Pl. Syst. Evol. 256: 35–53 (2006)
DOI 10.1007/s00606-005-0302-z
Morphological variation in Puya (Bromeliaceae):
an allometric study
C. Hornung-Leoni and V. Sosa
Instituto de Ecologı́a, Xalapa, Veracruz, Mexico
Received April 12, 2004; accepted December 14, 2004
Published online: September 26, 2005
Springer-Verlag 2005
Abstract. A group of representative species of the
genus Puya was studied to determine if there are
allometric relationships between vegetative and
floral parts, whether these relationships correlate
with their pollination system and if plant size is
correlated with elevation and latitude. Fifty-three
species representing the morphological variation
and distribution of the genus were studied. Total
plant height, as well as leaf, inflorescence, petal and
sepal length were measured and these data subjected to univariate and multivariate analyses. To
test for correlation between plant size and altitudinal and latitudinal distribution, ANOVAs were
performed. When the pollination system of a
species is known, additional multivariate and
univariate analyses were also performed. The
results indicate that the characters studied are
correlated with a size component, exhibiting
positive allometry for sepal and petal length and
negative allometry for leaf length. Inflorescence
length is an isometric character. There was no
significant correlation between plant size and altitudinal or latitudinal distribution. The ANOVAs
show that the only character correlated to pollinator type was petal length. Small plants with small
flowers are correlated to pollinators such as
insects, while medium to large plants with medium
to large-sized flowers are correlated with pollinators such as birds and bats. Large plants have small
flowers, that are more evident and attract more
pollinators.
Key words: Allometry, altitudinal and latitudinal
distribution, Bromeliaceae, plant size, pollination
system, Puya.
Data from different groups of angiosperms
indicate that in general, the size of branches,
leaves, flowers and inflorescences are allometrically related (White 1983, Primack 1987,
Bond and Midgley 1988, Midgley and Bond
1989, Maitre and Midgley 1991), as are plant
height and seed size (Kang and Primack 1999).
Allometric growth is characterized by differential growth in the different parts of a body
(Niklas 1994) and by a change of shape with
changes in size (Brookstein 1991).
Many hypotheses have been proposed to
explain both the allometric relationships
among plant parts and the differences in the
size of angiosperms. The correlation between
the size of plants and the size of inflorescences
has been attributed to pollinator visits
(Donnelly et al. 1998, Elle and Carney 2003),
and by way of explanation, it has been suggested
that pollinators come into contact with a greater
number of flowers in large inflorescences than
they do in small ones (Wilson and Price 1977).
The correlation between plant size and the
number of seeds has also been explained by
pollinator visitation (Robertson and Macnair
36
1995). Similarly, petal size is correlated with
the number of pollinator visits; visit rate
increasing as size increases (Delph 1994).
Variation in the size and form of flowers has
been attributed to differences in the pollinator
or in breeding system (Grant and Grant 1965,
Ornduff 1969, Faegri and van der Pijl 1971).
The size of the different floral parts can be
allometrically correlated with other floral parts
and with the type of pollinator (Faegri and van
der Pijl 1971).
Furthermore, some evidence suggests that
there is a correlation between the size of the
plants and altitude for some groups of angiosperms in some geographical areas (Hedberg
1970, Mabberley 1972, Monasterio 1980, Miller
and Selander 1981). Plant size increases at
higher elevations for different taxa in the Andes
where, for instance, for Espeletia (Asteraceae)
in the Venezuelan Andes, taller growth forms
have been observed to replace lower forms as
elevation increases (Monasterio 1980). The
same phenomenon occurs for Afroalpine flora
where taxa such as Lobelia (Lobeliaceae) and
Senecio (Asteraceae) have giant individuals
at high elevations (Hedberg 1969, Mabberley
1972).
We selected representative species of Puya
to determine if there are morphological patterns and allometric relationships between
vegetative and floral characters. We also
selected this group of Puya species to determine if these characters are correlated with
pollination system and if plant size is correlated with elevation or ecological unit. Puya is
part of an important group of Neotropical
plants, the Bromeliaceae, in which allometric
relationships have not yet been studied.
Puya is a genus of about 195 species almost
all of which are restricted to South America
(Smith and Downs 1974, Benzing 2000). It has
not yet been established if Puya is a monophyletic group. However, preliminary results
of a current phylogenetic study of the species
of the subgenus Puya suggest that the genus is
indeed monophyletic (Hornung-Leoni and
Sosa, unpublished data). Among the reasons
for choosing Puya is the remarkable variation
C. Hornung-Leoni and V. Sosa: Allometry in Puya
in the size of the plants, their inflorescences
and their flowers. For example, the largest
bromeliad in the world P. raimondi, that grows
as tall as 10 – 12 meters, is part of this group
(Foster 1950, Hornung-Leoni and Sosa 2004).
Moreover, Puya species are found along a
varied altitudinal range mainly in the Andes,
from sea level up to 4000 m asl (Smith and
Downs 1974). Pollinator specialization is common among species in Puya and different
pollinators, from hummingbirds to perching
birds and also diurnal moths, bees and bats
have been recorded for this genus (OrtizCrespo 1973, Varadarajan and Brown 1988,
Benzing 2000).
In order to understand the factors that
have influenced taller growth forms in a group
of 53 species representative of the genus Puya,
the objectives of this study are to determine: 1)
if there are morphological patterns and allometric relationships among the structural parts
(vegetative and floral) in this group, 2) if
morphological patterns are related to the
pollination system, and 3) if there is any
correlation between the size of the plants and
the altitude or the latitude at which they grow.
Materials and methods
Selection of taxa. Fifty-three of the approximately
195 species of Puya were selected to represent the
morphological variation and geographic distribution of the genus. Puya has been divided in two
subgenera: Puya and Puyopsis (Smith and Downs
1974) and we considered species from both subgenera; i.e. those with inflorescences that have a
sterile apex (Puya) and those with a fertile apex
(Puyopsis). Species of Puya with a stem or stemless,
with erect or prostrate stem, and with the different
types of inflorescences were included. Variation in
altitude and latitude (ecological units) was also
taken into account for the selection of species.
Measurements were taken from 272 herbarium
specimens (listed in the Appendix) from the COL,
F, HDCV, MERC, NY, SGO, US, USM and XAL
herbaria. Additional information was obtained
from the literature and with field observations.
Characters. For floral characters, we dissected
the flowers, took digital images and then used the
C. Hornung-Leoni and V. Sosa: Allometry in Puya
program Sigma Scan Pro 5 (SPSS Inc. 1999). Stem
was coded in analyses as a qualitative character
(present, absent, erect, prostrate) because herbarium collections usually lack stems. For each
species, we only used the maximum values of
characters, based on two criteria: (1) maximum
values for characters were more likely to reflect the
genetic potential of the species in good environments (Thompson and Rabinowitz 1989), and (2)
for certain characters such as plant height only the
maximum values were available. The characters are
indicated in Table 1 and in Fig. 1. Data on
elevation were obtained from herbarium specimens
and from field work, and these were used to
estimate the average (minimum and maximum
values). The latitudinal categories were based in
ecological units of the same latitude where species
are found were obtained from specimens and
databases. Coastal mountains, the Andes and
highlands were the ecological units for this study.
Data on pollinators were obtained from field
observations as well as from the literature. The
type of pollinator was coded as birds (hummingbirds and perching birds), insects, bats, or by two
kinds of pollinators.
Statistical analyses
Allometric relationships. The computer program
STATISTICA (StatSoft 2000) was used for all
statistical analyses. Variation among characters
was explored in a preliminary manner with univariate analysis, by box plots for every character.
Correlation among characters was determined by
regression analysis.
Principal Components Analysis (PCA) was
then carried out with the following characters: leaf
length, inflorescence length, and petal and sepal
length. Among the reasons for using PCA are: (1) it
provides an idea of the importance of both the
size factor and the shape factor (Jolicoeur and
Mosimann 1960, Sundberg 1989, Jackson 1991);
(2) it allows the majority of the variation for a set
of morphological data to be summarized in only a
few dimensions (Klingenberg and Zimmermann
1992, Klingenberg 1996); (3) the coefficients of each
variable of the first component describe the relative
growth rate of all the components simultaneously
(Shea 1985) and, (4) it allows us to compare the
relative growth of one part of an organism to that of
the whole organism (Jackson 1991). PCA was
performed on natural-log-transformed data in order
37
to improve symmetry of distribution across species
(Hoaglin et al. 1993) and to guarantee the connection with the methods of shape analysis (Brookstein
1991). We used the correlation matrix for the PCA,
in order to give the same weight to variables that
differ widely (with the means standardized to zero
and variances standardized to 1) (Jackson 1991,
Klingenberg 1996). This was also done because
variances were not homogeneous and this is a
criterion for the selection of the matrix (Harris
1985). To determine how many factors were considered, we follow Kaiser’s criterion (eigen values>1) and the percentage of Variance Explained
(Hair et al. 1998, Reyment and Jöreskog 1993). The
loading scores of PC1, the size component, were
used as a size factor to determine allometry (Strauss
1985). These loadings were rescaled to a mean of 1.0
and interpreted as allometric coefficients on total
plant size. Values greater than unity describe
positive allometry, while those less than unity
indicate negative allometry and values equal to
one indicate isometry (Strauss 1985).
Based on the PCA plot, six groups were
identified. A Discriminant Analysis was then
performed to check whether the species belonged
to those groups and the position of taxa was based
on this analysis. Two-way ANOVAs were performed to determine if there was a significant
difference in the characters we tested depending on
type of pollinator. Additionally, to map out species
according to pollinators on a multivariate plot, a
PCA was performed as above, including only those
species for which the pollinator is known. Two-way
ANOVAs were also performed to determine if the
type of stem or ecological unit had a significant
effect on the size of the species and to determine if
there was a significant relationship between plant
size and latitude. To determine whether altitudinal
distribution had a significant effect on the size of
the species studied, we used a regression analysis of
the first component (PC1) with the average elevation as well as a two-way ANOVAs. ANCOVAs
were performed to determine if the type of
pollination or the type of stem and elevation have
an effect on the overall size of the plant.
Results
Allometric relationships
Univariate analysis. There was a clear correlation between leaf length and inflorescence
Number
of
samples
Total
plant
height
Leaf
Type
of stem length
Inflorescence
length
Sepal
lengh
Petal
length
Altitudinal Ecological Alloaverage
units
metric
group
1. Puya alpestris Gay
2. P. angulonis L.B.Sm.
3. P. aristeguietae L.B.Sm.
4. P. berteroniana Mez
5. P. bicolor Mez
6. P. boliviensis Baker
7. P. boyacana Cuatrec.
8. P. brachystachya Mez
9. P. cardonae L.B.Sm.
10. P. castellanosii L.B.Sm.
11. P. chilensis Molina
12. P. coerulea Miers
13. P. cryptantha Cuatrec.
14. P. ctenorhyncha L.B.Sm.
15. P. cylindrica Mez
16. P. fastuosa Mez
17. P. ferruginea
(Ruiz & Pav.) L.B.Sm.
18. P. floccosa E. Morr.
19. P. glomerifera Mez & Sodiro
20. P. goudotiana Mez
21. P. grubbii L.B.Sm.
22. P. hamata L.B.Sm.
23. P. herrerae Harms
24. P. hofstenii Mez
25. P. humilis Mez
26. P. laccata Mez
27. P. lilloi Castellanos
28. P. lineata Mez
29. P. longisepala Mez
30. P. maculata L.B.Sm.
13
7
6
21
7
13
2
1
1
4
18
22
3
1
2
1
16
1.50
1.00
3.00
5.00
3.00
2.00
1.00
0.40
0.30
3.00
5.00
2.00
0.70
1.20
1.00
2.50
2.50
1
0
0
1
0
0
0
0
0
0
1
0
0
0
1
0
2
60
50
100
100
80
100
45
20
30
40
100
60
30
50
50
40
100
100.0
50.0
100.0
100.0
140.0
50.0
10.0
7.0
60.0
200.0
100.0
100.0
20.0
30.0
40.0
100.0
200.0
2.5
2.5
3.0
2.3
2.1
3.2
1.6
1.7
1.5
3.0
3.5
2.4
1.6
2.0
3.0
2.0
4.5
5.0
5.0
6.0
5.0
4.0
5.0
3.5
3.0
3.0
4.0
5.0
5.0
2.8
4.0
4.2
3.7
8.0
1000
3750
3050
1045
2715
685
3350
2850
3200
2900
25
1250
3150
3500
2550
3575
2800
C, A
C, A
A
C
A
C
A
A
A
A
C
A
A
A
C
A
A
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
3
4
8
1
10
1
2
5
1
1
2
1
1
2.00
2.00
5.00
1.00
4.00
1.50
1.00
0.25
1.00
1.00
0.80
2.00
0.60
0
2
0
0
0
0
0
0
0
2
0
0
0
100
60
170
30
150
100
50
25
40
100
60
75
40
100.0
35.0
200.0
30.0
200.0
40.0
70.0
8.0
40.0
40.0
14.0
50.0
15.0
3.3
2.5
3.0
2.0
2.3
2.5
2.6
1.3
2.1
1.4
2.1
3.0
2.1
4.0
5.0
6.0
3.5
4.0
5.5
4.0
2.0
3.7
2.0
4.2
5.5
4.0
1635
2675
3020
2050
3625
3625
3450
3500
3400
1400
2935
2100
3345
A, H, C
A
A
A
A
A
A
A
A
A
A
A
A
1
1
1
1
1
1
1
2
1
1
1
1
1
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Species
38
Table 1. Species of Puya studied, with measurements for their morphological characters and their altitudinal range. Stem character codes: 0 =
absent, 1 = erect, 2 = prostrate. Ecological units: A = Andes, C = coastal mountains, H = highlands
Species
Number
of
samples
Total
plant
height
Type
Leaf
of stem length
Inflorescence
length
Sepal
lengh
Petal
length
Altitudinal Ecological Alloaverage
units
metric
group
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
2
1
1
1
6
7
3
3
3
10
23
3
1
1
2
6
8
1
6
2
1
2
1
2.00
0.37
1.00
0.10
1.50
2.00
1.00
2.00
0.20
0.30
12
1.00
1.50
1.20
0.80
2.00
2.00
0.60
0.60
2.00
2.50
1.00
4.00
0
2
0
0
0
0
0
1
0
0
2
1
0
2
0
0
0
0
0
0
0
0
0
40.0
20.0
50.0
4.0
30.0
62.0
50.0
100.0
9.0
5.0
430.0
60.0
70.0
50.0
8.0
50.0
30.0
10.0
22.0
150.0
50.0
35.0
50.0
3.7
2.0
0.8
1.1
6.0
3.0
3.2
1.6
1.7
1.8
4.0
3.5
4.2
1.8
2.8
2.0
2.5
1.8
2.2
3.0
2.0
2.5
4.5
7.0
4.0
1.3
1.5
9.5
7.0
4.0
2.5
2.5
3.0
8.0
6.0
6.0
2.8
3.5
3.6
3.2
3.0
3.5
6.0
4.0
4.5
6.0
2200
3375
1700
2500
1670
3100
2200
2325
3150
3375
3225
3450
1350
2500
3500
3450
3750
3900
3060
2765
3075
2350
4000
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
P.
mariae L.B.Sm.
medica L.B.Sm.
micrantha Mez
minima L.B.Sm.
mirabilis (Mez) L.B.Sm.
nitida Mez
oxyantha Mez
parviflora L.B.Sm.
pratensis L.B.Sm.
pygmaea L.B.Sm.
raimondii Harms
reflexiflora Mez
riparia L.B.Sm.
roezlii E.Morr.
santanderensis Cuatrec.
santosii Cuatrec.
trianae Baker
tristis L.B.Sm.
venezuelana L.B.Sm.
vestita Ed.Andre
westii L.B.Sm.
wurdackii L.B.Sm.
yakespala Castellanos
100
20
90
12
60
60
120
45
20
17
125
80
50
100
30
37
28
30
34
30
100
30
50
A
C
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
C
1
1
4
3
5
1
1
1
2
2
6
1
1
1
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Table 1. (continued)
1
1
1
1
1
1
1
1
39
40
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Fig. 1. Characters measured for the species of Puya studied for the analysis of allometry. A) Leaf length,
B) Total height of plants with a stem, C) Total height of stemless plants (considered only for ANOVAs),
D) Inflorescence length, E) Petal length, F) Sepal length
length, as well as a correlation between the
length of the sepals and that of the petals
(Fig. 2).
Multivariate analysis. The eigenvalues of
the PCA (Table 2) show that there is a strong
plant size component (PC1). The score of this
component represents the overall plant size
and is defined as a linear combination of
character, which increased in all variables
simultaneously (leaf, inflorescence, sepal and
petal length) for those which have the same
factor (Table 3). For PC1 all the variables are
significant (with a = 0.05) (Hair et al. 1998,
Reyment and Jöreskog 1993), and they explain
68.71% of the variation (Tables 2, 3). For the
second component, leaf and petal length are
significant (Table 3). PC2 accounts for 21.44%
of the variation. Together, the two principal
components explain 90.15% of the variation.
The PCA plot (Table 3, Fig. 3) shows that
size (PC1) increases from left to right, and in
the second component, leaf length increases
from bottom to top as petal length decreases.
The PCA plot (Fig. 3) shows that inflorescence and petal and sepal sizes increase
from left to right. Among characters, only the
inflorescence is an isometric character. Leaf
length shows negative allometry, while sepal
and petal length show positive allometry
(Fig. 3, Table 3).
Based on the results of the PCA plot
(Fig. 3) we can characterize morphological
variation for the species studied. The size
component PC1 (leaf, inflorescence, petal,
C. Hornung-Leoni and V. Sosa: Allometry in Puya
41
Fig. 2. Regression of variables showing a significant correlation. A) Inflorescence length and leaf length.
B) Petal length and sepal length
Table 2. Eigenvalues for the principal components (PC1 and PC2) in which all taxa were considered
PC (factor)
Eigenvalues
% Total
variance
Cumulative
eigenvalues
Cumulative
%
1
2
2.748
0.857
68.706
21.440
2.748
3.606
68.706
90.145
42
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Table 3. Relative growth dimensions with the factor loadings for the principal components (PC1 and PC2)
and rescaled PC1 loadings (this values >1 indicates positive allometry, and values >1 negative allometry)
Variables
Factor 1
(PC1)
Factor 2
(PC2)
Rescaled loadings
Allometry
(loading/average PC1)
Log leaf length
Log inflorescence length
Log sepal length
Log petal length
Explained variance
Proportion total
0.757
0.815
0.862
0.887
2.516
0.629
0.500
0.464
)0.402
)0.549
0.921
0.230
0.92
0.99
1.05
1.07
Negative
Isometric
Positive
Positive
Fig. 3. Plot of Principal Components Analysis in which all the species of Puya studied were considered. AG =
Allometric group
and sepal length) and the shape component
PC2 (leaves and petal length) show six groups
of individuals (confirmed by the discriminant
analysis), with the following morphological
patterns: (1) Medium – Medium: Individuals
with medium-sized plants and medium-sized
leaves and medium-sized petals. (2) Small –
Medium: Individuals with small-sized plants
with medium-sized leaves and medium-sized
petals. (3) Medium – Small: Individuals with
C. Hornung-Leoni and V. Sosa: Allometry in Puya
medium-sized plants, large leaves and smallsized petals. (4) Dwarf – Medium: Individuals
with dwarf-sized plants, medium-sized leaves
and medium-sized petals. (5) Medium – Large:
Individuals with medium to large-sized plants,
small leaves and very large-sized petals. (6)
Giant – Medium: Giant individuals with
medium-sized leaves and reduced petals.
Morphological patterns and pollination systems
The PCA plot (Fig. 4) that includes only
species for which the pollinator is known
shows that the length of petal and sepal are
correlated with the type of pollinator. PC1
represented 50.41% of the variation and
includes all of the variables: leaf, inflorescence,
sepal and petal length (0.83, 0.70, 0.68, and
43
0.81 respectively). PC2 includes variables such
as inflorescence, sepal and petal length (0.65,
)0.66 and )0.50 respectively). The PCA plot
(Fig. 4) shows that the size component
increases from left to right (on the axis) and
that inflorescence (shape component) increases
from the bottom to the top, while petal and
sepal length decrease, indicating that when the
inflorescence is larger the flowers are reduced.
The PCA plot (Fig. 4) shows that species with
medium to large plants with medium to large
sepals and petals are correlated with pollinators such as bats and birds, while small plants
with medium petals and sepals (like P. trianae)
are visited by insects.
The results of the ANOVAs indicate that
there were no significant difference in the size
of species (PC1), leaf length, inflorescence
Fig. 4. Plot of Principal Components Analysis which included only the species for which their pollinators are
known. Symbols: insects; bats; birds (including perching birds); hummingbirds; moths; autogamous
44
C. Hornung-Leoni and V. Sosa: Allometry in Puya
length and sepal length with respect to the type
of pollinator or between ecological units and
the size of plants (Table 4). Petal length was
the only significant character related to pollinator type.
Correlation between plant size and elevation
None of the characters (leaf, inflorescence and
petal length and PC1) showed any significant
difference (p>0.01) in the size of the species
(PC1) with respect to elevation (Fig. 5). The
ANCOVAs also indicate that there were no
significant differences in the type of pollinator,
size of plants and elevation, or in the type of
stem, size of plants and elevation; nor was
there any interaction of pollinators, plant size
and elevation (Table 4).
Discussion
We found characters exhibiting positive and
others with negative allometry. This means
that the characters that exhibit positive allometry (sepal and petal length) with respect to
the first component (size) are relatively larger
(with respect to plant height) in large individuals, while those with negative allometry (leaf
length) are relatively smaller in large individuals. When a character is allometric, a change
in size produces a change in form (Brookstein
1991). For example, as seen in Puya, petal
length shows positive allometry while leaf
length shows negative allometry. Puya mirabilis
has a different form compared to Puya
hamata. In the first species the plants have
short leaves but the flowers have very large
petals, while in the second species, the plants
have a similar plant length but both leaf and
petal length are medium-sized, and plant form
is very different.
Most of Puya species studied show the
same morphological pattern of medium-sized
plants and medium-sized leaves and mediumsized flowers (AG1). Morphological patterns
are not related to infrageneric classification
because species of Puya from both subgenera
display different allometric patterns. These
subgenera were established based on the presence of an sterile apex in inflorescences (Smith
and Downs 1974). An interesting issue worthy
of further investigation is whether these patterns are dependent on phylogenetic relationships.
Our results indicate that there is a relationship between sepal and petal length, as well
as the length of the inflorescence and pollinator type. The size of these characters is related
to the type of pollinator, as has been widely
reported (e. g. Grant and Grant 1965, Faegri
and van der Pijl 1971). For example, the
majority of the species with medium to large
inflorescences and medium to small flowers are
mainly pollinated by birds (perching birds and
hummingbirds). Plants with small inflorescences
with large flowers have bats as pollinators and
Table 4. Fisher and P values in ANOVAs and ANCOVAs
Variables
Fisher test
P-level
Significance
Pollinator · Log plant height
Pollinator · Log leaf length
Pollinator · Log inflorescence length
Pollinator · Log sepal length
Pollinator · Log petal length
Pollinator · PC1
Stem · PC1
Ecological units · PC1
Pollinator · PC1 · elevation
Stem · PC1 · elevation
F
F
F
F
F
F
F
F
F
F
p>0.01 (0.525)
p>0.01 (0.058)
p>0.01 (0.308)
p> 0.01 (p = 0.411)
p> 0.01 (p = 0.049)
p> 0.01 (p = 1.112)
P< 0.01 (0.259)
P< 0.01 (0.551)
p> 0.01 (0.057)
p> 0.01 (0.493)
NS
NS
NS
NS
S
NS
NS
NS
NS
NS
=
=
=
=
=
=
=
=
=
=
0.813
4.058
1.447
1.100
4.369
2.889
1.388
0.710
4.445
0.717
C. Hornung-Leoni and V. Sosa: Allometry in Puya
45
Fig. 5. Regression of variables and elevation. (A) Leaf length and average elevation, (B) Inflorescence length
and elevation, (C) Petal length and elevation, (D) Size component (PC1) and average elevation
medium flowers in small plants and are
commonly visited by insects. Yet, for a number
of the Puya species we studied there are
records of several pollinators, mainly for those
species with medium-sized flowers. It has
been pointed out that the match between
pollinator and floral characteristics is often
loose and not so specific as to exclude other
animal taxa from visiting and pollinating the
flowers (Waser 1983, Herrera 1996, Hingston
and McQuillan 2000), and our study confirmed
this finding.
Our results also indicate that neither elevation nor ecological units found at a certain
latitude are correlated with plant size for the
species we studied. Puya raimondii, the largest
Bromeliad is found in altitudes between 24004080 m asl. However, other tall species such as
Puya chilensis are found at sea level. Therefore
in the group of these Bromeliaceae, giant
growth forms are not exclusive to high elevations as found for Espeletia in the Andes
(Monasterio 1980) or Lobelia and Senecio in
the Afroalpine flora (Hedberg 1969, 1970;
Mabberly 1972).
Alternative hypotheses have been suggested to explain taller growth forms in
Bromeliaceae based on other environmental
46
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Table 5. Pollinators for the species considered in the second PCA analysis
Species
Pollinators
Reference
Puya alpestris
Puya aristeguietae
Puya berteroniana
Perching birds and hummingbirds
Bats
Perching birds, diurnal
moths and hummingbirds
Perching birds, diurnal moth
and hummingbirds
Birds
Bats
Birds
Bats (also recorded as autogamous)
Bats
Hummingbirds and perching birds
Our observations
Benzing, 2000
Benzing, 2000; Our observations
Puya chilensis
Puya
Puya
Puya
Puya
Puya
Puya
coerulea
ferruginea
f loccosa
mirabilis
nitida
raimondii
Puya trianae
Insects
factors. According to Benzing (2000) lack of
humidity is a determinant of the increase in
plant size for this family. Givnish et al. (1997)
indicate that in other giant bromeliad taxa,
such as Brocchinia micrantha and B. paniculata
factors such as a cool, extremely wet climate
and exceedingly nutrient-poor soils can be
responsible of the increase in plant size. The
available information suggests that Puya
raimondii is found in conditions of cold, dry
climate and nutrient-poor soils (Foster 1950,
Benzing 2000, Kessler 2002). However, more is
required to determine if these environmental
factors influence the overall size in Puya. In
other monocotyledons such as palms, the
largest taxa are basal to other more recently
evolved groups (Henderson 2002). Therefore,
it would be interesting to know if taller growth
forms in Puya are the result of a phylogenetic
constraint. It has been suggested that in hapaxanthic plants (i.e. those that only produce
inflorescences and fruit once in their life, after
which they die), an increase in size increases
fitness (Metcalf et al. 2003). It would also be
interesting to know if the size of the hapaxanthic
Puya (among them, P. raimondii and P. hamata)
has increased.
The variability among the species of Puya
in plant size, number of inflorescences and
Scogin, 1988; Benzing, 2000;
Our observations
Scogin, 1988
Benzing, 2000; Kessler, 2002
Scogin, 1988
Benzing, 2000
Benzing, 2000
Foster, 1950; Rivera, 1985;
Benzing, 2000; Our observations
Chaparro, 2002
number of flowers, as well as the wide variety
of habitats it occupies – from mesic to xeric, its
broad altitudinal distribution and its morphological distribution is notable. Given this, Puya
is an interesting group for phylogenetic study
to determine the evolution of gigantism, to
understand the evolution of pollination systems and the role of other environmental
factors affecting the size of plants.
The first author extends grateful thanks to the
‘‘Red Latinoamericana de Botánica’’ (RLB) for
awarding her a doctoral scholarship and providing
special funds to carry out the research for this
thesis project (RLB D1-02). She also thanks the
International Association for Plant Taxonomists
(IAPT) for financial support to carry out revision
in the Chilean herbaria. She thanks Dr. Gloria
Montenegro and her staff for their help with
fieldwork in Chile. We thank Mery Suny and
Giovana Vadillo for their help in fieldwork in
Peru. We thank Dr. Mario M. Ojeda for his sound
advice on statistical analysis and Jorge López
Portillo, J. Antonio Guerrero, Ivón Ramı́rez and
two anonymous reviewers for their suggestions on
the manuscript. We are grateful to the curators of
the COL, F, HDCV, MERC, NY, SGO, US, USM
and XAL herbaria for their help with regard to
collections. Edmundo Saavedra kindly provide the
line drawings. We thank Bianca Delfosse for editing
the English version of this manuscript.
C. Hornung-Leoni and V. Sosa: Allometry in Puya
47
Appendix. Herbarium specimens used for this study. The number in parenthesis corresponds to the number
of specimens analyzed for every species
Puya alpestris Gay CHILE: J. West 4981, US; N. Floy Bracelin 2802, F; N. Floy Bracelin 2766, F; La
Hermida, cerca de Santiago, G. Looser 2104, F; F (739482); V/1884, US (Photo type
Puya whytei ); Prov. Concepción. Hills north to Concepción, J. West 4981, US; F
(1670523); Cerro San Cristóbal, Germain s.n., 1854, SGO; Coquimbo, Muñoz M.
909, SGO; Coquimbo, Muñoz C. & A. Coronel 1552, SGO; Coquimbo, Muñoz C. &
A. Coronel 1302, SGO; Cerro de Fray Jorge, Philippi s.n., SGO; Fray Jorge, Ovalle,
Philippi F. s.n., I/1883, SGO; Cordillera de La Dehesa, Philippi s.n., XI/1861, SGO.
Puya angulonis
PERU: Camino a Caro, M.I. La Torre 1908, USM; Prov. Cajamarca. Depto. CaL.B.Sm.
jamarca, P.C. Hutchinson 7035, USM; Dpto. Ayacucho, A. Cano 12079, USM;
Dpto. Cuzco, Prov. Calca, R. Ferreyra 9882, USM; Dpto. Cajamarca, Al Gentry
1988, USM; Prov. Cajamarca, D .N. Smith & R. Vásquez 3459, F; Prov. Cajamarca,
P.C. Hutchinson & J. Wright 7035, F.
Puya aristeguietae VENEZUELA: Edo. Mérida, Hornung, C., Tirado, C. & Quevedo, J. 208, MERC;
L.B.Sm.
Edo. Mérida, Hornung, C., Garbiso, C. & Quevedo, J. 161, MERC; Edo. Mérida,
Hornung, C., Gaviria, J.C & Tirado, C. 246, MERC; Edo.Trujillo, L. Aristeguieta
3539, US, NY; Edo. Trujillo, Dorr L.J. & L.C. Bernett 5000, NY.
Puya berteroniana BOLIVIA (?) : Near La Paz (probably the label is an error), Rusby 2850, US. CHILE:
Mez
IIX Región, Prov. Santiago, G. Montero O. 768, F; Prov. Curicó. Hacienda Monte
Grande, E. Wedermann 563, F; Bertero 115, F (779464); F (1026657); Viña del Mar,
F (678758); Limache, Belen s.n., F (633929); Cuesta de La Dorminda, Luis Gonzalez
s.n., 26/V/1983, HDCV; Hutchison P. 22, SGO; Los Molles, Landbeck L. s.n., XI/
1861, SGO; Altos de Tiltil, Reiche K. s.n., 01/XI/1897, SGO; Santiago: Laguna de
Aculeo, Navas E. s.n., 17/XII/1971, SGO; Las Tacas, Flores A. s.n., 15/VIII/1986,
SGO; Coquimbo, Muñoz M. 908, SGO; Maipu, Schlegel F. 1664, SGO; Reserva
Nacional Rio Clarillo, Yanez J. s.n., 03/VIII/1983, SGO; Rancagua, Bertero 115,US
(photo); Valparaiso, Weber 3252, US; Concón, Miers 347, US.
Puya bicolor Mez
COLOMBIA: Dpto. Boyacá, Cordillera Oriental, J. A. Ewen 15642, COL; J. Cuatrecasas 9641, COL, F; Dpto. de Boyacá, Mpio de Aquitania, O. Rangel & J. Aguirre
238, COL; Dpto. Boyacá, Pauna, H. Garcı́a B. 13218, COL; Dpto.Cundinamarca,
Suesca, M.B. Foster & R. Foster 1803, COL; Dpto. Cundinamarca, Huertas & Camargo 104, F.
Puya boliviensis
BOLIVIA (CHILE now): Cobija. Gaudichaud s.n., F (741178); Gaudichaud s.n., US
Baker
(21449992); Cobija, Gaudichau, Julliet s.n., 1836–1837, F; Gaudichau, F (1435081).
CHILE: Quebrada Paposo, ca. 5–7 km E of Caleta, M.O. Dillon & J.T.S. Teillier
5242, F; Antofagasta región II, Prov. Antofagasta. M.O. Dillon 5377, F; Constitución, Reiche K s.n., X/1893, SGO; Illapel, Philippi s.n., XII/1862, SGO; Santiago,
Philippi s.n., SGO (66008); Valparaiso, Hutchison P. 22, SGO; Reserva Nacional Rio
Clarillo, Yanez J s.n., 03/VIII/1983, SGO; Coquimbo, Muñoz M. 908, SGO; Hacienda Rinconada, Maipu, Schlegel F. 1664, SGO.
Puya boyacana
COLOMBIA: Dpto. Boyacá, Cordillera Oriental, J. Cuatrecasas & H. Garcı́a B.
Cuatrec.
9756, COL, US.
Puya brachystachya COLOMBIA: Dpto. Magdalena, San Sebastián, Foster & Smith 1458, US.
Mez
Puya cardonae
VENEZUELA: Edo. Táchira, Páramo de Tamá, Cardona 335 VEN.
L.B.Sm.
Puya castellanosii
ARGENTINA: Salta, Laguna del Brealito, Castellanos 45819, US; Prov. Salta,
L.B.Sm.
Catellanos s.n. 1897, US; Dpto. Molinos, Prov. Salta. Brealito, T. Meyer 9164, US;
Cachi, Prov. Salta, G.S. Varadarajan, U. Varadarajan & L. Novara 1476, US.
48
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Appendix. (continued)
Puya chilensis
Molina
Puya coerulea
Miers
Puya cryptantha
Cuatrec.
Puya ctenorhyncha
L.B.Sm.
Puya cylindrica Mez
Puya fastuosa Mez
Puya ferruginea
(Ruiz & Pav.)
L.B.Sm.
Puya floccosa
E. Morr.
Puya glomerifera
Mez & Sodiro
BOLIVIA (CHILE now): Gaudichaud, s.n. F (1370754). CHILE: Limache, Frisco, F
(633928); 1837, F (1435147); Viña del Mar, 7/IX/1922, F; Viña del Mar, 7/IX/1922, F;
Prov. Elqui.T.G. Lammers, C.M. Baeza P. & P. Peñalillo B. 7653, F; Angol, Philippi
s.n., I/1877, SGO; Constitución, Reiche K s.n., XII/1891, SGO; Fundo Fray Jorge,
Carrizo C., s.n., 10/III/1947, SGO; Cordillera de Colchagua, Landbeck L. s.n., XII/
1860, SGO; Cordillera de Colchagua, Philippi s.n., SGO (6425); Cerro, Sur Baños
Flacos, Espinosa M. s.n., XII/1937, SGO; Zapallar, Philippi s.n., IX/1875, SGO;
Aconcagua: Zapallar, Philippi s.n., IX/1865, SGO; Paposo, Reiche K. s.n., IX/1909,
SGO; Cordillera de Cauquenes, Reiche K. s.n., X/1907, SGO; Angol, Philippi s.n.,
1877, SGO; Constitución, Reiche K. s.n., XII/1891, SGO; Chillan, Philippi s.n., XII/
1869, SGO; Llico, Philippi s.n., XII/1861, SGO.
CHILE: Angostura de Praine, G. Looser s.n., 4/XII/1932, F; F (1668607); Angostura
de Praine. G. Looser 2553, F; Angostura de Praine, Looser 2550, F; Angostura de
Praine, G. Looser 2549, F; Angostura de Praine. G. Looser 2551, F; Prov. Santiago,
P.C. Hutchinson 202, F; Phillipii s.n. F (741265); Prov. Curicó: Hacienda Monte
Grande, E. Wedermann 539, F; F (835815); Rı́o Clarillo, HDCV (949); Baños de
Cauquenes, Philippi s.n., III/1875, SGO; Llico, Philippi s.n., XII/1861, SGO; Inter
Poblacion Et Cueva, Philippi s.n., SGO (46407); Chillan, Philippi s.n., XII/1869,
SGO; Cordillera de Popeta, Philippi s.n., I/1884, SGO (46413); Angol, Philippi s.n., I/
1877, SGO; Complejo Turistico La Leonera, Muñoz M. & S. Moreira 2393, SGO;
Complejo Turistico La Leonera, Muñoz M. & S. Moreira 2394, SGO; Las Tacas,
Flores A. s.n., 15/VIII/1986, SGO; Taltal, Hoffmann A. & Flores A. s.n., 03/XII/
1988, SGO; Quebrada Camino a Cifuncho, Hoffmann A. & Flores A. s.n., 01/XII/
1988, SGO.
COLOMBIA: Dpto. Cundinamarca, Cordillera Oriental. J. Cuatrecasas 9568, COL;
Dpto. Cundinamarca, La Calera. E. Forero 216, COL; Dpto. Cundinamarca, N.
Tryon & A. Tryon 6019, COL.
BOLIVIA: Anilaya-Lareca, La Paz, Cárdenas 4867, US.
PERU: J. Solomon 3065, USM, F.
PERU: Dpto. Cajamarca, near Hualgayoc, Weberbauer 4069, F.
PERU: Lima, Prov. Huarochiri. Rı́o Blanco, P.C. Hutchinson 573, F; Prov. Yanachaga, Gentry & D. Smith 35919, F; Cusco, Prov. Paucartambo, R. Foster & T.
Watcher 7482, F.; Prov. Huaura, Lomas de Lachay, A. Cano 7081, USM; Dpto.
Ancash; Prov. Huaylas, A. Cano 8882, USM; Dpto. Lima, Prov. Huarochirı́, E.
Cerrate 1775, USM; Pasco. Cordillera Yanachaga, A. Gentry 35919, USM; Prov.
Huaura, Lomas de Lachay, A. Cano 7081, USM; Prov. Chancay, A. Cano 7081,
USM; Dpto. Cuzco, Prov. Urubamba, E.W. Davis 1774, USM; Dpto. Cuzco, Prov.
Urubamba, W. Davis 1488, USM; Cuzco, Prov. Urubamba, Tupayachi, A. 931, NY.
BOLIVIA: Dpto. La Paz. Prov. Murillo, Solomon, J.C. 6093, NY; Dpto. La Paz.
Prov. Murillo, Nee, M. 34161, NY; Dpto. La Paz. Prov. Murillo, Buchtien, O. s.n.,
NY; Nequejahuira, Tate, G.H. 634, NY.
COLOMBIA: Dpto. Cundinamarca, Cordillera Oriental, J. Betancur, P. Franco, S.
Madriñan & M. Salazar 3975, COL. VENEZUELA: Edo. Mérida, Hornung, C.,
Tirado, C. & Quevedo, J., 206, MERC; Edo. Mérida Hornung, C., Garbiso, C. &
Quevedo, J., 81 MERC.
ECUADOR: Prov. El Chimborazo, M. Acosta Solı́s 7171, F ; Prov. Tungurahua, M.
Acosta Solı́s 9202, F; Prov. Pichincha, E. Asplund 17564, F; Prov. Pichincha, A.
Firmin s.n. 12-I-1928, F.
C. Hornung-Leoni and V. Sosa: Allometry in Puya
49
Appendix. (continued)
Puya goudotiana
Mez
Puya grubbii
L.B.Sm.
Puya hamata
L.B.Sm.
COLOMBIA: Dpto. Cundinamarca, Mpio. La Calera, C. Garcı́a R. 85, COL; Dpto.
Cundinamarca, Páramo de Chocontá, J. Cuatrecasas 9660, COL, F; Dpto. Amazonas, Páramo de Guasca, A. Fernández P. & R. Jaramillo H. 5760, COL; Dpto.
Cundinamarca, Páramo de Guasca, J. Cuatrecasas 13547, F; Dpto. Norte de Santander, Presidente, J. Cuatrecasas & H.Garcı́a 10039, F; Dpto. Cundinamarca, J.
Cuatrecasas 7969, F; Dpto. Cundinamarca, Macizo de Bogotá, J. Cuatrecasas 5162,
F.
COLOMBIA: Boyacá, Sierra Nevada de Cocuy, Grubb, Curry & Fernández 711, US.
COLOMBIA: Cordillera Central, Mpio. Totoro-Inza, J.Betancur & S. Churchill
2459, COL; Dpto. del Valle, J. Cuatrecasas 20578, COL, F; Cauca, Páramo de
Puracé, E.P. Killip & F.C. Lehmann 38587, COL; PERU: F (1641580); Cajamarca,
Prov. Chota, A. Sagástegui, D. Skillman, J. Mostacero & L. Ramirez 12881, F; Prov.
Cajamarca, Dpto. Cajamarca, P.C. Hutchinson 7036, USM; Putumayo, Mpio.
Santiago, Betancur, J. & J.L. Lutyen 7467, NY. ECUADOR: Prov. Cachi, Asplund,
E. 16997, NY; Prov. Cachi, Dorr L.J. & L.C. Barnett 6030, NY.
PERU: Valle de Apurimac, Herrera 1965, F.
Puya herrerae
Harms
Puya hofstenii Mez ARGENTINA: Yavi, Jujuy, Hoften 1710, F; Jujuy, Salinas Grande, Huathal 135, F.
Puya humilis Mez BOLIVIA: Wedermann 2006, F; Sucre, Near Guerraloma. M. Cárdenas 4128, US; S.
Voguel 468, US; 12/VI/1957, US; Prov. Tarata, M. Cárdenas, H. Cutler & H.
Gandarillas 7639, F.
Puya laccata Mez PERU: near Monzon, Weberbauer 3376, F.
Puya lilloi
ARGENTINA: Tucumán, La Hoyada, Schreiter 588, F.
Castellanos
Puya lineata Mez
COLOMBIA: Dpto. Cundinamarca, C. Saravia 02498A, COL; Apolinar, M. 527, F.
Puya longisepala
PERU: Puno, Weberbauer 550, F.
Mez
Puya maculata
ECUADOR: Prov. Azuay, Luther, Kress, Brown & Roesel 2703, F.
L.B.Sm.
Puya mariae
PERU: Amazonas, Wurdack 602, F; Amazonas, Chachapoyas, J. Wurdack 602, NY.
L.B.Sm.
Puya medica
PERU: Shorey, Prov. Santiago de Chuco, Sagástegui, Aldave, Fernandez & FuL.B.Sm.
kushima 6175; XAL.
Puya micrantha
ARGENTINA: Prov. Salta, Dpto. Anta, Parque Nacional El Rey, Brown & MalMez
mierco 1645, F.
Puya minima
BOLIVIA: Tarija, Cuesta de Sama, West, J. 8345, NY.
L.B.Sm.
Puya mirabilis
ARGENTINA: Jujuy, Venturi 1950, US; Tucumán, Venturi 5430, US; Venturi, 7621,
(Mez) L.B.Sm.
US. BOLIVIA: Fiebe 2420a, F; Cochabamba, Quillacollo, Cárdenas 3578, US; Cochabamba, Adolfo 177, US.
Puya nitida Mez
COLOMBIA: Dpto. Cundinamarca, J.Betancur, L.Montenegro & C. Rodrı́guez
2713, COL; Dpto. Cundinamarca, J. Cuatrecasas 9529, F; Macizo de Bogotá, Cerro
de Guadalupe, J. Cuatrecasas 7958, F; Macizo de Bogotá, J. Cuatrecasas 5629, F;
Dpto. Cundinamarca, Páramo de Guasca, J. Cuatrecasas 13539, F; Dpto. Boyacá, A.
M. Cleef 9740, COL; Dpto. Cundinamarca, G. Gutierrez 366, F.
Puya oxyantha Mez PERU: Prov. Sandia, Dpto. Puno, cerca de Sandia, R. Ferreyra & A. Vera 16654,
USM, US; Puno, Weberbauer 1058, F.
Puya parviflora
ECUADOR: El Oro, Espinosa E-2052, US; Loja to Loma, Harling 5724, US; Loja to
L.B.Sm.
Zamorana, Gilmartin 1134, US.
50
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Appendix. (continued)
Puya pratensis
L.B.Sm.
Puya pygmaea
L.B.Sm.
Puya raimondii
Harms
Puya reflexiflora
Mez
Puya riparia
L.B.Sm.
Puya roezlii
E.Morr.
Puya santanderensis
Cuatrec.
Puya santosii
Cuatrec.
PERU: Cajamarca, Cumbre Gavilán, Ferreyra 8580, USM; La Libertad, Huamachuco, López & Sagástegui 2862, US; USM.
BOLIVIA: Lchmamb 589, F. ECUADOR: Azuay, Páramo de Tinajillas, W.H Camp
2236, F, US; Prov. Morona - Santiago, G.S. Varadarajan, U. Varadarajan & V. Zak
1433, US; Prov. Azuay. Páramo de Tinajillas. S.E. Clemants, J.D. Boeke, N.H.
Holmgren & S. Crisafully 2199, US; PERU: Dpto. Cusco, Prov. Paucartambo. A.
Cano 4563, F; Prov. Paucartambo; Depto. Cusco, Tres Cruces P.N.M., A. Cano
3455, USM; Dpto. Cusco; Prov. Paucartambo, Altura de Teleban P.N.M., A. Cano
3782, USM; Dpto. Cusco; Prov. Paucartambo, A. Cano 4563, USM; Dpto. Cusco,
Prov. Paucartambo, Acjanaco. PN Manu, A. Cano 3393, USM.
BOLIVIA: Huacanqui, M. Cárdenas 4380, US; Comanche. M.B. Foster 2566, US;
Cochabamba, M.B. Foster 2546, US; Dpto. Cochabamba, Prov. Arani, G. Schmitt &
D. Schmitt 84, US; La Paz, Prov. Pacajes. Comanche, J.N. Rose & Mrs. Rose 18875,
US; La Paz, Pacajes: Comanche, J. Luteyn, L. Dorr, D. Smith & M.Buddensick
13840, US; Dpto. La Paz, Prov. Pacajes, Comanche, St.G. Beck 2353, US. PERU: La
Libertad, Prov. Otuzco. A. Sagástegui, S. Leiva & C. Tellez 14510, F; Weberbauer
2955, F; La Libertad, Prov. Otuzco. S. Leiva, P. Leiva & E. Zavaleta 292, F; Estation
30 miles of Huaraz. Pomopampa, Macbride & Featherstone s.n, 4/XII/1922, F; Dpto.
Puno, Prov. Melgar, H. Ilties & Don Ugent 1288, US; Prov. Bolognesi. Huishcashpampa, Emma Cerrate 2072, USM; Prov. Huaylas, Dpto. Ancash, A. Cano 6405,
USM; 1/2 km SE of Hacienda Santa Rosa de Achaco, USM (159980); Dpto.
Huancavelica, carretera Castrovirreyna- Ayacucho, USM (159979); Dpto. Ancash,
P.N. Huascarán, Hornung, C. 1118, USM; Dpto. Ancash, Canchayllo, Hornung, C.
1120, USM; Dpto. Ancash, P.N. Huascarán, Hornung, C. 1121, USM; Dpto. Ancash, P.N. Huascarán, Hornung, C. 1122, USM; Dpto. Ancash, P.N. Huascarán,
Pumapashimin, Hornung, C. 1123, USM; Dpto. Ancash, Cordillera Negra, Hornung,
C. 1124, USM; Dpto. Ancash, Hornung, C. 1126, USM.
PERU: P. Hutchison & M. Ricardi 4998, F; Weberbauer 3148, F; Dpto. Ancash;
Prov. Yungay, Llanganuco, R. Ferreyra 16500 A, USM.
BOLIVIA: La Paz, road to Hacienda Chaco, Foster 2578, US.
PERU: Dpto. Apurimac, C. Vargas 9789, F.
COLOMBIA: Dpto. Norte de Santander, páramo de Tamá, L.E.Mora 4641, COL;
Dpto. Norte de Santander, J. Cuatrecasas & N. Garcı́a - Barriga 10035, COL.
COLOMBIA: Dpto. Cundinamarca, Páramo de Cruz Verde, Cuatrecasas 9518, F;
Dpto. Cundinamarca, J. Cuatrecasas 9441, COL; Meta: Páramo de Sumapáz, A.
Cleef 1503, COL; Dpto. Cundinamarca, Páramo de Cruz Verde, J. Cuatrecasas
10468, COL; G. Gutierrez 386, F; Dpto. Cundinamarca, J. Cuatrecasas & R. Jaramillo 12042, F.
Puya trianae Baker COLOMBIA: Dpto. Cundinamarca, Páramo de Chisacá, J. Cuatrecasas & R. Jaramillo M. 25924, COL, F; Dpto. Cundinamarca, Andes de Bogotá, s.n, COL (03042);
Dpto. Cundinamarca, Páramo de Chisacá, T.R. Soderstrom 1286, Dpto. Boyacá,
A.M. Cleef 7430, COL, F.; Dpto. Cundinamarca, J. Cuatrecasas 10474, F; Antioquia,
Betancur, J. Et al 1166, F; Dpto. Cundinamarca, Chisacá, J. Hidrovo, H. Arias & M.
del Llano 6525, F.
Puya tristis
BOLIVIA: Cochabamba, between Cochabamba and Chapare, Cárdenas 6067, US.
L.B.Sm.
C. Hornung-Leoni and V. Sosa: Allometry in Puya
51
Appendix. (continued)
Puya venezuelana
L.B.Sm.
Puya vestita
Ed.Andre
Puya westii
L.B.Sm.
Puya wurdackii
L.B.Sm.
Puya yakespala
Castellanos
COLOMBIA: Dpto. del Arauca, A.M. Cleef 10.124, COL; Lı́nea divisoria entre
Dpto. Santander del Norte y Cesar, H. Garcı́a-Barriga & R. Jaramillo M. 19737,
COL. VENEZUELA: Edo. Mérida, Páramo de Pozo Negro, J. Steyermark 56285, F;
Edo. Mérida, Hornung, C., Tirado, C. & Quevedo, J., 205, MERC; Edo. Mérida,
Hornung, C., Tirado, C. & Quevedo, J., 204, MERC; Edo. Trujillo, L. Aristeguieta
3538, US.
COLOMBIA: Nariño-Putumayo, Páramo del Bordoncillo, A. Muñoz, N. Bernardo,
L. Peñafiel & Javier C. 34, COL; Comisarı́a del Putumayo: alta cuenca del rı́o Putumayo, J. Cuatrecasas 11737, F.
PERU: Dpto. La Libertad, P. Hutchinson, J. K. Wright & R. Straw 6146, F.
PERU: Jalca, between Yanayacu and Pomacocha, Wudarck 1066, US; USM.
ARGENTINA: Salta, Santa Victoria, Reynaga s.n. VII/1940, US.
References
Benzing D. H. (2000) Bromeliaceae profile of an
adaptive radiation. Cambridge University Press,
Cambridge.
Bond W. J., Midgley J. J. (1988) Allometry and
sexual differences in leaf size. Amer. Naturalist
131: 901–910.
Brookstein F. L. (1991) Morphometric tools for
landmark data. Cambridge University Press,
New York, USA.
Chaparro H. A. (2002) Efecto de la predación de
inflorescencias de Puya trianae por moscas
(Phoridae) sobre la producción de frutos en el
parque nacional natural Chingaza (Cundinamarca, Colombia). Resumen. VII Congreso Latinoamericano de Botánica. Cartagena de Indias,
Colombia.
Delph L. F. (1994) Flower size dimorphism in plants
with unisexual flowers. In: Lloyd D. G., Barrett S.
C. H. (eds.) Floral biology: Studies on floral
evolution in animal-pollinated plants. Chapman
and Hall, New York, USA, pp. 217–237.
Donnelly S. E., Lortie C. J., Aarssen L. W. (1998)
Pollination in Verbascum thapsus (Scrophulariaceae): The advantage of being tall. Amer. J. Bot.
85: 1618–1625.
Elle E., Carney R. (2003) Reproductive assurance
varies with flower size in Collinsia parviflora
(Scrophulariaceae). Amer. J. Bot. 90: 888–896.
Faegri K., van der Pijl L. (1971) The principles of
pollination ecology. Pergamon Press, Oxford,
UK.
Foster M. B. (1950) Puya, the pineapple’s Andean
ancestor. Natl. Geogr. Mag. 48: 463–480.
Givnish T. J., Sytsma K. L., Smith J. F., Hahn W.
J., Benzing D. H., Burkhardt E. M. (1997)
Molecular evolution and adaptive radiation in
Brocchinia (Bromeliaceae: Pitcairnioideae) atop
tepuis of the Guayana shield. In: Givnish T. J.,
Systma K. J. (eds.) Molecular evolution and
adaptive radiation. Cambridge University Press,
Cambridge, pp. 259–311.
Grant V., Grant K. A. (1965) Flower pollination in
the Phlox family. Columbia University Press,
New York, USA.
Hair J. F., Anderson R. E., Tatham R. L., Black
W. C. (1998) Multivariate data analysis. Upper
Saddle River, NJ: Prentice Hall, USA.
Harris P. (1985) Testing for variance homogeneity
of correlated variables. Biometrika 72: 103–107.
Hedberg O. (1969) Evolution and speciation in a
tropical high mountain flora. Biol. J. Linn. Soc.
1: 135–148.
Hedberg O. (1970) Evolution of the afroalpine
flora. Biotropica 2: 16–23.
Henderson A. J. (2002) Phenetic and phylogenetic
analyses Reinhardtia (Palmae). Amer. J. Bot. 89:
1491–1503.
Herrera C. (1996) Floral traits and plant adaptations to insect pollinators: a devil’s advocate
approach. In: Lloyd D. G., Barrett S. C. H.
(eds.) Floral biology: studies of floral evolution
in animal-pollinated plants. Chapman and Hall,
New York, pp. 65–87.
52
Hingston A. B., McQuillan P. B. (2000) Are
pollination syndromes useful predictors of floral
visitors in Tasmania? Austral. Ecology 25: 600 –
609.
Hoaglin D. C., Mosteller F., Tukey J. W. (1993)
Understanding robust and exploratory data
analysis. John Wiley & Sons, New York, USA.
Hornung-Leoni C., Sosa V. (2004) Uses of the
giant bromeliad, Puya raimondii. Journal Bromeliad Society 54: 3–8.
Jackson J. E. (1991) A user’s guide to principal
components. Wiley-Interscience, New York,
USA.
Jolicoeur P., Mosimann J. E. (1960) Size and shape
variation in the painted turtle, a principal
component analysis. Growth 24: 339–354.
Kang H., Primack R. B. (1999) Evolutionary
change in seed size among some legume species:
the effects of phylogeny. Pl. Syst. Evol. 219: 151–
164.
Kessler M. (2002) Environmental patterns and
ecological correlates of range size among bromeliad communities of Andean forests in Bolivia. Bot. Rev. 68: 100–127.
Klinbenberg C. P. (1996) Multivariate allometry.
In: Marcus L. F., Corti M., Loy A., Naylor G. J.
P., Slice D. E. (eds.) Advances in morphometrics. Plenum Press, New York, USA.
Klingenberg C. P., Zimmermann M. (1992) Static,
ontogenetic, and evolutionary allometry: A multivariate comparison in nine species of water
striders. Amer. Naturalist 140: 601–620.
Mabberley D. J. (1972) Evolution in the giant
groundsels. Kew Bull. 28: 61–69.
Maitre D. C., Midgley J. J. (1991) Allometric
relationship between leaf and inflorescence mass
in the genus Protea (Proteceaea): An analysis of
the exceptions to the rule. Funct. Ecol. 5: 476–484.
Metcalf J. C., Rose K. E., Rees M. (2003) Evolutionary demography of monocarpic perennials.
Trends Ecol. Evol. 18: 480.
Midgley J., Bond W. (1989) Leaf size and inflorescence size may be allometrically related traits.
Oecologia 78: 427–429.
Miller G. A., Selander J. A. (1991) Control of the
distribution of giant rosette species of Puya
(Bromeliaceae) in the ecuadorian páramos. Biotropica 23: 124–133.
Monasterio M. (1980) Las formaciones vegetales de
los páramos de Venezuela. In: Monasterio M.
(ed.) Estudios ecológicos en los páramos andinos.
C. Hornung-Leoni and V. Sosa: Allometry in Puya
Editorial Universidad de Los Andes, Mérida,
Venezuela.
Niklas K. J. (1994) Plant allometry: The scaling of
forms and process. University of Chicago Press,
USA.
Ornduff R. (1969) Reproductive biology in relation
to systematics. Taxon 18: 121–123.
Ortiz-Crespo F. I. (1973) Field studies in pollination of plants of the genus Puya. J. Brom. Soc.
23: 3–7, 54–58.
Primack R. B. (1987) Relationships among flowers,
fruits, and seeds. Annual Rev. Ecol. Syst. 18:
409–430.
Reyment R., Jöreskog K. C. (1993) Applied factor
analysis in the natural sciences. Cambridge
University Press, New York, USA.
Rivera C. A. (1985) Puya raimondii Harms. Boletı́n
Lima 7: 85–91.
Robertson A. W., Macnair M. R. (1995) The effects
of floral display size on pollinator service to
individual flowers of Myosotis and Mimulus.
Oikos 72: 106–114.
Scogin R. (1988) Floral anthocyanidins of birdvisited flowers. Bot. Gaz. 149: 437–442.
Shea B. T. (1985) Bivariate and multivariate
growth allometry: statistical and biological considerations. J. Zool. 206: 367–390.
Smith L. B., Downs R. W. (1974) Pitcairnioideae
(Bromeliaceae) Fl. Neotrop. Monogr. 14: 1–658.
USA.
SPSS Inc. (1999) SigmaScan Pro. V. 5. 0. USA.
Statsoft Inc. (2000) STATISTICA. Tulsa, Oklahoma, USA.
Strauss R. D. (1985) Evolutionary allometry and
variation in body form in the South American
catfish genus Corydoras (Callichthyidae). Syst.
Zool. 34: 381–396.
Sundberg P. (1989) Shape and size-constrained
principal component analysis. Syst. Zool. 38:
166–168.
Thompson K., Rabinowitz D. (1989) Do big plants
have big seeds. Amer. Naturalist 133: 722–728.
Varadarajan G. S., Brown G. K. (1988) Morphological variation of some floral features of the
subfamily Pitcairnioideae (Bromeliaceae) and
their significance in pollination biology. Bot.
Gaz. 149: 82–91.
Waser M. L. (1983) The adaptive nature of floral
traits: ideas and evidence. In: Real L. (ed.)
Pollination biology. Academic Press, Orlando,
USA, pp. 245–285.
C. Hornung-Leoni and V. Sosa: Allometry in Puya
White P. S. (1983) Corner’s rules in eastern
deciduous trees: allometry and its implications
for the adaptive architecture of trees. Bull.
Torrey Bot. Club 110: 203–212.
Wilson M. F., Price P. W. (1977) The evolution on
inflorescence size in Asclepias (Ascelapiadaceae).
Evolution 31: 495–511.
53
Address of the author’s: C. Hornung-Leoni
(e-mail: hornung@ecologia.edu.mx), V. Sosa (e-mail:
victoria@ecologia.edu.mx), Instituto de Ecologı́a,
A.C., Apartado Postal 63, 91000 Xalapa, Veracruz,
Mexico.
Claudia T, Hornung-Leoni