002). An analytical investigation of the dynamics of
inbreeding in multi-generation supportive breeding. Conserv. Genet. 3, 47–60.
Edmands, S. (2002). Does parental divergence predict reproductive compatibility? Trends
Ecol. Evol. 17, 520–527.
Edmands, S. and Timmerman, C. C. (2003). Modeling factors affecting the severity of
outbreeding depression. Conserv. Biol. 17, 883–892.
Eicher, J. L. (1970). Fish passage. In ‘‘A Century of Fisheries in North America’’
(N. G. Benson, ed.), pp. 161–173. American Fisheries Society, Washington.
Einum, S. and Fleming, I. A. (2000). Selection against late emergence and small offspring in
Atlantic salmon (Salmo salar). Evolution 54, 628–639.
Eldridge, W. and Killebrew, K. (2007). Genetic diversity over multiple generations of
supplementation: An example from Chinook salmon using microsatellite and demographic data. Conserv. Genet. Online first, DOI: 10.1007/s10592-007-9298-y.
Eldridge, W. H. and Naish, K. A. (2007). Long term effects of translocation and release
numbers on fine scale population structure among coho salmon (Oncorhynchus kisutch).
Mol. Ecol. 16, 2407–2421.
Elliott, D. G., Pascho, R. J., Jackson, L. M., Matthews, G. M. and Harmon, J. R. (1997).
Renibacterium salmoninarum in spring-summer Chinook salmon smolts at dams on the
Columbia and Snake Rivers. J. Aquat. Anim. Health 9, 114–126.
Emlen, J. M. (1991). Heterosis and outbreeding depression: A multi-locus model and an
application to salmon production. Fish. Res. 12, 187–212.
Englbrecht, C. C., Schliewen, U. and Tautz, D. (2002). The impact of stocking on the
genetic integrity of Arctic charr (Salvelinus) populations from the Alpine region. Mol.
Ecol. 11, 1017–1027.
Environment Agency (2004). ‘‘Fisheries Statistics 2004: Salmonid and Freshwater Fisheries
Statistics for England and Wales.’’ Environment Agency, Bristol, UK, www.
environment-agency.gov.uk.
Environment and Natural Resources Institute (2001). ‘‘Evaluating Alaska’s Ocean-Ranching
Salmon Hatcheries: Biologic and Management Issues.’’ University of Alaska, Anchorage.
Eriksson, T. and Eriksson, L. O. (1993). The status of wild and hatchery propagated Swedish
salmon stocks after 50 years of hatchery releases into the Baltic rivers. Fish. Res. 18, 147–159.
Essington, T. E., Quinn, T. P. and Ewert, V. E. (2000). Intra- and interspecific competition and
the reproductive success of sympatric Pacific salmon. Can. J. Fish. Aquat. Sci. 57, 205–213.
Evenden, M. D. (2004). ‘‘Fish Versus Power: An Environmental History of the Fraser
River.’’ Cambridge University Press, Cambridge.
Farrington, C. (2003). ‘‘Alaska Salmon Enhancement Programme 2002 Annual Report.’’
Rep. No. 5J03–05. Alaska Department of Fish and Game, Regional Information.
Fenster, C. B. and Galloway, L. F. (2000). Inbreeding and outbreeding depression in natural
populations of Chamaecrista fasciculata (Fabaceae). Conserv. Biol. 14, 1406–1412.
Flagg, T. A., Waknitz, F. W., Maynard, D. J., Milner, G. B. and Mahnken, C. V. W. (1995).
The effects of hatcheries on native coho salmon populations in the lower Columbia
River. Am. Fish. Soc. Symp. 15, 366–375.
Flagg, T. A., Berejikian, B. A., Colt, J. E., Dickhoff, W. W., Harrell, L. W., Maynard, D. J.,
Nash, C. E., Strom, M. E., Iwamoto, R. N. and Mahnken, C. V. W. (2000). ‘‘Ecological
176
Kerry A. Naish et al.
and Behavioral Impacts of Artificial Production Strategies on the Abundance of Wild
Salmon Populations—A Review of Practices in the Pacific Northwest.’’ Rep No.
NOAA Technical Memorandum NMFS-NWFSC-41. U.S. Department of Commerce,
Seattle, Washington.
Fleming, I. A. and Einum, S. (1997). Experimental tests of genetic divergence of farmed
from wild Atlantic salmon due to domestication. ICES J. Mar. Sci. 54, 1051–1063.
Fleming, I. A. and Gross, M. R. (1992). Reproductive behavior of hatchery and wild coho
salmon (Oncorhynchus kisutch)—Does it differ? Aquaculture 103, 101–121.
Fleming, I. A. and Gross, M. R. (1993). Breeding success of hatchery and wild coho salmon
(Oncorhynchus kisutch) in competition. Ecol. Appl. 3, 230–245.
Fleming, I. A., Jonsson, B. and Gross, M. R. (1994). Phenotypic divergence of SeaRanched, Farmed, and Wild Salmon. Can. J. Fish. Aquat. Sci. 51, 2808–2824.
Fleming, I. A., Jonsson, B., Gross, M. R. and Lamberg, A. (1996). An experimental study of
the reproductive behaviour and success of farmed and wild Atlantic salmon (Salmo salar).
J. Appl. Ecol. 33, 893–905.
Foerster, R. E. (1936). Sockeye salmon propagation in British Columbia. Biol. Board Can.
Bull. 53.
Foote, C. J., Wood, C. C., Clarke, W. C. and Blackburn, J. (1992). Circannual cycle of
seawater adaptability in Oncorhynchus nerka—Genetic differences between sympatric
sockeye salmon and kokanee. Can. J. Fish. Aquat. Sci. 49, 99–109.
Ford, M. J. (2002). Selection in captivity during supportive breeding may reduce fitness in
the wild. Conserv. Biol. 16, 815–825.
Ford, M. J., Teel, D., Van Doornik, D. M., Kuligowski, D. and Lawson, P. W. (2004).
Genetic population structure of central Oregon coast coho salmon (Oncorhynchus kisutch).
Conserv. Genet. 5, 797–812.
Ford, M. J., Fuss, H., Boelts, B., LaHood, E., Hard, J. and Miller, J. (2006). Changes in run
timing and natural smolt production in a naturally spawning coho salmon (Oncorhynchus
kisutch) population after 60 years of intensive hatchery supplementation. Can. J. Fish.
Aquat. Sci. 63, 2343–2355.
Frankham, R., Ballou, J. D. and Briscoe, D. A. (2002). ‘‘Introduction to Conservation
Genetics.’’ Cambridge University Press, Cambridge.
Franklin, I. R. (1980). Evolutionary changes in small populations. In ‘‘Conservation
Biology: An Evolutionary Ecological Perspective’’ (M. E. Soule, ed.), pp. 135–149.
Sinauer Associates, Sunderland, MA.
Fraser, G. A. and Friedlander, A. M. J. (1980). ‘‘Social Impacts and Economic Efficiency in
Resource Development Planning: The Case of Salmonid Enhancement on Canada’s
Pacific Coast.’’ Fisheries and Marine Service Technical Report 923, Vancouver, BC.
Fritzner, N. G., Hansen, M. M., Madsen, S. S. and Kristiansen, K. (2001). Use of microsatellite markers for identification of indigenous brown trout in a geographical region
heavily influenced by stocked domesticated trout. J. Fish Biol. 58, 1197–1210.
Galbreath, P. F., Adams, N. D., Guffey, S. Z., Moore, C. J. and West, J. L. (2001).
Persistence of native southern Appalachian brook trout populations in the Pigeon
River system, North Carolina. North Am. J. Fish. Manag. 21, 927–934.
Gallardo, J. A. and Neira, R. (2005). Environmental dependence of inbreeding depression in
cultured coho salmon (Oncorhynchus kisutch): Aggressiveness, dominance and intraspecific
competition. Heredity 95, 449–456.
Garcia-Marin, J. L., Sanz, N. and Pla, C. (1998). Proportions of native and introduced
brown trout in adjacent fished and unfished Spanish rivers. Conserv. Biol. 12, 313–319.
Garcia-Marin, J. L., Sanz, N. and Pla, C. (1999). Erosion of the native genetic resources of
brown trout in Spain. Ecol. Freshwater Fish 8, 151–158.
Evaluation of the Effect of Hatcheries on Wild Salmon
177
Garcia-Vazquez, E., Perez, J., Ayllon, F., Martinez, J. L., Glise, S. and Beall, E. (2004).
Asymmetry of post-F1 interspecific reproductive barriers among brown trout
(Salmo trutta) and Atlantic salmon (Salmo salar). Aquaculture 234, 77–84.
Gardner-Thorpe, C. (2001). Who was Frank Buckland? Biologist 48, 187–188.
Gargan, P. G., Poole, R. and Forde, G. (2006). A review of the status of Irish sea trout stocks.
In ‘‘Sea Trout: Biology, Conservation and Management. Proceedings of First International Sea Trout Symposium, Cardiff, July 2004’’ (G. S. Harris and N. J. Milner, eds),
pp. 25–44. Fishing News Books, Blackwell Scientific Publications, Oxford.
Gharrett, A. J., Smoker, W. W., Reisenbichler, R. R. and Taylor, S. G. (1999). Outbreeding depression in hybrids between odd- and even-broodyear pink salmon. Aquaculture
173, 117–129.
Gharrett, A. J. and Smoker, W. W. (1991). Two generations of hybrids between even-year
and odd-year pink salmon (Oncorhynchus gorbuscha)—A test for outbreeding depression.
Can. J. Fish. Aquat. Sci. 48, 1744–1749.
Gilk, S. E., Wang, I. A., Hoover, C. L., Smoker, W. W., Taylor, S. G., Gray, A. K. and
Gharrett, A. J. (2004). Outbreeding depression in hybrids between spatially separated
pink salmon, Oncorhynchus gorbuscha, populations: Marine survival, homing ability, and
variability in family size. Environ. Biol. Fishes 69, 287–297.
Gjerde, B., Gunnes, K. and Gjedrem, T. (1983). Effect of inbreeding on survival and growth
in rainbow trout. Aquaculture 34, 327–332.
Goode, G. B. (1881). Epochs in the history of fish culture. Trans. Am. Fish Cult. Soc. 10,
35–42.
Goodman, D. (2005). Selection equilibrium for hatchery and wild spawning fitness in
integrated breeding programmes. Can. J. Fish. Aquat. Sci. 62, 374–389.
Granath, K. L., Smoker, W. W., Gharrett, A. J. and Hard, J. J. (2004). Effects on embryo
development time and survival of intercrossing three geographically separate populations
of Southeast Alaska Coho Salmon, Oncorhynchus kisutch. Environ. Biol. Fish. 69, p. 299.
Great Britain (1861). ‘‘Statutes at Large of the United Kingdom of Great Britain and
Ireland.’’ Rep. No. 101. His Majesty’s Statute and Law Printers, London.
Great Britain (1862). ‘‘Statutes at Large of the United Kingdom of Great Britain and
Ireland.’’ Rep. No. 102. His Majesty’s Statute and Law Printers, London.
Great Britain (1863). ‘‘Statutes at Large of the United Kingdom of Great Britain and
Ireland.’’ Rep. No. 103. His Majesty’s Statute and Law Printers, London.
Great Britain (1865). ‘‘Statutes at Large of the United Kingdom of Great Britain and
Ireland.’’ Rep. No. 105. His Majesty’s Statute and Law Printers, London.
Gunnerd, T. B., Hvidsten, N. A. and Heggberget, T. G. (1988). Open sea releases of
Atlantic salmon smolts, Salmo salar, in central Norway, 1973–83. Can. J. Fish. Aquat. Sci.
45, 1340–1345.
Hansen, M. M. (2002). Estimating the long-term effects of stocking domesticated trout into
wild brown trout (Salmo trutta) populations: An approach using microsatellite DNA
analysis of historical and contemporary samples. Mol. Ecol. 11, 1003–1015.
Hansen, M. M., Nielsen, E. E., Ruzzante, D. E., Bouza, C. and Mensberg, K. L. D. (2000).
Genetic monitoring of supportive breeding in brown trout (Salmo trutta L.), using
microsatellite DNA markers. Can. J. Fish. Aquat. Sci. 57, 2130–2139.
Hansen, M. M., Nielsen, E. E., Bekkevold, D. and Mensberg, K. L. D. (2001). Admixture
analysis and stocking impact assessment in brown trout (Salmo trutta), estimated with
incomplete baseline data. Can. J. Fish. Aquat. Sci. 58, 1853–1860.
Hard, J. J. (1995). Genetic monitoring of life-history characters in salmon supplementation:
Problems and opportunities. Am. Fish. Sci. Symp. 15, 212–225.
Hard, J. J., Jones, R. P., Delarm, M. R. and Waples, R. S. (1992). Pacific salmon and
artificial propagation under the endangered species act. NOAA Technical Memorandum
NMFS-NWFSC-2, http://www.nwfsc.noaa.gov/pubs/tm/tm2/tm2.html.
178
Kerry A. Naish et al.
Hard, J. J., Berejikian, B. A., Tezak, E. P., Schroder, S. L., Knudsen, C. M. and Parker, L. T.
(2000). Evidence for morphometric differentiation of wild and captively reared adult
coho salmon: A geometric analysis. Environ. Biol. Fish. 58, 61–73.
Hargreaves, N. B. and LeBrasseur, R. J. (1986). Size selectivity of coho (Oncorhynchus kisutch)
preying on juvenile chum salmon (O. keta). Can. J. Fish. Aquat. Sci. 43, 581–586.
Harris, D. C. (2001). ‘‘Fish, Law, and Colonialism: The Legal Capture of Salmon in British
Columbia.’’ University of Toronto Press, Toronto.
Hayes, J. P., Guffey, S. Z., Kriegler, F. J., McCracken, G. F. and Parker, C. R. (1996). The
genetic diversity of native, stocked, and hybrid populations of brook trout in the southern
Appalachians. Conserv. Biol. 10, 1403–1412.
Healey, M. C. (1982a). Juvenile Pacific salmon in estuaries: The life support system. In ‘‘Estuarine
Comparisons’’ (V. S. Kennedy, ed.), pp. 315–341. Academic Press, New York.
Healey, M. C. (1982b). Timing and relative intensity of size-selective mortality of juvenile
chum salmon (Oncorhynchus keta) during early sea life. Can. J. Fish. Aquat. Sci. 39,
952–957.
Heard, W. R. (2001). Alaska salmon enhancement: A successful programme for hatchery
and wild stocks. In ‘‘30th Joint Meeting of the United States-Japan Cooperative
Programme in Natural Resources.’’ http://www.ufa-fish.org/taskforce/Doc/
hatcheries_ HASE112403.pdf, Sarasota, FL.
Heard, W. R., Nakamura, Y., McVey, J. P., Leber, K., Neidig, C., Fox, S. and Churchill, K.
(2003). Alaska salmon enhancement: A successful programme for hatchery and wild
stocks. In ‘‘Ecology of Aquaculture Species and Enhancement of Stocks. Proceedings
of the Thirtieth U.S.–Japan Meeting on Aquaculture’’ (Y. Nakamura, J. P. McVey,
K. Leber, C. Neidig, S. Fox and K. Churchill, eds). Mote Marine Laboratory, Sarasota,
FL.
Hedgecock, D., and Coykendall, K. (2007). Genetic risks of marine hatchery enhancement:
The good, the bad, and the unknown. In ‘‘Ecological and Genetic Implications of
Aquaculture Activities (T. Bert, ed.), pp. 85–101. Kluwer Academic Publisher, Norwell,
Massachusetts.
Hedrick, R. P. (1998). Relationships of the host, pathogen, and environment: Implications
for diseases of cultured and wild fish populations. J. Aquat. Anim. Health 10, 107–111.
Hedrick, P. W., Hedgecock, D. and Hamelberg, S. (1995). Effective population size in
winter-run chinook salmon. Conserv. Biol. 9, 615–624.
Hedrick, P. W., Rashbrook, V. K. and Hedgecock, D. (2000). Effective population size of
winter-run chinook salmon based on microsatellite analysis of returning spawners. Can.
J. Fish. Aquat. Sci. 57, 2368–2373.
Hedrick, R. P., McDowell, T. S., Mukkatira, K., Georgiadis, M. P. and MacConnell, E.
(2001). Susceptibility of three species of anadromous salmonids to experimentally
induced infections with Myxobolus cerebralis, the causative agent of whirling disease.
J. Aquat. Anim. Health 13, 43–50.
Hedrick, R. P., Batts, W. N., Yun, S., Traxler, G. S., Kaufman, J. and Winton, J. R. (2003).
Host and geographic range extensions of the North American strain of viral hemorrhagic
septicemia virus. Dis. Aquat. Organ. 55, 211–220.
Heggenes, J., Roed, K. H., Hoyheim, B. and Rosef, L. (2002). Microsatellite diversity
assessment of brown trout (Salmo trutta) population structure indicate limited genetic
impact of stocking in a Norwegian alpine lake. Ecol. Freshwater Fish 11, 93–100.
Heggenes, J., Beere, M., Tamkee, P. and Taylor, E. B. (2006). Genetic diversity in steelhead
before and after conservation hatchery operation in a coastal, boreal river. Trans. Am.
Fish. Soc. 135, 251–267.
Helle, J. H. (1981). Significance of the stock concept in artificial propagation of salmonids in
Alaska. Can. J. Fish. Aquat. Sci. 38, 1665–1671.
Evaluation of the Effect of Hatcheries on Wild Salmon
179
Hemmingsen, A. R., Holt, R. A., Ewing, R. D. and McIntyre, J. D. (1986). Susceptibility of
progeny from crosses among three stocks of coho salmon to infection by Ceratomyxa
shasta. Trans. Am. Fish. Soc. 115, 492–495.
Herbst, M. (2003). B.C. claims tighter fish-farming regs; Alaskans sceptical. In ‘‘Juneau Empire
State News,’’ Vol. 11. http://www.juneauempire.com/stories/031103/sta_fishregs.shtml
Herrman, M. (1993). Using an international econometric model to forecast Alaska Salmon
Revenues. Mar. Resour. Econ. 8, 249–272.
Hershberger, W. K., Myers, J. M., Iwamoto, R. N., McAuley, W. C. and Saxton, A. M.
(1990). Genetic changes in the growth of coho salmon (Oncorhynchus kisutch) in marine
net-pens, produced by 10 years of selection. Aquaculture 85, 187–197.
Hilborn, R. (1976). Optimal exploitation of multiple stocks by a common fishery: A new
methodology. J. Fish. Res. Board Canada 33, 1–5.
Hilborn, R. (1985a). Apparent stock recruitment relationships in mixed stock fisheries. Can.
J. Fish. Aquat. Sci. 42, 718–723.
Hilborn, R. (1985b). A comparison of harvest policies for mixed stock fisheries. In
‘‘Resource Management: Proceedings of the Second Ralf Yorque Workshop’’
(M. Mangel, ed.), Vol. 61, pp. 75–87. Springer-Verlag, Berlin.
Hilborn, R. (1992a). Hatcheries and the future of salmon in the northwest. Fisheries 17, 5–8.
Hilborn, R. (1992b). Institutional learning and spawning channels for sockeye salmon
(Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 49, 1126–1136.
Hilborn, R. (1999). Confessions of a reformed hatchery basher. Fisheries 24, 30–31.
Hilborn, R. and Eggers, D. (2000). A review of the hatchery programmes for pink salmon in
Prince William Sound and Kodiak Island, Alaska. Trans. Am. Fish. Soc. 129, 333–350.
Hilborn, R. and Eggers, D. (2001). A review of the hatchery programmes for pink salmon in
Prince William Sound and Kodiak Island, Alaska: Response to comment. Trans.
Am. Fish. Soc. 130, 720–724.
Hilborn, R. and Winton, J. (1993). Learning to enhance salmon production—Lessons from
the salmonid enhancement programme. Can. J. Fish. Aquat. Sci. 50, 2043–2056.
Hindar, K., Ryman, N. and Utter, F. (1991). Genetic effects of cultured fish on natural fish
populations. Can. J. Fish. Aquat. Sci. 48, 945–957.
Hoffman, G. L. (1999). ‘‘Parasites of North American Freshwater Fishes,’’ (2nd edn.)
Cornell University Press, Ithaca, NY.
Hume, R. D. (1893). ‘‘Salmon of the Pacific Coast.’’ Schmidt Label & Lithographic Co.,
San Francisco.
Hunter, J. G. (1959). Survival and production of pink and chum salmon in a coastal stream.
J. Fish. Res. Board Canada 16, 835–885.
Huntingford, F. A. and Garcia de Leaniz, C. (1997). Social dominance, prior residence and
the acquisition of profitable feeding sites in juvenile Atlantic salmon. J. Fish Biol. 51,
1009–1014.
Hyatt, K. D. and Stockner, J. G. (1985). Responses of sockeye salmon (Oncorhynchus nerka)
to fertilization of British Columbia coastal lakes. Can. J. Fish. Aquat. Sci. 42, 320–331.
ICES (2002). ‘‘Report of the Baltic Salmon and Trout Assessment Working Group.’’ ICES,
Copenhagen, Denmark.
ICES (2005).‘‘Report of the Baltic Salmon and Trout Assessment Working Group
(WGBAST),’’ 5–14 April 2005. Helsinki, Finland, ICES Copenhagen, Denmark.
IEAB (2002). ‘‘Artificial Production Review Phase I: Economic Analysis.’’ Northwest
Power and Conservation Council, Portland, OR. http://www.nwcouncil.org/library/
ieab/ieab2002-1.pdf
Imperial Fisheries Bureau (1904). ‘‘World’s Fair, St Louis, 1904: Japan Special Catalogue:
Fishing Industry Descriptive Catalogue.’’ Imperial Fisheries Bureau, Department of
Agriculture and Commerce, Tokyo.
180
Kerry A. Naish et al.
ISAB (2002). (Independent Scientific Advisory Board). Hatchery surpluses in the Pacific
Northwest. Fisheries 27, 16–27.
ISRP/IEAB (2005). ‘‘Review of the Select Area Fishery Evaluation Project 1993–2003,
Final Project Completion Report (SAFE Review).’’ Rep. No. ISRP/IEAB 2005-8.
Independent Scientific Review Panel/Independent Economic Analysis Board, Pacific
Northwest Power and Conservation Council, Portland, OR. http://www.nwcouncil.
org/library/.
Isaksson, A. (1988). Salmon ranching: A world review. Aquaculture 75, 1–33.
Isaksson, A. and Oskarsson, S. (2002). Icelandic salmon ranching: Problems and policy issues—
An historical perspective. ICES Annual Science Conference, Copenhagen, Denmark.
Isaksson, A., Pikitch, E. K., Huppert, D. D. and Sissenwine, M. P. (1997). Sea ranching of
Atlantic salmon with special reference to private ranching in Iceland. In ‘‘Global Trends:
Fisheries Management’’ (E. K. Pikitch, D. D. Huppert and M. P. Sissenwine, eds),
pp. 166–174. American Fisheries Society, Bethesda, Maryland.
Iwama, G. and Nakanishi, T. (1996). ‘‘The Fish Immune System: Organism, Pathogen, and
Environment.’’ Academic Press, San Diego, CA.
Jacobs, S. E. (1988). ‘‘Straying in Oregon by Adult Salmon of Hatchery Origin, 1985.’’
Oregon Department of Fish and Wildlife, Portland.
Jacoby, K. (2001). ‘‘Crimes Against Nature: Squatters, Poachers, Thieves, and the Hidden
History of American Conservation.’’ University of California Press, Berkeley.
Jansson, H. and Oest, T. (1997). Hybridization between Atlantic salmon (Salmo salar) and
brown trout (S. trutta) in a restored section of the River Dalaelven, Sweden. Can. J. Fish.
Aquat. Sci. 54, 2033–2039.
Johnsen, B. O. and Jensen, A. J. (1986). Infestations of Atlantic salmon, Salmo salar, by
Gyrodactylus salaris in Norwegian rivers. J. Fish Biol. 29, 233–241.
Johnson, O. S., Grant, W. S., Kope, R. G., Neely, K., Waknitz, F. W. and Waples, R. S.
(1997). Status Review of Chum Salmon from Washington, Oregon, and California.
Rep. No. NOAA Technical Memorandom NMFS-NWFSC-32. U.S. Department of
Commerce, Seattle.
Johnstone, K. (1977). ‘‘The Aquatic Explorers: A History of the Fisheries Research Board of
Canada.’’ University of Toronto Press, Toronto.
Jones, M. W., Danzmann, R. G. and Clay, D. (1997). Genetic relationships among populations
of wild resident, and wild and hatchery anadromous brook charr. J. Fish Biol. 51, 29–40.
Jonsson, B., Fleming, I. A. and Sundes, G. (1993). Enhancement of wild salmon populations.
In ‘‘Human Impact on Self-Recruiting Populations’’ (G. Sundes, ed.), pp. 209–238.
The Royal Norwegian Society of Sciences and Letters Foundation, Tapir Publishers,
Trondheim, Norway.
Jorgensen, J. and Berg, S. (1991). Stocking experiments with 0þ and 1þ trout parr, Salmo trutta
L., of wild and hatchery origin: 2. Post-stocking movements. J. Fish Biol. 39, 171–180.
Kaeriyama, M. (1998). Dynamics of chum salmon, Oncorhynchus keta, populations released
from Hokkaido, Japan. North Pac. Anadr. Fish Comm. Bull. 1, 90–102.
Kalinowski, S. T. and Hedrick, P. W. (1999). Detecting inbreeding depression is difficult in
captive endangered species. Anim. Conserv. 2, 131–136.
Kanda, N., Leary, R. F. and Allendorf, F. W. (2002). Evidence of introgressive hybridization
between bull trout and brook trout. Trans. Am. Fish. Soc. 131, 772–782.
Keeley, E. R. and Grant, J. W. A. (2001). Prey size of salmonid fishes in streams, lakes, and
oceans. Can. J. Fish. Aquat. Sci. 58, 1122–1132.
Keller, L. F. and Waller, D. M. (2002). Inbreeding effects in wild populations. Trends Ecol.
Evol. 17, 230–241.
Kerans, B. L., Rasmussen, C., Stevens, R., Colwell, A. E. L. and Winton, J. R. (2004).
Differential propagation of the metazoan parasite Myxobolus cerebralis by Limnodrilus
Evaluation of the Effect of Hatcheries on Wild Salmon
181
hoffmeisttri, Hyodrilus templetoni, and genetically distinct strains of Tubifex tubifex. J. Parasitol. 90, 1366–1373.
Khristoforov, O. L. and Murza, I. G. (2003). Status of populations and reproduction of
Atlantic salmon in the Russian part of the Baltic Sea. In ‘‘Atlantic Salmon: Biology,
Conservation and Restoration’’ (A. J. Veselov, E. P. Ieshko, N. N. Nemova,
O. P. Sterligova and Y. A. Shustov, eds), pp. 52–60. Russian Academy of Sciences,
Karelian Research Center, Institute of Biology, Petrozavodsk.
King, T. L., Kalinowski, S. T., Schill, W. B., Spidle, A. P. and Lubinski, B. A. (2001).
Population structure of Atlantic salmon (Salmo salar L.): A range-wide perspective from
microsatellite DNA variation. Mol. Ecol. 10, 807–821.
Kinsey, D. (2006). ‘Seeding the water as the earth’: The epicenter and peripheries of a
western aquaculture revolution. Environ. Hist. 11, 527–566.
Kocan, R. M., Hershberger, P. K. and Winton, J. (2004). Ichthyophoniasis: An emerging
disease of Chinook salmon in the Yukon River. J. Aquat. Anim. Health 16, 58–72.
Koenings, J. (2000).Commencement Address, University of Washington, College of Forest
Resources, 9 June 2000http://wdfw.wa.gov/depinfo/director/jun0900a.htm.
Koenings, J. P., Geiger, H. J. and Hasbrouck, J. J. (1993). Smolt-to-adult survival patterns of
sockeye salmon (Oncorhynchus nerka): Effects of smolt length and geographic latitude
when entering the sea. Can. J. Fish. Aquat. Sci. 50, 600–611.
Koljonen, M. L., Jansson, H., Paaver, T., Vasin, O. and Koskiniemi, J. (1999). Phylogeographic lineages and differentiation pattern of Atlantic salmon (Salmo salar) in the Baltic
Sea with management implications. Can. J. Fish. Aquat. Sci. 56, 1766–1780.
Koljonen, M. L., Tahtinen, J., Saisa, M. and Koskiniemi, J. (2002). Maintenance of genetic
diversity of Atlantic salmon (Salmo salar) by captive breeding programmes and the
geographic distribution of microsatellite variation. Aquaculture 212, 69–92.
Kope, R. G. (1992). Optimal harvest rates for mixed stocks of natural and hatchery fish. Can.
J. Fish. Aquat. Sci. 49, 931–938.
Kostow, K. E. (2004). Differences in juvenile phenotypes and survival between hatchery
stocks and a natural population provide evidence for modified selection due to captive
breeding. Can. J. Fish. Aquat. Sci. 61, 577–589.
Kostow, K. E., Marshall, A. R. and Phelps, S. R. (2003). Naturally spawning hatchery
steelhead contribute to smolt production but experience low reproductive success. Trans.
Am. Fish. Soc. 132, 780–790.
Krkosek, M., Lewis, M. A. and Volpe, J. P. (2005). Transmission dynamics of parasitic sea
lice from farm to wild salmon. Proc. R. Soc. Ser. B Biol. Sci. 272, 689–696.
Kurath, G., Garver, K. A., Troyer, R. M., Emmenegger, E. J., Einer-Jensen, K. and
Anderson, E. D. (2003). Phylogeography of infectious haematopoietic necrosis virus in
North America. J. Gen. Virol. 84, 803–814.
Labelle, M. (1992). Straying patterns of coho salmon (Oncorhynchus kisutch) stocks from
southeast Vancouver Island, British Columbia. Can. J. Fish. Aquat. Sci. 49, 1843–1855.
Lackey, R. T., Lach, D. H. and Duncan, S. L. (2006). Policy options to reverse the decline of
wild pacific salmon. Fisheries 31, 344–350.
Laikre, L. (1999). ‘‘Conservation Genetic Management of Brown Trout (Salmo trutta) in
Europe. Report of the Concerted Action on Identification, Management and
Exploitation of Genetic Resources in the Brown Trout (Salmo trutta).’’ Rep. No.
TROUTCONCERT EUFAIR CT97-3882.
Lampman, B. H. (1946). ‘‘The Coming of the Pond Fishes; An Account of the Introduction
of Certain Spiny-Rayed Fishes, and Other Exotic Species, into the Waters of the Lower
Columbia River Region and the Pacific Coast states.’’ Binfords & Mort, Portland.
Langston, N. (2003). ‘‘Where Land & Water Meet: A Western Landscape Transformed.’’
University of Washington Press, Seattle.
182
Kerry A. Naish et al.
Largiader, C. R. and Scholl, A. (1995). Effects of stocking on the genetic diversity of brown
trout populations of the Adriatic and Danubian drainages in Switzerland. J. Fish Biol.
47A, 209–225.
Largiader, C. R. and Scholl, A. (1996). Genetic introgression between native and introduced
brown trout Salmo trutta L populations in the Rhone River Basin. Mol. Ecol. 5, 417–426.
Larkin, P. A. (1974). Play it again Sam—Essay on salmon enhancement. J. Fish. Res. Board
Canada 31, 1433–1448.
Larsen, D. A., Beckman, B. R., Cooper, K. A., Barrett, D., Johnston, M., Swanson, P. and
Dickhoff, W. W. (2004). Assessment of high rates of precocious male maturation in a
spring Chinook salmon supplementation hatchery programme. Trans. Am. Fish. Soc. 133,
98–120.
Lasenby, T. R., Kerr, S. J. and Hooper, G. W. (2001). ‘‘Lake Whitefish Culture and
Stocking: An Annotated Bibliography and Literature Review.’’ Fish and Wildlife
Branch, Ontario Ministry of Natural Resources, Peterborough, Ont.
Lawlor, J. L. and Hutchings, J. A. (2004). Consequences of fitness related traits on hybridization between farmed and wild Atlantic salmon, Salmo salar. J. Fish Biol. 65, p. 319.
Lawson, P. W. and Sampson, D. B. (1996). Gear-related mortality in selective fisheries for
ocean salmon. North Am. J. Fish. Manag. 16, 512–520.
Leary, R. F., Allendorf, F. W. and Sage, G. K. (1995). Hybridization and introgression
between introduced and native fish. Am. Fish. Soc. Symp. 15, 91–101.
Leatherland, J. F. and Woo, P. T. K. (1998). ‘‘Fish Diseases and Disorders, Volume 2. NonInfectious Disorders.’’ CAB International, New York.
Leber, K. M., Kitada, S., Blankenship, H. L. and Svasand, T. (2005a). Preface. In ‘‘Stock
Enhancement and Sea Ranching’’ (K. M. Leber, S. Kitada, H. L. Blankenship and
T. Svasand, eds), pp. ix–xii. Blackwell, Oxford.
K. M. Leber, T. S. Kitada, T. H. L. Blankenship, T. and T. Svasand, T. ( eds), (2005b).
‘‘Stock Enhancement and Sea Ranching,’’ pp. 1–562. Blackwell, Oxford.
Lee, K. N. (1993). ‘‘Compass and Gyroscope: Integrating Science and Politics for the
Environment.’’ Island Press, Washington, DC.
Leider, S. A., Hulett, P. L., Loch, J. J. and Chilcote, M. W. (1990). Electrophoretic
comparison of the reproductive success of naturally spawning transplanted and wild
steelhead trout through the returning adult stage. Aquaculture 88, 239–252.
Leong, J. C. and Fryer, J. L. (1993). Viral vaccines for aquaculture. Annu. Rev. Fish Dis. 3,
225–240.
Levin, P. S. and Schiewe, M. H. (2001). Preserving salmon biodiversity. Am. Sci. 89,
220–227.
Levin, P. S. and Williams, J. G. (2002). Interspecific effects of artificially propagated fish:
An additional conservation risk for salmon. Conserv. Biol. 16, 1581–1587.
Levin, P. S., Zabel, R. W. and Williams, J. G. (2001). The road to extinction is paved with
good intentions: Negative association of fish hatcheries with threatened salmon. Proc. R.
Soc. Lond. Ser. B Biol. Sci. 268, 1153–1158.
Lichatowich, J. (1999). ‘‘Salmon Without Rivers: A History of the Pacific Salmon Crisis.’’
Island Press, Washington, DC.
Lichatowich, J. A. and McIntyre, J. D. (1987). Uses of hatcheries in the management of
Pacific anadromous salmonids. Am. Fish. Soc. Symp. 1, 131–136.
Loomis, J. (1996). Measuring the economic benefits of removing dams and restoring the
Elwha River: Results of a contingent valuation survey. Water Resour. Res. 32, 441–447.
Loomis, J. B. and Fix, P. (1999). A statistical approach to estimating costs of propagating
hatchery rainbow trout. North Am. J. Fish. Manag. 19, 110–119.
Lynch, M. (1991). The genetic interpretation of inbreeding depression and outbreeding
depression. Evolution 45, 622–629.
Evaluation of the Effect of Hatcheries on Wild Salmon
183
Lynch, M. and O’Hely, M. (2001). Captive breeding and the genetic fitness of natural
populations. Conserv. Genet. 2, 363–378.
Lynch, M. and Walsh, B. (1998). ‘‘Genetics and Analysis of Quantitative Traits.’’ Sinauer
Associates, Sunderland.
MacCrimmon, H. (1965). The beginning of salmon culture in Canada. Can. Geogr. J. 71,
96–103.
Machordom, A., Garcia-Marin, J. L., Sanz, N., Almodovar, A. and Pla, C. (1999). Allozyme
diversity in brown trout (Salmo trutta) from Central Spain: Genetic consequences of
restocking. Freshwater Biol. 41, 707–717.
Machordom, A., Suarez, J., Almodovar, A. and Bautista, J. M. (2000). Mitochondrial
haplotype variation and phylogeography of Iberian brown trout populations. Mol. Ecol.
9, 1324–1338.
Mackey, G., McLean, J. E. and Quinn, T. P. (2001). Comparisons of run timing, spatial
distribution, and length of wild and newly established hatchery populations of steelhead
in Forks Creek, Washington. North Am. J. Fish. Manag. 21, 717–724.
Magnusson, A. (2002). Survival rates of coho (Oncorhynchus kisutch) and chinook salmon
(O. tshawytscha) released from hatcheries on the U.S. and Canadian Pacific Coast, 1972–
1998, with respect to climate and habitat effects. p. 104. Master of Science thesis,
University of Washington, Seattle.
Mahnken, C., Ruggerone, G., Waknitz, W. and Flagg, T. (1998). A historical perspective
on salmonid production from Pacific Rim hatcheries. North Pac. Anadr. Fish. Comm. Bull.
1, 38–53.
Margan, S. H., Nurthen, R. K., Montgomery, M. E., Woodworth, L. M., Lowe, E. H.,
Briscoe, D. A. and Frankham, R. (1998). Single large or several small? Population
fragmentation in the captive management of endangered species. Zoo Biol. 17, 467–480.
Marsh, G. P. (1857). ‘‘Report Made Under the Authority of the Legislature of Vermont, on
the Artificial Propagation of Fish.’’ Free Press, Burlington.
Marshall, A. R., Blankenship, H. L. and Connor, W. P. (2000). Genetic characterization of
naturally spawned Snake River fall-run Chinook salmon. Trans. Am. Fish. Soc. 129,
680–698.
Marzano, F. N., Corradi, N., Papa, R., Tagliavini, J. and Gandolfi, G. (2003). Molecular
evidence for introgression and loss of genetic variability in Salmo (trutta) macrostigma as a
result of massive restocking of Apennine populations (Northern and Central Italy).
Environ. Biol. Fishes 68, 349–356.
Mason, J. C. (1976). Response of underyearling coho salmon to supplemental feeding in a
natural stream. J. Wildl. Manag. 40, 775–788.
Mathews, S. B. (1980). Trends in puget sound and Columbia River salmon. In ‘‘Salmonid
Ecosystems of the North Pacific’’ (W. J. McNeil and D. C. Himsworth, eds),
pp. 133–145. Oregon State University Press, Corvallis.
Matthews, M. A., Poole, W. R., Thompson, C. E., McKillen, J., Ferguson, A., Hindar, K.
and Wheelan, K. F. (2000). Incidence of hybridization between Atlantic salmon, Salmo
salar L., and brown trout, Salmo trutta L., in Ireland. Fish. Manag. Ecol. 7, 337–347.
McClelland, E. K. and Naish, K. A. (2007). What is the fitness outcome of crossing unrelated
fish populations? A meta-analysis and an evaluation of future research directions. Conserv.
Genet. 8, 397–416.
McClelland, E. K., Myers, J. M., Hard, J. J., Park, L. K. and Naish, K. A. (2005). Two
generations of outbreeding in coho salmon (Oncorhynchus kisutch): Effects on size and
growth. Can. J. Fish. Aquat. Sci. 62, 2538–2547.
McDonald, J. (1981). The stock concept and its application to British Columbia salmon
fisheries. Can. J. Fish. Aquat. Sci. 38, 1657–1664.
McElhany, P., Ruckelshaus, M. H., Ford, M. J., Wainwright, T. C. and Bjorkstedt, E. P.
(2000).‘‘Viable Salmon Populations and the Recovery of Evolutionarily Significant
184
Kerry A. Naish et al.
Units,’’156NOAA Technical Memorandum, NMFS-NWFSC-58http://www.nwfsc.
noaa.gov/publications
McEvoy, A. F. (1986). ‘‘The Fisherman’s Problem: Ecology and Law in the California
Fisheries, 1850–1980.’’ Cambridge University Press, New York.
McGinnity, P., Stone, C., Taggart, J. B., Cooke, D., Cotter, D., Hynes, R., McCamley, C.,
Cross, T. and Ferguson, A. (1997). Genetic impact of escaped farmed Atlantic salmon
(Salmo salar L.) on native populations: Use of DNA profiling to assess freshwater
performance of wild, farmed, and hybrid progeny in a natural river environment. ICES
J. Mar. Sci. 54, 998–1008.
McGinnity, P., Prodohl, P., Ferguson, K., Hynes, R., O’Maoileidigh, N., Baker, N.,
Cotter, D., O’Hea, B., Cooke, D., Rogan, G., Taggart, J. and Cross, T. (2003). Fitness
reduction and potential extinction of wild populations of Atlantic salmon, Salmo salar, as a
result of interactions with escaped farm salmon. Proc. R. Soc. Lond. Ser. B Biol. Sci. 270,
2443–2450.
McGinnity, P., Prodohl, P., Maoileidigh, N. O., Hynes, R., Cotter, D., Baker, N.,
O’Hea, B. and Ferguson, A. (2004). Differential lifetime success and performance of
native and non-native Atlantic salmon examined under communal natural conditions.
J. Fish Biol. 65, 173–187.
McGowan, C. and Davidson, W. S. (1992). Unidirectional natural hybridization between
brown trout (Salmo trutta) and Atlantic salmon (S. salar) in Newfoundland. Can. J. Fish.
Aquat. Sci. 49, 1953–1958.
McHugh, J. L. (1970). Trends in fishery research. In ‘‘A Century of Fisheries in North
America’’ (N. G. Benson, ed.), pp. 25–56. American Fisheries Society, Washington.
McKinnell, S. (1995). Age-specific effects of sockeye abundance on adult body size of
selected British Columbia sockeye stocks. Can. J. Fish. Aquat. Sci. 52, 1050–1063.
McPhee, J. A. (2002). ‘‘The Founding Fish,’’ (3rd edn.) Farrar, Straus and Giroux,
New York.
McVicar, A. H. (1997). Disease and parasite implications of the coexistence of wild and
cultured Atlantic salmon populations. ICES J. Mar. Sci. 54, 1093–1103.
Mead, R. W. and Woodall, W. L. (1968). ‘‘Comparison of Sockeye Salmon Fry produced
by Hatcheries, Artificial Channels and Natural Spawning Areas.’’ Rep. No. 20.
International Pacific Salmon Fisheries Commission Progress Report.
Meffe, G. K. (1992). Techno-arrogance and halfway technologies: Salmon hatcheries on the
Pacific coast of North America. Conserv. Biol. 6, 350–354.
Mezzera, M. and Largiader, C. R. (2001a). Comparative analysis of introgression at three
marker classes: A case study in a stocked population of brown trout. J. Fish Biol. 59,
289–305(sa).
Mezzera, M. and Largiader, C. R. (2001b). Evidence for selective angling of introduced
trout and their hybrids in a stocked brown trout population. J. Fish Biol. 59, 287–301.
Miller, K. M., Winton, J. R., Schulze, A. D., Purcell, M. K. and Ming, T. J. (2004). Major
histocompatibility complex loci are associated with susceptibility of Atlantic salmon to
infectious hematopoietic necrosis virus. Environ. Biol. Fish. 69, 307–316.
Milner, J. W. (1874). The progress of fish culture in the United States. In ‘‘Annual Report
of the Commissioner of Fish and Fisheries, 1872–1873,’’ pp. 524–527. Government
Printing Office, Washington, DC.
Milner, N. J., Russell, I. C., Aprahamian, M., Inverarity, R., Shelley, J. and Rippon, P.
(2004). In ‘‘The Role of Stocking in Recovery of River Tyne Salmon Fisheries.’’
Fisheries Technical Report No. 2004.1. Environment Agency, National Fisheries Technical Team, Cardiff.
Milstein, M. (2003). A tale of boom and bust. In ‘‘Portland Oregonian,’’ Vol. 21, p. A1A4.
Mitchum, D. L. and Sherman, L. E. (1981). Transmission of bacterial kidney disease from
wild to stocked hatchery trout. Can. J. Fish. Aquat. Sci. 38, 547–551.
Evaluation of the Effect of Hatcheries on Wild Salmon
185
Mobrand, L. E., Barr, J., Blankenship, L., Campton, D. E., Evelyn, T. T. P., Flagg, T. A.,
Mahnken, C. V. W., Seeb, L. W., Seidel, P. R. and Smoker, W. W. (2005). Hatchery
reform in Washington state: Principles and emerging issues. Fisheries 30, 11–23.
Mork, C., Hershberger, P., Kocan, R., Batts, W. and Winton, J. (2004). Isolation and
characterization of a rhabdovirus from starry flounder (Platichthys stellatus) collected from
the northern portion of Puget Sound, Washington, USA. J. Gen. Virol. 85, 495–505.
Morton, A., Routledge, R., Peet, C. and Ladwig, A. (2004). Sea lice (Lepeophtheirus salmonis)
infection rates on juvenile pink (Oncorhynchus gorbuscha) and chum (Oncorhynchus keta)
salmon in the nearshore marine environment of British Columbia, Canada. Can. J. Fish.
Aquat. Sci. 61, 147–157.
Mulcahy, D., Pascho, R. J. and Jenes, C. K. (1983). Detection of infectious haematopoietic
necrosis virus in river water and demonstration of waterborne transmission. J. Fish Dis. 6,
321–330.
Mundie, J. H., McKinnell, S. M. and Traber, R. E. (1983). Responses of stream zoobenthos
to enrichment of gravel substrates with cereal grain and soybean. Can. J. Fish. Aquat. Sci.
40, 1702–1712.
Mustafa, S. (2003). Stock enhancement and sea ranching: Objectives and potential. Rev. Fish
Biol. Fish. 13, 141–149.
Myers, J. M., Heggelund, P. O., Hudson, G. and Iwamoto, R. N. (2001). Genetics and
broodstock management of coho salmon. Aquaculture 197, 43–62.
Myers, J. M., Kope, R. G., Bryant, G. J., Teel, D. J., Lierheimer, L. J., Wainwright, T. C.,
Grant, W. S., Waknitz, F. W., Neely, K., Lindley, S. and Waples, R. S. (1998). ‘‘Status
review of Chinook salmon from Washington, Idaho, Oregon, and California’’. NOAA
Technical Memorandum, NMFS-NWFSC-35 US Department of Commerce, Seattle.
Myers, R. A., Levin, S. A., Lande, R., James, F. C., Murdoch, W. W. and Paine, R. T.
(2004). Hatcheries and endangered salmon. Science 303, p. 1980.
National Research Council (1996). ‘‘Upstream: Salmon and Society in the Pacific Northwest.’’ National Academy Press, Washington, DC.
Naylor, R. L., Hindar, K., Fleming, I. A., Goldberg, R., Mangel, M., Williams, S., Volpe, J.,
Whoriskey, F., Eagle, J. and Kelso, D. (2005). Fugitive salmon: Assessing risks of escaped
farmed fish from aquaculture. Bioscience 55, 427–437.
Neilson, J. D. and Geen, G. H. (1986). First-year growth rate of Sixes River Chinook
salmon as inferred from otoliths: Effects on mortality and age at maturity. Trans. Am. Fish.
Soc. 115, 28–33.
Netboy, A. (1968). ‘‘The Atlantic Salmon: A Vanishing Species?.’’ Houghton Mifflin,
Boston.
Netboy, A. (1980). ‘‘Salmon, the World’s Most Harassed Fish.’’ Winchester Press, Tulsa.
Newell, D. (1993). ‘‘Tangled Webs of History: Indians and the Law in Canada’s Pacific
Coast Fisheries.’’ University of Toronto Press, Toronto.
Newman, S. G. (1993). Bacterial vaccines for fish. Annu. Rev. Fish Dis. 3, 145–185.
Nicholas, J. W. and Downey, T. W. (1983). ‘‘Coastal Chinook Salmon Studies, 1980–83.’’
Rep. No. 81-ABD-ORIE and 82-ABD-ORIE. Oregon Department of Fish and
Wildlife, Portland.
Nicholas, J. W. and Van Dyke, L. (1982). ‘‘Straying of Adult Coho Salmon to and from a
Private Hatchery at Yaquina Bay, Oregon.’’ Rep. No. 82-10. Oregon Department of
Fish and Wildlife, Portland.
Nichols, K. M., Bartholomew, J. and Thorgaard, G. H. (2003). Mapping multiple genetic
loci associated with Ceratomyxa shasta resistance in Oncorhynchus mykiss. Dis. Aquat.
Organ. 56, 145–154.
Nickelson, T. E., Solazzi, M. F. and Johnson, S. L. (1986). Use of hatchery coho salmon
(Oncorhynchus kisutch) presmolts to rebuild wild populations in Oregon coastal streams.
Can. J. Fish. Aquat. Sci. 43, 2443–2449.
186
Kerry A. Naish et al.
Nielsen, J. L. (1992). Microhabitat-specific foraging behavior, diet, and growth of juvenile
coho salmon. Trans. Am. Fish. Soc. 121, 617–634.
Nielsen, J. L., Carpanzano, C., Fountain, M. C. and Gan, C. A. (1997). Mitochondrial DNA
and nuclear microsatellite diversity in hatchery and wild Oncorhynchus mykiss from
freshwater habitats in southern California. Trans. Am. Fish. Soc. 126, 397–417.
Nielsen, E. E., Hansen, M. M. and Loeschcke, V. (1999). Genetic variation in time and
space: Microsatellite analysis of extinct and extant populations of Atlantic salmon.
Evolution 53, 261–268.
Noga, E. J. (2000). ‘‘Fish Disease: Diagnosis and Treatment.’’ Iowa State University Press/
Ed. Iowa State University Press, Ames.
Norris, T. (1868). ‘‘American Fish-Culture, Embracing All the Details of Artificial Breeding
and Rearing of Trout; the Culture of Salmon, Shad and other Fishes.’’ Porter & Coates,
Philadelphia.
Norris, J. G., Hyun, S.-Y. and Anderson, J. J. (2000). Ocean distribution of Columbia
River upriver bright fall chinook salmon stocks. North Pac. Anadr. Fish Comm. Bull. 2,
221–232.
Nowak, G. M., Tabor, R. A., Warner, E. J., Fresh, K. L. and Quinn, T. P. (2004).
Ontogenetic shifts in habitat and diet of cutthroat trout in Lake Washington, Washington. North Am. J. Fish. Manag. 24, 624–635.
Office International des Epizooties (2003). ‘‘Manual of Diagnostic Tests for Aquatic Animals.’’ Office International des Epizooties, Paris.
Olivier, G. (2002). Disease interactions between wild and cultured fish—Perspectives from
the American Northeast (Atlantic Provinces). Bull. Eur. Assoc. Fish Pathol. 22, 103–109.
Oregon Secretary of State (2002). ‘‘Department of Fish and Wildlife: Hatchery CostEffectiveness State Fiscal Years 1994–1997. Audit Report No. 2002–39.’’ Oregon Audits
Division, Salem, OR.
Orr, C., Gallagher, P. and Penikett, J. (2002). ‘‘Hatcheries and the Protection of Wild
Salmon.’’ Simon Fraser University, Burnaby.
Otterstad, O. (1998). The exclusion of professional fisherman from Norwegian salmon
fishing: The distribution effects of fisheries management. In ‘‘Property Rights and
Regulatory Systems in Fisheries’’ (D. Symes, ed.), pp. 101–102. Fishing News Books,
Malden.
Paaver, T., Kupper, P. and Kangur, M. (2003). The effect of stocking on weak and
endangered salmon populations in Estonia. In ‘‘Atlantic Salmon: Biology, Conservation
and Restoration’’ (A. J. Veselov, E. P. Ieshko, N. N. Nemova, O. P. Sterligova and
Y. A. Shustov, eds), pp. 88–93. Russian Academy of Sciences, Karelian Research Center,
Institute of Biology, Petrozavodsk.
Pacific Salmon Commission Joint Chinook Technical Committee (2002). In ‘‘Annual
Exploitation Rate Analysis and Model Calibration, Report TCCHINOOK (02)-3.’’
Pacific Salmon Commission, Vancouver, BC, CanadaRep. No. TCCHINOOK (02)-3.
Palm, S., Dannewitz, J., Jarvi, T., Petersson, E., Prestegaard, T. and Ryman, N. (2003). Lack
of molecular genetic divergence between sea-ranched and wild sea trout (Salmo trutta).
Mol. Ecol. 12, 2057–2071.
Pante, M. J. R., Gjerde, B. and McMillan, I. (2001). Effect of inbreeding on body weight at
harvest in rainbow trout, Oncorhynchus mykiss. Aquaculture 192, 201–211.
Parent, G. (2003).‘‘Managing Marine Fisheries in a Changing Climate: Approaches for
Achieving Fisheries Sustainability.’’Western Division of American Fisheries Society, San
Diego15 April 2003. http://www.dfait-maeci.gc.ca/sustain/ambassador/sandie03-en.asp
Parenteau, B. (1998). ‘Care, control and supervision’: Native people in the Canadian
Atlantic salmon fishery, 1867–1900. Can. Hist. Rev. 79, 1–35.
Parker, R. R. (1971). Size selective predation among juvenile salmonid fishes in a British
Columbia inlet. J. Fish. Res. Board Can. 28, 1503–1510.
Evaluation of the Effect of Hatcheries on Wild Salmon
187
Parrish, D. L., Behnke, R. J., Gephard, S. R., McCormick, S. D. and Reeves, G. H. (1998).
Why aren’t there more Atlantic salmon (Salmo salar)? Can. J. Fish. Aquat. Sci. 55, (S1),
281–287.
Paulik, G. J., Hourston, A. S. and Larkin, P. A. (1967). Exploitation of multiple stocks by a
common fishery. J. Fish. Res. Board Can. 24, (12), 2527–2537.
Paulsen, C. M. and Wernstedt, K. (1995). Cost-effectiveness analysis for complex managed
hydrosystems: An application to the Columbia River Basin. J. Environ. Econ. Manag. 28,
388–400.
Pearse, P. H. (1994). ‘‘Salmon Enhancement: An Assessment of the Salmon Stock Development Programme on Canada’s Pacific Coast. Final Report of the Programme Review of
the Salmonid Enhancement Programme Internal Audit and Evaluation Branch.’’ Department of Fisheries and Oceans Canada, Ottawa, Canada.
Pearsons, T. N. and Fritts, A. L. (1999). Maximum size of chinook salmon consumed by
juvenile coho salmon. North Am. J. Fish. Manag. 19, 165–170.
Peeler, E. J. and Murray, A. G. (2004). Disease interaction between farmed and wild fish
populations. J. Fish Biol. 65, (Suppl. 1), 321–322.
Peeler, E. J. and Thrush, M. A. (2004). Qualitative analysis of the risk of introducing
Gyrodactylus salaris into the United Kingdom. Dis. Aquat. Organ. 62, 103–113.
Peterman, R. M. (1982). Nonlinear relation between smolts and adults in Babine Lake
sockeye salmon (Oncorhynchus nerka) and implications for other salmon populations. Can.
J. Fish. Aquat. Sci. 39, 904–913.
Peterson, D. G., Orr, C. and Penikett, J. (2002).A brief history of fish culture in BC and
the current status of the current provincial hatchery programme pp. 9–17. Simon Fraser
University, Burnaby, BC.
Pinkerton, E. (1994). Economic and management benefits from the coordination of capture
and culture fisheries: The case of Prince William Sound pink salmon. North Am. J. Fish.
Manag. 14, 262–277.
Pisani, D. J. (1984). Fish culture and the dawn of concern over water pollution in the United
States. Environ. Rev. 8, 117–131.
Plumb, J. A. (1999). ‘‘Health Maintenance and Principal Microbial Diseases of Cultured
Fishes,’’ (1st edn.) Iowa State University, Ames.
Poteaux, C., Beaudou, D. and Berrebi, P. (1998a). Temporal variations of genetic introgression in stocked brown trout populations. J. Fish Biol. 53, 701–713.
Poteaux, C., Bonhomme, F. and Berrebi, P. (1998b). Differences between nuclear and
mitochondrial introgressions of brown trout populations from a restocked main river and
its unrestocked tributary. Biol. J. Linnean Soc. 63, 379–392.
Poteaux, C., Berrebi, P. and Bonhomme, F. (2000). Allozymes, mtDNA and microsatellites
study introgression in a stocked trout population in France. Rev. Fish Biol. Fish. 10,
281–292.
Prat, J.-L. (1998). The status of fishery property rights in French law. In ‘‘Property Rights
and Regulatory Systems in Fisheries’’ (D. Symes, ed.), pp. 48–56. Fishing News Books,
Malden.
Prignon, C., Micha, J. C., Rimbaud, G. and Philippart, J. C. (1999). Rehabilitation efforts
for Atlantic salmon in the Meuse basin area: Synthesis 1983–1998. Hydrobiologia 410,
69–77.
Primmer, C. R., Aho, T., Piironen, J., Estoup, A., Cornuet, J. M. and Ranta, E. (1999).
Microsatellite analysis of hatchery stocks and natural populations of Arctic charr, Salvelinas
alpinus, from the Nordic region: Implications for conservation. Hereditas 130, 277–289.
Prince, E. E. (1900). Fish culture in Canada. Trans. Ottawa Literary Sci. Soc. 2, 163–182.
Pritchard, J. A. (1999). ‘‘Preserving Yellowstone’s Natural Conditions: Science and the
Perception of Nature.’’ University of Nebraska Press, Lincoln.
188
Kerry A. Naish et al.
Pritchard, S. B. (2001). ‘‘Recreating the Rhone: Nature and Technology in France Since
1945.’’ Ph.D. thesis, Stanford University.
Pyper, B. J. and Peterman, R. M. (1999). Relationship among adult body length,
abundance, and ocean temperature for British Columbia and Alaska sockeye salmon
(Oncorhynchus nerka), 1967–1997. Can. J. Fish. Aquat. Sci. 56, 1716–1720.
Quinn, T. P. (2005). ‘‘The Behavior and Ecology of Pacific Salmon and Trout.’’ University
of Washington Press, Seattle.
Quinn, T. P., Nemeth, R. S. and McIsaac, D. O. (1991). Homing and straying patterns of
fall chinook salmon in the lower Columbia river. Trans. Am. Fish. Soc. 120, 150–156.
Quinn, T. P., Peterson, J. A., Gallucci, V., Hershberger, W. K. and Brannon, E. L. (2002).
Artificial selection and environmental change: Countervailing factors affecting the timing
of spawning by coho and chinook salmon. Trans. Am. Fish. Soc. 131, 591–598.
Rahimian, H. and Thulin, J. (1996). Epizootiology of Ichthyophonus hoferi in herring populations off the Swedish west coast. Dis. Aquat. Organ. 27, 187–195.
Reed, D. H., Lowe, E. H., Briscoe, D. A. and Frankham, R. (2003). Inbreeding and
extinction: Effects of rate of inbreeding. Conserv. Genet. 4, 405–410.
Reichmann, D. (2003). Bush defends economic plan on Washington trip. In ‘‘The Olympian,’’ Vol. 23.. http://www.highbeam.com/doc/1P1-78815442.html
Reiger, J. F. (2001). ‘‘American Sportsmen and the Origins of Conservation,’’ (3rd edn.)
Oregon State University Press, Corvallis.
Reimers, P. E. (1971). ‘‘The Length of Residence of Juvenile Fall Chinook Salmon in Sixes
River, Oregon.’’ Oregon State University, Corvallis.
Reisenbichler, R. R. and McIntyre, J. D. (1977). Genetic differences in growth and survival
of juvenile hatchery and wild steelhead trout, Salmo gairdneri. J. Fish. Res. Board Canada
34, 123–128.
Reisenbichler, R. R. and Rubin, S. P. (1999). Genetic changes from artificial propagation of
Pacific salmon affect the productivity and viability of supplemented populations. ICES J.
Mar. Science 56, 459–466.
Rhodes, J. S. and Quinn, T. P. (1998). Factors affecting the outcome of territorial contests
between hatchery and naturally reared coho salmon parr in the laboratory. J. Fish Biol. 53,
1220–1230.
Rhodes, J. S. and Quinn, T. P. (1999). Comparative performance of genetically similar
hatchery and naturally-reared juvenile coho salmon in streams. North Am. J. Fish. Manage.
19, 670–677.
Rich, W. H. (1922). In ‘‘A Statistical Analysis of the Results of the Artificial Propagation of
Chinook Salmon, 1921–1922.’’ Vol. Reprint file. Northwest Fisheries Science Center
Library, National Oceanic and Atmospheric Administration, Seattle.
Ricker, W. E. (1958). Maximum sustained yields from fluctuating environments and mixed
stocks. J. Fish. Res. Board Can. 15, 991–1006.
Robbins, W. G. (2004). ‘‘Landscapes of Conflict: The Oregon Story, 1940–2000.’’ University of Washington Press, Seattle.
Roberts, R. J. (2001). ‘‘Fish Pathology,’’ (3rd edn.) Bailliere Tindall, London.
Rodriguez-Clark, K. M. (1999). Genetic theory and evidence supporting current practices
in captive breeding for conservation. In ‘‘Genetics and the Extinction of Species: DNA
and the Conservation of Biodiversity’’ (L. F. Landweber and A. P. Dobson, eds),
pp. 47–73. Princeton University Press, New Jersey.
Rogers, D. E. (1980). Density-dependent growth of Bristol Bay sockeye salmon. In
‘‘Salmonid Ecosystems of the North Pacific’’ (W. J. McNeil and D. C. Himsworth,
eds), pp. 267–283. Oregon State University Press, Corvallis.
Rogers, D. E. and Ruggerone, G. T. (1993). Factors affecting marine growth of Bristol Bay
sockeye salmon. Fish. Res. 18, 89–103.
Evaluation of the Effect of Hatcheries on Wild Salmon
189
Roni, P. and Quinn, T. P. (2001). Density and size of juvenile salmonids in response to
placement of large woody debris in western Oregon and Washington streams. Can. J.
Fish. Aquat. Sci. 58, 282–292.
Roos, J. E. (1991). ‘‘Restoring Fraser River Salmon: A History of the International Pacific
Salmon Fisheries Commission, 1937–1985.’’ The Pacific Salmon Commission,
Vancouver.
Roppel, P. (1982). ‘‘Alaska’s Salmon Hatcheries 1891–1959, Alaska Historical Commission
Studies in History, No. 20.’’ National Marine Fisheries Service, Portland, OR.
Rozwadowski, H. M. (2002). ‘‘The Sea Knows No Boundaries: A Century of Marine
Science Under ICES.’’ University of Washington Press, Seattle.
Ruggerone, G. T. and Rogers, D. E. (1984). Arctic char predation on sockeye salmon smolts
at Little Togiak River, Alaska. Fish. Bull. 82, 401–410.
Ruggerone, G. T. and Rogers, D. E. (1992). Predation on sockeye salmon fry by juvenile
coho salmon in the Chignik Lakes, Alaska: Implications for salmon management. North
Am. J. Fish. Manag. 12, 89–102.
Ruggerone, G. T., Zimmermann, M., Myers, K. W., Nielsen, J. L. and Rogers, D. E.
(2003). Competition between Asian pink salmon (Oncorhynchus gorbuscha) and Alaskan
sockeye salmon (O. nerka) in the North Pacific Ocean. Fish. Oceanogr. 12, 209–219.
Ruzzante, D. E., Hansen, M. M. and Meldrup, D. (2001). Distribution of individual
inbreeding coefficients, relatedness and influence of stocking on native anadromous
brown trout (Salmo trutta) population structure. Mol. Ecol. 10, 2107–2128.
Ruzzante, D. E., Hansen, M. M., Meldrup, D. and Ebert, K. M. (2004). Stocking impact
and migration pattern in an anadromous brown trout (Salmo trutta) complex: Where have
all the stocked spawning sea trout gone? Mol. Ecol. 13, 1433–1445.
Ryman, N. (1991). Conservation genetics in fishery management. J. Fish Biol. 39(A),
211–224.
Ryman, N. and Laikre, L. (1991). Effects of supportive breeding on the genetically effective
population-size. Conserv. Biol. 5, 325–329.
Saccheri, I. M., Kuussaari, M., Kankare, M., Vikman, P., Fortelius, W. and Hanski, I.
(1998). Inbreeding and extinction in a butterfly metapopulation. Nature 392, 491–494.
Saisa, M., Koljonen, M.-L. and Tahtinen, J. (2003). Genetic changes in Atlantic salmon
stocks since historical times and the effective population size of a long-term captive
breeding programme. Conserv. Genet. 4, 613–627.
Salminen, M. and Erkamo, R. (1998). Comparison of coastal and river releases of Atlantic
salmon smolts in the river Kokemaenjoki, Baltic Sea. ICES J. Mar. Sci. 55, 1071–1081.
Salminen, M., Kuikka, S. and Erkamo, E. (1995). Annual variability in survival of searanched Baltic salmon, Salmo salar L.: Significance of smolt size and marine conditions.
Fish. Manag. Ecol. 2, 171–184.
Schiewe, M. H., Flagg, T. A. and Berejikian, B. A. (1997). The use of captive broodstocks
for gene conservation of salmon in the Western United States. Bull. Natl. Res. Inst.
Aquacult. Suppl. 3, 29–34.
Schramm, H. L. (1996). Resurrecting hatchery bashing: A useless pastime. Fisheries 21,
26–26.
Schroeder, R. K., Lindsay, R. B. and Kenaston, K. R. (2001). Origin and straying of
hatchery winter steelhead in Oregon coastal rivers. Trans. Am. Fish. Soc. 130, 431–441.
Scribner, K. T., Page, K. S. and Bartron, M. L. (2000). Hybridization in freshwater fishes: A
review of case studies and cytonuclear methods of biological inference. Rev. Fish Biol.
Fish. 10, 293–323.
Sedgwick, S. D. (1982). ‘‘The Salmon Handbook: The Life and Cultivation of Fishes of the
Salmon Family.’’ Deutsch, London.
Seeb, J. E., Habicht, C., Templin, W. D., Seeb, L. W., Shaklee, J. B. and Utter, F. M.
(1999). Allozyme and mitochondrial DNA variation describe ecologically important
190
Kerry A. Naish et al.
genetic structure of even-year pink salmon inhabiting Prince William Sound, Alaska.
Ecol. Freshwater Fish 8, 122–140.
Seeb, L. W., Crane, P. A., Kondzela, C. M., Wilmot, R. L., Urawa, S., Varnavskaya, N. V.
and Seeb, J. E. (2004). Migration of Pacific Rim chum salmon on the high seas: Insights
from genetic data. Environ. Biol. Fishes 69, 21–36.
Shaklee, J. B., Beacham, T. D., Seeb, L. and White, B. A. (1999). Managing fisheries using
genetic data: Case studies from four species of pacific salmon. Fish. Res. 43, 45–78.
Sharma, R. and Hilborn, R. (2001). Empirical relationships between watershed characteristics and coho salmon (Oncorhynchus kisutch) smolt abundance in 14 western Washington
streams. Can. J. Fish. Aquat. Sci. 58, 1453–1463.
Shigeru, K. (1994). ‘‘Our Land was a Forest: An Ainu Memoir.’’ Westview Press, Boulder.
Shrimpton, J. M. and Heath, D. D. (2003). Census vs. effective population size in Chinook
salmon: Large- and small-scale environmental perturbation effects. Mol. Ecol. 12,
2571–2583.
Sibert, J. R. (1979). Detritus and juvenile salmon production in the Nanaimo estuary: II.
Meiofauna available as food to juvenile chum salmon (Oncorhynchus keta). J. Fish. Res.
Board Can. 36, 497–503.
Simenstad, C. A., Fresh, K. L. and Salo, E. O. (1982). The role of Puget sound and
Washington coastal estuaries in the life history of Pacific salmon: An unappreciated
function. In ‘‘Estuarine Comparisons’’ (V. S. Kennedy, ed.), pp. 343–364. Academic
Press, New York.
Skaala, O., Jorstad, K. E. and Borgstrom, R. (1996). Genetic impact on two wild brown
trout (Salmo trutta) populations after release of non-indigenous hatchery spawners. Can.
J. Fish. Aquat. Sci. 53, 2027–2035.
Smith, T. D. (1994). ‘‘Scaling Fisheries: The Science of Measuring the Effects of Fishing,
1855–1955.’’ Cambridge University Press, Cambridge.
Smoker, W. W., Wang, I. A., Gharrett, A. J. and Hard, J. J. (2004). Embryo survival and
smolt to adult survival in second-generation outbred coho salmon. J. Fish Biol. 65,
254–262.
Sonohara, T. (1997). Toward a genuine redress for an unjust past: The Nibutani dam case.
Murdoch Univ. Electr. J. Law 4, 1–43. http://www.murdoch.edu.au/elaw/issues/v4n2/
sonoha42.html
Srivastava, S., Sinha, R. and Roy, D. (2004). Toxicological effects of malachite green. Aquat.
Toxicol. 66, 319–329.
Steinberg, T. (1991). ‘‘Nature Incorporated: Industrialization and the Waters of New
England.’’ Cambridge University Press, Cambridge.
Stickney, R. R. (1996). ‘‘Aquaculture in the United States: A Historical Survey.’’ John
Wiley & Sons, New York.
Stockner, J. G. and MacIsaac, E. A. (1996). British Columbia lake enrichment programme:
Two decades of habitat enhancement for sockeye salmon. Regul. Rivers Res. Manage. 12,
547–561.
Stone, L. (1897). Some brief reminisces of the early days of fish-culture in the United States.
Bull. US Fish Comm. 17, 337–343.
Stoskopf, M. K. (1993). ‘‘Fish Medicine.’’ Saunders, Philadelphia.
Taylor, E. B. (1991). A review of local adaptation in Salmonidae, with particular reference to
Pacific and Atlantic salmon. Aquaculture 98, 185–207.
Taylor, J. E. (1998a). El Nino and vanishing salmon: Culture, nature, history, and the politics
of blame. West. Hist. Q. 29, 437–457.
Taylor, J. E. (1998b). Making salmon: The political economy of fishery science and the road
not taken. J. Hist. Biol. 31, 33–59.
Taylor, J. E. (1999a). ‘‘Making Salmon: An Environmental History of the Northwest
Fisheries Crisis.’’ University of Washington Press, Seattle.
Evaluation of the Effect of Hatcheries on Wild Salmon
191
Taylor, J. E. (1999b). ‘Politics is at the bottom of the whole thing’: Spatial relations of
power in Oregon salmon management. In ‘‘Power and Place in the North American
West’’ (R. White and J. M. Findlay, eds), pp. 233–263. University of Washington Press,
Seattle.
Taylor, J. E. (2004). Master of the seas? Herbert Hoover and the western fisheries. Oreg. Hist.
Q. 105, 40–61.
Templeton, A. R. (1986). Coadaptation and outbreeding depression. In ‘‘Conservation
Biology: The Science of Scarcity and Diversity’’ (M. E. Soule, ed.), pp. 105–116.
Sinauer, Sunderland, MA.
Tervet, D. J. (1981). The impact of fish farming on water quality. Water Pollut. Control 80,
571–581.
Tessier, N., Bernatchez, L. and Wright, J. M. (1997). Population structure and impact of
supportive breeding inferred from mitochondrial and microsatellite DNA analyses in
land-locked Atlantic salmon Salmo salar L. Mol. Ecol. 6, 735–750.
Theodorou, K. and Couvet, D. (2004). Introduction of captive breeders to the wild:
Harmful or beneficial? Conserv. Genet. 5, 1–12.
Thompson, E. P. (1975). ‘‘Whigs and Hunters: The Origin of the Black Act,’’ (1st edn.)
Pantheon Books, New York.
Thorpe, J. E. (1980). Salmon ranching in Britain. In ‘‘Salmon Ranching’’ (J. E. Thorpe, ed.),
pp. 199–211. Academic Press, London.
Thorpe, J. E. (1994). Salmonid fishes and the estuary environment. Estuaries 17, 76–93.
Tymchuck, W. E. and Devlin, R. H. (2005). Growth differences among first and second
generation hybrids of domesticated and wild rainbow trout (Oncorhynchus mykiss). Aquaculture 245, 295–300.
Tymchuck, W. E., Biagi, C. A., Whitler, R. E. and Devlin, R. H. (2006). Growth and
behavioural consequences of introgression of a domesticated aquaculture genotype into a
native strain of coho salmon. Trans. Am. Fish. Soc. 135, 442–455.
Todd, C. D., Walker, A. M., Ritchie, M. G., Graves, J. A. and Walker, A. F. (2004).
Population genetic differentiation of sea lice (Lepeophtheirus salmonis) parasitic on Atlantic
and Pacific salmonids: Analyses of microsatellite DNA variation among wild and farmed
hosts. Can. J. Fish. Aquat. Sci. 61, 1176–1190.
Tully, O., Gargan, P., Poole, W. and Whelan, K. (1999). Spatial and temporal variation in
the infectation of sea trout (Salmo trutta, L.) by the caligid copepod Lepeophtheirus salmonis
(Kroyer) in relation to sources of infection in Ireland. Parasitology 119, 41–51.
U.S. Atlantic Salmon Assessment Committee (2003). ‘‘Annual Report of the US Atlantic
Salmon Assessment Committee.’’ Rep. No. 15-2002 activities.
Unwin, M. J. and Glova, G. J. (1997). Changes in life history parameters in a naturally
spawning population of Chinook salmon (Oncorhynchus tshawytscha) associated with
releases of hatchery-reared fish. Can. J. Fish. Aquat. Sci. 54, 1235–1245.
Utter, F. (1998). Genetic problems of hatcher-reared progeny released into the wild, and
how to deal with them. Bull. Mar. Sci. 62, 623–640.
Utter, F. M. (2001). Patterns of subspecific anthropogenic introgression in two salmonid
genera. Rev. Fish Biol. Fish. 10, 265–279.
Utter, F. (2004). Population genetics, conservation and evolution in salmonids and other widely
cultured fishes: Some perspectives over six decades. Rev. Fish Biol. Fish. 14, 125–144.
Utter, F. and Epifanio, J. (2002). Marine aquaculture: Genetic potentialities and pitfalls. Rev.
Fish Biol. Fish. 12, 59–77.
Utter, F. M., Chapman, D. W. and Marshall, A. (1995). Genetic population structure and
history of chinook salmon of the upper Columbia River. In ‘‘Evolution and the Aquatic
Ecosystem: Defining Unique Units in Population Conservation’’ ( J. L. Nielsen, ed.),
Vol. 17, pp. 149–165. American Fisheries Society Symposium, Bethesda, Massachusetts.
192
Kerry A. Naish et al.
Vasemägi, A. and Primmer, C. R. (2005). Challenges for identifying functionally important
genetic variation: The promise of combining complementary research strategies. Mol.
Ecol. 14, 3623–3642.
Vasemägi, A., Gross, R., Paaver, T., Kangur, M., Nilsson, J. and Eriksson, L. O. (2001).
Identification of the origin of Atlantic salmon (Salmo salar L.) population in a recently
recolonized river in the Baltic Sea. Mol. Ecol. 10, 2877–2882.
Vasemägi, A., Gross, R., Paaver, T., Koljonen, M. L. and Nilsson, J. (2005). Extensive
immigration from compensatory hatchery releases into wild Atlantic salmon population
in the Baltic sea: Spatio-temporal analysis over 18 years. Heredity 95, 76–83.
Verspoor, E. and de Leaniz, C. G. (1997). Stocking success of Scottish Atlantic salmon in
two Spanish rivers. J. Fish Biol. 51, 1265–1269.
Vincent, E. R. (2002). Relative susceptibility of various salmonids to whirling disease with
emphasis on rainbow and cutthroat trout. ‘‘Whirling Disease: Reviews and Current
Topics, American Fisheries Society Symposium’’ (J. L. Bartholomew and J. C. Wilson,
eds), Vol. 29, pp. 109–115. American Fisheries Society, Bethesda, MD.
Wahle, R. J. and Vreeland, R. R. (1978). Bioeconomic contribution of Columbia River
hatchery fall chinook salmon, 1961 through 1964 broods, to the Pacific salmon fisheries.
Fish. Bull. 76, 179–208.
Wahle, R. J., Vreeland, R. R. and Lander, R. H. (1974). Bioeconomic contribution of
Columbia River hatchery coho salmon, 1965 and 1966 broods, to the Pacific salmon
fisheries. Fish. Bull. 72, 139–169.
Walker, B. L. (2001). ‘‘The Conquest of Ainu Lands: Ecology and Culture in Japanese
Expansion, 1590–1800.’’ University of California Press, Berkeley.
Walters, C. J. (1988). Mixed stock fisheries and the sustainability of enhancement production
for Chinook and coho salmon. In ‘‘Salmon Production, Management, and Allocation’’
(W. J. McNeil, ed.), pp. 109–116. Oregon State University Press, Corvallis, OR.
Wang, J. L. and Ryman, N. (2001). Genetic effects of multiple generations of supportive
breeding. Conserv. Biol. 15, 1619–1631.
Wang, S., Hard, J. J. and Utter, F. M. (2002). Salmonid inbreeding: A review. Rev. Fish Biol.
Fish. 11, 301–319.
Waples, R. S. (1990). Conservation genetics of Pacific salmon. 3. Estimating effective
population size. J. Hered. 81, 277–289.
Waples, R. S. (1991). Genetic interactions between hatchery and wild salmonids: Lessons
from the Pacific Northwest. Can. J. Fish. Aquat. Sci. 48, 124–133.
Waples, R. S. (1995). Evolutionarily significant units and the conservation of biological
diversity under the Endangered Species Act. In ‘‘Evolution and the Aquatic Ecosystem:
Defining Unique Units in Population Conservation Symposium 17.’’ (J. L. Nielsen, ed.),
pp. 8–27. American Fisheries Society, Bethesda, Maryland.
Waples, R. S. (1999). Dispelling some myths about hatcheries. Fisheries 24, 12–21.
Waples, R. S. (2002). Definition and estimation of effective population size in the conservation of endangered species. In ‘‘Population Viability Analyses’’ (S. R. Beissinger and
D. R. McCullough, eds), pp. 147–168. University of Chicago Press, Chicago.
Waples, R. S. (2004). Salmomid insights into effective population size. In ‘‘Evolution
Illuminated: Salmon and Their Relatives’’ (A. P. Hendry and S. C. Stearns, eds),
pp. 295–314. Oxford University Press, Oxford.
Waples, R. S. and Do, C. (1994). Genetic risk associated with supplementation of pacific
salmonids—Captive broodstock programmes. Can. J. Fish. Aquat. Sci. 51, 310–329.
Waples, R. S. and Drake, J. (2005). Risk/benefit considerations for marine stock enhancement: A Pacific salmon perspective. In ‘‘Stock Enhancement and Sea Ranching’’
(K. M. Leber, S. Kitada, H. L. Blankenship and T. Svasand, eds), pp. 260–306. Blackwell,
Oxford.
Evaluation of the Effect of Hatcheries on Wild Salmon
193
Waples, R. S., Ford, M. J. and Schmitt, D. (2007). Empirical results of salmon supplementation: a preliminary assessment. In ‘‘Ecological and Genetic Implications of Aquaculture
Activities.’’(T. Bert, ed.), pp. 383–403. Kluwer Academic Publishers, Norwell, MA.
Waples, R. S., Johnson, O. W., Aebersold, P. B., Shiflett, C. K., Van Doornik, D. M.,
Teel, D. J. and Cook, A. E. (1993). ‘‘A Genetic Monitoring and Evaluation Programme
for Supplemented Populations of Salmon and Steelhead in the Snake River Basin.’’ Rep.
No. DE-A179-89BP00911. Bonneville Power Administration, Portland, OR.
Waples, R. S., Gustafson, R. G., Weitkamp, L. A., Myers, J. M., Johnson, O. W.,
Busby, P. J., Hard, J. J., Bryant, G. J., Waknitz, F. W., Neely, K., Teel, D.
Grant, W. S. et al. (2001). Characterizing diversity in salmon from the Pacific Northwest.
J. Fish Biol. 59, 1–41.
Waples, R. S., Ford, M. J., and Schmitt, D. (In press). Empirical results of salmon supplementation: A preliminary assessment. In ‘‘Ecological and Genetic Implications of Aquaculture Activities’’ (T. Bert, ed.). Kluwer Academic Publishers.
Ward, R. D. (2006). The importance of identifying spatial population structure in restocking
and stock enhancement programmes. Fish. Res. 80, 9–18.
Was, A. and Wenne, R. (2002). Genetic differentiation in hatchery and wild sea trout (Salmo
trutta) in the Southern Baltic at microsatellite loci. Aquaculture 204, 493–506.
Was, A. and Wenne, R. (2003). Microsatellite DNA polymorphism in intensely enhanced
populations of sea trout (Salmo trutta) in the Southern Baltic. Mar. Biotechnol. 5, 234–243.
Washington Department of Fish and Wildlife (1997). ‘‘Final Environmental Impact Statement for the Wild Salmonid Policy.’’ Washington Department of Fish and Wildlife,
Olympia, WA.
Watanabe, R. A., Fryer, J. L. and Rohovec, J. S. (1988). Molecular filtration for recovery of
waterborne viruses of fish. Appl. Environ. Microbiol. 54, 1606–1609.
Weber, E. D. and Fausch, K. D. (2003). Interactions between hatchery and wild salmonids in
streams: Differences in biology and evidence for competition. Can. J. Fish. Aquat. Sci. 60,
1018–1036.
Wedemeyer, G. A. (1998). ‘‘Physiology of Fish in Intensive Culture Systems.’’ Chapman
and Hall, New York.
WGNAS (2003). ‘‘Report of the Working Group on North Atlantic Salmon.’’ ICES,
Copenhagen, Denmark.
Whelan, K. F. and Poole, W. R. (1996). The sea trout stock collapse, 1989–1992. In ‘‘The
Conservation of Aquatic Systems’’ (J. Reynolds, ed.), pp. 101–110. Royal Irish Academy, Dublin.
White, H. C. (1924). A quantitative determination of the number of survivors from planting
5,000 trout fry in each of two streams. Contrib. Can. Biol. 2, 135–150.
Williams, I. V. and Amend, D. F. (1976). A natural epizootic of infectious hematopoietic
necrosis in fry of sockeye salmon (Oncorhynchus nerka) at Chilko Lake, British Columbia.
J. Fish. Res. Board Canada 33, 1564–1567.
Williams, R. N., Shiozawa, D. K., Carter, J. E. and Leary, R. F. (1996). Genetic detection of
putative hybridization between native and introduced rainbow trout populations of the
upper Snake river. Trans. Am. Fish. Soc. 125, 387–401.
Williamson, K. S. and May, B. (2005). Homogenization of fall-run Chinook salmon gene
pools in the central valley of California, USA. North Am. J. Fish. Manag. 25, 993–1009.
Willis, D. B., Caldas, J., Frasier, M., Whittlesey, N. K. and Hamilton, J. R. (1998). The
effects of water rights and irrigation technology on streamflow augmentation cost in the
Snake River Basin. J. Agric. Resour. Econ. 23, 225–243.
Winton, J. R. (2001). Fish health management. In ‘‘Fish Hatchery Management’’
(G. Wedemeyer, ed.), pp. 559–640. American Fisheries Society, Bethesda, MD.
Wissmar, R. C. and Simenstad, C. A. (1988). Energetic constraints of juvenile chum salmon
(Oncorhynchus keta) migrating in estuaries. Can. J. Fish. Aquat. Sci. 45, 1555–1560.
194
Kerry A. Naish et al.
Wolf, K. (1988). ‘‘Fish Viruses and Fish Viral Diseases.’’ Cornell University Press, Ithaca,
NY.
Woo, P. T. K. (1995). ‘‘Fish Diseases and Disorders, Volume 1. Protozoan and Metazoan
Infections.’’ CAB International, Wallingford, UK.
Woo, P. T. K. and Bruno, D. (1998). ‘‘Fish Diseases and Disorders, Volume 3. Viral,
Bacterial and Fungal Infections.’’ CAB International, Wallingford, UK.
Wood, A., Orr, C. and Penikett, J. (2002). In ‘‘The History, Goals and Direction of the
Salmonid Enhancement Programme (SEP) in British Columbia,’’ pp. 5–9. Simon Fraser
University, Burnaby, BC.
Woodworth, L. M., Montgomery, M. E., Briscoe, D. A. and Frankham, R. (2002). Rapid
genetic deterioration in captive populations: Causes and conservation implications.
Conserv. Genet. 3, 277–288.
Wu, J., Adams, R. M. and Boggess, W. G. (2000). Cumulative effects and optimal targeting
of conservation efforts: Steelhead trout habitat enhancement in Oregon. Am. J. Agric.
Econ. 82, 400–413.
Zhou, S. J. (2002). Uncertainties in estimating fishing mortality in unmarked salmon in
mark-selective fisheries using double-index-tagging methods. North Am. J. Fish. Manag.
22, 480–493.
The Social Structure and Strategies
of Delphinids: Predictions Based on
an Ecological Framework
Shannon Gowans,*,† Bernd Würsig,* and Leszek Karczmarski*,‡
Contents
197
199
200
201
204
205
205
208
215
218
219
223
225
231
236
239
1. Biological Pressures on Social Strategies
1.1. Why do animals form groups?
1.2. Definitions and levels of grouping
1.3. What are social strategies?
1.4. How does ecology influence social strategies?
2. Dolphin Ecology
2.1. Distribution and habitat
2.2. Predation and predatory risk
2.3. Foraging behaviour and diet
2.4. Ranging patterns and daily movements
2.5. Socioecology
3. Resident Communities
3.1. Inshore bottlenose dolphins (Tursiops sp.)
3.2. Spinner dolphins (Stenella longirostris)
3.3. Comparisons with terrestrial mammals
4. Wide-Ranging Communities
4.1. Eastern Tropical Pacific dolphins (Delphinus and
Stenella sp.)
4.2. Coastal bottlenose dolphins (Tursiops sp.)
4.3. Dusky dolphins (Lagenorhynchus obscurus)
4.4. Comparisons with terrestrial mammals
5. Intermediate-Ranging Patterns
5.1. Humpback dolphin (Sousa sp.)
5.2. Killer whales (Orcinus orca)
5.3. Comparisons with terrestrial mammals
6. Demographic, Social and Cultural Influences
*
{
{
241
243
246
251
253
254
259
265
267
Texas A&M University, Galveston Texas 77551, USA
Eckerd College, Petersburg, Florida 33711, USA
University of Pretoria, Mammal Research Institute, South Africa
Advances in Marine Biology, Volume 53
ISSN 0065-2881, DOI: 10.1016/S0065-2881(07)53003-8
#
2008 Elsevier Ltd.
All rights reserved.
195
196
Shannon Gowans et al.
7. Comparisons with Other Cetaceans
7.1. Sperm whales (Physeter macrocephalus)
7.2. Northern bottlenose whales (Hyperoodon ampullatus)
7.3. Harbour porpoise (Phocoena phocoena)
7.4. Why are there no long-term bonds in baleen whales?
8. Conservation Implications
9. Concluding Comments
Acknowledgements
References
268
269
271
272
273
274
276
278
278
Abstract
Dolphins live in complex social groupings with a wide variety of social strategies. In this chapter we investigate the role that differing habitats and ecological conditions have played in the evolution of delphinid social strategies. We
propose a conceptual framework for understanding natural patterns of delphinid social structure in which the spatial and temporal predictability of resources
influences the ranging patterns of individuals and communities.
The framework predicts that when resources are spatially and temporally
predictable, dolphins should remain resident in relatively small areas. Predictable resources are often found in complex inshore environments where dolphins
may hide from predators or avoid areas with high predator density. Additionally,
available food resources may limit group size. Thus, we predict that there are few
benefits to forming large groups and potentially many benefits to being solitary
or in small groups. Males may be able to sequester solitary females, controlling
mating opportunities. Observations of inshore populations of bottlenose dolphins
(Tursiops sp.) and island-associated spinner dolphins (Stenella longirostris) seem
to fit this pattern well, along with forest-dwelling African antelope and primates
such as vervets (Cercopithicus aethiops), baboons (Papio sp.), macaques (Macaca
sp.) and chimpanzees (Pan troglodytes).
In contrast, the framework predicts that when resources such as food are
unpredictable, individuals must range further to find the necessary resources.
Forming groups may be the only strategy available to avoid predation, especially in the open ocean. Larger home ranges are likely to support a greater
number of individuals; however, prey is often sparsely distributed, which may
act to reduce foraging competition. Cooperative foraging and herding of prey
schools may be advantageous, potentially facilitating the formation of longterm bonds. Alternately, individuals may display many short-term affiliations.
These large groups make it difficult for a male or a small group of males to
sequester a female, and polygynandry is the most likely mating strategy. While
it is difficult to study wide-ranging delphinids to examine these predictions, this
ranging and behavioural pattern has been suggested for dusky dolphins
(Lagenorhynchus obscurus), coastal bottlenose dolphins (Tursiops sp.) and
mixed species of dolphins in the Eastern Tropical Pacific. These patterns also
resemble the ranging and social strategies of open savannah African antelopes
and desert-dwelling macropods.
The Social Structure and Strategies of Delphinids
197
Resource availability exists in a range of complex distributions and we predict
that delphinid ranging patterns will also vary. At intermediate-ranging patterns,
the framework predicts that individuals should form mid-sized groups balancing intra-group competition with predation protection. Humpback dolphins
(Sousa sp.) appear to fit this pattern, with some site fidelity over relatively
large ranges. They display fluid associations with other individuals. Predation
pressure is not sufficiently high to cause large groups to form, and individuals
probably reduce predation pressure more by hiding whenever possible. This
pattern is likely to prevent the formation of long-term complex bonds. In
contrast, killer whales (Orcinus orca) also display intermediate-ranging patterns, but have extremely strong social bonds within familial groups. Cooperative and altruistic behaviour in killer whales facilitate the formation of life-long
bonds, similar to those observations in sperm whales (Physeter macrocephalus)
and elephants (Loxodonta africana).
This conceptual framework remains largely untested, and for many species it
is not currently possible to describe ranging behaviours, anti-predator tactics or
social behaviour in sufficient detail for appropriate examination of these ideas.
Few studies on dolphins have been conducted to explicitly test this type of
framework; however, existing observations of delphinid social strategies and
communities are used throughout this chapter to examine this framework.
Additionally, we anticipate that the present framework may provide a starting
point to test hypotheses regarding the evolution of social strategies of
delphinids.
1. Biological Pressures on Social Strategies
Dolphins have intrigued human observers since at least the time of the
ancient Greeks (e.g., Reynolds et al., 2000), and this fascination has largely
been due to their obvious sociality. The popular literature is filled with
stories describing apparent intelligence, communicative abilities, acts of
altruism and social interactions of dolphins. The scientific literature tends
to be less effervescent; however, there are a growing number of studies
investigating the social lives of dolphins (e.g., Baird and Whitehead, 2000;
Connor and Norris, 1982; Connor et al., 1999; Karczmarski et al., 2005;
Krützen et al., 2003; Lusseau et al., 2003; Mann et al., 2000; Möller et al.,
2001; Slooten et al., 1993; Smolker et al., 1992; Würsig and Würsig, 1980).
The 36 recognized species in the family Delphinidae (Rice, 1998) are a
diverse assemblage of species, ranging in size from the 45 kg, 1.2–1.4 m
Hector’s dolphin (Cephalorhynchus hectori; Slooten and Dawson, 1994) to the
more than 5000 kg, 9.0 m male killer whale (Orcinus orca; Dahlheim and
Heyning, 1999; Fig. 3.1). The ecological habitats of delphinids are equally
diverse, with species such as the Irrawaddy dolphin (Orcaella brevirostris) and
tucuxi (Sotalia fluviatilis) present thousands of kilometres up river from the
198
Shannon Gowans et al.
A
B
Figure 3.1 Delphinids range widely in adult body size. (A) Hector’s dolphin Cephalorhynchus hectori is the smallest delphinid, while (B) the largest is the killer whale
Orcinus orca (see text). Both are disruptively coloured, mainly dark above and light
below. The killer whale has a large white eyespot and white lower jaw, probably both
for conspecific communication in murky waters and for confusing intended prey.
Killer whales are sexually dimorphic, with males larger and a very high erect, at times
wavy (as here), dorsal fin. (Photos courtesy of M.Wˇrsig, with permission.)
ocean (Arnold, 2002; da Silva and Best, 1994), while others such as the
striped and Clymene dolphin (Stenella coeruleoalba and Stenella clymene) tend
to occur thousands of kilometres from land in the open ocean (Perrin and
Mead, 1994; Perrin et al., 1994). The delphinids also display diversity in
group size and social structure, as some species typically occur in small fluid
groups (e.g., Hector’s dolphins; Bräger, 1999; Slooten et al., 1993), while
others are in large fluid groups (i.e., Hawaiian spinner dolphins Stenella
longirostris; Perrin and Gilpatrick, 1994) and still others occur in highly
structured, permanent groups (i.e., resident killer whales; Baird, 2000).
The Social Structure and Strategies of Delphinids
199
Additionally, even within the same species there can be clear differences in
diet, behaviour and social structuring of different groups (i.e., killer whales;
Baird, 2000).
Recent studies of several toothed whale species have given insight into
some of the forces leading to the evolution of sophisticated social groupings,
but the picture is far from clear (Connor, 2000). The focus of this chapter is
to investigate what has shaped the evolution of group living in delphinids,
the social strategies seen within groups of dolphins and to compare them
with other well-studied systems. In particular, we address the roles that
differing habitat and ecology have played in the evolution of social strategies. Research into the ecological basis of the evolution of social structuring
of delphinids enhances our understanding of delphinid biology and conservation as well as provides insight into the evolution of complex social
structuring in all mammals.
1.1. Why do animals form groups?
Animals form groups when survival and reproductive success of an individual is enhanced by group living. There are three major reasons individuals
form groups: (1) when susceptibility to predation is reduced by group
living, (2) when access to resources is increased by group living, and
(3) when distribution of resources promotes gregariousness; for example,
when there are only a few localized sites where resources such as water or
shelter can be found (Alexander, 1974; Bertram, 1978). Once animals are
grouped, then more sophisticated inter-animal social behaviours than sexual
and nurturant interactions can evolve (Alexander, 1974).
Group living is a trade-off between competing factors. Potential benefits
to group living include reduced predation, enhanced detection and capture
of prey, increased acquisition or defence of resources, enhanced reproduction, reduction in parasitism and the possibility of social interaction and
learning. However, group living can also result in increased predation,
reduced foraging efficiency, increased competition for resources, reduced
reproductive opportunities and increased transmission of parasites and disease. The matrix of cost–benefit ratios for an individual depends on its sex,
reproductive state and ecology (Bertram, 1978). Costs and benefits to
individuals are often measured in immediate returns such as rate of food
intake or estimates of predation risk. However, it may be more appropriate
to measure how group living and different social structures influence
lifetime reproductive success, the number of surviving offspring an individual produces and even inclusive reproductive success, which not only
includes individual surviving offspring but also considers shared genetic
heritage passed on by surviving offspring of relatives (Lucas et al., 1996).
Such analyses are difficult, especially for long-lived animals, and even more
200
Shannon Gowans et al.
so for marine mammals, where usually only a part of their lives can be traced
when they are at the water surface.
The ecological basis for group formation has been studied extensively in
terrestrial and aquatic environments (i.e., Beecham and Farnsworth, 1999;
Bonabeau et al., 1999; Wrangham and Rubenstein, 1986a), as have the
correlations between environmental variables and group size (i.e., Arcese
et al., 1995; Baird and Dill, 1996; Janson and Goldsmith, 1995; Macdonald,
1983; Wrangham et al., 1993). From this body of work, several overall
themes emerge in relation to predation and foraging. In complex, highly
structured environments, animals tend to be able to conceal themselves
from predators, and group sizes are often smaller than in open habitats, or
individuals may be predominantly solitary (i.e., Bednekoff and Lima, 1998;
Brashares et al., 2000; Wells et al., 1999a). Diet specificity tends to increase
competition between individuals and groups for limited resources, which in
turn limits group size (i.e., Brashares et al., 2000; Isbell, 1991). However, in
certain situations, cooperative foraging and hunting can reduce the disadvantages of living in large groups (i.e., Creel and Creel, 1995).
When the advantages of group living outweigh the costs and animals live
in groups, intra-group competition often occurs. However, when individuals within the group cooperate, competition can be repressed and cooperating groups out-compete non-cooperators (Frank, 2003). Thus,
repression of competition can result in highly coordinated and cooperative
groups, even when individuals are not related to each other. In fact,
individuals in many cooperative groups are not related to each other and
therefore kin selection cannot be used to explain these types of cooperation
(e.g., Lukas et al., 2005; Spong and Creel, 2004; Valsecchi et al., 2002; Van
Horn et al., 2004; Vucetich et al., 2004).
1.2. Definitions and levels of grouping
Although it may appear intuitive at first to define what constitutes a group of
individuals, in practice this has proved more difficult and represents a rather
contentious issue in the study of sociality. Some researchers define group to
include all individuals occupying similar space at the same time; often
expressed as a spatial scale (e.g., all individuals within 5 body lengths;
Connor et al., 2000b). However, this definition does not distinguish
between individuals simply aggregated together, and those that share a
common purpose. Other studies define a group to be all individuals in the
same place at the same time, which are acting in a similar manner
(e.g.,Connor et al., 2000a). These definitions can describe aggregations
and interactions that occur at a specific time, but do little to describe
long-term patterns of association. Some studies define groups as individuals
that spend proportionately more time together than with others (e.g., they
are grouped if they spend 50% of their time together; Bigg et al., 1990; Sailer
The Social Structure and Strategies of Delphinids
201
and Gaulin, 1984). Additionally, primate and cetacean researchers have also
used a variety of terms to denote grouping (e.g., subgroup, group, party,
school and pod) and rarely are definitions consistent between studies.
Therefore, it is essential to ascertain definitions used in each study before
comparisons can be made between studies.
Five major groupings have been identified; and an individual animal can
be considered to be in more than one at the same time. Population-level
groups can be defined as all individuals sharing a common home range.
Under this major level, feeding groups consist of individuals actively feeding
on the same food resource at the same time. Foraging groups represent
individuals that are searching or hunting for food together. Breeding groups
include individuals that mate with each other (Gittleman, 1989). Nursery
groups consist of mothers, their calves and at times their extended ‘other’
relations as well. Population level groups may split into subgroups for
feeding, foraging, mating or nurturant activities, and then coalesce sometime later. In these situations, individuals may associate with a number of
different individuals over time and relationships may appear rather fluid.
These types of associations are often termed fission–fusion and are common
in primates, cetaceans and some social carnivores (Connor et al., 1998;
Gittleman, 1989; Wrangham et al., 1993). ‘Community’ refers to regional
assemblage (society) of animals that share ranges, interact socially, but do not
represent closed reproductive units (sensu Wells and Scott, 1999), which is
similar to those in many primate studies (e.g., Boesch, 1996).
1.3. What are social strategies?
While extensive work has been carried out to examine the ecological
basis for group formation and the relative advantages of various group
sizes, relationships between individuals within groups can also vary with
ecology and are an important aspect of group living. When animals are
grouped in time and space, social behaviour can evolve, although this is not
necessarily always the case (Alexander, 1974). If individuals only encounter
each other once, then they do not have the potential to establish a knowledge base about the other individual. However, when associations persist
over time, individuals can repeatedly come into contact with each other and
establish a knowledge base of interactions (Hinde, 1976; Pusey and Packer,
1997). Such repeated interactions lead to the development of relationships
(e.g., dominance, cooperative foraging and mating coalitions), and a description of the various relationships in the group can lead to a description of
the social structure of the group (e.g., polygamous, female-bonded). Thus,
the social structure of the group is based upon an abstraction of the nature,
quality and patterning of relationships, and relationships are built upon a
series of interactions between individuals (Hinde, 1976).
202
Shannon Gowans et al.
While social systems can be viewed as an abstraction of interactions
between individuals, as well as the strategies that individuals employ when
interacting with each other, the underlying behaviour of an individual
is shaped by its ecology (i.e., distribution of risks and resources). However,
the social organization and demographics of a group also place constraints
on the behaviour of an individual, which create a feedback loop to the
types of interactions between individuals and the resulting social structure
(e.g., Kappeler and Van Schaik, 2002, and references therein).
1.3.1. Male versus female strategies
Social strategies vary not only with ecology but can also vary dramatically
between the sexes. A female mammal transfers a large portion of her
energetic resources to her offspring through lactation and gestation. Her
reproductive success is limited by metabolic costs of lactation and gestation;
therefore, a female may most effectively increase her reproductive success
by investing heavily in each offspring. This is especially true for large
mammals that typically produce a single offspring at a time. In contrast,
male mammals often provide little parental care, and most effectively
increase their reproductive success by maximizing the number of receptive
females with whom they mate (Emlen and Oring, 1977; Wrangham and
Rubenstein, 1986b).
This dichotomy between male and female strategies to increase reproductive success can lead to marked differences in distribution and social
strategies (Ruckstuhl and Neuhaus, 2005). As a female tries to ensure the
survival of each of her offspring, her distribution is often closely related to
high-quality food resources and habitats that have lower risk of predation.
The distribution of males, however, is often related more to female distribution than to availability of food resources (Emlen and Oring, 1977;
Wrangham and Rubenstein, 1986b). Male strategies are especially tuned
to the availability of reproductively active females, and can greatly be
influenced by the length and synchronicity of periods of oestrus (e.g., Say
et al., 2001; Whitehead, 1990).
1.3.2. Mating systems
Among mammals, females are the limiting sex as they can only produce a
restricted number of offspring in a single breeding season, while male
reproductive success is limited by the number of females he can successfully
monopolize (Emlen and Oring, 1977). Parental care by both males and
females is required to raise offspring successfully in only a few mammalian
species. In most mammals, females provide all the parental care, with little to
no male assistance. In these situations, the ability of males to defend
(from other males) a female, group of females, or a specific territory plays
an important role in determining the mating system. If the core area where
females range is defendable by a male or a group of males, then males are
The Social Structure and Strategies of Delphinids
203
typically territorial, excluding other males from a specific geographic location. If females within this range are solitary or found in small groups, a
single male may be able to defend the territory on his own. When females
form larger groups, males may form coalitions to collectively defend the
territory. Alternatively, when the core area where females range is not
defendable, male territoriality is not observed. In situations where females
form relatively stable groups, individual males or coalitions of males may
exclude other males from accessing the group of females. If females are
solitary or groups are widely dispersed, males may display a roving behaviour in which they search for receptive females and spend only short
periods with each female. However, if females are found in unstable groups,
males may form temporary mating territories or display lek behaviours
during the breeding season. Thus, mammalian mating systems are often
driven by male competition for access to females (Clutton-Brock, 1989),
although female choice is likely to be an important aspect of mammalian
mating systems, only recently being investigated. Male competition for
access to females is especially important in delphinids, as females can
only produce a single offspring at a time and have long interbirth intervals,
for much of which time they may not be in oestrus (Whitehead and
Mann, 2000).
Mating strategies have important implications for the dispersal of offspring. In almost all animal species, offspring of one sex disperse from
the natal area, or both sexes will disperse. Mammals are predisposed for
females to care for offspring, thus female philopatry is often favoured. For
females, there may be advantages to staying in the natal area (or with the
natal group), which can lead to the evolution of kin selection and altruism.
Typically, male mammals then disperse to avoid inbreeding (Greenwood,
1980).
1.3.3. Evolution of social complexity and variability
Animal social complexity can be viewed as attempts by individuals to form
cooperative solutions to challenges of survival and reproduction. In this
framework, animal social systems can be described as alliances between
individuals to simultaneously survive (find food, avoid predators) and
reproduce (find a mate, mate and rear offspring), and alliance memberships
and purpose may change with differing social and ecological conditions
(Dunbar, 1989).
Social behaviour can be viewed as a flexible response that varies in
relation to ecological and demographic factors. It is not possible to describe
categorically the social organization of most species, as changes in physical
and social environments of differing populations or communities may
influence their social structure (Kappeler and Van Schaik, 2002). However,
this does not preclude attempts to study social structure, or its evolution, as
there will undoubtedly be several unifying themes in the evolution of social
204
Shannon Gowans et al.
structure. Nevertheless, the focus should perhaps be on the investigation of
these themes, rather than attempts to describe any one particular social
structure of a species or subspecies.
1.4. How does ecology influence social strategies?
An individual’s social strategy can clearly be influenced by its ecology.
Resource and risk distribution are important factors (e.g., Cezilly and
Benhamou, 1996; Geffen et al., 1996; Johnson et al., 2002), affecting social
patterns and group size. Resource availability can vary both spatially (e.g.,
food patches that can occur in different areas such as mobile prey) and
temporally (e.g., food patches that only occur at certain times of year such as
fruiting trees). When resources are routinely available within a relatively
small area, individuals (and groups) may remain resident in the area and in
some situations can become territorial, defending their home range against
the intrusion of others. When resources are highly variable over a range of
scales, individuals must range widely and home ranges become very large
and not defensible as exclusive territories.
Resource availability also influences the ways individuals interact as
limited resources lead to competition. Scramble-type competition occurs
when individuals (or groups) cannot exclude others from access to the
resources and all individuals share equally in the resources (or lack thereof;
Milinski and Parker, 1991). In this case, individuals (or groups) should
distribute themselves such that they share resources equally (ideal free
distribution; Fretwell and Lucas, 1970). Contest-type competition (also
known as direct or interference competition) occurs when individuals
(or groups) can exclude others from access to the resources (Milinski and
Parker, 1991). Individuals or groups may establish territories, or simply
defend the actual resource. In this situation, certain individuals (or groups)
have greater access to resources than others. In groups with strong dominance hierarchies, or when individuals have great differences in competitive
ability, one individual may monopolize all the resources, leaving all other
members with none. The distribution and abundance of food resources
usually determines the type of competition. Scramble-type competition
tends to occur when the distribution of patches is homozygous and the
patches are relatively small, or the patches are very large and indefensible.
Contest-type competition tends to occur when patches are moderately sized
and defendable (Milinski and Parker, 1991).
Residency rates and measurements of day range, that is, the linear
movement of an individual or a group over a single day, can be used as
proximate measures of resource availability and predictability. As large-scale
movement may energetically be expensive or dangerous due to predation,
individuals would be expected to range only as widely as needed to obtain
required resources. Thus, individuals ranging widely are likely searching for
The Social Structure and Strategies of Delphinids
205
dispersed resources. If competition between group members for resources is
high, then group size is likely to be small, especially when resources are
widely distributed, and for many species mean day range can be correlated
with social grouping patterns (e.g., Wrangham et al., 1993).
1.4.1. Special features of the marine environment
There are several features of the marine environment which have the
potential to influence the relationships between a dolphin’s ecology and
its social structure. Dolphins live in a predominantly three-dimensional
(3D) world, with many species having a large component of their daily
movements in the vertical plane. It is unlikely that dolphins living in a
strongly 3D environment such as the open ocean would be able to defend a
territory, or even defend a large resource. It may be possible for individuals
or groups to defend relatively small resources (such as a receptive female;
see Connor et al., 2000a), or for coastal dolphins to defend a territory,
although this has not been reported (Connor, 2000; but see discussion of
spinner dolphins, Section 3.2). Although dolphins may not often be able to
defend territories, this deficiency may not incur the same costs as it would
for territorial animals, as swimming dolphins appear to be able to increase
their range of movements relatively easily, since the cost of locomotion is
relatively low in marine mammals (see Williams, 1999, for details). Additionally, the marine environment is typically characterized by highly clumped and
labile prey (Horwood and Cushing, 1978; Steele, 1985). Thus, contest-type
competition may not be as important in the marine environment as it is in the
terrestrial.
However, the 3D marine environment has implications for delphinid
anti-predator strategies. In the open ocean, there are few if any places to
hide and grouping may be the only effective anti-predatory strategy
(Norris and Dohl, 1980). Additionally, the requirement of dolphins to
routinely return to the surface to breathe further limits their anti-predator
strategies.
2. Dolphin Ecology
2.1. Distribution and habitat
Distributions of dolphins reflect a complex matrix of evolutionary history,
phylogeny, prey distribution, predation risk, thermal tolerance and habit as
learned from group members and ancestors. Some near-shore animals, such
as Cephalorhynchus of the southern hemisphere, are more geographically
restricted than many open ocean species that travel over relatively huge
206
Shannon Gowans et al.
areas, such as common dolphins (Delphinus sp.), pilot whales (Globicephala
sp.) and most species of the genus Stenella (Fig. 3.2).
Dolphins are found throughout the world’s oceans and even into several
large river systems. Broad habitat categorizations can be made, although
there are subtle differences within these categories. While there is no
consensus on exact habitat definitions, we use the following broad categories throughout this chapter. Riverine habitats consist of freshwater rivers
and their brackish estuaries. Dolphins in these habitats can strongly be
influenced by seasonal water levels and tidal cycle. Inshore habitats consist
of protected bays, estuaries and tidal marshes. In these typically shallowwater habitats, dolphins can be influenced by tidal cycles and seasonal
A
B
The Social Structure and Strategies of Delphinids
207
C
D
Figure 3.2 Examples of tropical and warm-water dolphins. (A) Short-finned pilot
whale Globicephala macrorhynchus, (B) Atlantic spotted dolphin Stenella frontalis, (C)
short-beaked common dolphin Delphinus delphis, and (D) spinner dolphin Stenella longirostris. Spotted dolphins develop their spots as they mature. Short-beaked common dolphins occur in nearshore and offshore waters of the Atlantic and Pacific oceans, while
their long-beaked congenerics Delphinus capensis occur as well in the Indian Ocean, but
always close to shore. (Photos courtesy of B.Wˇrsig, L. Karczmarski.)
208
Shannon Gowans et al.
changes in water temperature and salinity; but they are protected from most
storms. Coastal habitats describe open coastlines typically <100 m in depth
and relatively close to shore (within several kilometres), lacking barrier
islands, large protected bays or tidal marshes. Coastal habitats may be located
just offshore of inshore areas, as found throughout most of the Gulf of
Mexico, for example. These habitats are more exposed to storm systems, but
often experience less dramatic seasonal variations in temperature and salinity.
Neritic habitat describes the area over the continental shelf, which lies
offshore of the coastal habitat. This area is typically <200 m in depth but
does not include the surf zone or coastal waters. Dolphins in this habitat can
be influenced by storm systems and some seasonal changes in temperature
and salinity. Finally, the pelagic habitat represents deep-water areas over the
continental slope and open ocean. This habitat shows less seasonal variation
in temperature and salinity and has no barriers to movement. Dolphin
communities can occupy more than one habitat, on daily, periodic, or
seasonal basis and there can be overlap between categories.
The distribution of dolphin species can also vary between colder temperature to sub-polar and tropical waters. Typically, the more sub-polar
species tend to have larger, thicker bodies than tropical dolphins, and this
gradation can even be observed within species (e.g., bottlenose dolphins
(Tursiops sp.; Wells and Scott, 1999). However, some rapidly moving smallbodied dolphins such as the hourglass dolphin (Lagenorhynchus cruciger) and
Commerson’s dolphin (Cephalorhynchus commersonii) and northern and
southern right whale dolphins (Lissodelphis sp.) occur at high latitudes
(Brownell and Donahue, 1999; Goodall, 1994).
2.2. Predation and predatory risk
Predation is an important factor that promotes sociality in many different
species in a wide variety of habitats (e.g., Bertram, 1978; Jarman, 1974;
Norris and Schilt, 1988). Individuals in groups may have a better chance of
detecting predators, avoid being consumed, or cooperatively drive off
predators (Bertram, 1978). In some cases, however, living in groups may
actually increase predation risk to individuals if larger groups are more
attractive or easily detected by predators (e.g., Hebblewhite and Pletscher,
2002). Therefore, group living may have both costs and benefits, and the
net cost/benefit ratio depends on the specific predation conditions
(Bertram, 1978).
In most locations where dolphins are found, only limited information is
known about predation risk. Relatively few predatory attacks have been
observed (see Connor and Heithaus, 1996; Gibson, 2006; Maldini, 2003;
Mann and Barnett, 1999; Pitman et al., 2003, for examples). Shark-inflicted
scars on living dolphins can give some indication of predation risk, although
these scars only represent non-lethal attacks. Approximately 74% of
The Social Structure and Strategies of Delphinids
209
bottlenose dolphins (Tursiops sp.) in Shark Bay, Australia bear shark-inflicted
scars. Bottlenose dolphins in other area display lower scar rates (10–20% in
South Africa, 31% in Sarasota, Florida, 37% in Moreton Bay, Australia) while
dolphins in the open ocean and the Adriatic have few if any shark scars
(Heithaus, 2001a). While it is clear that that many dolphins face predation
by sharks and killer whales, the likelihood of predation and the strategies
that dolphins use to avoid predation are very poorly studied. Although it is
possible to draw some inferences about predation on dolphins, these remain
for the most part, untested conjecture based on limited observations. Future
studies will hopefully address and test predation questions more directly,
similar to the scope of studies currently conducted in Shark Bay, Australia
(e.g., Heithaus and Dill, 2002).
It is likely that predation is an important factor leading to the sociality of
delphinids (e.g., Connor, 2000; Norris and Dohl, 1980; Norris and Schilt,
1988). Group living in dolphins likely reduces predation risk in several
ways, although few studies have empirically examined delphinid group
living and predation risk. First, by grouping, the distribution of dolphins
becomes more heterogeneous in the environment. Thus, predation risk is
lowered as a predator must search a larger area to locate prey (the encounter
effect). However, aggregations of animals may be more easily detectable by
predators, and this could actually increase the encounter probability (e.g.,
Bertram, 1978; Turner and Pitcher, 1986). Second, an individual dolphin’s
chance of being preyed upon is reduced within the group, as predators
typically attack only one individual (the dilution effect; e.g., Childress and
Lung, 2003; Foster and Treherne, 1981; Hebblewhite and Pletscher, 2002;
Turner and Pitcher, 1986; Wisenden, 1999). Third and somewhat related to
the dilution effect is the concept that individual dolphins may hide behind
another individual (cover hypothesis; e.g., Bumann et al., 1997; Hamilton,
1971). Under this hypothesis, different positions in the group have different
associated risks, with the centre considered to be the safest (Bumann et al.,
1997). Fourth, being in a group can help in the detection of predators, as
there are more individuals on the lookout (vigilance; e.g., Elgar, 1989;
Pulliam, 1973). Additionally, dolphins in groups may cooperate by
integrated communication to detect a predator (what Norris and Schilt,
1988, termed a part of the Sensory Integration System; see also Norris et al.,
1994), to confuse a predator or to cooperatively attack it (e.g., Bertram,
1978; Krakauer, 1995; Ostreiher, 2003).
Predatory attacks are often difficult to observe and may lead to difficulties in estimating predation risk, as has been confirmed in studies on primate
predation (Stanford, 2002). It is especially difficult to estimate predation risk
faced by dolphins, as many attacks will likely occur below the water surface
and away from human observers. Major predators of dolphins include killer
whales and sharks (Connor, 2000; Heithaus, 2001b; Jefferson et al., 1991).
However, as relatively few predatory attacks have been observed, little is
210
Shannon Gowans et al.
known about how predation risk may vary between environments, species,
group sizes and group composition.
Habitat characteristics may influence general predation risk. Complex
habitats (such as enclosed bays, estuaries and tidal marshes) may provide
elements behind which the dolphins can hide and thereby reduce their
predation risk (e.g., Connor, 2000; Wells et al., 1999a). Clear water may
also help dolphins to detect predators (Heithaus, 2001b). The surf zone also
represents an area where predators may have difficulty detecting dolphins
(Heithaus, 2001b; Jefferson et al., 1991), and near-shore dusky dolphins
(Lagenorhynchus obscurus), for example, retreat to within the surf zone when
killer whales approach (Würsig and Würsig, 1980). Spinner dolphins rest in
clear shallow waters of island bays and lagoons, presumably to be able to scan
their entire environment, and to reduce the possibility of attacks from depth
(Fig. 3.3). They prefer to travel over expanses of sand, not variegated
bottom as presented by coral beds, presumably due to the greater possibility
of shark attacks in more complex substrata (Würsig et al., 1994b).
Killer whales are found throughout much (but not all) of the world’s
oceans and they probably prey upon most species of dolphin, except those
dolphins living far up river systems (e.g., Irrawaddy dolphin and tucuxi).
Several species of dolphins are known to be killer whale prey, including
long-finned pilot whales (Globicephala melas), common dolphins, dusky
dolphins and spotted dolphins (Stenella sp. Connor, 2000; Jefferson et al.,
1991; Pitman et al., 2003). Other species of dolphins, porpoises, mid-sized
toothed whales and baleen whales also comprise the killer whale’s prey
( Jefferson et al., 1991). There have been no documented predatory attacks
Figure 3.3 A resident group of spinner dolphins Stenella longirostris resting in shallowwater MidwayAtoll lagoon, far-western Hawaii. (Photo courtesy of L. Karczmarski.)
The Social Structure and Strategies of Delphinids
211
on killer whales by other killer whales, although they have conspecific scars
and toothrakes, these probably do not represent predation attempts. If killer
whales do prey upon conspecifics, it appears to be extremely rare (Baird,
2000).
Killer whale population density varies greatly worldwide, and some
dolphin populations are more at risk than others. Killer whales are most
abundant in coastal waters in the higher latitudes, so offshore and tropical
dolphins tend to face a lower predation risk from them (Dahlheim and
Heyning, 1999). Some inshore areas may represent protected refuge from
killer whales. The water may be too shallow for killer whales, or islands,
bays, the surf zone, or kelp beds provide areas to hide (Fig. 3.4). Documented killer whale attacks on delphinids are relatively rare but indicate that
groups of killer whales can effectively hunt seemingly healthy adult dolphins
(e.g., Constantine et al., 1998; Jefferson et al., 1991; Pitman et al., 2003).
Dolphins display a number of behavioural responses to the presence of
killer whales. In many situations, dolphins and killer whales coexist in the
same area, with no noticeable reaction of dolphins to killer whales. In fact in
many of these situations, killer whales make no attempt to attack and prey
species avoid fleeing from every potential attack (Jefferson et al., 1991), and
which is also commonly observed between terrestrial predators and ungulates (e.g., Caro et al., 2004). In other situations, dolphins have been
observed rapidly swimming away from the killer whales, forming tight
groups with coordinated movements, rapid expansion of groups with
animals swimming in many different directions, or hiding in shallow
water or kelp beds (e.g., Constantine et al., 1998; Jefferson et al., 1991;
Pitman et al., 2003). It is likely that the motivational state of predator and
prey will often determine the behavioural response of the prey (Lima and
Bednekoff, 1999; Stankowich and Blumstein, 2005).
In some non-delphinid marine mammals, deep diving behaviour
has been suggested to be an anti-predator strategy, with sperm whales
Figure 3.4 A group of Indo-Pacific bottlenose dolphins Tursiops aduncus in the surf
zone off the southeast coast of South Africa. (Photo courtesy of L. Karczmarski.)
212
Shannon Gowans et al.
(Physeter macrocephalus) and elephant seals (Mirounga sp.) spending much of
their lives deeper and for longer times than killer whales can dive. It is
unlikely that this is an effective strategy for many dolphin species, as most
dolphins appear to make similar depth dives as killer whales (Schreer and
Kovacs, 1997 and references therein). In addition to killer whales, dolphins
may also be preyed upon by false killer whales (Pseudorca crassidens) and pilot
whales (Perryman and Foster, 1980); these predators probably represent less
of a risk than killer whales worldwide, but may be important delphinid
predators in tropical pelagic zones.
Most observations of shark predation come from coastal areas; however,
this is also where the greatest research effort has occurred (Heithaus, 2001b).
It should not be assumed that offshore areas have lower predation risk; in
fact, predation risk may be highest in the open ocean where there is no place
for dolphins to refuge (Wells et al., 1999a). Indeed, the large dolphin groups
found in the open ocean may be anti-predator responses (Norris and Dohl,
1980; Wells et al., 1999a). Large-bodied sharks are capable of hunting adult
dolphins, although the very large delphinids (e.g., killer and pilot whales)
may only be vulnerable to shark predation as juveniles. Several different
species of sharks are known to be important predators of delphinids, including great white (Carcharodon carcharias), bull (Carcharhinus leucas), dusky
(Carcharhinus obscurus) and tiger sharks (Galeocerdo cuvier). Bull sharks range
far up large rivers and prey upon riverine dolphins such as the tucuxi
(Heithaus, 2001b). Several long-term studies of dolphins indicate that
shark predation is an important source of mortality, and many individuals
also bear the scars of shark attacks that have resulted in injury but not
mortality (Heithaus, 2001b; Mann and Barnett, 1999; Wells et al., 1987).
Several deep-water sharks are also likely predators of delphinids. The
six-gill shark (Hexanchus griseus) is a large deep-water shark found along
continental shelves and upper slopes. Off South Africa, it preys upon
dolphins, and given its worldwide distribution is likely a major delphinid
predator elsewhere as well (Heithaus, 2001b). Similarly, there is indirect
evidence that broadnose seven-gill sharks (Notorynchus cepedianus) prey upon
coastal dolphins, and these primitive sharks may be important predators in
some southern hemisphere areas. Other species such as dusky and oceanic
whitetip sharks (Carcharhinus longimanus) are probably occasional predators
on delphinids. Little is known about the distribution and prey on most of
the oceanic sharks, although short-fin mako (Isurus oxyrinchus), Pacific
sleeper (Somniosus pacificus) and Greenland sleeper sharks (Somniosus
microcephalus) are suspected dolphin predators, and further research will
probably identify more shark species as dolphin predators (Heithaus, 2001b).
Levels of shark predation on delphinid communities vary between
species and locations. In general, offshore cetaceans have fewer scars attributable to unsuccessful shark predation attempts than inshore cetaceans
(Fig. 3.5 for an example of a shark bite scar). While this could indicate
The Social Structure and Strategies of Delphinids
213
Figure 3.5 Common bottlenose dolphin Tursiops truncatus in Tampa Bay, Florida,
showing shark bite scars. (Photo courtesy of Eckerd College Dolphin Project).
lower predation risk in offshore waters, it also could indicate that inshore
dolphins are more likely to survive a shark attack (Heithaus, 2001b). Some
dolphin species appear more vulnerable to shark predation than others.
Off thecoastofKwaZulu-Natal,SouthAfrica,humpbackdolphins(Sousachinensis)
had higher scar rates than Indo-Pacific bottlenose dolphins (Tursiops aduncus)
in the same area, perhaps suggesting greater predation pressure on the inshore
humpback dolphins (Cockcroft, 1991). Predation rates may also vary by location
with relatively high scar rates found on bottlenose dolphins (Tursiops sp.,
see Section 3.1 for brief discussion of Tursiops taxonomy) off Shark Bay,
Australia, moderate scar rates off Moreton Bay, Australia and Sarasota, Florida,
and very low rates in the Adriatic Sea (Heithaus, 2001a,b). Unsuccessful
predation attempts can influence reproductive success. Injured female elephant
seals rarely gave birth or engaged in mating behaviour, and therefore lost
reproductive potential (Heithaus, 2001b). No similar evidence exists for the
costs of unsuccessful shark attacks on delphinids, but they likely do occur
(Heithaus, 2001b).
Dolphins also display a number of behavioural reactions to shark presence (reviewed in Heithaus, 2001b). Often, dolphins avoid areas where
sharks are present (e.g., Corkeron et al., 1987; Saayman and Tayler, 1979).
When predatory sharks approach dolphin groups, strong flight responses
can occur (e.g., Connor and Heithaus, 1996). In other situations, dolphin
groups mob a shark and chase it out of the area (e.g., Heithaus, 2001b;
Mann and Barnett, 1999; Saayman and Tayler, 1979). Predation is often
cited as the major reason for the formation of dolphin groups (see Norris
and Dohl, 1980; Wells et al., 1980, 1987), but few data have been collected
regarding this issue.
214
Shannon Gowans et al.
Finally, sharks and delphinid predators, such as killer, false killer and pilot
whales, are likely to elicit at least some differences in dolphin prey detection
and avoidance strategies. Sharks tend to strike independently and without
warning, generally coming from depth and in this context can be viewed as
essentially ambush predators. Killer whales and other predatory delphinids
are social, and they chase down dolphins that are generally well aware of the
predators’ presence long before an individual is taken. While differences in
detection and avoidance strategies have not been studied in detail, we
expect that dolphins generally group tightly and use coordinated sensory
awareness in the face of shark predation. To thwart predatory delphinid
attacks, they are more likely to use strategies of hiding in surf zones or other
areas of shoreline, and outmanoeuvre the generally larger social predator.
Dolphins may also zigzag away in an attempt to confuse and disorient the
mammalian predator.
It is likely that there is an overlap in the strategies to avoid each type of
predator and that strategies will vary with differing habitats and predation
pressures. For example, in South Africa, where killer whale predation is rare
(although it occurs) but where shark predation is substantial, T. aduncus and
Sousa sp. form tight groups and both species are seen just behind the
breaking waves (Karczmarski, 1996; Karczmarski et al., 2000a). Heaviside’s
dolphins (Cephalorhynchus heavisidii), when they are closer inshore in the
early part of the day (Elwen et al., 2006), also stay just behind the surf
zone. In each of these cases, both strategies of tight grouping and hiding
in the surf zone are most likely a defence against shark predation. This area
of predator–prey interactions is ripe for study, as it may help explain
differences in grouping strategies in different predatory environments.
Dolphins typically form dynamic groups, where group size and membership change frequently; this may represent individuals or subgroups
altering their group size and structure relative to their activity state (e.g.,
foraging, resting and social; Gero et al., 2005) and current habitat in relation
to predation risk (Heithaus, 2001b; Wells et al., 1999a). In Shark Bay,
Australia, Indo-Pacific bottlenose dolphins were found in larger groups
when foraging in shallow water (which also represented areas of higher
predation risk) and when resting (a risky behaviour which typically
occurred in deeper waters where shark densities were lower). When shark
density was high, the dolphins foraged less in the risky but more productive
shallow water (Heithaus and Dill, 2002). Age and sex differences were also
noted in habitat use. Juvenile male dolphins foraged the most in the risky
but productive shallow water, while females with dependent young avoided
these areas. High-growth rates may be important for juvenile males to reach
sexual maturity, and foraging in a risky habitat may be more profitable for
them than for other age and sex classes (Heithaus and Dill, 2002).
In Sarasota, Florida, male common bottlenose dolphins (Tursiops truncatus)
that formed pairs were more likely to be found in the coastal Gulf of Mexico
The Social Structure and Strategies of Delphinids
215
than solitary males. The Gulf of Mexico represents an area of high prey and
predator density, likely indicating that the paired males were better able to
safely forage in the riskier and more productive habitat than solitary males
(Owen, 2003).
While studies on the relationships between dolphins and predators are
currently being conducted, there are vast knowledge gaps about dolphin
predators, their behaviours, predator distribution and dolphin responses.
Dolphins likely face predation pressures in all habitats, with the possible
exception of areas with very recent high shark fishing mortality
(e.g., bottlenose dolphins in the Adriatic Sea; Bearzi et al., 1997). Dolphins
in varying habitats have apparently evolved ways to deal with predation;
often these are behavioural responses, and in the open ocean there may be
few options other than forming large groups. Competition for food or other
factors may limit the option of forming large groups in coastal habitats.
Further studies explicitly investigating these questions will aid our understanding of delphinid sociality.
2.3. Foraging behaviour and diet
While reducing predation risk clearly plays a role in the evolution of group
living, individuals may also form groups in order to increase access to food
resources (e.g., Beauchamp, 1998). Living in a group can increase access to
food resources in several different ways at different times over the foraging
cycle. By coordinating search effort, groups are often able to more readily
locate large prey patches and minimize search time (e.g., Beauchamp et al.,
1997; Buckley, 1997; Ryer and Olla, 1995). Group living can also assist
predators in capturing and handling large or difficult to catch prey
(e.g., Blundell et al., 2002; Creel and Creel, 1995). Grouping may also help
individuals to defend a food resource against conspecifics, or individuals
of other species (e.g., Boinski et al., 2002; Gittleman, 1989).
Fish and squid represent the major prey species for most dolphins,
although other prey types have been recorded. Diet is often correlated
with body size, dentition and habitat. The largest of the delphinids, killer
whales, false killer whales, pilot whales and Risso’s dolphins (Grampus
griseus), have relatively few large teeth and feed upon large fish or squid
(Wells et al., 1999a). Killer whales also prey upon other marine mammals,
and false killer whales and pilot whales occasionally prey on smaller dolphins
(see Section 5.1.2; Jefferson et al., 1991; Perryman and Foster, 1980). The
small pelagic dolphins (e.g., Stenella and Delphinus) typically have a high
number of small interdigitating teeth in both upper and lower jaw. These
teeth are well suited to capturing small schooling squid and fish. Swimming
speed appears to be more important in these species, with a more streamlined body shape (Wells et al., 1999a). Inshore and coastal delphinids
(e.g., Lagenorhynchus and coastal forms of Tursiops) tend to have larger teeth
216
Shannon Gowans et al.
that may assist in capturing larger prey. These species tend to be less streamlined and have larger flukes and fins than pelagic dolphins (Wells et al., 1999a).
Dolphins have traditionally been considered to be rather opportunistic
foragers, preying upon many different species. However, most of this work
has been based upon identifying prey from stomach contents. More detailed
work studying the foraging behaviour and diet of free-ranging dolphins has
indicated that different communities have different prey preferences, which
often reflect local prey distribution (Wells, 2003). For example, inshore
bottlenose dolphins in Sarasota Bay, Florida, feed predominantly on pinfish
(Lagodon rhomboides) and do not appear to eat cephalopods. In contrast,
coastal bottlenose dolphins in the adjacent Gulf of Mexico waters commonly prey upon cephalopods (Barros and Wells, 1998). Prey preference
differences can also vary seasonally, in response to both prey distribution
changes and seasonal changes in metabolic needs (Wells et al., 1999a).
Lactating female dolphins often have different prey preferences, perhaps
to maintain the increased metabolic demands of lactation, or to avoid
leaving an unprotected calf at the surface while she forages at depth
(Bernard and Hohn, 1989; Cockcroft and Ross, 1990; Wells et al.,
1999a). More refined data collection reveals strong individual differences,
with individual dolphins showing specific preferences for prey type and
foraging strategy, even within the same community (e.g., Mann and
Sargeant, 2003; Nowacek, 2002; Sargeant et al., 2007). Our understanding
of delphinid foraging strategies will be enhanced with additional studies that
focus on individual foraging behaviour.
Cooperative foraging has also been described or suggested in delphinids.
Dolphins in the open ocean often swim in long parallel lines, up to several
hundred metres wide. This may facilitate finding large and somewhat
sparsely distributed schools of fish or squid (Würsig, 1986). Dusky dolphins
off Argentina appear to cooperate to concentrate schools of Argentine
anchovy (Engraulis anchoita). When a small group (6–15 individuals) of
dusky dolphins locates a fish school, they swim around and under the fish
school in an apparent attempt to tighten the prey and herd it towards the
surface. If no other group of dolphins joins, the small group corrals prey for
only several minutes, and may not actually feed. It is likely that a small group
of dusky dolphins cannot both contain the prey ball and feed on it
(a disruptive activity), and for successful feeding more dolphins need to
recruit to the patch. If other groups join, group herding and feeding can
continue for hours and involve up to 300 dolphins. Underwater observations
indicate that dolphins are corralling the fish against the surface of the water,
and that feeding does not begin until the fish are sufficiently contained.
There is evidence that dolphins cooperate to feed and that larger group
sizes are more efficient at foraging in this manner (Würsig and Würsig,
1980). Similar types of cooperative encirclement of schools of fish have
been observed in long-beaked common (Delphinus capensis), Atlantic spotted
The Social Structure and Strategies of Delphinids
217
(Stenella frontalis), Frazer’s (Lagenodelphis hosei) and Pacific white-sided dolphins (Lagenorhynchus obliquidens; Wells et al., 1999a; Würsig and Würsig,
1980). Other species probably also cooperatively locate and hunt prey, but
descriptions are unclear. Killer whales that eat marine mammals clearly
cooperate in hunting large prey and it is unlikely that a solitary killer whale
could successfully kill a large baleen whale (Baird, 2000). When cooperatively foraging, dolphins often utilize large prey patches (e.g., a large school
of fish, or large baleen whale) that contain sufficient resources for many
individuals to share, and this likely reduces foraging competition.
Prey species in inshore and coastal areas may be found individually rather
than in large schools. In these areas, dolphins have a greater tendency to
forage individually with little to no cooperation (e.g., Nowacek, 2002).
However, bottlenose dolphins in some areas forage by intentionally stranding themselves and their prey on muddy banks simultaneously and in a
coordinated manner; however, they do so only in small groups (Hoese,
1971). Foraging behaviour may differ between locations, attributable to
habitat differences. Humpback dolphins in Algoa Bay, South Africa, display
little cooperation and forage individually over rocky reefs (Karczmarski and
Cockcroft, 1999; Karczmarski et al., 1997). In tidal channels of the Bazaruto
Archipelago, Mozambique, they are seen in small cooperative groups that
chase their prey against sand banks and sometimes intentionally strand
(Peddemors and Thompson, 1994).
Inshore and coastal dolphins that feed predominantly on non-schooling
fish probably experience competition with other group members for food
resources. It is likely that dolphins most often experience scramble-type
competition with no individual being able to exclude any other from the
resource and all individuals share equally in the resource and its depletion.
However, there have been a few observations that suggest contest-type
competition may occur. Off Argentina, a smaller pod of killer whales (two
individuals) was apparently displaced by a larger pod (seven individuals)
several times at a pinniped rookery (Hoelzel, 1991). Similar displacement of
a killer whale pod feeding on a school of herring (Clupea harengus) was also
observed off Norway (Bisther and Vangraven, 1995). An apparent contesttype competition for favourable resting sites was observed in spinner dolphins
associated with remote Midway Atoll, far-western Hawaii (Karczmarski et al.,
2005). Contest-type competition may also occur within groups. A dominance hierarchy appeared to form in bottlenose dolphins foraging near
trawlers in Moreton Bay, Australia, with males receiving the most food
(Corkeron et al., 1990), although it was not clear in this study if females
were avoiding the area nearest the trawlers because of increased risk of
entanglement (Fig. 3.6). Few delphinid studies have specifically investigated
the potential for scramble or contest-type competition between groups or
group members; therefore, these few examples do not necessarily represent
rare situations but instead examples of behaviour that is rarely examined.
218
Shannon Gowans et al.
Figure 3.6 Common bottlenose dolphins T. truncatus travelling behind a shrimping
vessel, surfacing here just above where the weighted net is touching the bottom.
Dolphins feed directly in the net, as well as on fishes and invertebrates stirred up from
the bottom by the net’s passing. (Photo courtesy of T. Henningsen, with permission.)
However, primate studies indicate the importance of both scramble-type and
contest-type competition in the evolution of sociality (e.g., Isbell, 1991; Isbell
and Young, 2002), and it would be valuable to conduct studies on delphinids
to explicitly examine competition levels.
Pelagic prey species (typically fish and cephalopods) are typically found
in rare but profitable patches, with large areas of habitat with little to no
available food resources (Horwood and Cushing, 1978; Steele, 1985).
Cooperative foraging to locate and corral this prey may help reduce competition levels, although little work has been done to investigate this issue.
Interspecific displacement through contest-type competition may also
occur. Many studies suggest that different species of dolphins segregate
their habitats either temporally or spatially (e.g., Gowans and Whitehead,
1995; Griffin and Griffin, 2003; Hamazaki, 2002; Würsig and Würsig,
1980) and competitive exclusion may play a role.
2.4. Ranging patterns and daily movements
Delphinids display a wide range of patterns of daily and seasonal movements. Some dolphins, such as bottlenose dolphins in Sarasota, Florida,
remain relatively resident year-round in a small home range, although
some seasonal changes in distribution can occur (Owen et al., 2002;
Wells, 2003). In contrast, other dolphins appear to have seasonal migrations
The Social Structure and Strategies of Delphinids
219
with long range movements (e.g., short-beaked common dolphins, Delphinus delphis; Goold, 1998, and Atlantic white-sided dolphins, Lagenorhynchus
acutus; Reeves et al., 1999), although relatively few studies have investigated
individual movement patterns in these communities. However, all dolphins
appear to have relatively high daily movements often exceeding 100 km d 1
(e.g., Mate et al., 1994, 2005; Wells et al., 1999b; Würsig and Würsig,
1980), in contrast to primates and terrestrial carnivores which rarely
travelled more than 2 km d 1 (Wrangham et al., 1993 and references
therein). Delphinids routinely travel throughout the day, rarely resting,
and can be exposed to large areas of habitat, or their entire home range
on a daily basis, while primates and carnivores rest much more frequently,
and do not cover as large an area on a daily basis.
Scramble-type competition can constrain group size, as in larger groups
there are more individuals utilizing the same food resources. Because of
difficulties in directly measuring levels of competition, indirect measures are
often used. With increasing food density, scramble-type competition is
predicted to decrease while group sizes increase. When travel costs between
food patches are low, scramble-type competition is also predicted to
decrease because it is energetically less expensive to find another food
patch. In primates and carnivores, increased levels of scramble-type competition, measured by population density (an indirect measure of food density)
and mean daily movements (an indirect measure of travel costs) are correlated with decreased group size. While the data do not discern whether
increased group size drives increased mean daily travel or mean daily travel
leads to increased group size, the hypothesis predicts that variation in daily
travel drives the variation in group size (Wrangham et al., 1993).
2.5. Socioecology
Socioecology investigates the evolution of social strategies in relation to
environmental and ecological factors. Socioecological models that relate
social structures to ecological parameters have been developed for several
terrestrial taxa; however, models have not been developed in the same detail
for marine animals. Thus we will present key concepts from these terrestrial
models first, before looking at how marine models may be developed.
Primate socioecology has been investigated in the greatest detail, probably due in large part by a human desire to understand our own social
behaviour. The general model relates female social strategies to availability
of food and competition with conspecifics, while male social strategies are
determined in large part by female distribution and bonding. When highquality food resources occur in small defendable patches that only a discrete
number of individuals can feed on at one time (e.g., a fruiting tree where
there are only a few branches on which individuals can sit), contest-type
competition occurs for access to the patches. When females can cooperate
220
Shannon Gowans et al.
to exclude other females from the patch, the cooperating females benefit
and long-term bonds and social groupings between females evolve
(Wrangham, 1980). Males may be territorial, or defend a group of females.
When food density is particularly high, a single male can defend a territory
and the females within it. An extra male in the group would be unlikely to
increase the defensibility of the territory, but would increase food competition within the group; therefore single male groups form. When food
density is lower, home ranges become too large to be defended as a territory
and males guard female groups. In these situations, cooperation between
males is beneficial in competitive interactions with neighbouring groups,
and multi-male groups occur (Wrangham, 1980). When long-term bonds
form between females, subadult males are often forced to leave their natal
group, and reproductive skew among males is high (see Koenig, 2002, for a
recent review of primate socioecology models).
Ungulate socioecology has also been well studied following a seminal
paper by Jarman (1974) that related African antelope social organization to
their feeding style, ranging behaviour and anti-predator behaviour. Food
availability and dispersion varies greatly between feeding styles and some
species may range widely, undergoing annual migrations to remain in areas
of optimal growth. Small groups of small antelope feed more selectively
than large groups of large bodied antelope. Similarly, smaller groups of
antelope tend to remain inconspicuous from predators, while larger groups
flee (see Brashares et al., 2000, for a recent review).
Most carnivores tend to be relatively asocial; grouping only during the
breeding season, but some species live in social groups year round
(Gittleman, 1989). Territory acquisition appears to drive the evolution of
group living in carnivores, such that animals may group to exploit a highquality territory that can support them during periods of low resource
availability (see Bekoff, 2001; Clutton-Brock, 2002; Johnson et al., 2002,
for a review of different theories). Once in groups, individuals can benefit
from anti-predatory strategies or cooperative exploitation of resources. Typically, the smaller carnivores receive greater anti-predator benefits than the
larger carnivores, although many small carnivores are solitary (Gittleman,
1989). For some larger-bodied carnivores, infanticide may be the main form
of predation, promoting the grouping of related females and male–female
bonds outside the breeding season (e.g., lions, Panthera leo; Packer et al.,
1990). Competition for food resources may influence group sizes and sociality. With low food availability, carnivores typically have small litters and may
be solitary or live in small groups. However, when food is abundant, large
groups form with larger litters (Geffen et al., 1996; Wrangham et al., 1993).
For most group living carnivores, grouping increases foraging success and in
particular, cooperative hunting may widen the available prey base as groups
of animals can successfully hunt larger animals than solitary individuals
(Gittleman, 1989). The benefits of cooperative hunting in carnivores has
The Social Structure and Strategies of Delphinids
221
been controversial, with some studies reporting little benefit (e.g., lions;
Packer and Caro, 1997; Packer et al., 1990, but see Stander, 1992) while
others argue that sociality is driven by the energetic benefits of communal
hunting (Creel, 1997, 2001; Creel and Creel, 1995); however, it is clear that
some species engage in cooperative hunting.
2.5.1. Delphinid socioecology
It is perhaps useful to think of delphinids as having the social intelligence/
thought processing capabilities of primates, the foraging requirements of
carnivores and the predation pressures of ungulates. Given this, we believe
there is much to be gained by incorporating aspects of models based on
other taxa and relating these specifically to delphinids. A conceptual framework may assist in the understanding of delphinid group size and social
structure. In particular we focus on the role that the spatial and temporal
predictability of resources plays in the determination of ranging patterns of
individuals and communities. As movement from one area to another may
be energetically expensive or expose an individual to predation risk, individuals will likely minimize movements to range only as widely as needed to
find sufficient resources. In many situations, the availability of food
resources will be the most critical factor, although other resources such as
habitat safe from predation, mating partners and appropriate thermal habitat
may also be important. Females, in particular, are likely to minimize their
movements, as they attempt to ensure the survival of each offspring. The
observed range of movements of individuals then influences social strategies, including anti-predator and foraging behaviour as well as levels of
competition between individuals.
We propose a socioecological model that predicts when resources are
spatially and temporally predictable, dolphins will remain resident in
relatively small areas. However, in these smaller areas, scramble-type competition may increase and thus we predict that group sizes should become
smaller. In contrast, we predict that when resources are spatially and temporally variable, dolphins cannot remain resident but must instead range
widely to find sufficient resources. This increase in ranging behaviour is
predicted to be correlated with increased group size. In larger groups,
scramble-type competition can be reduced by cooperative foraging to
more effectively find and exploit large prey schools (Fig. 3.7). Additionally,
larger groups provide anti-predator benefits to group members. Thus variable resources will lead to the formation of large groups of dolphins which
range over large spatial scales. In contrast, predictable resources will lead to
the formation of small resident groupings of dolphins. We will investigate
this model by comparing observations of delphinid social strategies in
different habitats. None of these studies were designed to empirically test
these hypotheses; however, they can yield some important insights into the
evolution of delphinid sociality.
222
Shannon Gowans et al.
Figure 3.7 Socioecological model of the influence of the predictability of resources
on ranging behaviour, scramble-type competition levels and group size.
This framework builds on an earlier descriptive model by Wells et al.
(1980, 1999a) which emphasized ecology and anti-predator strategies as
important factors driving the social structure. In coastal and offshore waters,
habitat structure is relatively simple, leaving few areas to hide from predators. In these areas, group formation provides the only effective
anti-predator strategy and thus large groups are favoured. However in
inshore waters, with barrier islands, tidal marshes and shallow areas, dolphins
are able to refuge from predators and therefore there is less selection pressure
to form large groups to avoid predation (Wells et al., 1999a). Anti-predatory
strategies are clearly an important component selecting for group living in
delphinids; resource predictability and competition levels, however, are at
least equally important. The conceptual framework proposed in this chapter
explicitly includes both resource acquisition and anti-predator strategies.
Additionally, this framework is built upon predictions about delphinid
social strategy that can be tested in the future.
Some delphinid communities can be categorized into either the resident
or wide-ranging pattern. Several well-studied communities of bottlenose
dolphins (e.g., Shark Bay, Australia and Sarasota Bay, Florida; Connor et al.,
2000b) clearly fit the resident pattern, while many of the offshore communities (e.g., spinner, spotted and bottlenose dolphins in the Eastern Tropical
Pacific; Johnson and Norris, 1994; Pryor and Kang Shallenberger, 1991;
Scott and Chivers, 1990) fit the wide-ranging pattern. However, there are
other communities that are not easily described by this simple dichotomy.
Their ranging pattern and social structure appears to be an intermediate
form between the two extremes, and we believe that delphinid ranging
The Social Structure and Strategies of Delphinids
223
behaviour and social structure forms a continuum, with most communities
considered within the context of resident, wide-ranging or intermediate
(Fig. 3.7). Currently, the discussion of these community descriptions can
only be conducted in a relatively superficial, qualitative manner as few
communities have been studied in sufficient detail for more than generalities. However, we anticipate that the proposed framework may provide a
starting point to examine the evolution of social strategies in delphinids in
more detail. We also hope that quantitative analyses of this and more refined
frameworks can be conducted in the future, similar to recent quantitative
tests conducted by Brashares et al. (2000) on Jarman’s original model for
describing African antelope sociality ( Jarman, 1974).
3. Resident Communities
The model we proposed suggests that when resources are predictable
in space and time, individuals remain resident in a relatively small area
(Fig. 3.8). Long-term residency leads to individuals having intimate knowledge of their habitat, and therefore knowledge about where food resources
and predators are most likely to be found. Thus, individuals may be able to
reduce predation risk by avoiding high risk areas. Many of the areas with
predictable resources are inshore areas with estuaries and marshes; in these
areas the habitat is suitably complex for individuals to hide from predatory
attack. Hence, in many of these resident communities there may be less
advantage in grouping as an anti-predator strategy and group sizes may be
smaller. Within a small resident area, a smaller quantity of food is likely to
be available, which supports a smaller community of individuals. In many of
these areas, prey tend to be found individually or in small schools, so there is
an advantage to foraging alone or in small groups to reduce competition
with little to no advantage to cooperative foraging. Predation risk and
resource availability then lead to small communities of individuals composed of small groups. It is also true that in many inshore areas, there is
simply not enough physical space for a large group of animals to exist,
similar to the restriction of space faced by forest antelopes as opposed to
open savannah antelopes.
The proposed model predicts that within these small resident communities, there will be little advantage to a female to be in a group as she can
reduce foraging competition by being solitary. However, when a female has
a dependent calf, predation risk increases dramatically and females often
form nursery groups to reduce predation. In this situation, females may form
loose associations with each other, and may preferentially associate with
other females in similar reproductive state. As females tend to not form
strong bonds with each other, it may be possible for males to sequester a
224
Shannon Gowans et al.
Figure 3.8 Model conceptual framework describing the influence of predictable
resources on delphinid social strategies. There are few benefits to forming large groups
in resident dolphins, either as a way to reduce predation risk or to increase access to
food. Thus, communities of resident dolphins are relatively small and composed of
small groups. It is unlikely females will form long-term bonds, although males may
form long-term bonds in order to sequester females.
female for an entire oestrus cycle. This leads to competition between males
to gain access to females (Fig. 3.9). In these situations, male–male cooperation may be helpful to guard females from other males as is often seen in
chimpanzees (Pan troglodytes; Watts, 1998 and lions; Packer and Pusey,
1982). This is likely to lead to the development of strong bonds between
males. In this scenario, maternal kin bonds are most probably known
within the community as adult individuals are likely to associate at least
on occasions with their mothers and her dependent offspring.
The Social Structure and Strategies of Delphinids
225
Figure 3.9 Common bottlenose dolphins T. truncatus near shore at Isla del Coco,
Costa Rica. Male bottlenose dolphins (as shown here) often mouth and rake each other
with their teeth, as displays of dominance or social play. (Photo courtesy of B. Wˇrsig.)
3.1. Inshore bottlenose dolphins (Tursiops sp.)
Bottlenose dolphins occur in warm temperate to tropical waters, very
close to shore, in bays and even mouths of rivers as well as in the open
ocean. They are almost cosmopolitan in distribution, perhaps the most
widespread delphinid alongside the killer whale (Wells and Scott, 1999).
Bottlenose dolphins occur in different forms from about 2–3.8 m in size and
from very light to almost black with a countershaded body colouration.
They are not disruptively coloured, which suggests that their countershading allows for stealth in both approaching prey and avoiding predators.
Smaller animals tend to live in warmer waters, and large animals occur
in colder waters and further offshore, at least in the southeastern United
States. Because of their almost continuous distribution along shores in
many parts of the world, it is no surprise that there are apparent clinal
morphological and genetical differences, as well as differences between
inshore and offshore forms (Hersh and Duffield, 1990; Leduc et al., 1999;
Wells et al., 1987).
The taxonomy of bottlenose dolphins is unsettled, but it is presently
recognized that there are at least two forms sufficiently different to deserve
species status, the common bottlenose dolphin (T. truncatus) that occurs
throughout most tropical to warm temperate parts of the major oceans; and
the Indo-Pacific bottlenose dolphin (T. aduncus), that occurs from the eastern
coast of Africa, into the Red Sea and Persian Gulf, and to the western Pacific
Ocean as far north and east as Taiwan (Ross and Cockcroft, 1990; Wang et al.,
1999). Some evidence suggests that the South African population should be
considered a possible third species, but it has not yet been recognized (Natoli
et al., 2004). At present, it is believed that T. aduncus is largely if not
226
Shannon Gowans et al.
exclusively coastal. It has a longer rostrum than that of T. truncatus, and tends
to develop ventral spotting as it becomes sexually mature (Natoli et al., 2004;
Ross and Cockcroft, 1990; Wells and Scott, 2002).
There have been numerous behavioural/social organization studies of
bottlenose dolphins worldwide, since early work that described a general
fission–fusion society within communities close to shore (Argentina:
Würsig, 1978; Würsig and Würsig, 1977; Florida: Wells et al., 1987;
Texas: Shane, 1980). The fission–fusion nature of dolphin societies has
been compared superficially to societies of chimpanzees and spider monkeys
(Ateles geoffroyi; originally by Würsig, 1978, and see Connor et al., 2000b;
Wells and Scott, 1999, for an update), but with the details of comparisons
lacking until quite recently. The most detailed descriptions of bottlenose
dolphin societies have come from two areas; for T. truncatus since 1970 in
Sarasota Bay, Florida, and for T. aduncus since 1982 in Shark Bay, Western
Australia (summarized in Connor et al., 2000b; Wells and Scott, 1999). We
present social strategy synopses of these two study areas, with comparisons
from other studies as well. We recognize that differences in society structure
described below might be due to phylogenetic differences, or an expression
of environmental pressures. However, as the taxonomic classification only
recently separated T. truncatus from T. aduncus, we present information for
both in the following paragraphs, and we hope that future research will
clarify whether the social evolution of these species should be considered
separately.
Most of the well-studied bottlenose dolphin populations are composed
of relatively small communities which remain resident in a small area over
very long times. In Sarasota Bay, Florida, approximately 100 individual
dolphins reside in an area about 125 km2 and these dolphins and their
offspring have been documented to have used this area for almost 30 years
(Wells, 2003; Wells et al., 1999a). This community’s home range consists of
shallow estuarine bays and channels behind a row of barrier islands
that separate it from the Gulf of Mexico. There are several deeper passages
to the Gulf of Mexico and the dolphins occasionally venture into the
more open waters on the gulf side of the barrier islands (Owen et al.,
2002). Similar-sized communities of bottlenose dolphins with similar
home ranges have been studied in protected areas near Galveston, Texas
(community size ¼ 28–34 individuals; home range ¼ about 100 km2; Irwin
and Würsig, 2004), Moray Firth, Scotland (community size ¼ 130; home
range ¼ under 100 km2; Wilson et al., 1997, 1999), Moreton Bay, Australia
(community size ¼ 250; home range ¼ 350 km2; Chilvers and Corkeron,
2001), and Fjordland, New Zealand (community size ¼ 65; home range ¼
85 km2; Lusseau et al., 2003). In these areas, groups are typically composed
of a small number of individuals (mean group sizes range from 3 to 10
individuals; Connor et al., 2000b, and references therein). However, most
of these communities are not isolated or closed. Instead other communities
The Social Structure and Strategies of Delphinids
227
of bottlenose dolphins occur in adjacent habitats, and non-resident individuals are often observed within one community’s core range area.
Predation risk is ever present for most bottlenose dolphins, and shark
predation has been well documented in Shark Bay, Sarasota, South Africa,
and elsewhere (Connor et al., 2000b). Interestingly, in the Moray Firth of
Scotland and in Fjordland, New Zealand, two areas of occurrence at the
very extremes of bottlenose dolphin ranges, the occurrence of sharks and
other predators appear to be very rare or non-existent. In Moray Firth,
dolphins habitually show strong aggression towards harbour porpoises
(Phocoena phocoena; Patterson et al., 1998), and may limit harbour porpoise
occurrence in the area by their aggression. It is unknown whether the
absence of predation risk allows them the flexibility (i.e., time and energy)
to show such strong aggression towards another species.
During the summer months when tiger sharks are present in Shark Bay,
Australia, dolphins occur in larger groups over shallow waters where sharks
are more common. Mothers with calves avoid shallow waters when sharks
are present, while juvenile males commonly enter these areas of rich food
resources (Heithaus and Dill, 2002; Mann and Watson-Capps, 2005).
Similarly, paired male dolphins off Sarasota, Florida, are more commonly
observed in coastal gulf waters, which are more productive, and these paired
males have higher shark predation risk than do solitary males (Owen et al.,
2002). Such observations indicate that dolphins in both communities have
intimate knowledge of their core habitat and alter their behaviour and
distribution in relation to predation risk and prey densities.
Bottlenose dolphins have diverse food aggregating and capturing techniques from feeding singly to highly coordinated groups. During solitary
foraging, they tend to feed on solitary, bottom- or rock-dwelling nearshore
prey; when feeding in coordinated fashion, it is often on schooling fishes
(Würsig, 1986). Bottlenose dolphins feed in association with shrimp and
other trawling activities worldwide, having learned to take advantage of
debilitated by-catch (Corkeron et al., 1990), as well as to take fishes and
other prey directly from nets (Cox et al., 2003). They even feed in association with humans, when dolphins drive prey towards fishers’ nets near
shore, and both dolphins and humans appear to benefit from the coordinated manner of interspecies activity (Pryor et al., 1990). In Shark Bay, some
dolphins wear sponges on their rostra, as an apparent tool to poke at prey in
crevices (Smolker et al., 1997); in the Bahamas, dolphins detect prey under
the sand, with their echolocation, and then ‘corkscrew’ into the substrate to
obtain the prey (Rossbach and Herzing, 1997); and in small estuaries and
marshes of the southeastern United States, dolphins coordinate lunges to
chase prey onto sand banks, where fish can be snapped up while they are out
of the water. The latter example is one of especially pertinent societal
learning, cooperation, role playing and potential reciprocity, but details of
interactions of known and potentially related individuals have not yet been
228
Shannon Gowans et al.
described (Petricig, 1995). Community level differences in foraging strategies appear to be common among bottlenose dolphins and are likely linked
to prey differences between areas (Mann and Sargeant, 2003; Sargeant et al.,
2007). However, strong individual differences in foraging strategies have
been documented in Sarasota, Florida (Nowacek, 2002) and Shark Bay,
Australia (Mann and Sargeant, 2003; Sargeant et al., 2007), and these
strategies may be passed from mother to offspring.
There is some evidence that bottlenose dolphins occur in smaller groups
when feeding on lone or scattered prey, and in larger aggregations when
feeding on schooling prey (often in deeper water and further from shore),
but these assertions have not been definitively tested (Acevedo-Gutierrez,
2002; Acevedo-Gutierrez and Parker, 2000; Würsig, 1978). The general
relationship is probably highly variable. Since larger groups tend to occur in
somewhat deeper offshore waters, the relationship to foraging on schooling
prey is confounded by a potential need to be in larger groups due to deepwater shark predation. Bearzi et al. (1997) noted that bottlenose dolphins of
the Adriatic Sea east of Italy do not occur in larger groups in offshore waters,
where there is very little to no shark predation as to sharks having been
overfished in that area of the Mediterranean. On the other hand, in Shark
Bay, Australia, shark predation is higher in coastal than offshore waters, and
dolphins occur in variably sized but generally small groups offshore
(Heithaus and Dill, 2002).
Many but not all females in Shark Bay and Sarasota occur in bands with
close associations, of several to about 13 individuals, although there is
fluctuation of band membership from day to day. This is because females
have a larger network of other females (and males) with whom they
associate with less frequency than with band members. Some females
occur solitarily more often than they are associated with others, and there
is some evidence that lone female or females and their most recent calves are
more vulnerable to shark predation than those occurring in bands or mixed
groups (Connor et al., 2000b).
In Shark Bay, males form alliances that last from days/weeks to several
years (up to 17 years, Connor et al., 1992; Krützen et al., 2004). First-order
alliances consist of long-term stable associations of generally two to three
males. These small alliances tend to be of related males, and they often
sequester or coerce certain maturing or adult females to stay with them, and
repeatedly copulate with them. Second-order alliances consist of several of
these first-order ones, to take females from other alliances or to protect
themselves from other male alliance raiders. These also tend to consist
of animals related to a higher degree than expected by chance. Highly
labile short-term ‘super alliances’ consist of one dozen or so males
that raid other alliances of females, but that, interestingly, do not have
greater-than-chance-expected relationships. Super alliances are short term,
and perhaps may be thought of as ‘gangs’ that coalesce to intimidate and
The Social Structure and Strategies of Delphinids
229
mate with females that might otherwise not be available to only one of the
individuals (Krützen et al., 2003). Alliance formation, presently described in
detail only for Shark Bay bottlenose dolphins (but known or suspected in
several other dolphin systems; see below), appears to have at least superficial
parallels in chimpanzee (de Waal, 1982; Nishida and Hosaka, 1996; Watts,
1998), spider monkey (Chapman, 1990; Chapman et al., 1995) and lion
(Packer and Pusey, 1982) male alliances.
Adult males in Sarasota are more often solitary (or in long-term closely
bonded pairs) than in Shark Bay, and more often in mixed-sex associations
with females and youngsters of variable ages. While there are often longterm pair bonds of unrelated males (Owen et al., 2002; Wells, 1991), there
may be no or few strong male alliances and there is less evidence of males
coercing unwilling females than in Shark Bay (Wells and Scott, 1999). It is
also possible that the tendency for more often being alone or in small groups
in Sarasota is due at least in part to a presumed lowered risk of predation by
sharks in Sarasota than in Shark Bay. The detailed descriptions of alliances in
Shark Bay (e.g., Connor et al., 2000b) may somewhat overshadow the fact
that males in Shark Bay are also found in association with mixed sex and age
groupings and that not all mating strategy by males rely on the formation of
alliances to sequester and coerce (at least at times) unwilling females.
Alliance formation is only one of several mating strategies, as males with a
variety of different association patterns successfully sire offspring (Krützen
et al., 2004), and there may be much more female choice and courtship
behaviour than has been described to date.
Male alliances have also been documented in Port Stephens, Australia
(Möller et al., 2001), and off the Bahamas (Parsons et al., 2003), but not in
the Moray Firth, Scotland (Wilson, 1995). Formation of male alliances may
be correlated with population density, as areas with high densities of
dolphins lead to increased competition between males for access to females,
and thus the only viable strategy for weaker males may be to form alliances
for a chance at reproductive success (Connor et al., 2000b). Observed
population densities of bottlenose dolphins support this hypothesis: in
Shark Bay, all or almost all males are found in some type of alliance and it
has the highest density of dolphins; while in Moray Firth, there is no
evidence for alliances, and density is low (Connor et al., 2000b). Recently,
modelling studies have been used to explore prevalence and group sizes of
alliances relative to rates of encounter of males and females, aspects of
resource utilization and presumed costs of switching alliances. These models
await detailed testing in nature, but appear particularly applicable to strongly
fission–fusion societies, as are most delphinids (Connor and Whitehead,
2005; Whitehead and Connor, 2005).
While all bottlenose dolphin societies studied to date show a varying
nature of affiliations and fluid group structure, this tendency is least strong in
a community of dolphins in Doubtful Sound, western coast of the South
230
Shannon Gowans et al.
Island, New Zealand (Lusseau et al., 2003). There, dolphins occur in a mean
school size of 17.2 individuals, appreciably larger than the generally <10
animals found in other areas, even after different definitions of ‘school’ or
‘group’ are taken into account. In Doubtful Sound, dolphins live in large
mixed-sex schools, with strong associations within and between sexes,
instead of the male alliances, lone males or lone females, and female bands
that are more common in Shark Bay and Sarasota. Many associations are
long lasting, with no permanent emigration or immigration into the community noted in seven years of study. The community structure is thus
relatively stable, and constant companionship is the most prevalent pattern
of association; however, group membership is still fluid, and groups fission
and fuse. Lusseau et al. (2003) hypothesize that this unprecedented
(for bottlenose dolphin societies studied to date) stability may be related to
the isolation of Doubtful Sound from other sounds and bays amenable
to dolphin habitation, and perhaps the low food productivity of the sound
system. Low interaction with others may be simply due to distance to other
communities, keeping the community of 100 or so dolphins in Doubtful
Sound relatively closed, and tighter associations form between those animals
that are present (Lusseau et al., 2003). Additionally, food requirements may
dictate that the dolphins spend more time foraging than time and energy
spent on sexual pursuits, alliance-related strategies and ‘politics’ of association (sensu de Waal, 1982; for chimpanzees) than they do in the more foodrich areas of Shark Bay, Sarasota, and several other bottlenose dolphin
habitats/locations. The interesting closed nature of the Doubtful Sound
community shows several similarities with what was recently discovered for
atoll-living spinner dolphins (Karczmarski et al., 2005 and Section 3.2).
In summary, bottlenose dolphins of Sarasota Bay live in rather small and
open society groups that travel little, probably due to the protected nature of
the inshore environment and with reliable food resources year-round. The
food resources are not so abundant, however, that they would support
larger schools of dolphins, as often seen in the open ocean. Perhaps we
might think of this as generally a scramble-type competition society.
Females and young travel little, and stay in the most protected bays, away
from storms and predators. Young and adult males travel more widely, and
it is assumed that adult males are in search of oestrus females in different areas
and communities. Bottlenose dolphins of Shark Bay have a somewhat
similar social system, except that males form alliances to sequester females.
Males may stay together in large part due to danger from sharks, and these
associations allow them to then overwhelm and ‘kidnap’ certain females
near or in oestrus. In the more isolated, predator-free and food-poor area of
Doubtful Sound, New Zealand, bottlenose dolphins form the most closed
mixed-sex society found in this genus to date. In general these observations
fit well with our predicted model (Fig. 3.8) forming relatively small resident
groups. The studies conducted in Sarasota and Shark Bay represent
The Social Structure and Strategies of Delphinids
231
the longest, most comprehensive studies of delphinid sociality and ecology.
However, there are still many gaps in our understanding of these dolphins.
Continued research on individual differences in foraging behaviour and
competition levels as well as predation risk will provide valuable tests of this
framework. Conducting similar long-term comprehensive studies investigating bottlenose dolphin sociality, foraging behaviour and predation risk in
other habitats will also further our understanding. The differences observed
between these communities of bottlenose dolphins may relate to differences
in predation pressure and resource predictability. Definitive tests of this
model await detailed measurements of predation levels and resource
availability.
3.2. Spinner dolphins (Stenella longirostris)
The spinner dolphin is a pantropical species, inhabiting tropical, subtropical
and some warm temperate waters in the Atlantic, Pacific and Indian oceans.
They are found from coastal to pelagic waters, but rarely they are located far
from deep-water access. Similar to most other members of the genus
Stenella, spinner dolphins are thin-bodied and streamlined, with a long
and thin rostrum, short flippers and small tail. They are countershaded
light below and dark above, with paintbrush stripes along their flanks
(Perrin, 1998; see Fig. 3.2).
Several subspecies are recognized (Perrin, 1998). The Gray’s (or longbeaked) spinner dolphin (S. longirostris longirostris) associates with tropical
island systems and represents a ‘semi-pelagic’ form. Two other subspecies,
the Central American (S. longirostris centroamericana) and eastern spinner
(S. longirostris orientalis), are known from the Pacific coast of meso-America
and pelagic Eastern Tropical Pacific, respectively (Perrin, 1998). Another
pelagic form, an apparent hybrid between S. longirostris orientalis and
S. longirostris longirostris, the so-called ‘white-belly spinner dolphin’ occurs
throughout most of the offshore Eastern Tropical Pacific (Perrin, 1998;
Perrin and Gilpatrick, 1994). Another subspecies, the dwarf spinner dolphin
(S. longirostris roseiventris), occurs in shallow coastal waters of Southeast Asia
from Malaysia to northern Australia (Perrin and Gilpatrick, 1994). The
degree of sexual dimorphism varies among the subspecies, with males in
general slightly larger than females, with the mean total length of most
populations around 220 cm for adults. In the eastern Pacific forms, adult
males develop a prominent post-anal keel and erect dorsal fin. In the whitebelly form, this sexual dimorphism is present but muted, and in the Gray’s
form, it is even less pronounced, but still distinguishable. Spinner dolphin
behavioural ecology has been studied off the west coast (known also as the
Kona coast) of the Big Island of Hawaii (Norris and Dohl, 1980; Norris et al.,
1994; Östman, 1994), off the island of O’ahu, Hawaii (Lammers, 2003), off
Moorea in French Polynesia (Poole, 1995), and most recently, in remote
232
Shannon Gowans et al.
atolls such as Kure and Midway Atolls at the far-western end of the
Hawaiian Island chain (Karczmarski et al., 2005). In all cases, the Gray’s
spinner dolphin was studied, a subspecies that in Hawaii is often nicknamed
‘Hawaiian spinner dolphin’ (hereafter referred to as ‘spinner dolphin’).
Population figures are known in some detail only for the remote atoll
communities, 120 and 260 dolphins at Kure and Midway Atoll, respectively
(Karczmarski et al., 2005), with each atoll approximately 10 km in diameter.
For the Big Island’s Kona coast, a population of at least 2000 dolphins
(possibly more) has been suggested (Würsig et al., 1994a), but a functional
community size and home ranges remain unknown.
In pelagic waters as well as near islands and atolls, spinner dolphins feed
at night on small mesopelagic fishes and squid that rise with the deep
scattering layer towards the ocean surface at night (Perrin and Gilpatrick,
1994); but at least one form, the dwarf spinner dolphin, feeds on benthic
fishes and invertebrates in generally shallow seas (Perrin et al., 1999).
The pelagic and semi-pelagic spinner dolphins probably face high shark
predation pressure, especially while feeding on the deep-scattering layer,
although this has not been clearly documented (Heithaus, 2001b). In all areas
studied, the semi-pelagic spinner dolphins use the inshore island habitat
(or atoll lagoons) for daytime rest and social interactions. It is probable
that sheltering in shallow water while resting during daytime reduces predation risk, and that the interaction between feeding in deep waters at night and
resting close to shore in day drives much of the observed social structure of
spinner dolphins in Hawaii (Norris and Dohl, 1980; Würsig et al., 1994b).
Off the Big Island of Hawaii, Gray’s spinner dolphins have a marked
fission–fusion system, with smaller groups of <100 individuals close to
shore during daytime, larger ones of several hundred (apparently aggregations of daytime groups) offshore at night, and then again in smaller
aggregations near shore on the next day. Group composition changes
from day to day, with some animals present on subsequent days, but others
having ‘switched to’ other groups in other areas, generally in adjacent or
nearby bays. Much of the earlier research was conducted in and near the
Kealake’akua Bay of the Kona coast; a shallow bay 2km wide by 1km long.
In this bay, mean group size was about 35 dolphins. The animals entered the
bay 0.5–2.0 h after sunrise, socialized and rested in the bay most of the day,
and left the bay with much social activity and leaping 1–3 h before sunset.
Thus, dolphins stayed in the bay longer in summer than winter, and
evidence from radio tracking (Würsig et al., 1994b) and stomach content
analyses (Norris and Dohl, 1980) indicates that dolphins travel to deeper
waters in the evening to feed on prey associated with the nighttime rising
deep-scattering layer. This supposition has recently been refined, with
information to indicate that while in a large aggregated school at night,
dolphins tend to forage in coordinated pairs at depth (Benoit-Bird, 2004;
Benoit-Bird and Au, 2003).
The Social Structure and Strategies of Delphinids
233
It has been suggested that spinner dolphin day–night behaviour and
group fission–fusion are related to efficiency of detecting and thereby
avoiding shark predation in deep water while concentrating on feeding at
night, and refuging from predators during social–sexual activities and resting
during daytime. In this scenario, it is best for dolphins to aggregate for
nighttime foraging, as more dolphins indicate greater chances of detecting
deep water sharks (Norris and Dohl, 1980; Norris et al., 1994; see Section 2.2
for predation theory). At the same time, bays provide excellent refuges during
the day, but bays only allow efficient use by a certain number of dolphins, thus
the fission–fusion system. Indeed, smaller bays than Kealake’akua Bay
tended to have smaller daytime resting schools, providing some support for
this idea. The situation may be somewhat similar to that of dusky dolphins off
the Kaikoura Canyon, New Zealand, that also feed on deep-scattering layerassociated prey at night and rest near shore during the day (Benoit-Bird et al.,
2004, see Section 4.3 for more details).
For spinner dolphins found off the coasts of the Big Island of Hawaii,
changes in affiliations of individual dolphins and bays used from day to day
may indicate that dolphins consider each other as part of the larger group or
community off a particular area, such as the Kona coast of the Big Island,
and that they do not need constant close affiliation with certain members of
that community. Although there might be some more tight bonds at the
intra-group level, beyond the obvious mother–calf bonds (and certain pairs
and trios of males, as suggested by Johnson and Norris, 1994; Östman,
1994), data gathered so far indicate a fluid society with dynamic groups of
constantly changing membership; naturally leading to easy changes in group
or school size from day to day.
Mating strategies of the Kona coast spinner dolphins remain little
known. Anecdotal observations and the muted sexual dimorphism indicate
that the basic system is one of multi-mate (polygynandrous), with both adult
females and males having multiple sexual partners, often within short
periods of minutes. If a dominance hierarchy exists, then we would expect
that a dominant animal may have greater access to certain other individuals,
but this is not clear from wild observations or from one detailed study in
captivity (Bateson, 1974). Constant social–sexual play, including homosexual activities by both sexes, indicates that close affiliations and mountings are
not necessarily reproductive, but may have a diverse social function that
includes strengthening of social bonds, social signalling, or simply play, as
known for several great apes (Stanford, 1998; White and Chapman, 1994).
Along with dusky dolphins off Kaikoura, New Zealand (Section 4.3) these
dolphins have protracted time to rest and socialize while not foraging or
feeding, and it is no surprise that they therefore engage in prolonged social–
sexual activities away from deep water dangers and food.
In island and atoll-living spinner dolphins, protected bays and lagoons
are important habitat for daytime rest and social activities away from shark
234
Shannon Gowans et al.
predation over deep water. There is some indication that size of rest area
determines approximate size of group, so that a large bay has a greater
carrying capacity than a smaller one (Wells and Norris, 1994). Since there
is fluidity in associations, a group that enters a resting bay on one day is
composed of somewhat different individuals than on days before and after.
Thus, there is considerable overlap in bays used along a larger island such as
the Big Island, Hawaii, although individuals appear to have core areas of
greater preference.
During the night, a consistent offshore food resource makes grouping an
effective anti-predatory strategy and may even allow for cooperative foraging (Benoit-Bird and Au, 2003). Thus, social structure of spinner dolphins
associated with a large island habitat, such as the Big Island of Hawaii, seems
to be driven by both resource acquisition and anti-predator strategies that
lead to a classic fission–fusion society with large, fluid foraging groups at
night and considerably smaller, but still fluid, resting groups during the day.
A considerably different social pattern has recently been described for
spinner dolphins associated with remote atolls at the western end of the
Hawaiian Archipelago, some 2000 km northwest of the main Hawaiian
Islands (Karczmarski et al., 2005). In these isolated small atolls, such as
Midway Atoll or Kure Atoll, spinner dolphins live in stable, bisexually
bonded societies of long-term associates. They are significantly genetically
differentiated from the spinner dolphins at the main Hawaiian Islands, with
greatly reduced genetic diversity compared to the dolphins at the main
islands (Andrews et al., 2006). In the atolls, both males and females form
preferential companionships, within and between sexes, and display complex social behaviours such as babysitting. Although the persistence
of specific associations (pairs and trios) is stronger for males than females
(L.K., unpublished data), the group membership seems remarkably stable.
Usually, the entire community of an atoll (approximately 120 dolphins at
Kure and 260 dolphins at Midway Atolls) occur in one coherent group,
with no obvious fission–fusion and no inter-individual changes in group
structure and fidelities from day to day (Karczmarski et al., 2005).
The protected lagoons of the atolls are used daily as resting grounds, and
there are well-defined favourite resting sites within each atoll. The dolphins
are well familiar with the bathymetry of the area, and use depth contours to
travel within the atoll lagoon and to move in and out of the atoll. Interactions between neighbouring atoll communities are rare and generally the
dolphins show high geographic fidelity to their specific atoll (Karczmarski
et al., 2005). Although movements between atolls approximately 100 km
apart were seen, they were infrequent, never involved the entire community or solitary individuals, but rather a subgroup, possibly a basic social
unit of about 30–60 individuals. These movements are rare enough that
social divergence between neighbouring atoll communities may develop
(Karczmarski et al., 2005), but still sufficiently frequent to facilitate gene
The Social Structure and Strategies of Delphinids
235
exchange (Andrews et al., 2006). At Midway Atoll in 1999, a group of
dolphins transferred from Kure Atoll, and individuals from the two communities remained socially discrete for several months. Apparent aggression
and/or aversion were observed, with the resident community chasing the
immigrant group away from primary resting sites into less favourable areas at
the atoll’s rim, suggesting contest-type competition for favourite habitats
(Karczmarski et al., 2005) and even the possibility of territorial behaviour.
Three years later, although seemingly integrated into one larger community, the ‘residents’ and ‘immigrants’ still showed considerably stronger
grouping within their original community memberships.
This social pattern of atoll-dwelling spinner dolphins is considerably
different from the pattern observed off the main Hawaiian Archipelago,
and seems more akin to such systems as killer whales and long-finned pilot
whales (reviewed in Connor, 2000; see also Section 5.2) where individuals
remain in long-term groups. These differences in spinner dolphin sociality
correspond to the geographic separation and habitat variation across the
Hawaiian Archipelago; Karczmarski et al. (2005) suggest that geographic
insularity and the availability of sheltered shallow-water daytime rest sites
influence this difference in overall society structure. In the main Hawaiian
Islands, each island provides a mosaic of closely located near-shore environments with several suitable resting habitats in close reach, and each with the
capability to hold a certain percentage of a nighttime feeding group.
Individuals frequently change daytime resting sites, depending on rest site
availability, and group membership varies over time (Würsig et al., 1994b).
In far-western Hawaii, suitable resting habitats are restricted to atoll lagoons,
limited in size, and separated by large stretches of open pelagic waters with
potentially high risk of shark predation. Individuals do not switch resting
sites, as it becomes energetically advantageous to spinner dolphins in the
remote Hawaiian atolls to remain philopatric rather than travel to other
atolls. Thus, there is stability in the community membership and little or no
fission–fusion, with long-term inter-individual bonds and complex networks of social interactions. The geographic isolation and small size of
remote atolls trigger a process in which the fluidity of the fission–fusion
spinner dolphin society is replaced with long-term group fidelity and social
stability (Karczmarski et al., 2005).
In the open ocean, spinner dolphins occur in schools of hundreds to a
few thousand, often in multi-species associations with pantropical spotted
(Stenella attenuata), striped (S. coeruleoalba) and common (Delphinus sp.)
dolphins (Perrin and Gilpatrick, 1994; see Section 4.1). The group dynamics and social patterns of the more offshore forms of spinner dolphins remain
largely unknown, although considerable geographic variation in the mating
system has been suggested based on testis size (Perrin and Mesnick, 2003).
These spinner dolphins do not depend on the available resting sites for their
daily activities, as they are frequently long distances from the nearest island
236
Shannon Gowans et al.
or atoll. Therefore, we might expect a different pattern of association than
those patterns observed off the Hawaiian mainland and atolls.
In summary, three very different patterns of social structure have been
described for spinner dolphins. Long-term social bonds and geographic
fidelity occur in communities associated with remote far-western Hawaiian
atolls, where critical resource (resting sites) is limited but available on very
predictable (non-varying) basis. It is advantageous for these individuals to
remain at a known and secure location forming relatively small communities of resident dolphins as predicted by our model (Fig. 3.8). Off the
main islands of Hawaii, individuals frequently switch daytime resting sites,
and association patterns can be characterized as a fluid, fission–fusion pattern. We suggest that the daytime resting sites may not always be available
(if they have already been filled by other dolphins); therefore, the more
variable availability of daytime resting sites may lead to wider ranging
behaviour and more fluid social structure. This pattern fits with our overall
model (Fig. 3.7); however as resource availability is less predictable for the
main Hawaiian Island spinner dolphins, they display some of the characteristics of more wide-ranging communities (as outlined in Section 4). Finally,
offshore spinner dolphins are found in large, often multi-species groups in
the open ocean where no daytime resting sites are available. The unpredictable availability of food resources leads these dolphins to display a very
different pattern of social organization (as predicted by overall model, see
Section 4 for more details).
Most of the research on spinner dolphin social strategies has been
conducted around Hawaii. Expanding research to other areas would facilitate an investigation of the role that daytime resting site variability plays in
the evolution of spinner dolphin social structure.
3.3. Comparisons with terrestrial mammals
While it is clear from the previous sections that there are differences in social
structure in resident dolphin communities, several general trends are evident. Community sizes tend to be relatively small, with individuals mostly
associating with only about 100 individuals. Group sizes also tend to be
smaller, typically <10 individuals are found within the same group. Foraging is typically either solitary or involves small cooperative groups, and
individuals do not tend to form large groups as an anti-predator strategy
(Fig. 3.8). This trend does not follow in the case of insular spinner dolphin
communities in the far-western Hawaiian atolls, most likely due to other
factors such as geographic isolation, the specific nature of the critical
resource (shelter habitat rather than food), and the proximity of pelagic
waters with both pelagic predators and prey.
Predictable resources have led to long-term residency in many terrestrial
mammals as well. Some of the best-studied examples are primates and
The Social Structure and Strategies of Delphinids
237
African antelope. While comparisons between bottlenose dolphins and
chimpanzee social structure have previously been made (e.g., Connor
et al., 2000b), we believe it is useful to extend these comparisons between
resident delphinids and resident terrestrial mammals in general. However,
several major differences exist between terrestrial and marine systems
(see Section 1.3.1 for more details), in particular the lack of territoriality
(but note a possible exception for atoll spinner dolphin communities, Section
3.2), and the decreased probability of infanticide in the marine system.
Among primates, competition for food resources appears to be an important factor influencing social structure (see Isbell and Young, 2002, for a
review). In particular, high levels of contest-type competition (between and
within group) can lead to the evolution of social structure resembling that
found in resident dolphin groups. Various models of primate social ecology
place different emphasis on the role of predation, population density and the
costs of dispersal. However, all models agree that clumped distribution of
food resources (such as a fruiting tree) can lead to contest-type competition, as
individual primates attempt to exclude others from accessing the resource.
When specific individuals or groups can exclude others from accessing the
resource, aggressive interactions are common and stable dominance hierarchies can evolve. Females cooperate with kin to improve or maintain status
within these hierarchies and female philopatry is common (see Isbell and
Young, 2002, and references therein for primate socioecology models).
Primates that fit this general pattern include vervets (Cercopithicus aethiops),
baboons (Papio sp.), macaques (Macaca sp.) and chimpanzees.
Because similar types of primate societies may be driven at least in part by
contest-type competition, what is the potential for contest-type competition to occur in resident delphinids? Male herding and sequestering
of female bottlenose dolphins represent one of the clearest examples of
contest-type competition, where males compete for access to females, and
actively attempt to exclude other males from the resource (Connor et al.,
1992), although male coalitions are not the only successful strategy in this
community (Krützen et al., 2004). Similar contest-type competition may
occur in spinner dolphins for access to suitable rest areas, especially in the
Hawaiian atolls where groups acted antagonistically towards each other
(Karczmarski et al., 2005). It has been expressed (Würsig et al., 1994a) that
spinner dolphin groups avoided a rest area already occupied by another
group off the Kona coast of Hawaii, and this avoidance could be due to
perceived or real competition from those already in the resting area.
Another interesting possibility of contest-type competition for profitable
habitat occurs in bottlenose dolphins along the southeast coast of the United
States. Much of this coastline is fringed with barrier islands, and behind the
barrier islands are relatively small, isolated communities of dolphins, best
studied in Sarasota Bay. However, a larger ecotype of bottlenose dolphin
resides just offshore in these coastal waters. These dolphins range more
238
Shannon Gowans et al.
widely than the inshore dolphins and have different diets (see Section 4.2 for
more details). Occasionally, the inshore and coastal dolphins will interact,
but usually they are separated by the barrier islands. It is possible, although
not tested, that the inshore dolphins exclude the coastal dolphins from the
shallower waters, which may represent a safer habitat with more options to
avoid predation. Alternatively, the coastal dolphins may exclude the inshore
dolphins from the more productive offshore areas. Overall, it is unlikely that
contest-type competition plays as important a role in delphinid society as it
does in primates, given the difficulties of defending resources in the marine
environment, and based on the limited observations made to date.
Many species of African antelope show some similar patterns of residency and sociality, as the resident dolphins described above. Antelopes
such as reedbucks (Redunca redunca), bushbucks (Tragelaphus scriptus) and
oribi (Ourebia ourebi) tend to feed selectively on specific plant parts found
within their small home range. These antelope do not have to range widely
to find sufficient resources, and their relatively small group size limits
competition for food resources. Their habitat is relatively complex, and
they attempt to avoid detection by predators. Female groups do not appear
to be territorial, nor do they form long-term bonds. Breeding males appear
to hold territories and actively exclude other breeding males (Brashares
et al., 2000; Jarman, 1974). Resident dolphins appear to have relatively
selective diets, preying specifically on only a few species of fish, which may
locally be abundant or profitable to catch. Additionally, individual resident
dolphins appear to be highly selective, with individual preferences for prey
type and foraging strategy (Mann and Sargeant, 2003; Nowacek, 2002;
Sargeant et al., 2007). This dietary selectivity may serve to reduce competition between community members, although this assertion has not been
tested. Small African antelope have many strategies to avoid being detected
by predators (Caro et al., 2004). Bottlenose dolphins may use similar
strategies to reduce predation. In Sarasota Bay, bottlenose dolphins echolocate at lower rates than in other areas (Jones and Sayigh, 2002) perhaps to
avoid attracting sharks. Instead, the dolphins appear to use passive acoustics
to find prey (Gannon et al., 2005) and relatively rarely echolocate.
Three hypotheses have been proposed to explain the evolution of group
living in carnivores, and all relate to the acquisition of territories (Bekoff,
2001) and thus are unlikely to give much insight into the evolution of social
groups in delphinids. However, once carnivores form groups, individuals
can benefit from anti-predatory strategies and/or cooperative exploitation
of resources, and these studies may assist us in examining social strategies in
delphinids. Many of the cat species are solitary, as competition for food
resource likely prohibits group formation. Resident delphinids appear to be
predominantly individual foragers, with little to gain from cooperative
foraging. However, as dolphins face much lower transportation costs than
terrestrial species they may be able to forage solitarily but then aggregate for
The Social Structure and Strategies of Delphinids
239
social behaviour or anti-predator protection, while terrestrial carnivores
which forage solitarily usually remain solitary except for breeding.
Resident dolphins tend to live in relatively small communities composed
of small fluid groups. These dolphins do not range widely, probably due to
the reliability of food resources. Although food resources are reliably
located, they are not abundant and therefore the community size is limited
and individuals are more likely to benefit from solitary foraging. In this
protected inshore habitat, they rely on their ability to hide and avoid
predators, rather than using grouping as an anti-predator strategy. The
characteristics of this habitat result in few benefits to forming large groups,
and many benefits to being solitary or in small groups. There are relatively
few advantages to females forming long-term bonds, for cooperative foraging, babysitting or predator avoidance. This permits the possibility that
males may be able to sequester reproductively active females, and prevent
other males from mating with her. When competition for females is high,
males may benefit from cooperating to herd females, as has been observed in
bottlenose dolphins. Characteristics of the habitat of these resident dolphins
are similar to some African antelope and primates such as vervets, baboons,
macaques and chimpanzees. Explicit examination of the role of contest
versus scramble-type competition in resident delphinids and further investigation of anti-predator strategies may increase our understanding of the
role of ecological characteristics in the evolution of resident delphinid social
strategies and their similarities with terrestrial mammals.
4. Wide-Ranging Communities
When resources such as food are unpredictable or at lower concentration, individuals must range further to find sufficient resources. The lability
of resources in offshore areas may be due to fluctuations in oceanographic
conditions over varying time scales such as the short-term presence of
eddies, seasonal fluctuations or long-term variability such as El Niño Southern
Oscillation (ENSO) events. The home range of individuals will depend on the
persistence and severity of these fluctuations. However, it is likely that predation risk and foraging habitat are less predictable over the entire range, and
individuals are less likely to have intimate detailed knowledge of every aspect of
their home range. Thus, it may be more difficult to reduce predation risk,
simply by avoiding high risk areas. In fact, in these situations, predators are
likely to be as mobile as their prey and be less predictable in distribution.
Additionally, most of these wider ranging communities are found in
more open unstructured environments where there are few places for refuge,
and forming groups may be the only reliable (or available) anti-predator
strategy (Fig. 3.10).
240
Shannon Gowans et al.
Figure 3.10 Model conceptual framework describing the influence of unpredictable
resources on delphinid social strategies. Large group sizes are favoured to reduce predation risk and access food resources.Within these large groups, male bonds are unlikely
to form; however, females may form long-term bonds if cooperation between females
increases the survival of their offspring.
Larger home ranges are likely to support a greater number of individuals;
however, prey is often found in larger schools with vast areas in between
schools with little to no prey. In these situations, cooperative foraging and
herding of fish or cephalopod schools may be advantageous. When prey
schools are very large, foraging competition may be reduced as food is
superabundant. If cooperative foraging is present, then individuals may
form long-term bonds either with kin (kin selection) or non-kin (reciprocal
altruism). These foraging and anti-predator strategies may lead to larger
communities of larger groups of dolphins.
In these situations, males and females receive benefits from being in
groups, and the observed groups are often very large with males and females
The Social Structure and Strategies of Delphinids
241
present. These large groups make it difficult for a male or a small group of
males to sequester a female throughout an entire oestrus cycle, and polygynandry is the most likely mating strategy. As it is difficult to sequester
females, there is little advantage to forming strong bonds between males,
and male–male associations may be more weak and labile. In contrast,
females may form long-term strong bonds if cooperation is required to
successfully raise offspring, as has been suggested for sperm whales
(Whitehead, 1996). However, if cooperative foraging does not occur, or
does not require extensive coordination of individuals, then females
(and males) may form no long-term associations, but simply display many
short-term affiliations with a large number of individuals.
4.1. Eastern Tropical Pacific dolphins (Delphinus and
Stenella sp.)
In the open ocean, the smaller delphinids, such as common, spinner, striped
and spotted dolphins, tend to be in very large groupings of hundreds to
thousands. Especially in the tropical Pacific, there is a tendency for several
species to travel together, and it has been hypothesized that such multispecies aggregations may more efficiently detect sharks while feeding at different depths or time of day. For example, pantropical spotted dolphins tend to
feed during daytime and spinner dolphins at night, so it has been surmised
that aggregations composed of the two species essentially trade times of
increased vigilance, while muting competitive interactions related to
finding and securing food (Norris and Dohl, 1980). There is a general
but not complete tendency for the larger delphinids, such as Risso’s dolphins, pilot whales and false killer whales, to occur in smaller schools, and it
is generally surmised that this is possible for them as they are not as
vulnerable to shark predation due to their larger size (summarized by
Wells et al., 1999a).
Little is known about community structure in these large, wide-ranging
groupings of offshore dolphins. While techniques, such as photoidentification and genetic analysis of biopsy samples, have revealed great
insight into the lives of coastal dolphins, these techniques are difficult to
apply to offshore animals. It is difficult and expensive to locate and follow
groups of dolphins over long periods. It is also difficult to apply photoidentification techniques to large open populations. Therefore, our knowledge of offshore dolphin societies is much restricted. Insights into offshore
dolphin sociality have come from brief observations and inferences drawn
from morphometric analysis of by-caught dolphins (e.g., Perrin and
Mesnick, 2003; Pryor and Kang Shallenberger, 1991).
Open ocean dolphins travel great distances of hundreds (to low
thousands) of kilometres, often in remarkably short periods (Reilly, 1990).
It is a widely held view that large scale travel by large schools allows efficient
242
Shannon Gowans et al.
uptake of limited resources in the relatively low productivity of tropical seas,
much as large herds of Serengeti Plains migrating grazers ( Jarman, 1974)
need to keep moving so as not to deplete their resources in one area.
However, another reason for moving may have to do with not attracting
predators such as large sharks and (potentially) killer whales to a specific
location, as has been suggested for many systems (Brown et al., 1999; Ripple
and Beschta, 2004). Dolphins in the open ocean travel together as a mixed
sex and age school, and it is likely that all social–sexual interactions go on
within the large school, but with subgroupings of certain age and sex
spread throughout the school, and with fission–fusion interactions among
subgroups (Dohl et al., 1986; Pryor and Kang Shallenberger, 1991).
Most dolphins of the open ocean show muted sexual dimorphism,
indicative of a polygynandrous (or multi-mate) society. Recently, Perrin
and Mesnick (2003) provided an evaluation of two geographic forms of
spinner dolphins, the highly sexually dimorphic eastern form, where adult
males develop a tall forward canted dorsal fin and prominent postanal keel;
and the whitebelly form that shows some slight sexual dimorphism
(see Section 3.2 for details on different forms of spinner dolphins). They
reanalysed morphometric data (Perrin et al., 1991) in conjunction with
testes sizes (Perrin and Henderson, 1984) in these two forms, from an
extensive database of thousands of animals killed in the tuna-purse seine
fishery. The sexually dimorphic eastern form has small testes, while the
more monomorphic form has large testes, and the combination of data
allows the reasonable interpretation that there is more polygyny (i.e.,
males competing for access to females in some manner) in the eastern
form, and more of a multi-mate strategy of potential sperm competition
(Brownell and Ralls, 1986; Kenagy and Trombulak, 1986) in the whitebelly
form. The eastern spinner form lives in one of the most productive tropical
ocean regimes, just north of the equator (Ballance et al., 1997). Perrin and
Mesnick (2003) hypothesize that the high productivity may allow females to
form close-knit, long-term bonds (this supposition lacks observational data
at this point), and may have led to males needing to compete physically for
females. Furthermore, the high productivity may have released males from
foraging time to make polygynous competition possible. In the less productive waters occupied by whitebelly spinners, perhaps female bonds are less
tight, there is more overall coordination of protection from predators and
foraging together, and a polygynandrous strategy of multi-mating (sperm
competition) is the norm. The least sexually dimorphic form of spinner
dolphins occurs in Hawaii, other island and atoll situations, and much of the
western and southern tropical Pacific. In Hawaii, it has also been suggested
from behavioural observations ( Johnson and Norris, 1994) that the system
tends towards polygynandry (see Section 3.2).
Beyond reasonable inferences taken from morphological data on sexual
dimorphism and testes size, there is little direct information about social
The Social Structure and Strategies of Delphinids
243
strategies in slim-bodied open ocean pelagic dolphins such as spinner,
pantropical spotted and common dolphins. High resolution photography
from the air shows that such schools have subgroup structure with partial
separation of reproductive females, apparent adult males and mixed age/sex
groupings (Scott and Perryman, 1991). It is unknown how long associations
last, beyond the mother–calf association. However, it is unlikely that there are
tight bonds among the members of all or most of the school, and fission–fusion
associations exist within the overall school.
With formation of a large school the entire life of an individual may
remain within that ever-moving large community, broadly analogous to the
migration of wildebeest (Connochates tauriuns), caribou (Rangifer tarandus)
and human hunter-gatherer tribes (Berger, 2004). In the large school,
cooperation can develop, although we presently do not have the behavioural details to substantiate this likelihood of cooperation, nor knowledge
of persistence of long-term bonds beyond the mother–calf bond.
While there is limited information available about the foraging behaviour, social structure and predation risk of offshore Eastern Tropical Pacific
dolphins, what is known appears to support the proposed framework. These
dolphins form very large groups and range over wide areas. In the
future, detailed information on the foraging behaviour, social structure
and predation risk faced by dolphins of these communities would provide
a valuable test of this proposed framework.
4.2. Coastal bottlenose dolphins (Tursiops sp.)
Bottlenose dolphins (T. aduncus and T. truncatus) are found in coastal waters,
often close to shore, but also offshore of barrier islands. The inshore and
offshore ecotypes can be differentiated by morphological features (including
larger body size in the offshore type), cranial morphology, haematology and
parasitic infections (Hersh and Duffield, 1990; Van Waerebeek et al., 1990;
Walker, 1981). Genetic differences have been found in both the mitochondrial and nuclear genome between offshore and inshore dolphins, further
indicating their reproductive isolation from each other (Hoelzel et al.,
1998). The offshore ecotype has a relatively cosmopolitan distribution,
found in the Atlantic, Pacific and Indian oceans (Duffield et al., 1983;
Ross, 1984; Ross and Cockcroft, 1990; Van Waerebeek et al., 1990;
Walker, 1981).
In comparison to the inshore ecotype (see description in Section 3.1),
relatively few studies have been conducted on the offshore ecotype of
bottlenose dolphins. This is in large part due to the difficulties of studying
these wide-ranging dolphins, which do not stay within a small, logistically
manageable study area. Photo-identification studies have been initiated off
the coast of California (Defran and Weller, 1999; Defran et al., 1999) and
Florida (Caldwell, 2001); however, none of them have been able to resight
244
Shannon Gowans et al.
routinely either large number of individuals or re-identify them throughout
their range. It has also been difficult to estimate the number of dolphins in
these communities.
Prey preferences of coastal bottlenose dolphins have not been studied in
the same detail as inshore bottlenose dolphins, but some differences have
been noted. Off Florida, stomach contents of stranded dolphins show that
coastal and offshore specimens have higher proportions of cephalopods than
inshore forms (Barros and Wells, 1998). Off the California coast, bottlenose
dolphins appear to prey predominantly on surfperch (Embioticidae sp.) or
croakers (Sciaenidae sp.; Bearzi, 2003), whereas off South Africa, deep-water
reef fish were an important part of the diet of offshore forms (Ross, 1977).
Predation risk is poorly understood in coastal areas, but it is likely that these
dolphins are at least occasionally preyed upon by sharks (Heithaus, 2001b)
and killer whales (Jefferson et al., 1991), and these coastal habitats provide
few refuge opportunities.
Although the data are sparse, offshore bottlenose dolphins tend to display
a very different pattern of residency and social structure than the more
inshore bottlenose dolphins (described in detail in Section 3.1). These more
offshore bottlenose dolphins appear to be found in much larger open
communities which range widely and show relatively few long-term associations. For example, bottlenose dolphins off California (an area without an
inshore bottlenose dolphin community due to the lack of barrier islands) are
typically found in groups of 10–20 or more dolphins (Bearzi, 2003; Defran
and Weller, 1999). Very few individuals were resighted with previous
associates, indicating weak and fluid association patterns (Weller, 1991).
In the area off San Diego, it was estimated that approximately 250 bottlenose dolphins were present annually; however, low resighting rates
indicate that this is only one portion of a much larger open population
(Defran and Weller, 1999). Movements of identified dolphins support this
idea, as several long-distance re-sightings were recorded, including a rapid
movement of dolphins identified off San Diego which were photographed
north of Santa Barbara, some 286 km further north, only 14 days later
(Defran and Weller, 1999; Defran et al., 1999).
Studies of coastal bottlenose dolphins off the coast of Jacksonville,
Florida, show a similar pattern to those observed off California. In the
coastal waters of Jacksonville, bottlenose dolphins were observed in groups
of about 17 individuals that showed no signs of residency in the area. Some
of the dolphins photographed off Jacksonville were photographed 200 km
further north off Hilton Head, South Carolina, while others were photographed 440 km further north off Myrtle Beach, South Carolina. Some
individuals were sighted together on different occasions; however, data
were too sparse to quantitatively analyse social structure and most individuals did not display long-term associations (Caldwell, 2001). In contrast to
the California coast, barrier islands exist along the east coast of Florida, and a
The Social Structure and Strategies of Delphinids
245
different community of dolphins were observed behind the barrier islands.
Inshore and coastal dolphins did not interact, and genetic studies of these
dolphins indicated little mixing. The social structure of the inshore bottlenose dolphins near Jacksonville more closely resembles other communities
of inshore groups (see Section 3.1; Caldwell, 2001).
Detailed movement patterns of a few offshore type bottlenose dolphins
have been described based on satellite telemetry. Wells et al. (1999b) tracked
two males that had initially stranded along Florida’s coast after they were
successfully rehabilitated. After release, one male moved over 2000 km in
43 days from the Gulf of Mexico to offshore of Cape Hatteras, North
Carolina. The other male moved 4200 km in 47 days from Cape Canaveral
Florida, across deep water (5000-m maximum depth) to northeast of the
Virgin Islands. In contrast, two offshore bottlenose dolphins that were
captured, satellite tagged and released off eastern Australia remained over
shallow but offshore water (typically <50 m) and rarely dived below surface
waters (most dives were <5 m; Corkeron and Martin, 2004).
Bottlenose dolphins are found in coastal waters in other parts of the
world, including the east coast of South Africa. Here, Indo-Pacific bottlenose dolphins (T. aduncus) appear to show similar social structure to those
observed off Florida and California, with large groups several tens to a few
hundred individuals which range widely and do not show clear patterns of
long-term associations (L.K., unpublished data). However, it is much more
difficult to study wide-ranging communities as research effort must be
spread over a far greater geographic area. Thus, our understanding of
dynamics in these communities is much more limited and additional social
structure may be present, but undetected at this time.
In Bahamian waters, there are few inshore locations, but a community of
coastal ecotype of bottlenose dolphins (T. truncatus) has been well studied in
the shallow waters of Little Bahama Bank. Their physical characteristics,
social structure, ranging behaviour and site fidelity are similar to the inshore
bottlenose dolphins of Florida (Rogers et al., 2004; Rossbach and Herzing,
1999). Groups are typically composed of 3–5 dolphins, and show fission–
fusion. The strongest bonds were between mothers and calves, while female
associations varied with reproductive state. Long-term associations were
formed by some males (Rogers et al., 2004) and appear to be kinship based
(Parsons et al., 2003). This community appears to resemble the inshore
bottlenose dolphin pattern despite its offshore location. This group of
dolphins tends to forage individually on benthic prey (Rossbach and
Herzing, 1997), suggesting that when individuals forage on solitary
prey, dolphins do not form large cooperative groups, as observed in other
offshore bottlenose dolphin communities.
Bottlenose dolphins are also found further offshore in neritic and
pelagic waters. Little is known about the ranging behaviour, diet, predation
risk or social organization of these dolphins. Group sizes are typically
246
Shannon Gowans et al.
smaller (Connor et al., 2000b) than some of the other offshore dolphins
(see Section 4.1), but it is difficult to draw any conclusions from these
differences.
The behaviour of offshore bottlenose dolphins (Tursiops sp.) provides
further support for this framework; however, many details are lacking to
fully test the proposed framework. The work initiated off California and
Florida will provide a valuable test of this hypothesis. The observations of
bottlenose dolphins off the Bahamas provide valuable support for this
framework as the association patterns of these dolphins resemble the patterns
observed in inshore bottlenose dolphins (see Section 3.1) despite their
offshore habitat. However, bottlenose dolphins off the Bahamas and in
Sarasota and Shark Bay appear to forage individually on predominantly
non-schooling prey which tend to be predictably available (Mann and
Sargeant, 2003; Nowacek, 2002; Rossbach and Herzing, 1997; Sargeant
et al., 2007). In contrast, the framework proposed by Wells et al. (1999a),
which focused on habitat complexity and predation pressure, suggests that
the bottlenose dolphins off the Bahamas should form large groups to avoid
predation in their open ocean habitat.
4.3. Dusky dolphins (Lagenorhynchus obscurus)
Dusky dolphins occur discontinuously on both sides of southern South
America (in Peru, to as far north as about 7 S; in Argentina, to at least
about 38 S), the south-western coast of South Africa, the south island and
southern half of the north island of New Zealand, occasionally off Tasmania
and southern Australia, and off several south Atlantic and southern Indian
Ocean Islands (Brownell and Cipriano, 1999; Van Waerebeek and Würsig,
2002). They are a semi-pelagic species, preferring water no deeper than
about 2000 m, generally near shore, on or near continental or island
submarine shelves. They are found over expansive non-pelagic shelves
between New Zealand and the Chatham Islands (Gaskin, 1968), and
>200 km from shore east of Argentina (Dans et al., 1997). Yet they also
occur in bays and inlets, although only as seasonal residents and not as yearround resident communities (Würsig and Würsig, 1980 for Golfo San José,
Argentina; Markowitz et al., 2004 for Admiralty Bay, New Zealand).
Therefore, while we place them among the ‘wide-ranging’ communities,
they have elements of seasonal residency in some areas and elements of
intermediate-ranging patterns, as will be described in more detail below.
Their closest generic relative is the northern hemisphere Pacific white-sided
dolphin, and it is presently unclear how closely they are related to other
southern hemisphere and North Atlantic Lagenorhynchus species (Cassens
et al., 2003; Leduc et al., 1999).
Dusky dolphins are a small delphinid, generally <200-cm long (with
some specimens off Peru slightly >200 cm). Males are slightly larger than
females at the same age, and there is a tendency for males to have thicker
The Social Structure and Strategies of Delphinids
247
and more curving dorsal fins than females. This somewhat muted sexual
dimorphism hints at a polygynous society, but details are unknown.
Dusky dolphins tend to feed on fishes and squid of a large variety of
species, but generally with prey body lengths less than about 30 cm
(Cipriano, 1992). They are gregarious wherever they occur, in groups
from less than one dozen to well over 1000 individuals (Würsig et al.,
1997). Deep-water sharks, such as broadnose sevengill, great white and
shortfin mako, probably pose risks in much of their range (Crespi-Abril
et al., 2003; Heithaus, 2001b; Van Waerebeek and Würsig, 2002); and killer
whales take them off Argentina (Würsig and Würsig, 1980) and the east
coast of the south island of New Zealand (Constantine et al., 1998). Large
group sizes may be at least a partial response to predation, and use of near
shore habitats (see below) may minimize both shark and killer whale attacks
(Würsig and Würsig, 1980).
Dusky dolphin foraging and social strategies have been studied in some
detail in shallow waters of Golfo San José, Argentina (Würsig, 1982, 1986;
Würsig and Würsig, 1980), and Admiralty Bay, Marlborough Sounds, New
Zealand (Markowitz, 2004; Markowitz et al., 2004; McFadden, 2003), as
well as in deep waters of the Kaikoura Canyon, New Zealand (Benoit-Bird
et al., 2004; Markowitz, 2004; Würsig et al., 1997). Group structures and
foraging strategies are remarkably different in these two disparate habitats.
In shallow waters (to about 60 m) off Argentina, dolphins in summer
have a marked fission–fusion society. Approximately 300 dolphins live in
the community of Golfo San José and groups range about 20–100 km d 1
usually in zigzag movements, as they search for food. However, they may
move to other adjacent areas, up to about 300 km away, for days to weeks at
a time (Würsig, 1982). The general pattern for the community is to occur in
about 30 small subgroups (6–10 dolphins) in early morning, with the
subgroups covering an area over 100 km2 as they slowly travel within
about 1 km from the nearest other subgroup. It appears that when one
subgroup finds a school of fish (anchovy, Engraulis sp.), aerial leaping
behaviours increase, birds aggregate and dolphins from other subgroups
recruit to that area. It is thus apparent that dolphins detect activities of
other groups, and perhaps communicate, across distances approximating
1 km or more. The group size grows, at times with all 30 or so subgroups
coalesced after one or more hours. Dolphin subgroups appear to make
decisions as to which other subgroups to approach, presumably by an
assessment of intensity of activity with fish and birds. Some subgroups that
begin herding (or baitballing) are not approached, and these activities tend
to die out after 10–15 min (sometimes less), suggesting dolphins require
greater group size to efficiently herd the prey into a tight baitball that is
stable over time. Typically, dolphins that herd fish do not begin to feed until
the fish school is tightly balled. Dolphins rapidly move under and around fish
schools, often emitting bubble blasts from their blowholes to help herd
248
Shannon Gowans et al.
the school. They use the surface as a wall against which a fish school is pinned
(Würsig and Würsig, 1980).
As feeding progresses and group size becomes larger due to fusion of
smaller subgroups, dolphins display activities such as highly acrobatic leaps
(Fig. 3.11), head-outs (raising their heads out of the water, without jumping
clear of the surface), chasing each other and copulating (including homosexual activities within both sexes), and mouthing or rubbing a wide variety
of objects including the legs of marine birds, large fishes, small sharks,
humans in the water, the side of a research boat and inanimate floating
objects such as kelp or debris. These activities after intense feeding are most
frequent when most or all dolphins within an area are together, and this
high level of sociality can last for up to an hour or more. Generally late in
the afternoon, dolphins fission back into smaller subgroups, and stay in small
subgroups, close to shore, until the next day.
It is surmised that this diel cycle occurs so that at night dolphins rest in
shallow waters close to shore, away from shark and killer whale predation
and in groups not easily discovered. Fissioned groups thus allow for efficient
rest but also allow for enhanced capability of finding food in the morning, as
they range over a wide area. Fusion of groups allows for enhanced herding
and feeding, and during/after feeding the social activity is probably important for social bonding. During daytime feeding, mothers and calves tend to
feed early on, but segregate from the feeding-socializing group as play and
mating activities increase. It is unclear whether there is strong sexual
segregation in subgroups, but there appears to be a tendency for mothers
and calves to form separate subgroups (or nursery groups); as well as some
subgroups to be composed of all males and others to be of mixed sex and age
Figure 3.11 Dusky dolphins Lagenorhynchus obscurus are highly aerially acrobatic. Here
a male chases a female off Kaikoura, New Zealand. Coordinated, rapid salmon-like
leaps of this kind are often directly followed by a quick copulation. (Photo courtesy of
B. Wˇrsig.)
The Social Structure and Strategies of Delphinids
249
(Würsig and Würsig, 1980). While subgroups tend to have at least some
different memberships from day to day, one pair of dolphins that was
marked together in one subgroup in 1975 was sighted together almost
eight years later (Würsig and Bastida, 1986).
Among dusky dolphins in winter in Argentina, there is less fission–fusion
(and lower occurrence of Argentine anchovy), and most subgroups of 6–10
animals stay in shallow water day and night. It is surmised that winter nonaggregating feeding is of a more individual than highly cooperative nature
(Würsig and Würsig, 1980), but more work needs to be done to adequately
describe the social patterns of this community throughout the annual cycle.
Off New Zealand, dusky dolphins occurring in winter in shallow waters of
Admiralty Bay feed largely on schooling pilchards (Sardinops sagax), yelloweyed mullet (Aldrichetta forsteri) and sprat (Sprattus antipodum), also in association
with birds (Würsig et al., 2007). A community of about 150 dolphins at any one
time is present in Admiralty Bay, but typically split into subgroups of 3–12
animals (Markowitz, 2004). In this locale, dolphin groups range little on a daily
basis, moving within an area of only about 10–20 km in diameter. Unlike in
Argentina, there may be no concerted effort to move fish schools towards the
surface, and much feeding appears to occur deeper than 10 m below the
surface. This may be due to different behaviours of the dolphins or their
prey. Furthermore, there is no consistent and clear fusion of subgroups,
although they move towards each other at times. Certain known individuals
return to Admiralty Bay year after year (Markowitz, 2004). It is presently not
known why only particular dolphins make the trip, and it has been suggested
that use of this winter foraging habitat may be a cultural phenomenon passed
from one generation to the next among some, but not all, dolphins of the larger
society (Whitehead et al., 2004).
Dusky dolphins occur throughout the year in and near the Kaikoura
Canyon, on the east coast of New Zealand’s south island. This is an open
ocean environment, not unlike non-bay coastal areas of Hawaii, as
described for spinner dolphins. There can be between 500 to several
thousand dusky dolphins in the waters near Kaikoura at a time, the largest
groups usually in late autumn to early winter. Here, dolphins move north
and south along shore, travelling back and forth in an approximate daily
range of 20–50 km, but sometimes up to 100 km. Different dolphins are
observed near Kaikoura in the summer and winter. Some of the dolphins
identified in the summer near Kaikoura have been photographed in winter
in Admiralty Bay, several hundred kilometres to the north. Near Kaikoura,
group sizes are highly variable, from subgroups fewer than one dozen to
well over 1000 animals in one cohesive school. Subgroups, composed of
nursery groups, apparently all males, and others of mixed sex and age, often
travel on the periphery of the larger group of several hundred animals, as
far as 100 m to 10 km or so from the main group. There is no strong or
time-predictable fission–fusion (Würsig et al., 2007).
250
Shannon Gowans et al.
During the day, dusky dolphins near the Kaikoura Canyon travel back
and forth within several kilometres of shore, rest and socialize. There is very
little daytime feeding, with baitballing and associated birds seen only rarely;
on average less than once per year in the past 10 years of research. At dusk,
dolphins spread out into small subgroups, separated from others by about
20–100 m, but stay cohesive as a school, all travelling in the same direction.
They head into deeper water, generally along the escarpment of the
Kaikoura Canyon. There is a well-developed diel-vertically migrating deepscattering layer that rises towards the surface in these oceanic waters. Sonar
used to track both the deep-scattering layer and individual dolphins
(Benoit-Bird et al., 2004) shows that dusky dolphins do not begin to feed
until the layer is within 130 m of the surface, and prefer to feed at depths
<100 m. The deep-scattering layer-associated organisms are available for
the dolphins throughout all or most of the night and then the dolphins
return closer to shore just before or during dawn, when the deep-scattering
layer descends below their preferred diving range. We suspect that the
daytime preference for shallow water is to avoid the higher risk of
predation in deep water (as suspected for island-associated spinner dolphins,
Section 3.2).
While dusky dolphins off Argentina have intensive bouts of social–
sexual activity closely linked to a post-prandial state in both Admiralty
Bay and off Kaikoura, such bouts are less intense and are spread throughout
the daytime in sporadic brief (10–20 min) social activities. However, the
amount of socializing during the night for any of the three studied areas is
not known.
The mating system or systems are unknown for dusky dolphins. Slight
sexual dimorphism suggests that males may compete for access to females
and that some males gain more copulations of females in oestrus than do
others. However, the sexual dimorphism is indeed so slight that the system
may tend more towards poygynandry, or multi-mate, and that sperm
competition may be important.
In summary, dusky dolphins feed both in daytime on schooling fishes
and at night on mesopelagic fishes and squid, depending on habitat and food
availability patterns. In Argentina they have a fission–fusion society related
to foraging and refuge from predation risk, but do not have such strong and
constantly fluctuating subgroup membership in the shallow- and deepwater habitats in New Zealand. In deep water off the Kaikoura Canyon,
it is probably of advantage to remain in a large school both day and night,
since sharks and killer whales are present in these oceanic waters even
though the dolphins are quite close to shore (generally within 10 km). In
both the Argentine and Admiralty Bay inshore situations, groups can be
quite small without overt fear of predation. In Argentina, the smallest
groups, of generally fewer than 12 animals, stay very close to shore to rest
and socialize, and have been seen to retreat into the surf zone when killer
The Social Structure and Strategies of Delphinids
251
whales approach. However, the fission–fusion nature of feeding societies,
especially so during summer in Argentina when feeding on schooling fishes,
indicates that the larger group size may be important to efficient feeding at
least as much as to predator detection/avoidance strategies.
Overall, dusky dolphins show at least three different grouping patterns:
Off Kaikoura New Zealand deep-canyon waters they occur as a large
seasonally resident society; off Argentina over an expansive continental
slope, they occur in strong fission–fusion from small to large groups, with
much travel; and in Admiralty Bay New Zealand, they occur in small
seasonally resident groups. These patterns appear to track our general
framework very well. In Argentina, prey are unpredictably distributed and
dolphins need to range widely to find enough food. They form large groups
related to apparent cooperative foraging, but split into small groups and
move nearshore in order to rest at night, presumably to minimize predation
while they do not need to be in open water areas to feed. The large daytime
groups may also be of advantage to help detect predators, and to react faster/
more efficiently than in small groups in open water, but this is presently
unknown. Off Kaikoura, the large school envelope both day and night
would seem to allow greater vigilance of predators and may help enhance
long-term bonds and cooperative foraging. Here, prey are predictably
found at the edges of the Kaikoura Canyon, and movements are restricted
to within about 50 km along the shore. In the Admiralty Bay situation,
predictable food resources seem to be available in the complex inshore
environment of the area. Dolphins can hide nearshore in reaction to predators, but may be limited to the small groups sizes in which they are found
because of limits of food in this smaller-extent environment. Interestingly,
many of the same dolphins that feed at night in the Kaikoura Canyon in
summer in large groups move to Admiralty Bay to feed in daytime in winter
in small groups. They thus change strategies and grouping patterns in
flexible manner, and generally consistent with the framework.
4.4. Comparisons with terrestrial mammals
As described above, wide-ranging dolphins tend to be found in larger
groups than more resident dolphins, although much variation exists. Typically, these dolphins feed on sparsely distributed but abundant food, thus
reducing competition. Therefore, it is beneficial for individuals to form
large groups to increase the likelihood of detecting prey patches and or
reduce predation risk. Within these large groups, social affiliations are likely,
however, researchers have found these difficult to document.
By comparison, primate food resources rarely fit the pattern of rare but
abundant and instead high levels of competition (either scramble or direct)
restrict group sizes (Isbell, 1991; Isbell and Young, 2002). Travel costs are
high for primates, and consequently wide-ranging behaviour is rarely
252
Shannon Gowans et al.
observed (Isbell et al., 1999). However, patterns of social structure and
movement similar to that of wide-ranging delphinids can be observed in
large-grazing ungulates in open grasslands. Jarman (1974) suggested that
among African antelopes, those found in larger groups had a non-selective
diet of a broad range of grasses, fled or counterattacked predators and had
large body size. Their need for large quantities of low-quality food, which is
often seasonally variable, led to large home ranges on the open savannah.
However, when grasses are available they are typically fairly abundant, and
therefore competition for grass patches is relatively rare. Additionally, as
there are relatively few places to hide in the African savannah, increased
group size is an effective way to reduce predation risk. Recent quantitative
analysis of social structure and movements in African ungulates indicated
that pattern of large body mass, non-selective diet, large group sizes and
anti-predator behaviour holds true even when phylogenetic effects are
removed (Brashares et al., 2000).
Home range size often scales with body size ( Jetz et al., 2004; LaBarbera,
1989; Peters, 1983) as larger animals require greater food resources. However, attempts to correlate home range with body size in macropod marsupials revealed an interesting deviation, with home range size closely
associated with climate (especially annual rainfall) rather than body size.
Female marsupials living in the rainforest with high-annual rainfall had small
home ranges and foraged on predictable food resources. In these habitats,
males had larger ranges, which overlapped with the home ranges of several
females indicating that females were minimizing energetic costs related to
travel, while males maximized exposure to the number of females (Fisher
and Owens, 2000). In contrast, females living in the Australian desert, in
areas with low-annual rainfall, had large home ranges and tended to forage
on much less predictable resources. Males had similar sized home ranges as
did females, perhaps indicating that these home ranges cannot be increased
because of travel costs and males were not able to increase contact with
more females (Fisher and Owens, 2000). In some cases, these species will
form mixed sex groups, although sexual segregation is also common (Fisher
and Owens, 2000; MacFarlane and Coulson, 2005).
The unpredictable nature of food resources of the Australian desert and
the African savannah shares some similarities with the open ocean. While
ungulates and macropod marsupials have often been considered convergent
dominant herbivore species (Fisher and Owens, 2000), delphinids are
clearly not herbivores. However, patches of grasslands and large fish schools
do have some remarkable similarities. Locations of both resources can be
patchy, unpredictable and ephemeral. The effect of a small-localized rain
shower on an arid location can rapidly lead to plant growth, providing high
yield resources in relation to the surrounding area. Grazers can then able to
deplete the patch, or the patch dies back naturally without further rainfall.
Similarly, the arrival of a school of fish can suddenly increase food
The Social Structure and Strategies of Delphinids
253
availability in a discrete location and disappear quickly as it is either consumed by predators or moves to another location (or, for smaller delphinids,
below their preferred dive depth). In these situations, resources are locally
abundant and competition between group members (either herbivores or
delphinids) is reduced, and therefore relatively large groups can form.
In contrast, terrestrial carnivores rarely encounter highly dense but
sparsely distributed food patches. While food resources are patchy, they
are rarely sufficiently abundant that competition could be reduced. Instead,
competition for these resources restricts group sizes and in most cases
constrains carnivores to a solitary lifestyle (Geffen et al., 1996; Gittleman,
1989). Approximately 10–15% of the species within the carnivore class are
social (live in groups outside the breeding season); however, most live in
relatively small groups (<10 members; Gittleman, 1989). Although groups
cooperating to hunt can bring down larger prey than solitary individuals,
even large prey represent a limited food resource and competition for food
resources likely limits these groups from becoming larger (Creel, 1997;
Geffen et al., 1996). Travel costs are likely to limit terrestrial carnivores
from increasing their ranges to support more individuals in a group. In
contrast, low travel costs for cetaceans permit movement to disparate but
new superabundant food patches, and delphinid grouping can occur on
scales generally inaccessible to terrestrial carnivores.
While there is considerable support that large group sizes are an adaptive
response to living in open habitats with unpredictable food resources (e.g.,
Brashares et al., 2000; Fisher and Owens, 2000; Jarman, 1974), Gerard et al.
(2002) argue that observations of increased group size in open habitats is an
emergent property, resulting simply from the increased visibility of other
groups in open habitat and an attraction between individuals in different
groups. In closed and complex habitats, individuals and groups are less likely
to detect each other and therefore groups remain small (including solitary
individuals). In open habitats, individuals can perceive another individual or
group at greater distances and therefore are more likely to join, resulting in
larger group sizes. While some observations of terrestrial ungulates support
this hypothesis (Gerard et al., 2002), it is unclear if similar trends would
occur in the open ocean, where perception of other groups is likely aural
(acoustic) rather than visual.
5. Intermediate-Ranging Patterns
The observed social strategies of the delphinids described above
(Sections 3 and 4) clearly indicate that even within a species a great deal
of variation occurs. While it is possible to fit some species into the broad
categories of resident and wide-ranging patterns, other species clearly do not
254
Shannon Gowans et al.
Figure 3.12 The influence of ecology, in particular the predictability of resources on
the evolution of social structure in delphinids, such as ranging behaviour, group size
and formation of associates and long-term bonds.
fit these patterns. We perceive that resource availability exists in a graded
continuum from predictable to extremely variable and less predictable
(Fig. 3.7). Therefore, we expect that delphinid ranging patterns, which
we believe are determined in large part by resource availability, are also
likely to vary along a graded continuum (Fig. 3.12). At intermediateranging patterns, individuals form mid-sized groups, which reduces intragroup competition for resources while still providing protection against
predation. It is within intermediate-ranging patterns of delphinid groups
that we may find the strongest support and challenges to our proposed
hypothesis, as examinations of exceptions to the rule often result in different
interpretations of a dataset (e.g., Henzi and Barrett, 2003).
Killer whales (Section 5.2) may be one of the most important exceptions
in this respect, as it has been suggested that they face limited or no predation
pressure, especially as adults. Additionally, many groups are strongly kin
based, and we expect that cooperation among kin may play a much larger
role in group dynamics rather than the repression of competition.
5.1. Humpback dolphin (Sousa sp.)
Humpback dolphins (genus Sousa) inhabit coastal waters of Indian and
western Pacific oceans and tropical West Africa. They are medium-sized
(up to 280 cm in length) and robust, with a maximum weight of
250–280 kg (Ross et al., 1994). The common name originates from their
most characteristic feature—a dorsal ridge (or hump) of connective tissue in
the middle of the upper back, with a small dorsal fin on the dorsal hump
(Fig. 3.13). In the Indo-Pacific region, the shape and size of dorsal fin and
hump, and overall body colouration vary, with large broad-based hump
The Social Structure and Strategies of Delphinids
255
Figure 3.13 Humpback dolphin mother and one year-old calf, showing conspicuous
hump of the East African form Sousa plumbea. (Photo courtesy of L. Karczmarski.)
and grey-coloured adults in western Indian Ocean, and gently sloping ridge
(no hump) and white-pinkish adults in Southeast Asia and western Pacific.
In southern African waters, humpback dolphins are sexually dimorphic in
length, with males larger than females (Ross et al., 1994), although in Hong
Kong and the South China Sea, sexual dimorphism is far less evident
( Jefferson, 2000).
The taxonomic status of the genus remains unresolved. Between one
and five nominal species have been proposed, but the most commonly
accepted taxonomy recognizes only two species, the Indo-Pacific humpback dolphin (S. chinensis) and the Atlantic humpback dolphin (Sousa teuszi;
e.g., Jefferson and Karczmarski, 2001). Taxonomic separation into three
species, with humpback dolphins occurring in the western Indian Ocean
referred to as Sousa plumbea (Indian Ocean humpback dolphin) is also used
(e.g., Rice, 1998; Ross et al., 1994). S. teuszi remains one of the leastknown delphinids (e.g., Van Waerebeek et al., 2004), and most of the
current knowledge of humpback dolphin social systems comes from studies
of the Indian Ocean Sousa in South Africa (Karczmarski, 1996) and west
Pacific Sousa in Hong Kong ( Jefferson, 2000).
Indo-Pacific humpback dolphins occur in shallow coastal waters, generally <20-m deep and within 1 km of shore, often near large estuaries (Ross
et al., 1994). In southern Africa, they are usually just outside breaking waves,
<500 m from shore and 15-m deep (Karczmarski et al., 2000a), in protected
parts of bays or estuaries, sometimes following tidal channels into coastal
lagoons (Guissamulo and Cockcroft, 2004; Karczmarski, 2000; Keith et al.,
2002). They display no apparent preference for clear or turbid waters, and
have been seen in a variety of coastal habitats including sandy coasts,
enclosed bays and coastal lagoons, mangrove channels, seagrass meadows,
rocky and coral reefs and turbid estuarine waters (reviewed in Jefferson and
Karczmarski, 2001; Ross et al., 1994). Although in some areas they venture
further offshore (up to 55 km from shore, Corkeron et al., 1997; Durham,
256
Shannon Gowans et al.
1994; Jefferson, 2000), this happens only if the water remains shallow, and
in general water depth seems to be the main physical factor limiting their
offshore distribution (Karczmarski et al., 2000a).
Off the Eastern Cape, South Africa, humpback dolphins are mostly seen
in the morning and, to a lesser extent, in the evening (Karczmarski et al.,
2000b) when they forage around shallow rocky reefs (Karczmarski and
Cockcroft, 1999; Karczmarski et al., 2000a). These rocky reefs represent
critical feeding areas for humpback dolphins within their restricted nearshore distribution (Karczmarski et al., 2000a). In Algoa Bay, South Africa,
dolphin activities follow a well-defined daylight pattern that varies little
with seasons and tides (Karczmarski and Cockcroft, 1999), although elsewhere tidal influences on behaviour seem more noticeable (e.g., in Hong
Kong, Parsons, 1998; Maputo Bay, Mozambique, A. T. Guissamulo,
unpublished data).
In the Algoa Bay region, humpback dolphins forage mostly individually
with little or no cooperation (Karczmarski et al., 1997). However, in
Maputo Bay, cooperative foraging is frequent, and often involves a coordinated chasing of schooling fish against the slopes of sandbanks and tidal
channels. Further north along the Mozambique coast, in the Bazaruto
Archipelago, humpback dolphins deliberately beach themselves on sandbanks while in pursuit of small fish, individually and sometimes cooperatively in small groups (Peddemors and Thompson, 1994).
In Hong Kong, humpback dolphins are restricted to the immediate
vicinity of large estuaries with linear movements of only a few tens of
kilometres (Hung and Jefferson, 2004). In the Pearl River Estuary, individual home range sizes vary between approximately 25 and 300 km2, with an
average of 100 km2, which covers only part of the overall range
(>1800 km2) of a population of more than 1000 animals (Hung and
Jefferson, 2004; Jefferson, 2000).
Off the South African coast, although they do not undergo large-scale
migrations, movements of several tens of kilometres are common, and
movements of some hundreds of kilometres are likely (Karczmarski,
1996). In the Algoa Bay region, approximately 10% of the dolphins seen
in the Bay are relatively resident, but most others range widely within a
narrow band along the coast (Karczmarski, 1999). Seasonal variation in
occurrence, abundance and group size is considerable (Karczmarski et al.,
1999a) and results from seasonal immigration of humpback dolphins into,
and emigration from, the Algoa Bay region in summer (Karczmarski et al.,
1999b). Similar summer influx of transient individuals was also seen in
Maputo Bay, Mozambique (Guissamulo and Cockcroft, 2004). All studies
conducted to date along the southern African coast indicate the same
prevailing pattern in which some individuals are resident (possibly longterm resident), and many others display long-distance movements with
ranges that might cover hundreds of kilometres of coastline (Atkins et al.,
The Social Structure and Strategies of Delphinids
257
2004; Guissamulo and Cockcroft, 2004; Karczmarski, 1999; Karczmarski
et al., 1999a,b).
Population estimates vary between 450 and 480 dolphins in the Algoa Bay
region (Karczmarski et al., 1999b), between about 240 and 260 at Richards
Bay, KwaZulu-Natal coast of South Africa (Atkins et al., 2004), and approximately 105 dolphins in Maputo Bay (Guissamulo and Cockcroft, 2004). The
overall percentage of dolphins that appear to be resident is comparable
between locations, although it is noticeably higher in relatively sheltered
Maputo Bay than it is in the predominantly exposed Algoa Bay.
Grouping patterns of humpback dolphins seem relatively simple, with
the dolphins either solitary or in small groups. In the western Indian Ocean,
mean group sizes range between 6–7 dolphins in the exposed bays of the
Eastern Cape coast, South Africa (Karczmarski, 1999; Karczmarski et al.,
1999a; Saayman and Tayler, 1979), and 15 dolphins in Maputo Bay
(Guissamulo and Cockcroft, 2004). In Hong Kong, an average group
contains 3–4 individuals ( Jefferson, 2000; Parsons, 1998). The smallest
groups are known from Moreton Bay, Australia (mean group size 2.4;
Corkeron, 1990) and the Goa coast of India (mean group size 2.6;
Parsons, 1998). Unusually large groups of 50–100 individuals have been
seen in the Persian region, but these sightings were very unusual and it
appeared that the large groupings represented aggregations of several smaller
groups; generally the groups do not exceed 20 dolphins, with a mean size of 11
(Baldwin et al., 2004). Nursery groups are larger and generally more
cohesive than non-calf groups, at least in South African waters
(Karczmarski, 1999; Saayman and Tayler, 1979), which may reflect an
antipredator strategy to protect vulnerable calves. Most calves are born in
summer, and gestation lasts approximately 10–12 months ( Jefferson and
Karczmarski, 2001). Little to no seasonal variation in group sizes occurs in
Hong Kong waters ( Jefferson, 2000), while in the dynamic coastal habitats
of the Eastern Cape, South Africa, there are well-pronounced peaks in
group sizes during summer, which corresponds with peak calving, and is
probably attributed to increased inshore food abundance (Karczmarski,
1999; Karczmarski et al., 1999a).
Social dynamics of humpback dolphins are fluid, with only casual and
short duration affiliations. Strong bonds between individuals other than
mothers and calves are uncommon, and fluid group membership represents
the general pattern in both Hong Kong and South African waters ( Jefferson,
2000; Karczmarski, 1999; Keith et al., 2002). However, preliminary reports
from Maputo Bay, Mozambique, indicate a relatively high number of
stronger affiliations (Guissamulo and Cockcroft, 2004), suggesting a possibly
more stable society of humpback dolphins in more protected inshore
habitats of the Bay. Unfortunately, in all studies to date, the sex of only a
small number of individuals was identified, which hinders more in-depth
analyses of group dynamics and social structure.
258
Shannon Gowans et al.
An apparent temporal segregation between sexes (and possibly age
classes) was seen among humpback dolphins in Algoa Bay, South Africa,
and mate-searching behaviour that may involve long-distance travel was
suggested to be the most likely reproductive strategy of males along the
exposed Eastern Cape coastline (Karczmarski, 1999). Little evidence of
alliance formation was seen in Algoa Bay, and it appeared that most fully
grown males travelled either solitarily or in groups that contained both
sexes. The social relationships among males might be more complex,
however, and differ with habitat. Coordinated herding of a female by a
group of three males and subsequent mating, similar to the pattern seen
among Indo-Pacific bottlenose dolphins in Shark Bay (Connor et al., 1992),
was seen in Maputo Bay, although once only (A. T. Guissamulo and
L. Karczmarski, unpublished data). Even though generally this behaviour
seems rare, it indicates that there is likely much more diversity to humpback
dolphin social strategies than currently described.
The travel range of females and their grouping pattern differ with
reproductive stage, as they concentrate their activities over more restricted
areas (nurseries) during lactation (Karczmarski, 1999). During that time,
allomaternal grouping and care-giving behaviours are frequent (Karczmarski
et al., 1997), as also reported from Hong Kong (Parsons, 1998). The stretch
of coast used as a nursery encompasses several feeding grounds, and its size
seems to depend on resource availability (size and number of feeding
grounds) and the ability of the calf to travel. As older calves better cope with
difficulties of long-range movements, the stretch of coastline used as a nursery
increases, and gradually the more resident female can switch to a wide-ranging
pattern, until her next motherhood (Karczmarski, 1999). It is likely that male
calves follow their mother’s ranging and grouping pattern through adolescence, but increase their travel range at the onset of adulthood. In the generally
fluid society, their interactions with other males are probably limited by
competition for females, unless the distribution and availability of females is
more predictable, as it seemingly is in relatively protected Maputo Bay.
Otherwise, with progressing age, males may lead a more solitary and wideranging life, as they apparently do along the Eastern Cape coast, perhaps
somewhat similar to the life pattern of male elephants (Loxodonta africana;
Poole, 1994).
Although the current knowledge of humpback dolphin social structure
is far from conclusive, and relatively little is known about their foraging
behaviour and predation risk, it is apparent that for this shallow-water
coastal species, their group dynamics are driven primarily by resource
availability, parental needs of nursing females, and mating opportunities of
males. The limited inshore resource abundance and intra-group scrambletype competition limits the group size, and the patchy distribution of
foraging grounds determines the ranging pattern of the group. The demands
of motherhood limit the distribution, ranging behaviour and grouping
The Social Structure and Strategies of Delphinids
259
pattern of females, which in turn affect social and sexual strategies of males.
The pattern of social relationships is fluid and has elements of fission–fusion,
although considerably different than the classic model described earlier for
large-island-associated spinner dolphins. These observations of humpback
dolphin sociality fit with our predictions, with small groups of dolphins
forming medium-sized communities and individuals within these communities ranging somewhat widely for available resources. There appear to be
few strong bonds between individuals; however, our ability to study these
dolphins is restricted by their ranging behaviour.
5.2. Killer whales (Orcinus orca)
The killer whale is the largest of the delphinids, with geographic variability
in size and subtle aspects of colour shadings (see Fig. 3.1). It has been
suggested that this disruptive colouration is useful in disorienting the
killer whales’ prey (i.e., a confusion effect), comprising everything from
large schools of small fishes to large fishes such as salmon (Oncorhynchus sp.),
to intermediately sized marine mammals such as pinnipeds and small
odontocetes, and to the large whales such as blue (Balaenoptera musculus)
and grey whales (Eschrichtius robustus). Killer whales (originally named
‘whale killers’ by whalers) are probably one of only a few cetaceans that
have little or no predation risk, and instead have been able to refine their
social systems relative to efficient hunting and the social–sexual system
(Baird, 2000).
Killer whales occur in all major oceans, including Arctic and Antarctic
seas, with somewhat less occurrence in much of the tropics, especially on
the high seas. They usually occur in groupings of fewer than 50 individuals,
although in the North Atlantic and the Antarctic, groups of hundreds appear
to be common. It is not known whether such large groupings are composed
of permanent schools or whether they represent temporary aggregations of
several smaller units.
Most behavioural ecology (and other) research has been carried out in
the US–Canadian Pacific Northwest near Vancouver Island, British
Columbia (summarized by Baird, 2000; Ford et al., 1998a), with other
less-detailed studies off Norway (Similä, 1997), Argentina (Hoelzel, 1991)
and the Crozet Islands of the southern Indian Ocean (Guinet and Bouvier,
1995). All of these studies have been of killer whales that occur within
several kilometres of shore, and only recently have studies commenced on
pelagic North Pacific and Bering Sea societies (Burdin et al., 2001).
In the US–Canadian Pacific Northwest there are several different communities, consisting of two major types of killer whales: resident and
transient. Actually, both are resident to certain areas of the nearshore, but
transients tend to move faster, in straighter lines, and over greater overall
day-to-day distances than residents. Residents feed largely on salmon and
260
Shannon Gowans et al.
transients feed largely on marine mammals including pinnipeds, sea otters
(Enhydra lutris) and cetaceans (Baird, 2000). There are three different communities of resident killer whales. The southern community consists of
approximately 80 killer whales which frequent the Straits of Juan de Fuca
and waters surrounding southern Vancouver Island and northern Washington (Ford et al., 2000). The northern community consists of approximately
210 whales ranging from the northern parts of Vancouver Island to
the Queen Charlotte Islands and southeastern Alaska (Ford et al., 2000).
A third community of residents, the southern Alaskan resident community,
consists of at least 350 individuals which range from the Kenai Peninsula
to the southern Alaska Panhandle (Matkin et al., 1999a). Each of these
communities is split into several different pods that frequently interact;
however, rarely do individuals interact with members of different communities (Barrett-Lennard, 2000). There are also three different communities
of transient killer whales; the AT1 community of 11 individuals inhabits
the Prince William Sound area of Alaska (Matkin et al., 1999b), the west
coast transients comprise 220 individuals which range from the Alaskan
Panhandle to northern California, and the Gulf of Alaska community has at
least 60 members which ranges along the Pacific Alaskan coast (Ford
and Ellis, 1999). Pods within each community interact, but no associations
between community members have been observed (Barrett-Lennard,
2000). A third group of killer whales has been identified offshore in the
Pacific Northwest, but this group has been poorly studied and little
is known about its range, diet or social structure (Ford et al., 2000).
Community size and ranging patterns have not been investigated in detail
in other areas.
Both resident and transient killer whales have the basic social unit of a
matriline, consisting of one older female, her sons and daughters and the
offspring of her daughters. Since killer whales live for many years, females
up to 80–90 years, a matriline can consist of about four generations of
maternally related whales. This is so especially for resident matrilines, which
tend to be socially closed, that is, female and male offspring stay with mother
throughout life. In the transient matriline, on the other hand, some offspring of both sexes may leave for long periods or life, to other matrilines or
live alone, making transient matrilines smaller than those of residents, with
more frequently changing association patterns (Baird, 2000). Nevertheless,
transient matrilines associate with each other throughout a community, and
are linked by a network of closely and more distantly related individuals that
at times interact. Transient killer whales also share similar vocal calls within
the community, indicating that they are learning from each other and that
vocal recognition may be of importance in patterns of affiliation (Baird,
2000). The more open nature of transient society probably relates at least in
part to their specializing on marine mammal prey, with fewer whales
needed to take smaller animals such as sea otters or seals, and larger
The Social Structure and Strategies of Delphinids
261
groupings of one dozen or more required for taking a blue whale (Tarpy,
1979) or sperm whale (Pitman et al., 2001), for example. Baird and Dill
(1996) showed that groups of three killer whales had the highest energy
intake per whale when feeding primarily on harbour seals (Phoca vitulina).
In resident salmon-feeding killer whales of the Pacific Northwest, the
next higher organization level from the closed matriline is the pod, made up
of several such matrilines with recent maternal ancestry. Matrilines of one
pod often but not always travel together. The next level is the clan,
consisting of pods that share similarity of acoustic dialect. It is assumed
that clans have such similarity due to ancestral splitting into related pods
(Ford, 1991). The next and highest definable social level ( just below that of
population) is that of the entire community, consisting of pods that regularly
associate with each other but never with those of another community
(although, interestingly, communities may overlap; Baird, 2000).
These killer whales that have been studied to date show clear and
generally long-term matriarchal societies. In resident Pacific Northwest
animals this reaches an extreme with offspring staying with their mothers
for life. Although actual mating has not been observed, genetic information
shows that mating occurs outside their matrilines (Barrett-Lennard, 2000),
probably during brief aggregations of several pods of a community. Therefore, sibling relatedness is unlikely to be more than about one-half, yet
such close-knit bonds are very likely to lead to both kin and reciprocal
altruism behaviour (Trivers, 1971). It is believed that such altruism, including prey sharing, is exhibited wherever killer whales have been studied
(Baird, 2000).
Cooperative hunting occurs in resident and transient killer whales in the
Pacific Northwest (Baird, 2000), with transients generally vocally quiet
while foraging on acoustically sensitive marine mammals, and fish hunting
residents usually highly loquacious (Ford et al., 1998b). In fjords of Norway
(Similä, 1997), killer whales coordinate swimming in tight circles, tailslapping and bubble blasting in order to tighten herring schools, and bring
them towards the surface. The basic herding techniques are quite similar to
those of dusky dolphins herding anchovy schools in near-shore Argentina
(Würsig and Würsig, 1980; Section 4.3). In New Zealand, killer whales at
times forage on sting rays hidden beneath the sand (perhaps quite similar to
feeding by Bahamian bottlenose dolphins on flatfish beneath the sand;
Rossbach and Herzing, 1997), and this single-prey feeding appears to be
carried out individually without apparent cooperation (Visser, 1999). In
Argentina and the Crozet Islands (Guinet and Bouvier, 1995; Hoelzel,
1991, respectively), killer whales display complex strategies of taking southern elephant seals (Mirounga leonina) and South American sea lions (Otaria
flavescens) near and on shore, the latter by the predators briefly beaching
themselves. Often, several whales cut off potential escape routes while one
whale charges the prey. Beaching appears to be a highly skilled behaviour
262
Shannon Gowans et al.
that requires practice, and there is some evidence that adults teach young the
skills that are needed (Guinet and Bouvier, 1995; Lopez and Lopez, 1985).
It is clear that killer whales have been able to develop sophisticated
foraging behaviours presumably, in part, because they are relatively free
from predation. This sophistication is likely enhanced by long-term
social affiliations of both kin and non-kin. It is not yet clear how individuals
respond within, or order themselves by sexual–activity relationships,
although the highly sexually dimorphic size differences of males compared
with females (Fig. 3.14) suggest that some males may obtain more matings
than others do, and as such, there may be a strong tendency towards
polygyny (sensu Kenagy and Trombulak, 1986; Lopez and Lopez, 1985).
It is not known to what degree males order relationships among themselves,
or indeed how much female choice is involved. It is perhaps reasonable to
conjecture that in matriarchal societies of killer whales, where females and
extended matrilines take care of young in a network of long-term associations, female choice of mating partners is likely to be of great importance.
Overall, killer whales particularly demonstrate food-related social organization and behaviour clearly, and which is not confounded by predationrisk effects. Killer whales that feed on salmon in the inshore waters of the
Pacific Northwest occur in matriarchal societies that change little in group
size (although matrilines coalesce and super-pods associate, the latter
most probably for genetic exchange). Food is predictable and relatively
abundant, and the whales are apparently resident to a remarkably small
area even though they are apex predators. Yet matrilines are not so large
as to have animals overtly competing with each other when hunting salmon
schools.
Figure 3.14 Killer whale O. orca young adult male on the left and an immature male
(or possibly mature female) on the right, off Friday Harbor,Washington State. (Photo
courtesy of M.Wˇrsig, with permission.)
The Social Structure and Strategies of Delphinids
263
On the other hand, killer whales that feed on marine mammals have a
much more labile group structure, occurring from single animals to associations of 12 or more. As was mentioned previously, a group size of three
appears to be most efficient for taking medium-sized seals (Baird and Dill,
1995), while it is likely that many more animals are needed to debilitate a
large whale. It is also important for killer whales to keep the large whale
alive while they are feeding on it, at least in deep waters, as a dead whale
quickly sinks (Guinet et al., 2000). When feeding on marine mammals the
element of surprise is highly important, therefore mammal-eating killer
whales may need to travel longer distances to encounter vulnerable prey.
The exception is killer whales that feed on young pinnipeds that wander
from shore rookeries; the inexperienced young prey are generally unaware
of how to avoid killer whales, and the predators can station themselves at a
rookery for several days to weeks at a time (Lopez and Lopez, 1985). At any
rate, mammal feeders have more labile group sizes and affiliations as dictated
by the need for efficiency of feeding.
The absence of predation has probably freed killer whales from the
increased costs of grouping in inshore waters, unlike the costs believed to
be incurred by, for example, inshore bottlenose dolphins (see Wells et al.,
1999a, and Section 3.1 for details). So, killer whales may be more likely to
form large groups in inshore waters. It is likely that life-long association with
offspring allows for highly sophisticated learned behaviours of food occurrence patterns and acquisition techniques, further encouraging the formation of larger groups. Additionally, cooperative foraging techniques
probably limit competition among group members by enhancing the food
resources available to the group. Beyond this, however, it may also be useful
for close matrilineal kinship bonds to have animals help each other in
aggregations of matrilines and ‘superpods’, during which associations it is
assumed most male displays and mating take place. Killer whale males are
much larger than females and show a striking dorsal fin secondary sexual
characteristic. It is possible that males display to each other and/or females in
order to gain access to females in oestrus, although this has yet to be
confirmed directly. Matrilineal pods may be in part a bonding technique
to facilitate female choice (and rejection) and confer protection from forced
copulations by males.
Although cooperation can evolve among unrelated group members
(Frank, 2003), the added reward of indirect reproductive success likely
enhances cooperation when members of the group are related. Given
the strong natal philopatry of killer whales, we would expect that cooperation with kin is a strong driving force in the evolution of killer whale
social strategies, and especially in the evolution of highly cooperative
foraging.
Genetics and life history information from long-finned and short-finned
pilot whales (Globicephala sp.: Amos, 1993; Kasuya and Tai, 1993, respectively),
264
Shannon Gowans et al.
and behavioural information on short-finned pilot whales (HeimlichBoran, 1993), indicate that group structure and matriarchy may be quite
similar to those of resident and transient killer whales in the Pacific Northwest. Pilot whale pods may be matrilineal, including apparent life-long
fidelity of male offspring to the maternal pod. Pilot whales are also relatively
free of predation and are sexually dimorphic. However, unlike most killer
whale societies, pilot whales can occur in schools of hundreds of animals, as
they move in the open ocean, feeding in large part on squid for which they
need to dive to great depths. The tight matriarchal bonds in this case may
strongly be linked to diving and needing to leave calves and youngsters at
the surface. Thus, large group formation means individuals benefit from kin
and close associates through ‘babysitting’ leaving mothers free to feed at
depth. This system may therefore be more attuned to sperm whale societies
than killer whales, per se (Whitehead, 2003). Evidence from photoidentification and behavioural observations of long-finned pilot whales
from Nova Scotia, Canada, supports the sperm whale analogy, with at
least some pilot whales forming long-term associations in relatively small
groups (12 individuals), while interacting briefly with many other associates (20 individuals). However, these long-finned pilot whales do not
appear to be long-term residents of a small area, but instead may be ranging
fairly widely, although they return to the same general area off Nova Scotia
at approximately the same time of year (Ottensmeyer and Whitehead,
2003), probably tracking food resources.
In summary, there are differences in killer whale social structure with
resident (fish-eating) killer whales forming relatively large, closely related
groups, foraging predominantly on individual or schooling fish. While
these killer whales range much more widely than inshore bottlenose
dolphins, they range less widely than some other dolphins and less than
transient (marine mammal eating) killer whales that form smaller groups in
order to cooperatively hunt their prey. In transients, group size is limited
by competition for food within the group. Killer whales are ‘freed’ from
serious predation and therefore resident killer whales can form larger
groups than other ‘resident’ dolphins (see Section 3); however, competition for food limits group size in transients when they are foraging on
relatively small prey items such as harbour seals. In contrast, transients form
larger groups, when greater cooperation is needed to forage on large and
difficult to catch prey items such as mysticete whales. Observations of killer
whale sociality test the predicted framework, as predation does not exert a
strong evolutionary pressure. However, it appears that even in killer
whales, the availability of resources and competition for those resources
are important in driving their sociality. Further revisions of this model may
be required in order to fully apply it to communities with strong matrilineal bonds as have been observed in killer whales and suggested in pilot
whales.
The Social Structure and Strategies of Delphinids
265
5.3. Comparisons with terrestrial mammals
It is inherently difficult to make generalizations about the intermediate
classification pattern, as it falls somewhere on the continuum between
local and wide-ranging communities. However, some generalizations can
be made. It is probable that many delphinid species fit into this intermediate
pattern, although few have been investigated in sufficient detail due to the
difficulties in studying animals that live in small- to mid-sized groups and
range over fairly wide areas of sea. While humpback dolphins have been
studied in a number of areas (see Section 5.1), these studies represent relatively short-term snapshots of a specific area and understandably therefore do
not give a very complete picture of their lives. We speculate that humpback
dolphins live in a world of moderately predictable resources, with some site
fidelity, but mostly over a much larger area than the inshore resident dolphins. They display fluid associations with other individuals, and the need to
keep moving to find food probably limits long-term contact with other
individuals and strong bonding. Predation pressure is not sufficiently high
to cause large groups to form, and in fact individuals probably reduce
predation pressure more by refuging in complex habitats or surf zones
when they are available. Food resources are not always available, nor are
they entirely clustered in relatively rare but superabundant patches. Therefore, solitary individuals or smaller groups are often most effective for locating
food, while at the same time reducing competition. Additionally, if different
individuals have different goals (e.g., immediate need for food, protection of
young or mating), then this will increase the fluidity of associations as groups
fission commonly and regroup with different membership. This pattern is
likely to prevent the formation of long-term complex bonds.
The resource dispersion hypothesis (RDH) suggests that when resources
are patchy but reasonably abundant, the costs of group living may be
reduced and groups form even if there are no direct benefits to the individuals ( Johnson et al., 2002). Typically, when resources become more
abundant, individuals should reduce their territory (or home range) to
reduce travel and territorial maintenance costs. However, if resources are
not evenly distributed (in space or time), then a reduced home range may
not contain all of the required resources. Instead, it would be beneficial for
an individual to maintain a larger home range that encompasses sufficient
resources, even if at times this range could support more individuals. Thus
in habitats with patchy but rich resources, competition between individuals
may be reduced and individual ranges may support additional animals in the
range at relatively little cost to the primary animals (Johnson et al., 2002).
This model may be relevant to describe the evolution of sociality in several
carnivore species, especially those that do not cooperate extensively within
these groups. There have been numerous difficulties in testing this model,
many of which relate to accurate measures of territory size as well as
resource dispersion and availability ( Johnson et al., 2001).
266
Shannon Gowans et al.
Humpback dolphins and other ‘intermediate’ ranging dolphins are clustered in groups that do not appear to be generally cooperative and they
exploit resources that are heterogeneously distributed. The RDH may well
describe the sociality of these dolphins; however, it remains to be tested.
The evolution of cooperative and altruistic behaviours in killer whales
has led to a very different pattern of sociality than observed in humpback
dolphins. Among cooperative terrestrial animals, individuals in larger
groups appear to have higher reproductive success as larger groups are better
able to cooperate (e.g., meerkats, Suricata suricatta; Clutton-Brock et al.,
1999). Thus there is selection for larger groups in highly cooperative
animals. Pilot and killer whale group sizes mostly support this suggestion,
as they tend to be found in larger groups than more resident or other
intermediate-ranging dolphins (compare Sections 3 and 5.1 with 5.2).
However, scramble-type competition for food resources may limit group
size (see Baird and Dill, 1996). Group size may be maintained by the eviction
of subordinate individuals (as observed for meerkats; Stephens et al., 2005) and
suggested in transient killer whales (Baird and Whitehead, 2000). Further
investigation of the similarities and differences between cooperative terrestrial
mammals and killer whale sociality may yield important insights.
While there are clear differences between killer whale and elephant
societies, there are many similarities. As in killer whales, adult elephants
face little predation pressure and form a multi-tiered social structure consisting of matrilines that will associate with other related matrilines (Wittemyer
et al., 2005). While elephants feed on a wide variety of plants, their large body
size and low digestive efficiency means that they require very large quantities
of food to sustain themselves. Therefore, they range fairly widely to find food,
and resource competition limits group size. Family units consist of related
females and their dependent young. These units remain together throughout
the year (dry and wet season), indicating that these groups may be the largest
that can be supported by the environment without leading to resource
competition. New family units appear to form after the death of the matriarch, when the matriarch’s surviving daughters split into separate family units.
These new family units will frequently merge (especially during the wet
season when food resources are more abundant); however, such associations
are relatively brief (Wittemyer et al., 2005). A similar pattern is seen in killer
whales with family-based units, which frequently associate with close kin and
less frequently associate with more distantly related groups (Baird, 2000).
Among delphinids, the killer whale pattern of sociality may relatively be
rare and driven in large part by extensive cooperative behaviour between
kin. Thus far it appears restricted to the subfamily Globicephalinae, which
includes the pilot and killer whales as well as false and pygmy killer whales
(Feresa attenuata) as well as the melon-headed whale (Peponocephala electra).
Unfortunately, relatively little is known about the social structure of the
melon-headed whale as well as false and pygmy killer whales.
The Social Structure and Strategies of Delphinids
267
6. Demographic, Social and Cultural Influences
Societies tend to some stability in space and time. A one- or even fiveyear description of a group of bottlenose dolphins in a bay system, for
example, shows births, deaths, variations in affiliations, but no large-scale
change in habitat use or social structure. It is not until societies are studied
for longer that we observe either slow changes or the occasional rapid
wholescale invasion of new habitat (e.g., bottlenose dolphins abruptly
moving north off the California coast; Wells et al., 1990), disappearance
from an area (e.g., pilot whales disappearing from the Catalina Island area;
Shane, 1995), or large variation in community size and structure (Wells
et al., 1999a). It has generally been assumed for dolphins that such changes
are due to extrinsic environmental factors, but studies of chimpanzee
societies show that much community structure and habitat use can be
ascribed to behavioural interactions, or ‘internal politics’ of individual
associations in the community (Wrangham and Peterson, 1996).
In the light of chimpanzee and similar studies on land, it appears
worthwhile for us to re-examine the causes of subtle long-term and abrupt
changes in habitat use and social affiliations in delphinids. For example, pilot
whales were known to feed off the southern California Channel Islands in
the late 1970s and early 1980s, but in 1982 abruptly disappeared, with
Risso’s dolphins frequenting the pilot whales’ former areas (Shane, 1995).
The most likely explanation for the distributional change is the effects of
ecological factors, such as the apparent concomitant decline of nearshore
market squid (Loligo sp.) upon which pilot whales feed. However, we
simply do not know whether intrinsic (to the community) social factors
might have been at least partially responsible, such as one community
converging with the Channel Island one, and other alternative travel and
habits being adopted by the latter. Competitive exclusion between Risso’s
dolphins and pilot whales could be another factor. There is evidence for
rapid community-wide adoption of a new behavioural pattern (a particular
male song display) in humpback whales (Megaptera novaengliae; Noad et al.,
2000), but this has not been clearly demonstrated in delphinids. One of us
(Karczmarski et al., 2005) recently showed that two communities of Hawaiian spinner dolphins converged at Midway Island, with an abrupt habitat
shift by one that had apparently been resident in an atoll elsewhere. The shift
was possibly caused by ecological factors, but also possibly by group splitting
due to increasing group size in one community, competitive exclusion by
others elsewhere, or any number of social interaction factors. While the
habitat shift was abrupt, there was a well-documented slower, greater than
one year, gradual acceptance by the resident community of the newcomers.
We can presently only speculate as to the resultant social changes of affiliation, mating and calf rearing. It is to be hoped that present and future studies
268
Shannon Gowans et al.
of such special cases like this will help describe the complexity of ecological
and social interactions responsible for demographic and societal change.
It is probable that much of the observed stability in space and time is due
to animals occupying habitat that fulfils their requirements, and the likelihood that adjacent habitats are already occupied by conspecifics or others
using resources there. Continued use of the same habitat likely results in
individuals adapting to it and an increased chance of adopting behaviour
patterns best suited to the habitat. Such behavioural specificity, as observed
in sympatric communities of killer whales that feed either on fish or marine
mammals (summarized by Baird, 2000), is probably perpetuated by transmission of information within a generation (‘horizontal’ transfer, among
similar-age peers) and from generation to generation (‘vertical’ transfer from
mothers and ‘oblique’ from older non-relatives) by, in a broad sense, culture
(Rendell and Whitehead, 2001). Once different cultures are established, it
becomes unlikely that animals of one culture adopt that of another one,
even though some gene flow may be occurring between them. It is also
possible that cultural differences allow sympatric communities to coexist
without, or with only muted, competitive exclusion. Lack of competition
would depend on amount of overlap of required spatial resources for
foraging, rest away from predators and inclement weather, for example.
Furthermore, it is possible in delphinids with differing culturally transmitted
ways of living, that ‘gene-cultural hitch-hiking’ (when selection favours
individuals that share a culturally transmitted behaviour simultaneously
selects the genes these individuals possess, without the genes possessing
any inherent benefit) may result in a somewhat different genetic make-up
of communities or populations (as indicated by Laland, 1992, and suggested
for cetaceans by Whitehead, 1998).
7. Comparisons with Other Cetaceans
While delphinids comprise the most diverse group of cetaceans, several other families exist and the social structure of some of these species has
been studied in some detail. We suspect that many of the same forces
influencing delphinid social structure also operated on the evolution of
other odontocete social strategies. Only the porpoises (family Phocoenidae)
and the four different families of river dolphins (family Platanistidae, Iniidae,
Lipotidae and Pontoporiidae) are similar in size or smaller than most of the
dolphins. It is likely that predation pressure strongly influences the social
structuring of these taxa, although there may be relatively few predators for
some of the river dolphin species. In contrast, the beaked whales (family
Ziphiidae), sperm whales (family Physeteridae and Kogiidae) as well as
the beluga (Delphinapterus leucas) and narwhal (Monodon monoceros; family
The Social Structure and Strategies of Delphinids
269
Monodontidae) are larger than most of the Delphinidae. Thus, many of
these species may be less vulnerable to predation from sharks ( Jefferson et al.,
1991; Heithaus, 2001b) especially as adults, although killer whales are
capable of killing even adult sperm whales (Pitman et al., 2001).
7.1. Sperm whales (Physeter macrocephalus)
Sperm whales are the largest of the odontocetes and have one of the most
complex social strategies described so far. They are also one of the most
sexually dimorphic species with mature females rarely longer than 11 m,
while mature males may be longer than 18 m. Sperm whales are found
throughout the world’s oceans, usually in waters deeper than 1000 m.
Females and their young are typically in equatorial and subtropical waters,
rarely ranging higher than 40 North or South. Mature males tend to forage
at high latitudes, often near the ice edge, although they will migrate to
equatorial waters to breed. Sperm whales feed opportunistically on a wide
variety of squid in the deep-scattering layer, routinely diving 400–600 m.
The social structure of sperm whales has been best studied in the Eastern
Tropical Pacific, focusing on sperm whales identified near the Galapagos
and the west coast of South America (see review in Whitehead, 2003).
Female groups appear to range over large areas (on the order of
1000 km), although they remain in these large areas for long periods, on
the order of decades. Longer range movements have been recorded, in
particular many females appear to have left the Galapagos area for example,
and have been re-identified along the west coast of Central and South
America, from the Gulf of California to northern Chile (Whitehead,
2003). On a daily basis, females can move quickly over large areas or remain
in a small area for hours to days, although the average mean displacement
(straight line distance) was 70 km d 1 (Whitehead, 2003).
Adult sperm whales are probably not vulnerable to predation by sharks,
and may be much less vulnerable to killer whale attacks than the smaller
delphinids; however, adult sperm whales have been successfully hunted by
killer whales (Pitman et al., 2001). Calves do appear to be particularly
vulnerable to predation, especially as young calves do not seem to be
capable of diving to the mother’s foraging depths (300–800 m) or for the
time required to successfully forage (dive times average 40 min; Gordon,
1987; Papastavrou et al., 1989). Whitehead (1996) outlines how this
vulnerability of calves may have driven the evolution of sperm whale social
structure.
Calves remain with their mothers until approximately age 6 when males
begin to emigrate from their natal group. As the males mature, they cluster
in progressively smaller groups consisting of other subadult males, and they
forage at increasingly higher latitudes. By the time the males are sexually
mature (approximately age 20) they tend to be solitary and forage in Arctic
270
Shannon Gowans et al.
and Antarctic waters, often near the ice edge. Around age 25, males begin to
migrate to equatorial waters where female groups are located. These large,
sexually and socially mature males tend to avoid each other, and rove
between groups of females, spending relatively short periods (hours to
days) with each group, presumably mating with all receptive females in
the group. Genetic paternity testing indicates that some males father more
than one offspring in a group, likely indicating that at least two females were
receptive during the male’s brief visit. After mating, a mature male will leave
the group, presumably to attempt to locate another female group or return
to high latitude feeding grounds (Whitehead, 2003, and references therein).
Male–male competition for access to female groups appears to be based
on size or vocalizations (which may be an honest signal of male size),
although conflicts do also occur (Whitehead, 2003). The sexually based
distribution of sperm whales has probably evolved through the requirement
of males for increased food resources to achieve and support their larger
body size. Thus, sexual selection for large body size in males may have
restricted the development of large groups composed of mature adults of
both species.
Females lead a more complex, stable social existence. Female offspring
remain with their mothers past sexual maturation, forming long-term bonds
with 8–12 other females, which are often but not always kin. These ‘units’
(following the terminology of Whitehead, 2003) remain together for a long
time; recorded associations persist over 10 years, although units can on
occasion merge, split or individuals transfer between units. Units routinely
join to form a ‘group’, an association that typically lasts 1 to 2 weeks. In fact,
sperm whales are most often encountered in a group composed of 20–24
females and dependent males. Often there is not a mature male present, but
if there is it is typically only one male (Whitehead, 2003).
When there are no young calves present in the group, all group members
tend to surface and dive synchronously, such that whales are present at the
surface for approximately 8–10 min, followed by about 40 min with no
whales at the surface. However when a young calf is present, the group
members tend to dive asynchronously, such that at least one adult-sized
individual remains at the surface at all times, and the calf appears to associate
with a number of different individuals, over the course of its mother’s
foraging dive. The evolution of babysitting behaviour can be explained
through kin selection (as most of the females have some maternal kin
present in the group) or through reciprocal altruism, where babysitting
services may be returned at some later time, or a combination of the two
factors (Whitehead, 1996).
The presence of an adult-sized babysitter likely provides increased
vigilance to detect a predator, and may deter attacks by sharks. Killer whales
will attack sperm whale groups, and it appears that the adults will cooperate
to defend the young by forming a protective circle around the young.
The Social Structure and Strategies of Delphinids
271
In addition to providing babysitting, it has been suggested that individual
females will nurse not only their own calves but also the calves of other
females (Whitehead, 2003).
Therefore, a female’s need to reduce the risk of predation of her calf may
have selected for the formation of long-term bonds between females,
especially between kin to provide babysitting services. Females may also
be able to limit competition for food resources by specializing on rich,
patchily distributed squid, and by spreading out and forming foraging ranks
to increase detection rates of these rare, but abundant patches. However,
sperm whales’ large body size and subsequent energetic demands may
restrict groups to much smaller sizes than seen in Eastern Tropical Pacific
dolphins (Section 4.1). Competition for food resources may play an important role in the temporary fission and fusion of units into groups, and in the
relatively rare, but important cases of unit membership change.
Sperm whale social structure seems to fit the pattern described for wideranging delphinid communities (Section 4). However, competition may
limit group size, and increase the formation of bisexual groups. Additionally, female cooperation appears to be an important aspect in protecting
offspring from predation and may have lead to the formation of long-term
bonds, and a predominantly matriarchal society. There are additional levels
of grouping and social complexity observed in sperm whales, including
acoustically based clan groups similar to killer whales (Section 5.2), which
are beyond the scope of this chapter on delphinid social structure
(see Whitehead, 2003, and references therein for more detail).
7.2. Northern bottlenose whales (Hyperoodon ampullatus)
Northern bottlenose whales are a relatively large beaked whale (7–9 m in
length); approximately the same size as male killer whales. They are found
only in the North Atlantic, typically in deep water (>1000 m) from Nova
Scotia, Canada (44 N) to Scotland, with a potentially disparate concentration near the Azores. Northern bottlenose whales have a specialized diet,
focusing mainly on squid from the genus Gonatus. They are deep divers,
routinely diving deeper than 800 m. Little is known about predation risk
faced by bottlenose whales. Few individuals bear scars easily identifiable as
unsuccessful shark or killer whale attacks. Their large body size likely means
that bottlenose whale adults are relatively safe from shark predation; however, calves are probably still vulnerable and all ages are likely to be
vulnerable to killer whale predation. Their biology and social structure
has been studied in detail in the ‘Gully’, a submarine canyon located
offshore of Nova Scotia, Canada (Gowans, 2002) and which serves as a
useful example for comparison with the principal features of delphinid social
groups.
272
Shannon Gowans et al.
This submarine canyon and the surrounding area (100 km along the
shelf break) support a resident population of 160 (Whitehead and Wimmer,
2005). Given that the whales are present in relatively large numbers, it is
presumed food is predictably available in the canyon, and given that the
whales concentrate in the Gully, and to a lesser extent in other smaller
canyons in the area, that the canyons represent a locally higher concentration of resources. The social structure of northern bottlenose whales resembles that of inshore bottlenose dolphins (Section 3.1). Northern bottlenose
whales are typically found in small groups (average 3, rarely more than 10
individuals), females appear to have no preferential associates, while mature
males form long-term bonds (Gowans et al., 2001). Nursery groups were
not clearly identified, perhaps because young calves were only occasionally
sighted and there was no clear evidence of babysitting. Little is known about
the diving abilities of young bottlenose whales, although young animals
were not observed at the surface for longer periods than adults (Gowans,
1999). Males were not observed herding or attempting to sequester females,
nor has mating behaviour been observed. Male–male competition was
observed once, with two males (who had in previous years been longterm associates) head-butting each other in several bouts. Shortly after the
head-butting behaviour, one of the participating males was observed in
association with a sexually mature female (Gowans and Rendell, 1999).
This suggests that male–male bonds are probably not the only route to male
reproductive success which is similar to observations of bottlenose dolphins
in Shark Bay, Australia (Krützen et al., 2004).
The social structure of northern bottlenose whales, at least in the Gully,
resembles the pattern described for resident bottlenose dolphin communities (Section 3). Competition for the limited food resources available in
the canyon may limit group sizes and the overall community size. Predation
is either not a strong enough force to lead to the formation of large groups
or the need to reduce food competition prohibits routine formation of large
groups. Within this resident community, males may be able to sequester
females, leading to the formation of long-term bonds between males to
increase access for females. It is not clear, however, if the observations in the
Gully are generally applicable to all northern bottlenose whale communities, or if this community’s reliance on a reliable food source in the
submarine canyon has led to the evolution of a relatively unique social
structure.
7.3. Harbour porpoise (Phocoena phocoena)
Harbour porpoises are smaller than most delphinids (1.5–1.7 m). They are
found along the coastlines of the North Atlantic and North Pacific, as well as
in the Black Sea. Their small size means that they are vulnerable to
predation by a number of species of sharks as well as killer whales. They
The Social Structure and Strategies of Delphinids
273
typically feed individually on benthic fish. The social structure of harbour
porpoises has not been studied in detail, largely because they are mostly
solitary. While small groups (two to five) individuals are observed, the
species is generally considered asocial (Bjrge and Tolley, 2002). Harbour
porpoises may predominantly follow a refuging from predators’ strategy,
remaining in relatively complex inshore and coastal habitats and avoiding
groups as a way to reduce predator detection (similar to forest-dwelling
oribi; Brashares and Arcese, 2002; Section 3.3). The relative asociality of
harbour porpoises and several other members of the porpoise family may
indicate that predation risk on very small species may be too great to lead to
the formation of groups. However, a paucity of studies means this question
has yet to be investigated fully.
7.4. Why are there no long-term bonds in baleen whales?
Baleen whales (Sub-order Mysticeti) adopted a different style of living from
that of predatory odontocetes due to the formers’ capability of batch feeding
quite low on the trophic level. This resulted in a concomitant increase in
mouth and body size, large energy reserves, and therefore seasonal feeding
together with the capacity for migrations away from feeding areas for safe
mating/calving. Migratory mysticetes do not seem to have the long-term
and complicated social bonds that appear normal among delphinids. Instead,
they loosely affiliate on the feeding grounds with only short-term bonds,
such as tandem-feeding in blue whales (Schoenherr, 1988), bubble-netting
in humpback whales (summarized by Clapham, 2000), and echelon-feeding
bowhead whales (Balaena mysticetus; Würsig et al., 1985). In general, they do
not need to coordinate foraging strategies to take large swarms of mysids or
stands of ampeliscids (grey whales), euphausiid crustaceans (blue and fin
whales; Balaenoptera physalus), and fishes and squid (the majority of balaenopterids). Their size also makes them relatively free of predation as adults,
with the notable exception of occasional encounters with large groups of
killer whales (Tarpy, 1979). Therefore, there is no need for the coordinated
(perhaps, cooperative) social matrix of familiarity for feeding, migration or
predator avoidance. However, humpback whales in Chatham Strait, Alaska,
routinely forage cooperatively to form bubble nets to encircle schooling
herring. Instead of only forming brief associtions to cooperatively build
bubble nets (as observed in most other areas; Clapham, 2000), 22 of these
humpback whales formed a core group which routinely cooperatively foraged together, displaying associations that lasted over subsequent years
(Sharpe, 2001). Individuals within this group tended to take on certain
specific roles (e.g., bubble blower, vocalizer), suggesting that when individuals routinely cooperatively forage, formation of long-term bonds and task
specializion may be important, even among mysticetes.
274
Shannon Gowans et al.
Migrating mysticetes meet on the mating/calving grounds and seasonal
affiliations, displays of dominance among males (apparently in large part
mediated by sound) and some variable gregariousness among females (right
whales Eubalaena glacialis; Payne, 1995) do exist. Although social/sexual
systems have been inadequately described, it is unlikely that tight bonds of
affiliation exist, at least beyond one season. There are no matriarchal or
other long-term associations beyond the mother–calf bond that lasts for one
year or slightly less, depending on species.
Non-migratory mysticetes, such as some populations of minke (Balaenoptera acutorostrata; Dorsey, 1983; Dorsey et al., 1990), Bryde’s (B. brydei)
and pygmy right whales (Caperea marginata), may more closely resemble
some odontocete social groupings, with longer term small group affiliations,
but research to substantiate or refute this possibility has simply not been
carried out. The non-migrating mysticetes are generally the smaller of this
suborder, tend to feed throughout the year, feed in a less individual batch
style than do the larger species, and have much to gain from being in
association with others, such as for food and predator avoidance. This area
is ripe for dedicated research, and we look forward to comparisons with
smaller mysticetes and some delphinid societies. However, we make this
observation in advance of such studies: odontocetes are macropredators
with tight social strategies interwoven with large behavioural flexibility
and large brains. Mysticetes might be thought of as being in line with
the ungulate model of behavioural strategy, with certainly impressive
behavioural capabilities, but little evolution of sophisticated foraging and
therefore social strategies.
8. Conservation Implications
As highlighted in previous sections of this chapter, social systems are
not fixed features of species but often show intraspecific variations that are a
result of individual attempts to maximize fitness under local environmental
circumstances. Collectively, these studies indicate that as environments vary
in the conditions and resources they provide, so do delphinid social systems.
As social systems frequently place constraints on which individuals are able
to breed, and thereby determine the reproductive success of individuals and
effective population sizes, understanding how environmental variation
affects social structure is of major concern in conservation management
(Durant, 1988; Komdeur and Deerenberg, 1997; Parker and Waite, 1997;
Pettifor et al., 2000). Therefore, it is important to consider the incorporation of social structure into predictive models of populations (where
changes in certain parameters suggest changes in management approach;
Pettifor et al., 2000) and population viability analyses (Durant, 2000). This is
The Social Structure and Strategies of Delphinids
275
particularly so when a predictive model is used to explore the effects of
management perturbations, because these may influence the population
indirectly through their effects on social structure, as well as directly
through their effects on survival or fecundity (Pettifor et al., 2000), as was
the case with elephants in Lake Manyara National Park, Tanzania (Prins
et al., 1994).
The effective size of populations, the most sensitive predictor of the
ability of any populations to maintain genetic variability and persist through
time, is greatly affected by the breeding sex ratio, and social and mating
systems of animals (Parker and Waite, 1997). Understanding of animal social
strategies and reliable modelling of the distribution of reproductive success
are key components for understanding population dynamics. For the needs
of predictive conservation modelling, behavioural description of social
systems may be sufficient to characterize the mating system, even if the
underlying genetic structure is difficult to investigate, as is often the case in
studies of delphinids.
In everyday management practice, it is important to know what aspects
of a population determine its vulnerability. However, it is through understanding the behaviour of individuals that we can understand populations,
and predict their potential trajectories in response to demographic changes
or anthropogenic impacts. A number of recent studies, both terrestrial and
marine, have incorporated social structure into population viability analyses
of small mammalian populations (e.g., Martien et al., 1999; Vucetich et al.,
1997; Young and Isbell, 1994). These models consider such factors as social
grouping, territoriality and the effective availability of mates, providing a
framework by which key elements of social structure can be incorporated
into essentially demographic models (Pettifor et al., 2000), and facilitate
better management practices. However, in most cases these models do not
assume individual behavioural strategies to change in response to changing
conditions, which becomes a problem when predicting responses of large
and/or wide-ranging populations. For some land mammals, where the
flexibility of social systems is better understood, sensitivity analyses have
been conducted allowing ‘what if ’ scenarios to be explored. As a result,
management strategies have been developed to manipulate these systems by
varying the environmental determinants of the variation (Komdeur and
Deerenberg, 1997; Pettifor et al., 2000). A better understanding of delphinid social system might have similar application in future management
considerations, especially during the design and management planning of
coastal Marine Protected Areas.
Studies of sperm whales in the Eastern Tropical Pacific highlight the
crucial importance of social strategies to effective conservation. Stock
structures of marine species have been very difficult to determine, especially
in highly mobile species such as sperm whales. When stock structures were
incorporated into management plans, they were often defined more on
276
Shannon Gowans et al.
easy-to-identify geographic boundaries rather than on knowledge of how
the animals were distributed. Sperm whaling in the 1960s and 1970s off the
coast of Chile targeted mature males and ended (at least in part) when
whales could no longer be profitably encountered. This pattern of sexbiased whaling may have led to the elimination of most of the mature males
that could potentially migrate to the Galapagos to breed with females. In
fact, Whitehead et al. (1997) found very few mature male sperm whales in
the Galapagos in the 1980s and 1990s, and also found very few young calves.
While some males were being born into the population, the prolonged
delayed sexual and social maturity in sperm whales and sexual segregation of
males meant that there was a 25–35 year lag between the birth of a male and
his presence back on the mating grounds to potentially father offspring
(see Section 7.1 for more details). Survival of a sufficient number of mature
males through the period of intense whaling in the 1960s and 1970s may
relate to a potential for improved birth rate and recovery from whaling in
the present.
9. Concluding Comments
In conclusion, our predictive framework fits well with most of the
observed patterns of delphinid sociality. Small communities of dolphins
tend to remain resident in small ranges. These dolphins form labile networks
of associates and male bonds sometimes form in order to gain access to
females. Observations suggest that these dolphins tend to forage individually
on relatively solitary prey, and reduce their predation risk by hiding from
predators or avoiding risky areas of their habitat (Section 3). In contrast,
large groups of dolphins tend to form in coastal areas and the open ocean
where grouping may be the only strategy to reduce predation risk. Membership in these large groups may change over time, with most individuals
forming only brief associations. These dolphins range widely and may
cooperatively forage for rare but abundant prey such as large schools of
fish (Section 4).
While these observations lend support for the framework proposed here,
no tests of this framework have been conducted to date. In fact few studies
on delphinid social structure have included testable predictions to support
or disprove a suggested hypothesis. Additionally, this chapter highlights the
many gaps in our understanding of dolphin behaviour, especially our
understanding of foraging behaviour and predation risk. However by establishing a predictive framework that attempts to explain the observed social
strategies, we can now design studies that will test these predictions. Alterations to this model will likely be made in the future (or perhaps a new model
will be needed); however, great advances in studying the evolution of
The Social Structure and Strategies of Delphinids
277
sociality in the terrestrial environment were made once predictive frameworks were established (such as Jarman, 1974; Wrangham, 1980).
It is also clear in this chapter that although studies have begun to
elucidate many of the details of delphinid social strategy, important questions remain. The potential complexities of these systems are vast and
varied. Further research into the social strategies of relatively unstudied
species may provide valuable tests of the concept underlying the framework
outlined in this chapter. Variations in social strategies among different
individuals and different communities will also test this framework. It is
likely that the conceptual framework we propose is an oversimplification of
the multitude of different forces acting on dolphin social strategies. However, we hope that it may serve as a starting point to test some of these
theories, and it may be especially valuable as a way to compare different
communities of the same species as well as between different species.
There is an inherent bias against studying wide-ranging open ocean
species of dolphins, as it is simply more costly and difficult to locate and
follow them. Current advances in remote sensing make it possible to track
the movements of large marine animals using satellite-linked tags which
transmit information about an individual’s location, behaviour and physiology as well as features of its environment (Ropert-Coudert and Wilson,
2005). As these electronic tags become smaller, easier to attach and less
expensive, it may be possible to sample a large number of individual
dolphins from a single pod and ascertain whether they remain together for
long periods. Additionally, we may be able to measure when these individuals are feeding, and on what. If a large number of predators are also
tagged, then we will gain valuable insight into the predation risk faced by
dolphins. Passive acoustic monitoring will also become a more valuable
technique for examining details of dolphin distribution and movements. By
conducting large-scale, multifaceted research projects that incorporate
ocean ecology as well as cetacean sociality, we will gain valuable insight
into the remarkable lives of delphinids.
Cetaceans have clearly diverged from terrestrial mammals in the 55-plus
million years since a group of ungulates adopted a predominantly aquatic
lifestyle (Gingerich, 1998; Thewissen, 1998). However, it is clear that they
maintain many mammalian characteristics that have exerted different evolutionary forces than fish and other marine organisms face. Thus delphinids
share many similarities with other mammals, despite their marine lifestyle.
Above all, we need to view dolphins and other odontocete cetaceans as
large brained social predators (similar to carnivores) that face challenging
predation risks (similar to ungulates) and complex social strategies and
relationships between individuals (similar to primates). By doing so, our
understanding of dolphin social strategies may approach that of some
terrestrial mammals including humans.
278
Shannon Gowans et al.
ACKNOWLEDGEMENTS
Preparation of this chapter has involved countless discussions with a number of different
people, including those who specialize in dolphin sociality as well as terrestrial animal
sociality. This chapter has been improved by the comments of David Sims, Peter Simard
and an anonymous reviewer. The photographs taken in Hawaii were authorized by the
permit conducted under NOAA Fisheries scientific research permit no. 1007-1629-01 and
US Fish and Wildlife Service special use permit no. 12520-01014. Photographs from Tampa
Bay were conducted under NOAA permit number 1077-1794.
REFERENCES
Acevedo-Gutierrez, A. (2002). Interactions between marine predators: Dolphin food intake
is related to number of sharks. Mar. Ecol. Prog. Ser. 240, 267–271.
Acevedo-Gutierrez, A. and Parker, N. (2000). Surface behaviour of bottlenose dolphins is
related to spatial arrangement of prey. Mar. Mamm. Sci. 16, 287–298.
Alexander, R. D. (1974). The evolution of social behaviour. Annu. Rev. Ecol. Syst. 5,
325–383.
Amos, B. (1993). Use of molecular probes to analyse pilot whale pod structure: Two novel
analytical approaches. Symp. Zool. Soc., London 66, 33–48.
Andrews, K., Karczmarski, L., Au, W. W. L., Rickards, S., Vanderlip, C. A. and
Toonen, R. J. (2006). Patterns of genetic diversity in the Hawaiian spinner dolphin
(Stenella longirostris). Atoll Res. Bull. 543, 63–73.
Arcese, P., Jongejan, G. and Sinclair, A. R. E. (1995). Behavioural flexibility in a small
African antelope: Group size and composition in the oribi (Ourebia ourebi, Bovidae).
Ethology 99, 1–23.
Arnold, P. W. (2002). Irrawady dolphin Orcaella brevirostris. In ‘‘Encyclopedia of Marine
Mammals’’ (W. F. Perrin, B. Würsig and J. G. M. Thewissen, eds), pp. 652–654.
Academic Press, San Diego.
Atkins, S., Pillay, N. and Peddemors, V. M. (2004). Spatial distribution of Indo-Pacific
humpback dolphins (Sousa chinensis) at Richards Bay, South Africa: Environmental
influences and behavioural patterns. Aquat. Mamm. 30, 84–93.
Baird, R. W. (2000). The killer whale—foraging specializations and group hunting.
In ‘‘Cetacean Societies: Field Studies of Dolphins and Whales’’ ( J. Mann, R. C. Connor,
P. Tyack and H. Whitehead, eds), pp. 127–153. University of Chicago Press, Chicago.
Baird, R. W. and Dill, L. M. (1995). Occurrence and behavior of transient killer whales:
Seasonal and pod-specific variability, foraging behavior and prey handling. Can. J. Zool.
73, 1300–1311.
Baird, R. W. and Dill, L. M. (1996). Ecological and social determinants of group size in
transient killer whales. Behavioral Ecology 7, 408–416.
Baird, R. W. and Whitehead, H. (2000). Social organization of mammal-eating killer
whales: Group stability and dispersal patterns. Can. J. Zool. 78, 2096–2105.
Baldwin, R. M., Collins, M., Van Waerebeek, K. and Minton, G. (2004). The Indo-Pacific
humpback dolphin of the Arabian region: A status review. Aquat. Mamm. 30, 111–124.
Ballance, L. T., Pitman, R. L. and Reilly, S. B. (1997). Seabird community structure along a
productive gradient: Importance of competition and energetic constraint. Ecology 78,
1502–1518.
Barrett-Lennard, L. G. (2000). Population structure and mating patterns of killer whales
(Orcinus orca) as revealed by DNA analysis. Ph.D. Thesis, University of British Columbia.
Barros, N. B. and Wells, R. W. (1998). Prey and feeding patterns of resident bottlenose
dolphins (Tursiops truncatus) in Sarasota Bay, Florida. J. Mammol. 79, 1045–1059.
The Social Structure and Strategies of Delphinids
279
Bateson, G. (1974). Observations of a cetacean community. In ‘‘Mind in the Waters’’
( J. McIntyre, ed.), pp. 146–165. C. Scribner’s and Sons, New York.
Bearzi, G., Notarbartolo di Sciara, G. and Politi, E. (1997). Social ecology of bottlenose
dolphins in the Kvarneric (northern Adriatic Sea). Mar. Mamm. Sci. 13, 650–668.
Bearzi, M. (2003). Behavioural ecology of the marine mammals of Santa Monica Bay,
California, Ph.D. Thesis, University of California.
Beauchamp, G. (1998). The effect of group size on mean food intake rate in birds. Biol. Rev.
73, 449–472.
Beauchamp, G., Belisle, M. and Girarldeau, L. A. (1997). Influence of conspecific attraction on
the spatial distribution of learning foragers in a patchy habitat. J. Anim. Ecol. 66, 671–692.
Bednekoff, P. A. and Lima, S. L. (1998). Re-examining safety in numbers: Interactions
between risk dilution and collective detection depend upon predator targeting behaviour. Proc. R. Soc. Lond., B 265, 2021–2026.
Beecham, J. A. and Farnsworth, K. D. (1999). Animal group forces resulting from predator
avoidance and competition minimization. J. Theor. Biol. 198, 533–548.
Bekoff, M. (2001). Cunning coyotes: Tireless tricksters, protean predators. In ‘‘Model Systems
in Behavioral Ecology: Integrating Conceptual, Theoretical, and Empirical Approaches’’
(L. A. Dugatkin, ed.), pp. 381–407. Princeton University Press, Princeton, NJ.
Benoit-Bird, K. J. (2004). Prey caloric value and predator energy needs: Foraging predictions
for wild spinner dolphins. Mar. Biol. 145, 435–444.
Benoit-Bird, K. J. and Au, W. W. L. (2003). Prey dynamics affect foraging by a pelagic
predator (Stenella longirostris) over a range of spatial and temporal scales. Behav. Ecol.
Sociobiol. 53, 364–373.
Benoit-Bird, K. J., Würsig, B. and McFadden, C. J. (2004). Dusky dolphins (Lagenorhynchus
obscurus) foraging in two different habitats: Active acoustic detection of dolphins and their
prey. Mar. Mamm. Sci. 20, 215–231.
Berger, J. (2004). The last mile: How to sustain long-distance migration in mammals.
Conserv. Biol. 18, 320–331.
Bernard, H. J. and Hohn, A. A. (1989). Differences in feeding habits between pregnant and
lactating spotted dolphins (Stenella attenuata). J. Mammal. 70, 211–215.
Bertram, B. C. R. (1978). Living in groups: Predators and prey. In ‘‘Behavioural Ecology:
An Evolutionary Approach’’ (C. J. Krebs and N. B. Davies, eds), pp. 64–96. Blackwell
Scientific Publications, London.
Bigg, M. A., Olesiuk, P. F., Ellis, G. M., Ford, J. K. B. and Balcomb, K. C. (1990). Social
organization and genealogy of resident killer whales (Orcinus orca) in the coastal waters of
British Columbia and Washington State. Rep. Int. Whaling Comm. (special issue) 12,
383–405.
Bisther, A. and Vangraven, D. (1995). Studies of the social ecology of Norwegian killer
whales (Orcinus orca). In ‘‘Whales, Seals and Man’’ (A. S. Blix, L. Walle and Ulltang,
eds), pp. 169–176. Elsevier Science, Amsterdam.
Bjrge, A. and Tolley, K. A. (2002). Harbor porpoise. In ‘‘Encyclopedia of Marine Mammals’’ (W. F. Perrin, B. Würsig and J. G. M. Thewissen, eds), pp. 549–551. Academic
Press, San Diego.
Blundell, G. M., Ben-David, M. and Bowyer, R. T. (2002). Sociality in river otters:
Cooperative foraging or reproductive strategies. Behav. Ecol. 13, 134–141.
Boesch, C. (1996). Social groupings in chimpanzees. In ‘‘Great Ape Societies’’
(W. C. McGrew, L. F. Marchant and T. Nishida, eds), pp. 101–113. Cambridge
University Press, Cambridge.
Boinski, S., Sughrue, K., Selvaggi, L., Quatrone, R., Henry, M. and Cropp, S. (2002). An
expanded test of the ecological model of primate social evolution: Competitive regimes
and female bonding in three species of squirrel monkeys (Saimiri oerstedii, S. boliviensis,
and S. sciureus). Behaviour 139, 227–261.
280
Shannon Gowans et al.
Bonabeau, E., Dagorn, L. and Freon, P. (1999). Scaling in animal group-size distributions.
Proc. Natl. Acad. Sci. USA 96, 4472–4477.
Bräger, S. (1999). Association patterns in three populations of Hector’s dolphin, Cephalorhynchus hectori. Can. J. Zool. 77, 13–18.
Brashares, J. S. and Arcese, P. (2002). Role of forage, habitat and predation in the behavioural plasticity of a small African antelope. J. Anim. Ecol. 71, 626–638.
Brashares, J. S., Garland, T. and Arcese, P. (2000). Phylogentic analysis of coadaptation in
behavior, diet and body size in the African antelope. Behav. Ecol. 11, 452–463.
Brown, J. S., Laundre, J. W. and MGurung, M. (1999). The ecology of fear: Optimal
foraging, game theory, and trophic interactions. J. Mammal. 80, 385–399.
Brownell, R. L. and Cipriano, F. (1999). Dusky dolphin Lagenorhynchus obscurus (Gray,
1828). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 85–104. Academic Press, London.
Brownell, R. L. and Donahue, M. A. (1999). Hourglass dolphin Lagenorhynchus cruciger
(Quoy and Gaimard, 1824). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and
R. Harrison, eds), pp. 121–135. Academic Press, London.
Brownell, R. L. and Ralls, K. (1986). Potential for sperm competition in baleen whales. Rep.
Int. Whaling Comm. (special issue) 8, 97–112.
Buckley, N. J. (1997). Spatial-concentration effects and the importance of local enhancement in the evolution of colonial breeding in seabirds. Am. Nat. 149, 1091–1112.
Bumann, D., Krause, J. and Rubenstein, D. (1997). Mortality risk of spatial positions in
animal groups: The danger of being in the front. Behaviour 134, 1063–1076.
Burdin, A. M., Hoyt, E., Sato, H., Filatova, O. and Tarasyan, K. (2001). The ecology of
Orcinus orca in southeast Kamchatka, Russia. In ‘‘14th Biannual Conference on the
Biology of Marine Mammals’’ Vancouver, BC, Canada.
Caldwell, M. (2001). Social and genetic structure of bottlenose dolphin (Tursiops truncatus) in
Jacksonville Florida, Ph.D. Thesis, University of Miami.
Caro, T. M., Graham, C. M., Stoner, C. J. and Vargas, J. K. (2004). Adaptive significance of
anti-predator behaviour in artiodactyls. Anim. Behav. 67, 205–228.
Cassens, I., Van Waerebeek, K., Best, P. B., Crespo, E. A., Reyes, J. and Milinkovitch, M. C.
(2003). The phylogeography of dusky dolphins (Lagenorhynchus obscurus): A critical examination of network methods and rooting procedures. Mol. Ecol. 12, 1781–1792.
Cezilly, F. and Benhamou, S. (1996). Optimal foraging strategies: A review. Revue D
Ecologie-La Terre Et La Vie 51, 43–86.
Chapman, C. A. (1990). Association patterns of spider monkeys: The influence of ecology
and sex on social organization. Behav. Ecol. Sociobiol. 26, 409–414.
Chapman, C. A., Wrangham, R. W. and Chapman, L. J. (1995). Ecological constraints
on group size: An analysis of spider monkey and chimpanzee subgroups. Behav. Ecol.
Sociobiol. 36, 59–70.
Childress, M. J. and Lung, M. A. (2003). Predation risk, gender and the group size effect:
Does elk vigilance depend upon the behaviour of conspecifics? Anim. Behav. 66, 389–398.
Chilvers, B. L. and Corkeron, P. J. (2001). Trawling and bottlenose dolphins’ social
structure. Proc. R. Soc. Lond., B 268, 1901–1905.
Cipriano, F. (1992). Behavior, feeding ecology, and life history of dusky dolphins
(Lagenorhynchus obscurus) off Kaikoura, New Zealand. Ph.D. Thesis, University of Arizona.
Clapham, P. J. (2000). The humpback whale: Seasonal feeding and breeding in a baleen
whale. In ‘‘Cetacean Societies: Field Studies of Dolphins and Whales’’ ( J. Mann,
R. C. Connor, P. L. Tyack and H. Whitehead, eds), pp. 173–196. University of Chicago
Press, Chicago.
Clutton-Brock, T. (2002). Behavioral ecology—Breeding together: Kin selection and
mutualism in cooperative vertebrates. Science 296, 69–72.
The Social Structure and Strategies of Delphinids
281
Clutton-Brock, T. H. (1989). Mammalian mating systems. Proc. R. Soc. Lond., B 236,
339–372.
Clutton-Brock, T. H., Gaynor, D., Mcllrath, G. M., Maccoll, A. D. C., Kansky, R.,
Chadwick, P., Manser, M., Skinner, J. D. and Brotherton, P. N. M. (1999). Predation,
group size and mortality in a cooperative mongoose, Suricata suricatta. J. Anim. Ecol. 68,
672–683.
Cockcroft, V. G. (1991). Incidence of shark bites on Indian Ocean Hump-backed dolphins
(Sousa plumbea) off Natal, South Africa. In ‘‘Cetaceans and Cetacean Research in the
Indian Ocean Sanctuary. Marine Mammal Technical Report No. 3’’ ( J. S. Leatherwood
and G. P. Donovan, eds). United Nations Environment Programme, Nairobi, Kenya.
Cockcroft, V. G. and Ross, G. J. B. (1990). Food and feeding of the Indian Ocean bottlenose
dolphins off Natal, South Africa. In ‘‘The Bottlenose Dolphin’’ ( J. S. Leatherwood and
R. R. Reeves, eds), pp. 295–308. Academic Press, San Diego, CA.
Connor, R. C. (2000). Group living in whales and dolphins. In ‘‘Cetacean Societies: Field
Studies of Dolphins and Whales’’ ( J. Mann, R. Connor, P. L. Tyack and H. Whitehead,
eds), pp. 199–218. University of Chicago Press, Chicago.
Connor, R. C. and Heithaus, M. R. (1996). Approach by great white shark elicits flight
response in bottlenose dolphins. Mar. Mamm. Sci. 12, 602–606.
Connor, R. C. and Norris, K. S. (1982). Are dolphins reciprocal altruists? Am. Nat. 119,
358–374.
Connor, R. C. and Whitehead, H. (2005). Alliances II. Rates of encounter during resource
utilization: A general model of intrasexual alliance formation in fission-fusion societies.
Anim. Behav. 69, 127–132.
Connor, R. C., Smolker, R. A. and Richards, A. F. (1992). Two levels of alliance formation
among male bottlenose dolphins (Tursiops sp.). Proc. Natl. Acad. Sci. USA 89, 987–990.
Connor, R. C., Mann, J., Tyack, P. L. and Whitehead, H. (1998). Social evolution in
toothed whales. Trends Ecol. Evol. 13, 228–232.
Connor, R. C., Heithaus, M. R. and Barre, L. M. (1999). Superalliance of bottlenose
dolphins. Nature 397, 571–572.
Connor, R. C., Read, A. J. and Wrangham, R. W. (2000a). Male reproductive strategies
and male bonds. In ‘‘Cetacean Societies: Field Studies of Dolphins and Whales’’
( J. Mann, R. Connor, P. L. Tyack and H. Whitehead, eds), pp. 247–269. University
of Chicago Press, Chicago.
Connor, R. C., Wells, R. S., Mann, J. and Read, A. J. (2000b). The bottlenose dolphin:
Social relationships in a fission-fusion society. In ‘‘Cetacean Societies: Field Studies of
Dolphins and Whales’’ ( J. Mann, R. C. Connor, P. Tyack and H. Whitehead, eds),
pp. 91–126. University of Chicago Press, Chicago.
Constantine, R., Visser, I., Buurman, D., Buurman, R. and McFadden, B. (1998). Killer
whale (Orcinus orca) predation on dusky dolphins (Lagenorhynchus obscurus) in
Kaikoura, New Zealand. Mar. Mamm. Sci. 14, 324–330.
Corkeron, P., Morris, R. J. and Bryden, M. M. (1987). Interactions between bottlenose
dolphins and sharks in Moreton Bay. Aquat. Mamm. 13, 109–113.
Corkeron, P. J. (1990). Aspects of the behavioral ecology of inshore dolphins Tursiops
truncatus and Sousa chinensis in Moreton Bay, Australia. In ‘‘The Bottlenose Dolphins’’
(S. Leatherwood and R. R. Reeves, eds), pp. 285–293. Academic Press, London.
Corkeron, P. J. and Martin, A. R. (2004). Ranging and diving behaviour of two ‘‘offshore’’
bottlenose dolphins, Tursiops sp., off eastern Australia. J. Mar. Biol. Assoc. UK 84,
465–468.
Corkeron, P. J., Bryden, M. M. and Hedstrom, K. E. (1990). Feeding by bottlenose dolphins
in association with trawling operations in Moreton Bay, Australia. In ‘‘The Bottlenose
Dolphin’’ (S. Leatherwood and R. R. Reeves, eds), pp. 329–336. Academic Press,
San Diego, CA.
282
Shannon Gowans et al.
Corkeron, P. J., Morrissette, N. M., Porter, L. and Marsh, H. (1997). Distribution and status of
hump-backed dolphins, Sousa chinensis, in Australian waters. Asian Mar. Biol. 14, 49–59.
Cox, T. F., Read, A. J., Swanner, D., Urian, K. W. and Waples, D. M. (2003). Behavioral
responses of bottlenose dolphins, Tursiops truncatus, to gillnets and acoustic alarms. Biol.
Conserv. 115, 203–212.
Creel, S. (1997). Cooperative hunting and group size: Assumptions and currencies. Anim.
Behav. 54, 1319–1324.
Creel, S. (2001). Cooperative hunting and sociality in African wild dogs, Lycaon pictus. In ‘‘Model
Systems in Behavioral Ecology: Integrating Conceptual, Theoretical, and Empirical
Approaches’’ (L. A. Dugatkin, ed.), pp. 466–490. Princeton University Press, Princeton, NJ.
Creel, S. and Creel, N. M. (1995). Communal hunting and pack size in African wild dogs,
Lycaon pictus. Anim. Behav. 50, 1325–1339.
Crespi-Abril, A. C., Garcı́a, N. A., Crespo, E. A. and Coscarella, M. A. (2003). Consumption of marine mammals by broadnose sevengill shark Notorynchus cepedianus in the
northern and central Patagonian Shelf. Lat. Am. J. Aquat. Mamm. 2, 101–107.
da Silva, V. M. F. and Best, R. C. (1994). Tucuxi Sotalia fluviatilis (Gervais, 1853).
In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 43–69. Academic Press, London.
Dahlheim, M. E. and Heyning, J. E. (1999). Killer whale Orcinus orca (Linnaeus, 1758).
In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 281–322. Academic press, London.
Dans, S. L., Crespo, E. A., Garcia, N. A., Reyes, L. M., Pedraza, S. N. and Alonso, M. K.
(1997). Incidental mortality of Patagonian dusky dolphins in mid-water trawling: Retrospective effects from the early 1980s. Rep. Int. Whaling Comm. 47, 699–703.
de Waal, F. B. M. (1982). ‘‘Chimpanzee Politics: Power and Sex Among Apes’’. Harper &
Row, New York.
Defran, R. H. and Weller, D. W. (1999). Occurrence, distribution, site fidelity, and school
size of bottlenose dolphins (Tursiops truncatus) off San Diego, California. Mar. Mamm. Sci.
15, 366–380.
Defran, R. H., Weller, D. W., Kelly, D. L. and Espinosa, M. A. (1999). Range characteristics of Pacific coast bottlenose dolphins (Tursiops truncatus) in the Southern California
Bight. Mar. Mamm. Sci. 15, 381–393.
Dohl, T. P., Bonnell, M. L. and Ford, R. G. (1986). Distribution and abundance of common
dolphin Delphinus delphis, in the southern California bight: A quantitative assessment
based upon aerial transect data. Fish. Bull. 84, 333–343.
Dorsey, E. M. (1983). Exclusive adjoining home ranges in individually identified minke
whales (Balaenoptera acutorostrata) in Washington State. Can. J. Zool. 61, 174–181.
Dorsey, E. M., Stern, J. S., Hoelzel, A. R. and Jacobsen, J. (1990). Minke whales
(Balaenoptera acutorostrata) from the west coast of North America: Individual recognition
and small-scale site fidelity. Rep. Int. Whaling Comm. (special issue) 12, 357–368.
Duffield, D. A., Ridgeway, S. H. and Cornell, L. H. (1983). Hematology distinguishes
coastal and offshore forms of dolphins (Tursiops). Can. J. Zool. 61, 930–933.
Dunbar, R. I. M. (1989). Social systems as optimal strategy sets: The costs and benefits of
sociality. In ‘‘Comparative Socioecology: The Behavioural Ecology of Humans and
Other Mammals’’ (V. Standen and R. A. Foley, eds), pp. 131–149. Blackwell Scientific
Publications, Oxford.
Durant, S. (1988). A minimum intervention approach to conservation: The influence of
social structure. In ‘‘Behavioral Ecology and Conservation Biology’’ (T. Caro, ed.),
pp. 105–129. Oxford University Press, Oxford.
Durant, S. (2000). Dispersal patterns, social organization and population viability. In ‘‘Behaviour and Conservation’’ (L. M. Gosling and W. J. Sutherland, eds), pp. 172–197.
Cambridge University Press, Cambridge.
The Social Structure and Strategies of Delphinids
283
Durham, B. (1994). The distribution and abundance of the humpback dolphins (Sousa
chinensis) along the Natal coast, South Africa. M.Sc. Thesis, University of Natal.
Elgar, M. A. (1989). Predator vigilance and group size in mammals and birds. Biol. Rev. 64,
13–33.
Elwen, S., Meÿer, M. A., Best, P. B., Kotze, P. G. H., Thornton, M. and Swanson, S.
(2006). Range and movements of female Heaviside’s dolphins (Cephalorhynchus heavisidii),
as determined by satellite-linked telemetry. J. Mammal. 87, 866–877.
Emlen, S. T. and Oring, L. W. (1977). Ecology, sexual selection, and the evolution of
mating systems. Science 197, 215–222.
Fisher, D. O. and Owens, I. P. F. (2000). Female home range size and the evolution of social
organization in macropod marsupials. J. Anim. Ecol. 69, 1083–1098.
Ford, J. K. B. (1991). Vocal traditions among resident killer whales (Orcinus orca) in coastal
waters of British Columbia. Can. J. Zool. 69, 1454–1483.
Ford, J. K. B. and Ellis, G. M. (1999). ‘‘Transients.’’ University of British Columbia Press,
Vancouver, BC.
Ford, J. K. B., Ellis, G. M. and Balcomb, K. C. (1998a). ‘‘Killer Whales: The Natural
History and Genealogy of Orcinus orca in British Columbia and Washington’’. UBC
Press, Vancouver, Canada.
Ford, J. K. B., Ellis, G. M., Barrett-Lennard, L. G., Morton, A. B., Palm, R. S. and
Balcomb, K. C. (1998b). Dietary specialization in two sympatric populations of killer
whales (Orcinus orca) in coastal British Columbia and adjacent waters. Can. J. Zool. 76,
1456–1471.
Ford, J. K. B., Ellis, G. M. and Balcomb, K. C. (2000). ‘‘Killer Whales: The Natural History
and Genealogy of Orcinus orca in British Columbia and Washington’’. UBC Press,
Vancouver, Canada.
Foster, W. A. and Treherne, J. E. (1981). Evidence for the dilution effect in the selfish herd
from fish predation on a marine insect. Nature 293, 466–467.
Frank, S. A. (2003). Perspective: Repression of competition and the evolution of cooperation. Evolution 57, 693–705.
Fretwell, S. D. and Lucas, H. L. (1970). On territorial behavior and other factors influencing
habitat distribution in birds. Acta Biotheor. 19, 16–36.
Gannon, D. P., Barros, N. B., Nowacek, D. P., Read, A. J., Waples, D. M. and Wells, R. S.
(2005). Prey detection by bottlenose dolphins, Tursiops truncatus: An experimental test of
the passive listening hypothesis. Anim. Behav. 69, 709–720.
Gaskin, D. E. (1968). Distribution of delphinidae (Cetacea) in relation to sea surface temperatures off eastern and southern New Zealand. N. Z. J. Mar. Freshwater Res. 2,
527–534.
Geffen, E., Gompper, M. E., Gittleman, J. L., Luh, H. K., MacDonald, D. W. and
Wayne, R. K. (1996). Size, life-history traits and social organization in the canidae:
A reevaluation. Am. Nat. 147, 140–160.
Gerard, J. F., Bideau, E., Maublanc, M. L., Loisel, P. and Marchal, C. (2002). Herd size in
large herbivores: Encoded in the individual or emergent? Biol. Bull. 202, 275–282.
Gero, S., Bejder, L., Whitehead, H. and Connor, R. C. (2005). Behaviourally specific
preferred associations in bottlenose dolphins, Tursiops spp. Can. J. Zool. 83, 1566–1573.
Gibson, Q. A. (2006). Non-lethal shark attack on a bottlenose dolphin (Tursiops sp.) calf.
Mar. Mamm. Sci. 22, 190–197.
Gingerich, P. D. (1998). Paleobiological perspectives on Mesonychia, Archaeiceti, and the
origin of whales. In ‘‘The Emergence of Whales: Evolutionary Patterns in the Origins of
Cetacea’’ ( J. G. M. Thewissen, ed.), pp. 423–449. Plenum Press, New York.
Gittleman, J. L. (1989). Carnivore group living: Comparative trends. In ‘‘Carnivore Behavior,
Ecology and Evolution’’ ( J. L. Gittleman, ed.), pp. 183–207. Cornell University Press,
Ithaca.
284
Shannon Gowans et al.
Goodall, R. N. P. (1994). Commerson’s dolphin Cephalorhynchus commersonii (Lacépède
1804). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 241–267. Academic Press, London.
Goold, J. C. (1998). Acoustic assessment of populations of common dolphin off the west
Wales coast, with perspectives from satellite infrared imagery. J. Mar. Biol. Assoc. UK 78,
1353–1368.
Gordon, J. C. D. (1987). Behaviour and ecology of sperm whales off Sri Lanka. Ph.D. Thesis,
University of Cambridge.
Gowans, S. (1999). Social organization and population structure of northern bottlenose
whales in the Gully. Ph.D. Thesis, Dalhousie University.
Gowans, S. (2002). Bottlenose whales. In ‘‘Encyclopedia of Marine Mammals’’ (W. F. Perrin,
B. Würsig and J. G. M. Thewissen, eds), pp. 128–129. Academic Press, San Diego.
Gowans, S. and Rendell, L. (1999). Head-butting in northern bottlenose whales (Hyperoodon
ampullatus): A possible function for big heads. Mar. Mamm. Sci. 15, 1342–1350.
Gowans, S. and Whitehead, H. (1995). Distribution and habitat partitioning by small odontocetes in the Gully, a submarine canyon on the Scotian Shelf. Can. J. Zool. 73, 1599–1608.
Gowans, S., Whitehead, H. and Hooker, S. K. (2001). Social organization in northern
bottlenose whales (Hyperoodon ampullatus): Not driven by deep water foraging? Anim.
Behav. 62, 369–377.
Greenwood, P. J. (1980). Mating systems, philopatry and dispersal in birds and mammals.
Anim. Behav. 28, 1140–1162.
Griffin, R. B. and Griffin, N. J. (2003). Distribution, habitat partitioning, and abundance of
Atlantic spotted dolphins, bottlenose dolphins and loggerhead sea turtles on the eastern
Gulf of Mexico continental shelf. Gulf Mex. Sci. 21, 25–34.
Guinet, C. and Bouvier, J. (1995). Development of intentional stranding hunting techniques
in killer whale (Orcinus orca) calves at Crozet Archipelago. Can. J. Zool. 73, 27–33.
Guinet, C., Barret-Lennard, L. G. and Loyer, B. (2000). Co-ordinated attack behaviour and
prey sharing by killer whales at Crozet Archipelago: Strageties for feeding on negativelybuoyant prey. Mar. Mamm. Sci. 16, 828–834.
Guissamulo, A. T. and Cockcroft, V. G. (2004). Ecology and population estimates of the
Indo-Pacific humpback dolphins (Sousa chinensis) in Maputo Bay, Mozambique. Aquat.
Mamm. 30, 94–102.
Hamazaki, T. (2002). Spatiotemporal prediction models of cetacean habitats in the midwestern North Atlantic ocean (from Cape Hatteras, North Carolina, USA to Nova
Scotia, Canada). Mar. Mamm. Sci. 18, 920–939.
Hamilton, W. D. (1971). Geometry for the selfish herd. J. Theor. Biol. 31, 295–311.
Hebblewhite, M. and Pletscher, D. H. (2002). Effects of elk group size on predation by
wolves. Can. J. Zool. 80, 800–809.
Heimlich-Boran, J. R. (1993). Social organization of the short-finned pilot whale,
Globicephala macrorhynchus, with special reference to the comparative social ecology of
delphinids. Ph.D. Thesis, Cambridge University.
Heithaus, M. R. (2001a). Shark attacks on bottlenose dolphins (Tursiops aduncus) in Shark
Bay, Western Australia: Attack rate, bite scar frequencies, and attack seasonality. Mar.
Mamm. Sci. 17, 526–593.
Heithaus, M. R. (2001b). Predator-prey and competitive interactions between sharks (order
Selachii) and dolphins (suborder Odontoceti): A review. J. Zool. (Lond.) 253, 53–68.
Heithaus, M. R. and Dill, L. M. (2002). Food availability and tiger shark predation risk
influence bottlenose dolphin habitat use. Ecology 83, 480–491.
Henzi, P. and Barrett, L. (2003). Evolutionary ecology, sexual conflict, and behavioral
differentiation among baboon populations. Evol. Anthropol. 12, 217–230.
Hersh, S. L. and Duffield, D. A. (1990). Distinction between northwest Atlantic offshore and
coastal bottlenose dolphins based on hemoglobin profile and morphometry. In ‘‘The
The Social Structure and Strategies of Delphinids
285
Bottlenose Dolphin’’ (S. Leatherwood and R. R. Reeves, eds), pp. 129–139. Academic
Press, San Diego.
Hinde, R. A. (1976). Interactions, relationships and social structure. Man 11, 1–17.
Hoelzel, A. R. (1991). Killer whale predation on marine mammals at Punta Norte, Argentina; food sharing, provisioning and foraging strategy. Behav. Ecol. Sociobiol. 29, 197–204.
Hoelzel, A. R., Potter, C. W. and Best, P. B. (1998). Genetic differentiation between
parapatric ‘nearshore’ and ‘offshore’ populations of bottlenose dolphins. Proc. R. Soc.
Lond., B 265, 1177–1183.
Hoese, H. D. (1971). Dolphin feeding out of water in a salt marsh. J. Mammal. 52, 222–223.
Horwood, J. W. and Cushing, D. H. (1978). Spatial distributions and ecology of pelagic fish.
In ‘‘Spatial Patterns in Plankton Communities’’ ( J. H. Steele, ed.), pp. 355–383. Plenum
Press, New York.
Hung, S. K. and Jefferson, T. A. (2004). Ranging patterns of Indo-Pacific humpback
dolphins (Sousa chinensis) in the Pearl River estuary, People’s Republic of China.
Aquat. Mamm. 30, 159–174.
Irwin, L. J. and Würsig, B. (2004). A small resident community of bottlenose dolphins,
Tursiops truncatus, in Texas: Monitoring recommendations. Gulf Mex. Sci. 2004, 13–21.
Isbell, L. A. (1991). Contest and scrabble competition: Patterns of female aggression and
ranging behavior among primates. Behav. Ecol. 2, 143–155.
Isbell, L. A. and Young, T. R. (2002). Ecological models of female social relationships in
primates: Similarities, disparities and some directions for future clarity. Behaviour 139,
177–202.
Isbell, L. A., Pruetz, J. D., Nzuma, B. M. and Young, T. P. (1999). Comparing measures of
travel distances in primates: Methodological considerations and socioecological implications. Am. J. Primatol. 48, 87–98.
Janson, C. H. and Goldsmith, M. L. (1995). Predicting group size in primates: Foraging costs
and predation risks. Behav. Ecol. 6, 326–336.
Jarman, P. J. (1974). The social organization of antelope in relation to their ecology.
Behaviour 48, 215–267.
Jefferson, T. A. (2000). Population biology of the Indo-Pacific hump-backed dolphin in
Hong Kong waters. Wildl. Monogr. 144, 1–65.
Jefferson, T. A. and Karczmarski, L. (2001). Sousa chinensis. Mammal. Species 655, 1–9.
Jefferson, T. A., Stacey, P. J. and Baird, R. W. (1991). A review of killer whale interactions
with other marine mammals: Predation to co-existence. Mamm. Rev. 4, 151–180.
Jetz, W., Carbone, C., Fulford, J. and Brown, J. H. (2004). The scaling of animal space use.
Science 306, 266–268.
Johnson, C. M. and Norris, K. S. (1994). Social behavior. In ‘‘The Hawaiian Spinner
Dolphin’’ (K. S. Norris, B. Würsig, R. S. Wells and M. Würsig, eds), pp. 243–286.
University of California Press, Berkley.
Johnson, D. D. P., Macdonald, D. W., Newman, C. and Morecroft, M. D. (2001). Group
size versus territory size in group-living badgers: A large-sample field test of the Resource
Dispersion Hypothesis. Oikos 95, 265–274.
Johnson, D. D. P., Kays, R., Blackwell, P. G. and Macdonald, D. W. (2002). Does the
resource dispersion hypothesis explain group living? Trends Ecol. Evol. 17, 563–570.
Jones, G. J. and Sayigh, L. S. (2002). Geographic variation in rates of vocal production of
free-ranging bottlenose dolphins. Mar. Mamm. Sci. 18, 374–393.
Kappeler, P. M. and Van Schaik, C. P. (2002). Evolution of primate social systems. Int. J.
Primatol. 23, 707–740.
Karczmarski, L. (1996). Ecological studies of humpback dolphins Sousa chinensis in the Algoa
Bay region, Eastern Cape, South Africa. Ph.D. Thesis, University of Port Elizabeth.
Karczmarski, L. (1999). Group dynamics of humpback dolphins Sousa chinensis in the Algoa
Bay region, South Africa. J. Zool. (Lond.) 249, 283–293.
286
Shannon Gowans et al.
Karczmarski, L. (2000). Conservation and management of humpback dolphins: The South
African perspective. Oryx 34, 207–216.
Karczmarski, L. and Cockcroft, V. G. (1999). Daylight behaviour of Humpback dolphins
Sousa chinensis in Algoa Bay, South Africa. Z. Saugetierkunde 64, 19–29.
Karczmarski, L., Thornton, M. and Cockcroft, V. G. (1997). Description of selected
behaviours of humpback dolphins Sousa chinensis. Aquat. Mamm. 23, 127–133.
Karczmarski, L., Cockcroft, V. G. and McLachlan, A. (1999a). Group size and seasonal
pattern of occurrence of humpback dolphins Sousa chinensis in Algoa Bay, South Africa.
S. Afr. J. Mar. Sci. 21, 89–97.
Karczmarski, L., Winter, P. E. D., Cockcroft, V. G. and McLachlan, A. (1999b). Population
analysis of Indo-Pacific humpback dolphins Sousa chinensis in Algoa Bay, Eastern Cape,
South Africa. Mar. Mamm. Sci. 15, 1115–1123.
Karczmarski, L., Cockcroft, V. G. and McLachlan, A. (2000a). Habitat use and preferences
of Indo-Pacific humpback dolphins Sousa chinensis in Algoa Bay, South Africa. Mar.
Mamm. Sci. 16, 65–79.
Karczmarski, L., Thornton, M. and Cockcroft, V. G. (2000b). Daylight occurrence of
humpback dolphins Sousa chinensis in Algoa Bay, South Africa. Afr. J. Ecol. 38, 86–90.
Karczmarski, L., Würsig, B., Gailey, G., Larson, K. W. and Vanderlip, C. (2005). Spinner
dolphins in a remote Hawaiian atoll: Social grouping and population structure. Behav.
Ecol. 16, 675–685.
Kasuya, T. and Tai, S. (1993). Life history of short-finned pilot whale stocks off Japan and a
description of the fishery. Rep. Int. Whaling Comm. (special issue) 14, 439–473.
Keith, M., Peddemors, V. M., Bester, M. N. and Ferguson, J. W. H. (2002). Population
characteristics of Indo-Pacific humpback dolphins at Richards Bay, South Africa: Implications of incidental capture in shark nets. S. Afr. J. Wildl. Res. 32, 153–162.
Kenagy, G. J. and Trombulak, S. C. (1986). Size and function of the mammalian testes in
relation to body size. J. Mammal. 67, 1–22.
Koenig, A. (2002). Competition for resources and its behavioral consequences among female
primates. Int. J. Primatol. 23, 759–783.
Komdeur, J. and Deerenberg, C. (1997). The importance of social behavior studies for conservation. In ‘‘Behavioral Approaches to Conservation in the Wild’’ ( J. R. Clemmons and
R. Buchholz, eds), pp. 262–276. Cambridge University Press, Cambridge.
Krakauer, D. C. (1995). Groups confuse predators by exploiting perceptual bottleneck:
A connection model of the confusion effect. Behav. Ecol. Sociobiol. 36, 421–429.
Krützen, M., Sherwin, W. B., Connor, R. C., Barre, L. M., Van de Casteele, T., Mann, J. and
Brooks, R. (2003). Contrasting relatedness patterns in bottlenose dolphins (Tursiops sp.)
with different alliance strategies. Proc. R. Soc. Lond., B 270, 497–502.
Krützen, M., Barré, L. M., Connor, R. C., Mann, J. and Sherwin, W. B. (2004). ‘O father:
Where art thou?’—Paternity assessment in an open fission-fusion society of wild bottlenose dolphins (Tursiops sp.) in Shark Bay, Western Australia. Mol. Ecol. 13, 1975–1990.
LaBarbera, M. (1989). Analyzing body size as a factor in ecology and evolution. Annu. Rev.
Ecol. Syst. 20, 97–117.
Laland, K. N. (1992). A theoretical investigation of the role of social transmission in
evolution. Ethol. Sociobiol. 13, 87–113.
Lammers, M. O. (2003). The behaviour and broadband acoustic signalling of Hawaiian
spinner dolphins (Stenella longirostris). Ph.D. Thesis, University of Hawaii.
Leduc, R. G., Perrin, W. F. and Dizon, A. E. (1999). Phylogenetic relationships among the
dephinid cetaceans based on full cytochrome b sequences. Mar. Mamm. Sci. 15, 619–648.
Lima, S. L. and Bednekoff, P. A. (1999). Temporal variation in danger drives antipredator
behaviour: The predation risk allocation hypothesis. Am. Nat. 153, 649–659.
Lopez, J. C. and Lopez, D. (1985). Killer whales (Orcinus orca) of Patagonia, and their
behavior of intentional stranding while hunting nearshore. J. Mammal. 66, 181–183.
The Social Structure and Strategies of Delphinids
287
Lucas, J. R., Creel, S. R. and Waser, P. M. (1996). How to measure inclusive fitness,
revisited. Anim. Behav. 51, 225–228.
Lukas, D., Reynolds, V., Boesch, C. and Vigiland, L. (2005). To what extent does living in a
group mean living with kin? Mol. Ecol. 14, 2181–2196.
Lusseau, D., Schneider, K., Boisseau, O. J., Haase, P., Slooten, E. and Dawson, D. (2003).
The bottlenose dolphin community of Doubtful Sound features a large proportion of
long-lasting associations: Can geographic isolation explain this unique trait? Behav. Ecol.
Sociobiol. 54, 396–405.
Macdonald, D. W. (1983). The ecology of carnivore social behaviour. Nature 301, 379–384.
MacFarlane, A. M. and Coulson, G. (2005). Synchrony and timing of breeding influences
sexual segregation in western grey and red kangaroos (Macropus fuliginosus and M. rufus).
J. Zool. (Lond.) 267, 419–429.
Maldini, D. (2003). Evidence of predation by a tiger shark (Galeocerdo cuvier) on a spotted
dolphin (Stenella attenuata) off O’ahu, Hawai’i. Aquat. Mamm. 29, 84–87.
Mann, J. and Barnett, H. (1999). Lethal tiger shark (Galeocerdo cuvier) attack on bottlenose
dolphin (Tursiops sp.) calf: Defense and reactions by the mother. Mar. Mamm. Sci. 15,
568–575.
Mann, J. and Sargeant, B. (2003). Like mother, like calf: The ontogeny of foraging traditions
in wild Indian ocean bottlenose dolphins (Tursiops sp.). In ‘‘The Biology of Traditions:
Models and Evidence’’ (D. M. Fragazy and S. Perry, eds), pp. 236–266. Cambridge
University Press, Cambridge, UK.
Mann, J. and Watson-Capps, J. J. (2005). Surviving at sea: Ecological and behavioral
predictors of calf mortality in Indian Ocean bottlenose dolphins, Tursiops sp. Anim.
Behav. 69, 899–909.
Mann, J., Connor, R. C., Tyack, P. L. and Whitehead, H. (2000). ‘‘Cetacean Societies’’.
University of Chicago Press, Chicago.
Markowitz, T. M. (2004). Social organization of the New Zealand dusky dolphin. Ph.D.
Thesis, Texas A&M University.
Markowitz, T. M., Harlin, A. D., Würsig, B. and McFadden, C. J. (2004). Dusky dolphin
foraging habitat: Overlap with aquaculture in New Zealand. Aquat. Conserv.: Mar.
Freshwat. Ecosyst. 13, 133–149.
Martien, K. K., Taylor, B. L., Slooten, E. and Dawson, S. M. (1999). A sensitivity analysis to
guide research and management for Hector’s dolphin. Biol. Conserv. 90, 183–191.
Mate, B. R., Stafford, K. M., Nawojchick, R. and Dunn, J. L. (1994). Movements and dive
behavior off a satellite-monitored Atlantic white-sided dolphin (Lagenorhynchus acutus) in
the Gulf of Maine. Mar. Mamm. Sci. 10, 116–121.
Mate, B. R., Lagerquist, B. A., Winsor, M., Geraci, J. B. and Prescott, J. H. (2005).
Movements and dive habitats of a satellite-monitored longfinned pilot whale (Globicephala melas) in the Northwest Atlantic. Mar. Mamm. Sci. 21, 136–144.
Matkin, C. O., Ellis, G., Olesiuk, P. and Saulitis, E. (1999a). Association patterns and
inferred genealogies of resident killer whales, Orcinus orca, in Prince William Sound,
Alaska. Fish. Bull. 97, 900–919.
Matkin, C. O., Ellis, G. M., Saulitis, E. L., Barrett-Lennard, L. G. and Matkin, D. R. (1999b).
‘‘Killer Whales of Southern Alaska’’. North Gulf Oceanic Society, Homer, Alaska.
McFadden, C. J. (2003). Behavioural flexibility of feeding dusky dolphins (Lagenorhynchus
obscurus) in Admiralty Bay, New Zealand. M.Sc. Thesis, Texas A&M University.
Milinski, M. and Parker, G. A. (1991). Competition for resources. In ‘‘Behavioural Ecology:
An Evolutionary Approach’’ ( J. R. Krebs and N. B. Davis, eds), pp. 137–168. Blackwell
Scientific Publications, London.
Möller, L. M., Behergaray, L. B., Harcourt, B. G. and Krützen, M. (2001). Alliance
membership and kinship in wild male bottlenose dolphins (Tursiops aduncus) of southeastern Australia. Proc. R. Soc. Lond., B 268, 1941–1947.
288
Shannon Gowans et al.
Natoli, A., Peddemors, V. and Hoelzel, A. R. (2004). Population structure and speciation in
the genus Tursiops based on microsatellite and mitochondrial DNA analyses. J. Evol. Biol.
17, 363–375.
Nishida, T. and Hosaka, K. (1996). Coalitions strategies among adult male chimpanzees of
the Mahale Mountains, Tanzania. In ‘‘Great Ape Societies’’ (W. C. McGrew,
L. F. Marchant and T. Nishida, eds), pp. 114–134. University of Cambridge Press,
Cambridge.
Noad, M. J., Cato, D. H., Bryden, M. M., Jenner, M. N. and Jenner, K. C. S. (2000).
Cultural revolution in whale songs. Nature 408, p. 537.
Norris, K. S. and Dohl, T. P. (1980). The structure and functions of cetacean schools.
In ‘‘Cetacean Behavior: Mechanisms and Functions’’ (L. M. Herman, ed.), pp. 211–261.
Wiley-Interscience, New York.
Norris, K. S. and Schilt, C. R. (1988). Cooperative societies in three dimensional space: On
the origins of aggregations, flocks, and schools with special reference to dolphins and fish.
Ethol. Sociobiol. 9, 149–179.
Norris, K. S., Würsig, B., Wells, R. S. and Würsig, M. (1994). ‘‘The Hawaiian Spinner
Dolphin’’. University of California Press, Los Angeles.
Nowacek, D. P. (2002). Sequential foraging behaviour of bottlenose dolphins, Tursiops
truncatus, in Sarasota Bay, Florida. Behaviour 139, 1125–1145.
Östman, J. S. O. (1994). Social organization and social behavior of Hawai’ian spinner
dolphins (Stenella longirostris). Ph.D. Thesis, University of California.
Ostreiher, R. (2003). Is mobbing altruistic or selfish behaviour? Anim. Behav. 66, 145–149.
Ottensmeyer, C. A. and Whitehead, H. (2003). Behavioural evidence for social units in
long-finned pilot whales. Can. J. Zool. 81, 1327–1338.
Owen, E. (2003). The reproductive and ecological functions of the pair-bond between
allied, adult male bottlenose dolphins, Tursiops truncatus, in Sarasota Bay, Florida. Ph.D.
Thesis, University of California, Santa Cruz.
Owen, E. C. G., Wells, R. S. and Hofmann, S. (2002). Ranging and association patterns of
paired and unpaired adult male Atlantic bottlenose dolphins, Tursiops truncatus, in
Sarasota, Florida, provide no evidence for alternative male strategies. Can. J. Zool. 80,
2072–2089.
Packer, C. and Caro, T. M. (1997). Foraging costs in social carnivores. Anim. Behav. 54,
1317–1318.
Packer, C. and Pusey, A. E. (1982). Cooperation and competition within coalitions of male
lions: Kin selection or game theory. Nature 296, 740–742.
Packer, C., Scheel, D. and Pusey, A. E. (1990). Why lions form groups: Food is not enough.
Am. Nat. 136, 1–19.
Papastavrou, V., Smith, S. C. and Whitehead, H. (1989). Diving behaviour of the sperm
whale, Physeter macrocephalus, off the Galapagos Islands. Can. J. Zool. 67, 839–846.
Parker, P. G. and Waite, T. A. (1997). Mating systems, effective population size, and
conservation of natural populations. In ‘‘Behavioral Approaches to Conservation in the
Wild’’ ( J. R. Clemmons and R. Buchholz, eds), pp. 243–262. Cambridge University
Press, Cambridge.
Parsons, E. C. M. (1998). The behaviour of Hong Kong’s resident cetaceans: The IndoPacific hump-backed dolphins and the finless porpoise. Aquat. Mamm. 24, 91–110.
Parsons, K. M., Durban, J. W., Claridge, D. E., Balcomb, K. C., Noble, L. R. and
Thompson, P. M. (2003). Kinship as a basis for alliance formation between male
bottlenose dolphins, Tursiops truncatus, in the Bahamas. Anim. Behav. 66, 185–194.
Patterson, I. A. P., Reid, R. J., Wilson, B., Grellier, K., Ross, H. M. and Thompson, P. M.
(1998). Evidence for infanticide in bottlenose dolphins: An explanation for violent
interactions with harbour porpoises? Proc. R. Soc. Lond., B 265, 1167–1170.
Payne, R. (1995). ‘‘Among Whales’’. Scribner, New York.
The Social Structure and Strategies of Delphinids
289
Peddemors, V. M. and Thompson, G. (1994). Beaching behaviour during shallow water
feeding by humpback dolphins Sousa plumbea. Aquat. Mamm. 20, 65–67.
Perrin, W. F. (1998). Stenella longirostris. Mammal. Species 599, 1–7.
Perrin, W. F. and Gilpatrick, J. W. (1994). Spinner dolphin Stenella longirostris (Gray, 1828).
In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds), pp. 99–128.
Academic Press, London.
Perrin, W. F. and Henderson, J. R. (1984). Growth and reproductive rates in two populations of spinner dolphins, Stenella longirostris, with different histories of exploitation. Rep.
Int. Whaling Comm. (special issue) 6, 417–430.
Perrin, W. F. and Mead, J. G. (1994). Clymene dolphin—Stenella clymene (Gray, 1846).
In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 161–171. Academic Press, London.
Perrin, W. F. and Mesnick, S. L. (2003). Sexual ecology of the spinner dolphins: Geographic
variation in mating system. Mar. Mamm. Sci. 19, 462–483.
Perrin, W. F., Akin, P. A. and Kashiwada, J. V. (1991). Geographic variation in external
morphology of the spinner dolphin Stenella longirostris in the eastern Pacific and implications for conservation. Fish. Bull. 89, 411–428.
Perrin, W. F., Wilson, C. E. and Archer, F. I. (1994). Striped dolphin Stenella coeruleoalba
(Meyen, 1833). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison,
eds), pp. 129–159. Academic Press, London.
Perrin, W. F., Dolar, M. L. L. and Robineau, D. (1999). Spinner dolphins (Stenella
longirostris) of the western Pacific and Southeast Asia: Pelagic and shallow-water forms.
Mar. Mamm. Sci. 15, 1029–1053.
Perryman, W. L. and Foster, T. C. (1980). ‘‘Preliminary Report on Predation by Small
Whales, Mainly the False Killer Whale, Pseudorca Crassidens, on Dolphins Stenella spp. and
Delphinus Delphis in the Eastern Tropical Pacific’’. National Marine Fisheries Service,
Southwest Fisheries Center.
Peters, R. H. (1983). ‘‘The Ecological Implications of Body Size’’. Cambridge University
Press, Cambridge, UK.
Petricig, R. O. (1995). Bottlenose dolphins (Tursiops truncatus) in Bull Creek, South
Carolina. Ph.D. Thesis, University of Rhode Island.
Pettifor, R. A., Norris, K. J. and Rowcliffe, J. M. (2000). Incorporating behaviour in
predictive models for conservation. In ‘‘Behaviour and Conservation’’ (L. M. Gosling
and W. J. Sutherland, eds), pp. 198–220. Cambridge University Press, Cambridge.
Pitman, R. L., Ballance, L. T., Mesnick, S. I. and Chivers, S. J. (2001). Killer whale
predation on sperm whales: Observations and implications. Mar. Mamm. Sci. 17,
494–507.
Pitman, R. L., O’Sullivan, S. and Mase, B. (2003). Killer whales (Orcinus orca) attack a school
of pantropical spotted dolphins (Stenella attenuata) in the Gulf of Mexico. Aquat. Mamm.
29, 321–324.
Poole, J. H. (1994). Sex differences in the behaviour of African elephants. In ‘‘Sex Differences
in Behaviour’’ (R. V. Short and E. Balaban, eds), pp. 331–346. Cambridge University
Press, Cambridge.
Poole, M. M. (1995). Aspects of behavioral ecology of spinner dolphins (Stenella longirostris)
in the nearshore waters of Moorea, French Polynesia. Ph.D. Thesis, University of
California.
Prins, H. H. T., van der Jeugd, H. P. and Beekman, J. H. (1994). Elephant decline in Lake
Manyara National Park, Tanzania. J. Afr. Ecol. 32, 185–191.
Pryor, K. and Kang Shallenberger, I. (1991). Social structure in spotted dolphins (Stenella
attenuata) in the tuna purse seine fishery in the Eastern Tropical Pacific. In ‘‘Dolphin
Societies: Discoveries and Puzzles’’ (K. Pryor and K. S. Norris, eds), pp. 160–196.
University of California Press, Berkley.
290
Shannon Gowans et al.
Pryor, K., Lindberg, J., Lindberg, S. and Milano, R. (1990). A dolphin-human fishing
cooperative in Brazil. Mar. Mamm. Sci. 6, 77–82.
Pulliam, H. R. (1973). On the advantages of flocking. J. Theor. Biol. 38, 419–422.
Pusey, A. E. and Packer, C. (1997). The ecology of relationships. In ‘‘Behavioural Ecology:
An Evolutionary Approach’’ ( J. R. Krebs and N. B. Davies, eds), pp. 254–283. Blackwell
Science, London.
Reeves, R. R., Smeenk, C. S., Brownell, R. L. and Kinze, C. C. (1999). White-sided
dolphin, Lagenorhynchus acutus (Gray 1828). In ‘‘Handbook of Marine Mammals’’
(S. H. Ridgway and R. Harrison, eds), pp. 31–56. Academic Press, San Diego.
Reilly, S. B. (1990). Seasonal changes in distribution and habitat differences among dolphins
in the eastern tropical Pacific. Mar. Ecol. Prog. Ser. 66, 1–11.
Rendell, L. and Whitehead, H. (2001). Culture in whales and dolphins. Behav. Brain Sci. 24,
309–342.
Reynolds, J. E., III, Wells, R. S. and Eide, S. D. (2000). ‘‘The Bottlenose Dolphin: Biology
and Conservation’’. University of Florida Press, Gainesville.
Rice, D. W. (1998). ‘‘Marine Mammals of the World: Systematics and Distribution’’. The
Society for Marine Mammalogy, Lawrence Kansas.
Ripple, W. J. and Beschta, R. L. (2004). Wolves and the ecology of fear: Can predation risk
structure ecosystems? Bioscience 54, 755–766.
Rogers, C. A., Brunnick, B. J., Herzing, D. L. and Baldwin, J. D. (2004). The social
structure of bottlenose dolphins, Tursiops truncatus, in the Bahamas. Mar. Mamm. Sci.
20, 688–708.
Ropert-Coudert, Y. and Wilson, R. P. (2005). Trends and perspectives in animal-attached
remote sensing. Front. Ecol. Environ 3, 437–444.
Ross, G. J. B. (1977). The taxonomy of bottlenosed dolphins Tursiops species in south
African waters, with notes on their biology. Ann. Cape Provincial Museums: Nat. Hist. 11,
135–194.
Ross, G. J. B. (1984). The smaller cetaceans of the south east coast of southern Africa. Ann.
Cape Provincial Museums: Nat. Hist. 15, 173–410.
Ross, G. J. B. and Cockcroft, V. G. (1990). Comments on Australian bottlenose dolphins
and the taxonomic status of Tursiops aduncus (Ehrenberd, 1832). In ‘‘The Bottlenose
Dolphin’’ ( J. S. Leatherwood and R. R. Reeves, eds), pp. 101–128. Academic Press,
San Diego.
Ross, G. J. B., Heinsohn, G. E. and Cockcroft, V. G. (1994). Humpback dolphins Sousa
chinensis (Osbeck, 1765), Sousa pumbea (G. Cuvier, 1829) and Sousa teuszi (Kukenthal,
1892). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 23–42. Academic Press, London.
Rossbach, K. A. and Herzing, D. L. (1997). Underwater observations of benthic feeding
bottlenose dolphins (Tursiops truncatus) near Grand Bahamas Island, Bahamas. Mar.
Mamm. Sci. 13, 498–503.
Rossbach, K. A. and Herzing, D. L. (1999). Inshore and offshore bottlenose dolphin
(Tursiops truncatus) communities distinguished by association patterns near Grand Bahama
Island, Bahamas. Can. J. Zool. 77, 581–592.
Ruckstuhl, K. E. and Neuhaus, P. (2005). (eds) ‘‘Sexual Segregation in Vertebrates: Ecology
of the Two Sexes’’. Cambridge University Press, Cambridge, UK.
Ryer, C. H. and Olla, B. L. (1995). Influences of food distribution on fish foraging
behaviour. Anim. Behav. 49, 411–418.
Saayman, G. S. and Tayler, C. K. (1979). The socioecology of humpback dolphins (Sousa sp.).
In ‘‘Behavior of Marine Animals’’ (H. E. Winn and B. L. Olla, eds), pp. 165–226. Plenum
Press, New York.
Sailer, L. D. and Gaulin, S. J. C. (1984). Proximity, Sociality and Observation: The
definition of social groups. Am. Anthropol. 86, 91–98.
The Social Structure and Strategies of Delphinids
291
Sargeant, B. L., Wirsing, A. J., Heithaus, M. R. and Mann, J. (2007). Can environmental heterogeneity explain individual foraging variation in wild bottlenose dolphins
(Tursiops sp.)? Behav. Ecol. Sociobiol. 61, 679–688.
Say, L., Pontier, D. and Natoli, E. (2001). Influence of oestrus synchronization on male
reproductive success in the domestic cat (Felis catus L.). Proc. R. Soc. Lond., B 268, 1049–1053.
Schoenherr, J. H. (1988). Blue whales feeding on high concentrations of euphausiids around
Monterey submarine canyon. M.Sc. Thesis, California State University, San Jose.
Schreer, J. F. and Kovacs, K. M. (1997). Allometry of diving capacity in air-breathing
vertebrates. Can. J. Zool. 75, 339–358.
Scott, M. D. and Chivers, S. J. (1990). Distribution and herd structure of bottlenose dolphins
in the eastern tropical Pacific Ocean. In ‘‘The Bottlenose Dolphin’’ (S. Leatherwood and
R. R. Reeves, eds), pp. 387–402. Academic Press, San Diego.
Scott, M. D. and Perryman, W. L. (1991). Using aerial photogrammetry to study dolphins
school structure. In ‘‘Dolphin Societies: Discoveries and Puzzles’’ (K. Pryor and
K. S. Norris, eds), pp. 227–244. University of California Press, Berkley.
Shane, S. H. (1980). Occurrence, movements and distribution of bottlenose dolphin,
Tursiops truncatus, in Southern Texas. Fish. Bull. US 78, 593–601.
Shane, S. H. (1995). Relationship between pilot whales and Risso’s dolphins at Santa
Catalina Island, California USA. Mar. Ecol. Prog. Ser. 123, 5–11.
Sharpe, F. A. (2001). Social foraging of the southeast Alaskan humpback whale, Megaptera
novaeangliae. Ph.D. Thesis, Simon Fraser University.
Similä, T. (1997). Sonar observations if killer whales (Orcinus orca) feeding on herring
schools. Aquat. Mamm. 23, 119–126.
Slooten, E. and Dawson, S. M. (1994). Hectors dolphin—Cephalorhynchus hectori (van
Benden, 1881). In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and
R. Harrison, eds), pp. 311–333. Academic Press, London.
Slooten, E., Dawson, S. M. and Whitehead, H. (1993). Associations among photographically identified Hector’s dolphins. Can. J. Zool. 71, 2311–2318.
Smolker, R. A., Richards, A. F., Connor, R. C. and Pepper, J. W. (1992). Sex differences in
patterns of association among Indian Ocean bottlenose dolphins. Behaviour 123, 38–69.
Smolker, R. A., Richards, A. F., Connor, R. C., Mann, J. and Berggren, P. (1997). Spongecarrying by Indian Ocean bottlenose dolphins: Possible tool-use by a delphinid. Ethology
103, 454–465.
Spong, G. and Creel, S. (2004). Effects of kinship on territorial conflicts among lions,
Panthera leo. Behav. Ecol. Sociobiol. 55, 325–331.
Stander, P. E. (1992). Cooperative hunting in lions—the role of the individual. Behav. Ecol.
Sociobiol. 29, 445–454.
Stanford, C. B. (1998). The social behaviour or chimpanzees and bonobos. Curr. Anthropol.
39, 399–420.
Stanford, C. B. (2002). Avoiding predators: Expectations and evidence in primate antipredator behavior. Int. J. Primatol. 23, 741–757.
Stankowich, T. and Blumstein, D. T. (2005). Fear in animals: A meta-analysis and review of
risk assessment. Proc. R. Soc. Lond., B 272, 2627–2634.
Steele, J. H. (1985). A comparison of terrestrial and marine ecological systems. Nature 313,
355–358.
Stephens, P. A., Russell, A. F., Young, A. J., Sutherland, W. J. and Clutton-Brock, T. H.
(2005). Dispersal, eviction and conflict in meerkats (Suricata suricatta): An evolutionarily
stable strategy model. Am. Nat. 165, 120–135.
Tarpy, C. (1979). Killer whale attack. Natl. Geogr. 155, 542–545.
Thewissen, J. G. M. (1998). Cetacean origins: Evolutionary turmoil during the invasion
of the oceans. In ‘‘The Emergence of Whales: Evolutionary Patterns in the Origins of
Cetacea’’ ( J. G. M. Thewissen, ed.), pp. 451–464. Plenum Press, New York.
292
Shannon Gowans et al.
Trivers, R. L. (1971). The evolution of reciprocal altruism. Q. Rev. Biol. 46, 35–57.
Turner, G. F. and Pitcher, T. J. (1986). Attack abatement: A model for group protection by
combined avoidance and dilution. Am. Nat. 128, 228–240.
Valsecchi, E., Hale, P. T., Corkeron, P. and Amos, W. (2002). Social structure in migrating
humpback whales (Megaptera novaeangliae). Mol. Ecol. 11, 507–518.
Van Horn, R. C., Engh, A. L., Scribner, K. T., Funk, S. M. and Holekamp, K. E. (2004).
Behavioural structuring of relatedness in the spotted hyena (Crocuta crocuta) suggests direct
fitness benefits of clan-level cooperation. Mol. Ecol. 13, 449–458.
Van Waerebeek, K. and Würsig, B. (2002). Pacific white-sided dolphin and dusky dolphin
Lagenorhynchus obliquidens and L. obscurus. In ‘‘Encyclopedia of Marine Mammals’’
(W. F. Perrin, B. Würsig and J. G. M. Thewissen, eds), pp. 859–861. Academic Press,
San Diego.
Van Waerebeek, K., Reyes, J., Read, A. J. and McKinnon, J. (1990). Preliminary observations of bottlenose dolphins from the Pacific coast of South America. In ‘‘The Bottlenose
Dolphin’’ (S. Leatherwood and R. R. Reeves, eds), pp. 143–154. Academic Press,
New York.
Van Waerebeek, K., Barnett, L., Camara, A., Cham, A., Diallo, M., Djiba, A., Jallow, A. O.,
Ndiaye, E., Ould Bilial, A. O. S. and Bamy, I. J. (2004). Distribution, status and biology
of the Atlantic humpback dolphins Sousa teuszi (Kukenthal, 1892). Aquat. Mamm. 30,
56–83.
Visser, I. (1999). Benthic foraging on stingrays by killer whales (Orcinus orca) in New Zealand
waters. Mar. Mamm. Sci. 15, 220–227.
Vucetich, J. A., Peterson, R. O. and Waite, T. A. (1997). Effects of social structure and prey
dynamics on extinction risk in gray wolves. Conserv. Biol. 11, 957–965.
Vucetich, J. A., Peterson, R. O. and Waite, T. A. (2004). Raven scavenging favours group
foraging in wolves. Anim. Behav. 67, 1117–1126.
Walker, W. A. (1981). ‘‘Geographic Variation in Morphology and Biology of Bottlenose
Dolphins in the Eastern North Pacific’’. National Marine Fisheries Service, Southwest
Fisheries Center Administrative Report no. LJ-81-03c.
Wang, J. Y., Chou, L. S. and White, B. N. (1999). Mitochondrial DNA analysis of sympatric
morphotypes of bottlenose dolphins (genus: Tursiops) in Chinese waters. Mol. Ecol. 8,
1603–1612.
Watts, D. P. (1998). Coalitionary mate guarding by male chimpanzees at Ngogo, Kibale
National Park, Uganda. Behav. Ecol. Sociobiol. 44, 43–55.
Weller, D. W. (1991). The social ecology of Pacific Coast bottlenose dolphins. M.Sc.
Thesis, San Diego State University.
Wells, R. S. (1991). The role of long-term study in understanding the social structure of a
bottlenose dolphin community. In ‘‘Dolphin Societies: Discoveries and Puzzles’’
(K. Pryor and K. S. Norris, eds), pp. 199–225. University of California Press, Berkley.
Wells, R. S. (2003). Dolphin social complexity: Lessons from long term study and life
history. In ‘‘Animal Social Complexity: Intelligence, Culture and Individualized Societies’’ (F. B. M. de Waal and P. L. Tyack, eds), pp. 32–56. Harvard University Press,
Cambridge, Massachusetts.
Wells, R. S. and Norris, K. S. (1994). The island habitat. In ‘‘The Hawaiian Spinner
Dolphin’’ (K. S. Norris, B. Würsig, R. S. Wells and M. Würsig, eds), pp. 31–53.
University of California Press, Berkeley.
Wells, R. S. and Scott, M. D. (1999). Bottlenose dolphin Tursiops truncatus (Montagu, 1821).
In ‘‘Handbook of Marine Mammals’’ (S. H. Ridgway and R. Harrison, eds),
pp. 137–182. Academic Press, San Diego.
Wells, R. S. and Scott, M. D. (2002). Bottlenose dolphins Tursiops truncatus and
T. aduncus. In ‘‘Encyclopedia of Marine Mammals’’ (W. F. Perrin, B. Würsig and
J. G. M. Thewissen, eds), pp. 123–128. Academic Press, San Diego.
The Social Structure and Strategies of Delphinids
293
Wells, R. S., Irvine, A. B. and Scott, M. D. (1980). The social ecology of inshore
odontocetes. In ‘‘Cetacean Behaviour: Mechanisms and Functions’’ (L. M. Herman,
ed.), pp. 263–317. John Wiley & Sons, Toronto.
Wells, R. S., Scott, M. D. and Irvine, A. B. (1987). The social structure of free-ranging
bottlenose dolphins. In ‘‘Current Mammalogy’’ (H. H. Genoways, ed.), pp. 247–305.
Plenum Press, New York.
Wells, R. S., Hansen, L. J., Ladridge, A., Dohl, T. P., Kelly, D. L. and Defran, R. H. (1990).
Northward extension of the range of bottlenose dolphins along the California coast. In
‘‘The Bottlenose Dolphin’’ (S. Leatherwood and R. R. Reeves, eds), pp. 421–431.
Academic Press, San Diego.
Wells, R. S., Boness, D. J. and Rathburn, G. B. (1999a). Behavior. In ‘‘Biology of Marine
Mammals’’ ( J. E. Reynolds, III and S. A. Rommell, eds), pp. 324–422. Smithsonian
Institution Press, Washington.
Wells, R. S., Rhinehart, H. L., Cunningham, P., Whaley, J., Baran, M., Koberna, C. and
Costa, D. P. (1999b). Long distance offshore movements of bottlenose dolphins. Mar.
Mamm. Sci. 15, 1098–1114.
White, F. J. and Chapman, C. A. (1994). Contrasting chimpanzees and bonobos: Nearest
neighbor distances and choices. Folia Primatol. 6, 181–191.
Whitehead, H. (1990). Rules for roving males. J. Theor. Biol. 145, 355–368.
Whitehead, H. (1996). Babysitting, dive synchrony, and indications of alloparental care in
sperm whales. Behav. Ecol. Sociobiol. 38, 237–244.
Whitehead, H. (1998). Cultural selection and genetic diversity in matrilineal whales. Science
282, 1708–1711.
Whitehead, H. (2003). ‘‘Sperm Whales: Social Evolution in the Ocean’’. Chicago University Press, Chicago, IL.
Whitehead, H. and Connor, R. C. (2005). Alliances 1. How large should alliances be? Anim.
Behav. 69, 117–126.
Whitehead, H. and Mann, J. (2000). Female reproductive strategies of cetaceans. In ‘‘Cetacean
Societies: Field Studies of Dolphins and Whales’’ ( J. Mann, R. Connor, P. L. Tyack and
H. Whitehead, eds), pp. 154–172. University of Chicago Press, Chicago.
Whitehead, H. and Wimmer, T. (2005). Heterogeneity and the mark-recapture assessment
of the Scotian Shelf population of northern bottlenose whales (Hyperoodon ampullatus).
Can. J. Fish. Aquat. Sci. 62, 2573–2585.
Whitehead, H., Christal, J. and Dufault, S. (1997). Past and distant whaling and the rapid
decline of sperm whales off the Galapagos Islands. Conserv. Biol. 11, 1387–1396.
Whitehead, H., Rendell, L., Osborne, R. W. and Würsig, B. (2004). Culture and conservation of non-humans with reference to whales and dolphins: Review and new directions. Biol. Conserv. 120, 427–437.
Williams, T. M. (1999). The evolution of cost efficient swimming in marine mammals:
Limits to energetic optimization. Philos. Trans. R. Soc.: Biol. Sci. 354, 193–201.
Wilson, B., Thompson, P. M. and Hammond, P. S. (1997). Habitat use by bottlenose
dolphins: Seasonal distribution and stratified movement patterns in the Moray Firth,
Scotland. J. Appl. Ecol. 34, 1365–1374.
Wilson, B., Hammond, P. S. and Thompson, P. M. (1999). Estimating size and assessing
trends in a coastal bottlenose dolphin population. Ecol. Appl. 9, 288–300.
Wilson, D. R. B. (1995). The ecology of bottlenose dolphins in the Moray Firth, Scotland:
A population at the northern extreme of the species’ range. Ph.D. Thesis, Aberdeen
University.
Wisenden, B. D. (1999). Alloparental care in fishes. Rev. Fish Biol. Fish. 9, 45–70.
Wittemyer, G., Douglas-Hamilton, I. and Getz, W. M. (2005). The socioecology of
elephants: Analysis of the processes creating multi-tiered social structures. Anim. Behav.
69, 1357–1371.
294
Shannon Gowans et al.
Wrangham, R. W. (1980). An ecological model of female-bonded primate groups. Behaviour
75, 262–300.
Wrangham, R. W. and Peterson, D. (1996). ‘‘Demonic Males: Apes and the Origins of
Human Violence’’. Houghton Mifflin Press, Boston, MA.
Wrangham, R. W. and Rubenstein, D. I. (1986a). ‘‘Ecological Aspects of Social Evolution’’. Princeton University Press, Princeton, NJ.
Wrangham, R. W. and Rubenstein, D. I. (1986b). Social evolution in birds and mammals.
In ‘‘Ecological Aspects of Social Evolution’’ (D. I. Rubenstein and R. W. Wrangham,
eds), pp. 452–470. Princeton University Press, Princeton, NJ.
Wrangham, R. W., Gittleman, J. L. and Chapman, C. A. (1993). Constraints on group size
in primates and carnivores: Population density and day-range as assays of exploitation
competition. Behav. Ecol. Sociobiol. 32, 199–209.
Würsig, B. (1978). Occurrence and group organization of Atlantic bottlenose porpoises
(Tursiops truncatus) in an Argentine bay. Biol. Bull. 154, 348–359.
Würsig, B. (1982). Radio tracking dusky dolphins in the South Atlantic. Mamm. Seas, FAO
Fish. Ser. 5, 145–160.
Würsig, B. (1986). Delphinid foraging strategies. In ‘‘Dolphin Cognition and Behavior: A
Comparative Approach’’ (R. J. Schusterman, J. A. Thomas and F. G. Wood, eds),
pp. 347–359. Lawrence Erlbaum Associates, Hillwood, NJ.
Würsig, B. and Bastida, R. (1986). Long-range movement and individual associations of two
dusky dolphins (Lagenorhynchus obscurus) off Argentina. J. Mammal. 67, 773–774.
Würsig, B. and Würsig, M. (1977). The photographic determination of group size, composition, and stability of coastal porpoises (Tursiops truncatus). Science 198, 755–756.
Würsig, B. and Würsig, M. (1980). Behavior and ecology of the dusky dolphin,
Lagenorhynchus obscurus in the south Atlantic. Fish. Bull. US 77, 871–890.
Würsig, B., Dorsey, E. M., Fraker, M. A., Payne, R. S. and Richardson, W. J. (1985).
Behavior of bowhead whales, Balaena mysticetus, summering in the Beaufort Sea:
A description. Fish. Bull. 83, 357–377.
Würsig, B., Wells, R. S., Würsig, M. and Norris, K. S. (1994a). Population structure.
In ‘‘The Hawaiian Spinner Dolphin’’ (K. S. Norris, B. Würsig, R. S. Wells and
M. Würsig, eds), pp. 122–140. University of California Press, Berkeley.
Würsig, B., Wells, R. S., Würsig, M. and Norris, K. S. (1994b). A spinner dolphin’s day.
In ‘‘The Hawaiian Spinner Dolphin’’ (K. S. Norris, B. Würsig, R. S. Wells and
M. Würsig, eds), pp. 65–102. University of California Press, Berkley.
Würsig, B., Cipriano, F., Slooten, E., Constantine, R., Barr, K. and Yin, S. (1997). Dusky
dolphins (Lagenorhynchus obscurus) of New Zealand: Status of present knowledge. Reports
of the International Whaling Commission 47, 715–722.
Würsig, B., Duprey, N. and Weir, J. (2007). ‘‘Dusky Dolphins (Lagenorhynchus obscurus) off
Kaikoura New Zealand: Present Knowledge and Research Goals. New Zealand Department of Conservation (DOC)’’ Science and Research Publication, Wellington,
New Zealand.
Young, T. P. and Isbell, L. A. (1994). Minimum group size and other conservation lessons
exemplified by a declining primate population. Biol. Conserv. 68, 129–134.
TAXONOMIC INDEX
Acanthina spirata, 48
Acipenser transmontanus, 75
Acropora hyacinthus, 47
Acropora millepora, 47
Acropora spathulata, 47
Adalaria proxima, 13, 48
Aldrichetta forsteri, 249
Alosa pseudoharengus, 75
Alosa sapidissima, 73
Ameiurus catus, 74
Anguilla rostrata, 75
Arbacia lixula, 17, 22, 49
Asterina minor, 49
Astreopora myriophthalma, 47
Astrobrachion constrictum, 49
Astropecten gisselbrechti, 49
Ateles geoffroyi, 226
Balaenoptera brydei, 274
Babylonia areolata, 48
Balaena mysticetus, 273
Balaenoptera acutorostrata, 274
Balaenoptera musculus, 259
Balaenoptera physalus, 273
Balanus balanoides, 32, 48
Botrylloides violaceus, 24–6
Brachidontes virgiliae, 47
Buccinum cyaneum, 48
Buccinum undatum, 30
Bugula neritina, 24–6, 29, 49
Bugula simplex, 49
Bugula stolonifera, 30, 49
Bugula turrita, 49
Bulla gouldiana, 48
Bushiella abnormis, 47
Calanus helgolandicus, 16
Calliostoma zizyphinum, 48
Cantharidus callichroa, 47
Caperea marginata, 274
Carcharhinus leucas, 212
Carcharhinus longimanus, 212
Carcharhinus obscurus, 212
Carcharodon carcharias, 212
Cephalorhynchus hectori, 198
Cephalorhynchus commersonii, 208
Cephalorhynchus heavisidii, 214
Cephalorhynchus hectori, 197
Cephalorhynchus, 205
Ceratomyxa Shasta, 146
Ceratoserolis trilobitoides, 30
Cercopithicus aethiops, 196, 237
Chlamys asperrima, 47
Chlamys bifrons, 47
Chorismus antarticus, 34
Chthamalus dentatus, 48
Ciona intestinalis, 18, 24–5, 31, 44, 49
Circeis armoricana, 47
Clupea harengus, 217
Clypeaster rosaceus, 49
Clypeaster subdepressus, 49
Connochates tauriuns, 243
Conus marmoreus, 29
Conus sp., 30, 35
Crepidula adunca, 48
Crepidula dilatata, 30, 48
Crucibulum quirqinae, 48
Cymatium corrugatum, 48
Cymatium cutaceum, 48
Cypraea caputdraconis, 47
Cypraecassis testiculus, 48
Cyprinus carpio carpio, 73
Delphinapterus leucas, 268
Delphinus capensis, 207, 216
Delphinus sp., 206, 215, 235, 241
Dendraster excentricus, 49
Dendropoma corrodens, 47
Dendropoma petraeum, 48
Diplasterias brucei, 49
Diplosoma listerianum, 24–5, 49
Drupella cornus, 47
Embioticidae sp., 244
Emerita analoga, 30
Engoniophos unicinctus, 47
Engraulis anchoita, 216
Engraulis sp., 247
Enhydra lutris, 260
Eschrichtius robustus, 259
Esox lucius, 74
Eubalaena glacialis, 274
Eulus gaimadrii, 34
Euprymna tasmanica, 20, 30
Euterpina acutifrons, 48
295
296
Favites halicora, 47
Feresa attenuata, 266
Flavobacterium, 142
Galeocerdo cuvier, 212
Gammarus duebeni, 30
Geryon Chaceon fenneri, 48
Geryon Chaceon quinquedens, 48
Globicephala sp., 206, 263
Globicephala macrorhynchus, 207
Globicephala melas, 210
Gonatus, 271
Goniastrea retiformis, 47
Grampus griseus, 215
Gyrodactylus salaris, 144
Haloa japonica, 23–4, 30
Haminoea vesicular, 48
Heliocidaris erythrogramma, 18, 23–4
Heliopora coerulea, 47
Henricia sp., 34
Holothuria scabra, 49
Homarus americanus, 30
Hydroides dianthus, 47
Hyperoodon ampullatus, 271
Ictalurus furcatus, 74
Ictalurus punctatus, 74
Isurus oxyrinchus, 212
Lagenodelphis hosei, 217
Lagenorhynchus acutus, 219
Lagenorhynchus cruciger, 208
Lagenorhynchus obliquidens, 217
Lagenorhynchus obscurus, 196, 210, 246, 248
Lagenorhynchus sp., 215, 246
Lagodon rhomboids, 216
Lepeophtheirus salmonis, 145
Leptosynapta clarki, 49
Ligia oceanica, 30
Lissodelphis sp., 208
Listonella anguillarum, 104
Loligo sp., 267
Loxodonta Africana, 197, 258
Luidia foliolata, 49
Luidia maculate, 49
Luidia quinaria, 49
Macaca sp., 196, 237
Macoma mitchelli, 47
Mediaster aequalis, 34
Megaptera novaengliae, 267
Meridiastra calcarm, 4
Micropterus dolomieu, 74
Micropterus salmoides, 74
Mirounga leonine, 261
Taxonomic Index
Mirounga sp., 212
Monodon monoceros, 268
Montipora capitata, 29
Montipora digitata, 47
Morone americana, 74
Morone saxatilis, 73
Mytilus edulis, 28
Myxobolus cerebralis, 104, 144
Notocrangon antarticus, 34
Notorynchus cepedianus, 212
Nucella crassilabrum, 48
Nucella lapillus, 48
Nucella ostrina, 24
Octomeris angulosa, 48
Odostomia Columbiana, 48
Oncorhynchus, 67
Oncorhynchus clarki, 67, 122
Oncorhynchus clarkii, 120
Oncorhynchus gorbuscha, 66–7, 92–5, 97, 134
Oncorhynchus gorbuschi, 121
Oncorhynchus keta, 66–7, 92–5, 99, 134
Oncorhynchus kisutch, 66–7, 92, 95, 97, 99, 135
Onchorhynchus masou, 67, 93
Oncorhynchus mykiss, 66–7, 74, 96, 99, 103, 105,
120, 122
Oncorhynchus nerka, 66–67, 91–2, 95, 97, 134
Oncorhynchus sp., 95, 97, 259
Oncorhynchus tshawytscha, 66–7, 92, 95, 97, 99,
120–1, 134–5, 138–9
Orcaella brevirostris, 197
Orcinus orca, 197, 259, 262
Ostrea edulis, 47
Otaria flavescens, 261
Ourebia ourebi, 238
Pachyseris speciosa, 47
Pagurus hirsutiuculus, 27
Pagurus longicarpus, 30, 48
Palaemon gravieri, 30
Pan troglodytes, 196, 224
Panthera leo, 220
Papio sp., 196, 237
Paracentrotus lividus, 17
Paradexiospira vitrea, 47
Parvulustra exigua, 4
Parvulustra parvivipara, 4
Patiriella regularis, 4, 49
Peponocephala electra, 266
Perca flavescens, 74
Petaloconchus montereyensis, 48
Phallusia obesa, 18
Phoca vitulina, 261
Phocoena phocoena, 227, 272
Phragmatopoma lapidosa, 30, 47
297
Taxonomic Index
Phyllacanthus imperialis, 49
Physeter macrocephalus, 197, 212
Pileolaria berkelyana, 47
Pisaster brevispinus, 49
Pisaster ochraceus, 49
Pocillopora damicornis, 47
Polinices lewisii, 48
Pomoxis nigromaculatus, 74
Porania Antarctica, 49
Porania sp., 49
Protolaeospira exima, 47
Pseudocalanus minutus, 13
Pseudorca crassidens, 212
Psolidium bullatum, 49
Psolus chitonoides, 49
Psychrophilum, 142
Pteraster militaris, 49
Pteraster tesselatus, 34
Pyura fissa, 49
Pyura stolonifera, 30, 49
Rangifer tarandus, 243
Redunca redunca, 238
Renibacterium salmoninarum, 142
Salmo marmoratus, 124
Salmo salar, 66–67, 84–90, 119, 121
Salmo trutta, 67, 69, 74, 90–1, 117, 121
Salvelinas alpinus, 67
Salvelinus confluentus, 124
Salvelinus fontinalis, 124
Salvelinus nyamacush, 121
Salvelinus umbla, 120, 122
Salvelinus, 67
Sander vitreus, 74–5
Sardinops sagax, 249
Sciaenidae sp., 244
Scottolana cnadensis, 13
Searlesia dira, 24, 27, 48
Sepioteuthis australis, 48
Serolis polita, 30
Solaster dawsoni, 34
Solaster endeca, 34
Somniosus microcephalus, 212
Somniosus pacificus, 212
Sotalia fluviatilis, 197
Sousa chinensis, 213, 255
Sousa plumbea, 255
Sousa sp., 197, 254
Sousa teuszi, 255
Sprattus antipodum, 249
Stenella attenuata, 235
Stenella clymene, 198
Stenella coeruleoalba, 198, 235
Stenella frontalis, 207, 217
Stenella longirostris, 196, 198, 207, 210, 231
Stenella longirostris centroamericana, 231
Stenella longirostris longirostris, 231
Stenella longirostris orientalis, 231
Stenella longirostris roseiventris, 231
Stenella sp., 206, 210, 215, 241
Streblospio benedicti, 30
Strombina francesae, 47
Strombina pumilio, 47
Strombus costatus, 47
Strombus gigas, 47
Strombus raninus, 47
Strongylocentrotus droebachiensis, 17–18
Styela plicata, 49
Stylochus ellipticus, 47
Suricata suricatta, 266
Thais emarginata, 48
Tragelaphus scriptus, 238
Tridacna squamosa, 47
Tubularia mesenbryanthemum, 47
Tursiops aduncus, 211, 213, 225–26, 243, 245
Tursiops sp., 196, 209, 213, 246
Tursiops truncates, 213–14, 218, 225–26,
243, 245
Uniophoragranifera, 31
Vermetus sp., 47
Verruca stroemia, 48
Watersipora subtorquata, 24–5, 49
This page intentionally left blank
SUBJECT INDEX
African antelope. See Resident communities
American Fisheries Society, 74–5, 166
Anadromous pacific salmon, 67
Animals grouping
ecological basis and levels of grouping, 200–1
major reasons, 199
social behaviour and relatioships, 201–2
ANOVA, testing offspring size within
populations, 7
Aquaculture classification, 68
Atlantic salmonids (Salmo), 67, 69
Atlantic white-sided dolphins, 219
Audits Division of Oregon’s Secretary
of State, 154
Baltic Sea, 76, 88–91
Belgium’s Meuse River basin, 77
Benefit-cost analysis (BCA), 151, 153–4,
156–8, 167
Big Qualicum hatchery, 139
Biological Board of Canada, 74–5
Bottlenose dolphins, 208, 211, 213, 215, 217,
222, 225, 227, 238, 245, 258
Bulletin of the Bureau of Fisheries, 75
Bull sharks, 212
Canada’s Department of Fisheries and Oceans
(DFO), 77
Canadian fish culture, 72
Captive broodstock hatcheries, 69
Captive propagation programmes, 65, 169
Cetaceans, comparisons with
baleen whales and, 273–4
harbour porpoise (Phocoena phocoena), 272–3
northern bottlenose whales (Hyperoodon
ampullatus), 271–2
sperm whales (Physeter macrocephalus), 269–71
Chinook hatchery, 154
Chinook salmon (O. tshawytscha)
exploitation rates on, 139
in North American Great Lakes, 70
stocks, in mixed-stock fisheries, 138
Chum salmon programme, 70
Clatsop County Economic Development
Council, 155
Coefficients of variation (CVs), 7
Columbia river, 115, 130
basin, 70
drainage system, 69
gillnet fishery, 155
hatchery, 156
Commerson’s dolphin, 208
Community. See Animals grouping
Conservation implications, 274–6
Conus species, 8, 14, 16, 20, 35
Cost–benefit ratios, 199
Cost-effectiveness analysis (CEA), 151–4, 167
Council’s Fish and Wildlife Programme, 153
Cultured organisms, 63
Demographic, social and cultural influences,
267–8
Direct developers, 20, 42–3
encapsulated and marine, 37
intrapopulation variation in offspring size, 8–10
offspring size–performance relationships,
42, 44
strong effects on postmetamorphic
performance, 21, 26
Dolphin ecology
distribution and habitat, 205–8
coastal and neritic, 208
riverine and inshore, 206
evolution of social strategies, 219–23
delphinid socioecology, 221–3
dolphin social structure, 222
foraging behaviour and diet, 215–18
cooperative foraging, 216, 218
pelagic prey species, 218
prey species in inshore and coastal areas, 217
predation and predatory risk, 208–15
anti-predator responses in open ocean, 212
behavioural responses, 211, 213
Gulf of Mexico, area of high prey and
predator density, 215
levels of shark predation on delphinid
communities, 212–13
predator–prey interactions, 214
by sharks and killer whales, 209
range of patterns of daily and seasonal
movements, 218–19
Dolphins
cetaceans, comparisons with, 268–73
conservation implications, 274
demographic, social and cultural
influences, 267
299
300
Dolphins (cont.)
foraging behaviour and diet in, 215–17
intermediate-ranging patterns, 253–65
predation and predatory risk in, 208–14
ranging patterns and daily movements in, 218
residentc communities, 223–36
wide-ranging communities, 239–51
Dusky dolphins, 210, 216, 233, 246, 248, 250
Economic perspectives, on hatchery programmes
BCA of hatchery programmes, 153–6
complicating factors, 156–9
cost-effectiveness of hatchery programmes,
151–3
measuring costs, effectiveness and benefits, 150–1
Egg incubation box, 66
Egg size, 5, 8
Egg volume and diameter, CVs in, 7
Elephant seals, 212–13, 261
Elk River Hatchery, 131
ElNiño Southern Oscillation (ENSO), 239
Elokomin hatchery, 156
Endangered Species Act, 65, 155
Entiat River, 155
Facultative planktotrophs, 6
False killer whales, 212, 215, 241
Fisheries Rehabilitation Enhancement and
Development (FRED), 157
Fishery enhancement
economic issues, 167
scientific and social dimensions, 168
Fishery enhancement hatcheries. See Production
hatcheries
Fishery enhancement, objectives and
activities, 161
Fishery management, 63
Foraging competition, 196, 217, 223, 240. See also
Dolphin ecology, foraging behaviour
and diet
Foraging groups. See Animals grouping
Game Commission, planting bass and walleye, 75
Genetic drift, 102–3, 109–12, 162. See also
Salmon hatchery programmes, genetic
risks in
Geographical extent of activities, salmonid
enhancement of, 84
Grand Coulee Dam, 70
Greenland sleeper sharks, 212
Gross National Product (GNP), 154
Habitat enhancement projects, 66
Harbour porpoise, 227, 272–3
Hatcheries
captive broodstock and supplementation, 69,
111–12, 114, 150
Subject Index
classification, 68
in Columbia basin and Oregon, varying costs
(CEA), 167
and commercial fishing, benefit-cost
analysis, 153
ecological and social implications of
salmonid, 65
fishery enhancement to conservation, 64, 145
mitigation, 69, 84
motivations and objectives of, 68–70
non-indigenous fisheries and, 70
Pacific salmon and, 67
production, 69, 77, 116, 122
programmes, political dynamics of, 78–84
reduce effects of disease in, 147
related disputes, 65
unexposed stocks release from, 147
and wild fish, interactions between, 162–6
Hatchery activities, historical survey, 71–8
Alosa sapidissima and Morone saxatilis,
colonizing, 73
American Fisheries Society, 74
Baltic Sea, hatchery programmes in, 76
Belgium’s Meuse River basin, 77
Biological Board of Canada, 74–5
Canada’s Department of Fisheries and Oceans
(DFO) role in production hatcheries, 77
Canadian fish culture, 72
Chilean salmon aquaculture industry and, 77
Cyprinus carpio carpio cultivation, 73
fishbreeding station at Huningen, 71
fish culture problems and compensating
policies, New England, 74
fish-ways and hatcheries, 75
Game Commission, planting bass and
walleye, 75
genetic homogenization in wild Atlantic
salmon populations, 77
home-stream theory, salmon and fishing
treaties, 76
institutionalized approach by France, 71
International Committee on Exploration of
Seas (ICES), 75
International Pacific Salmon Fisheries
Commission (IPSFC) and, 77
North American fishery agencies, 73
Norwegian salmon farming programme, 77
Oregon’s Fish Commission role, 75
railways and steamships in hatchery
programmes, 73
scientific research in, 75
species hybridization, Atlantic salmon and trout
populations in Sweden, 77
US Bureau of Fisheries (USBF) and, 74
US Fish Commission (USFC) and, 72
western salmonid fish culture and, 71
wild salmonid stocks in peril and, 78
Wisconsin’s Commissioners of Fisheries, 73
301
Subject Index
Hatchery fish, 70, 81, 89, 92, 96, 105, 114, 122,
128, 134, 146, 154, 165, 169
genetic risks associated with, 100–1
Hatchery programmes
economic perspectives on, 150–60
BCA of, 156–9
costs, effectiveness and benefits
measurement, 151–5
evaluating hatcheries, by broodstock origin
operation, 150
market prices for salmon, complication in, 159
Oregon cost-effectiveness audit, 155
SEP, role of, 158–9
trade-offs in designing and authorizing
projects, 160
genetic impacts evidence for, 111–26
captive broodstock, 111–12
introduction of new species, 124
management practises negate genetic
impacts, 124–6
production hatcheries, wild stocks and,
116–23
supportive breeding and, 112–16
political dynamics of, 78–83
contested prizes, salmonid hatcheries, 81
dam-building programmes in Washington
State, 82
declining stocks of wild salmonids, genetic
implications of, 81
fish culture policies, 83
Hokkaido Development Agency, role in,
82
private, non-profit’ (PNP) hatcheries, 81
social and economic politics and, 78
social conflicts, fish stocks and modern
development and, 80
Vermont’s legislature in, 79
Hawaiian spinner dolphins, 198, 267
Heaviside’s dolphins, 214
Hector’s dolphin, 197–8
Hokkaido hatcheries, 130. See also Hatcheries
Hourglass dolphin, 208
Humpback dolphin (Sousa sp.), 213, 217,
254–9, 265–6
in Algoa Bay region and Hong Kong, 256
characteristic feature of, 254
Indo-Pacific humpback dolphins
occurrence, 255
population estimates, 257
social dynamics and grouping patterns, 257
South African coast and, 256–7
taxonomic status of, 255
temporal segregation and travel range, 258
IHNV, in anadromous salmonids. See Infectious
diseases
Independent Economic Analysis Board (IEAB),
153, 155
Indigenous salmonids, propagation and
conservation
Atlantic salmon (S. salar), 84–90
baltic sea region, 88–90
eastern atlantic, 86–8
western atlantic, 84–6
Pacific salmon (Oncorhynchus spp.), 91
Alaska, 94–6
British Columbia, 91–2
Japan, 92–3
Russia, 93–4
United States, lower pacific states, 96–9
sea trout (S. trutta), 90–1
Indo-Pacific bottlenose dolphins, 211
Infectious diseases
of salmonids, 142
in wild and hatchery salmonids, 142–3
Inshore bottlenose dolphins, 225–31
Interception fisheries, 136, 165. See also Salmon
harvesting
International Baltic Sea Fishery Commission, 69
International Committee on the Exploration of
Seas (ICES), 75
International Pacific Salmon Fisheries
Commission (IPSFC), 77
Irrawaddy dolphin, 210
Japanese chum salmon, 130
Juvenile salmon growth, 66
Killer whales (Orcinus orca), 199, 209–12, 214,
217, 235, 241–2, 244, 259–64, 268,
271, 273
cooperative hunting in, 261
food-related social organization and behaviour,
262
matriline and vocal calls within community,
259–60
occurrence, communities, and behavioural
ecology, 259
population density, 211
in Prince William Sound area of Alaska, 260
sexual–activity relationships, 262
Lake Manyara National Park, 275
Larval activity levels, 5
Larval energetic reserves and nutrition, 5
Lecithotrophs
data analysis for Asteroidea (Echinodermata), 14
internal fertilizers and, 9
juvenile size in, 21
offspring size within populations, CV for, 8
survival differences observation in, 27
Malachite green, 141
Mammalian mating systems. See Social strategies,
mating system
302
Marine invertebrate
intraspecific variation and, 6–5
maternal size and offspring size in, 30–1
quantity and quality of larvae effects on, 5
range of offspring sizes, 4, 8
variation in offspring size in, 46–9
Maternal size, offspring size in marine
invertebrates and, 29–31
Mating strategies. See Social strategies, mating
system
Maximum sustainable yield (MSY), 134
Meta-analysis, offspring size within species, 6
Mitigation hatcheries, 69, 70
Mixed-stock fisheries, 70, 76, 83, 135, 137,
139, 153
Mixed-stock harvesting problem. See Wild
salmon populations, effects of harvest on
Ne/Nc ratios
in hatchery and captive population, 102,
111, 113
in wild salmon populations, 101
Nez Perce tribal hatchery, 154–5
Non-feeding taxa, 42
Non-indigenous fisheries, 70
Non-indigenous salmon and trout, enhancement
of, 99–100
North American fishery agencies, 73
Northern bottlenose whales, 271–2
Northwest Power Planning Council, 153
Norwegian salmon farming programme, 77
Ocean ranching, 69
Offspring size
degree of variation in, 6–10
and energetic content, 34
evolutionary implications, 43–4
evolutionary pressures acting on, 3
fitness function, 36–7
intraspecific variation in, 5, 8
measures of, 7
optimality models, marine invertebrates for, 37
variation, ecological implications, 43
Offspring-size models
offspring size-fitness function, 36–7
offspring size-number trade-off, 33–6
reconciling within-clutch variation, 38–9
Offspring sizes, effect on
developmental time, 13–20
direct developers, 20
non-feeding larvae, 18–19
planktotrophs, 13–18
fertilization and, 10–12
post-metamorphosis, 21–6
direct developers, 26–7
marine invertebrates and, 24
non-feeding larvae, 22–6
Subject Index
planktotrophs, 21–2
Offspring size, sources of variation in
among populations
habitat quality, 31–2
latitudinal clines, 32
within populations, 28–31
maternal size, 29–31
stresses, 28–9
Oregon Aqua-Foods hatchery, 131
Oregon’s Fish Commission role, 75
Oregon’s hatchery programmes, for
cost-effectiveness examination, 154
Outbreeding depression, 112, 116, 125–6, 163
hybridization and, 108–110
Pacific Salmon Commission Joint Chinook
Technical Committee, 139
Pacific salmon hatchery management, 67
Pacific salmonids, 67
Pacific sleeper, 212
Pelagic dolphins, 215
Pilot whales, 206, 210, 212, 214–15, 235, 241,
264, 267
Pink salmon fishing industry, 69
Planktotrophs, 13–16, 19–22, 40–2
entire planktonic period, interspecific
comparisons, 14–17
entire planktonic period, intraspecific
comparisons, 17–18
fertilization effect, on level of variation in
offspring size, 9
offspring size and fertilization, CV for, 8, 9
offspring size effects on, 20
pre-feeding period, 13–14
Population dynamics, 3
Population-level groups. See Animals grouping
Potential consequences, of enhancement activities
behavioural and ecological interactions,
100–26
disease effects of, 141
effects of harvest, 133–40
genetic risks associated with, 100
Prince William Sound, 69
Proceedings of the American Fish- Cultural
Association, 75
Production hatcheries, 69. See also Hatcheries
Propagule size effects, 3
Puget Sound, 130
Q10 values, 14–17
Rainbow trout, in Southern Hemisphere, 70
Redfish Lake, 69
Reproductive modes
in marine invertebrates, 36
offspring size within, 9
types for, 7
Subject Index
Resident communities
advantages to female, 223
inshore bottlenose dolphins (Tursiops sp.)
affiliations and fluid group structure, 229–30
associations and alliances within, 228–9
behavioural/social organization studies of, 226
fission–fusion nature of dolphin societies, 226
food aggregating and capturing
techniques, 227
foraging strategies, community level
differences in, 228
predation risk, 227
of Sarasota Bay and Shark Bay, 229–30
male–male cooperation, 224
resources on delphinid social strategies, 223–4
spinner dolphins (Stenella longirostris), 231–6
anti-predatory strategy and bisexually
bonded societies of, 234
behavioural ecology study, 231
contest competition for favourite
habitats, 235
fission–fusion system, 232–3
forage in coordinated pairs at depth, 232
habitat for daytime rest and social
activities, 233
high shark predation pressure to, 232
mating strategies of Kona coast, 233
in multi-species associations, 235
population figures of, 232
social pattern of atoll-dwelling spinner
dolphins, 235
subspecies of, 231
terrestrial mammals, comparisons with, 236–9
African antelope, patterns of residency and
sociality, 238
community sizes, 236
competition for food resources, primates, 237
contest competition, 237–8
evolution of group living in carnivores,
238–9
predictable resources, 236–7
small fluid groups, resident dolphins, 239
territoriality and infanticide in, 237
Risso’s dolphins, 215, 241, 267
Ryman–Laikre effect, 113
Salmon Action Plan, 69
Salmon enhancement activities, 64
Salmon growth projects, 66
Salmon harvesting, 133, 136
Salmon hatchery programmes, 64. See also
Salmonid enhancement, disease effects of
disease risks associated with, 143
altering ecology of watershed, 147
disease resistance of wild stocks, 146
endemic pathogens in hatchery fish, 144–5
exotic pathogens, 143–4
infected fish, contact wild stocks, 145
303
large numbers release, of uninfected fish, 147
reservoir for exposure of wild fish, 145–6
economic issues, 167
genetic risks associated with, 100, 124
Salmon hatchery programmes, genetic
risks in, 100–1
disease risks with salmon hatchery
programmes, 143–7
effects of hatcheries on, 101–106
hatchery regime effects on wild fish, 106–111
Salmonid enhancement, disease effects of, 141–9
approaches to control in hatchery fish, 147–8
disease risks with salmon hatchery
programmes, 143–7
fundamental problems with, 83
infectious diseases in wild and hatchery
salmonids, 142–3
possible implications of marine carrying
capacity for, 130
tools of risk assessment and control, 148
Salmonid hatcheries, 77–8, 81, 83, 143. See also
Hatcheries
ecological and social implications of, 65
Salmonid management, 65
Sensory Integration System. See Dolphin ecology;
predation and predatory risk
Seven-gill sharks, 212
Sex ratios, 102, 275
Short-fin mako, 212
Six-gill shark, 212
Smolt-to-adult survival, 134, 153–4
Smolt-to-adult survival rates (SARs), 154
Social strategies
biological pressures on, 197–9
male vs. female strategies, 202
marine environment influencing dolphin’s
ecology and social structure, 205
mating system, 202–3
predictable resources on delphinid social
strategies, 223–4
resource and risk distribution, ecological
factors in, 204–5
social complexity and variability, 203–4
unpredictable resources on delphinid social
strategies, 240
Social strategies, biological pressures on
animal grouping, 197–9
influence of ecology, 204
levels and definitions, 200
Socioecology
delphinid, 221–3
primate, 219–20
ungulate, 220–1
Sockeye salmon, 69
Spawning channels, 66
Sperm whales, 197, 211, 241, 268–71, 276
Spinner dolphins, 198, 205, 210, 217, 230–6,
241–2, 249–50, 259, 267
304
Spring Creek hatchery, 156
Supplementation hatcheries, 68–9
Terminal fisheries, 87, 136, 165. See also Wild
salmon populations, effects of harvest on
Terrestrial mammals, comparisons with, 265–6
resource dispersion hypothesis (RDH), 265–6
scramble competition for food resources, 266
Tiger sharks, 212, 227
Total sustainable yield (TSY), 137
Transactions of the American Fisheries Society, 75
Tucuxi, 197, 210, 212
Upriver Brights (URB), 137
US Bureau of Fisheries (USBF), 74
US Fish Commission (USFC), 72
Washington Department of Fish, 140
Western salmonid fish culture, 71
Wide-ranging communities
coastal bottlenose dolphins (Tursiops sp.), 243–6
photo-identification studies, 243
prey preferences of, 244
rapid movement of, 244
satellite telemetry, detailed movement
patterns for, 245
social structure and groups, 245
in waters of Little Bahama Bank, 245–6
dusky dolphins (Lagenorhynchus obscurus),
246–51
acrobatic leaps and diel cycle, 248
foraging, group size and social strategies, 247
less fission–fusion and social– sexual activity
in Argentina, 249, 250
mating system, 250
occurrence, size and generic relative,
246, 249
eastern tropical pacific dolphins (Delphinus and
Stenella sp.), 241–3
morphometric analysis of by-caught
dolphins, 241
multi-mate strategy of potential sperm
competition, 242
open ocean dolphins, travelling and muted
sexual dimorphism, 241–2
strategies in slim-bodied open ocean pelagic
dolphins, 243
terrestrial mammals, comparisons with, 251–3
Subject Index
African savannah and antelopes, 252
patterns of social structure and
movement, 252
primate food resources and wide-ranging
behaviour, 251
terrestrial carnivores and group sizes, 253
ungulates and macropod marsupials, 252
unpredictable resources on delphinid social
strategies, 240
Wild and hatchery-produced salmon
behavioural and ecological interactions,
127–33
competition among juveniles, 128
competition for food and space in
streams, 129
dominance relationships and effects of
increased density, 128
emergence of all salmonid species and
territorial behaviour, 127
estuary habitat, role in salmonid
ecology, 129
hatchery steelhead approach for
juveniles, 132
interactions on spawning grounds, 131
mortality in salmonids, 127
phenotype and genotype traits, implications
for, 131
predation, 132–3
predator swamping effects, 130
stock complexes of sockeye salmon, 130
Wild fish
genetic effects of hatchery fish on, 116–21
hatchery regime effects on, 106–11
effective population size changes, 106–8
hybridization and outbreeding depression,
108–10
structure of population, 110–11
populations, 63
Wildlife’s Wild Salmonid Policy, 140
Wild salmon populations, effects of harvest on,
133–41
and impacts of hatchery fish on wild fish, 136–40
selective harvesting, 140
sustainable harvest in mixed-stock fisheries,
135–6
Wisconsin’s Commissioners of Fisheries, 73
Yaquina River, 131