Population dynamics and management response of N2-fixing indigenous
legumes for soil fertility restoration in smallholder farming systems of
Zimbabwe
By
Tonny Phirilani TAURO
A Thesis submitted in partial fulfillment for the Degree of Master of Philosophy in
Agriculture
Department of Soil Science and Agricultural Engineering
Faculty of Agriculture
University of Zimbabwe
September 2009
Dedication
This work is dedicated to my family and my loved one. Glory be to the Almighty for all
the blessings granted unto me.
ii
Abstract
The main problem undermining food security in most smallholder farming systems in
Zimbabwe is the limited range of alternative nutrient resources for a wide range of farmers
and soils. This study was initiated to investigate field establishment patterns, population
dynamics and management dynamics of potential indigenous legume species found in
smallholder farming systems of Zimbabwe. Germination patterns were determined under
laboratory conditions following various scarification procedures using a germination test
technique while field studies were conducted under low (450-650 mm yr-1) to high (>800
mm yr-1) rainfall conditions in Zimbabwe on nutrient depleted soils. Indigenous legume
seeds were broadcast on disturbed soil in mixtures at 120 seeds m-2 species-1 over two
growing seasons: 2004/05 and 2005/06 rainfall seasons. Low emergence rates among
species were attributed to seed hardness and low seed viability, accounting for > 50 % and
10 - 30 % of germination failure, respectively. Acid scarification significantly (P<0.05)
increased germination while hot water treatment had no effect on germination. Crotalaria
pallida and C. ochroleuca were the first to emerge within 14 days from seeding while
emergence for other species was spread over two months. Eriosema ellipticum had a
highest population of 42 plants m-2 (8 %) while Crotalaria pallida, Crotalaria
cylindrostachys, Crotalaria ochroleuca and Indigofera arrecta had populations of < 23
plants m-2 (5 %). Species biomass contribution into the system was spread across time
and in space characterized by slow initial growth rate compared to that of non-legume
species. Three months after sowing, annual herbs such as C. cylindrostachys and C.
glauca attained peak biomass of 0.5 t ha-1 (dry weight) while other annuals, C. ochroleuca
and C. pallida, attained maximum biomass (> 5 t ha-1) within a growth period of six
months. Most biennials only attained maximum biomass in the second season with
seasonal yields not exceeding 2 t ha-1. Application of single super phosphate (SSP) or
Dorowa phosphate rock (DPR) had no significant (P>0.05) effect on legume species
composition and establishment patterns but SSP significantly increased biomass
productivity by 20-60 %. Non-legume species responded to SSP application and recovered
more phosphorus than legumes. Following addition of SSP, the amount of N fixed by
Crotalaria cylindrostachys, C. ochroleuca and C. pallida increased by 2, 27 and 51 kg N
ha-1 respectively, resulting in increased total N fixed within the system. Over time in the
first season, the abundance of predominant non-legume species such as Cynodon dactylon
(a grass species) and Richardia scabra (a broadleaf species) was suppressed to 17 % and 7
% respectively, while the abundance of the same species in natural fallows increased.
There was insignificant change in legume species diversity in the second season
characterized by persistent growth of about 50 % of the sown species. However, nonlegume species diversity decreased as C. dactylon dominated apparently associated with
increased N availability. These findings suggest that the indigenous legumes could
contribute in restoring productivity of soils continuously cultivated fields with little or no
nutrient inputs in most parts of Zimbabwe and similar agro-ecologies in Sub-Saharan
Africa. However, low germination of these indigenous legumes still remains a challenge.
Acknowledgements
The study was funded by The Rockefeller Foundation through Grant 2004 FS 105 to the
University of Zimbabwe. I would like to thank my supervisors and principal investigors,
Dr. P. Mapfumo and Dr. F. Mtambanengwe for their time, guidance and encouragement
during the study. I am also grateful to the Soil Fertility Consortium for Southern Africa
(SOFECSA) and CIMMYT for their financial, technical and networking support. I would
also like to thank Chemistry and Soil Research Institute for allowing me to have the
chance for career development. I also extend my thanks to my colleague Hatirarami
Nezomba for his help, encouragement and moral support during the whole study.
I wish to thank Dr. D.K.C Dhliwayo, Dr. R. Chikowo and my colleagues at Chemistry and
Soil Research Institute for their assistance and contributions to my study. Many thanks to
Eliah Mbizah, Timothy Mapfumo, Joyce Ushe, James Makuvire, Tongai Mutangadura and
Linda Mtali for technical assistance with laboratory analyses and field work. I also thank
Seed Services and National Herbarium & Botanic Gardens in the Department of
Agricultural Research for Development of the Ministry of Agriculture for their assistance.
Cooperation and participation of farmers in the study areas is highly appreciated.My
special thanks go to my family and friends for believing in me.
iv
TABLE OF CONTENTS
Dedication............................................................................................................................. ii
Abstract................................................................................................................................iii
Acknowledgements ............................................................................................................. iv
TABLE OF CONTENTS ..................................................................................................... v
LIST OF TABLES ............................................................................................................viii
LIST OF FIGURES............................................................................................................. ix
LIST OF APPENDICES ...................................................................................................... x
CHAPTER 1............................................................................................................................ 1
INTRODUCTION .................................................................................................................. 1
1.0 Background..................................................................................................................... 1
1.1 Hypotheses ..................................................................................................................... 5
1.2 Overall objective ............................................................................................................ 6
1.3 Specific objectives.......................................................................................................... 6
CHAPTER 2............................................................................................................................ 7
LITERATURE REVIEW ...................................................................................................... 7
2.0 Soil fertility management in Zimbabwe ......................................................................... 7
2.1 Use of legumes in soil fertility restoration ..................................................................... 9
2.2 Species germination and emergence profiles ............................................................... 10
2.3 Factors affecting nitrogen fixation and legumes growth.............................................. 11
2.4 Biomass productivity of legumes in smallholder farming systems.............................. 13
2.5 The benefits associated with mixed legume fallows .................................................... 14
2.6 Species interaction in mixed stands.............................................................................. 15
2.7 Ecology of legumes and grass species in relation to N economy................................. 16
2.8 Utilization of phosphate rock (PR) as a source of P in cropping systems.................... 17
CHAPTER 3.......................................................................................................................... 18
GENERAL MATERIALS AND METHODS .................................................................... 18
3.0 Introduction .................................................................................................................. 18
3.1.0 Field study sites and farming systems ....................................................................... 18
3.1.1 Initial soil sampling and preparation ..................................................................... 21
3.1.2 Determination of soil texture................................................................................. 21
3.1.3 Determination of soil pH ....................................................................................... 21
3.1.4 Determination of soil organic carbon .................................................................... 22
3.1.5 Determination of exchangeable bases ................................................................... 22
3.1.6 Determination of available soil phosphorus .......................................................... 23
3.1.7 Determination of total soil nitrogen ...................................................................... 23
3.2 Evaluation of indigenous legume seed germination..................................................... 24
3.3 Field determination of species emergence ................................................................... 26
3.4 Species biomass productivity and abundance .............................................................. 26
3.5 Estimation of N2-fixation of indigenous legume species under field conditions ......... 27
3.6 Determination of total plant nitrogen ........................................................................... 27
3.7 Determination of total plant phosphorus ...................................................................... 28
v
CHAPTER 4.......................................................................................................................... 29
ESTABLISHMENT AND BIOMASS ACCUMULATION OF INDIGENOUS
LEGUME SPECIES UNDER INDIFALLOW SYSTEMS ON NUTRIENT
DEPLETED SOILS 1.2 ......................................................................................................... 29
4.1 Abstract......................................................................................................................... 29
4.2 Introduction .................................................................................................................. 30
4.3.0 Materials and Methods .............................................................................................. 32
4.3.1 Field sites............................................................................................................... 32
4.3.2 Seed germination tests ........................................................................................... 32
4.3.3 Determination of species emergence patterns ....................................................... 33
4.3.4 Monitoring of species growth habits ..................................................................... 33
4.3.5 Statistical analysis ................................................................................................. 34
4.4.0 Results ....................................................................................................................... 34
4.4.1 Germination of indigenous legume species........................................................... 34
4.4.2 Species emergence in indifallow systems ............................................................. 36
4.4.3 Species growth habits ............................................................................................ 39
4.5.0 Discussion.................................................................................................................. 44
4.5.1 Species germination and emergence under indifallows ........................................ 44
4.5.2 Species growth habits ............................................................................................ 46
4.6 Conclusions .................................................................................................................. 48
CHAPTER 5.......................................................................................................................... 49
CHANGES IN PLANT SPECIES COMPOSITION AND ABUNDANCE IN
INDIFALLOW SYSTEMS.................................................................................................. 49
5.1 Abstract......................................................................................................................... 49
5.2 Introduction .................................................................................................................. 50
5.3.0 Materials and methods............................................................................................... 52
5.3.1 Determination of species abundance ..................................................................... 52
5.3.2 Monitoring changes in species composition over time ......................................... 52
5.3.4 Statistical analysis ..................................................................................................... 53
5.4. Results ......................................................................................................................... 53
5.4.1 Changes in indigenous legume species composition in indifallows...................... 53
5.4.2 The effect of indigenous legume species on non-legume species composition .... 60
5.5.0 Discussion.................................................................................................................. 62
5.5.1 Species dynamics in an indifallow system ............................................................ 62
5.5.2 Influence of indigenous legume and non-legume species in cropping systems. ... 63
5.6 Conclusions .................................................................................................................. 65
CHAPTER 6.......................................................................................................................... 66
INFLUENCE OF PHOSPHORUS IN INDIFALLOWS MANAGED FOR
RESTORING FERTILITY OF DEGRADED SOILS ON SMALLHOLDER
FARMS1................................................................................................................................. 66
6.1 Abstract......................................................................................................................... 66
6.2 Introduction .................................................................................................................. 67
6.3.0 Materials and methods............................................................................................... 69
6.3.1 Evaluating species response to phosphorus application under indifallow and
natural fallow systems ........................................................................................................ 69
6.3.2 Determining the effect of P application on species abundance and biomass
productivity..................................................................................................................... 70
vi
6.3.3 Determining the effect of P application on BNF................................................... 70
6.3.4 Statistical analysis ................................................................................................. 71
6.4.0 Results ....................................................................................................................... 71
6.4.1 Influence of P fertilization on species composition in indifallows and natural
fallows ................................................................................................................................ 71
6.4.2 Influence of P application on indifallow and natural fallow productivity............. 78
6.4.3 Effects of P application on N2-fixation of indigenous legumes ............................ 82
6.5.0 Discussion.................................................................................................................. 84
6.5.1 Species composition as affected by P application in indifallows.......................... 84
6.5.2 Influence of P on biomass productivity under indifallow system ......................... 85
6.5.3 Importance of P in N2-fixation of indigenous legumes ......................................... 86
6.6 Conclusions .................................................................................................................. 87
CHAPTER 7.......................................................................................................................... 88
OVERALL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS .............. 88
7.1 Introduction .................................................................................................................. 88
7.2 Establishment of indigenous species ............................................................................ 88
7.3 Biomass accumulation of indigenous legume species.................................................. 89
7.4 Population dynamics within indifallow systems .......................................................... 90
7.5 Response of indigenous legumes to P application ....................................................... 91
7.6 Area for further research .............................................................................................. 93
References .......................................................................................................................... 94
vii
LIST OF TABLES
Table 3.1 Physical and chemical characteristics of soils at study sites where indigenous
legumes were established. .............................................................................................. 20
Table 4.1 Germination (%) of indigenous legume species following different scarification
methods under laboratory conditions ............................................................................. 35
Table 4.2. Field emergence (%) of selected indigenous legume species on sandy soils
under different rainfall regions in Zimbabwe................................................................. 37
Table 4.3. Growth habits and patterns of different indigenous legume species studied
under smallholder farm conditions in Zimbabwe........................................................... 41
Table 5.1. Changes in indigenous legume species population over time at Domboshawa,
Zimbabwe. ...................................................................................................................... 56
Table 5.2. Changes in indigenous legume species population over time at Mudange,
Zimbabwe. ...................................................................................................................... 59
Table 5.3. Biomass productivity of indigenous legumes species at 6 months after sowing,
under high and medium rainfall conditions in Zimbabwe during the 2005/06 rainfall
season.............................................................................................................................. 60
Table 6.1. Shannon-Wiener diversity index (H’) and Evenness index (E) for non-legume
species at 3 and 6 months after establishment under indifallow and natural fallow
system with or without P at Domboshawa, Makoholi and Chikwaka in the 2005/6
rainfall season. ................................................................................................................ 76
Table 6.2. Effect of phosphorus application on biomass productivity (t ha-1) of
indigenous legumes at 6 months after establishment under low rainfall (Makoholi)
and high rainfall (Domboshawa), in the 2005-06 season. .............................................. 80
Table 6.3. Total amount of N2-fixed (kg N ha-1) in indifallows plots under different
rainfall conditions at 3 months after establishment in the 2005-06 season. ................... 83
Table 6.4. N2-fixation patterns of different indigenous legume species between 3 and 6
months of growth on poor soils, with and without SSP application under different
rainfall conditions in Zimbabwe..................................................................................... 83
viii
LIST OF FIGURES
Figure 2.1 Changes in the relative importance of N2-fixation at different stages in
vegetation succession (After Gorham et al., 1979). ....................................................... 16
Figure 4.1. Effects of soil disturbance on species emergence of early planted species,
under high rainfall conditions on-station at Domboshawa (a) and on-farm Chikwaka
(b), Zimbabwe during 2006/07 seaso. ............................................................................ 38
Figure 4.2. Biomass accumulation patterns of indigenous legume species over two seasons
under high rainfall conditions at Domboshawa, Zimbabwe during 2004/05 and
2005/06 seasons.............................................................................................................. 42
Figure 4.3. Biomass accumulation patterns of total leguminous and non-leguminous
species over two seasons under high rainfall conditions (>750 mm yr-1), at
Domboshawa, Zimbabwe during 2004/05 and 2005/06 seasons. .................................. 43
Figure 5.1. Relative abundance of legume and non-legume species at 3 months after
establishment of indifallows at Domboshawa (>800 mm yr-1), in the 2005/06 season.. 55
Figure 5.2. Relative abundance of legume and non-legume species at 15 months of a two
year indifallow Domboshawa (annual rainfall = 1400 mm yr-1), in the 2005/06
season.............................................................................................................................. 57
Figure 6.1 Relative abundance of indigenous legume and non-legume species at 6 months
in (a) indifallow and (b) indifallow + SSP systems under low rainfall (450-650 mm
yr-1), Makoholi in the 2005/06 season. ........................................................................... 72
Figure 6.2. Relative abundance of indigenous legume and non-legume species at 3 months
in indifallow + SSP systems under high rainfall (>800 mm yr-1), Domboshawa in the
2005/06 season. .............................................................................................................. 75
Figure 6.3. Biomass accumulation patterns of leguminous and non-leguminous species
biomass under indifallow + SSP over two seasons at Domboshawa (>800 mm yr-1),
Zimbabwe during 2005/06 and 2006/07 seasons. .......................................................... 78
ix
LIST OF APPENDICES
Appendix A. Rainfall data for 2005/06 and 2006/07 season................................................ 106
Appendix B. List of publications from this thesis ................................................................ 108
Appendix C. Statistical analysis outputs .............................................................................. 109
x
CHAPTER 1
INTRODUCTION
1.0 Background
Natural fallows are a common feature of farming systems in the sub-humid and semi-arid
zones of Southern Africa, with over 30% of available land fallowed in any one season
(Mafongoya et al., 1998; Mapfumo and Giller, 2001). Low soil fertility and lack of
fertilizers have been cited as some of the reasons for fallowing (Nyakanda et al., 2002).
Resource-constrained farmers largely depend on natural fallowing for soil fertility
restoration (Mtambanengwe and Mapfumo, 2005). However, the current fallowing cycles
have been revealed to be too short to build the fertility of soil to levels that can sustain
meaningful crop production (Hikwa et al., 1998; Mapfumo and Giller, 2001). Efforts to
improve the quality of fallows through agroforestry tree crops such as Leucaena, Sesbania
and Acacia spp have often been hindered by high establishment costs and lack of
immediate benefits to the farmer (Kwesiga and Coe, 1994; Snapp et al., 1998). There is a
need for continued exploration of technologies that can reduce fallow periods, fallow
establishment costs and at the same time contribute sufficiently to generation of soil
fertility benefits.
In most smallholder farming systems of Sub-Saharan Africa, resource-constrained farmers
depend almost entirely on locally available nutrient resources for soil fertility restoration
(Mapfumo and Giller, 2001). However, the farmers often do not apply the recommended
rates of manure and mineral fertilizers due to a number of reasons (Mugwira, 1985;
Chuma et al., 2000), which reduces the potential benefits associated with relatively large
1
quantities of manure or fertilizers. There is a need to focus on alternative options for
increasing crop production, drawing from indigenous knowledge and resources.
Legumes have been known to make realistic nitrogen (N) addition to the cropping systems
through biological nitrogen fixation (BNF) (Giller et al., 1997; Mapfumo et al., 1999;
Gathumbi et al., 2002). However, most grain legume species such as chickpea (Cicer
arietinum), soybean (Glycine max), faba bean (Vicia faba) and groundnut (Arachis
hypogea), have failed to make significant N contribution to the systems partly due to high
harvest indices (> 70%) and poor biomass productivity (< 2 t ha-1) (Sanchez et al., 1997;
Gilbert, 1998; Mapfumo and Giller, 2001). Furthermore, some N2-fixing legumes have
failed to significantly improve soil fertility of nutrient-depleted sandy soils with < 0.5%
organic carbon and < 5 mg kg-1 available phosphorus partly due to poor stand
establishment (Mapfumo et al., 1999; Chikowo et al., 2004). Kamanga and Shamudzarira
(2001) also reported that at least 60% of the smallholder farmers selectively use their
poorest fields for grain legume production resulting in poor germination and stand
establishment. Thus, it is important to develop and evaluate alternative establishment
options. For example, establishment of mucuna (Mucuna pruriens) on sandy soils in
Zimbabwe has been reported to be highly productive when phosphorus is applied (Hikwa
et al., 1998).
Apart from poor soil fertility, other causes of poor legume establishment include planting
method and depth, quality of seed (seed dormancy and hard seedness), competition from
grass and other plant species and low rhizobial population in soils (Clatworthy and
Thomas, 1972; Grant, 1979; Mpepereki and Makonese, 1998). Successful establishment
of legumes is determined by biomass productivity and amount of N fixed in a given
2
environment (Giller and Wilson, 1991; Gilbert, 1998). In natural ecosystems, N2-fixing
legumes have been revealed to be the most important species in early successional phases
(Gorham et al., 1979) but little is known about such successional patterns in natural grass
fallows (natural fallows). Most studies on legumes have focused on biomass productivity,
N2-fixing capacity and residual fertility benefits of legumes (Hooper and Vitousek, 1998;
Mapfumo, 1999; Chikowo et al., 2004; Mafongoya et al., 2006) with little attention on
population dynamics of the legumes.
Phosphorus (P) deficiency is often a major factor limiting biomass production and N2fixation of legumes in both cropping and natural systems (Giller and Mapfumo, 2002).
There is a paucity of information on the influence of P application on plant species
composition and abundance in managed fallows. The main sources of P in cropping
systems are from weathering of soil minerals, release from soil sorption sites and
mineralization of soil organic matter. However, most soils in sub-Saharan Africa are old
and highly weathered resulting in low available P (Giller, 2001) and most locally available
nutrient resources have insufficient P to meet the requirement for sustainable crop
production.
A strategic option to increasing available P on nutrient depleted soils is through
application of soluble inorganic phosphate fertilizers. However, the high cost of soluble
inorganic phosphate fertilizers and their inaccessibility has generated considerable interest
in the utilization of phosphate rock (PR) (Akande et al., 1998; Nandi and Haque, 1998;
Dhliwayo, 1999). Legumes such as lupin (Lupinus angustifolius) have been shown to
increase the dissolution and utilization of PR compared with non-legumes mainly due to
rhizosphere acidification (Kamh et al., 1999; Horst et al., 2001).
3
After the advocacy for ecological approaches to agriculture (Giller and Cadisch, 1995;
Mafongoya et al., 2006), most studies targeted agroforestry tree species for soil fertility
management. Despite some of these agroforestry species having great potential, most
resource-constrained farmers often failed to adopt them partly due to the high costs
involved in the establishment process. Some of the establishment costs include purchase
of seed, raising seedlings and meeting labour demands. Most of the agroforestry tree
species were also non-coppicing leading to reduced additional organic inputs derived each
year from regrowth and an associated decline in maize yields after the second year of
cropping (Mafongoya et al., 2006).
Continued establishment failure of single species fallows, mainly attributed to adverse
weather conditions (drought, water logging) and low fertility justifies the need for mixed
species fallows. The advantages of mixed-stand fallows include increased biodiversity,
sustainability of the fallow system, insurance against failure, production of multiple
products (fire wood or stakes for climbing crops and light construction wood), improved
utilization of available growth resources and reduction of pest pressure (Gathumbi et al.,
2004). There is a need for continued exploration of legume-based technologies that are
appropriate for all the ranges of farmers and soils.
Over 36 different indigenous legumes species mainly of the genera Crotalaria, Indigofera,
Rothia and Tephrosia have been identified across different agro-regions in Zimbabwe
(Mapfumo et al., 2005). The capacity of these indigenous legumes to establish and grow
on nutrient-depleted soils where maize was failing to yield any grain, and where most
cultivated annual legumes are usually constrained, is indicative of their potential for
4
fertility management. This led to the introduction of indigenous fallows (indifallows)
concept
which involve harnessing self-generating indigenous herbaceous legumes in
dense stands in order to improve the N economy of natural fallows (Mapfumo et al.,
2005). Building an understanding on the indigenous legume species might be a starting
point in the utilisation of a possible nutrient resources existing in our agro-ecosystem for
most resource-poor farmers in Southern Africa. However, these indigenous legumes
remain unexploited with very little information available regarding their propagation,
establishment pattern, population dynamics and management responses. This study
investigated the growth performance of indigenous legume species on nutrient depleted
soils under smallholder farmer management conditions.
1.1 Hypotheses
The hypotheses that guided this research were:
1. Mixing indigenous legume species of different growth habits ensures successful
establishment of indifallow systems that accumulate more biomass to influence soil
fertility than natural fallows.
2. Indigenous legume species established in mixtures under one year indifallows are able
to persist into the second year without change of species composition or influence on
the growth of non-leguminous companion species.
3. Phosphorus application on nutrient depleted soils influences persistence, N2-fixation
and biomass productivity of indigenous legume species in indifallow systems.
5
1.2 Overall objective
To investigate the establishment patterns, population dynamics and management response
of N2-fixing indigenous legume species for restoration of nutrient depleted soils on
smallholder farms.
1.3 Specific objectives
The study had the following specific objectives:
1. To establish germination profiles, emergence and establishment patterns of indigenous
legumes under field and laboratory conditions.
2. To determine biomass accumulation patterns of different indigenous legumes and nonleguminous species grown in mixtures under indifallow systems in comparison with
natural fallows.
3. To study changes in composition and abundance of indigenous legumes and nonleguminous species grown in indifallow systems over two years in relation to natural
fallows under smallholder farm conditions.
4. To evaluate the effect of P application and source on persistence, N2-fixation and
biomass productivity of indigenous legume species under indifallow systems.
6
CHAPTER 2
LITERATURE REVIEW
2.0 Soil fertility management in Zimbabwe
Over 70% of soils in agro-ecosystems of Zimbabwe have severe fertility constraints,
generally yielding less than 1t ha-1 of maize grain (Zea mays L.) (Grant, 1981; Mapfumo
et al., 2005). Nitrogen and phosphorus are the most limiting nutrients in the predominantly
granite derived soils of Zimbabwe (Nyamapfene, 1991). Efforts to increase the N and P
availability in these soils using locally available nutrient resources is limited by low
quality materials, leading to low supply of N and P to crop production (Giller et al., 1997).
For example, miombo litter has been shown to immobilize N instead of supplying N
immediately to the crops (Mtambanengwe and Kirchmann, 1995; Mafongoya and Nair,
1997). Mineral fertilizers are accepted as a way to supply N and P to increase productivity
(Mapfumo and Giller, 2001). However, fertilizers are being applied at suboptimal rates
due to high cost and unavailability in smallholder farming areas. For most smallholders,
fertilizer use is as low as 5 -30 kg ha-1 yr-1 (Gerner and Harris, 1993; Chuma et al., 2000).
Many of the farm land in smallholder farming systems are often abandoned by farmers
due to lack of appropriate technologies to revitalize them (Mapfumo and Giller, 2001). A
wide range of locally derived fertilizers (animal manure, leaf litter, termitaria) and crop
residues have been used by farmers but without much impact usually due to inaccessibility
and unavailability (Carter and Murwira, 1995; Campbell et al., 1998). Most crop residues
are usually fed to livestock in the dry season or burnt during land preparations resulting in
limited use of crop residues in soil fertility management. Despite grass fallows being
7
unproductive, still most resource-constrained farmers desperately resort to natural
fallowing as a soil restorative practice.
Studies have revealed strong relationships between farmer resource endowment and
adoption of soil fertility management options. For example, cattle manure can be the
cheaper source of organic fertilizer but is a priviledge (accessible and available) for cattle
owners. Manure application improves soil fertility through supply of nutrients (calcium
(Ca), magnesium (Mg), nitrogen (N) and phosphorus (P)), improvement of soil structure,
increasing water holding capacity and neutralizes soil acidity (Nzuma et al., 1998).
However, most farmers can not apply the recommended 10 t ha-1 of manure (Mugwira,
1985) thereby reducing the potential benefits associated with manure for soil fertility
improvement.Annual woodland litter fall may be as high as 5 t ha-1 but the total amount of
leaf litter incorporated in the field is often < 1.2 t household-1 (Nyathi and Campbell,
1993) due to lack of labour and draft power needed for collection and transportation,
respectively. Most communal areas are also in severe state of deforestation which further
reduces the availability of leaf litter to households. Although termitarium soil enhances
some soil physical properties and benefits can last for 4 - 5 years (Chibudu et al., 1995;
Mapfumo et al., 1997), the high labour associated with digging and application of
termitarium soil is a major limitation to resource-constrained farmers. The current
limitations of most locally available resources suggest that there is need to identify
alternative organic nutrient sources for smallholder conditions. Recently, over 36 different
indigenous legume species have been identified in Zimbabwe but still remain untapped
(Mapfumo et al., 2005). These indigenous legumes were found to grow on abandoned
soils on which other legumes failed to make impact and maize crops failed to yield any
grain.
8
2.1 Use of legumes in soil fertility restoration
In most smallholder farming systems of Zimbabwe, systematic grain legume-cereal
rotations are not feasible due to the difference in areas under legumes and cereals. Studies
have demonstrated that legumes in fallows improve nutrient cycling through litter fall and
retrieval of subsoil nitrogen and biological N2-fixation (Palm, 1995; Gathumbi et al.,
2004; Mafongoya et al., 2006). Improved fallows with Sesbania sesban and Tephrosia
vogelii have been shown to give maize grain yields of 3 to 4 t ha-1 without supply of
mineral fertilizers (Dzowela, 1998). Despite the potential of improved fallows, adoption in
most smallholder farming areas remains low and insignificant due to a number of reasons.
The high labour demands required in cutting, transporting and managing prunings make
improved fallows unpopular with farmers (Jama et al., 1998). Other successful fallow
species such as Sesbania sesban and Crotalaria grahamiania have increasingly become
susceptible to pest and diseases or led to outbreaks of new pest and disease (Cadisch et al.,
2002).
Despite green manures increasing maize grain yields following incorporation (Marura et
al., 2000) adoption still remains low particularly under unimodal rainfall pattern compared
to bimodal rainfall pattern. Apart from the possible N losses, seed availability and labour
requirements for land preparation and ploughing-under of the green manure become major
constraints for most farmers (Muza, 2003). Most soils cultivated by smallholder farmers
are exhausted and production of legumes is constrained by poor availability of nutrients,
especially P (Giller, 2001). Hikwa et al. (1998) showed that rehabilitation of such soils by
Mucuna pruriens was only possible when P fertilizers were supplied, and in some cases P
alone was not enough to stimulate good growth (Zingore et al., 2007). Given that
9
indigenous legume seed can easily be collected by farmers and that these species are
adapted to low P soil the cost associated with seed and P fertilizers might be reduced.
2.2 Species germination and emergence profiles
The necessary activities for a seed to complete the process of germination include water
imbibition, germination, radicle entry to the soil and commencement of root growth
(Kusekwa and Lwoga, 1985). This is a simple process in a cultivated land where the soil is
moist. However, limited soil-seed contacts, light condition on the surface and water
availability are some environmental conditions that may limit germination (Ghorbani et
al., 1999). For surface sown seeds the relative humidity of the micro-environment
surrounding the seed is important. Germination of some seed also requires light but is
normally inhibited by vegetation-filtered (far-red) light and enhanced by fluctuations
temperature (Alteri and Liebman, 1988). Seed germination at greater sowing depth in the
soil may also be prevented by inadequate oxygen, lack of light and low temperature, while
delayed emergence is due to the larger distance to extend the coleoptile to the soil surface.
If seed is placed just below the surface and receive adequate water, it would emerge using
the limited carbohydrate reserves (Ghorbani et al., 1999). Germination and growth
amongst relatively small-seeded species such as Antyliis, Hippocrepis and Cistus may be
improved by supplying external source of N, P and K because their internal nutrient
reserves are low (Fenner and Lee, 1989). Addition of some fertilizers is often essential at
sowing as the seedlings have a high requirement for nutrients but have insufficient root
growth to compete effectively with the associated grasses (Giller and Wilson, 1991).
Phosphorus is the most commonly required nutrient at early root development. However,
larger seeded species such as the pine tree (Pinus brutta) contain enough essential
10
minerals to supply the seedling through out the early establishment period. These larger
seeded species can also emerge from greater depth.
While emergence profiles may vary from year to year depending upon environmental
conditions, the emergence patterns of different species remain relatively consistent in
relation to each other (Hartzler, 2000). The size of food reserve in the seed (seed size)
provides information on both the length of emergence and distribution of emergence.
Some species germinate readily soon after seeding while others germinate a little bit later
or do not germinate at all due to difference in genetic makeup. For instance, under
laboratory and glasshouse conditions indigenous legume species generally had low
germination. Studies by Venge (2003) showed that Eriosema ellipticum, a perennial
legume in Zimbabwe had the highest while some species such as Rothia hirsuta and
Tephrosia radicans did not germinate at all. Some leguminous species have hard seeded
coats, which need to be scarified for them to germinate or to increase germination. This
study might provide more information on the factors affecting establishment and
populations under natural environment and managed conditions.
2.3 Factors affecting nitrogen fixation and legumes growth
Soil nutrient availability is a primary factor driving nitrogen fixation and legume
establishment (Giller, 2001; O’Hara, 2001). Legume-rhizobium symbiosis is often limited
by phosphorus, molybdenum (Mo), sulphur (S) and zinc (Zn) deficiencies in soils
deficiencies. Calcium is important for root infection and nodule formation in legume BNF
while cobalt is required by N2 fixing bacteria for electron transport. Among the potential
nutritional constraints to legume growth, availability of P is often overriding in both
agricultural and natural ecosystems (Giller and Wilson, 1991). Phosphorus deficiency
11
appears to be the most limiting nutrient to legume biomass productivity and N2-fixation
leading to low yield of subsequent cereal crops (Giller, 2001). Phosphate fertilizer
application is necessary to support the luxurious growth of green manures. For example,
application of 8 kg P ha-1 + 4 kg S ha-1 was shown to improve the biomass of Mucuna
pruriens, Cajanus cajan and Tephrosia vogelii in Malawi (Malwanda et al., 2002).
In soil with low N levels, symbiosis is initiated between the legume roots and the rhizobia,
causing nodulation and nitrogen fixation to occur. However, in systems dominated by
legumes, such as fodder banks, rates of N2 fixation decline as the system accumulate N in
the soil (Vance, 2001). According to Munns (1977) high plant available N in the soil
inhibits nodule development, nitrogenase activity and accelerates nodule senescence in the
short term. Over long periods, this high available N would promote plant growth which
would then deplete the soil mineral N resulting in renodulation of the legumes (Dart,
1977). Toxic levels of aluminum (Al), manganese (Mn) and other heavy metals have been
reported to interfere with the process of nodule formation in legumes and survival of N2fixing bacteria in acid soils (Giller 2001; Metcalf, 2005). When Al concentration in the
soil is high, dry matter production of tropical pastures dependent on symbiotically fixed N
was shown to decrease (Carvalho et al., 1981). However, application of 4 t ha-1 of millet
straw to a groundnut crop reduced the imbalance caused by soil acidity (Al and Mn
toxicity) coupled with increased availability of molybdenum (Mo) resulting in the
doubling of accumulated N (Rebafka et al., 1993).
Soil moisture also affects legume performance with both water-logging and drought
conditions reducing legume growth (Muza and Mapfumo, 1998). Inadequate rainfall has
been reported to affect legume biomass productivity of Tephrosia vogelii (Rutunga et al.,
12
1999). Low soil water can reduce nodulation by as much as 50 % and continued water
stress can irreversibly damage nodules and bacteria, permanently stopping N2-fixation
(Abdel-Wahab et al., 2002). Low or high temperature also delays the establishment and
functioning of the symbioses.
2.4 Biomass productivity of legumes in smallholder farming systems
Most legume-based soil fertility improving technologies have failed to make the desired
impact on nutrient-depleted soils partly due to low N2-fixation rates and poor biomass
productivity (Mapfumo, 2000; Nhamo et al., 2002). The impact of legumes on soil fertility
restoration depends largely on the amount of biomass the legumes produce. The minimum
legume biomass required to make a significant impact on soil fertility is above 2 t ha-1
which generates about 60 kg N ha-1 (Gilbert, 1998). In Western Kenya, Ojiem et al (1998)
reported dry matter accumulation of up to 9 t ha-1 for all green manure legumes
(Crotalaria ochroleuca, Crotalaria grahamiana, Crotalaria incana and Mucuna) while
soybean accumulated dry matter of about 2 t ha-1. Kamanga and Shamudzarira (2001)
recorded dry matter yield of < 1 t ha-1 for C. grahamiana, Vigna subterranea and Vigna
unguiculata in Zimuto Communal Area in Zimbabwe. Such poor biomass production is a
major factor undermining the capacity of grain legumes to improve the fertility of sandy
soils in Zimbabwe (Mapfumo et al., 1999). Low soil fertility on abandoned fields
contributed to poor legume performance compared to homestead fields which are
relatively fertile due to a history of manure and household waste applications (Kamanga
and Shamudzarira, 2001).
According to Giller (2003) an initial application of 20-30 kg P ha-1 is critical in biomass
productivity in most soils. Application of plant residues or manure will provide some P
13
but mineral P fertilizers are the most effective. However, legumes differ in their ability to
utilize available forms of P in the soil and large differences exist between plant species
and cultivars within species (Rao et al., 1999). Species growth habits can also influence
the biomass contribution in a system despite the soil being fertile. For example Ojiem et al
(2000) noted that Calopogonium mucunoides and Mucuna pruriens produced low biomass
due to their low initial growth rates, which increase substantially 2-3 months after
planting. This study might bring back abandoned field into production by managing
indigenous legume species which are adapted to nutrient depleted soils where other
legumes often fail to yield significant biomass.
2.5 The benefits associated with mixed legume fallows
Developing integrated pest management through mixing species has been considered to be
an option for increasing benefits associated with most legume based soil fertility
technologies (Sileshi et al., 2002). Furthermore, mixing species in fallows provides a
better risk management strategy through compensatory biomass and nutrient production
gains obtained from the strongly competing species (Gathumbi et al., 2004). For example,
indigenous legumes in sole stands would contribute about 3-17 % of the total biomass
translating to < 5 kg N ha-1, while mixing species increases contribution up to 40 % of
total biomass which translates to potential N inputs of 6-53 kg N ha-1 (Mapfumo et al.,
2005). Mixing species also enhances the uptake of soil water and nutrients within the soil
profiles due to differences in rooting depth and ability to exploit resources (Farley and
Fitter, 1999; Nordin et al., 2001). Mulder et al (2002) showed that the percent legume in
plots was the single most important factor-controlling %N, total N and δ
15
N. Increased
diversity could lead to greater niche complementarity, influencing the N release patterns in
the system as a result of different quality of biomass, and extents the time of residual
14
effect. Mixed fallows of sprouting species such as Gliricidia sepium and a non-sprouting
species S. sesban significantly increased grain yield to 4.8 t ha-1 from single species yield
of 3.7 t ha-1 and 4.3 t ha-1, respectively (Mafongoya et al., 2006). Species richness could
result in greater microbial diversity and higher decomposition rates, although there is no
empirical data to support this hypothesis. The indifallow concept involves using a mixture
of indigenous legume species for soil fertility management which might reduce the risk of
single species.
2.6 Species interaction in mixed stands
The effects of species interaction in mixed stands can either be negative or positive
depending on the competitive utilization of available resources (above and below ground)
in relation to the companion species. For example, the accumulated N yield of Sesbania
was decreased by 54 % and 67 % when mixed with pigeon pea and crotalaria, respectively
(Gathumbi et al., 2004). This was attributed to poor initial establishment of sesbania and
in about 60 % reduction in biomass productivity due to interspecific competition from the
surounding crotalaria or pigeon pea plants which were stronger competitors in the
mixtures. Plantation studies conducted in the humid tropics of Costa Rica also showed that
biomass of Genipa Americana and Vochysia ferruginea trees were reduced by 55-60 % in
mixture than in pure stands. In contrast, Hieronyma alchorneoides and Pithecellobium
elegans trees grew 40-50 % more rapidly in mixture (Stanley and Montagnini, 1999).
Species interaction in mixtures can be greatly influenced by initial establishment of
species, as demonstrated by the low biomass and N yield observed for S. sesban in mixed
species fallows. Poorly established species in mixtures compete less vigorously than fast
growing species and this greatly influences the final benefit of the fallow (Gathumbi et al.,
2004).
15
2.7 Ecology of legumes and grass species in relation to N economy
In natural ecosystems, N2-fixation is more important in early successional phases unless
other nutrients are strongly limiting, but becomes relatively unimportant as the amount of
N in the soil gradually accumulates (Figure 2.1). However, at later stages of primary or
secondary succession when the system becomes more abundant in C and limited in N,
there would be a resurge of N2-fixation (Gorham et al., 1979). Agricultural systems are
also influenced in the same way, and rate of N2 fixation are largest when N is limiting but
other nutrients plentiful.
N fixation rate
Primary succession
Secondary succession
Successional time
Figure 2.1 Changes in the relative importance of N2-fixation at different stages in
vegetation succession (After Gorham et al., 1979).
Persistence of a species refers to its ability to repeatedly invade an environment even
when it was apparently removed from the scene by man (or any other agent). Both
legumes and non-legume species have different mechanisms of persistence. For example
high persistence of weeds is attributed to their prolific seed production, dormancy of seed,
vegetative propagation and rapid dispersal. The difference in rooting systems of species
can also affect persistence of species in a system (Oberson et al., 2001). Currently there is
16
limited information on the mechanisms that promote indigenous legume persistence in
fallow systems and such information is required to fully utilize these legumes for soil
fertility management.
2.8 Utilization of phosphate rock (PR) as a source of P in cropping systems
The use of local resources such as Dorowa phosphate rock may be a feasible alternative to
satisfy the farmers’ needs, both agronomically and economically. However, direct use of
Dorowa phosphate rock (DPR) as a P source is limited due to low solubility in neutral and
alkaline conditions (Dhliwayo, 1999). Ground phosphate rock has been reported to be
beneficial to crops only in acidic soils (Nnadi and Haque, 1988). Current efforts have been
directed at composting of phosphate rock with agricultural waste, and results have shown
an increase in solubility of phosphate rock (Akande et al, 1998; Akande et al., 2005).
Akande et al., (2003) reported an increase in okra (Abelmoschus esculentusus) yield
grown under combination of phosphate rock and poultry manure. However, availability
and accessibility of poultry manure is dependent on the number of birds on the household
and can be the major limitation to adoption.
Legumes have also been shown to increase the dissolution and utilization of phosphate
rock compared with non-legumes mainly due to rhizosphere processes. The combined
effect of rhizosphere acidification and higher Ca uptake by lupin (Lupinus angustifolius)
increased the release of P from PR (Kamh et al., 1999; Horst et al., 2001). There is limited
use of Dorowa phosphate rock in Zimbabwe due to its ingenous nature and low solubility
despite the vast deposit of the rock. This study seeks to generate information on the role of
indigenous
legume
species
in
influencing
17
soil
P
availability
from
PR.
CHAPTER 3
GENERAL MATERIALS AND METHODS
3.0 Introduction
This chapter gives details of the common methods used in the various parts of this thesis.
The study was conducted in two concurrent sets of experiments: (a) laboratory
germination tests, and (b) field experiments. The germination tests were conducted at the
Seed Services in the Department of Agricultural Research for Development of the
Ministry of Agriculture.
3.1.0 Field study sites and farming systems
The study was conducted in 2005/06 and 2006/07 rainfall seasons under smallholder
farming conditions at Chinyika Resettlement Area (18013’S, 32022’E) and Chikwaka
Communal Area (17044’S, 31029’E). Parallel studies were also conducted on-station at
Domboshawa Training Centre (Domboshawa) (17035’S, 31014’E) and Makoholi
Experimental Station (Makoholi) (19047’S, 30045’E). Domboshawa and Chikwaka which
are 30 km and 80 km northeast of Harare respectively are in Agro-ecological Region (NR)
II of Zimbabwe which receives over 800 mm of rainfall annually. Chinyika is 250 km east
of Harare in NR III (650-750 mm yr-1) and has been under smallholder cultivation since
1983 when farmers were resettled there. The area had been under commercial croplivestock farming before then. In Chikwaka, experiments were hosted by farmer
Nyamayaro while in Chinyika three farmers hosted the experiments namely Mudange,
Chikodzo and Masara. Makoholi experiment station is about 300 km southeast of Harare
18
and in NR IV (450-650 mm of rainfall per annum). Rainfall at all sites follows a unimodal
pattern with rains received between November and March.
The soils at all sites are granite-derived sands to loamy sands that are classified as Lixisol
under the World Reference Base system (World Reference Base, 1998). The soils are
largely depleted of nutrients (Table 3.1). The soils are acidic with pH 4.1 – 4.6 (CaCl2),
low organic carbon (C) = 0.4 – 0.5 %, low available P = 3 – 5 ppm, exchangeable Ca = 0.3
– 0.7 cmolc kg-1, exchangeable Mg = 0.3 – 0.7 cmolc kg-1, exchangeable K = < 0.2 cmolc
kg-1 and clay < 8 % (Table 3.1). The soils at Domboshawa and at Chikodzo farm in
Chinyika were relatively fertile with available P = 9 - 15 ppm, 22 % and 10 % clay
content, respectively (Table 3.1). Apart from Domboshawa and Chikodzo soils, most of
the soils had < 0.05 ppm total N (Table 3.1). There is a relationship between high organic
C and the mineralizable N in these soils. For example, Domboshawa soil produced the
highest mineralizable N (35 mg ka-1) form 0.7 % organic C while Makoholi only manage
to give 17 mg kg-1 mineral N from 0.4 % organic C (Table 3.1).
The farming systems in all the areas are maize-based, characterized by crop-livestock
interactions and with low legume utilization. Groundnut (Arachis hypogaea), bambara
(Vigna subterranea) and cowpea (Vigna unguiculata) are most widely grown as sole
legumes but often on small pieces of land (e.g. < 0.3 ha) (Mapfumo et al., 2005). Cattle
manure, leaf litter, mineral fertilizers and fallowing are the predominant methods used for
soil fertility management in such areas (Mapfumo and Giller, 2001.
19
Table 3.1 Physical and chemical characteristics of soils at study sites where indigenous legumes were established.
Site
Clay
Sand
Organic C
Total N
Available P
Mineral N
pH
(%)
(%)
(%)
(%)
(ppm) a
(mg kg-1) *
(0.01 M CaCl2)
Domboshawa 22
72
0.7 (0.08)
0.06 (0.003)
15 (0.3)
4.8 (0.15)
4.8 (0.15)
0.8 (0.06)
0.4 (0.01)
0.1 (0.005)
Makoholi
8
81
0.4 (0.02)
0.03 (0.005)
3 (0.5)
4.6 (0.13)
4.6 (0.13)
0.3 (0.02)
0.3 (0.02)
0.1 (0.004)
Chikwaka
8
79
0.4 (0.06)
0.04 (0.006)
4 (0.1)
4.5 (0.11)
4.5 (0.11)
0.4 (0.01)
0.5 (0.01)
0.2 (0.001)
Mudange**
7
89
0.5 (0.05)
0.05 (0.005)
5 (0.4)
4.1 (0.05)
4.1 (0.05)
0.7 (0.01)
0.5 (0.02)
0.2 (0.005)
Chikodzo**
10
86
0.6 (0.04)
0.06 (0.002)
9 (0.2)
4.5 (0.12)
4.5 (0.12)
1.0 (0.02)
0.8 (0.04)
0.3 (0.004)
Masara**
8
75
0.4 (0.05)
0.04 (0.007)
4 (0.6)
4.8 (0.13)
4.8 (0.13)
0.6 (0.03)
0.7 (0.03)
0.1 (0.003)
Chikwaka = Nyamayaro field; ** = fields in Chinyika
*
Mineralizable N after two weeks of anaerobic incubation.
a
Available P measured using the Olsen method.
Figures in parentheses denote standard error.
20
Ca
Mg
K
(cmolc kg-1)
3.1.1 Initial soil sampling and preparation
Soils were sampled from respective field sites by collecting a total of 5 sub-samples from
0-20 cm depth using an auger. The sub-samples were thoroughly mixed in a clean
polystyrene bucket, after which a 1 kg composite sample was withdrawn and put into a
polythene bag for laboratory analysis. The air-dried soil samples were characterized for
texture, pH, organic carbon (OC), bases, Cation Exchange Capacity (CEC), P and N.
3.1.2 Determination of soil texture
Soil texture was measured using the Hydrometer method (Gee and Bauder, 1986). Airdried soil (50 g) was weighed into a 600 ml beaker and 100 ml of sodium
hexametaphosphate (calgon) added. The beaker was filled to the 500 ml mark with
distilled water and the mixture was put on an automatic shaker overnight, and then left to
stand for at least 10 minutes. The mixture was transferred into a dispersing cup of an
electric mixer and stirred for 5 minutes. Using a plunger, the suspension was thoroughly
mixed by moving it up and down for 1 minute. The hydrometer was gently placed into the
cylinder, and its reading noted at 40 seconds and 5 hours after plunging. The temperature
of the solution was also noted. The hydrometer and temperature readings were then used
to calculate the corrected hydrometer.
3.1.3 Determination of soil pH
Soil pH was determined using the CaCl2 method (Thomas, 1996). To a soil sample
weighing 10 g, 25 ml of 0.01M CaCl2 solution was added. The mixture was shaken for 30
minutes. Prior to taking the pH reading from the pH meter the electrode was thoroughly
rinsed free of buffer solution. The suspension was stirred up and the electrode immersed
21
into the suspension making sure it did not touch the base of the beaker. When the pH
reading was stable, the displayed value was recorded (McNeal, 1982).
3.1.4 Determination of soil organic carbon
A modified Walkley-Black dry combustion method was used to determine the percentage
of organic carbon in the soil samples (Anderson and Ingram, 1983). About 15 g sucrose
was dried at 105 oC for 2 hours and 11.87 g of this were dissolved in water. The solution
was made up to 100 ml in a volumetric flask, to make a 50 mg C ml-1 solution. Two and
half to 25 ml of the solution were pipetted into labeled 100 ml volumetric flasks, at 7.5 ml
intervals, and made to the mark with deionized water. Two milliliters of each of these
working standards were pipetted into labeled 150 ml conical flask and dried at 105 oC in
an oven.
One gram of air-dried soil sample was weighed into a 150 ml flask to which 10 ml of 1N
K2Cr2O7 was added and the contents were gently swirled until the sample was completely
wet. This was followed by the addition of 20 ml of concentrated H2SO4 with further
swirling to ensure through mixing of the solution in a fume cupboard. After cooling the
solution, 50 ml of 0.4 % BaCl2 was added. The solution was left to stand over night and
the absorbance of the samples and the standards were read at 600 nm on UV visible
spectrophotometer (Spectronic-20-Bausch and Lamb).
3.1.5 Determination of exchangeable bases
Total exchangeable bases were determined following extraction of soil by 1M acidified
ammonium acetate (Anderson and Ingram, 1993). Air-dried soil samples weighing 10 g
were put into 150 ml plastic containers to which 40 ml of ammonium acetate were added.
22
The containers were tightly closed and put on a reciprocating shaker (Stuart Flask model)
for 1 hour. The solution was then filtered through a Whatman number 125 filter paper into
250 ml flasks. The remaining soil was washed using ammonium acetate making the
volume in the collecting flask to about 90 ml. Fresh ammonium acetate was used to make
the flask to the mark. After leaching the bases, potassium (K) and sodium (Na) emission
were read on a flame spectrophotometer with the filter set on either K or Na while
magnesium (Mg) and calcium (Ca) absorption were read on an atomic absorption
spectrophotometer (AAS) (SpectrAA50 model) using the respective lamps (Thomas,
1982). The concentrations of the bases were then calculated from a standard curve.
3.1.6 Determination of available soil phosphorus
Available soil P was measured using the Olsen method (Olsen et al., 1954). An air-dried
soil sample weighing 2.5 g was placed into a 250 ml polythene shaking bottle followed by
addition of Olsen’s extracting solution (0.5 M NaHCO3 at pH 8.5). The mixture was
shaken on a mechanical shaker for 30 minutes. The resultant suspension was filtered
through Whatman No. 42 filter paper. To the filtrate, 5 ml of 0.8 M boric acid and 10 ml
of ascorbic acid reagent were added and allowed to stand for 1 hr. The phosphorus content
of the sample was then determined colorimetrically from a phosphorusmolybdate complex
formed by addition of acidified ammonium molybdate (Okalebo et al., 2002). The
absorbance of sample and standards were read at 880 nm wavelength.
3.1.7 Determination of total soil nitrogen
Total N was determined using the semi-Kjeldhal method as described by Anderson and
Ingram (1983). Soil samples were digested using a digestion solution made up of selenium
powder (being the catalysts), lithium sulphate (to raise the boiling temperature) and in
23
concentrated sulphuric acid. An air-dried soil sample weighing 0.5 g (0.1 g for ground
plant material) was weighed into a digestion tube followed by addition of 4.4 ml of
digestion mixture. The resultant mixture was placed on a digester at 3600C for 2 hours.
Three reagent blanks were included to each batch of the samples. Afterwards, the solution
was allowed to cool and 25 ml of distilled water added. A further 75 ml of distilled water
was added and the solution allowed to settle.
Standards were prepared by oven drying 7g of ammonium sulphate ((NH4)2SO4) at 105 oC
for 2 hours. About 4.714 g of dry ammonium sulphate was dissolved in 1000 ml of
deionized water (1000 mg N/litre) and 50 ml of the solution was pipetted into a 500 ml
volumetric flask. Zero, 2.5, 5.0, 7.5, 10, 15, 20 to 25 ml of the 50 ml solution in
volumetric flask was poured into 100 ml. About 2.5 ml of digested blank was added to
each flask before marking it, to the mark with distilled water. To 0.1 ml of the sample or
standard solution, 5 ml of reagent N1 (34 g sodium salicylate, 25 g sodium citrate and 0.12
g sodium nitroprusside dissolved 1000 ml deionised water) was added. After 15 minutes, 5
ml of reagent N2 (30 g NaOH, 750 ml water and 10 ml sodium hypochlorite in 240 ml
deionised water) was added. The solution was mixed well and left for color development.
Total N in the sample was then determined colorimetrically from the clear solution against
a set of standards at an absorbance of 655 nm (Anderson and Ingram, 1993).
3.2 Evaluation of indigenous legume seed germination
Fifteen species targeted for establishment of indigenous legume fallows (indifallows),
namely, Crotalaria laburnifolia (L.), C. ochroleuca G. Don, C. cylindrostachys Welw.ex
Baker, C. pallida (L.), C. pisicarpa Welw.ex Baker, C. glauca Willd, Eriosema ellipticum
Welw.ex Baker, Neonotonia wightii (Wight & Arn.) J.A. Lackey, Chamaecrista
24
mimosoides (L.), C. absus (L.), Indigofera arrecta Hochst. Ex A. Rich., I. astragalina DC
and Rothia hirsuta, Tephrosia radicans Welw.ex Baker and T. purpurea (L.) were tested
for germination under laboratory conditions.
Before the germination tests, different
scarification treatments were employed:
i.
Sulphuric acid (H2SO4) submersion
ii.
Hot water at 70 oC for 15 minutes and at 80 oC for 10 minutes
iii.
Control
Preliminary results on H2SO4 treatments have shown no significant differences in time of
seed submersion into the acid. Sulphuric acid treatment involved submersion of 400 seeds
in 100 ml of 98 % concentrated H2SO4 for 30 minutes at room temperature. The mixture
was occasionally stirred using a glass rod. The seeds were then removed from sulphuric
acid and rinsed under running water for one minute before being air-dried (Rodrigues et
al., 1990).
Hot water treatment involved subjecting 400 seeds to 70 oC for 15 minutes and to 80 oC
for 10 minutes (Kak et al., 2006), after which the seeds were air dried. Unscarified seeds
were included as a control. A sub-sample of 50 seeds per species per treatment was tested
for germination. Germination tests involved evenly distributing seeds on top of wet filter
papers and covering with cups prior to incubation at 20 – 30 oC in a water bath under light
(two florescence lamps, daylight type, 40w) (International Rules for Seed Testing, 2007).
The experiment was laid out in randomised complete block design (RCBD) with four
replicates, with an existing gradient of light intensity being the reason for blocking.
Germinating species were counted by destructive sampling, with all the germinated seeds
removed on each day of counting.
25
3.3 Field determination of species emergence
Plant emergence was estimated from a 0.5 m x 0.5 m quadrat replicated three times within
each plot (Mapfumo et al., 2005). Three quadrats were randomly selected from a total of
144 grids from each plot. Species emergence was determined at 3-days intervals for the
first 3 weeks and then at 1-week interval for a further 4 weeks, effectively spreading the
monitoring period over two months. Individual plants belonging to the same species were
counted and recorded for each quadrat. To evaluate the effects of soil disturbance
indifallow establishment, indigenous legume species were broadcast on the surface of
disturbed and undisturbed soil, and their emergence patterns also monitored.
3.4 Species biomass productivity and abundance
The identification of the indigenous legumes species had already been done in previous
studies (Mapfumo et al., 2005). The scientific identification of grass and non-legume
broadleaved species was done with the assistance of botanists from the National
Herbarium and Botanic Gardens under the Department of Agricultural Research for
Development of the Zimbabwe’s Ministry of Agriculture. To determine the biomass
productivity, a random quadrat sampling strategy was used. The quadrat measuring 0.5 m
x 0.5 m was made using a 5 mm diameter wire. The quadrat was used to randomly sample
for species (legumes and non-legumes) biomass productivity with at least 3 replicates
sampled from each plot. Legume biomass production was monitored twice between 3 and
6 months after indifallow establishment.
All biomass in the quadrats was cut just above the soil surface, separated into individual
species and then oven-dried to constant weight at 60 oC before the biomass yield was
26
calculated. To estimate species abundance within the system, the species yield was
divided with the total system biomass.
3.5 Estimation of N2-fixation of indigenous legume species under field conditions
The nitrogen difference (ND) method was used for estimating N2-fixation of indigenous
legumes between 3 and 6 months after seeding. Non-leguminous species from natural
fallow plots were used as reference plants. The ND method measures N2-fixation by
comparing N accumulated in a N2-fixing plant with that taken up by a non-fixing
reference plant of corresponding growth habit. The difference between N accumulated by
the fixing legume and a non-fixing reference crop is considered as the N derived from
fixation (Peoples et al., 1989). Because different legume species exhibited different
growth patterns, several reference plants of corresponding growth habits of individual
species were used in this study. Plant species were cut above the soil surface line and
separated into species before being oven-dried to constant at 60 oC. The dried biomass
samples were then ground in a Wiley Mill to pass through a 1 mm sieve. A semi-Kjeldhal
digestion method was used to determine the total N content in the samples (Anderson and
Ingram, 1993). The difference between N accumulated by the fixing legumes and a nonfixing reference plant was assumed to be the N derived from fixation.
3.6 Determination of total plant nitrogen
Total plant N was determined using the Kjeldhal method as describe by Anderson and
Ingram (1983). To a mass of 0.02 g of the sample, 5 ml of concentrated sulphuric acid and
selenium mixture was added. The mixture was digested at about 70 o C until the solution
was clear to pale yellow and then cooled for 10 minutes. The contents of the digestion
flask were transferred into the distillation flask by rinsing twice with distilled water and
27
the sidearm replaced. To the distillate, 20 ml of 50% NaOH was steadily introduced to
liberate NH3 from the solution which was cooled and absorbed in 5 ml of boric acid
indicator until the solution reaches the 50 ml mark. Total nitrogen was then determined by
titration with standardised sulphuric acid (H2SO4). The titre value is the N content of the
sample.
3.7 Determination of total plant phosphorus
About 0.5 g of ground plant samples were digested using the digested method described in
section 3.17. To a 5 ml sub-sample of the digested sample and standard in a 50 ml
volumetric flask, 20 ml of distilled water followed with 10 ml of ascorbic acid were added
starting with standards. The volumetric flask was then filled to the mark with distilled
water. Total phosphorus was then determined colorimetrically at 880 nm wavelength
(Anderson and Ingram, 1993).
28
CHAPTER 4
ESTABLISHMENT AND BIOMASS ACCUMULATION OF
INDIGENOUS LEGUME SPECIES UNDER INDIFALLOW
SYSTEMS ON NUTRIENT DEPLETED SOILS1.2
4.1 Abstract
Studies in Southern Africa have shown the potential of indigenous N2-fixing legumes to
improve soil productivity and reclaim nutrient-depleted fields abandoned by smallholder
farmers. The emergence and growth patterns of fifteen selected indigenous legume species
were evaluated under smallholder farm conditions in Zimbabwe. Crotalaria pallida and C.
ochroleuca were the first to emerge within 14 days from seeding while emergence for
other species was late and spread over two months. Emergence for Eriosema ellipticum,
Crotalaria ochroleuca and C. pallida was above 15 % compared with lowest values of <
10 % for Tephrosia radicans and Indigofera astragalina. Low emergence rates were
attributed to the seed hardness and seed viability, which accounted for > 50 % and 10 - 30
% of germination failure, respectively. Scarification with concentrated sulphuric acid
significantly (P < 0.05) increased germination by > 60 %, reducing seed hardness to < 15
% for most species. Responses to hot water treatment were insignificant in all species.
Annual herbs such as C. cylindrostachys and C. glauca had the shortest vegetative period
of 3 months, yielding about 0.5 t ha-1 (dry weight) biomass. In contrast, fast growing erect
annuals, C. ochroleuca and C. pallida, attained maximum biomass of about 5 and 9 t ha-1,
respectively within a growth period of six months. Biennials (Neonotonia wightii, E.
ellipticum and Tephrosia radicans) exhibited slow growth rate and only attained
maximum biomass in the second season with seasonal yields not exceeding 2 t ha-1. An
exception was Indigofera arrecta which accumulated up to 5 t ha-1 in the second season.
About 50 % of the sown species successfully persisted and established on undisturbed soil
in the second season. Successful establishment of indigenous legume species was
influenced by wide variability of species germination profiles, emergence patterns which
were spread over time and different growth habits resulting in distribution of species
biomass into the system in space and time.
This chapter has been published as follows:
1
T.P. Tauro, H. Nezomba, F. Mtambanengwe, and P. Mapfumo, 2007. Field emergence and establishment
of indigenous N2-fixing legumes for soil fertility restoration. African Crop Science Conference Proceedings
8: 1929-1935.
2
Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2009. Germination, field establishment
patterns and nitrogen fixation of indigenous legumes on nutrient-depleted soils, Symbiosis, 48:92-101.
29
4.2 Introduction
Soils in the smallholder farming area of Zimbabwe are inherently poor in fertility, being
deficient in both macro- and micronutrients, are often abandoned or fallowed after some
years of cropping. Most of the fallowed land is composed of common grass species which
can not build up meaningful soil fertility over a short fallowing cycle (Mapfumo and
Giller, 2001). Efforts to build fertility of these soils using livestock manure, leaf litter and
termitarium soil is limited by low quantity and poor quality resources available to farmers.
In the wake increasing population pressure, the main challenge is to find technologies that
can regenerate some of the fallowed land for crop production.
Introduction of legumes in fallows have been shown to improve the fallow system through
nutrient cycling, litterfall, biomass input and biological nitrogen fixation (Palm, 1995;
Gathumbi et al., 2004; Mafongoya et al., 2006). Although some legume-based soil fertility
restoration options have shown great potential in Sub-Saharan Africa (Jama et al., 1998),
most of them have not been integrated in smallholder farming system due to various
reasons. The use of leguminous species (trees, shrubs and herbs) have failed to be an
integral pathway for revitalizing degraded soil in Sub-Saharan Africa due to the costs
involved for adoption.
According to Giller (2001), successful establishment of legumes is determined by
adaptation, biomass productivity, amount of N2-fixed, weed suppression and persistence in
a given environment. However, the major drawback of some grain legume species have
been poor stand establishment, giving low biomass of < 2 t ha-1 due to severe fertility
constraints in fields often cultivated by smallholder farmers (Mapfumo et al., 1999;
30
Kamanga and Shamudzarira, 2001).
In such soils, productivity of legumes and N2-
fixation is limited by availability of nutrients especially P (Giller, 2001), thus reducing the
biomass and N inputs into the system. Since most grain legumes have failed to make a
significant contribution to fertility restoration of depleted soils, new legume innovations
are required. In Zimbabwe, selection of legumes for use in most farming systems has
focused on grain legumes for human consumption and pasture reinforcement options for
livestock feed (Clatwothy and Thomas, 1972; Maclaurin and Grant, 1985) with little
attention on legumes for soil fertility management.
Recently, work in Zimbabwe by Mapfumo et al. (2005) has identified over 36 different
indigenous legumes species mainly of the genera Crotalaria, Indigofera and Tephrosia
across different agro-regions. These indigenous legumes established and grew on nutrientdepleted soils where common agricultural crops often fail to yield anything. Seed
collection for most of the identified indigenous legume species was done by farmers from
natural stands following some training on identification procedures (Mapfumo et al.,
2005).This reduced the constraints related to seed costs and availability associated with
other legume-based technologies. Integrating N2-fixing indigenous legumes has potential
to contribute to the alleviation of major agricultural constraints associated with nitrogen
(N) inputs in smallholder farming in many parts of Sub-Saharan Africa. However, there is
limited information on the establishment patterns, growth habits and biomass
accumulation of these species when planted in mixtures on nutrient depleted soils. This
chapter focuses on the following specific objectives:
1. To establish germination profiles, emergence and establishment patterns of indigenous
legumes under field and laboratory conditions
31
2. To determine biomass accumulation patterns of different indigenous legumes and nonleguminous species grown in mixtures under indifallow (indigenous legume fallow)
systems in comparison with natural fallows
4.3.0 Materials and Methods
4.3.1 Field sites
Fields abandoned by farmers due to poor soil fertility were identified and used for the
study using a criterion developed by Mtambanengwe and Mapfumo (2005). Field sites
included Domboshawa, Makoholi, Chikwaka and Chinyika, which have been described in
Chapter 3. In Chikwaka, experiments were hosted by farmer Nyamayaro while farmers
Mudange, Chikodzo and Masara hosted the experiment in Chinyika.
The study was conducted in two concurrent experiments that were: (a) a field investigation
on establishment patterns and biomass accumulation, and (b) a laboratory germination test
conducted at Seed Services under the Department of Agricultural Research for
Development of Zimbabwe’s Ministry of Agriculture.
4.3.2 Seed germination tests
Seed was subjected to various scarification methods before germination tests were done
under laboratory conditions. Apart from the control, the scarification methods used
included sulphuric acid submersion and hot water treatments (at 70 oC for 15 minutes and
at 80 oC for 10 minutes) as described in section 3.2. Germinating species were counted by
destructive sampling, where all the germinated seeds were removed on each day of
counting.
32
4.3.3 Determination of species emergence patterns
Indifallows were established by broadcasting mixtures of Crotalaria laburnifolia, C.
ochroleuca, C. pallida, C. pisicarpa, C. cylindrostachys, Eriosema ellipticum, Neonotonia
wightii, Chamaecrista mimosoides, C. absus, Indigofera astragalina and I. arrecta seeds
at a seeding rate of 120 seeds m-2 species-1 during the 2005/06 rainy season. Seeding rates
were based on preliminary seed viability results by Venge (2003). Planting were done on:
(a) plots where soil had been disturbed by hand hoeing, and (b) plots with undisturbed
soil. The experiment was laid out as a randomised complete block design (RCBD) with
three replicates per treatment. The treatments were established in plots measuring 6 m by
4.5 m. Emergence was estimated in quadrats randomly selected within the plots and this
involved counting individual plants belonging to the same species.
4.3.4 Monitoring of species growth habits
Species were monitored from emergence to senescence over two years using an existing
observation scheme. The scheme included recording qualitative description of growth
patterns, time to sprout, days to physiological maturity and measurement of biomass
accumulation as described in section 3.4. Identification of non-legumes in indifallows was
done with the assistance of botanist while biomass of non-legumes was also quantified
using the quadrat method. The following treatments were effected:
i.
1-year Indifallow (established during 2005/06 season)
ii.
2-year Indifallow (established during 2004/05 season)
33
4.3.5 Statistical analysis
Data was analyzed using GENSTAT statistical package (GENSTAT, 2005). Mean
comparisons for abundance and productivity of legumes and grasses were done using ttest. All the mean comparisons were done at 0.05 % significance level and all the biomass
is reported on a dry matter basis.
4.4.0 Results
4.4.1 Germination of indigenous legume species
Eriosema ellipticum and Crotalaria pallida had the highest germination of >40%
compared with 10-20% for C. ochroleuca, C. laburnifolia, Chamaecrista mimosoides and
I. astragalina. Crotalaria piscarpa, C. glauca, C. absus, N. wightii, I. arrecta and R.
hirsute had the lowest final germination of <10 % (Table 4.1). Seed hardness was a major
constraint to germination for most species, accounting for >50% of cases of germination
failure. Apart from seed hardness, lack of viability accounted for 10-30% of species
germination failure (Table 4.1). Scarification using concentrated sulphuric acid
significantly (P<0.05) increased germination by ~60% and reduced seed hardness to less
than 15% for most species. However, submersing E. ellipticum and C. mimosoides for 60
minutes in acid caused germination failure. Hot water treatment had no significant
(P>0.05) effect on species germination (Table 4.1).
34
Table 4.1 Germination (%) of indigenous legume species following different
scarification methods under laboratory conditions
Treatment
Control
Concentrated
Hot water
Hot water
Sulphuric
70oC for 15 80oC for 10
acid*
min.
min.
91(1.7)
16(2.5)
6(2.7)
16(2.5)
Crotalaria pisicarpa
90(1.7)
45(5.5)
40(2.5)
33(0.1)
Crotalaria pallida
79(1.9)
20(5.9)
10(1.9)
7(1.0)
Indigofera astragalina
64(4.7)
31(4.7)
3(1.3)
4(1.4)
Indigofera arrecta
0
38(7.4)
57(4.1)
38(7.4)
Eriosema ellipticum
59(3.5)
21(2.1)
5(1.0)
8(1.6)
Rothia hirsuta
0
0
6(1.0)
0
Chamaecrista absus
84(2.6)
7(1.0)
7(1.9)
6(1.3)
Crotalaria cylindrostachys
87(2.6)
nd
10(3.2)
nd
Crotalaria laburnifolia
73(3.9)
nd
13(1.9)
nd
Crotalaria ochroleuca
74(3.6)
nd
7(1.0)
nd
Crotalaria glauca
nd
8(1.8)
nd
50(6.0)
Neonotonia wightii
7(0.5)
19(4.3)
nd
0
Chamaecrista mimosoides
16(1.8)
nd
19(2.2)
nd
Tephrosia radicans
18(4.2)
nd
21(2.1)
nd
Tephrosia purpurea
nd = not determined; * = species seed scarified in acid for 30 minutes; min = minutes;
Figures in parentheses denote standard errors.
Indigenous legume species
35
4.4.2 Species emergence in indifallow systems
There was wide variability in indigenous legume species emergence and establishment.
Crotalaria pallida and C. ochroleuca were the first to emerge in about 14 days, while
emergence was spread over two months for other species. By 21 days, 90% of Eriosema
ellipticum emerging seedlings had been recorded. For Tephrosia radicans, Indigofera
astragalina and Indigofera arrecta 90% of the emerging seedlings were recorded 5 weeks
after sowing.
Across all sites, Eriosema ellipticum had the highest emergence (>20 %). Crotalaria
ochroleuca and C. pallida had medium emergence of 3 – 15% compared with lowest
values of 1 – 10% for Tephrosia radicans, Indigofera astragalina and Indigofera arrecta
(Table 4.2). Soil disturbance significantly (P<0.05) increased emergence of Crotalaria
pisicarpa, C. laburnifolia, Eriosema ellipticum, Indigofera arrecta and I. astragalina both
at Domboshawa and Chikwaka (Figure 4.2). However, the emergence of C. pallida, C.
cylindrostachys, Tephrosia radicans, and Chamaecrista mimosoides was not significantly
affected by soil disturbance (Figure 4.2).
36
Table 4.2. Field emergence (%) of selected indigenous legume species on sandy soils
under different rainfall regions in Zimbabwe.
Indigenous legume species
Makoholi
Chinyika
(450-650 mm yr-1)
(650-750 mm yr-1)
(>750 mm yr-1)
(>750mm yr-1)
(608 mm)**
(583 mm)**
(1042 mm)**
(600 mm)**
Domboshawa
Chikwaka
40(6.3)
45(7.0)
18(4.0)
22(4.0)
Crotalaria ochroleuca
3(0.4)
15(3.0)
13(3.3)
8(3.0)
Crotalaria pallida
4(1.5)
nd
9(2.6)
10(4.1)
Crotalaria cylindrostachys
4(0.4)
3(0.1)
8(1.4)
7(0.9)
Indigofera arrecta
1(0.4)
8(1.8)
10(1.8)
10(1.4)
Indigofera astragalina
2(0.8)
5(1.1)
0
3(0.8)
Tephrosia radicans
2(0.5)
5(1.3)
2(1.1)
3(1.5)
Eriosema ellipticum
nd = not determined; ** = rainfall that had been received at time of evaluation; Figures in
parentheses denotes standard error
37
Figure 4.1. Effects of soil disturbance on species emergence of early planted species,
under high rainfall conditions on-station at Domboshawa (a) and onfarm Chikwaka (b), Zimbabwe during 2006/07 season. SE denotes
standard error.
38
4.4.3 Species growth habits
The indifallow system was a mixture of annuals, biennials and perennials of both legume
and non-legumes species of complementary growth habits resulting in the distribution of
biomass inputs in space and time into the system (Table 4.3). Of the three groups, there
was variation of growth rates and traits. Annuals included Crotalaria pallida, C.
ochroleuca, C cylindrostachys, C. glauca, C. pisicarpa, Indigofera astragalina,
Chamaecrista mimosoides, C. absus, Tephrosia longipes and T. purpurea. Crotalaria
pallida and C. ochroleuca are erect legume shrubs that exhibited fast growth habits and
attained maximum biomass of about 9 and 5 t ha-1, respectively within a growth period of
six months (Figure 4.2). They were distinctively taller in stature exceeding a height of 1.5
m, and had a longer vegetative phase compared to all other erect legumes in the indifallow
system. While the two had similar growth habits, C. ochroleuca had prolific branching at
the base of main stem just above the ground, while C. pallida started branching midway
along the main stem all the way to the growing tip (Table 4.3).
Crotalaria cylindrostachys, C. pisicarpa and C. glauca are annual herbs with a vegetative
period that only lasted 3 months, yielding about 0.5 t ha-1 biomass on a dry weight basis
(Figure 4.2). However, C. cylindrostachys had more leafy biomass and it completely
senescenced all its biomass as seed matured. Crotalaria pisicarpa is a slow growing
legume and exhibited a dual growth habit. The lower branches had a prostate growth habit
while the upper branches and main stem were erect. It also reached its peak biomass after
3 months. Indigofera astragalina, Chamaecrista mimosoides, C. absus, Tephrosia
longipes and T. purpurea are annual herbs which accumulated an average of 350-450 kg
39
biomass per month over a growth period of 6 months. In the second season, Indigofera
astragalina, Tephrosia longipes and T. purpurea resprouted from the root crown.
Biennials included Neonotonia wightii, I. arrecta, Tephrosia radicans and E. ellipticum.
Most of these indigenous legumes had a slow initial growth rate, dying back after six
months of establishment and only resprouting from stems to reach a maximum biomass of
up to 2 t ha-1 in the second season. An exception was I. arrecta, which attained about 5 t
ha-1 in the second season (Figure 4.2). During the second season, these biennials
vigorously resprouted and had a relatively faster growth rate as compared to that observed
in the first season, mostly likely due to an improved rooting environment (Figure 4.2).
While most of the species sprouted from the root crown, Indigofera arrecta and E.
ellipticum sprouted from the standing stems. Tephrosia radicans and Neonotonia wightii
had prostrate and tailing growth habits, respectively. The perennial species was Crotalaria
laburnifolia which had an erect growth habit, reaching a height of ~1 m and also resprouting from stems. It attained maximum biomass of about 3.5 t ha-1 in the second
season (Figure 4.2).
The majority of non-legume species were annual grasses, sedges and broadleaf weeds that
had fast initial growth rates, attaining peak biomass at 3 months and then dying back
(Figure 4.2). Maximum total legume biomass of up to ~13 t ha-1 was attained under 1-year
indifallow (Figure 4.3). Non-legumes species only managed to attain a maximum of ~ 4 t
ha-1 under 1-year indifallow. However, in the second season, non-legumes species
biomass reached a peak of ~ 7.5 t ha-1, out-competing most indigenous legume species
which attained < 5 t ha-1 of biomass (Figure 4.3).
40
Table 4.3. Growth habits and patterns of different indigenous legume species studied under smallholder farm conditions
in Zimbabwe
Species
Description Growth habit
Time to flowering
Sprouting
period
Annual
Erect, about 1 metre tall in stature
180
ns
Crotalaria pallida
Crotalaria laburnifolia
Crotalaria ochroleuca
perennial
Annual
119 and 147 days
77 days
Mid October
ns
Annual
Annual
Annual
Biennial
Biennial
Erect, about 1 metre tall in stature
Erect, about 2 metre tall in stature,
prolific branching
Erect and prostate
Herb
Herb
Erect
Trailing
Crotalaria pisicarpa
Crotalaria cylindrostachys
Crotalaria glauca
Eriosema ellipticum
Neonotonia wightii
65 days
70 days
60 days
In second season
In second season
Chamaecrista mimosoides
Chamaecrista absus
Indigofera arrecta
Annual
Annual
Biennial
Herb
Herb
Erect
90 days
90 days
In second season
I. astragalina
Annual
119 day
Tephrosia radicans
Biennial
Erect herbs first season then erect and
prostate in second season
Prostate
In second season
Tephrosia longipes
Annual
Herb, with longer internodes
90 days
Tephrosia purpurea
Annual
Herb, with shorter internodes
90 days
ns
ns
ns
Mid October
December, after
the rains
ns
ns
Early
November
December, after
the rains
December, after
the rains
December, after
the rains
December, after
the rains
ns= non-sprouting species
41
12
10
Biomass productivity (t ha-1)
Crotalaria pallida
8
Non-leguminous biomass
Indigofera arrecta
6
Crotalaria ochroleuca
4
Crotalaria laburnifolia
2
Eriosema ellipticum
Neonotonia wightii
0
N
ov
em
be
r2
00
Ja
4
nu
ar
y
20
M
05
ar
ch
20
05
M
ay
20
05
Ju
ly
Se
20
pt
05
em
be
r2
N
ov
00
em
5
be
r2
Ja
nu 005
ar
y
20
06
M
ar
ch
20
06
M
ay
20
06
Crotalaria cylindrostachys
Time (months)
Figure 4.2. Biomass accumulation patterns of indigenous legume species over two
seasons under high rainfall conditions at Domboshawa, Zimbabwe
during 2004/05 and 2005/06 seasons.
42
16
Total legume biomass
14
-1
Biomass productivity (t ha )
12
10
8
6
4
2
Total non-legume biomass
N
ov
em
be
r2
Ja
00
nu
4
ar
y
20
M
ar 05
ch
20
05
M
ay
20
05
Ju
ly
Se
2
pt
em 005
be
N
r2
ov
00
em
5
be
r2
Ja
00
nu
5
ar
y
20
06
M
ar
ch
20
06
M
ay
20
06
0
Time (months)
Figure 4.3. Biomass accumulation patterns of total leguminous and non-leguminous
species over two seasons under high rainfall conditions (>750 mm yr-1),
at Domboshawa, Zimbabwe during 2004/05 and 2005/06 seasons.
43
4.5.0 Discussion
4.5.1 Species germination and emergence under indifallows
The low rate of emergence of the majority of the indigenous legumes may be attributed to
the seed hardness caused by the impermeability of seed coat that maintains the quiescence
of the embryo by preventing imbibition of water (Russ, 1993). Germination tests of these
species showed that seed hardness was the major constraint for germination of most
indigenous species, accounting for >50 % of cases of germination failure for most species.
Consistent with findings from this study Loi et al (1999) also reported the existence of
seed hardness among species even of the same genera. Most wild or indigenous legumes
such as Crotalaria laburnifolia, C. medicaginea, C. pallida and C. retusa have been
shown to possess physical dormancy (Venge, 2003; Kak et al., 2006). Such dormancy
results in erratic stand establishment as shown by the low emergence under field
conditions. Apart from seed hardness, seed mortality accounted for 10 – 15 % of species
failure under field conditions.
Sulphuric acid scarification significantly increased germination. Similar results have been
reported (Rodrigues et al., 1990; Mackay et al., 1995; Mapfumo, 1999). However, the
practicality of this procedure under smallholder farming conditions poses a challenge. For
instance, a shorter time of soaking in sulphuric acid may be required for Chamaecrista
absus and C. mimosoides to avoid loss of seed viability. As the intensity of seed hardness
varies among legume species (Kelly et al., 1992), it is likely that indigenous legume
species under study responded differently to acid treatments. Hot water scarification might
have been a practical option for most farmers; however this simple treatment had no effect
on species germination. The results are contrary to findings by Kak et al (2006), who
44
reported enhancement germination of C. laburnifolia, C. medicaginea, C. pallida and C.
retusa using hot water treatment. Therefore for practical purposes farmers may need to
further increase their seeding rate to cater for the seed physical dormancy. There also need
to monitor if natural process of heat and fluctuating moisture conditions may influence
germination.
The high emergence observed both under field and laboratory conditions for E. ellipticum
suggest that it is free from physical dormancy and does not need scarification. Sunnhemp
(Crotalaria juncea), which has found widespread use as a green manure and cover crop in
cropping systems is known to have high germination (Kak et al., 2006). The variation in
emergence of species over time ensured successful establishment of indifallows against
species failure from adverse weather conditions or poor early growth. The result also
indicated that increasing the seeding rate (Venge, 2003) alone may not produce a good
stand hence the need to consider mechanisms to reduce seed hardness.
Soil disturbance enhanced the emergence of indigenous legume species suggesting that
there was better seed-soil contact for species germination. Clatworthy and Thomas (1972)
had previously highlighted the importance of disturbing the soil to set back grass growth
and competition during pasture establishment. However, there was no significant effect of
disturbing the soil on species emergence of C. pallida, C. cylindrostachys, T. radicans,
and C. mimosoides in the high rainfall areas of Domboshawa and Chikwaka and this was
attributed to adequate surface soil moisture at planting and during the germination phase.
This is in agreement with observations made by Lwoga (1983) who noted that when
surface soil moisture was not limiting, oversown seeds germinated well.
45
4.5.2 Species growth habits
In general, Crotalaria species were fast growing and performed better in terms of biomass
accumulation than Indigofera and Tephrosia species suggesting that Crotalaria species
were stronger competitors in the mixtures. However, species such as C. ochroleuca only
made a significant contribution in the first season and disappeared in the second season,
and for that reason there is a need to have persistent species such as E. ellipticum and I.
arrecta in indifallows. Crotalaria laburnifolia regenerated through reseeding and
resprouting, and therefore persisted well. The ability of some species to persist into the
second season also suggests that these species required little maintenance once they have
established, conforming to findings by Metcalf (2005). According to Piano (1993), short
lived species such as C. cylindrostachys, C. pisicarpa and C. glauca have an advantage in
marginal areas where rainfall ends abruptly since they are able to produce mature seed
faster than slow-flowering and long maturity species. Conversely, in high potential areas,
high rainfall after the vegetative cycle would favour pathogen and saprophyte attack on
mature seeds of short duration species while late maturing species are likely to escape
attack (Sulas et al., 2000).
Biennial species such as N. wightii and E. ellipticum apparently exhibited the “suffretex”
habit (Maruzane and Zirobwa, 2002). Suffretex habit enables seedling otherwise thought
to be dead to recover when conditions improve, owing to the persistence of rootstock in
the ground. These species had an early season die back as they developed a good rooting
system to support the shoot development in the next season. As biennials, N. wightii and
E. ellipticum would not be suitable for 1-year fallows. For longer term fallows, mixing
46
species with different growth habits would ensure that species biomass contributions into
the system are spread over time.
Mixing indigenous legume species of different growth habits ensured successful
establishment of indifallows by safeguarding against single species failure and
guaranteeing distribution of species contribution into the system in space and time. Mixing
species with complementary growth traits has also been reported to produce a more
diverse fallow system and also maximize biomass productivity (Cadisch et al., 2002).
Gathumbi et al. (2004) showed that mixed species fallows provided a better risk
management strategy through compensatory biomass and nutrient production gains
obtained from the strongly competing species. Crotalaria laburnifolia and C. ochroleuca
produces a dense canopy cover, which can shade off other species such as T. radicans and
C. pisicarpa thereby reducing their contribution in the system. Therefore in terms of
establishment options, species with a prostate growth habit such as T. radicans and a slow
growth rate such as C. pisicarpa should be seeded at a higher rate and early before fast
growing species such as C. laburnifolia and C. ochroleuca.
Over time, indigenous legumes out-competed non-legume species despite having a slow
initial growth rate because the system was N deficient which affected non-legume growth
while these legumes were using biological nitrogen fixation to meet their N needs. The
dominance of grasses which then eclipsed most annual legumes in the second season of
indifallow might be associated with an improved N environment following decomposing
legume residues from the first season. With this indifallow system, it then possible to
reduce the labour related to weeding associated with other legume-based technologies
during establishment as non-legumes have no effect on the legumes biomass inputs.
47
4.6 Conclusions
Indigenous legume species had different establishment patterns being influenced by wide
variability of species germination profiles, emergence patterns and different growth habits.
Low germination rate are related to seed coat hardness but increasing seeding rate can
result in desired plant population. Acid scarification significantly increased species
germination by ~60% where as hot water pre-treatment had no effect. Mixing indigenous
legume species resulted in distribution of species biomass into the system across time,
with Crotalaria cylindrostachys, C. pisicarpa and C. glauca attaining peak biomass (0.5 t
ha-1) over 3 months, while others (Crotalaria pallida and C. ochroleuca) gave maximum
biomass (5-9 t ha-1) at 6 months and Indigofera arrecta attained maximum biomass (5 t ha1
) in the second season after resprouting. Total legume biomass out-yielded non-legumes
in indifallow but non-legumes reached a peak of 7.5 t ha-1 in the second season as a result
of rotational benefits from the N inputs from decomposition of N-rich litter of short lived
indigenous legume species. Mixing indigenous legume species of different growth habits
ensured successful establishment of indifallows by safeguarding against single species
failure and guaranteeing distribution of species biomass contribution into the system in
space and time.
48
CHAPTER 5
CHANGES IN PLANT SPECIES COMPOSITION AND ABUNDANCE IN
INDIFALLOW SYSTEMS
5.1 Abstract
Legume-based soil fertility innovations that offer alternative N nutrient sources to address
soil fertility challenges for resource-constrained farmers are being sought by smallholder
farmers in many sub-Saharan countries. Studies were conducted under low (450-650 mm
yr-1) to high (> 800 mm yr-1) rainfall areas in Zimbabwe to investigate the changes in
species composition within indigenous fallows (indifallows) established on nutrient
depleted soils. The indigenous legume seed was broadcasted in mixtures at 120 seed m-2
species-1 over two growing seasons. In the first season, Eriosema ellipticum was the most
dominant legume species having a population of 42 plants m-2 (8 %) while Crotalaria
pallida, Crotalaria cylindrostachys, Crotalaria ochroleuca and Indigofera arrecta each
had population of < 23 plants m-2 (5 %). Over time there was a significant (P< 0.05)
decrease in the population of E. ellipticum and Neonotonia wightii as a result of low
survival rates as competition increased. Introduction of indigenous legume species
suppressed the abundance of predominant non-legume species such as Cynodon dactylon
and Richardia scabra to 17 % and 7 % respectively, while the abundance of the same
species in natural fallows increased to 47 % and 11%, respectively. Overall, the most
abundant species were C. pallida and C. ochroleuca having dense canopy cover. The two
species yielded 9 and 5 t ha-1 of biomass respectively, translating to approximately 48%
and 26% of indifallow biomass. There was a significant change in non-leguminous species
composition and abundance in the second season. Abundance of annual legume species
such as C. pallida, C. ochroleuca and C. cylindrostachys decreased to < 2% in the second
season while there was a disappearance of Richardia scabra, Bulbostylis hispidula and
Nicandra physalodes. There was significant (T(obs)>1.96) decrease in legume species
diversity into the second season. Non-legume species diversity decreased as C. dactylon
dominated in the second season apparently associated with improved N economy where
indigenous legumes once dominated. This study suggest that establishment of these
species in degraded soils can influence changes in non-legume composition from legume
competition and resultant increase of N in the system while the changes in legumes
population and abundance was influenced by species emergence, competitiveness and
growth trait.
49
5.2 Introduction
In many parts of tropical Africa, much of the land cleared for agriculture is abandoned
after a few years, due to invasion of aggressive weeds or a drastic decline in soil fertility
(Donfack et al., 1995). Resources for the management of soil fertility and weeds are
generally scarce in most smallholder farming areas raising the need for soil fertility
options that complement weed management. However, most resource-constrained farmers
depend on traditional technologies such as natural grass fallows and other locally derived
nutrient resources to manage soil fertility (Mtambanengwe and Mapfumo, 2005;
Mafongoya et al., 2006). Most of these methods have consistently shown little impact
while efforts to use some potential legumes such as mucuna (Mucuna pruriens) and
pigeon pea (Cajanus cajan) have been hindered by poor establishment (Hikwa et al.,
1998; Mapfumo et al., 1999).
Successful establishment of legumes is determined by weed suppression and persistence in
a given environment (Peoples et al., 1995). Past studies on legume-based technologies
have focused on biomass productivity and residual fertility benefits (Hooper and Vitousek,
1998; Mafongoya et al., 2006), with little attention on the population dynamics of both
legumes and non-legume species contributing to the total system biomass. Species
diversity and functional composition have been confined to forage legumes in pastures
(Clatworthy and Thomas, 1972; Barker et al., 2002; Guretzky et al., 2005), but there has
been little of such information on unpalatable legumes for soil fertility management.
Mapfumo et al. (2005) advocated for the use of well-adapted indigenous legume species
found across different agro-regions in Zimbabwe for soil fertility management option for
poor households. Most of such legumes have been ignored in the past as a result of being
50
unpalatable despite their ability to grow on degraded soils on which other legumes failed
to make impact. However, when unchecked, such wild plant species can become
troublesome weeds once integrated in farming systems. For example, species such as
Lantana camara and Pinus patula in Zimbabwe have posed serious problems to
conservationists, forestry managers, farmers and land owners because of their
invasiveness (Maroyi, 2006). It is therefore imperative to gain more information about
indigenous legume species dynamics and factors driving persistence under the indifallow
system.
There has been a general focus on implications of N management on species composition
and community structure in natural systems and agricultural systems (Wilson and Tilman,
1993; Clements et al., 1994). Little information exists on species competitive ability and
shading characteristics within indigenous fallows (indifallows). Before integrating
indifallows into farming system, there is need to understand how species composition and
abundance changes over time within indifallows and how best to manage indigenous
legume species for use as a soil fertility management resource. This study investigated the
changes in species composition under 1 and 2 year indifallows on nutrient depleted soils.
This chapter focuses on the following specific objectives:
1. To investigate changes in abundance of N2-fixing indigenous legumes seeded in
mixtures and grown over one- to two-cropping seasons.
2. To evaluate the influence of mixed indigenous legume stands on composition and
abundance of non-leguminous species after two years in comparison to natural
fallows
51
5.3.0 Materials and methods
5.3.1 Determination of species abundance
Relative species abundance under indifallow and natural fallow systems was estimated on
the basis of individual species biomass productivity in relation to total system biomass.
Species biomass sampling was done 3, 6, 15 and 20 months spreading over two rainy
seasons using a random quadrat method. Following individual species biomass
quantification, species abundance was calculated (Mapfumo et al., 2005).
5.3.2 Monitoring changes in species composition over time
The following treatments were effected to investigate changes in species composition over
time in both indifallow and natural fallow systems:
i.
1-year Indifallow
ii.
2-year Indifallow
iii.
1-year natural fallow
iv.
2-year natural fallow
Indifallows were established at Domboshawa and Makoholi through mixtures of
indigenous legume species as earlier described in Chapter 3. Control plots were disturbed
by ploughing and left to natural fallow. The experiment was laid out in randomized
complete block design (RCBD) with three replicates per treatment. Plots were 36 m2 and
blocking was based on soil fertility within the fields. Species population and biomass was
quantified by sampling three replicate quadrats per plot between 3 and 20 months
spreading over two rainy seasons. The quadrats measured 0.5 m x 0.5 m and all legume
species plants were counted before all plant biomass was cut at soil surface, dried and
weighed as earlier described in Chapter 3 (section 3.4).
52
5.3.4 Statistical analysis
Data was analyzed using GENSTAT statistical package (GENSTAT, 2005). Mean
comparison for legumes and grasses abundances and productivity was done by t-test. The
Shannon-Wiener diversity index (H`) was used to test for diversity changes over time and
comparisons of system diversity was done at critical t value of 1.96 and 774 degrees of
freedom (Guretzky et al., 2005). Species evenness was calculated with the formula: E =
H’/ ln S, where S was the number of species present. All the mean comparisons were done
at 0.05 % significance level and all the biomass is reported on a dry matter basis while
species abundance is reported in percentages of plant species per plot.
5.4. Results
5.4.1 Changes in indigenous legume species composition in indifallows
The indifallow system at Domboshawa was comprised of fourteen indigenous legume
species. The predominant non-legume species in the system included the grasses Cynodon
dactylon, Eragrostis patens, Bulbostylis hispidula, a sedge Cyperus esculentus and
broadleaved species such as Commelina benghalensis, Bidens pilosa and Richardia scabra
(Figures 5.1). Eriosema ellipticum was the most dominant legume species having a
population of 42 plants m-2 (8%) while Crotalaria pallida, Crotalaria cylindrostachys,
Crotalaria ochroleuca and Indigofera arrecta each had populations of < 23 plants m-2
(5%) (Table 5.1). However, E. ellipticum accumulated insignificant biomass resulting in
C. pallida and C. ochroleuca being the most abundant species which attained 9 and 5 t ha1
of biomass respectively, representing 48% and 26% of total indifallow biomass.
Indifallow at 3 months had the highest Shannon-Wiener index (H’) and evenness (E) than
53
at 6, 15 and 18 months after establishment indicating more diverse and greater legume
species evenness (Table 5.1).
Over time there was a significant (T observed
(obs)>1.96)
decrease in legume species
diversity from 6 months after establishment into two-year indifallow. However, there was
no significant (T (obs) <1.96) difference in legume species diversity at 6 and 15 months
after establishment. There was a substantial decrease in the population of E. ellipticum, C.
pallida, C. ochroleuca and C. cylindrostachys into the second season. Indigofera
astragalina, T. radicans and C. pisicarpa did not persist into the second cropping season
(Table 5.1). Annual legume species such as C. pallida, C. ochroleuca and C.
cylindrostachys decreased in abundance from 48% to 0.9%, 26% to 2.2% and 3.5% to
0.2% respectively into the second season (Figure 5.2). However, I. arrecta, N. wightii and
Chamaecrista mimosoides persisted into the second season without any changes in species
population over the growing seasons. Overall, about 50 % of the sown species
successfully persisted and established on undisturbed soil in the second season.
54
Ageratum conyzoides Eragrostis patens
6.0%
1.7%
Eleusine indica
Oldenlandia herbacea
1.7%
0.4%
Bulbostylis hispidula
0.1%
Commelina benghalensis
5.1%
Bidens pilosa
0.3%
Solanum aculeastrum
10.9%
Crotalaria pallida
33.6%
Eriosema ellipticum
2.4%
Richardia scabra
12.5%
Galinsoga parviflora
2.8%
Cyperus esculentus
3.3%
Cynodon dactylon
4.6%
Crotalaria cylindrostachys
3.5%
Crotalaria ochroleuca
7.9%
Crotalaria laburnifolia
0.5%
Indigofera arrecta
2.6% Neonotonia wightii
0.0%
Figure 5.1. Relative abundance of legume and non-legume species at 3 months after
establishment of indifallows at Domboshawa (>800 mm yr-1), in the
2005/06 season.
55
Table 5.1. Changes in indigenous legume species population over time at
Domboshawa, Zimbabwe.
Population in indifallow system (plants m-2)
Indigenous legume species
Months after planting
3
Crotalaria laburnifolia
Crotalaria pisicarpa
6
12(2.1)
3(1.3)
15
18
4(2.3)
38(12.7)
17(5.7)
ni
ni
ni
Crotalaria pallida
26(7.1)
40(5.4)
8(1.2)
4(2.3)
Crotalaria ochroleuca
26(12.3)
10(2.0)
7(3.5)
3(1.6)
Crotalaria cylindrostachys
14(4.4)
8(1.2)
4(2.4)
ni
Neonotonia wightii
13(5.3)
4(2.3)
9(2.9)
8(4.1)
ni
ni
ni
Indigofera astragalina
3(1.3)
Indigofera arrecta
20(5.3)
15(8.1)
14(5.3)
14(1.3)
Eriosema ellipticum
42(12.2)
17(2.8)
13(0.8)
8(2.7)
Chamaecrista mimosoides
3(1.3)
3(1.7)
4(0.3)
6(2.0)
Tephrosia radicans
6(4.2)
ni
ni
ni
Tephrosia purpurea
12
11
8(0.4)
7(2.4)
Tephrosia longipes
8
7
4(2.3)
6(1.2)
Shannon-Wiener diversity
2.29
2.00
2.01
2.07
0.89
0.87
0.88
0.88
index (H’)
Evenness (E)
ni = not identified in the indifallow system; Figures in parentheses denotes standard errors.
56
Other non-legume
species
Crotalaria pallida
7.9%
0.9%
Digitaria decumbens
3.3%
Crotalaria laburnfolia
19.2%
Ageratum conyzoides
4.6%
Eriosema ellipticum
8.0%
Crotalaria
cylindostachys
0.2%
Cynodon dactylon
16.1%
Crotalaria ochrelueca
2.2%
Chameacrista
mimosodies
0.2%
Macrotylomia daltonii
1.1%
Neonotonia wightii
4.0%
Tephrosia longipes
9.2%
Tephrosia purpurea
23.2%
Figure 5.2. Relative abundance of legume and non-legume species at 15 months of a
two year indifallow Domboshawa (annual rainfall = 1400 mm yr-1), in the
2005/06 season.
On a nutrient depleted soil at Mudange farm, the indifallow was comprised of seventeen
species, eight being non-leguminous species while the remaining nine were indigenous
legume species. The predominant non-leguminous species included Richardia scabra,
Eragrostis patens, Bulbostylis hispidula, Commelina benghalensis, Hibiscus meeusei and
Cyperus esculentus, showing close similarities with observations at Domboshawa despite
the relatively less annual rainfall in Chinyika. Domboshawa has an average annual rainfall
of > 800 mm yr-1 compared with 650 - 750 mm yr-1 for Chinyika. Consistently, Eriosema
ellipticum was the most dominant legume species having a population of 79 plant m-2 (16
%) while C. ochroleuca, C. cylindrostachys, and I. arrecta had population of < 20 plant m2
(5 %) each (Table 5.2). Overall, the most abundant species in the indifallow system was
57
C. ochroleuca, which attained a maximum of 7.7 t ha-1 (Table 5.3). Similarly, indifallow
at 3 months had the highest Shannon-Wiener index (H’) indicating more diverse system
but indifallow after 15 months had greater species evenness.
Indigofera arrecta, I. astragalina, T. radicans and Chamaecrista mimosoides persisted
into the second season without any apparent changes in species population while
Crotalaria ochroleuca and C. glauca did not persist (Table 5.2), most likely linked to
growth habits and competition from non-legume biomass. There was a significant (T (obs)
>1.96) decrease in H’ over time into the second (Table 5.2). There was a significant
(P<0.05) decrease in E. ellipticum and N. wightii populations as a result of low survival
rates over the growing season, while the population of C. cylindrostachys significantly
(P<0.05) increased in the second season (Table 5.2) probably from the existing soil seed
reserve since the species in prevalent in Chinyika. About 50 % of the sown species
successfully persisted and established on undisturbed soil in the second season.
58
Table 5.2. Changes in indigenous legume species population over time at Mudange,
Zimbabwe.
Population in indifallow system (plants m-2)
Indigenous legume species
Months after planting
3
6
20(9.2)
28(11.6)
ni
Crotalaria glauca
8(2.3)
12(2.1)
ni
Crotalaria cylindrostachys
5(1.3)
3(1.3)
Neonotonia wightii
6(3.5)
ni
ni
Indigofera astragalina
5(3.5)
5(1.3)
8(2.3)
Indigofera arrecta
13(5.2)
5(0.7)
12(4.1)
Eriosema ellipticum
74(16.7)
48(5.1)
25(6.6)
Chamaecrista mimosoides
10(2.3)
6(0.8)
5(2.9)
Tephrosia radicans
11(1.3)
7(2.7)
5(1.3)
Shannon-Wiener diversity
1.70
1.64
1.63
0.78
0.79
0.91
Crotalaria ochroleuca
15
14(1.6)
index (H’)
Evenness (E )
ni = not identified in the indifallow system; Figure in parentheses denotes standard error
59
Table 5.3. Biomass productivity of indigenous legumes species at 6 months after
sowing, under high and medium rainfall conditions in Zimbabwe during
the 2005/06 rainfall season.
Biomass productivity (t ha-1)
Indigenous legume species
(> 750 mm yr-1)
(650 -750 mm yr-1)
Domboshawa
Chikwaka
Chinyika
Crotalaria pallida
8.9
3.4
np
Crotalaria ochroleuca
4.1
0.5
7.7
Crotalaria cylindrostachys
0.2
0.5
0.2
Eriosema ellipticum
0.7
0.1
0.07
<0.1
ni
ni
Indigofera astragalina
ni
ni
0.3
LSD
3.1
2.2
5.2
Indigofera arrecta
np = not planted during establishment of indifallow; ni = not identified in the indifallow
system
5.4.2 The effect of indigenous legume species on non-legume species composition
In the first year, introduction of indigenous legumes resulted in the abundance of common
species being suppressed. For example, at Domboshawa the abundance of Cynodon
dactylon and Richardia scabra was ~5% and ~13 % in indifallows while the same species
constituted 12% and 18% of total biomass under natural fallow (Figure 5.3). Similar
trends were also observed at Makoholi, Chikwaka and Chinyika despite having different
non-legume compositions. Within one season, abundance of Cynodon dactylon increased
by 12% in indifallows, while that of Eleusine indica, Acanthospermum hispidum and
Cyperus esculentus decreased by 6 - 9%.
60
In the second season, there was a disappearance of species that included Richardia scabra,
Bulbostylis hispidula and Nicandra physalodes while the abundance of Cynodon dactylon
increased by 42%. In contrast, Cynodon dactylon abundance only increased by 2% in the
second season under natural fallow. The abundance of Cynodon dactylon translated to
increase in non-legume biomass in the second season which resulted in a significant
decrease in non-legume diversity.
Digitaria decumbens
6.1%
Eleusine indica
7.2%
Eragrostis patens
4.9%
Cynodon dactylon
12.1%
Cyperus esculent
6.7%
Oldenlandia herbacea
6.9%
Bulbostylis hispidula
1.8%
Richardia scabra
17.6%
Commelina benghalensis
6.9%
Acanthospermum
hispidum
29.8%
Figure 5.3. Relative abundance of non-legume species at 3 months in natural fallow
systems, expressed as % of total biomass per unit area at Domboshawa
(>800 mm yr-1), in the 2005/06 season.
61
5.5.0 Discussion
5.5.1 Species dynamics in an indifallow system
Changes in indigenous species composition were influenced by seed germination,
competitiveness and regenerative ability of species within the indifallow. The high
population of Eriosema ellipticum in indifallow than other species across all study sites
was consistent with germination tests findings (Venge, 2003; Tauro et al., 2007). The
decrease in the population of E. ellipticum observed in the second season might indicate
its low survival potential implying reduced species biomass contribution into the system
during this phase. According to Cook (1980), survival of individual legumes is influenced
by its ability to withstand competition from the existing vegetation and to tolerate other
environmental stress. Disturbing the soil during planting suppressed the growth of nonlegume species while promoting the emergence of indigenous legume species, but was
apparent that competition would occur as soon as the native vegetation began to emerge
resulting in species such as E. ellipticum dieing off within the system.
Despite Crotalaria species having low population density in the indifallow system, they
had superior survival ability compared with E. ellipticum, suggesting that Crotalaria
species were stronger competitors in the mixtures. Comparable biomass accumulations
amounting to 9 t ha-1 for Crotalaria ochroleuca, C. grahaminia and C. incana have been
reported in western Kenya (Ojiem et al., 1998). The decrease in population and abundance
during the second season for annual legume species such as C. ochroleuca, C. pallida and
C. cylindrostachys could be attributed to their growth traits and poor emergence.
Crotalaria cylindrostachys and C. ochroleuca are non-sprouting and have high seed
dormancy which resulted in a reduction in species abundance and population in the second
year. Similar results have been reported in Sesbania sesban improved fallows where a
62
negative biomass accumulation was observed after fallow termination (Mafongoya et al.,
2006). The significant increase in population of C. cylindrostachys at Mudange might be
from the existing soil seed banks. Crotalaria cylindrostachys had previously been reported
to contribute about 25% of the total legume biomass under natural system in Chinyika
area (Mapfumo et al., 2005). Apart from seed dormancy and species growth traits, the
decrease of legume species population might be associated with increased N availability
(Nezomba, 2009) leading to non-legume species dominance. Application of N in pasture
was also reported to result in selective invasion of non-legume species as it affects nodule
formation and inhibits N2-fixation of legumes (Heichel, 1985; Guretzky et al., 2005).
Indigofera arrecta, E. ellipticum and C. laburnifolia were able to persist into the second
season by respouting from standing stems while T. radicans, T. longipes, T. purpurea and
N. wightii regenerated from rootstocks.
This resulted in little change in species
composition in the second season. The die back of most annuals then provided a window
for most biennials and perennials apparently exhibiting the suffretex habit to persist into
the system. These resprouting species also lead to maintenance of indigenous species
diversity under indifallow systems, showing the potential of indifallow systems than
improved fallow of Sesbania sesban, Tephosia vogellii, Tephosia candida and Cajanus
cajan which depend on fresh supply of seedling or seed (Mafongoya et al., 2006).
5.5.2 Influence of indigenous legume and non-legume species in cropping systems.
Indigenous legume species were able to suppress non-leguminous species within the
system as previously reported in sole and mixed legume species fallows (Anoka et al.,
1991; Niang et al., 2002). Some problem weeds such as Eleusine indica, Acanthospermum
63
hispdum and Cyperus esculentus that exist in most cropping system disappeared under
indifallows. The increase in populations of some non-legume species in natural fallows
also confirmed the ability of indigenous legume species to suppress weeds. Crotalaria
pallida and C. ochroleuca which attained peak biomass at six months (Tauro et al., 2007)
suppressed non-legume species during the establishment. Furthermore, resprouting species
such as C. laburnifolia, I. arrecta, Tephrosia longipes, and T. purpurea further reduced
non-legume species diversity in the second season.
At the beginning of the indifallow, the soils were N-depleted and legumes were more
competitive than non-legume species because of their ability to fix nitrogen. According to
Barrios et al., (2000) natural and agricultural systems respond simultaneously to
degradation and regeneration processes through natural succession. Similarly, the peak
biomass productivity and increase in abundance of non-legumes which eclipsed most
annual legumes in the second season of indifallow establishment might be associated with
N release from decomposing legume residues. The dominance of grass, Cynodon dactylon
in the system was associated to the soil N availability and its growth traits. Similarly,
nitrogen addition in mixed species grassland has been reported to change species
composition and increase the dominance of Cynodon dactylon with an overall loss in
species richness (Humphrey, 1977; Mamolos et al., 1995). According to Horowitz (1973),
C. dactylon was considered as a problem weed due to its open growth pattern and lack of
seasonality. Depsite C. dactylon being termed a problem weed, it could be a useful
indicator species for fertility restoration under indifallow systems. According to Mairura
et al., (2007), tapping such knowledge may be used as a cheaper alternative by resourceconstrained farmers since laboratory procedures are time consuming and costly.
64
5.6 Conclusions
Changes in indigenous legumes species population, composition and abundance within the
indifallow system were influenced by species emergence, competitiveness and growth
traits. High emergence of Eriosema ellipticum ensured high species population of > 42
plants m-2 but contributed insignificant biomass compared to Crotalaria pallida and C.
orcholeuca which had a low species population (< 23 plants m-2). Low survival rate of E.
ellipticum and the disappearance of C. pallida, C. orcholeuca and C. cylindrostachys
contributed to decrease in species populations into the second season. Re-sprouting
species such as Indigofera arrecta, E. ellipticum and C. laburnifolia, Tephrosia radicans,
T. longipes, T. purpurea and N. wightii maintained species diversity under the indifallow
system. The most appropriate legume species would br those with high survival,
persistence and at the same time contributing more biomass into the system. Introduction
of indigenous legume species on nutrient depleted soils contributed to changes in nonlegume species composition with well adapted species such as Cynodon dactylon
becoming dominant resulting from legume competition effects and changes in soil
characteristics, particularly increased soil N where indigenous legumes once dominated.
This study suggest that establishment of these species in degraded soils can influence
changes in non-legume composition from legume competition and resultant increase of N
in the system while the changes in legumes population was influenced by species
emergence, competitiveness and growth trait.
65
CHAPTER 6
INFLUENCE OF PHOSPHORUS IN INDIFALLOWS MANAGED FOR
RESTORING FERTILITY OF DEGRADED SOILS ON SMALLHOLDER
FARMS1
6.1 Abstract
Developing alternative soil fertility management options is vital for increasing food
security in many parts of Africa where soil productivity is constrained by nitrogen (N) and
phosphorus (P) deficiencies. This study investigated the P response of indigenous legumes
as organic nutrient sources in smallholder farming systems of Zimbabwe. Single super
phosphate (SSP) and Dorowa phosphate rock (DPR) were broadcast and incorporated to
20 cm depth at 26 kg P ha-1 before mixtures of indigenous legume species were broadcast
on the surface at 120 seeds m-2 species-1 on disturbed soil. The study was conducted during
2005/06 and 2006/07 rainfall season under low (450-650 mm yr-1) to high (>800 mm yr-1)
rainfall conditions. Indigenous legumes responded significantly (P<0.05) better to SSP as
a source of P than DPR. Application of SSP or DPR had no significant (Tobs<1.96) effect
on legume species composition and establishment patterns but SSP significantly increased
biomass productivity by between 20-60%. Single super phosphate influenced (Tobs>1.96)
the composition and biomass yield of non-legume species. Abundance of non-legume
common broad-leaves weeds Commelina benghalensis, Richardia scabra, and Solanum
aculeastrum was reduced by between 5-20% while that of Cynodon dactylon a grass,
increased with SSP application, leading to a significant (P<0.05) increase in overall nonlegume biomass. Indigenous legumes derived 61 – 90% of their N from biological
nitrogen fixation with amounts fixed ranging between 5 – 120 kg N ha-1 under semi-arid
conditions and between 78 – 267 kg N ha-1 under high rainfall. There was selective
response to SSP application among indigenous legume species, with only Crotalaria
cylindrostachys, C. ochroleuca and C. pallida increasing biomass accumulation thereby
increasing the amount of N fixed by 2, 27 and 51 kg N ha-1, respectively, hence raising the
total N fixed within the system. However, non-legumes species responded to SSP
application and recovered more P apparently due to their rooting system. The combined
effect of P and N from decomposing legume residues resulted in high biomass
productivity of non-legumes in the second season thereby out-competing legumes.
Application of SSP is critical to increase the N2-fixation of indigenous legume species.
This chapter has been published as follows:
1
Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2008. Responses of indigenous legumes
species to mineral P application for soil fertility restoration in smallholder farming systems of Zimbabwe.
African Network for Soil Biology and Fertility (Afnet)-SOFECSA International Symposium titled
“Innovations as Key to the Green Revolution in Africa: Exploring the Scientific Facts”, 17-21 September,
2007, Arusha, Tanzania.
66
6.2 Introduction
Apart from mineral fertilizers, legumes have been known to make net N additions to
cropping systems through biological nitrogen-fixation (BNF) and increase maize yields by
two to three times (Niang et al., 2002). However, the productivity and nitrogen fixation of
most legumes in smallholder farming system of Zimbabwe are constrained by poor
nutrients availability, especially phosphorus (P) (Giller, 2001). Biomass generated from
such soils is less than 2 t ha-1 and yielding < 60 kg N ha-1 required to make significant
impact on productivity (Gilbert, 1998). Supply of P would boost early root development
resulting in take up of other soil nutrients for plant development. Most of the locally
available resources for smallholder farmers such as cattle manure, crop residues, termitaria
and leaf litter supply no more than 15 kg N t-1 and 1.5 kg P t-1 (Giller et al., 1997;
Mapfumo and Giller, 2001). When these resources are applied in modest amounts of 3-5 t
ha-1 dry matter, N supplied is sufficient to produce a 2 tonne maize crop but their N: P
ratio is too high to supply required P for legumes production (Vanlauwe and Giller, 2006).
In this case, addressing such soil fertility challenges for smallholder farmers of Southern
Africa requires external supply of P or new technical innovations.
Recent findings have identified indigenous legume species capable of establishing on soil
with low clay content (< 100 g kg-1 soil), low soil organic carbon (~4 g C kg-1 soil) and <
5 mg kg-1 available P, (Mapfumo et al., 2005). Such soils are often abandoned by farmers
due to frequent failure of crops such as maize and soyabean. One way of increasing the N
economy of most depleted soils could be through harnessing of these indigenous legumes
into cropping system. It is likely that the problem of P deficiencies would not be
adequately addressed. Unlike N, biological mechanisms to improve the availability of P
67
are limited thus application of P containing fertilizers is vital during the restoration of
degraded soils. Despite the potential of these indigenous legumes, there is paucity of
information on the influence of P application on population dynamics, N2-fixation and
biomass productivity.
In the wake of increasing cost of P fertilizers, a lot of interest has emerged for the use of
phosphate rock available in many countries in Sub-Saharan Africa (Akande et al., 1998).
According to Dhliwayo (1999), utilization of Dorowa phosphate rock (DPR) satisfies both
agronomic and economic farmers’ needs, particularly the resource-constrained. Since the
indifallow is a low cost technology in terms of establishment and management, use of
DPR would integrate well into this concept. Indifallows (indigenous fallows) involves
establishing various naturalized indigenous legume species in dense stands in order to
improve soil N of natural fallows. Encouraging results have been reported from
composting of rock phosphate with agricultural waste (Akande et al., 2005) resulting in
increased solubility of phosphate rock but little has been done in utilization of phosphate
rock in improved fallows.
Most studies have focused on the influence of P on biomass productivity in improved
fallows and yield of the subsequent crop (Chikowo et al., 1999; Akande et al, 2005), with
little attention on population dynamics and N2-fixation of the legumes in the system. This
study investigated the population dynamics and P response of indigenous legumes as
organic nutrient sources in smallholder farming systems of Zimbabwe. This chapter
focused on the following specific objectives:
1. To evaluate the differential response of leguminous and non-leguminous plant species
to P application under indifallow and natural systems.
68
2. To determine the effect of Single Super Phosphate (SSP) and Dorowa Phosphate Rock
(DPR) on N2-fixation by indigenous legume species and biomass productivity of
indifallow systems.
6.3.0 Materials and methods
6.3.1 Evaluating species response to phosphorus application under indifallow and
natural fallow systems
The following treatments were effected to investigate response of species under indifallow
and natural fallow systems to P application:
i.
Indifallow
ii.
Indifallow + Single Super Phosphate (SSP)
iii.
Indifallow + Dorowa Phosphate Rock (DPR)
iv.
Natural fallow
v.
Natural fallow + SSP
vi.
Natural fallow + DPR
The experiment was carried out at Domboshawa and Makoholi experimental stations
using both SSP and DPR while at Chikwaka and Chinyika smallholder farming areas SSP
was the only P source used. The plots were established by broadcasting and incorporating
SSP or DPR to a 20 cm depth at 26 kg P ha-1. A rate between 20 and 30 kg P ha-1 has been
used for other grain legumes (Ojiem et al., 2000). Natural fallows were established by
disturbing the land by hand hoeing and then allowing naturalized species to reestablish.
The experiment was laid out in randomized complete block design (RCBD) with three
replicates per treatment. Plots were 36 m2 and blocking was based on soil fertility within
the fields.
69
6.3.2 Determining the effect of P application on species abundance and biomass
productivity
A 0.5 m x 0.5 m quadrat was used to randomly determine species biomass productivity at
3, 6, 15 and 20 months spreading over two rainy seasons as described in Chapter 3
(section 3.4). The species biomass yield was then divided with the total system biomass to
estimate species abundance. After abundance estimation, the dried legume and nonlegumes plant biomass were separately ground in a Wiley Mill to pass through a 1 mm
sieve followed by total P analysis using the colorimetric method (Anderson and Ingram,
1993). Phosphorus uptake by legume and non-legume species was obtained by
multiplying the tissue P concentration by the corresponding biomass. Following individual
species P uptake quantification, system total P uptake was calculated as:
System total P uptake = Average % P of the system * Total system biomass
(1)
6.3.3 Determining the effect of P application on BNF
The N-difference (ND) method, as described in Chapter 3 was used for estimating N2fixation of indigenous legumes between 3 and 6 months after seeding. Appropriate nonleguminous reference plants for the different indigenous legumes species were selected
from natural fallow plots. The tissue N concentration in both legume and selected
reference species were determined using the semi-Kjeldhal digestion method (Anderson
and Ingram, 1993). The accumulated N was obtained by multiplying the tissue N
concentration by the corresponding species biomass quantity. The difference between N
accumulated by the fixing legume and a non-fixing reference plant was considered as the
N derived from fixation (Peoples et al., 1995).
70
6.3.4 Statistical analysis
Data was analyzed by analysis of variance and mean comparisons using GENSTAT
statistical package (GENSTAT, 2005). Mean comparison for legumes and non-legume
productivity were done by t-test. The Shannon-Wiener diversity index (H`) was used to
test for diversity changes and comparisons of system diversity (Guretzky et al., 2005).
Comparisons of system diversity were done at t critical of 1.96 and 774 degrees of
freedom. All the mean comparisons were done at 0.05 % significance levels. All the
biomass is reported on a dry matter basis while species abundance is reported in
percentages of individual plant species per plot.
6.4.0 Results
6.4.1 Influence of P fertilization on species composition in indifallows and natural
fallows
Application of SSP or DPR had no significant effect on legume species composition. The
abundance of C. pallida was increased by ~ 30% at Makoholi and ~ 20% at Domboshawa
with SSP application (Figure 6.1), though there was no significant (P>0.05) increase in
biomass yield.
71
a) Indifallow
Other non-legume
species
6.0%
Digitaria decumbens
16.5%
Digitaria eriantha
5.3%
Crotalaria pallida
54.8%
Indigofera astragalina
0.1%
Tephrosia radicans
0.2%
Crotalaria
ochroleuca
0.5% Crotalaria
cylindrostachys
Eriosema ellipticum
15.9%
0.8%
b) Indifallow + SSP
Other non-legume
species
4.5%
Digitaria decumbens
2.7%
Cynodon dactylon
8.1%
Eriosema ellipticum
1.1%
Crotalaria pallida
83.7%
Figure 6.1 Relative abundance of indigenous legume and non-legume species at 6
months in (a) indifallow and (b) indifallow + SSP systems under low
rainfall (450-650 mm yr-1), Makoholi in the 2005/06 season.
72
At Domboshawa and Chikwaka, there was no significant difference (T (obs) <1.96) in nonlegume species diversity in indifallow system regardless of P application after 3 months of
establishment (Table 6.1). However, application of SSP increased the abundance of
Eleusine indica and Cynodon dactylon by about 10% and 4% in indifallow resulting in
abundance decline of Richardia scabra, and Solanum aculeastrum by about 6% and 8%,
respectively (Figure 6.2). Six months after establishment, natural fallow was significantly
(T (obs) >1.96) high in non-legumes species diversity than the indifallow system. Following
SSP application, both indifallow and natural fallow systems had the highest ShannonWiener index (H’) and evenness (E) indicating highest diversity and greater non-legumes
species evenness, respectively though not significantly different (Table 6.1). Indifallow
with SSP was more diverse than indifallow and indifallow + DPR while there was no
significant difference in non-legumes species composition among natural fallow system
regardless of P source. Consistently, at Chikwaka indifallow + SSP had significantly
(Tobs>1.96) high species diversity than natural fallow + SSP six months after
establishment (Table 6.1). This was due to increased non-legume species diversity under
indifallow while non-legume species diversity under natural fallow was reduced following
SSP application.
At Makoholi, natural fallow had the highest H’ and E than indifallow 3 months after
establishment indicating more diverse and greater species evenness, though not
significantly (Tobs<1.96) different from indifallow system. Application of either SSP or
DPR had no significant (Tobs<1.96) effect on non-legume species diversity at 3 months
after establishment under the indifallow system while application of SSP significantly
(Tobs>1.96) increased species diversity under natural fallow system (Table 6.1). Six
months after establishment, there was significant (Tobs>1.96) difference in non-legumes
73
species diversity under indifallow and natural fallow system. Indifallow had the highest H’
and E indicating a more diverse and greater species evenness. Application of SSP
significantly (Tobs>1.96) increased and reduced species diversity under natural fallow and
indifallow system, respectively. Non-legume species diversity was significantly
(Tobs>1.96) reduced following DPR application under both natural fallow and indifallow
systems (Table 6.1). Species diversity under natural fallow + SSP was significantly
different from indifallow + SSP. Generally, application of SSP increased the abundance of
most non-legume species under indifallow and also led to the appearance of new species
namely Conyza sumatrensis and Ageratum conyzoides.
74
a) Indifallow
Other non-legume
species
14.7%
Commelina
benghalensis
5.1%
Crotalaria pallida
33.6%
Solanum aculeastrum
10.9%
Richardia scabra
12.5%
Eriosema ellipticum
2.4%
Crotalaria cylindrostachys
3.5%
Crotalaria ochroleuca
7.9%
Crotalaria laburnifolia
0.5%
Neonotonia wightii
Eleusine indica
1.7%
Cynodon dactylon
0.0%
4.6%
Indigofera arrecta
2.6%
b) Indifallow + SSP
Other non-legume
species
16.9%
Crotalaria pallida
27.9%
Commelina
benghalensis
4.6%
Solanum aculeastrum
2.9%
Eriosema ellipticum
1.4%
Crotalaria
cylindrostachys
1.0%
Richardia scabra
7.0%
Crotalaria ochroleuca
13.0%
Indigofera arrecta
2.4%
Eleusine indica
11.1%
Cynodon dactylon
8.2%
Macrotyloma daltonii
3.6%
Figure 6.2. Relative abundance of indigenous legume and non-legume species at 3
months in (a) indifallow (b) indifallow + SSP systems under high rainfall
(>800 mm yr-1), Domboshawa in the 2005/06 season.
75
Table 6.1. Shannon-Wiener diversity index (H’) and Evenness index (E) for non-legume species at 3 and 6
months after establishment under indifallow and natural fallow system with or without P at
Domboshawa, Makoholi and Chikwaka in the 2005/6 rainfall season.
Treatment
Indifallow
Indifallow + SSP
Indifallow + DPR
Natural fallow
Natural fallow + SSP
Natural fallow + DPR
Domboshawa
3 months
6 months
H’
E
H’
E
1.78
0.62 0.49
0.25
1.63
0.56 1.27
0.55
1.80
0.62 1.08
0.49
2.00
0.69 1.10
0.44
1.79
0.62 1.41
0.51
1.67
0.58 1.12
0.47
Makoholi
3 months
6 months
H’
E
H’
E
1.20
0.67 1.74 0.97
0.83
0.46 1.59 0.77
1.04
0.58 1.06 0.60
1.57
0.71 1.65 0.69
1.84
0.71 1.71 0.88
0.69
0.38 1.58 0.64
nt = not established, H’= Shannon-Wiener diversity index, E= Species eveness index
76
Chikwaka
3 months
6 months
H’
E
H’
E
1.81 0.79 1.69 0.77
1.55 0.74 1.99 0.90
nt
nt
nt
nt
1.91 0.80 2.01 0.87
1.78 0.77 1.88 0.76
nt
nt
nt
nt
In the first growing season, non-legume species had peak biomass at 3 months while
legumes attained maximum yield at 6 months after sowing thereby suppressing most of
the non-legume species. During the second growing season, the abundances of Cyperus
esculentus and R. scabra was reduced possibly due to competition from increased
abundance of C. dactylon. In the second season non-legumes species out-competed
indigenous legumes species as a result of P application (Figure 6.3). Non-legume species
were dominant in the system through-out the second growing season (Figure 6.3).
77
Figure 6.3. Biomass accumulation patterns of leguminous and non-leguminous
species biomass under indifallow + SSP over two seasons at
Domboshawa (>800 mm yr-1), Zimbabwe during 2005/06 and 2006/07
seasons.
6.4.2 Influence of P application on indifallow and natural fallow productivity
At 3 month, P application regardless of source had no significant effect on total legume
biomass productivity under indifallows while there was a significant increase in biomass
under natural fallows. Of the three legume species (Crotalaria cylidrostrachys, C. glauca
78
and C. pisicarpa) which had peak biomass at 3 month, only C. cylidrostrachys responded
to SSP application at Chikwaka. Application of SSP increased C. cylidrostrachys biomass
from 0.2 t ha-1 to 0.4 t ha-1. At 6 months after establishment indifallow biomass
productivity increased significantly between 20% and 60% as a result of SSP application
across sites. Crotalaria pallida significantly responded to SSP application while slow
growing biennials such N. wightii, I. arrecta and E. ellipticum did not respond (Table 6.2).
At both Makoholi and Domboshawa, C. pallida increased yield by 0.2 and 0.3 t ha-1,
respectively with SSP application (Table 6.2). Generally indifallow biomass productivity
(legume and non-legume) increased with the addition of SSP at both sites. Similary six
months after establishment, Crotalaria pallida and C. cylindrostachys increased biomass
by about 1.0 and 0.8 t ha-1 at Chikwaka, respectively, following SSP application.
Response to SSP addition was also evident for C. ochroleuca which accumulated an
additional 0.1 t ha-1 at Chikwaka due to poor initial establishment and 3.1 t ha-1 at
Mudange.
Dorowa phosphate rock showed contrasting effects on biomass productivity across sites.
There was no significant (P>0.05) difference in biomass productivity with application of
SSP or DPR at 6 months after establishment at Domboshawa (Table 6.2). Conversely at
Makoholi, addition of DPR to indifallows resulted in low yield of indigenous legume
species compared with yield in indifallow with SSP over the 6 months period. Crotalaria
pallida and C. cylindrostachys only accumulated 0.6 t ha-1 and 0.9 t ha-1 following DPR
application compared with 7.6 t ha-1 and 0.9 t ha-1 without supply of P (Table 6.2).
Shoot P concentration of the legumes was particularly low, ranging from a mean value of
0.05 % for C. cylindrostachys to 0.09 % for E. ellipticum and C. pallida. In contrast, non79
legume species had slightly higher shoot P concentration of 0.12 %. Regardless of P
source, there was no significant difference (P>0.05) in the total P uptake of the indigenous
legume species (Figure 6.4). In contrast, there was an increase in total P uptake for nonlegume species in both indifallow and natural fallow systems as a result of DPR or SSP
application at Domboshawa (Figure 6.4). Despite the application of P, the accumulated P
ranged from 9 to 11 kg ha-1 and 2 to 7 kg ha-1 at Domboshawa and Makoholi, respectively.
The total P uptake of non-legumes was reduced with DPR application in both indifallow
and natural fallow system at Makoholi a more depleted soil (Figure 6. 4). At Makoholi,
application of SSP increased the total P uptake in indifallow from 4 kg ha-1 to 7 kg ha-1
while the total P uptake indifallow was reduced to <1.5 kg ha-1 with application of DPR
(Figure 6.4).
Table 6.2. Effect of phosphorus application on biomass productivity (t ha-1) of
indigenous legumes at 6 months after establishment under low rainfall
(Makoholi) and high rainfall (Domboshawa), in the 2005-06 season.
Makoholi
Indigenous legume species
Domboshawa
SSP
DPR
No P
SED
SSP
DPR
No P
SED
Crotalaria pallida
7.8
0.6
7.6
2.10
9.2
9.2
8.9
2.70
Crotalaria ochroleuca
np
np
np
-
0.3
4.5
4.0
1.88
Crotalaria cylindrostachys
0.5
0.1
0.9
0.10
0.7
0.07
0.2
0.53
Eriosema ellipticum
0.1
0.1
0.1
0.04
0.04
0.01
0.5
0.27
Indigofera arrecta
ni
ni
ni
-
ni
0.08
0.04
-
0.08
0.2
0.1
0.03
ni
ni
ni
-
Indigofera astragalina
np = not planted during establishment of indifallow; ni = not identified in the indifallow
system
80
18
a) Domboshawa
16
LSD =4.67
14
12
10
8
Total P uptake (kg ha-1)
6
4
2
0
18
b) Makoholi
16
LSD = 2.50
14
12
10
8
6
4
P
SS
+
fa
al
ur
at
N
N
at
ur
N
al
at
fa
ur
ll o
al
ll o
w
w
fa
+
l lo
D
w
P
SS
+
w
llo
fa
di
In
In
di
fa
In
l lo
di
w
fa
+
ll o
D
w
PR
0
PR
2
Treatments
Figure 6.4. Total phosphorus uptake in indifallow and natural fallow biomass after 6
months of establishment under (a) high rainfall conditions at
Domboshawa (750 mm yr-1) and (b) low rainfall conditions at Makoholi
(450-650 mm yr-1) in the 2005-06 season.
81
6.4.3 Effects of P application on N2-fixation of indigenous legumes
Generally, addition of SSP increased N fixation by indigenous legume species across the
rainfall zones, except at Makoholi which was severely depleted (Table 6.3). Addition of
SSP increased the total amount of N fixed at Domboshawa and Chinyika by 43 %,
yielding 120 kg N ha-1 and 81 % yielding 26 kg N ha-1, respectively. The amount of N
fixed at Makoholi following application of SSP decreased by 47 % yielding 63 Kg N ha-1
instead of 90 kg N ha-1. At Domboshawa, there were no significant (P>0.05) differences in
amount of N fixed as a result of SSP or DPR application. In contrast, DPR application
significantly (P<0.05) reduced the amount of N fixed at Makoholi. The greatest increase
in N fixation was at Chikwaka where application of SSP to indifallows led to a 141 %
increase in N fixation (Table 6.3).
Most indigenous legumes significantly (P<0.05) fixed more N after the addition of SSP
(Table 6.3). Addition of SSP significantly increased the amount of N fixed by C.
ochroleuca (18 %) and C. pallida (26 %) (Table 6.4). However, addition of SSP did not
significantly increase the amounts of N fixed by the legume species in cases where overall
biomass accumulation remained poor, suggesting that there were other limiting factors
apart from P.
82
Table 6.3. Total amount of N2-fixed (kg N ha-1) in indifallows plots under different
rainfall conditions at 3 months after establishment in the 2005-06 season.
Site
Treatments
Indifallow
Indifallow
Indifallow
SED
Plus SSP
Plus DPR
Domboshawa
84
120
123
21.0
(>750 mm yr-1)
Makoholi
91
63
11
37
(450-650 mm yr-1)
**Chikwaka
5.4
13
np
7.3
-1
(>750 mm yr )
*Chinyika
14.4
26.1
np
6.5
-1)
(650-750 mm yr
**Nyamayaro, *Mudange, np = not planted during establishment of indifallow
Table 6.4. N2-fixation patterns of different indigenous legume species between 3 and
6 months of growth on poor soils, with and without SSP application under
different rainfall conditions in Zimbabwe
Study site (rainfall zone)/
Indigenous legume species
Chinyika (450-650 mm yr-1)
Crotalaria pallida
Crotalaria ochroleuca
Crotalaria cylindrostachys
Eriosema ellepticum
Indigofera arrecta
Indigofera astragalina
Tephrosia radicans
Chamaecrista mimosoides
*Domboshawa (>750 mm yr-1)
Crotalaria pallida
Crotalaria ochroleuca
Crotalaria cylindrostachys
Eriosema ellepticum
Indigofera arrecta
Chikwaka (>750 mm yr-1)
Crotalaria pallida
Crotalaria ochroleuca
Crotalaria cylindrostachys
Eriosema ellepticum
Indigofera arrecta
N2-fixed (kg N ha-1)
% N from
fixation
(%Ndfa)
Reference plant
Minus P
Plus P
140(23.8)
150(31.5)
7(1.20)
15(1.4)
0.5(0.1)
1.5(0.3)
1.4(0.1)
3.4(2.4)
nd
177(4.0)
9(0.5)
9(0.2)
2(0.7)
5(0.3)
nd
1(0.2)
99
95-97
68-74
93-95
61-78
59-75
99
62-87
Ageratum conyzoides
Leucus martinicensis
Richardia scabra
Ageratum conyzoides
Ageratum conyzoides
Richardia scabra
Ageratum conyzoides
Ageratum conyzoides
193(6.2)
70(2.8)
3(1.05)
11(0.2)
6(0.3)
244(27.7)
76(2.0)
nd
3(0.2)
5(0.5)
99
93-99
66
99
70-84
Ageratum conyzoides
Ageratum conyzoides
Richardia scabra
Ageratum conyzoides
89(10.2)
2.6 (0.3)
5.2 (0.3)
nd
nd
82(14.3)
3.9 (0.9)
8.8 (0.2)
nd
nd
87-95
72-80
84-89
nd
nd
Bidens pilosa
Bidens pilosa
Bidens pilosa
Richardia scabra
* Domboshawa soils have 22 % clay compared with <10 % for Chikwaka; nd = not
determined; Figure in parentheses denotes standard errors
83
6.5.0 Discussion
6.5.1 Species composition as affected by P application in indifallows
The response of non-legume species such as Cynodon dactylon and Eleusine indica to P
application can be attributed to their fibrous rooting systems, which enable their ready
access to soil P (Cadisch et al., 1998). This could account for increased non-legume
species diversity under both indifallow and natural fallow following P addition. However,
the increase in non-legume species diversity under indifallow can also be partly due
increased availabiliity of N as a result of contributions from N2 fixation by the legumes.
Lack of seasonality and open growth pattern of C. dactylon also attributed to the response
to P application while other species where being out-competed in the first season. The
dominance and competitiveness of C. dactylon contributed to the reduction in abundance
of Cyperus esculentus and R. scabra in the indifallow. The combined effect of P and N
from decomposing legume residues could have resulted in high biomass productivity of
non-legumes in the second season. However, at Makoholi lack of response by non-legume
species to P application at seemingly high rate of 26 kg P ha-1 suggests that nutrients other
than P were also limiting. The soils at Makoholi are also highly prone to nutrient leaching
(Mtambanengwe and Mapfumo, 2006), which might have attributed to lack of
responsiveness of non-legume species.
According to Tauro et al., (2007) indigenous legume species composition is governed by
germination and emergence but these results showed that the application of P influenced
species composition in the the system. Changes in species abundance following P
application contributed to changes in species composition. Even though legumes have
84
high P requirement than grasses (Cassman et al., 1981; Giller and Wilson, 1991), most of
indigenous legumes had a slow initial growth rate, resulting in poor early response to SSP
application. The result also showed that most indigenous legume species except C.
cylidrostrachys, C. pallida and C. ochroleuca are adapted to low soil available P.
6.5.2 Influence of P on biomass productivity under indifallow system
Indifallows produced high biomass at Domboshawa (~15 t ha-1) and Chinyika (Chikodzo)
(~7 t ha-1) sites despite these soils having 5 mg kg-1 available P, suggesting that such low
P levels might be adequate to support indigenous legume species establishment and
growth. Furthermore, the particularly low P content (P< 0.09%) of these indigenous
legume species and low total P uptake of indifallow systems confirm that these legumes
utilize low P from the soil and may not respond to P application. Application of either SSP
or DPR at Domboshawa had no significant effect on biomass accumulation. These results
support similar findings by Mapfumo et al. (2005) who identified these species to be
growing in P deficient soils where maize could not yield 1 tonne of grain.
Despite ground phosphate rock being beneficial in acidic soils (Nandi and Haque, 1988),
application of DPR depressed the growth of the legumes species such as C. pallida at
Makoholi. The low supply of P for legume utilization might be attributed to the low
solubility of the material (Dhliwayo, 1999) and short time of fallowing which was
unlikely to beneficiate the DPR. In another study by Muza and Feresu (1998), velvet bean,
sunnhemp and cowpea did not respond to direct application of DPR or Pelleted Phosphate
Rock (PPR). According to Dhliwayo (1999), DPR might need to be beneficiated with
organic materials prior to use in order to release the locked P implying that despite being a
85
cheaper source of P, DPR might not provide the immediate benefits which most farmers
expect. However, DPR still remains a cheap source for improving soil P capita for most
resource-constraint farmers. Therefore there is need to understand the long term effects of
applying DPR to indifallows.
Varied responses among legume species to SSP application can be attributed to their
growth traits and different P requirements. Slow growing biennials species such as I.
arrecta, N. wightii and E. ellipticum did not respond to SSP as they had early die back
compared to Crotalaria pallida and C. ochroleuca. Such a phenology may help to explain
why the effect of SSP on legume biomass productivity was evident at 6 months after
legume establishment. The responsiveness of C. pallida and C. ochroleuca to P
application among the Crotalaria genera provides evidence of different P response even
among species of the same genera. Similarly, Rao et al. (1999) reported that plants differ
in their ability to utilize available forms of P in the soil and large differences exist among
species even of the same genera.
6.5.3 Importance of P in N2-fixation of indigenous legumes
The increases in N2-fixation with application of SSP at Chikwaka suggest that P was a
limiting nutrient to N2-fixation of indigenous legumes. Phosphorus application is a major
requirement for N2-fixation and biomass productivity in both cropping and natural systems
when soil P is low (Giller, 2001). Despite most indigenous legume species adapted to soil
with low available P, application of P increased biomass productivity and N2-fixation of
C. ochroleuca and C. pallida. Indifallows without P will not attain the addition benefits
from responding species suggesting that P application is critical. The poor response in N2-
86
fixation of indigenous legumes following P application at Makoholi suggest that
indigenous legumes may be limited by nutrients other than N, which is consistent with
findings by Grime and Curtis (1976) and Tilman (1982). Similar challenges have been
cited in efforts to enhance productivity of soyabean and other grain legumes under
severely depleted soil in sub-Saharan Africa (Snapp et al., 2002; Giller, 2003). According
to Ritchie and Tilman (1995), other nutrients such as calcium and manganese may be
necessary for optimum performance of wild legumes. To increase fixation of indigenous
legumes on such depleted soils, cattle manure or other organic resources might need to be
applied to improve the supply of bases, micronutrient and reduce soil acidity. Working on
similar depleted soils, Zingore et al. (2007) reported a yield of 44 kg N ha-1 for soyabean
amended with manure compared to 25 kg N ha-1 following SSP application.
6.6 Conclusions
Indigenous legume species responded significantly to SSP as a source of P than DPR.
Phosphorus application to indigenous legume fallows influenced non-legume species
composition but did not affect establishment and composition of the legume species.
There was an apparent low P requirement by indigenous legume species. Different legume
species of the same genera responded differently to P application, exhibiting the advantage
of sowing them in mixtures. Application of P significantly increased total biomass
productivity and N2-fixation of indifallows. The study demonstrates that phosphorus
remains a strategic nutrient input in stimulating restoration of sand soil productivity using
indigenous legume species. However, the realisation of P benefits is apparently limited by
multi-nutrient deficiency in sand soils.
87
CHAPTER 7
OVERALL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS
7.1 Introduction
Much of the previous soil fertility research has targeted legume-based technologies to
increase maize yields with little attention on restoration of nutrient depleted soils. This
study evaluated the establishment and management of indigenous legumes in indifallow
systems, which is a potential integrated soil fertility management (ISFM) option. The data
presented here is from three different agro-ecological regions of Zimbabwe and may apply
to other farming systems in sub-Saharan Africa where such indigenous legume resource
can be found.
7.2 Establishment of indigenous species
This study has demonstrated that mixing indigenous legume species ensured successful
establishment of indifallows for soil fertility amelioration as it provides insurance against
germination or emergence failure and competition. Acid scarification of the seed and
disturbing the soil at planting increased the emergence of indigenous legume species
resulting in successful indifallow establishment. However, there is need to explore
alternative mechanisms that are acceptable to farmers for increasing species emergence
apart from acid scarification, as acid is not readily available in rural outlets. It was evident
that Eriosema ellipticum showed no signs of dormancy while other species had seed
dormancy being expressed. Apart form seed dormancy, lack of viability also affected
species germination in the system resulting in poor stand establishment. The most ideal
planting time for the species is at the onset of the rains when there is adequate moisture for
seed germination.
88
Most legume-based technologies have failed to make an impact due to high establishment
and maintenance cost. However, this new innovation requires no weeding once the species
have established. Since labour shortage is common in communal areas, this might be a
good option for soil fertility management particularly for resource-constrained farmers.
Establishment of the 2-year indifallows is mainly through resprouting species with little
influence of seed from the first season. Disturbing the soil in the second season might
improve species germination due to improved seed soil contact. On the other hand,
collection of indigenous seed from the 1-year indifallows might increase the availability of
the seed for use on other plots and at the same time save time of collecting seed from the
wild.
7.3 Biomass accumulation of indigenous legume species
Mixing various indigenous legume species in the indifallow systems ensured species
contribution to biomass could be spread over time. Crotalaria species such as C. pallida,
C. ochroleuca and C. laburnifolia attained > 5 t ha-1, showing superiority in biomass
productivity in relation to Indigofera and Chamaecrista species, which might be related to
their high N2-fixation capacity. However, most of these species did not persist into the
second season and needed to be complemented. Most biennial species such Neonotonia
wightii, Tephrosia radicans and E. ellipticum only managed to attain about 2 t ha-1, which
might be low amounts but critical contributions in increasing residual benefit associated
with biomass inputs in the system.
Given that most locally derived organic nutrients are often of poor quality, particularly
low in N, synchronization of nutrient release and crop demand is normally poor. In
addition, unavailability and inaccessibility to most nutrients resource limits farmer’s
89
capacity to manipulate the synchronization. The indifallow concept can potentially address
such challenges through the inputs of different quality biomass into the system. A regular
supply of high and low quality organic materials to sandy soils have been shown to
improve the fertility status and improve maize yields (Mtambanengwe and Mapfumo,
2005). An accumulation of 7-15 t ha-1 of biomass without P application and low
requirement of P by indigenous legumes makes them suitable for use in fertility
restoration purposes in most depleted soils of Zimbabwe.
7.4 Population dynamics within indifallow systems
In the first season, Eriosema ellipticum was the most dominant legume species having a
population of 42 plants m-2 while Crotalaria pallida, C. cylindrostachys, C. ochroleuca
and Indigofera arrecta each had populations of < 23 plants m-2. As the season progressed,
there was a significant decrease in the population of E. ellipticum and Neonotonia wightii
due to low survival rates as competition increased. This showed that population of
indigenous legumes species is governed by species germination, emergence, ability to
withstand competion from non-legume spsecies and growth habit. Monitoring of changes
in species composition in the indifallow systems revealed that annuals are abundant in the
first season and could not persist into the second season. On the contrary biennials and
perennials persisted, attaining their maximum biomass in the second season. The
generative capacity of indigenous legume species through resprouting resulted in
maintainance of legume species diversity in the systems. Indigenous legume species
influenced changes in non-legume species diversity. This was indicated by the smothering
effect of the species, suggesting that weeding might not be necessary thereby reducing the
maintenance costs of the indifallow systems.
90
On the other hand, the increase in the N economy of the system resulted in the increase in
abundance of non-leguminous species in the second season. The rapid dominance of
Cynodon dactylon, a grass weed with high nutrient requirements suggests that it took
advantage of increased nutrient levels particularly N. Such competitive dynamics for N
may act as indicators for the restoration process of degraded soils. According to
Promsakha na Sakonnakhon et al. (2005) non-legumes species are likely to immobilize
applied nitrogen due to their high C: N ratio. However, the quality of non-legumes species
under indifallow systems is likely to be high due to accumulation of nitrogen from
decomposing legumes which reduces the C: N ratio and can release nitrogen to subsequent
crop. Overall, the combination of N and organic matter inputs could be vital in addressing
the depleted soils in Sub-Saharan Africa to bring them back into productivity.
7.5 Response of indigenous legumes to P application
The two sources of P (SSP and DPR) varied in quality and their effectiveness within the
soil matrix was probably governed by their solubility and availability to crops. Both
sources had no effect on the abundance and diversity on indigenous legume species in the
systems. It was evident that these indigenous legume species are adapted to low P soil and
most did not respond to P application in term of biomass productivity. The increase in
abundance of non-legume species with SSP application might be linked to the readily
available P from SSP compared to DPR. Despite having no effect on legume species
composition, addition of SSP increased the amount of N fixed by most indigenous legume
species across sites except at Makoholi where deficiencies of other nutrients could have
been limiting N2-fixation.
91
The practicality of P application in indifallows under smallholder farming conditions
poses a challenge given that fertilizers are expensive and limited for most resourceconstrained farmers. However, the P application is critical under indifallow as it increases
the benefit associated with indigenous legume species by increasing biomass productivity
and N2-fixation of selective species. Farmers tend to allocate the limited fertilizers to
cereal crops which might not ensure household food security when the soils are depleted.
Since P is not lost through leaching the farmer will benefit more after the N economy of
the system has improved by indigenous legume species. According to Muza (2003), the
application of P to green manure crops such as sunnhemp and velvet bean almost doubled
the subsequent maize grain yields. Phosphorus application is a pre-request in increasing
the benefits associated with indigenous legumes species and in restoration of abandoned
soils.
Short term benefits of DPR were not evident in the study, which could be attributed to its
low solubility and short time of fallowing. Further research is required to quantify
potential P benefits to subsequent cereal crops and to determine the long term effects of
DPR application in indifallow systems. Apparently, beneficiating the DPR with organic
materials or manure might be the best way to improve the release of the P into the system
(Dhliwayo, 1999). Most of these indigenous legumes had low P uptake from the system
and since P is not very mobile in soils DPR can then contribute to the P recapitalization in
most of depleted soils.
92
7.6 Area for further research
The indifallow systems can potentially address most challenges of resource constrained
farmers in restoring fertility of their abandoned fields. An understanding of indigenous
legume species establishment patterns, population dynamics and P management response
in indifallows is useful in the integration of indifallows into farming systems. However,
there is still a scarcity of information on:
•
The effect of indigenous legume species competition on biomass productivity
and N fixation within the indifallow systems. To understand the nature of
competition among species in the indifallow there is need to study the species
in their sole stands.
•
Factors affecting emergence of indigenous legume seed in the second season,
despite the prolific seed production in the first season.
•
Ways to make indigenous legume seed available to most farmers and at the
same time improving the quality of seed to ensure high emergence during
establishment.
•
The potential P benefits to soil fertility and the subsequent cereal crops.
•
Opportunities for incorporating legumes in land rehabilitation programs.
•
Issues of species invasiveness.
93
References
Abdel-Wahab, A.B., Shaheb, M.S.A., Younis, M.A.M. 2002. Studies on the effect of
salinity, drought stress and soil type on nodule activity of Lablab purpureus (L.)
sweet (kashrangeeg). Journal of Arid Environment, 51: 587-602
Akande, M.O, Aduayi, E.A, Olayiuka, A and Sobula, R.A. 1998. Efficiency of Sokoto
rock phosphate as a fertilizer source for maize production in South Western
Nigeria. Journal of Plant Nutrition. 21: 1339-1353.
Akande, M.O., Adediran, J.A. and Oluwatoyinbo, F.I. 2005. Effects of rock phosphate
amended with poultry manure on soil available p and yield of maize and cowpea.
African Journal of Biotechnology. Vol.4 (5): pp444-448.
Akande, M.O., Oluwatoyinbo, F.I., Adediran, J.A., Buari, K.W. and Yusuf, O.I. 2003. Soil
amendment affects the release of P from rock phosphate and development of okra.
Journal of Vegetable Crop Production. 9 (2): 3-9.
Alteri, M.A. and Liebman, M. 1988. Weed management in agroecosystems: ecological
approaches. CSC press. 331-337.
Anderson, JM. and Ingram, J.S.I. 1993. Tropical Soil Biology and fertility: A Handbook
of methods, second Ed, CAB International, Wallingford, UK.
Anoka, U.A., Akobundu, I.O. and Okonkwo, S.N.C. 1991. Effects of Gliricidia sepium
and Leucaena leucocephala on growth and development of imperata cylindrical.
Agroforestry. 16: 1-12.
Barker, D.J., Moore, J.K. and Guretzky, J.A. 2002. Spatial variation in species richness
under contrasting topographies and grazing regimes. Proceedings of American
Forage Grassland Conference. 11: 222-225
Barrios, E., Bekunda, M., Delve, R., Esilaba, A. and Mowo, J. 2000. Identifying and
classifying local indicators of soil quality. Methodologies for decision making in
natural resource management. CIAT: Cali.
Cadisch, G., de Oliveira, O.C., Cantarutti, R., Carvalho, E. and Urquiaga, S. 1998. The
role of legume quality in soil carbon dynamics in Savannah Ecosystems. In:
Bergstrom, L. and Kirchmann, H. (Eds.). Carbon and Nutrient Dynamics in
Natural and Agricultural Tropical Ecosystems. CAB International, Wallingford,
U.K. pp 51 – 53.
Cadisch, G., Gathumbi, S.M., Ndufa, K.J. and Giller, K.E. 2002. Resource acquisition of
mixed species fallows – competition or complementarity? In: Vanlauwe, B., Diels,
J., Sanginga, N. and Merckx, (Eds.). Integrated Plant Nutrients Management in
Sub-Saharan Africa. CAB, Wallingford, U.K. 143-154.
94
Campbell, B.M., Frost, P., Kirchmann, H. and Swift, M. 1998. A survey of soil fertility
management in small-scale farming systems on Northeastern Zimbabwe, Journal
of Sustainable agriculture. 11: 19-39.
Carter, S.E. and Murwira, H.K. 1995. Spatial variability in soil fertility management and
crop response in Mutoko Communal Area, Zimbabwe, Ambio. 24: 77-84.
Carvalho, M.M.de., Edwards, D.G., Andrew, C.S. and Asher, C.J. 1981. Aluminum
toxicity, nodulation and growth of Stylosanthes species.Agronomy Journal 73:
261-265.
Cassman, F.G., Whitney, A.S. and Fox, R.L. 1981. Phosphorus requirement of cowpea
and soyabeans as affected by mode of N nutrition. Agronomy Journal. 73:17-22.
Chibudu, C., Mombeshora, B.G. and Mavedzenge, B. 1995. A study on the nutrient
cycling in farming systems of the Communal Areas of Zimbabwe (Chivi and
Mangwende), Proceedings of the International Institute for Environment and
development (IIED) Farm Africa Cycling workshop, 26 November to 3 December
1995, Soddo Walayta, Ethopia, London, UK:IIED. 387-409.
Chien, S.H. and Hammond, L.L. 1988. Agronomic evaluation of partially acidulated
phosphate rocks in the tropics. IFDC’s experience. IFDC paper series P-7.
International Fertilizer Development Centre, Muscle Shoals, AL.
Chikowo, R., Mapfumo, P., Nyamugafata, P. and Giller, K.E. 2004. Maize productivity
and mineral N dynamics following different soil fertility management practices on
sandy soil in Zimbabwe. Agriculture Ecosystem Environment. 102: 119-131.
Chikowo, R., Tagwira, F. and Piha, M. 1999. Agronomic effectiveness of poor quality
manure supplemented with phosphate fertilizer on maize and groundnut in maizegroundnut rotation. African Crop Science Journal 7: 383-395.
Chuma, E., Mombeshora, B.G., Murwira, H.K. and Chikuvire, J. 2000. The dynamics of
soil fertility management in communal areas of Zimbabwe. In: Hilhorst, T. and
Muchena, F.M. (Eds.). Nutrients on the move-Soil fertility dynamics in African
Farming systems. International Institute for Environment and Development,
London, 45-63.
Clatworthy, J.N. 1985. Pasture research in Zimbabwe (1964-1984). In: Kategile, J.A. (Ed).
Proceedings of the IDRC/SADGC workshop on pasture improvement research in
eastern and southern Africa, Harare, Zimbabwe, 17-21 September, 1984. 25-58.
Clatworthy, J.N. and Thomas, P.I. 1972. Establishment of Stylosanthes guianensis in
Marondera sandveld. Proceedings of the Grassland Society of Southern Africa. 7:
76-83.
Clements, D.R.., Weise, S.F. and Swanton, C.J. 1994. Integrated weed management and
weed species diversity. Phytoprotection. 75: 1-18.
95
Cook, S.J. 1980. Establishing pasture species in existing swards. A review. Tropical
Grasslands. 14(3): 181.
Dart, P. 1977. Infection and development of leguminous nodules. In: Hardy, R.W.F. and
Silver, W.S (Eds.). A treatise on nitrogen fixation.3. Biology. John Wiley and
Sons, New York, 366-472.
Dhliwayo, D.K.C. 1999. Evaluation of the agronomic potential and effectiveness of
Zimbabwe (Dorowa) phosphate rock – based phosphate fertilizer materials. D.Phil
Thesis. Department of Soil Science and Agricultural Engineering. University of
Zimbabwe, Harare. 248 pp.
Donfack, P., Floret, C.H. and Pontainer, R. 1995. Secondary succession in abandoned
fields of dry tropical Northern Cameroon. Journal of Vegetation Science 6: 499508.
Dzowela, B.H. 1998. Environmental Adaptation and Performance of various agroforestry
tree species in Zimbabwe. Transactions of the Zimbabwe scientific Association. 72
(supplement): 2-9.
Farley, R.A and Fitter, A.H. 1999. The responses of seven co-occurring woodland
herbaceous perennials to localized nutrient-rich patches, Journal of Ecology 87:
849-859.
Fenner, M., and Lee, W.G. 1989. Growth of seedlings of pasture grasses and legume
deprived of single mineral nutrients. Journal of Applied Ecology. 26: 223-232.
Gathumbi, S.M., Cadisch, G. and Giller, K.E. 2004. Improved fallows: effects of species
interaction on growth and productivity in monoculture and mixed stands, Forestry
Ecology and management. 187: 267-280.
Gathumbi, S.M., Ndufa, J.K., Giller, K.E. and Cadisch, G. 2002. Do species mixtures
increase above- and below ground resource capture in woody and herbaceous
tropical legumes? Agronomy Journal. 94:518-526.
Gee, G.W. and Bauder, J.W. 1986. Particle-size analysis. In: A. Klute (Ed.). Methods of
Soil Analysis. Part 1, 2nd Edition, Madison: W1: ASA and SSSA. 383-411.
GENSTAT. 2005. GENSTAT Edition 2 Reference Manual. Laws Agricultural Trust
(Rothamsted Experimental Station). Clarendon Press, Oxford Science
Publications, UK.
Gerner, H. and Harris, G. 1993. The use and supply of fertilizer in Sub-Saharan Africa. In:
Reuler, H.V. and Prins, W. (Eds). The role of plant Nutrients for Sustainable Food
Crop Production in Sub-Saharan Africa. Vereniging van Kunstmest Producenten,
Leidschendam, The Netherlands
96
Ghorbani, R., Seel, W. and Leifert, C. 1999. Effect of environmental factors on
germination and emergence of Amaranthus retroflexus. Weed Science 47: 505510.
Gilbert, R. 1998. Undersowing green manures for soil fertility enhancement in the maizebased cropping systems of Malawi. In: Waddington, S.R., Murwira, H.K.,
Kumwenda, J.D.T., Hikwa, D. and Tagwira, F. (Eds.). The soil fertility network for
maize-based cropping systems in Malawi and Zimbabwe. Proceedings of the
SoilFertNet Results and Planning Workshop held from 7 to 11 July 1997 at Africa
University, Mutare, Zimbabwe. SoilFertNet and CIMMYT-Zimbabwe, 73-80.
Giller, K.E, Cadisch, G., Ehaliotis, C., Adams, E, Sakala, W.D. and Mafongoya, P.L.
1997. Building soil nitrogen capital in Africa. In: Buresh, R.J., Sanchez, P.A. and
Calhoun, F. (Eds.). Replenishing soil fertility in Africa. Soil Science Society of
America Special Publication 51.SSSA and ASA, Madison, W1, 151-192.
Giller, K.E. 2003. Kick-starting legumes. Leisa magazine. December 2003.
Giller, K.E. and Cadisch, G. 1995. Future benefits from biological nitrogen fixation: An
ecological approach to agriculture. Plant and Soil. 174: 255-277.
Giller, K.E. and Mapfumo, P. 2002, Biological nitrogen fixation, forms and regulation
factors. In: Lal, R. (Ed). Encyclopedia of soil science. Marcel Dekker, New York,
107-109.
Giller, K.E. and Wilson, K.F. 1991. Nitrogen fixation in tropical cropping system,
Redwood Press, United Kingdom 313pp.
Giller, K.E. 2001. Nitrogen fixation in tropical cropping system, second edition, CAB
International, Wallingford
Gorham, E., Vitousek, P.H. and Reiners, W.A. 1979. The regulation of chemical budget
over the course of terrestrial ecosystem succession, Ann, Rev, Ecol, Syst, 10: 5384.
Grant, P.J. 1979. Mechanical scarification of Stylosanthes guianensis cv Oxley seed.
Proceedings of the Grassland Society of Southern Africa 14: 137-141.
Grant, P.M. 1981. The fertilization of sandy soils in peasant agriculture in Zimbabwe,
Zimbabwe Agriculture Journal 78: 169-175.
Grime, J.P. and Curtis, A.V. 1976. The interaction of drought and mineral nutrient stress
in calcareous grassland. Journal of Ecology. 64: 976-998.
Guretzky, J.A., Moore, J.K., Brummer, C. and Burras, C.L. 2005. Species diversity and
functional composition of pastures that vary in landscape position and grazing
management. Crop Science Society of America. 45: 282-289.
97
Guretzky, J.A., Moore, J.K., Knaap, A.D. and Brummer, E.C. 2004. Emergence and
survival of legume seeded into pasture varying in landscape position. Crop
Science. 44: 227-233.
Hartzler, B. 2000. Weed population dynamics, department of Agronomy, IOWA State
University.
Heichel, G.H., Vance, C.P., Barnes, D.K. and Henjura, K.I. 1985. Dinitrogen fixation, N
and dry matter distribution during a year stands of birdsfoot trefoil and red clover.
Crop Science. 25: 101-105.
Hikwa, D., Murata, F., Tagwira, F., Chiduza, C., Murwira, H., Muza, L. and Waddington
S.R. 1998. Performance of green manure legumes on exhausted soils in Northern
Zimbabwe: A soil fertility Network trial. In: Waddington, S.R, Murwira, H.K,
Kumwenda, J.D.T, Hikwa, D. and Tagwira, F. (Eds.). The soil fertility Network
for Maize-based cropping system in Malawi and Zimbabwe, 81-84.
Hopper, D.U. and Vitousek, P.M. 1998. Effects of plant composition and diversity on
nutrient cycling. Ecological Monogram 68: 121-149.
Horowitz, M.1973. Spatial growth of Sorghum halepense. Weed Research 13:pp200-208.
Horst, W.J., Kamh, M., Jibrin, J.M. and Chide, V.O. 2001. Agronomic measures for
increasing P availability to crops, Plant and Soil 237: 211-223.
Humphrey, R. 1977. Arizona range grasses, their description, forage value, and
management. University of Arizona Press, Tucson, Arizona. 159pp.
International Rules for Seed Testing Edition, 2007. Adapted at the ordinary meeting 2006,
Glattbrugg/Zurich, Switzerland. The International Seed Testing Association
(ISTA), 5: 1 -50.
Jama, B., Buresh, R.J. and Place, F.M. 1998. Sesbania trees on phosphorus deficient sites:
maize yield and financial benefits. Agronomy Journal, 90:717-726.
Jensen, S. and Cavalieri, A. 1983. Drought tolerance in US maize. Agricultural water
management. 7: 223-236.
Kak, A., Devi, C. and Gupta, T. 2006. Response of seed treatment on seed germination in
wild crotalaria species. In proceedings of the First International Conference on
Indigenous Vegetables and Legumes. December 12–15, 2006, ICRISAT Campus,
Patancheru, Hyderabad, India. 66.
Kamanga, B.C.G. and Shamudzarira, Z. 2001. On-farm legume experimentation to
improve soil fertility in Zimuto Communal Area: Farmer perceptions and feed
back. 7th Eastern and Southern Africa Maize Conference. 11th -15th February. 495507.
98
Kamh, M., Horst, W.J., Amer, F., Mostafa, H. and Maier P. 1999. Mobilization of soil and
fertilizer phosphate by cover crops. Plant and Soil 211:19-27.
Kusekwa, M.L. and Lwoga, A.B. 1985. Establishment and early survival of nine pasture
legumes oversown into natural pastures in Northern Tanzania. In: Hague, I., Jutzi,
S. and Neatie, P.J.H. (Eds.). Potential of forage legumes in farming systems of
sub-Saharan Africa. Proceedings of a workshop held at ILCA, Addis Ababa,
Ethopia. 16th -19th September: 490-504.
Kwesiga, F. and Coe, R. 1994. The effect of short rotation Sesbania sesban planted
fallows on maize yields. Forestry Ecology and Management. 64:199-208.
Loi, A., Cocks, P.S., Howieson, J.G., Carr, S.J. 1999. Hard seededness and the pattern of
softening in Biserrula pelecinus L., Ornithopus compressus L. and Trifolim
subterraneum L. Australian Journal of Agricultural Research. 50: 1073-1081.
Lwoga, A.B. 1983. The potential of forage legumes in the exploitation of grazing
resources in Tanzania. Paper presented at a symposium on the role of biology in
the development and utilization of natural resources in Tanzania, Dar-er-Salaam,
26 - 29 September 1983.
Mackay, W.A., Davis, T.D. and Sankhla, D. 1995. Influence of scarification and
temperature treatments on seed germination of Lupinus havardii. Seed Science and
Technology 23: 815-822.
Maclaurin, A.R. and Grant, P.J. 1985. Research experience of reinforcing veld with
legumes in Zimbabwe. Grassland Society of Zimbabwe Symposium on Forage
Legumes. 26 September 1985.
Mafongoya, P.L, Bationo, A., Kihara, J., Waswa, B.S. 2006. Appropriate technologies to
replenish soil fertility in Southern Africa. Nutrient Cycling Agroecosystems. 76:
137-151.
Mafongoya, P.L. and Nair, P.K.R. 1997. Multipurpose tree prunnings as source of
nitrogen to maize under semi-arid conditions in Zimbabwe: interactions of pruning
quality and time and method of application on nitrogen recovery by maize in two
soil types, Agroforestry systems. 37: 1-14.
Mafongoya, P.L., Giller, K.E. and Palm, C.A. 1998. Decomposition and nitrogen release
patterns of tree prunings and litter. Agroforestry Systems 38: 77-97.
Mairura, F.S., Mugendi, D.N., Mwanje, J.I., Ramisch, J.J., Mbugua, P.A. and Chianu, J.N.
2007. Scientific evaluation of smallholder land use knowledge in central Kenya.
Land degradation and development. Volume 19, Issue 1: 77-90.10.1002/idr.815
Malwanda, A.B., Mughogho, S.K., Sakala, W. and Saka, A.R. 2002. The effect of
phosphorus and sulphur and the yield of subsequent maize in Northern Malawi. In:
Waddington, S.R. (Eds). Grain legumes and green manures for soil fertility in
Southern Africa: Taking stock of progress. Proceedings of the soil fertility network
99
for maize-based cropping systems in Malawi and Zimbabwe. October 8-11, 2002,
Leopard Rock Hotel, Vumba, Zimbabwe. 197-204.
Mamolos, A.P, Veresoglou, D.S. and Barbayiannis, N. 1995. Plant species abundances
and tissue concentration of limiting nutrient in low-nutrient grassland: a test of
competition theory, Journal of Ecology. 83: 485-495.
Mapfumo, P. 1999. Legumes: An under-utilised resource in communal areas, Zimbabwe
Science news. 32: 54-57.
Mapfumo, P. 2000. Potential contribution of legumes to soil fertility management in
smallholder farming systems of Zimbabwe: The case of pigeonpea (Cajanus
cajan{L} millsp), Dphil Thesis Department of soil science and Agricultural
Engineering, University of Zimbabwe, Harare, Zimbabwe, 198pp.
Mapfumo, P. and Giller, K.E. 2001. Soil Fertility Management Strategies and Practices by
Smallholder Farmers in Semi-Arid Areas of Zimbabwe. International Crops
Research Institute for the Semi-Arid Tropics (ICRISAT), 1-43.
Mapfumo, P., Giller, K.E., Mpepereki, S. and Mafangoya, P.L.1999. Dinitrogen fixation
by pigeon pea of different maturity types on granitic sandy soils in Zimbabwe.
Symbiosis. 27: 305-318.
Mapfumo, P., Mtambanengwe, F., Giller, K.E. and Mpepereki, S. 2005. Tapping
indigenous herbaceous legumes for soil fertility management by resource-poor
farmer in Zimbabwe. Agriculture Ecosystem and Environment. 109: 221-233.
Maroyi, A., 2006. Preliminary checklist of introduced and naturalized plants in
Zimbabwe. The Zimbabwe Journal of Botany, Volume 18, Part 2: 177-246.
Maruzane, D. and Zirobwa, M. 2002. Early performance of Acacia Senegal var. Senegal
in Zimbabwe. In: Rao, M.R. and Kwesiga, F.R. (Eds.). Agroforestry impacts on
livelihoods in Southern Africa: Putting Research into Practice, Aventura
Warmbaths, South Africa, 20 -24 May 2002, 101-106.
McNeal, E.O. 1982. Soil pH and lime requirement. In: Page, A.L., Miller, R.H., Keeney,
D.R. (Eds.). Method of soil analysis part 2. Chemical and microbiological
properties, ASA inc. SSSA Inc Publishers, NY. USA, 199-224.
Metcalf, S.J. 2005. Symbiotic nitrogen fixation and establishment of six Montana native
legume species. Masters thesis, Montana State University. 127pp.
Mpeperiki, S. and Makonese, F. 1998. Seasonal rhizobial population fluctuations under
field conditions in Zimbabwean soils. In: Mpepereki, S and F, Makonese (Eds.).
Harnessing Biological Nitrogen Fixation in African Agriculture. Challenges and
opportunities. University of Zimbabwe Publications, Harare, Zimbabwe. 109-115.
100
Mtambanengwe, F. and Kirchmann, H. 1995. Litter from a tropical savanna woodland
(miombo): Chemical composition and C and N mineralization. Soil Biology and
Biochemistry, Elsevier Science Ltd. 1639-1651
Mtambanengwe, F. and Mapfumo. P. 2005. Organic matter management as an underlying
cause for soil fertility gradients on smallholder farms in Zimbabwe. Nutrient
Cycling in Agroecosystems. 73: 227-243.
Mugwira, L.M. 1985. The effects of supplementing communal area manure with lime and
fertilizer in plant growth and nutrient uptake. Zimbabwe Agriculture Journal. 82:
153-160.
Mulder, C.P.H, Jumpponen, A., Hogberg, K. and Huss-Danell, K. 2002. How plant
diversity and legumes affect nitrogen dynamics in experimental grassland
communities, Community ecology, Springer-Verlag. 133: 412-421.
Munns, D.N. 1977. Mineral nutrition and the legume symbiosis. In: Hardy, R.W.F.,
Gibson, A.H (Eds.). A treatise on dinitrogen fixation Section IV. Agronomy and
Ecology, John Wiley and Sons, New York, 353-391.
Murata, M., Waddington, S. and Murwira, H. 2000. Rehabilitation of degraded sandy soils
using annual green manures in Zimbabwe. In: Target (The Newsletter of the Soil
Fertility Research Network for Maize-Based Cropping Systems in Countries of
Southern Africa) 21:4.
Muza, L. 2003. Green manuring in Zimbabwe from 1900 to 2002. In: Waddington, S.R.
(Ed). Grain Legumes and Green Manures for Soil Fertility in Southern Africa:
taking Stock of Progress. Proceedings of a Conference held 8-11th October 2002 at
the Leopard rock Hotel, Vumba, Zimbabwe. SoilFertNet and CIMMYTZimbabwe, Harare, Zimbabwe.103 – 112.
Muza, L. and Mapfumo, P. 1998. Constraints and opportunities for legumes in the fertility
enhancement of sandy soil in Zimbabwe. Proc, of the 6th Eastern and Southern
Regional Maize conference, held in Addis Ababa Ethiopia 21-25 September 1998,
pp214-217.
Muza, L., and Feresu, S. 1998. Effect of phosphorus source and level on green manure
legumes, biomass production and nodulation under green house conditions.
Proceedings of 8th congress of African Association for Biological Nitrogen
Fixation, 23-27 November 1998, University of Cape Town, Rondebosch, South
Africa. 153-154.
Nandi, L.A, Haque, I. 1998. Agronomic effectiveness of Rock phosphate in an Andept of
Ethopia. Comm. Soil Sci. and Plant Anal. 19: 79-90.
Nezomba, H. 2009. Integrating N2-fixing indigenous legumes in smallholder farming
systems of Zimbabwe: The Indifallow concept. Mphil Thesis, Department of soil
science and Agricultural Engineering, University of Zimbabwe.
101
Nhamo, N., Mupangwa, W., Gatsi, T. and Chikazunga, D. 2002. The role of cowpea
(Vigna unguiculata) and other grain legumes in the management of soil fertility in
the smallholder. In Waddington S.R. (Ed.), Grain legumes and green manures for
soil fertility in Southern Africa: Taking stock of progress. Proceedings of the soil
fertility network for maize-based cropping systems in Malawi and Zimbabwe.
October 8-11, 2002, Leopard Rock Hotel, Vumba, Zimbabwe. 119-127.
Niang, A.I. Amodalo, B.A., Wolf, J.de and Gathumbi, S.M. 2002. Species screening for
short-term planted fallows in the highlands of western Kenya. Agroforestry
Systems 56: 145-154.
Nordin, A., Högberg, P. and Nasholim, T., 2001. Soil N form availability and plant N
uptake N uptake along boreal forest productivity gradient Oecologia. 129:125-132.
Nyakanda, C., Mariga, I.K., Murwira, H.K., Nyamugafata, P. and Salimfu, D. 2002. The
status of existing set aside fallows in smallholder farming: A case of two
contrasting agricultural land-use system in Zimbabwe. In: Rao, M.R. and Kwesiga,
F.R. (Eds.). Proceedings of the Agroforestry impacts on livelihoods in Southern
Africa: Putting Research into Practice. May 20-24, 2002, Aventura Warmbaths,
South Africa. 1-10.
Nyamapfene, K.W. 1991. Soils of Zimbabwe, Nehanda publisher, Harare.
Nyathi, P. and Campbell, B.M. 1993. Influence of tree leaf litter on nutrient status of
sandy soil. African Crop science proceedings. 3: 413-419.
Nzuma, J.K., Murwira, H.K. and Mpepereki, S. 1998. Cattle manure options for reducing
nutrient losses: Farmers perceptions and solutions in Mangwende, Zimbabwe. In:
Waddington, S.R., Murwira, S.R., Kumwenda, J.D.T., Hikwa, D. and Tagwira, F.
(Eds.). Soil fertility research for maize-based farming systems in Malawi and
Zimbabwe. Harare, Zimbabwe: Soil Fertility Network for Maize-based Cropping
Systems in Countries of Southern Africa, Centro Internationla De Mejoramiento de
Maiz y Trigo (CIMMYT), 183-190.
O’Hara, G.W. 2001. Nutritional constraints on root nodule bacteria affecting symbiotic
nitrogen fixation, Aust J, Exp, Agric, 417-433.
Oberson, A., Friesen, D.K., Rao, I.M., Buihler, S. and Frossard, E. 2001. Phosphorous
status and cycling in native savanna and improved pastures on an acid low-P
Colombian soil, Nutrient Cycling in Agroecosystems. 55: 77-88.
Ojiem, J.O., Mureith, J.G., Maobe, S.N., Okwuosa, A. and Nekesa, C. 2000. Effect of
phosphorus on nodulation and dry matter yield of legumes green manure species in
the humid highlands of Kenya. Proceedings of the 2nd scientific conference of the
soil management and legume research network projects, June 2000, Mombasa,
Kenya.
Ojiem, J.O., Ransom, J.K., Odongo, O.M. and. Okwuossa, E.A. 1998. Chemical
characterization and agronomical evaluation of potential green manure species in
102
western Kenya. Proceedings of the 6th, Eastern and Southern Africa CIMMYT
maize conference.
Okalebo, J.R., Gathua, K.W. and Woomer, P.L. 2002. Laboratory methods of soil and
plant analysis: A working Manual. The Second Edition. TSBF-CIAT and
SACRED Africa, Nairobi, Kenya, 128 pp.
Olsen, S.R., Cole, C.V., Watanabe, F.S. and Dean, L.A. 1954. Estimation of available
phosphorus in water and sodium bicarbonate extracts from soils. Soil Science
America Proceedings. 29:677-678.
Palm, C.A. 1995. Contribution of agroforestry trees to nutrient requirements of
intercropped plants. Agroforestry Systems. 30: 105-124.
Peoples, M.B., Faizah, B., Rerkasen and Herridge, D.H. 1989. Methods of evaluating
Nitrogen fixation by nodulated legumes in the field. Australian Centre for
International Agriculture Research, Canberra.
Piano, E. 1993. Developing annual pasture legumes for Mediterranean farming systems.
In: Christiansen, S., Materon, L., Falcinelli, M. and Cocks, P. (Eds.). Introducing
ley farming to the Mediterranean basin, Icarda. 208.
Promsakha S., Sakonnaknon Na., Cadisch G., Toomsan B., Vityakon P., Limpinuntana V.,
Jogloy S. and Patanothai A. 2006. Weeds-friend or foe? The role of weed
composition on stover nutrient recycling efficiency. Field Crops Research. 97:
238-247.
Rao, I.M., Friessen, D.K. and Hoorst, W.J. 1999. Opportunities for variety germpalsm
selection to influence phosphorus acquisition from low phosphorus soils, Agro
forestry Forum 9(4):13-16.
Rebafka, F.P., Ndunguru, B.J. and Marschner, H., 1993. Crop residues application
increase nitrogen fixation and dry matter production in groundnuts (Archis
hypogaea L.) grown in acid soil. Plant and Soil. 150: 213-222.
Ritchie, M.E. and Tilman, M. 1995. Response to Herbivores and nutrients during
succession on nitrogen-poor soil, Ecology, Volume 76, No. 8:2648-2655.
Rodrigues, E.A., Aguiar, I.B. and Sader, R. 1990. Treatments to break seed dormancy in
Cassia sp. Revista Brasileira de Sementes. 12: 17-27.
Russ, L. 1993. Preliminary observations on seed coat hardness in six pasture legumes.
Developing annual pasture legumes for Mediterranean farming systems. In:
Christiansen, S., Materon, L., Falcinelli, M. and Cocks, P. (Eds.). Introducing ley
farming to the Mediterranean basin, ICARDA, 289-293.
Rutunga, V., Karaja, N.K., Gachene, C.K.K. and Palm, C. 1999. Biomass production and
nutrient accumulation by Tephrosia vogelii (Hemsley) A, Gray and Tithonia
103
diversifolia Hook F. fallows during the six-month growth period at Maseno,
Western Kenya. Biotechnol. Agron. Soc. Environ 3 (4): 237-246.
Sanchez, P.A., Shepherd, K.D., Soule, M.J., Place, F.M., Buresh, R.J., Izac, A.M.N.,
okwunye, A.U., Kwesiga, F.R., Ndiritu, C.G. and Woomer, P.L.1997. Soil fertility
replenishment in Africa: an investment in natural resource capital. In: Buresh. R.J.,
Sanchez, P.A. and Calhoun F. (Eds.). Replenishment soil fertility in Africa. Soil
Science Society of America Special Publication 51.SSSA and ASA, Madison, W1,
1-46.
Sileshi, G., Mafongoya, P.L. and Kwesiga, F. 2002. Performance of Sesbania sesban
infested by the defoliating beetle Mesoplatys ochroptera in Zambia. Agroforestry
systems. 56: 117-184.
Smaling, E.M.A., Nandwa, S.M. and Janssen. B.H. 1997. Soil fertility in Africa is at stake.
In: Buresh, R.J and Sanchez, P.A. (Eds.). Replenishing soil fertility in Africa. Soil
Science Society of America (SSSA) Special Publication 51. Madison, Wisconsin,
USA. pp 47-61.
Snapp S.S., Rohrbach D.D., Simtowe F. and Freeman H.A. 2002. Sustainable soil
management options for Malawi: can smallholder farmers grow more legumes?
Agric. Ecosyst. Environ. 91: 159-174.doi. 10.1016/s0167-8809(01)00238-9
Snapp, S.S., Mafongoya, P.L.and Waddington, S. 1998. Organic matter technologies for
integrated nutrient management in smallholder cropping systems of southern
Africa. Agriculture, Ecosystem and Environment.71: 185-200.
Stanley, W.G. and Montagnini, F. 1999. Biomass and nutrient accumulation in pure and
mixed plantations of indigenous tree species grown on poor soils in the humid
tropics of Costa Rica. Forestry Ecological Management. 113: 91–103.
Sulas, L., Franca, A. and Caredda, A. 2000. Persistence and regeneration mechanisms in
forage legumes, Proceedings of the 10th meeting of Mediterranean Sub-Network of
the FAO_CIHEAM inter-regional cooperative research and development network
on pasture and fodder crops, Sassori, Italy, 4-9 April.1-8.
Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2007. Field emergence
and establishment of indigenous N2-fixing legumes for soil fertility restoration.
African Crop Science Conference Proceedings 8: 1929-1935.
Thomas, G.W. 1982. Methods of Soil Analysis, Part 2, Chemical and Microbiological
Properties-Agronomy Monograph No. 9 Second Edition.
Thomas, G.W. 1996. Soil pH and Soil Acidity. In: Sparks, D.L, Page, A.L., Helmke,
P.A., Loeppert, R.H., Soltanpour, P.N., Tabatabai, M.A., Johnston, C.T. and M.E.
Summer (Eds.). Methods of Soil Analysis, Part 3, Chemical Methods. Soil Science
Society of America and American Society of Agronomy. Madison, Wisconsin,
USA. 475-490.
104
Tilman, D. 1982. Resource competition and community structure, Princeton, Princeton
University, Press.
Tilman, D. 1987. Secondary succession and the pattern of plant dominance along
experimental nitrogen gradient, Ecological monographs 57: 189-214.
Vance, C.P. 2001. Symbiotic nitrogen fixation and phosphorus acquisition: plant nutrition
in a world of declining renewable resources. Plant Physiology. 127: 390-397.
Vanlauwe, B. and Giller, K.E. 2006. Popular myths around soil fertility management in
sub-Saharan Africa. Agriculture, Ecosystems and Environment. 116: 34-46.
Venge, R. 2003, Population dynamics of selected non-cultivated indigenous legume
species in Zimbabwe, BSc dissertation, University of Zimbabwe. 30pp
Wilson, S.D. and Tilman, D. 1993. Plant competition and resource availability in response
to disturbance and fertilisation. Ecology. 74: 599-611.
World Reference Base for Soils, FAO/ISRIC/ISSS. 1998. World Soil Resources Report
No. 84. Food and Agriculture Organization, Rome
Zagal D., Rydberg I. and Martensson A. 2001. Carbon distribution and variations in
nitrogen-uptake between catch crop species in pot experiments. Soil Biology
Biochemistry 33, 523-532.
Zingore S., Murwira H.K, Delve R.J, Giller K.E. 2007. Variable grain legume yields,
responses to phosphorus and rotational effects on maize across soil fertility
gradients on African smallholder farms. Nutrient Cycling Agroecosystem 101:296305. doi 10.10007/s10705-007-9117-3
105
Appendix A. Rainfall data for 2005/06 and 2006/07 season
600
a) Domboshawa
2005/06
2006/07
b) Makoholi
2005/06
2006/07
500
400
300
100
0
600
500
400
300
200
100
Months
106
Ap
ril
M
ar
ch
0
O
ct
ob
er
N
ov
em
be
r
D
ec
em
be
r
Ja
nu
ar
y
Fe
br
ua
ry
Rainfall (mm)
200
600
c) Chinyika
2005/06
2006/07
d) Chikwaka
2005/06
2006/07
500
400
300
200
0
600
500
400
300
200
100
Months
107
Ap
ril
M
ar
ch
er
N
ov
em
be
r
D
ec
em
be
r
Ja
nu
ar
y
Fe
br
ua
ry
0
O
ct
ob
Rainfall (mm)
100
Appendix B. List of publications from this thesis
1. Mapfumo, P., Nezomba, H., Tauro, T.P. and Mtambanengwe, F. 2006. N2-fixing
indigenous legumes: From the wild to farmers’ fields. Abstract presented at the 1st
International conference on indigenous vegetables and legumes, ICRISAT
Campus, Patancheru, Hyderabad, India, 12-15 December 2006.
2. Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2007. Field
emergence and establishment of indigenous N2-fixing legumes for soil fertility
restoration. African Crop Science Conference Proceedings 8: 1929-1935.
3. Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2007. Responses
of indigenous legumes species to mineral P application for soil fertility restoration
in smallholder farming systems of Zimbabwe. Abstract published in the
proceedings of African Network for Soil Biology and Fertility (Afnet) -SOFECSA
International Symposium titled “Innovations as Key to the Green Revolution in
Africa: Exploring the Scientific Facts”, 17-21 September, 2007, Arusha, Tanzania.
4. Tauro, T.P., Nezomba, H., Mtambanengwe, F. and Mapfumo, P. 2009.
Germination, field establishment patterns and nitrogen fixation of indigenous
legumes on nutrient-depleted soils, Symbiosis, 48:92-101.
108
Appendix C. Statistical analysis outputs
(i)Domboshawa
establishment
biomass
C.pallida
productivity
at
6
months
F pr.
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
5.23
2.62
0.24
Block.*Units* stratum
Treatment
Residual
2
4
0.16
43.85
0.08
10.96
0.01
Total
8
49.25
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
9.1
1
8.9
2
9.2
3
9.2
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
1.91
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
2.70
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
7.51
109
0.993
after
(ii) Domboshawa C.ochroleuca biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
6.472
3.236
0.61
Block.*Units* stratum
Treatment
Residual
2
4
24.763
21.251
12.382
5.313
2.33
Total
8
52.486
***** Tables of means *****
Variate: Biomass
Grand mean 2.64
Treatment
1
4.02
2
0.31
3
3.59
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
1.331
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
1.882
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
5.225
110
F pr.
0.213
(iii) Domboshawa E.ellipticum biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
0.3192
0.1596
1.50
Block.*Units* stratum
Treatment
Residual
2
4
0.3084
0.4258
0.1542
0.1065
1.45
Total
8
1.0534
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
0.19
1
0.45
2
0.04
3
0.09
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
0.188
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
0.266
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
0.740
111
F pr.
0.336
(iv) Domboshawa C.cylindrostachys biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
0.1609
0.0804
0.19
Block.*Units* stratum
Treatment
Residual
2
4
0.3592
1.6553
0.1796
0.4138
0.43
Total
8
2.1754
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
0.39
1
0.16
2
0.65
3
0.36
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
0.371
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
0.525
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
1.458
112
F pr.
0.675
(v) Makoholi C.pallida biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
7.334
3.667
2.01
Block.*Units* stratum
Treatment
Residual
2
4
23.200
7.315
11.600
1.829
6.34
Total
8
37.850
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
2.71
1
3.81
2
3.88
3
0.44
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
0.781
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
1.104
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
3.066
113
F pr.
0.057
(vi) Makoholi I.astragalina biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
0.0020792
0.0010396
2.93
Block.*Units* stratum
Treatment
Residual
2
4
0.0077415
0.0014180
0.0038707
0.0003545
10.92
Total
8
0.0112387
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
0.0397
1
0.0038
2
0.0397
3
0.0756
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
0.01087
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
0.01537
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
0.04268
114
F pr.
0.024
(vii) Makoholi E.ellipticum biomass productivity at 6 months after
establishment
TREATMENTS
1. Indifallow
2. Indifallow + SSP
3. Indifallow + RP
***** Analysis of variance *****
Variate: Biomass
Source of variation
d.f.
s.s.
m.s.
v.r.
Block stratum
2
0.0015423
0.0007712
1.13
Block.*Units* stratum
Treatment
Residual
2
4
0.0006344
0.0027247
0.0003172
0.0006812
0.47
Total
8
0.0049014
***** Tables of means *****
Variate: Biomass
Grand mean
Treatment
0.0527
1
0.0501
2
0.0641
3
0.0440
*** Standard errors of means ***
Table
rep.
d.f.
e.s.e.
Treatment
3
4
0.01507
*** Standard errors of differences of means ***
Table
rep.
d.f.
s.e.d.
Treatment
3
4
0.02131
*** Least significant differences of means (5% level) ***
Table
rep.
d.f.
l.s.d.
Treatment
3
4
0.05916
115
F pr.
0.658
Tonny Phirilani Tauro