INDIGENOUS
KNOWLEDGE AND EFFECTS OF
.
,
INTEGRATED SOIL FERTILITY MANAGEMENT ON
GROWTH, GRAIN YIELD AND RHIZOBIA GENETICS OF
SELECTED COWPEA VARIETIES I(
JACINT A MALIAIKIMITI
Reg. No. 184/7687/00
A Thesis Submitted in Partial Fulfillment of the Requirement for the Award of the
Degree of Doctor of Philosophy in the School of Pure and Applied Sciences of
Kenyatta University
Department of Plant and Microbial Sciences
December 2008
©Tl'lr.intl'l
Malia Kimiti
Kamiti, Jacinta Malia
Indigenous know/edge
and effects of
111111111111111
2009/336643
u
E
LI HARY1'
DECLARATION
This thesis is my original work and has not been presented for a degree in any other
university or any other award
JACINTA MALIA KIMITI
Reg No. 184/7687/000
oR
d--\ ~
Signature
DECLARATIO
Date
BY SUPERVISORS
We confirm that the work reported in this thesis was carried out by the candidate
under our supervision
Dr. Gitonga N. M.
Department of Plant and Microbial Sciences, Kenyatta University
P.O Box 43844, Nairobi
11 6 J 0 i
Signature
D~te
JwcJI
I
Dr. Odee D. W.
KEFRI, P.O. Box 20412-00200 Nairobi.
,
Signature
Date
Dr. Vanlauwe B.
TSBF·CIAT, Clo ICRAF P.O Box 30677, Nairobi.
~~
Signature
~/~/~1
Date
11
DEDICATION
This thesis is dedicated to Theresia Mumbua
III
ACKNOWLEDGEME
TS
I am grateful for financial support from my employer, Kenya Forestry Research
Institute (KEFRI) without which this study would have been impossible. I also thank
KEFRI for granting me a study leave so as to undertake this study. I greatly appreciate
financial support from The Directorate of Personnel Management (DPM) that paid for
my tuition fees at Kenyatta University. I am indebted to AfNet (TSBF/CIAT) through
Desert Margins Project (DMP) for the financial support that helped me to start off my
PhD. research activities. I am very grateful to Dr. Andre' Bationo who ensured that I
got some seed money from AfNet to start off my research work. I am grateful for the
assistance from DMP Secretariat of KARI during data collection. Further, am grateful
to the management of the KEFRI Muguga Regional Research Centre for their tireless
support with transport to travel to the research sites. Biotechnology
laboratory staff
greatly assisted me and it is worthwhile mentioning the support accorded me both in
the laboratory and in the field by Dr. David W. Odee, James Otieno, John Ochieng
and John Gicheru. I am grateful to the KEFRI taxonomist, Mr. Francis Gachathi who
gave botanical names to plant samples that I collected during my research activities. I
am grateful to my supervisory committee for providing the guidance needed to make
this study a reality. In addition, I am indeed grateful to Kenyatta University for
registering me as their PhD. Student. Finally, I am grateful to my family for prayers
and support throughout my study period. To crown it all I thank the Almighty God for
doing everything well for me, may my God take all the glory in Jesus' great name.
Amen.
IV
TABLE OF CO TE TS
DECLARATION
i
DEDICATION
ii
ACKNOWLEDGEMENTS
iii
TABLE OF CO TE TS
iv
ABSTRACT
vii
LIST OF FIGURES
viii
LIST OF TABLES
.ix
LIST OF APPENDICES
xi
LIST OF ABBREVIATIONS
xii
CHAPTER ONE
1
1.
1
GENERAL INTRODUCTION
1.1
Research problem
3
1.2
Justification
4
1.3
Research questions
4
1.4
Study hypotheses
5
1.5
Research objectives
5
CHAPTER TWO
8
2.
8
LITERATURE REVIEW
2.1
Participatory rural appraisal (PRA) and questionnaire techniques
8
2.2
Integrated soil fertility management (ISFM)
9
2.3
Cowpea (Vigna unguiculata)
10
2.4
Rhizobia
11
CHAPTER THREE
18
3.
18
GENERAL MATERIALS A D METHODS
v
3.1
Reconnaissance
3.2
Cowpea screening and selection
18
3.3
School demonstration
19
3.4
On-farm
3.5
surveys, community
meetings and on-farm
urveys
18
plots
trials
19
itrogen fixation
19
3.6
Rhizobia populations
3.7
Determination
in soil samples
19
of rhizobia diversity
20
CHAPTER FOUR
4.
FARMER
21
AWARE
AVAILABILITY
ESS 0
SOIL FERTILITY
STATUS
A D FOOD
IN MAKUENI DISTRICT
21
4.1
Introduction
21
4.2
Materials
23
4.3
Results
24
4.4
Discussion
31
and Methods
CHAPTER FIVE
AREA
5.
35
UNDER
GRAIN
LEGUMES
AND
PROBLEMS
FACED
BY
FARMERS IN LEGUME PRODUCTIO
35
5.1
Introduction
35
5.2
Materials
5.3
Results
41
5.4
Discussion
46
and Methods
CHAPTER SIX
SCREE
6.
6.1
ING
.40
50
EW COWPEA VARIETIES FOR DR YLA DS OF EASTER
KE YA
50
Introduction
50
VI
6.2
Materials and Methods
52
6.3
Results and Discussion
54
CHAPTER SEVEN
7.
68
EFFECTS OF ISFM ON NODULATION, GROWTH AND GRAIN YIELD OF
SELECTED COWPEA VARIETIES
68
7.1
Introduction
68
7.2
Materials and Methods
71
7.3
Results
73
7.4
Discussion
81
CHAPTER EIGHT
8.
NITROGEN
86
FIXATION,
POPULATION
AND DIVERSITY
RHIZOBIA UNDER ISFM
86
8.1
Introduction
86
8.2
Materials and Methods
89
8.3
Results
94
8.4
Discussion
101
CHAPTER NINE
9.
OF COWPEA
CONCLUSIONS
104
AND RECOMMENDATIONS
104
9.1
Conclusions
104
9.2
Recommendations
105
LITERATURE CITED
106
APPENDICES
131
Vll
ABSTRACT
The main objectives of this study were; (1) To find out whether farmers in Makueni
District were aware of soil fertility status in their farms and annual food availability,
(2) To find out the proportions of cultivated areas under grain legume production and
the problems faced by farmers in grain legume production, (3) To screen and select
high yielding cowpea varieties for dryland Makueni District, (4) To determine the
effects of integrated soil fertility management (ISFM) on nodulation, growth and
grain yield of selected cowpea varieties, and (5) To determine the effects of ISFM on
nitrogen fixation, indigenous soil rhizobia populations and rhizobia diversity. Farmer
participatory meetings were used to establish whether farmers recognized soil fertility
as a problem in legume production. Results obtained revealed that farmers in the
selected sites recognized soil fertility as a problem and included it in the list of
general problems affecting them. Participating farmers indicated that only 2% of the
cultivated farms in the study sites had fertile soils. To document grain yields, area
under legume cultivation and problems faced by farmers in grain legume production,
a structured questionnaire was used to collect information from farmers. Results
obtained showed that grain yields ranged from 30 kg/ha to 416 kg/ha and area under
legume cultivation from 48% to 92%. Problems faced by farmers in legume
production included low soil fertility, inadequate farm inputs, weeds, pests and
diseases. To select pioneer cowpea varieties, 34 cowpea varieties were selected and screened
for two seasons at Kiboko Dryland Research Station. Some of the cowpea parameters
assessed included pod length, plant biomass, grain yield and weights of 100 seeds. From the
screening studies, nine cowpea varieties were selected for on-farm trials. To determine the
effects of integrated soil fertility management (ISFM) on nodulation growth and
grain yield of selected cowpea varieties, on-farm trials were established at two sites.
The nine cowpea varieties that had been selected during the screening studies were
planted in the trials. Treatments applied included a control, farmyard manure at 2.5
t/ha, phosphorus as triple superphosphate (TSP) (P20S, 0:46:0) at 15 kg/ha and a
combination of both manure and TSP at the singly applied rates. Data collected
included nodule and shoot biomass, and grain yields. Results obtained revealed that
treatment application enhanced nodule and shoot biomass, and grain yields. Nitrogen
fixation was estimated using J sN natural abundance method whi le rhizobia
populations were determined using most probable number (MPN) experiment.
Rhizobia diversity was determined using culture characterization and direct PCRRFLP of the 16S-23S rRNA intergenic spacer region (IOR) of rhizobia genome.
Results of nitrogen fixation showed that 46-53% nitrogen (N) was fixed at a wetter
site, while no N fixation took place at a drier site. Results from rhizobia population
assessment revealed population counts of 4.89x I02 to 2.0x 104 cells/gram of soil with
lower rhizobia counts at planting relative to the harvesting time while high rhizobia
counts were recorded in amended soils relative to the controls. Further, restriction of
eighteen rhizobia isolates from cowpea nodules with MspI restriction endonuclease
revealed four rhizobia IOS groups.
Vlll
LIST OF FIGURES
Figure 1.1
Map of Makueni District showing sub-locations with study sites
Figure 4.1
Food availability calendars of Mbitini and Nguu Divisions
30
Figure 8.1
Amount of nitrogen fixed (%) at Kavuthu during the long rains
94
Figure 8.2
15N
Figure 8.3
15N values for site by variety during the long rains
95
Figure 8.4
Nodule biomass of nodules recovered from MPN experiment..
97
Figure 8.5
Shoot biomass of cowpea plants harvested from MP
98
Figure 8.6
Four PeR-amplified
in sites 1 and 2 during the long rains
16S-23S rRNA
6
95
IGS patterns
experiment..
obtained
restriction of rhizobia strains with MpsI restriction endonuclease
after
99
IX
LIST OF TABLES
Table 4.1
Population densities and number of households at Kavuthu, Matiku,
Yikivumbu and Ndunguni Sub-locations in 1999 and 2003
Table 4.2
General problems faced by farmers in the selected sites of Makueni
District.
Table 4.3
22
,
25
Soil characterization and fertility status as described by farmers in
selected sites of Makueni District..
26
Table 4.4
Soil fertility management across farmer wealth classes
28
Table 4.5
Crop preference in Nguu and Mbitini Divisions of Makueni
District
29
Table 4.6
Household sources of income and expenditure in the study sites
30
Table 5.1
Area under grain legume production in selected sub-locations
.41
Table 5.2
Average legume yields (kg/ha) in selected sub-locations
.42
Table 5.3
Soil fertility status of farms in the selected sites of Makueni
District.
42
Table 5.4
Farmers (%) using inputs to enhance soil fertility
Table 5.5
Sources of animal manure used by farmers
43
Table 5.6
Farmers (%) that reported common weeds in their farms
44
Table 5.7
Farmers (%) that reported pests and diseases in their farms
44
Table 5.8
Level of formal education (%) of household heads
45
Table 5.9
Food availability and source (%) during long dry spells
45
Table 6.1
Pod characters assessed during the short rains
57
Table 6.2
Pod characters assessed during the long rains
58
Table 6.3
Variety characters assessed during the short rains
60
Table 6.4
Variety characteristics assessed during the long rains
61
.43
x
Table 6.5
Character ranking of varieties selected for on-farm trials
Table 7.1
Chemical characteristics of soils collected at on-farm trial sites and
animal manure applied to the on-farm trial..
Table 7.2
72
Nodule biomass (mg/plant) at 50% flowering during the long
rains
Table 7.3
67
74
Nodule biomass (mg/plant) at 50% flowering during the short
rains
75
Table 7.4
Shoot dry weight (kg/ha) at crop maturity during the long rains
77
Table 7.5
Shoot dry weight (Kg/ha) at crop maturity during the short
rains
78
Table 7.6
Grain yield at Kavuthu during the long rains
79
Table 7.7
Grain yield (Kg/ha) during the short rains
80
Table 8.1
Rhizobia estimates (rhizobia cells/gram of soil) in soils collected at the
beginning and at the end oflong rain season
97
Table 8.2
Rhizobia isolate characterization using culture
99
Table 8.3
Fragment sizes in base pairs (bp) obtained after rhizobia
restriction
Table 8.4
100
Strain group, rRNA IGS pattern, cowpea varieties and treatments used
in the PCR-RFLP analysis of 16S-23S rDNA IGS
100
Xl
LIST OF APPENDICES
Appendix 1
Maps of Yikivumbu, Ndunguni, Kavuthu and Matiku Sub-locations
drawn by farmers during farmer meetings
131
Appendix 2
Questionnaire
135
Appendix 3
Experimental design (ISFM trial)
139
Appendix 4
Rhizobia cultures isolated from cowpea plant nodules
140
Xll
LIST OF ABBREVIATIONS
AfNet
The African Network for Soil Biology and Fertility
ASALs
Arid and semi arid lands
CIAT
International Centre for Tropical Agriculture
cm
Centimeters
Eds.
Editors
g
Grammes
ISFM
Integrated soil fertility management
kg/ha
Kilogrammes per hectare
Kgha-l yr'
Kilogrammes per hectare per year
Km
Kilometer
Km2
Square kilometer
Litres
M.a.s.l
Metres above sea level
mg
Milligrammes
mg/ml
Microgrammes per millilitre
mg/plant
Milligrams per plant
mgl'
Milligrammes per litre
ml
Millilitres
mM
Micromolar
mm
Millimetres
s
Seconds
TSBF
Tropical Soil Biology and Fertility
USA
United States of America
CHAPTER
1.
ONE
GENERAL INTRODUCTIO
Soil nutrient depletion of arable lands is a major constraint to crop production in most
parts of Africa. It results from continuous cropping with little or no inputs to replenish
soil fertility, removal of crop residues to feed animals, overgrazing
between the
cropping seasons and soil erosion (Sanchez et al., 1997; Smaling et al., 1997;
Stoorvogel et al., 1993). Soil nutrient depletion lowers the returns to agricultural
investment, reduces food security through low crop yields, increases food prices and
reduces government revenue through reduced taxes collected on agricultural goods
(Sanchez et al., 1997).
Nutrient depletion is not uniform and varies with soil characteristics,
and is usually
greater in sandy soils, although total nutrient loss is greater in clay soils (Swift et al.,
1994). In addition,
management
of an individual
farm directly affects nutrient
depletion rates of that farm (Sanchez et al., 1997) and therefore nutrient depletion is
not uniform across cultivated fields.
Over the history of farming, farmers have used inputs to replenish soil fertility and
animal manure has been used over a long time in tropical Africa to replenish soil
fertility and enhance crop production (Dennison, 1961; Giller et al., 1997; Hartley,
1937; McCown et al., 1992). For example, in the arid and semi-arid lands (ASALs) of
eastern Kenya, smallholder
farmers use animal manure to improve soil fertility
(Mathuva et al., 1996; Probert et al., 1995). However, animal manure, especially
cattle manure, which is commonly used by farmers, is of low nutrient content and the
KE VAT A IJNIVFR.~frY
I H1RllAV}
2
amounts available are not usually adequate (Giller et al., 1997; Kimani and Lekasi,
2004; Probert et al., 1995). In addition, chemical composition of manure is highly
varied depending
on the diet fed to the livestock, collection and storage methods
(Gill er et al., 1997; Powell, 1986). For example in Kenya nitrogen (N) content of
cattle manure ranges from 0.2%-2.2%, while phosphorus
(P) content ranges from
0.08% to 0.95% (Kimani and Lekasi, 2004).
Inorganic fertilizers can supplement low nutrient animal manures but their prices
especially to smallholder farmers on staple food crops are uneconomical (lama et al.,
1997). For example in the arid and semi arid lands (ASALs) of eastern Kenya,
adoption of inorganic fertilizers is constrained by high costs, low farm returns, and
lack of right fertilizers to the resource poor farmers (Ikombo, 1984).
Organic inputs in form of plant residues, especially legumes offer a cheap alternative
source of N to expensive
fertilizers for smallholder
estimated that legume-Rhizobium
farmers. For example, it is
symbioses in tropical soils can fix up to almost 600
kg ha-1 yr' N (Gibson et al., 1982). However,
unlike N, which may be fixed
biologically, P can only be supplied to crops through addition of manures, compost,
rock phosphates
and inorganic fertilizers (Ayaga and Brookes, 2005). In addition,
organic materials are low in P (Ayaga and Brookes, 2005; Giller et al., 1997; Palm
1995; Sanchez
et al., 1997) and they should be supplemented
with inorganic
fertilizers (lama et al., 2000; Vanlauwe et al., 2001). For example, Janssen (1994) and
Smaling and Braun (1996), found that application
of combined inorganic P with
farmyard manure significantly increased yields compared to when either of the inputs
were applied as single dressing.
3
Soil nutrient depletion of arable lands is a major constrain to crop production in the
drylands of Eastern Kenya (Siderius and Muchena,
1977). Need for increased food
production has led to intensive agriculture, associated with ecological disturbance and
soil fertility decline (Sande et al., 2004), which has resulted to reduced crop, yields.
Soil productivity can be improved by use of inorganic fertilizers but the fertilizers are
expensive
and have negative effects on environment.
Organic amendments,
animal
manure, crop residues and compost are used but are poor in nutrients and not adequate
(Prober et al., 1995). Combinations
of organic residues and inorganic fertilizers have
been found to increase crop yields (Nziguheba et al., 2004, Ojiem et al., 2004 and
Okalebo
et al., 2004).
combination
Therefore
the key to soil fertility
improvement
is the
between organic amendments and mineral fertilizers. The combination of
organic and inorganic fertilizers in a management system is termed as integrated soil
fertility management (ISFM).
1.1
Research problem
Soils in eastern Kenya drylands are low in nutrients especially phosphorus
nitrogen
(N) but the magnitude
Makueni
District
knowledge.
of the soil infertility
has not been well documented
using local farmers'
In addition, there exists limited information
cow pea varieties in comparison
distribution
on selection
(P) and
especially
in
indigenous
of improved
to the locally grown ones in the drylands. Further
studies on ISFM effects on N fixation and rhizobia diversity
drylands of eastern Kenya have not been reported.
on cowpea
in the
4
1.2
Justification
There is need to document
the magnitude of soil fertility status using indigenous
knowledge from the local farmers so as to understand the actual soil infertility status
at the farm level. Cowpea is the most important legume in the eastern Kenya drylands
and there is therefore need to select high yielding varieties for the drylands so as to
boost cowpea production
in the region. In addition, it will be necessary to carry out
ISFM studies on cowpea so as to understand the best nutrient management
options for
the crop under dryland conditions. There is also need to carry out studies on nitrogen
fixation and rhizobia diversity of cowpea so as to document the effects of ISFM on
these parameters.
1.3
Research questions
1) Are the farmers at Makueni District aware of soil fertility status in their farms
and annual food availability?
2) What proportions of the cultivated areas do grain legumes occupy and what
are the problems faced by farmers in grain legume production?
3) Are there better legume varieties, especially of cowpea, which can be selected
for field trials at dryland Makueni District?
4) What effects would integrated soil fertility management (lSFM) have on
nodule formation, growth and yield of selected cowpea varieties?
5) How would ISFM affect nitrogen fixation, indigenous rhizobia populations
and diversity?
5
1.4
Study hypotheses
1) Farmers in Makueni District are not aware of soil fertility status in their farms
and annual food availability.
2) Grain legumes dot not occupy large proportions of cultivated areas in Makueni
District and farmers do not face any problems in grain legume production in
the district.
3) Locally grown cowpea varieties cannot perform as well as the improved
cowpea varieties in dryland Makueni District.
4) Integrated soil fertility management (ISFM) has no effect on nodulation,
growth and grain yield of selected cowpea varieties.
5) Integrated soil fertility management has no effect on nitrogen fixation and,
indigenous rhizobia population and diversity in the drylands ofMakueni
District.
1.5
Research
objectives
1) To find out whether farmers in Makueni District were aware of soil fertility
status in their farms and annual food availability.
2) To find out the proportions of cultivated areas under grain legumes production
and the problems faced by farmers in grain legume production.
3) To screen and select high yielding cow pea varieties
for dryland Makueni
District.
4) To determine the effects of integrated soil fertility management
(ISFM) on
nodulation, growth and grain yield of selected cowpea varieties.
5) To determine
the effects
of ISFM on nitrogen
rhizobia populations and rhizobia diversity.
fixation,
indigenous
soil
6
This study was implemented
(Figure
1.1). Objectives
namely, Yikivumbu
Mbitini Division.
111
the ASALs of eastern Kenya in Makueni
one and two were
and Ndunguni
Objective
111
implemented
in Nguu Division
three was implemented
District
in four sub-locations,
and Kavuthu
at Kiboko
Station (not shown in the map) while the rest of the objectives
and Matiku in
Dryland
Research
were implemented
using field trials that were established at Kavuthu and Ndunguni Sub-locations.
I~
I
I
:;.,
Legend
•
Study sites
•
Market centres
RO<1dclasses
~~
'b
o
~ivl slon bound.ry
Location
boundary
a sue-recanons
sub-tce atton
with study situ
II
boundary
~
_.
Figure 1.1
Map ofMakueni
The study is represented
conclusions
.__ ...._.~_ ~_~_~_4~
District showing sub-locations
8 Kilomete:
J
with study sites
in chapters four, five, six, seven and eight. A summary of
and recommendations
of the study is also included as chapter nine. In
chapter four, results from farmer participatory
meetings, that were held in four sub-
locations in Makueni District are documented.
Chapter five documents
results from
7
farm surveys using a structured
questionnaire.
The surveys were carried out in the
four sub-locations where farmer participatory meetings had been previously held.
Chapter six documents
uguiculata
Agricultural
commonly
cultivated
results obtained from on-station screening of cowpea (Vigna
(L.) Walp) varieties
Research
grown
Institute
legume
at Kiboko, a dryland
(KARI).
after common
In Kenya,
research
station of Kenya
cow pea is the second most
beans (Phaseolus
vulgaris
for food (leaves, young pods and grain), while secondary
fodder and soil fertility improvement.
(L.) and is
uses include
It has high protein content with mean crude
protein of leaves, grain and crop residues estimated at 32-34%, 23-35% and 11-12%,
respectively
(Imungi and Porter, 1983). Cowpea is mostly grown in the ASALs and
about 85% is cultivated
in eastern province (Muli and Saha, 2000). Chapter seven
documents" the effects of integrated
formation,
biomass production
soil fertility management
and grain yields of nine selected cowpea varieties
under on-farm conditions in contrasting rainfall conditions
chapter eight documents
(ISFM) on nodule
over two seasons. Lastly,
the effects of ISFM on nitrogen fixation and, soil rhizobia
populations and diversity in the ISFM trial plots.
8
CHAPTER
2.
2.1
Participatory
2.1.1
PRA
Participatory
LITERATURE
rural appraisal
rural appraisal
REVIEW
(PRA) and questionnaire
was developed
including Rapid Rural Appraisal
developed
TWO
from several
(RRA) and Agroecosystem
in Britain by the Institute for Development
Institute for Environment
and development
Participatory
and Development
institutions
(Conway
techniques
participatory
Analysis,
which were
studies and the International
(IIED) and several agricultural
and McCracken,
National
Environment
Secretariat
and Clark
University.
approach has developed and expanded to other countries (Bronson,
aim of PRA is to change the attitude of development
view participation
own problems
about rural
efforts of the
Since
then the
1995). The main
partners and change agents to
as an end rather than a means. Participatory
attempts to make communities
in
1988).
The term PRA was coined in Kenya in early 1986 through collaborative
Kenya
1983).
techniques
that it shortens the process of data collecting and analysing information
(Conway and McCracken,
research
1988; Chambers,
rural appraisal differs from the previous data collecting
communities
methods
Rural Appraisal
directly involved in and responsible for assessing their
and arriving at a consensus
of action that needs to be taken. The
assumption
of PRA is that unless people feel a project is theirs, there will be no
commitment
and the outside experts must consult on the community needs since they
do not know enough on the community
participation
discussions,
is ensured
by gathering
visual aids, and observations
in question. Under PRA approach,
data in group discussions,
(K WAP, 1993).
farmer
using informal
9
Farmer participation
has been used in many development
projects in the sub-Saharan
Africa. For example, in Southern Malawi it was used to determine farmer preferences
for bean varieties by combining
ranking and scoring by farmers (Abeyasekera
2004), while in Kenya farmer participation was used in crop prioritization
et al.,
(Onyango
et al., (2000) and Okoko and Makwaro (2000) to establish the most preferred crop by
selected communities.
2.1.2
Questionnaire
In this technique structured questions targeting particular subjects are developed. Data
is collected through visits to targeted communities or institutions. The disadvantage of
this technique is that it is expensive and time consuming (PRA Programme
University, 1995) because of the time taken to administer questionnaire
Egerton
and to analyse
the information collected.
2.2
Continuous
Integrated soil fertility management (ISFM)
cropping,
removal
of crop residues to feed animals
and overgrazing
between cropping seasons with little or no external inputs have reduced the productive
capacity of arable
lands and thus threaten the sustainability
systems in Sub-Saharan
of food production
Africa (Sanchez et al., 1997; Stoorvogel
Kenya, decline in crop yields has been a major problem-facing
et al., 1993). In
smallholder farmers
(Mathuva et aI., 1996). This is attributed to the high costs of inputs that make the use
of inorganic
fertilizers
on staple food crops uneconomical
for most smallholder
farmers (Jama et al., 1997). Use of organic inputs as an external
source of soil
nutrients is a logical cheap alternative to expensive fertilizers to smallholder farmers.
However, organic
inputs are low in nutrient concentration
compared
to inorganic
10
fertilizers (Sanchez et al., 1997). Despite the fact that organic inputs are low in
nutrients (Gill er et al., 1997), cattle manure is an integral component of soil fertility
management, and manure application is one of the most commonly used and effective
way of soil fertility improvement
for crop production in Africa (Dennison, 1961;
Hartley, 1937). For example, in the semi-arid areas of eastern Kenya where nitrogen
and phosphorus limit crop production, manure has been used to enhance soil fertility
and crop production (Gibberd, 1995; Ikombo, 1984; Kihanda et al., 2004). Recent
research has shown that combination of organic and inorganic inputs enhances crop
production and reduces cost of inorganic fertilizers ((Nziguheba et al., 2004; Ojiem et
al., 2004; Okalebo et al., 2004). The combination of organic and inorganic inputs is
termed integrated soil fertility management (ISFM).
Studies on integrated nutrient management are common. For example, in Tanzania,
Tithonia
diversifolia
superphosphate
(Hems!.)
improved
A. Gray and Mijingu
phosphorus
phosphate
rock or triple
content and increased maize yields (Ikera,
2006). Further in the coastal lowlands of Kenya, a combination of inorganic fertilizers
and manure showed that half of the recommended rates of organic manure and half of
the inorganic fertilizers in combination had same effect on maize yields as the full
recommended inputs when not combined (Saha and Muli, 2000).
2.3
Cow pea (Vigna unguiculatay
Cowpea (Vigna unguiculata L. Walp.) is an annual or bi-annual grain cereal legume
commonly referred to as cowpeas, southern pea, black- eyed pea, Crowder pea, lubia,
niebe, coupe or frijole. It is native to Africa, Asia and the Middle East. It is grown in
the Savanna regions of the tropics and subtropics. It is widely grown in Africa, Latin
11
America,
cowpea
Southeast
is grown
commonly
Asia and Southern United States (Davis et al., 1991). Most
in west and central African
countries.
In Kenya,
cowpea
is
grown in the arid and semi-arid areas (ASALs) of mostly eastern Kenya.
Cowpea value lies in its high protein content, its ability to tolerate drought and fix
atmospheric N, which allows it to grow in and improve poor soils (Bressani, 1985).
Cowpea is used for food, fodder and as a source of income. Leaves, young pods and
grain are the parts of the plant used for food. The same plant parts are sold to generate
cash for farmers. Cowpea has an added advantage to soil fertility improvement
it is a nitrogen-fixing
through residual
in that
legume and can grow in very poor soils and replenish them
effect on a subsequent
crop. Studies on cowpea are common
in
Africa. For example, in Nigeria, Saidou et aI., (2006) evaluated cowpea tolerance and
response to external application of phosphorus both in the field and in the greenhouse.
In Kenya, Muli and Saha (2000) investigated adaptation
and yield performance
of
cowpea cultivars along the coast of Kenya.
2.4
Rhizobia
Rhizobia are aerobic gram-negative
legumes
have
Bradyrhizobium.
been
grown.
There
bacteria that live freely in soil, especially where
are two types
The term Rhizobium
of rhizobia,
refers to the fast-growing
Rhizobium
and
types whereas
Bradyrhizobium
is used for the slow-growing types (Date and Halliday, 1987). They
form specialised
structures on roots of leguminous
the sites of symbiotic
energy requiring
plants called nodules, which are
nitrogen fixation (Lara et al., 1988). Nitrogen fixation is an
process and utilises phosphorus
1990). To fix one molecule
of nitrogen
in form of ATP (Bockman et al.,
16 ATPs are required.
Thus,
legume
12
association
aspect
with
rhizobia
of symbiosis
number,
nodule
nitrogen
fixation.
2.4.1
contributes
is enhanced
weight,
to high
phosphorus
by phosphorus.
plant dry matter
This
production,
uptake.
includes
Virtually
traits
plant nitrogen
every
such a nodule
concentration
and
Rhizobia diversity
Population
diversity
stability,
is an important
population
be investigated
Ribosomal
aspect
must be as diverse
using
differences
of population
as possible.
genetics.
The microsymbiont
in 16S ribosomal
ribonucleic
RNA genes are the best targets for studying
phylogenetic
of an "evolutionary
clock".
all bacteria,
and more
domains
highly
conserved
variable
diversity
acid (rRNA)
rRNA one can find all the properties
have
For population
can
genes.
In
relationships.
The rRNA present
in
et al.,
(Vandamme
1996).
The
5S
rRNA
(Vandamme
was
the
et al., 1996),
sequence
comparison
sequences
can (Ochman
investigated
used
first
marker
to a lesser
molecule
but
cannot
it is only
provide
and Wilson,
molecules
extent
and
to
be sequenced
120 nucleotides
information
only
partial
sequences
et al., 1997). The 16S sequences
and above
(Young,
1996),
whereas
sequence
change
topology
of the 16S and 23S rRNA
in the 23S rRNA
is faster and it can therefore
of rRNA
gene
dendrograms
has become
bacteria
therefore
as
and most
the
the
larger
extensively
The later has so far been
of some
are reliabie
molecule,
rhizobia
have
been
at the level of genus
the average
for close relationships.
is however,
obligatory
and
accurately
useful
be valuable
numerous
long
are the 16S and 23S rRNAs.
(Tesfaye
Sequencing
as
1987). The most
determined
1995).
for
similar
(Ludwig
and together
with
rate of
The
et al.,
DNA-
13
DNA reassociation
they form the key methods
that have to be used when new
rhizobial species or genes are to be described. Several methods have been used in the
past to assess
composition,
genetic
(2)
diversity.
Isozyme
Restriction
Fragment
PCR-based
fingerprinting
on repetitive
Some of the methods
electrophoresis,
length polymorphism,
elements,
(5) Pulsed-field
hybridization,
gel electrophoresis,
(8) Amplification
fragment
length polymorphism,
(4)
(6)
based
and (9)
of amplified rRNA gene fragments.
DNA-base composition
The unique taxonomic
cytosine content
organisms
DNA-DNA
(1) DNA base
with random or arbitrary primers, (7) Amplification
Restriction fragment length polymorphism
2.4.1.1
(3)
include:
feature used in this method was mole percent guanine plus
(Mol% G+C). However,
the disadvantage
of this method is that
having the same Mol% G+C values are not necessarily
since Mol% G+C does not take into account the linear arrangement
in the DNA (Johnson,
1984). However,
the technique
closely related
of the nucleotides
is still used as part of the
standard description of bacterial taxa (Vandamme et al., 1996)
2.4.1.2
DNA-DNA hybridization
The property of DNA to dissociate and reassociate is used in hybridization
The complementary
strands are separated and then let to associate
another single-stranded
DNA (Johnson,
used are: the hydroxyapatite
nuclease method (Vandamme
reassociation
analysis.
to itself or to
199 I). Some of the reassociation
methods
method, the optical renaturation rates method and the SI
et al., 1996). The hydroxyapatite
method was the first
method used in rhizobia taxonomy (Wedlock and Jarvis, 1986; Crow et
al., 1981; Hollis et al., 1981). In this method, the DNA of the reference organism is
14
labeled with
32p
(in vivo) and let to associate with an excess of unlabelled DNA from
the test organism. Then, the double stranded DNA is separated
DNA with hydroxyapatite
to which the double-stranded
into single-stranded
DNA binds. The single-
stranded DNA can then be eluted from the solution. In the optical renaturation
method,
the DNA
spectrophotometer
is not
labeled
and the reassociation
at 260 nm (Johnson,
difference in the temperature
between the homologous
1991). However,
is measured
the determination
rates
with a
of the
corresponding to half the increase in relative absorbance
and heterozygous hybrids formed under standard conditions
(Vandamme et al., 1996). In the S I nuclease method, the DNA is labeled with tritium
eH). The reassociation
mixture
is treated with single-strand
specific nuclease SI,
which degrades single strand regions including the loops. Double stranded DNA is
collected by precipitation
the association
or by binding to nitrocellulose
is calculated
counter (Johnson,
from the radioactivity
membrane.
The percent of
values obtained by a scintillation
1991). Currently, the SI nuclease method (Nour et 01., 1995; Nour
et al., 1994; Laguerre et 01., 1993) has replaced the hydroxyapatite
used high levels of radioactivity.
The disadvantage
of DNA-DNA
method, which
hybridization
method is that, it is laborious and when rhizobial species are many, only those strains,
which are phenologically
closely related, can be analysed (de Lajudie et al., 1994,
Nour et al., 1994).
2.4.1.3
Restriction
fragment
length polymorphism
(RFLP)
This method is usually used together with southern blot hybridization,
DNA fragments are used as probes (Mason and Williams,
used with plasmids
but the disadvantage
where, specific
1985). RFLP can also be
is that different strains of bacteria do not
always contain or keep their plasmids (Vandamme et al., 1996). RFLP analysis was
15
good typing method before the routine application
rhizobial taxonomy
(Kaijalainen
of polymerase
it has been used to study chromosomally
and Lindstrorn,
1989), especially hybridization
chain reaction. In
encoded polymorphism
patterns of nod and nif
genes which are often plasmid coded (Eardly et al., 1992; Laguerre et al., 1993). The
RFLP analysis
using symbiotic
genes is still very useful (UI1Z and Elkan, 1996;
Paffetti et al., 1996).
2.4.1.4
Pulse-field gel electrophoresis
(PFGE)
In PFGE the D A molecules are forced to change their direction of migration with
modification
of current and their speed of reorientation
varies as a function of their
size. Resolution of DNA molecules of up to 9 Mb can be obtained. Using this method,
rhizobia can be differentiated
at species, biovar and strain level. The PFGE method
has mainly been used to study the genome rearrangements
(Sobral et al., 1991) and in
estimating the genome size (Huber and Selenska-Pobell,
1994; Sobral et al., 1991).
However, PFGE has not been the method of choice in taxonomy,
the methods
based
on amplification
by polymerase
chain
probably because
reaction
are more
convenient.
2.4.1.5
PeR-based
fingerprinting
with random or arbitrary
This method is based on the theory that the distribution
oligonucleotide
primers
of any randomly chosen
sequence will vary within the genome of different individuals. Thus
when a single randomly
chosen sequence
is used as a primer,
the amplification
product of individual organisms will be different (Harrison et aI., 1992). This method
has mainly been used to study genetic diversity within a single rhizobial
species
(Sikora et al., 1997; Paffetti et al., 1996; Dye et al., 1995; Selenska-Pobell
et al.,
16
1995). The main disadvantage
of this method is the low sensitivity to the reaction
conditions (Vos et al., 1995).
2.4.1.6
Amplification
based
011
repetitive
elements
This method is based on the principle that prokaryotic
repetitive
sequences
(rep-Pf.R)
genomes contain interspersed
(Stern et al., 1984; Higgins et al., 1982; Sharples and L1oyd,
1990; Hulton et al., 1991; Martin et al., 1992; Versalovic et al., 1994). PCR-primers
are designed from these elements and they enable the simultaneous
amplification
of
many DNA fragments of different sizes, originating from the sequences lying between
the repetitive
within
elements.
a single
This method has only been used to study the differences
rhizobial
species using primers
of only REP or ERIC element
(Laquerre et al., 1996).
2.4.1.7
Amplification
Amplification
fragment length polymorphism
the fingerprinting
fragment
length polymorphism
(AFLP)
(AFLP) is the newest technique among
methods. In AFLP, the genomic DNA is digested with a rare cutting
and a frequent cutting restriction enzyme. Adapters are then ligated to the ends of the
fragments to generate template DNA for amplification.
In this technique, primers are
used. They consist of three parts: a core sequence, an enzyme-specific
sequence and a
selective
to the adapter
extension
of three
nucleotides,
and
is complementary
sequence and to the adjacent restriction site. The selective extension makes sure that
only a subset of the fragments
fingerprint
is amplified and a specific and highly reproducible
is obtained. The advantage of this method is that no prior knowledge of
nucleotide sequence is needed. From 50 to 100 amplified fragments are separated on
polyacrylamide
gel and visualised
autoradiographically
as one of the primers
is
17
labeled
radioactively
visualization
(Vos et al.,
1995). If a non-radioactive
can be made possible
bradyrhizobial
and mesorhizobial
by silver staining.
AFLP
method
is used,
fingerprinting
of
strains have been found to be highly reproducible
and the results in agreement with rep-PCR and with RFLP analyses of the 16S and
23S rDNA amplified by PCR.
2.4.1.8
Restriction fragment length polymorphism
gene fragments
of amplified rRNA
There are three rRNA molecules of different sizes, 5S, 16S and 23S rRNA. However,
the RFLP analysis
of PCR-amplified
rRNA fragment
is based on the two large
molecules, the 16S rDNA and 23S rDNA. The 16S molecule has also been amplified
with the 16S-23S rDNA intergenic spacer region (IGS). Analysis of PCR-RFLP, also
called amplified
identification
ribosomal
method
DNA-restriction
for bacteria.
analysis, has been found to be a rapid
The procedure
involves
amplification
of PCR
product with universal primers located in the conserved regions of the rRNA genes.
The amplified
enzymes,
fragment
and
reproducible
the
is then digested with several, selected,
profiles
and composed
thus
obtained
are mainly
of 2 to 10 fragments
similarity of the bacteria can be visualised
frequently
species-specific,
(Vandamme
by a dendrogram
cutting
highly
et al., 1996). The
constructed
from the
patterns. The PCR-RFLP analysis of the 1.5 kb 16S rDNA has been the most popular
DNA fingerprinting
identification
method. Studies have shown that, this method is not only a rapid
method but also a simple tool in taxonomy and gives an estimation of
genetic relationships
at species and higher levels (Laguerre et al., 1994).
18
CHAPTER THREE
3.
3.1
GENERAL MATERIALS AND METHODS
Reconnaissance surveys, community meetings and on-farm surveys
A reconnaissance
survey was done in Makueni District with the help of agricultural
extension officers. Two divisions were selected and in each division two sub-locations
were identified
administrators
as study sites. After site identification,
visits were made to the
(chiefs and assistant chiefs) to brief them on the aim of the project.
During the same visits dates for farmer participatory
meetings
were set. During
farmer participatory
meetings, farmers formed discussion groups so as to allow all the
farmers participating
farmers to contribute to the discussions. All discussions were led
by one of the community
from farmer discussions
farm surveys,
members using a checklist provided to each group. Results
were documented on flip charts in the local language. For on-
a questionnaire
was formulated
so as to seek specific
answers to
selected questions from farmers in the selected sites. Individual interviews were done
with household heads on random visits to the homesteads.
3.2
Cowpea screening and selection
Thirty-four cow pea varieties, with 30 varieties from International Institute of Tropical
Agriculture
(lIT A) and 4 varieties from local sources were multiplied and screened at
Kiboko, a Kenya Agricultural
Research Institute
(KARI) centre. The centre is a
dry land research station located about 30 km from the research sites. The cowpea
screening activities
rains of2005.
were implemented
during the short rains of 2004 and the long
After screening, seven cow pea varieties were selected for on-farm trials
after assessment
of some yield parameters such as plant biomass production,
yield, pod length and pod numbers.
grain
19
3.3
School demonstration
School demonstration
plots
trials were established during the short rains of 2005 in two
primary schools, Kavuthu and Ndunguni. All the 34 cowpea varieties were planted
with four treatments (control, manure at 2.5 t/ha, phosphorus
as TSP (P20S: 0:46:0) at
15 kg/ha and munure+ TSP at the singly applied rates) and replicated
From the demonstration
plots, two cowpea varieties
in 3 blocks.
that were recommended
by
farmers were used in on-farm trials (Chapter 7).
3.4
On-farm
trials
Cow pea varieties selected on-station
(Kiboko) and those selected by farmers were
tested using integrated soil fertility management (ISFM) in on-farm trials established
at Kavuthu and Ndunguni.
Treatments that were used in the trials were the same as
those used in the school demonstration
3.5
Nitrogen
Nitrogen
fixation
plots.
fixation
was estimated
using ISN natural
abundance
method.
Selected
cowpea varieties were sent to USA for ISN analysis.
3.6
Rhizobia
populations
in soil samples
Soils for this study were sampled from the integrated soil fertility management
trials.
The soils were collected at the beginning of the trials and during harvesting time in
the first season of the trials. To determine rhizobia populations
in the sampled soils,
most probable number (MPN) method was used. Two cowpea varieties selected from
among the nine varieties used in the ISFM trials were used as test crops. A fivefold
dilution series replicated four times at each dilution level was used. The dilution series
20
ranged from s' to y6. A 1 ml volume from each dilution level was applied directly to
the cowpea root systems using sterile 1.0 ml pipettes. Leonard jars were used and
infection counts were assessed after 28 days. Data sheets were prepared to allow for
the entry of each experimental
unit displaying
either positive
or negative nodule
formation.
3.7
Determination
of rhizobia diversity
3.7.1
Using rhizobia
cultures
Rhizobia cultures were obtained from root nodules of two selected cowpea varieties.
Using culture characteristics
such as colony size, shape, colour and mucus production,
nine rhizobia strain groups were identified. The rhizobia strains were stored in a 15%
glycerol yeast manitol at -70°C for future studies.
3.7.2
Direct polymerase
Dried beads
chain reaction (Pf.R) amplification
(one for each amplification
process)
Ready-to-go
of 16s rRNA
PCR beads (GE
Healthcare illustra™) containing nucleotides and buffers, and combined with primers,
were used to amplify rhizobia genome from fully-grown
amplification,
a programmable
rhizobia culture cells. For
Thermal Controller (PTC-100TM
MJ Research Inc.,
Watertown, MA) was used as described here: Initial denaturation
at 93°C for 2 min;
35 cycles of denaturation
at 72
0c)
(45 s at 93°C), annealing (45 s at 62°C), extension (2 min
and final extension
at 72
-c
for 5 min. Amplified
rRNA products were
subjected to restriction enzyme and separated by horizontal gel electrophoresis
agarose gel, stained with ethidium bromide and photographed
under UV.
in 1%
21
CHAPTER FOUR
4.
FARMER AWARENESS ON SOIL FERTILITY STATUS AND FOOD
AVAILABILITY IN MAKUENI DISTRICT
4.1
Introduction
Makueni district is one of the districts in Eastern Province of Kenya and occupies a
total area of7,966
km", this being 5.16% of Eastern Province which has a total area of
154,354 km", and 2.69% of the total land area of Kenya (581,677 krn '). The district
was curved from Machakos
district and borders Kitui district to the east, Machakos
district to the north and west and Kajiado district and Taita Taveta districts to the
south.
Makueni
District rises between
1000-1500
m above sea level and has vegetation
generally described as bush and thicket. It has a bimodal rainfall that has an annual
average
of 200-400
November/December.
Makueni
is described
mm, varying
with
altitude,
Average annual temperature
as recent sedimentary
and falls
in March/May
and
of the district is 28°C. Geology of
rocks and major soils are generally
described as dark-red sandy loams (The MacmiJlan Atlas, 1997).
The 1999 population census carried out in Kenya indicated that the total population of
the country was 28, 686,607, with 6,371,370 households and a population density of
49 persons
population
per square kilometre.
At the same time Makueni
of 77 I ,545 with 144,320 households
district had a total
and a population
density of 97
persons per square kilometre (Republic of Kenya, 2001). Some of the sub-locations
Makueni District are Kavuthu and Matiku in Mbitini Division and, Yikivumbu
le
in
and
22
Ndunguni in Nguu Division. According to Republic of Kenya (2001), some of these
sub-locations had very high population densities in 1999 (Table 4.1).
Table 4.1
Population densities and number of households at Kavuthu, Matiku,
Yikivumbu
Sublocation
Population
Kavuthu
Matiku
Ndunguni
Yikivumbu
1,060
2,273
957
1,321
and Ndunguni Sub-locations
Area of the
sub-location
(km2)
7.8
8.7
8.5
21.4
in 1999 and 2003
Population densities
(persons/krrr') in
1999
136
261
113
62
Number of
households
in 1999
171
372
165
205
Smaling et al., (1997) reported that loss of soil nutrients leading to soil depletion is a
worldwide problem affecting
135 million hectares, most of them in South America
and African, and that the problem of soil infertility in sub-Saharan
continuous cropping without sufficient replenishment
depletion.
Sanchez
of nutrients leading to nutrient
et al., (1997) further reported that, soil fertility
smallholder farms is the fundamental
food production
Africa is due to
in sub-Saharan
depletion
in
biophysical root cause for declining per capita
Africa.
Continuous
cropping,
removal
of crop
residues to feed animals and overgrazing between cropping seasons with little or no
external inputs reduce the productive capacity of arable lands and thus threaten the
sustainability of food production systems in sub-Saharan Africa (Sanchez et aI., 1997;
Stoorvogel et al., 1993). In Kenya, decline in crop yields is a major problem facing
smallholder farmers (Mathuva et al., 1996), which has been attributed to high costs of
inputs that make the use of inorganic fertilizers on staple food crops uneconomical
most smallholder farmers (lama et al., 1997).
for
23
4.2
A reconnaissance
and Methods
survey was carried out in Makueni district to select study sites.
Division agricultural
site identification.
existence
Materials
extension officers from Mbitini and Nguu Divisions assisted in
The criteria for site selection included accessibility
of on-going
donor
funded projects. Chiefs
to the sites and
and assistant
chiefs of the
selected sites advised on sites where no other projects were going on to avoid conflict
of interest. Two divisions,
Mbitini and Nguu, were identified and in each division,
two study sites were selected. The sites selected included Kavuthu and Matiku sublocations in Mbitini division, and Yikivumbu and Ndunguni
division. After site selection,
help of the extension
location. The assistant
officers
mobilized
announcements
farmer participatory
sub-locations
meetings were arranged with the
officers and the local chiefs and their assistants,
chiefs of each sub-location
the communities
in public meetings,
in Nguu
in each sub-
and the agricultural
to attend the meetings
through
extension
posters and
locally called 'barazas'.
Meetings were held in
central points, in local primary schools, since the schools
had closed for August
holidays. Participating
of between
representation
farmers democratically divided themselves
eight to thirty
on attendance,
and with an equal
of gender. Each group was assigned a topic to discuss and issued with a
flip chart where to document
general problems
economics.
people depending
into groups of sizes
affecting
group discussion outcomes. Topics discussed included
farmers,
soil fertility, crop production
and household
24
4.3
4.3.1
General
Participating
problems
affecting farmers in each study site
farmers in all selected sites identified and discussed general problems
facing them. Problems
identified
products, water accessibility,
and communication,
decreasing
Results
were poor soils, seed scarcity,
livestock feed scarcity, decreasing
crop pests and diseases, unreliable
marketing
farm
land, poor transport
rainfall,
education
costs,
crop yields, distant hospitals and poor services, social problems, lack of
post primary education, low livestock prices, soil erosion, poor planting methods, lack
of farm implements,
problems
inadequate labour and noxious weeds. In each site, all identified
were ranked
University,
using pairwise
ranking matrix
(PRA Programme
Egerton
1995). However for easy comparison of the site problem prioritization, the
results from pairwise ranking matrixes from each site were summarized
(Table 4.2).
Problems identified and priorities varied with the study sites. However, from farmer
discussions,
priority problems
were accessibility
to water, unreliable
soils, seed scarcity due to famines, decreasing land, education
rainfall, poor
costs, poor planting
methods, pests and diseases and distant hospitals.
4.3.2
Soil types, distribution
Soil identification
maps (Appendix
and women
and fertility status
and characterization
was made possible by drawing of sub-location
1). In all the study sites, sub-location maps were drawn by both men
representatives
and used them to identify
and quantity
various soil types. This became possible when the farmers
location
into villages,
distribution
the extent of
sub-divided
each sub-
which helped them to discuss the extent of soil type and
in detail. Soil types identified were red soils, sandy soils, black cotton
soils, sandy gravel and stony soils. Soil fertility status were described
as fertile,
25
average,
low or very low (Table 4.3). Low and very low fertility
study sites and covered
and Yikivumbu,
were described
Table 4.2
40%,
respectively.
100%, 75% and 98% of Kavuthu,
soils dominated
Ndunguni,
Only 2% of the soils, and at Yikivumbu
by the participating
General problems
Matiku
Sub-location,
farmers as fertile.
faced by farmers in the selected
sites of Makueni
District
Problem
Poor soils
Crop pests and diseases
Unreliable rainfall
Seed scarcity due to famine
Marketing farm products
Water accessibility
Lack of livestock feeds and
diseases
Education costs
Decreasing land
Decreasing crop yields
Poor transport and
comm unication
Distant hospitals and poor
services
Social problems
Lack of post primary
institutions
Low livestock prices
Soil erosion
Poor planting methods
Lack of farm implements
Inadequate labour
Noxious weeds
Site and Rank
Kavuthu
3
4
N/A
7
9
I
8
Matiku
7
N/A
N/A
5
10
1
Ndunguni
3
8
2
11
10
1
Yikivumbu
7
8
1
4
6
2
9
3
N/A
N/A
2
11
2
3
5
5
5
N/A
N/A
N/A
12
8
7
12
N/A
the
6
4
6
11
N/A
N/A
N/A
N/A
10
NIA
N/A
N/A
N/A
NI A
NI A
NI A
12
9
6
N/A
NI A
NI A
12
NI A
N/A
4
N/A
NI A
NI A
NI A
NI A
N/A
N/A
9
11
10
NI A means the item was not discussed at the selected site, while 1 and 12 means most
important and 12 least important, respectively.
26
Table 4.3
Soil characterization
and fertility status as described by farmers in
selected sites of Makueni District
Sublocation
Kavuthu
Ndunguni
Matiku
Yikivumbu
Soil type
Important soil characteristics
'I1ivi (Black
cotton soils)
'Kitune' (Red
Soils)
'Nthangathi'
(Sandy Soils)
Black 111 colour, high water
holding capacity
Red, productive, good water
holding capacity
Whitish, low water holding
capacity, poor In nutrients,
dries quickly
a) Cracking type: black, cracks
b) Non-cracking
cotton soil,
high water holding capacity
a) Hard red soils ('Kitune
Kyumu'). Gives low yields
b)
Soft
red
soils
(' lturn bekethe')
High water holding capacity
and good for crop production
Whitish or greyish in colour,
poor moisture retention
Black, high water holding
capacity
Red, good water retention
capacity
Whitish
III colour,
easily
drained
Reddish white in colour, have
gravel, low water retention
capacity
Black 111 colour, high water
holding capacity
'Yumba' or
'Ilivi' (Black
cotton soils)
'Kitune' (Red
Soils)
'Nthangathi'
(Sandy soils)
'I1ivi' (Black
cotton soil)
'Kitune' (Red
Soils)
'Nthangathi'
(Sandy soil)
'Kivuthi'
(Sandy)
gravel
'Yumba' or
'ilivi'
(Black cotton
soil)
'Kitune' (Red
soils)
'Nthangathi'
Sandy soil
'Mavia',
(Stony soil)
Red 111 colour, Low water
holding capacity
Greyish in colour, poor water
retention
Rocky, stony and shallow, low
water holding capacity
Fertility
Status
Average
Cover
(%)
60
Low
25
Very
low
15
Low
30
Low
60
Low
10
Low
15
Average
25
Low
45
Very
low
15
Fertile
2
Low
15
Low
75
Very
low
8
27
4.3.3
Wealth in relation to soil fertility management
Farm size was an important
conservation
indicator of riches and soil fertility
and food production.
farmer classes,
Farmers in the four study sites identified various
farm sizes they owned and soil fertility
employed in each farmer-class
improvement,
management
strategies
(Table 4.4). The richest farmers, class 1 farmers, had
largest farm sizes and commonly
bought animal manure from farmer class 3 and 4.
However, farm sizes owned by various farmer-classes
varied with site. For example
farmer class 1 at Kavuthu owned 10 to 15 acres of land while the same farmer class at
Yikivumbu owned more than 20 acres.
4.3.4
Crop preference
Main cereals
grown
in the study sites were maize (Zea mays L.), pearl millet
tPennisetum glaucum (L.) R.Br.), sorghum (Sorghum bicolor (L.) Moench) and finger
millet tEleusine
corcana
(L.) Gaertn), while grain legumes included common bean
(Phaseolus vulgaris L.), cowpea (Vigna unguiculata
(L.) Walp), pigeonpea (Cajanus
cajan (L.) Millsp), green gram (Vigna radiata (L.) R. Wilcz) and dolichos (Lablab
purpures L.). Using pairwise ranking crop preference was prioritized
the results summarized
the most preferred
important
in all sites and
(Table 4.5). Crop preference ranking showed that maize was
crop followed
legume preferred
by common
bean. Pigeon
pea was the second
by farmers while finger millet was the second most
preferred cereal. Pearl millet and Dolichos were only discussed at Yikivumbu
location.
sub-
28
Table 4.4
Site
2
3
4
Soil fertility management across farmer wealth classes
Wealth classes and Soil management strategy
Class 4
Class 1
Class 2
Class 3
N/A
Own less than 7
Own 10-15 acres -Own 7-9 acres
of land, all farms of land, most acres of land,
farms terraced, rarely
terrace
are well terraced,
buy their farms, sells
manures
all may
farms,
buys manure
from their manure to
manure
from class 3
class 1 and 2
class 3
N/A
Own 20-50 acres Own 5-9 acres Own less than 5
of land, heavily of
land, acres of land,
limited
manures
farms manures farms, have
may
purchase manure use, a few
and
purchases
manure
from manure
from terrace
their
class
3,
high class
3, farms, farms are
is not fenced
and
degree
of terracing
terracing,
well common,
most are
commonly
fenced
farms, farms fenced,
overgrazed
by
farms have fertile most
farms roaming animals,
have
fertile Soil infertility is
soils.
soils
very common.
N/A
Own more than Own
10-15 Own less than 10
15 acres,
have acres,
most acres
of
land,
well- fenced
terraced farms, provide labour for
farms,
well- many
fence class 2 and 1 and
terraced
farms, their
farms, their farms
are
high harvest due some have good not manured nor
to fertile soils
harvest due to terraced
and
fertile soils
commonly
no
harvest.
Own more than Own
10-19 Own 5-9 acres, Own less than 5
20 acres, all farms acres, all their have
scattered
acres of land,
are terraced,
farms
are braches and sisal farms not fenced,
manures
farms terraced,
fencing, have few do not terrace
and buys manure
manures
and terraces,
does their
farms,
from class 3 and buys
manure limited manuring,
limited manuring,
4,
from class 3 and limited
harvest little
if
any
high
harvest 4, high harvest due to poor soils
harvest
because of fertile due to fertile
soils
soils
Site (sub-location); 1- Kavuthu, 2- Matiku, 3- Ndunguni and 4 Yikivumbu.
Class 1 consisted of the richest farmers while class 4 was the poorest farmers.
NI A means farmer class was not described in the site.
29
Table 4.5
Crop preference in Nguu and Mbitini Divisions ofMakueni
District
Site and crop priority position
Crop
Yikivumbu
Ndunguni
Kavuthu
Matiku
Maize
]
1
1
I
Common bean
2
2
2
4
Green gram
4
6
5
7
Pigeonpea
5
3
3
2
Cowpea
3
5
6
3
Finger millet
6
4
4
5
Sorghum
9
7
7
6
Dolichos
8
N/A
N/A
N/A
Pearl millet
7
N/A
N/A
N/A
N/A means the item was not discussed at the selected site, whilel means most
important and 9 means least important.
4.3.5
Participating
Annual food availability
farmers discussed food security in all the sub-locations.
Results obtained
from the study showed that food availability peaked in February and slowly dropped
over the year in all sites except Matiku where it rose between June and August
(Figure 4.1). At Yikivumbu,
from September
through
Ndunguni and Kavuthu, food availability dropped to zero
November.
However,
from November
food availability
started rising again at Kavuthu due to availability of green vegetables
from cowpea
that provided food from November, a month following the on-set of short rains.
4.3.6
Household income and expenditure
Results obtained from the farmer participatory meetings revealed that crop production
and livestock keeping were the main source of income in the study sites (Table 4.6).
Income from crops was 55%, 10%, 40% and 60% at Yikivumbu,
and Matiku, respectively
Ndunguni, Kavuthu
compared to income from livestock that was 25%, 70%,
30% and 15% at Yikivumbu,
Ndunguni, Kavuthu and Matiku, respectively. However,
30
the percentage
expenditure
on food purchases
was 40%, 60%, 40% and 50% at
Yikivumbu, Ndunguni, Kavuthu and Matiku, respectively.
120
-----,---
----
.--.-.----.----._------------
.........-Matiku
-Kavuthu
100
C
_Ndunguni
____ Yikivumbu
80
"0
0
<2
0)
60
>
40
:0
-'.<;;"
'"
<
20
0
Jan
Feb
Mar
Apr
Ma
JWl
Jul
Aug
Sept
Oct
Nov
Dec
Months of the year
Figure 4.1
Food availability calendars ofMbitini
Table 4.6
Household sources of income and expenditure in the study sites
Sub-location
Yikivumbu
Ndunguni
Kavuthu
Matiku
Income
Crops
Livestock
Others
Contribution (%)
55
25
20
Crops
Livestock
Others
10
70
20
Crops
Livestock
Monthly
salary
Casual labour
Crops
Livestock
Employment
Others
40
30
10
20
60
15
15
10
and Nguu Divisions
Expenditure
Food
Farm inputs
Health
Others
Food
Farm inputs
School fees
Others
Education
Food
Others
Education
Food
Health
Others
Contribution
(%)
40
IS
20
25
60
II
5
24
30
40
30
IS
50
18
17
n
'
31
4.4
Accessibility
Discussion
to water was the most important problem in three out of the four study
sites, where it took position number I. According to the farmers in all the sites, there
was no clean water for domestic and livestock use because all water in the sublocations came from shallow wells sunk in seasonal rivers. The farmers said that the
wells occasionally
sometimes
dried up in dry weather
far away from their homesteads
forcing
them to sink deeper
wells
that caused them to spend most time
fetching water for domestic use. However, rainfall unreliability was very important at
Yikivumbu and Ndunguni sub-locations where it ranks position 1 and 2, respectively.
It was noted that the two sites had experienced
Related to the rainfall unreliability
priority
position
commonly
planted
4 at Yikivumbu.
drought between
1998 and 2003.
was the problem of seed scarcity that was given
It was noted that farmers
in the study sites
local crop varieties most of which had been wiped out by the
frequent droughts.
The problem of decreasing land was more important at Kavuthu and Matiku, where it
took positions 2 and 3, respectively. It was observed that most farm units were smaller
than in the other sites. The farmers said that in addition to population increases there
were influxes of farmers from drier sites who purchased
because the sub-locations
land in the sub-locations
were wetter than most parts of the district. In addition, soil
fertility problem was more important at Kavuthu and Ndunguni
sub-locations
than
other sites, where the problem took position 3. It was noted that the farmers in the two
sub-locations
practiced
irrigation
of vegetables
and cereals along nearby seasonal
rivers and had noted that the soils in these sub-locations
had low fertility status.
However, crop pests and diseases were important at Kavuthu where it took position 4
32
out of the listed 11 problems. Farmers in this site cited pest problems in cowpea and
pigeonpea, which they said, lowered crop yields. However at Ndunguni poor planting
methods
was listed at position
number 4 where farmers
cited lack of equipment
especially ploughs that are commonly used for cultivation in the sub-location.
Distant hospital and poor services at the hospitals took position 4 out of 12 problems
at Matiku. It was noted that the nearby shopping centre, Matiku market, lacked health
care facilities,
clinics or health centres and farmers took their sick to hospital that
were over 20 km away for treatment, where the services where poorly offered. It was
also noted that the sub-location was hilly that made transportation
The commonest
difficulty.
soil types were black cotton, red and sandy soils with means of
26.8%, 31.3% and 36.3% respectively across the sites. In addition, dominant soil type
at Kavuthu was black cotton soil (60%), Ndunguni was red soil (60%), Matiku and
Yikivumbu
was sandy soils with estimated cover of 45% and 75%, respectively.
According to participating
farmers, some of the characteristics
of infertile soils are
powdery nature of a soil, presence of bright red colour, soil hardness to till. The soils
in all the study sites were described as infertile with low to very low fertile soils
covering 40%, 100%, 75% and 98% of Kavuthu, Ndunguni, Matiku and Yikivumbu,
respectively. These results can be compared to the findings of Mach aria (2003) where
farmers classified
classification,
soils at Kasikeu division of Makueni
Unlike scientific
which considers whole soi I profile, farmer classification
based on physical characteristics
hardness.
district.
of soils was
of the soil, such as surface colour, drainage and
33
Soil fertility status in the sites could be tied to soil fertility management
practices and
farmer beliefs. From the results in soil fertility status, it appeared that most farmers in
the study sites were in farmer-class
3 or 4 of farmers, who rarely terraced or used
animal manure to enhance soil fertility in their farms. In addition, farmers, animal
manure was the main input used by farmers to enhance soil fertility. A similar finding
was also documented
by Probert et a/., (1995), who found out that in the drylands of
eastern Kenya animal manure was the main farm input and was applied without
measurement.
A related finding was documented
in West Pokot where farmers used
farmyard manure for crop production without measuring the amount of manure they
used (Wanjekeche
et a/., 2000).
Maize was their staple food crop and of highest priority
111
all sites followed by
common bean that took position 2 in all sites except at Matiku where bean took
position 4, because farmers experience was that pigeonpea
and cowpea yields had
j
i
i,j
been better than those of bean in the site. Pigeon pea took position 3 at Kavuthu and
Ndunguni and was therefore the second most important
gram was the third most important legume at Yikivumbu
legume in the sites. Green
after common
bean and
cowpea because farmers planted the legume largely for income generation. It was also
found out that in all the study sites legumes were sold to generate income because
they fetched better cash in the markets than cereals. Finger millet was the second most
important cereal after maize in all sites because it was used for domestic consumption
and for income generation.
Farmer crop prioritization
was also reported by Onyango
et al., (2000); Okoko and Makwaro (2000) who found out that maize was the most
preferred crop in other parts of Kenya.
34
Food availability to households was highest between February and April because this
was the harvesting season for the short rains crop. The farmers also revealed that this
was the most reliable rain season compared to the long rains. After the harvesting
season, household food availability dropped in all sites, with sharp drops at Ndunguni
and Yikivumbu.
According
to the agricultural
extension
officers, the two sites had
high incidences of crop failures due to rainfall unreliability.
the farmers from Matiku Sub-location,
the sub-location
However, according to
occasionally
got a crop in the
short rains that matured in August and boosted household food availability in the sublocation. The participating
sites by estimating
farmers measured household food availability in the study
proportions
of farmers buying food from the food stores in the
market centres at the study sites. Some farmers started purchasing
food as from June
and their numbers increased as the year progressed and peaked at September, October
and November at Yikivumbu, Ndunguni and Kavuthu, respectively.
Farmers obtained their livelihood from crop growing and livestock keeping. However,
contribution
to household
income from crops on average (41 %) was higher than that
from livestock (35%). This largely indicated that most farmers grew crops compared
to keeping livestock. The farmers also spend most of their income to purchase food,
thus confirming the annual food deficit in the study sites.
35
CHAPTER FIVE
5.
AREA UNDER GRAIN LEGUMES AND PROBLEMS FACED BY
FARMERS IN LEGUME PRODUCTION
5.1
Introduction
Grain legumes are valued for their multiple uses as food, green manure, fodder and
cover for crops. Besides
these direct benefits,
legumes
generate
cash and grain
legumes fetch more cash when sold compared to cereals such as maize. In Kenya,
many types of grain legumes are grown in various parts of the country. However
common
legumes cultivated
in the country include bean (Phaseolus
pigeon pea (Cajanus cajan (L.) Millsp), cowpea (Vigna unguiculata
vulgaris L.),
(L.) Walp) and
green grams (Vigna radiata Wilczek).
5.1.1
Bean (Common bean) (Phaseolus
Bean is the commonest
legume incorporated
vulgaris L.)
in different cropping systems in Kenya
(Chui and Nadar, 1984). The main bean production areas in the country are Eastern,
Central, Western and Nyanza Provinces
at altitudes varying from 1,500 to 2,500 m
above sea level. In addition, common beans are grown in Eastern Province,
there are two rainy seasons
(Masumba,
with a mean of between
1984; Stoetzer and Waite,
dietary protein and contributes
500 to 800 mm annually
1984). Like other legumes,
to the maintenance
where
bean provides
of soil fertility (Chui and Nadar,
1984).
Both pests and diseases attack beans either in the field or during storage. Common
insect pests of economic of importance to bean include:
36
•
Bean fly or bean stem maggot (Opyomyia phaseoli),
usually common in the
seedling stage where it feeds on the stem base from inside the bean plant.
•
Black bean aphid (Aphis fabae Scop.), found usually during cool dry periods
and sacks sap from pods, leaves, leaf stalks and stems.
•
Bean
leafhopper
(Empoasca
solana
Delong),
common
during
vegetative
stages where it feeds on leaves and flowers.
•
Bean weevil (Acanthoscelides
obtectus Say.), common in seed storage, where
it feeds on the grain.
Dimethoate is used to control bean fly, black aphid and leafhopper, and planting early
is recommended
to avoid insect infestation. However, bean weevil is controlled by
dusting bean seeds with super atelic at 50 g per bag before storage. Common diseases
of bean in the production
Synonym:
Uromyces
appendiculatus
vignae BarcI.), that cause defoliation
stem blight (Macrophomina
monocultures
areas are rust (Uromyces
phaseolina)
Pars.; Frios
and yield losses; Ashy
and Southern blight (Scleoatinia
roflsii). In
of bean, most diseases are severer than when bean are intercropped with
maize. These common diseases are controlled using Benomyl and copper oxychloride,
as well as rotating beans with cereals. (Van Rheenen et al., 1981; Audi et al., 1996).
Some of the improved
bean varieties
from Kenya Agricultural
Research
Institute
(KARI) are Katumani bean 1, Katumani bean 2, Katumani bean 9 and Kat-x-56. With
an exception
of Kat-x-56,
that matures within 62-68 days other varieties
mature
within 60-65 days. The potential yield of Katumani bean I, Katumani bean 9 and Katx-56 is 1440-1980 kg/ha while that of Katumani bean 2 is 1350-1530 kg/ha (Audi et
al., 1996).
37
5.1.2
Pigeon pea (Cajanus cajan (L.) MiIIsp)
Kenya ranks second in pigeon pea production after India in the world (Remanandan et
al., 1982). Eastern, Coast and Central provinces are the major pigeon pea growing
areas of Kenya. Eastern Province leads in hectarage and about 90% of the total area
under pigeon pea is in the province. In Kenya, both green and dry grain of pigeon pea
is used as food while the plants are used for animal feed, green manure and as a cover
crop (Khan,
(Sheldrake
1973). Pigeon
and Narayana,
pea has a residual
nitrogen
(N) of about 40 kg/ha
1979). Both diseases and pests attack Pigeon pea. For
example wilt that is caused by Fusarium undum Butler is the commonest
disease of
pigeon pea in Kenya, while serious pigeon pea pests in the country include thrips
(Megaluro-thrips
rjostedti Tryhom), Heliothis tHeliothis armigera)
and pod-sucking
bug (Clavigralla gibbosa) (Shakoor et al., 1984a).
Pigeon pea has several improved varieties that include Kat 60/8, Kat 81/3/3, Kat 777
and ICPL 89091
that mature
within
136-150,
170-185,
160-180
and 120 days
respectively. Their yield potentials are 1200-1500 kg/ha, 1400-2500 kg/ha, 1400-2200
kg/ha and 1000 kg/ha respectively (Audi et al., 1996).
5.1.3
Green gram (Vigna radiata Wilczek)
Green grams (mung bean) are an important crop in the warm dry parts of eastern
Kenya where it is grown for both subsistence and as a cash crop (Shakoor et al.,
1984b). Dry whole grain is used for food, although the Asian community,
the largest
consumer of the crop, cooks it as split grains (Dhai). Grain protein content of green
gram varies from 21-29% depending on the variety and environment
where the crop
has been grown. Mung bean is generally free from flatulence inducing factors that are
38
common in many grain legumes. Sulphur amino acids, methionine and cystine content
are low as in other legumes and lysine is high (8gll OOg protein dry weight) (Shakoor
et 01., 1984b)
Mung bean is pan-tropical
and is able to grow in adverse conditions.
It can escape
drought through its early maturing ability (Rowe, 1980) and some varieties are more
resistant to drought than cow pea (Waite et al., 1984). Pests of green gram are thrips,
aphids, pod-sucking
bugs, apion beetle and bruchids while common diseases include
powdery mildew and yellow mosaic virus (Audi et al., 1996). In addition, improved
varieties of green gram include KVR 22 and KVR 26 with potential yields of 1000 to
l300 kg/ha (Audi et al., 1996).
5.1.4
Cowpea (Vigna unguiculata (L.) Walp)
Cowpea is an annual or bi-annual grain legume commonly
referred to as cowpeas. In
Kenya, it is the third most important grain legume after beans and pigeon pea and
covers about 18000 ha, excluding the cowpea grown in home gardens (Muthamia and
Kanampiu, 1996). About 85% of the total area under cowpea is in arid and semi-arid
lands (ASALs)
of Eastern Province
and 15% in the Coast, Western and Central
Provinces (Muruli et al., 1980; Muthamia and Kanampiu,
both indeterminate
predominant
1996). In Eastern Province,
and semi-erect types are grown, while indeterminate
in Western and Central Provinces. Chaturverdi
that indeterminate
types
give
better yields under
drought
land races are
et aI., (1980) indicated
conditions
than the
determinate types. Cowpea is commonly grown in mixtures with maize, sorghum or
pigeon pea. In Eastern Province cowpea is grown for both grain and leaves while in
Western and Central Provinces
it is mainly grown for its leaves (Shakoor et al.,
39
1984a). Cowpea is used for food, fodder and as a source of income. Leaves, young
pods and grain are the parts of the plant used for food. The same plant parts are sold
to generate cash for farmers. Cowpea value lies in its high protein content of 10% 35% (Imungi and Porter 1983), its ability to tolerate drought and fix atmospheric N,
which allows it to grow and improve poor soils. It has a well-developed
deep root
system and grows well under drought conditions (Shakoor et al., 1984a; Muruli et al.,
1980).
Major
insect
leafhoppers
pests
of
(Empoasca
cow pea
are thrips
(Megalurothrips
sp), legume pod borers (Maruca
armigera), aphids (Aphis craccivora Koch), pod-sucking
and apion beetle (Apion solenortum).
rjostedt
testulalis
Tryhom),
and Heliothis
bugs (Clavigralla
gibbosa)
Bruchid is a pest of dry cowpea grain (Shakoor
et al., 1984a; Audi et al., 1996). Some of the improved cowpea varieties from KARI
include Machakos 66, Katumani
80, KVU-419 and KVU HB 48EI0. These varieties
matures within 85-95, 75-85, 65-72 and 85-95 days, respectively,
and have yield
potentials of 1200-1800 kg/ha, 1500-1800 kg/ha, 1170 kg/ha and 1200-1500 kg/ha,
respectively (Audi et al., 1996).
Legume production in the dryland is very low partly because soils are commonly
nutrient deficient, especially of nitrogen (N) and phosphorus (P). This is iargely due
to continuous cropping without external inputs (McCown et al., 1992). However
many farmers use manure because they are aware of its benefits but the quanties
avilable are insufficient and of poor quality (Probert et al., 1995). As a result, crop
yields are low and yields of grain legumes rarely exceed 500 kg/ha (Mathuva et al.,
1996).
40
5.2
Materials and Methods
On-farm surveys using a structured questionnaire
four sub-locations
covering
Kenya. The sub-locations
Kavuthu
and
Matiku
reconnaissance
survey
Agricultural
two divisions
(Appendix
of dryland
Makueni
included Yikivumbu and Ndunguni
In Mbitini
division.
The
2) were carried out in
sites
had
on the basis of their accessibility
district
in eastern
in Nguu division, and
been
during
selected
rainy seasons.
extension officers of each division assisted in site identification
the reconnaissance
surveys. To implement questionnaires,
random visits to homesteads.
The number of household
in a
during
the process involved single
During the visits, household
heads were interviewed.
heads interviewed were 32, 23, 21 and 43 at Yikivumbu,
Ndunguni, Kavuthu and Matiku, respectively, and were determined with the help of a
biometrician
Ndunguni,
participatory
from the total households
Kavuthu
meetings
and
Matiku,
(Chapter
(350, 234, 216 and 443 at Yikivumbu,
respectively)
documented
during
farmer
4). Samples of weeds, pests and disease-affected
crops were taken from Ndunguni and Kavuthu during the on-farm trials and brought
to Kenya Forestry Research Institute (KEFRI) laboratories where a plant taxonomist,
an entomologist
and a pathologist
identified relevant samples. Data collected was
analysed using SPSS for windows Release 10.0 of 1999. The analysis was done by
sub-location to allow for comparison between sites.
41
5.3
5.3.1
Results
Grain legume production and area under legumes
Yikivumbu had largest farm sizes followed by Ndunguni, Kavuthu and finally Matiku
and the total cultivated
operated generally
farm sizes were in the same order. The number of farms
increased
with decreasing total farm size. In addition, the area
occupied by cultivated legumes increased with decreasing farm size and ranged from
48% to 92% (Table 5.1).
Table 5.1
Site
Yikivumbu
Ndunguni
Kavuthu
Matiku
SED
Area under grain legume production in selected sub-locations
Area
Under
legumes
(ha)
1.2
l.1
Proportions ofthe
Cultivated area
under grain
legumes
(%)
48
55
l.5
83
1.2
l.l
92
0.69
0.57
Average
Total
Cultivated
number of farm size farm size
(ha)
fields
(ha)
1.1
4.9
2.5
1.4
4.4
2.0
l.8
2.1
2.3
2.0
0.25
1.5
1.28
Main legumes grown in the study sites included common bean, cowpea, green gram
and pigeon pea; with grain yield ranges of 30 to 416 kg/ha. Yikivumbu
yields of cowpea
had highest
and green gram of 239 kg/ha and 416 kg/ha, respectively.
addition highest amounts
In
of common bean and pigeon pea of 250 kg/ha and 189
kg/ha, respectively, were recorded at Ndunguni (Table 5.2).
42
Table 5.2
Average legume yields (kg/ha) in selected sub-locations
Sub-location and legume yield
Crop
Kavuthu
Matiku
Ndunguni
Bean
42
91
250
Pigeon pea
81
178
189
Cowpea
107
130
102
Green grams
178
30
60
Totals
260
459
719
SED means standard errors of differences of means
5.3.2
Soil fertility distribution,
According
to farmers'
indigenous
Yikivum
bu
63
187
239
416
905
Grain yield
totals
446
635
578
684
SED
29
35
42
88
farm inputs and their sources
knowledge, soil fertility status in the study sites
could be described as fertile, moderate or poor (Table 5.3). Most of the farms in the
study sites had moderate
fertility status. With exception
of Kavuthu, fertile soils in
other sites covered less than 10% of the cultivated areas. In addition, Kavuthu and
Matiku had highest percentages
of farms with poor soils that covered about 24% and
23%, respectively.
Table 5.3
Soil fertility status of farms in the selected sites of Makueni District
Sub-location
Yikivumbu
Ndunguni
Kavuthu
Matiku
Means
Fertile
0
21.7
5
7
8.4
Soil Fertility Status (%)
Moderate
100
69.6
71
70
77.7
Poor
0
8.7
24
23
13.9
Farmers in the study sites used farmyard manure (animal manure), compost and crop
residues and inorganic fertilizers to enhance soil fertility (Table 5.4). Animal manure
was the main farm input and between 84% and 97% of the farmers interviewed said
KENYATTA UNiVERSITY LIBRARY}
!
43
they used it, while farmers using crop residues ranged from 6% to 39%, and those
using compost ranged from 0 to 70%. Use of inorganic fertilizers was only recorded
at Ndunguni.
In addition, between 84% and 100% of the farmers interviewed used
animal manure from their own animal kraals (Table 5.5).
Table 5.4
Farmers (%) using inputs to enhance soil fertility
Sub-location
Yikivumbu
Ndunguni
Kavuthu
Matiku
Means
Table 5.5
Fertilizer
6
3
o
39
70
9
o
o
o
Crop residues
87
91
10
14
17
16
22
84
90
2
Sources of animal manure used by farmers
Sub-location
Yikivumbu
Ndunguni
Kavuthu
Matiku
Means
5.3.3
Compost
Manure
97
Manure sources (%)
Bought
Owned
100
3
3
35
91
100
84
94
Borrowed
17
5
o
2
11
9
7
Problem weeds, diseases and pests
There were two main types of problem weeds recorded, grass and non-grass types.
Non-grass
weeds were 'Mukuutu'
'Mung'oi'
(Acanthospernul11
(Trichodesma
hispidum
zeylanicum
(DC», 'Munzee'
(Burm.f.)
R. Br.),
black jack (Bidens pilosa
(L.» and 'Uthunga'
(Launaea cornuta (Oliv. & Hiern) O. Jeffrey). The grass weeds
included 'Kithangai'
couch grass tDigitaria scalarum (Schweinf.) Chiov.), 'Ikoka'
star grass (Cynodon dactylon (L) Pers.) and 'Mbiu' (Cyperus rotundus (1.». Presence
of weeds in farmers' fields varied across sites. For example all farmers interviewed
(100%) at Yikivumbu said that 'Mung'oi'
and 'uthunga'
were present in their farms
44
while at Kavuthu 91 % of the farmers interviewed said that 'Mukuutu',
'Mung'oi'
'Munzee'
and
were present in their farms (Table 5.6).
Pests in this study were divided into two, field and storage pests. Field pests included
animals such as squirrels and monkeys, birds and insects such as aphids and white
flies (Bemisia tabaci Gennadius),
while weevils were the storage pests. Documented
diseases were rust and powdery mildew (Erysiphe polygoni
5.7). All sites experienced
DC ex St.-Am) (Table
pest and disease infestation at varying levels. For example
monkeys were commonly repotted at Ndunguni (39%) while squirrels and birds were
commonly reported at Yikivumbu and Ndunguni.
Table 5.6
Farmers (%) that reported common weeds in their farms
Site
1
2
3
4
5
6
7
Yikivumbu
9
28
97
41
78
100
100
Ndunguni
52
74
96
83
96
96
96
Kavuthu
62
62
91
91
67
81
91
Matiku
40
49
81
72
61
61
81
Means
41
53
91
72
76
85
92
1 - Digitaria scalarum, 2- Cynodon dacylon, 3 - Trichodesma zeylanicum, 4 Bidens
pilosa, 5- Cyperus rotundus, 6- Launaea cornuta and 7- Acanthospernum hispidum
Table 5.7
Farmers (%) that reported pests and diseases in their farms
Storage
Field pest
Site
Monkeys
Squirrels
Birds
Pests
Aphi
ds
White
flies
Weevil
s
6
lOO
lOO
97
97
2
lOO
39
lOO
74
74
3
6
95
86
86
95
4
72
84
9
97
61
Mean
15
98
90
85
82
1) - Yikivumbu, 2)- Ndunguni, 3- Kavuthu and 4)- Matiku
I
Diseases
Powdery
mildew
Rust
lOO
78
88
74
86
88
88
91
81
84
100
lOO
74
lOO
94
45
5.3.4
Household
head education
and food sources and availability
Majority of the household heads (between 43% and 70%) had primary education only
(Table 5.8). In addition, about 90% of households experienced
food shortage during
the long dry spells and on average over 95 % of the household heads purchased food
during the long dry spells experienced in the drylands (Table 5.9).
Table 5.8
Level of formal education (%) of household heads
Sub-location
None
Yikivumbu
15.6
Ndunguni
4
Kavuthu
19
Matiku
12
Means
13
N/A means not available.
Table 5.9
Site
Yikivumbu
Ndunguni
Kavuthu
Matiku
Means
Primary
65.6
70
43
58
59
Secondary
15.6
26
38
30
28
University
3.1
N/A
N/A
N/A
0.75
Food availability and source (%) during long dry spells
Food shortage was
experienced during the long dry
spells
Yes
No
100
91.3
8.7
95.2
4.8
95.3.
4.7
95.5
4.6
o
Purchasing was the main source
of food during long dry spells
Yes
100
87
100
No
95.3
95.6
4.7
4.4
o
13
o
46
5.4
Discussion
Grain legume cover ranged from 48% to 90% increasing with decrease in cultivated
farm
size.
The
intercropping
increase
in land under
of grain legumes
legume
with cereals,
cover
was
commonly
due
maize.
legume and cereal crops is a common practice of smallholder
to increased
Intercropping
of
farmers throughout the
tropics (Sakala et al., 2000) and in East Africa maize is commonly
intercropped or
rotated with grain legumes (Pilbeam et al., 1995).
Sites with relatively larger cultivated areas, Yikivurnbu and Ndunguni
higher grain yields probably
important
had relatively
indicating that farm sizes in the study sites were more
than the area under grain legume cultivation.
Commonly
grown grain
legumes had mean grain yields recorded as 250 kg/ha, 189 kg/ha, 239 kg/ha, and 4 I6
kg/ha for bean, pigeon pea, cowpea and green grams, respectively, which were below
the potential grain legume yields documented by KARI, where potential grain yield of
common bean lies between
1350- 1980 kg/ha, that of pigeon pea between
1200-2500
kglha, green gram between 1000- 1500 kg/ha and that of cowpea between 1200-1800
kg/ha (Audi et al., 1996).
Most farms in the study sites had moderate soil fertility status followed by poor soils
and finally fertile soils with overall means of 77.7%, 13.9% and 8.4%, respectively.
This implied that fertile soils covered less than 10% of the cultivated
farms in the
selected sites. These
participatory
results
agreed with the findings
from farmer
meetings in the same study sites (Chapter 4), which revealed that, fertile soils covered
less than 10% of cultivated farms. According to the farmers, soil erosion was the main
cause of soil infertility and 81 %, 22%, 62% and 71 % of the farmers interviewed
at
47
Yikivumbu,
Ndunguni,
they had problems
Kavuthu
and Matiku Sub-locations,
of soil erosion
reported in the sub-locations
respectively,
said that
in their farms. The high erosion observations
were largely due to free-range mode of livestock keeping
in the study sites that was noted at 97%, 87%, 81 % and 84% at Yikivumbu,
Ndunguni,
Kavuthu
animal keeping
and Matiku
often causes overgrazing
However,
higher
respectively.
between seasons
Free-range
(Sanchez
mode of
et al., 1997;
et al., 1993) that may cause soil erosion and soil
Smaling et al., 1997; Stoorvogel
degradation.
Sub-locations,
proportions
of poor soils at Kavuthu
(24%) and
Ndunguni (23%) were most probably due to continuous cropping without farm inputs,
removal of nutrients in harvested crops, runoff and erosion, and leaching (Sanchez et
al., 1997; Smaling et al., 1997)
Farmers improved their soils mainly by the use of farmyard manure (manure). The
manure was readily available
because
most households
kept livestock.
This was
confirmed by the incidence of livestock ownership of 100%, 96%, 100% and 98% at
Yikivumbu,
Ndunguni, Kavuthu and Matiku, respectively.
and 86% of farmers at Yikivumbu,
respectively,
manure
Ndunguni,
Kavuthu
However, 98%, 91 %,67%
and Matiku sub-locations,
said that animal manure was not adequate.
from their own animal
Farmers
commonly
used
kraals and only a few farmers,
especially
from
Ndunguni, bought or borrowed manure from their neighbours.
Other inputs were crop
residues, compost and fertilizers, that were used at small amounts and commonly at
Ndunguni.
Ndunguni
sub-location
borders Muoni River and the residents practiced
irrigation of vegetables that required a lot of farm inputs, which explained high farm
inputs and higher proportion of fertile soils in the sub-location.
suggested
that low soil fertility
and low use of organic
Bekunda et al., (1997)
and inorganic
mineral
48
fertilizers
are
productivity
the
greatest
biophysical
constraints
to
increasing
agricultural
in the farming systems in the semiarid regions of Africa.
There were seven problem weeds recorded in the study sites of both grass and nongrass types. According
to the farmers these weeds were very competitive
controlled could significantly
and if not
lower legume yields. They said some of the weeds were
most difficult to control especially 'Kithangai'
and 'Mbiu'
rhizomes. Two of the non-grass weeds 'Mukuutu'
which have underground
and 'Mung'oi'
became aggressive
after first weeding and farmers were forced to do about three weedings to ensure good
legume yields. The 'Munzee'
and according
and 'Uthunga'
to the farmers weeding
germinate
immediately
after planting
had to be done two weeks after planting.
Majority of the farmers used hand hoes to control weeding after legume germination.
However, where the legumes were grown in rows, the farmers used ox-ploughs
remove weeds between the crop rows. Farmers
'Mukuutu'
Farmers
and 'Mung'oi'
at Ndunguni
also said they occasionally
to
burnt
during the dry seasons, commonly before planting seasons.
said that they culturally
controlled
'Kithangai'
by planting
pumpkin (Cucurbita maxima Duchesne) plants that chocks the weed.
Pest infestation
varied across sites because of variation
in location and vegetation
cover. For example monkeys were commonest
in Ndunguni because the sub-location
bordered
called Nguu where
an unsettled
and forested
Squirrels and birds were commonest
scheme
in Yikivumbu
monkeys
lived.
and Ndunguni because the sub-
locations had large uncultivated and bushy areas, which provided hiding places for the
birds and the squirrels. Monkeys, birds and squirrels were controlled by scaring them
away and sometimes
by use of locally made traps. According to farmers, pests and
49
diseases were controlled by use of chemicals although cultural control was also used.
Farmers
at Yikivumbu
mildew
(Erysiphe
sub-location
po lygoni
De.
said they used wood ash to control powdery
Es St.-Am.)
and white
flies
(Bemisia
tabaci
Gennadius). However, farmers in all the sites said that they control rust by use of crop
rotation.
Most household
Ndunguni
heads had attained
70%, Kavuthu
primary
school education
(Yikivumbu
43% and Matiku 58%) and commonly
66%,
relied on crop
growing and livestock keeping. It was noted that 88%, 87%, 95% and 70% of the
household
heads at Yikivumbu,
Ndunguni,
Kavuthu and Matiku, respectively,
were
engaged in livestock keeping and crop growing. This most probably resulted to the
food shortage documented
in the study sites and consequent buying food. Poshiwa et
aI., (2006) found out that additional years of schooling have significant positive effect
on farm output and gross values of farm production. It was interesting to note that not
all farmers who experienced food shortage at Ndunguni and Matiku bought food most
probably
because Ndunguni
borders
Muoni River and most farmers
plant crops,
especially maize, that provide food to households during the long dry spells. It was
observed that (Chapter 4) farmers at Matiku got a crop during the long rains that
boosted household food availability during the long dry spells.
50
CHAPTER SIX
6.
SCREENING NEW COWPEA VARIETIES FOR DRYLANDS OF
EASTERN KENYA
6.1
Introduction
Cowpea (Vigna unguiculata [L.] walp) belongs to the tribe Phaseoleae
Van der Maesen,
1985). It was domesticated
Africa (Ng and Marechal,
in Sub-Saharan
1995) and is a major component
(Polhil and
Africa, probably West
of traditional
cropping
systems in the drier parts of the tropics. Cowpea is important because of its multiple
uses, which
suppression
include
improving
soil fertility through
biological
nitrogen
fixation,
of weeds, the green pods and dry grains are eaten as well as leaves and
green pods. Green leaves may be consumed or sold to generate household
income
(Muli and Saha, 2000). It is also grown for forage and for use as a green manure
(Tarawali et al., 1997). Cowpea is a source of protein with mean crude protein of
leaves,
grains
and crop residues
ranging
from 32-34%,
respectively,
and contains 62% soluble carbohydrates.
carbohydrate
content cowpea
is used in nutritional
23-35%
and
11-25%,
Due to its high protein and
products
(Imungi
and Porter,
1983).
In Africa, West Africa is the key cowpea producing zone mainly in the dry Savannah
and semi-arid
agro-ecological
zones. Nigeria is the largest cowpea
producer
and
consumer in the world while Niger is the largest cowpea exporter in the world with an
estimated 215, 000 metric tones (MT) exported annually mainly to Nigeria. World
cowpea
production
was estimated
at 3,319,375
metric tones
with 75% of that
production being from Africa, 21 % from South America, 1% from Europe, 2% from
Asia, and 1% from North America
(FAOST AT, 2000). West and Central Africa
51
account for over 64% of the estimated
wide followed
12.5 million ha cultivated
by Central and South America
to cowpea world
(19%), Asia (10%) and East and
Southern Africa (6%) (Singh et al., 1997).
Cowpea
is commonly
grown in the arid and semi-arid
second to pigeon pea in productivity
areas of Kenya and rates
(Waite et al., 1984). The area under cowpea in
Kenya is 1800 ha excluding cowpea in home garden, with about 85% of the area
under cowpea production
and Kanampiu,
as sorghum
being in the arid and semi-arid lands (ASALs) (Muthamia
1996). Cowpea is commonly grown in mixtures with other crops such
(Sorghum bieolor (L.) Moench),
(Manihot eseulenta Crantz) (Mortimore
advantages of intercropping
maize (Zea mays (L.), and cassava
et al 1997; and Van Ek et al., 1997). The
are that it provides crops that enhance balanced diet and
reduces labour demand; it minimizes crop failure risk and adverse effects of pests,
maximizes
returns per unit area of land and reduces soil erosion
(Nadar,
1984).
However, cowpea grown in mixtures has low yields than those in pure stands. For
example
Muleba
and Ezumah
(1995) found that grain yields of cowpea
in field
intercrops ranged between 0-133 kg/ha compared to its potential yield of 1500 kg/ha
to 3000 kg/ha.
Factors limiting cowpea yields are low plant population density, low yield potential of
local cultivars,
insect pests and diseases, shading by the cereals, drought stress and
low soil fertility.
For example
cow pea growth is retarded
especially low phosphorus and micronutrients
by poor soil fertility
(Bationo et al., 1991).
52
6.2
Materials
and Methods
A total of 34 improved cowpea varieties were used in this study. Of the 34 varieties,
30 varieties were obtained from (lJTA) and consisted
maturing varieties;
of 15 early and 15 medium
two varieties from Maseno, TSBF -CIA T trials, and 2 varieties
were obtained from a local farmer. Early maturing varieties obtained from lIT A with
given codes
were:
IT97K-568-18
IT97K-461-4
(E4),
IT99K-1060
IT97K-568-18
(E5), IT97K-570-18
IT99K-II22
(E8),
IT98K-429-2
(EI2), IT96K-6IO (EI3), IT98K-428-3
Medium maturing
IT98K-I31-2
(El),
(E9), IT98K-506-1
(E2), IT97K-494-3
(E3),
(E6), IT97K-499-38
(E7),
(EIO), IT97K-356-1
(El I),
(EI4) and IT98K-1399
(EI5).
varieties with given codes included IT99K-491-7
(M1), IT95K-
193-12 (M2), IT98K-503-1
(M3), IT95K-207-22
(M4), IT99K-205-9
(M5), IT98K-
128-4 (M6), IT97K-499-35
(M7), IT95K-1073-57
(M8), IT97K-818-35
(M9), IT97K-
556-4 (MIO), IT98K-463-7
(MI I), IT97K-I021-24
(MI2),
IT97K-I075-7
(M13),
IT95K-52-34
(M14) and IT97K-564-1 (MI5). Varieties from Maseno included IT98K
247-2 (CPI4)
and IT97K -1068-7 (CP21), while local varieties used were Kathoka
and Kang'au.
The experiment
Institute (KARI) dryland
was established
at a Kenya Agricultural
research centre, Kiboko,
Research
960m above sea level, 2015'S;
37°45'E and ran for two seasons, during the short and the long rain seasons. The
experiment
was set up as a completely randomized
design (CRD) with each variety
replicated 3 times, in rows (Muli and Saha, 2000). Each row had 48 plants spaced at
20 x 60 cm (Audi et al., 1996), covering an area of 5.76 m2. The experiment was both
rain-fed and irrigated. Overhead sprinklers that were set to run for 3 hours at each
irrigation cycle, discharging an equivalent of about 32 mm rainfall, were used. In the
first season, planting was done on IOIl 11 2004 and irrigation was done when rainfall
was limiting.
In the second season, planting was done on 6/4/2005 and irrigation was
53
done as in short rains. In both seasons and at planting time, phosphorus
triple superphosphate
(TSP) (P20S,
was not limiting. Weeding
in the form of
0:46:0) was applied at 20kg/ha to ensure that P
was done 3 times each season. At flower onset, early
podding and at full podding in both seasons, the crop was sprayed with Karate® at a
rate of20 ml per 20 litres of water to control white flies and aphids. To eliminate edge
effect, only 40 plants were harvested leaving 4 plants on either side of the row. Data
collected included pod numbers, pod length, seeded pod weights, number of seeds in
each pod, seed weight per pod, number of branches, plant biomass production, grain
yields and, shell biomass, total above ground and weights of 100 seeds. To determine
grain dry weights, shell and shoot biomass, sub-samples
and dried in a Sanyo Connection Oven (MOV-2l2F,
were taken to the laboratory
Sanyo Electric Company, Japan)
at 60°C for 72 hours. However, to select pioneer species for further studies, means of
the assessed characters
were selected
were totaled. Cowpea varieties that had highest mean totals
for on-farm
trials. The data obtained
was analysed
Discovery Edition 1, GenStat Procedure Library Release PL12.2.
using GenStat
54
6.3
Results and Discussion
Pod weights, seeds per pod and seed weight per pod were generally higher during the
short rains than during the long rains (Table 6.1 and 6.2). This was probably because
of an infestation
(Nephtys
of the cowpea
spp). In addition,
length, seeded pod weights,
plants during early growth stages by cat worms
medium
maturing
varieties
had generally
higher pod
seeds per pod and seed weight per pod than the early
maturing varieties in both the short and long rain seasons (Table 6.1 and 6.2). Pod
lengths ranged from 11.2 cm to 17.2 cm and from 10.6 cm to 17.3 cm, respectively,
during the short and the long rains. These pod lengths were close to the range of 13.0
cm to 19.8 cm obtained
by Muli and Saha (2000) in a cowpea screening
trial at
Msabaha, in the Kenyan Coast. The pod lengths obtained in this study were also close
to the range of 7.3 cm to 15 cm recorded by Amanullah
and Hatam (2000) in
Pakistan. During the short and the long rains, mean pod length for all 34 varieties
were 13.4 cm and 13.5 cm, respectively,
and varieties that had above average pod
lengths in both seasons were El, E2, E6, E7, E9, EIO, M2, M4, M6, M7, M8, M9,
MIO, MII,
M13, MI4,
MI5,
Kathoka
and CP21. Of these varieties
58% were
medium maturing and only 32% were early maturing varieties. Longest pod lengths
were recorded in E6 (17.2 cm) and Kathoka (17.3 cm) during the short and long rain
seasons, respectively.
This implied that the local variety, Kathoka, had potential for
long pod formation that was comparable to that of the improved varieties.
Pod weight ranged from 1.1g to 2.2 g during the short rains and 0.9 g to 2.5 g during
the long rains with overall means of 1.7 g and 1.5 g, respectively (Table 6.1 and 6.2).
During the two rain seasons, varieties that had above average pod weights were El,
E2, E7, E8, EIO, M2, M4, M7, M8, M9, MIO, MII, MI2, M13, MI4, MI5, Kathoka,
55
CPI4 and CP21 with 58% of the varieties being medium maturing and 26% early
maturing varieties. However,
varieties with heaviest pods of 2.2g were E8, M9 and
CP21 during the short rains and M 13 (2.5 g) during the long rains. Kathoka had a pod
weight of2.3 g during the long rain, which was higher than the highest pod weight in
the short rains implying that the local variety had a high potential to produce heavy
pods. The correlations
respectively,
between
pod length and pod weight were 54% and 53%,
during the short and the long rains, implying that pod length influenced
pod weight in both seasons.
During both the short and the long rains, seed numbers ranged from 4 to 12. However,
the average seeds per pod (seed numbers) for the 34 varieties studied were 8 and 7,
respectively,
during the short and long rain (Table 6.1 and 6.2). These seed numbers
were comparable to cowpea seeds/pod recorded by Amanullah
and Hatam (2000) that
ranged from 7 to 15 with a mean of 9 seeds/pod. Varieties that had above average
seed numbers during both the short and the long rains seasons were E2, E7, E8, M2,
M4, M7, M8, M9, Mll,
M13, M14, Kathoka and CP21. Of these varieties 62% were
medium maturing while 23 % were early maturing varieties.
Highest seed numbers
(12 seeds per pod) were recorded in Kathoka and E8 during the short rains, and in
Kathoka (12 seeds per pod) during the long rains. This meant that the local variety,
Kathoka, was a heavy seeder and was comparable to the improved variety E8 with
respect to the number of seeds produced. Pod length influenced
the number of seeds
as reflected by significant (p<0.05) correlations between pod length and seed numbers
of 61 % and 69%, respectively,
during the short and the long rains. Seed numbers and
pod weights
correlations
had significant
of 72% (p<O.OI) and 61 % (p<0.05),
respectively, during the short and the long rains. Significant correlations between seed
56
numbers
and pod length were also reported by Nakawuka
and Adipala (1999)
111
Uganda.
Seed weight per individual
pod ranged from 0.9 to 1.7 g during the short rains and
from 0.6 to 1.8 g during the long rains. Average seed weights per pod for 34 varieties
were 1.3 g and 1.0 g, respectively,
during the short and the long rains (table 6.1 and
6.2). During the two seasons varieties that had above average seed weights were El,
E2, E5, E7, E8, ElO, E13, M2, M4, M7, M8, M9, MIO, Mll,
M12, M13, M14,
Kathoka, CP14 and CP21. Of these varieties medium maturing varieties contributed
50% while early maturing ones had 35%. This was the highest contribution
by the
early maturing varieties in the parameters so far measured (pod length, pod weight,
seed numbers and seed weight per pod). Highest seed weight per pod was recorded in
M8 and M9 (1.7) and M13 (1.8 g) during the short and long rains, respectively.
variety, Kathoka had seed weight per pod of 1.6 g and 1.5 g, respectively,
Local
during the
short and the long rains, implying that the variety had high potential for heavy seed
production. There were significant (p<0.05) correlations between seed weight and pod
lengths of 58% and 65%, respectively,
weights were also significantly
of 71 and 62 %, respectively,
significant
during the short and the long rains. Seed
(p<O.Ol) correlated to pod weights, with correlations
during the short and the long rains. The strongest and
(p<O.01) correlations
of 81% and 83%, respectively,
during the short and
the long rains were between seed number and seed weight. This implied that seed
weight was strongly influenced by seed numbers in both seasons.
57
Table 6.1
Pod characters assessed during the short rains
Variety
El
E2
E3
E4
E5
E6
E7
E8
E9
EI0
Ell
E12
E13
E14
E15
Ml
M2
M3
M4
M5
M6
M7
M8
M9
M10
Ml1
M12
M13
M14
MI5
CP14
CP21
Kathoka
Kang'au
Pod length
(cm)
12.5
12.5
13.3
11.2
11.6
17.2
11.9
12.8
13.8
14.2
11.3
13.1
12.9
12.7
12.6
11.8
15.1
12.9
14.2
11.6
13.8
13.6
11.8
15.0
15.4
13.6
13.3
12.3
Mean of 34 varieties
15.3
13.5
13.3
16.3
15.2
12.7
13.4
SED
0.42
Seeded pod
weight (g)
1.3
1.4
1.5
1.1
1.3
1.6
1.7
2.2
1.4
1.9
1.4
1.3
1.7
1.4
1.7
1.3
2.0
1.6
1.6
1.4
1.5
1.8
2.0
2.2
2.0
2.0
2.0
1.7
1.9
1.9
1.7
2.2
2.1
1.6
1.7
Seeds
numbers
Seed weight
per pod (g)
7
7
7
4
0.9
1.1
1.2
6
8
8
12
7
8
6
7
7
6
8
8
9
7
6
7
8
8
8
9
7
9
7
10
9
7
8
9
12
8
8
0.44
0.9
0.9
1.1
1.3
1.6
1.1
1.6
1.1
1.0
1.4
1.0
1.4
1.0
1.6
1.2
1.2
1.2
1.2
1.5
1.7
1.7
1.5
1.5
1.4
1.3
1.4
1.3
1.3
1.5
1.6
1.3
1.3
0.12
0.09
Significance
**
**
**
**
** Means p<O.OOI. E and M means early and medium maturing cowpea varieties,
respectively. Data values are means of 3 replicates.
58
Table 6.2
Pod characters assessed during the long rains
Variety
El
E2
E3
E4
E5
E6
E7
E8
E9
EI0
Ell
E12
E13
E14
E15
Ml
M2
M3
M4
MS
M6
M7
Pod length
(cm)
14
15.5
12.9
10.8
12.3
15.7
14.9
10.6
14.2
13.9
12.2
11.7
12.6
12.9
11.9
11.6
14.3
11.7
13.8
11.4
16.0
15.0
13.8
14.1
15.7
13.7
11.9
M8
M9
M10
M11
M12
M13
M14
MI5
CP14
CP21
Kathoka
Kang'au
15.9
17.3
11.7
Mean of 34 varieties
13.5
SED
0.75
15.3
15.3
12.7
13.2
Seeded pod
weight (g)
1.7
2.0
1.2
0.9
1.2
1.1
1.6
1.3
1.4
1.4
1.4
1.2
1.4
1.0
1.3
1.2
1.7
1.0
1.9
1.0
1.3
1.9
2.1
2.3
1.6
1.4
1.0
2.5
1.5
1.2
1.7
1.8
Seeds per
pod
7
9
6
4
6
6
8
7
6
7
5
6
6
5
6
6
9
6
8
6
7
8
10
6
7
10
4
Seed weight
per pod (g)
1.1
1.5
0.9
0.7
1.5
0.7
1.1
1.0
1.0
1.1
0.7
0.7
0.8
0.8
1.0
0.9
1.3
0.8
1.2
0.7
0.9
1.4
1.6
0.9
1.1
1.0
0.6
1.8
1.0
1
1.5
11
6
6
7
9
12
7
7
0.7
1.0
0.32
0.93
0.18
2.3
0.8
1.2
1.2
1.5
Significance
**
**
**
**
Means p<0.001. E and M means early and medium maturing cow pea varieties,
respectively. Data values are means of 3 replicates.
**
59
Branch numbers, pods per plant, plant biomass, grain yield, shell weight, total above
ground and weight of 100 seeds were generally higher during the short rains than
during the long rains. In addition, medium maturing varieties had generally higher
values for all parameters
measured during both the short and the long rain seasons
compared to the early maturing varieties (Table 6.3 and 6.4). Cowpea branching is an
adaptation for high and efficient translocation
development
branches,
by
of a substantial root system (Olufajo and Singh, 2000). The numbers of
nodes
intercropping
system that is usually accompanied
and
internodes
length
(Nelson and Robichaux,
are plant
traits
that are
important
in
1997). Thus cultivars with a bushy type habit
are high yielding under sole cropping, whereas the cultivars with a spreading habit are
higher yielding under intercropping
(Nelson and Robichaux,
1997). Branch numbers
varied from 2 to 6 during the short rains and from 0 to 5 during the long rains (Table
6.3 and 6.4). These branch numbers were lower than number of branches of between
6 and 14 documented
by Amanullah and Hatam (2000) at Pakistan, the differences
being probably due to differences in cowpea genotypes used in the two studies. Out of
the 34 varieties,
mean branch numbers were 3 in both seasons. Varieties that had
branches above average were El, E4, E7, Ell,
E14, Ml, M2, M9, MIO, M12, Ml3,
MI4, MI5, Kathoka
and Kang'au.
varieties
53% and 33%, respectively.
contributed
Of these varieties
medium and early maturing
Highest
branch numbers
were
recorded in the local variety, Kang'au (6 branches) during the short rains, and in El (5
branches) during the long rains. The later observation
probably implied that Kang'au
had a high potential of branch formation than most of the improved varieties.
60
Table 6.3
Variety charactersassessedduring the shortrains
Variety
El
E2
E3
E4
E5
E6
E7
E8
E9
EI0
Ell
E12
E13
E14
EIS
Ml
M2
M3
M4
MS
M6
M7
M8
M9
MI0
Mll
M12
M13
M14
MI5
CP14
CP21
Kathoka
Kang'au
Mean of 34 varieties
SED
Significance
BN
"
2
"
4
3
3
4
2
"
3
3
3
3
4
3
"
3
3
3
"
3
3
3
4
4
2
4
4
4
4
3
3
4
6
3
0.5
**
.)
.)
.)
.)
.)
PN
19
15
20
16
15
12
24
15
20
16
15
16
14
24
12
22
10
19
12
15
13
17
12
8
17
17
13
15
18
21
17
17
15
46
17
3
**
PB
1005
676
984
647
436
638
2881
803
1306
1773
986
687
784
1654
623
1226
559
1121
678
729
690
1539
1324
1178
2382
2232
791
2855
2280
1630
761
1386
1961
3933
1328
201
**
GY
1563
1359
1995
1190
1310
1094
2466
2023
1875
2213
1265
1378
1600
2052
1398
1867
1197
1845
1236
1424
1279
1955
1722
1079
1978
1917
1569
1567
2132
2318
1901
2203
1970
3435
1746
359
**
SWt
421
347
529
311
372
460
661
576
490
513
372
384
352
658
324
517
262
495
345
336
357
461
336
270
658
496
404
553
795
850
513
765
636
1110
489
lIS
**
TAG
2989
2383
3508
2148
2117
2192
6007
3401
3671
4500
2623
2449
2736
4363
2345
3610
2018
3461
2259
2488
2327
3955
3382
2526
5018
4644
2763
4975
5206
4798
3175
4354
4569
8479
3572
597
**
WI00
17
16
16
22
18
16
16
14
18
18
19
17
18
16
17
14
16
17
20
16
17
19
21
19
19
16
19
14
19
18
15
18
14
18
17
2
**
** Means p<O.OOl. E and M means early and medium maturing cowpea varieties, respectively. BNbranch numbers; PN- Pod numbers; PB- plant biomass kg/ha; GY - Grain yield kg/ha; SWt- shell
biomass kg/ha; TAG- Total aboveground kg/ha and WIOO- weight of 100 seeds.
61
Table 6.4
Variety
El
E2
E3
E4
E5
E6
E7
E8
E9
E10
Ell
E12
E13
E14
E15
M1
M2
M3
M4
M5
M6
M7
M8
M9
MI0
Ml1
M12
M13
M14
MI5
CP14
CP21
Kathoka
Kang'au
Mean of34
SED
Significance
Variety characteristics
assessedduring the long rains
BN
5
3
2
4
0
3
3
3
3
2
4
3
2
4
3
4
4
3
2
2
3
3
3
2
4
3
2
3
3
3
3
3
3
4
3
0.7
**
PN
30
18
14
15
4
14
16
13
12
13
13
10
10
16
13
6
10
12
8
10
13
12
18
6
19
8
7
15
12
14
10
13
8
10
12
3
**
PB
1762
1077
354
614
98
494
758
535
825
737
548
352
362
929
358
329
508
527
531
441
681
693
1320
737
1329
367
171
1479
691
462
423
746
1332
469
678
201
**
GY
2686
2245
)066
854
304
816
1379
1070
835
1214
756
585
631
460
1050
431
1035
841
693
581
1026
1445
2484
412
1722
718
367
2220
1264
985
956
1277
1031
643
1061
359
**
SWt
1120
754
352
289
106
456
646
371
371
337
312
221
206
375
267
148
298
258
254
221
427
439
579
202
704
252
133
873
569
389
299
572
508
229
398
115
>!<*
TAG
5569
4075
1772
1749
508
1766
2882
1976
2030
2289
1616
1156
1200
1764
1674
908
1841
1626
1478
1243
2130
2578
4384
1352
3755
1337
671
4571
2524
1834
1679
2595
2872
1341
2140
597
**
WI00
17
17
16
20
17
17
16
14
20
17
17
14
17
15
17
14
19
19
21
18
18
18
20
14
19
14
17
21
19
16
17
18
11
18
17
9
**
** Means p<O.OOI. E and M means early and medium maturing cowpea varieties, respectively. BNbranch numbers; PN- Pod numbers; PB- plant biomass kg/ha; GY - Grain yield kg/ha; SWl- shell
biomass kg/ha; TAG- Total aboveground kg/ha and WIOO- weight of lOO seeds. Data values are means
of3 replicates.
I
•
62
Pod numbers ranged from 8 to 46 during the short rains but dropped to a range of 4 to
30 during long rains. The ranges were higher than those obtained by Muli and Saha
(2000) of 8 to 18 at Mtwapa and 7 to 12 at Msabaha, at the Kenyan Coast. However,
pod numbers obtained were lower than those obtained by Owolade et al., (2006), at
Ibadan, Nigeria, that ranged from 50 to 75 pods per plant. Of the varieties tested,
varieties that had above average pod numbers were El, E2, E3, E4, E6, E7, E8, E9,
EIO, Ell,
E13, E14, E15, MI, M3, M6, M8, MIO, M13, M14, MI5, Kang'au
and
CP21. Of these varieties, 57% were early maturing while 35% were medium maturing
varieties. This observation
probably indicated that early maturing varieties produce
high number of pods compared to medium maturing ones. In addition, highest pod
numbers of 46 pods per plant during the short rains was recorded in the local variety,
Kang'au while during the long rains El had the highest pods per plant (of 30). This
observation probably meant that the local variety was superior to the tested improved
varieties in pod production
production.
during the short rains and had high potential
Branch and pod numbers were significantly
(p<0.05) correlated
for pod
(55%)
during the short rains but the correlation dropped to 46% during the short rains. A
positive correlation
between branch and pod numbers of 46% was also documented
by Nakawuka and Adipala (1999) in Uganda.
Plant biomass of the tested varieties ranged from 436 to 3933 kg/ha during the short
rains and from 98 to 1762 kg/ha during the long rains. This biomass was comparable
to the biomass production
of between 960 to 3590 kg/ha obtained by Olufajo and
Singh (2000) in Nigeria. Average biomass of the 34 varieties was 1372 kg/ha during
the short rains and 678 kg/ha during the long rains. Varieties that had above average
biomass in both seasons were El, E2, E7, E9, EIO, E14, M6, M7, M8, M9, MIO,
63
M11, M13,
contributed
produced
Kang'au
M14,
MI5,
Kathoka,
Kang'au
33% while medium-maturing
and CP21.
varieties
Early maturing
varieties
had 50% of the varieties that
biomass above average. Highest biomass of 3933 kg/ha was recorded in
during the short rains but E I produced highest biomass (1762 kg/ha) during
the long rains. This observation
implied that Kang'au
biomass
to the tested
production
significant
compared
(p<0.05) correlation
had a very high potential for
improved
varieties.
between biomass production
There
was a
and branch numbers of
62% during the sh0I1 rains that dropped to 45% during the short rains. However, the
correlations
between
biomass
production
and pod numbers
were significant
71 %
(p<O.Ol) and 66 % (p<0.05), respectively, during the short and the long rains.
Grain yield ranged from 1079 to 3435 kg/ha during the short rains and from 304 to
2686 kg/ha during the long rains. These ranges were close to the seed yields obtained
by Muli at Saha (2000) of 1600 to 2800 kg/ha at Mtwapa and 1300 to 1980 kg/ha at
Msabaha along the Kenyan Coast. The yields were also close to grain yields of 1499
to 2739 kg/ha obtained by Olufajo and Singh (2000) at Nigeria. The grain yields were
also close to the yields obtained by Owolade et al., (2006) of between 983 to 1498
kg/ha. Average grain yield for the 34 varieties was 1746 kg/ha during the short rains
but dropped to 1061 kg/ha during the long rains. Varieties that had above average
grain yields in both rain seasons were El, E2, E3, E7, E8, E9, EIO, E14, MI, M3, M7,
M8, MIO, M11, M13, M14, Kathoka, Kang'au, CP14 and CP21. Of these varieties
early maturing
varieties
contributed
40% compared
to 40% contribution
from the
medium yielding varieties. During the short rains, highest grain yield (3435 kg/ha)
was recorded in the local variety, Kang'au while in the long rains highest grain yield
(2686 kg/ha) was recorded in variety El. This observation indicated a very high grain
64
yield potential
of the local variety, Kang'au,
varieties.
observation
The
significantly
during
the long
compared
rains
diversification
correlation
during
Kang'au
improved
grain
yields
of the local varieties to
of local varieties accompanied
of the locally grown varieties. Grain yield had a significant
by
(p<0.05)
(59%) with branch numbers during the short rains that dropped to 32%
the long rains. Pod numbers
correlations
where
dropped could have indicated the susceptibility
pests and therefore need for efficient management
to the tested
and grain yields
of 89% and 73%, respectively,
had significant
(p<O.Ol)
during the short and the long rains. In
addition, grain yield had significant (p<O.O1) correlations with biomass production of
80% and
correlations
73%,
respectively,
between
during
and the
grain yield and branch numbers,
yields of 86% and 89% respectively
(1999) in Uganda.
the short
were documented
long rains.
Significant
and pod numbers and grain
by Nakawuka
and Adipala
From the data obtained in the study, it appeared that cat worm
attack not only affected yield parameters but also their correlation coefficients.
Shells (pod shells) remain after seeds are removed from pods. Unlike leaves, most of
which disintegrate during threshing, shells make a good proportion of legume biomass
that is fed to livestock after grain is removed. Shells are therefore
considering
legume remains after grains are removed.
important when
In this study, shell biomass
(biomass) ranged from 262 to 1110 kg/ha during the short rains and from 106 to 1120
kg/ha during the long rains. However, average shell biomass for the 34 varieties was
489 kg/ha during the short rains and 398 kg/ha during the long rains. Varieties that
had shell biomass above average in both seasons were El, E2, E3, E6, E7, E8, E9,
EIO, E14, M1, M3, M6, M7, M8, M10, Mll,
M13, M14, MI5, Kathoka, Kang'au,
CP14 and CP21. Of these varieties 39% were early maturing
varieties
and 43%
65
medium
yielding
varieties.
Highest
shell biornass
(1110 kg/ha) was recorded
Kang'au
during the short rains while El had highest shell biomass
in
(1120 kg/ha)
during the long rains. These are the same varieties that had highest grain yield during
the short and the long rains, respectively.
potential for shell biornass production.
Again the local variety
showed high
Shell weight had positive correlations
with
branch num bers of 57% (p<0.05) and 39%, respectively, during the long and the short
rains. However,
significant
respectively.
the relationship
between shell weight and biornass produced was
(p<O.OI) being 74% and 80% during
A higher and significant
the short and the long rains,
(p<O.OI) relationship
shell biornass and grain yield of 89% and 85%, respectively,
long rains. Further,
there was a significant
(p<0.05)
biomass and pod numbers of 72 and 66%, respectively
was obtained between
during the short and the
relationship
between
shell
during the short and the long
rams.
Total aboveground
was the sum of plant biornass, seed yield and shell biornass. It
ranged from 2018 to 8479 kg/ha during the short rains and from 508 to 5569 kg/ha.
Mean of the 34 varieties was 3572 kg/ha during the short rains and 2140 kg/ha during
the long rains. Varieties with above average total aboveground
(TAG) were El, E2,
E7, E9, EI0, E14, MI, M7, M8, MIO, MII, M13, M14, MI5, Kathoka, Kang'au and
CP21. Medium maturing varieties made up 47% while early yielding varieties made
35% of the varieties with TAG in both seasons. Highest TAG (8479 kg/ha) during the
short rains was recorded in Kang'au while highest TAG (5561 kg/ha) during the long
rains was obtained in El. Kang'au again showed superiority
in TAG.
over improved varieties
66
Weight of 100 seeds (WIOO) ranged from 14 to 22 g during the short rains and 11 to
21 g during the long rains. This seed weight range was close to the seed range of I I to
20 g documented
by Ogbonnaya
et al., (2003) in Nigeria. Mean W I00 of the 34
varieties was 17 in both seasons. Varieties that had above average W 100 were E4, E5,
E9, EIO, Ell,
Kang'au
EI3, M2, M3, M4, M5, M6, M7, M8, M9, M 10, Ml2 M14, MI5,
and CP2I.
Of these varieties 30% were early maturing and 60% medium
maturing. During the short rains highest (22 g) W I00 was recorded in E4 and in the
long rains M4 and M13 had highest (21 g) WIOO. In addition, local variety, Kang'au
was among the above average varieties, again placing the local variety among the
superior cowpea varieties.
To determine
best performing
best cowpea
varieties,
all characters
varieties
that were assessed
that were identified
in order
were
summed
up. Nine
included
Kang'au,
M13, E7, MIO, El, M8, MI4, Kathoka and CP21 (Table 6.5). However,
M13 was found to be very susceptible to drought and insect attack and was therefore
not included
in the list of the nine cowpea
varieties
selected
for on-farm
trials.
Kathoka was also omitted to give way to CP2] that was obtained from Maseno and
closely resembled
Kathoka,
showed resistance
to drought in a rain-fed school demonstration
recommended
and had close character
totals. Varieties
M7 and E6
trial and farmers
the two cowpea varieties for further testing. Therefore,
nine cow pea
varieties that included Kang'au, lT97K-568-18
(E 1), IT97K-570-18
(E6), IT97K-499-
38 (E7), IT97K-499-35
(M8), IT97K-556-4
(MlO), IT95K-52-
(M7), IT95K-I073-57
34 (MI4) and IT97K-1068-7
(CP21), were selected for on-farm trials.
67
Table 6.5
Character ranking of varieties selected for on-farm trials
Varieties selected
Kang'au
E7
MIO
El
M8
MI4
CP21
M7
E6
Variety
Kang'au
M13
E7
MIO
El
M8
M14
Kathoka
CP21
EI0
MI5
M7
E2
E14
Ml1
E9
E8
E3
M3
CP14
Ml
M6
Ell
E15
E6
E13
E4
M9
M2
M4
M5
E12
M12
E5
Position
Character total
I
2
19785
19220.9
17807.5
17679.3
17251.5
3
4
5
6
7
8
15659
15587.4
14998
14026.9
13694.1
13386.4
13188.2
9
10
I1
12
13
14
15
13037
12374.8
12075.2
11524.9
10864.5
10675
10289.2
9819.4
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
9140.8
9033.7
8588.1
8147.9
8032.4
7978.8
7916.6
7860.2
7834
7587.9
7567.3
7317
.6972.2
33
34
5348.8
IlJ:NVATTll IIMI\/::P.~lTV I 1r:lIlADVJ
!
68
CHAPTER SEVEN
EFFECTS OF ISFM ON NODULATION, GROWTH AND GRAIN
YIELD OF SELECTED COWPEA VARIETIES
7.1
Cowpea (Vigna unguiculata
Introduction
(L) Walp.) is one of the earliest plants cultivated by man
and the third most important grain legume in Kenya after common beans and pigeon
pea. It is grown in the semi-arid areas, with most of it in eastern province. Cowpea is
used for fodder, food and for soil fertility improvement
(Muthamia
and Kanampiu,
1996; Tarawali et aI., 1997).
Cowpea growth is retarded by poor soil fertility especially
micronutrients
(Bationo
et al., 199 I). However,
low phosphorus
its growth
(P) and
can be improved
addition of inorganic P and nitrogen (N) (Audi et al., 1996). Phosphorus
by
can be added
at a rate of 13-25 Kg/ha as diammonium phosphate or as triple superphosphate
and N
at 10-15 kg/ha (Audi et al., 1996; Shetty et al., 1995; Subarao et al., 1999).
Phosphorus
is an essential
component
of all living systems and in higher plants it
functions
as a constituent
of nucleic acids and proteins
important
in cell division
and induces root growth, promotes
increases disease resistance
Nelson, 1975). Phosphorus
anthocyanin
(Marschner,
1996; Owolade
seed formation
and
et al., 2006; Tisdale and
of purple
pigment and stagnant growth. The condition is more pronounced during
seed
Deficiency
1996). It is
deficiency in young plants leads to accumulation
the periods of adverse weather when growth processes
during
(Marschner,
formation,
the leaves
turn yellow
are slow. If P is deficient
starting
with the older
leaves.
of P is also shown by formation of small, dark green leaves that stand
69
more upright than do the normal leaves (Marschner,
1996). Crops growing In P
deficient soils also exhibit delayed crop maturity, and reduced quality and quantity of
crop yields (Fairhurst et al., 1999).
The primary source of P is mineral apatite found in primary rocks and apatite is the
most commonly occurring phosphate mineral in rocks. Other sources of P are organic
matter and secondary
and complex compounds in the soil. The amount of organic P
varies greatly among soils and within soil profile and processes that lead to increased
soil organic matter also increase P content of the soil (Wild, 1988). Phosphorus
absorbed
into plants in ionic form, mainly as H2P04-
availability
phosphates
is greatest
precipitate
between
(Marschner,
1986) and its
pH 6.0 to 7.2. Iron (Fe) and aluminium
at low pH «5.5)
and calcium (Ca) phosphates
is
(AI)
at high pH (>
8). However, the available phosphorus supply in the soils depends on the amounts and
forms ofP present in the soil (Kamprath, 1991).
Phosphorus
deficiency
is widespread covering an area estimated at over 2 billion ha
(lama et al., 1997; Smaling, 1993) and it may result from 10w-P status of the parent
material,
weathering,
between
nutrient
(Fairhurst
long-term
anthropogenic
mismanagement
inputs and exports, and P loss by erosion
et al., 1999). However, soil P deficiency
primarily
through
imbalance
and surface run-off
resuits from either
inherent low P levels or depletion of soils (Buresh et al., 1997) but soil P can be
replenished
by addition of inorganic fertilizers, organic matter in form of plant and
animal residues or phosphate
(Chien and Menon, 1995).
rocks such as Busumbu and Mijingu phosphate rocks
70
Continuous
cropping,
removal of crop residues
to feed animals
and overgrazmg
between cropping seasons with little or no external inputs have reduced the productive
capacity of arable lands and thus threatened
systems
in sub-Saharan
Kenya,
decline
the sustainability
of food production
Africa (Sanchez et al., 1997; Stoorvogel
in crop yields
is a major
problem
facing
et al., 1993). In
smallholder
farmers
(Mathuva et al., 1996). This is attributed to the high costs of inputs that make the use
of inorganic
fertilizers
on staple food crops uneconomical
for most smallholder
farmers (Jama et al., 1997).
Use of organic
alternative
inputs as an external
to expensive
source of soil nutrients
fertilizers to smallholder
are low in nutrient concentration
is a logical cheap
farmers. However,
organic inputs
compared with inorganic fertilizers (Sanchez et al.,
1997) although cattle manure has been one of the most commonly used ways of soil
fertility improvement
for crop production
1997). For example,
in the semi-arid
phosphorus
1961; Giller et al.,
areas of eastern Kenya where nitrogen and
limit crop production, farmyard manure is commonly used to enhance soil
fertility and crop production (Gibberd,
Recent
in Africa (Dennison,
research
1995; Ikombo,
has shown that a combination
1984; Kihanda et al., 2004).
of organic
and inorganic
inputs
enhances crop production and reduces cost of inorganic fertilizers (Ojiem et al., 2004;
Okalebo et al., 2004). The combination
of organic and inorganic
inputs is termed
integrated soil fertility management (ISFM). Therefore the objective of this study was
to determine the effect of integrated soil fertility management
and grain yield of selected
amounts in Makueni District.
improved
cowpea varieties
on nodulation, growth,
under contrasting
rainfall
71
7.2
Materials and Methods
Two farms were selected
in two sites with contrasting
were located at Kavuthu
Sub-location
7"S, 037°25'23.6"E,
in Mbitini division,
and at Ndunguni Sub-location
02°04'44. 2"S, 037°34'44.3"E.
rainfall amounts. The farms
1223 m.a.s.l, 0 I°59' 52.
in Nguu division,
1082 m.a.s.l,
The farms were selected based on the willingness of
the farmers to relinquish their farms for trials, farm uniformity and availability of the
required farm sizes. First planting was done in the long rains of 2006 and a second
one in the short rains in the same year. Ox-ploughs
were used to prepared land for
planting during the long rains but hand hoes were used to prepare land during the
short rains to avoid mixing up of the treatments applied during the long rains. Before
planting, soils were collected in the upper 20 cm at 10 locations in each plot, mixed
and sub-samples
taken for analyses (Mathuva
et al, 1996). The soils along with
animal manure used in on-farm trials were analysed as described by Okalebo et al.,
(2002) (Table 7.1). Each trial site had 3 blocks and 36 sub-plots in each block. Each
sub-plot measured 3 m2 (100 cm x 300 cm). A distance of 60 cm and 80 cm separated
the sub-plots and blocks, respectively.
treatments
within
the blocks
GenStat programme
and varieties
within
was used to randomize
the treatments
(Appendix
3).
Treatments applied consisted of; I) A control, no inputs were applied (T!), 2) Animal
manure at 2.5 t/ha (T2), 3) P as TSP (P20S, 0:46:0) at 15 kg/ha (T3) and 4) Animal
manure + TSP (T4) at the above rates. Treatment
application
application
was done using spot
method to ensure that the crop was in contact with the nutrients. Nine
cowpea varieties, eight of which were improved varieties (Chapter 6); IT97K-568- 18,
IT97K-570-18,
IT95K-52-34
IT97K-499-38,
Kang'au
IT97K-499-35,
and IT97K-I068-7
Kang'au and CP21, respectively,
IT97K-I073-57,
IT95K-556-4,
coded El, E6, E7, M7, M8, MIO, M14,
were planted in each trial site. The experiment was
72
set up as a split plot design with nine varieties and four treatments that were replicated
in three blocks. Cowpea was planted at a spacing of 20 cm x 60 cm within sub-plots.
During planting, 3-4 seeds were planted per hole and thinned to one plant per hole
after 2 weeks. Weeding was done as required to ensure the plots remained clear of
weeds over the cropping
seasons. In both seasons, spraying was done at 50% crop
flowering to control insect pests using Bestox® 100EC at a rate of 10 ml per 20 litre
water/ha. Plant samples were taken at 50% flowering and at crop maturity. At 50%
flowering, plant samples were carefully dug out, cut at the root color, the root system
wrapped
in a paper, stored in a cooling box and brought to the laboratory.
laboratory,
nodules were dried and nodule biomass was recorded.
weight determination,
Connection
the samples
In the
For nodule dry
were dried at 60°C for 72 hours in a Sanyo
oven (Model MOV-212
F, Sanyo Electric Company,
Japan). At crop
maturity, mature pods were separated from the plants and the plants were cut at the at
root color. In the laboratory, pods were shelled and the seeds and shoot system were
dried and weighed as above. Data obtained was analysed using GenStat Discovery
Edition
1, GenStat
Procedure
Library Release PL12.2,
and treatment
means were
separated using standard errors of differences of means (SED).
Table 7.1
Chemical
characteristics
of soils collected
at on-farm trial sites and
animal manure applied to the on-farm trials
Source
Manure
On-farm
Ndunguni
On-farm
Kavuthu
PHin
water
8.8
Conductivity
Nitrogen
(mmhos ern")
(%)
1.4
4.2 I
Carbon
(%)
18
Organic
matter (%)
30.9
Phosphorus
(ppm)
(Olsen P)
5
7.6
6.0
0.09
12
19.9
4
7.7
2.9
0.05
6.5
11.1
3
73
7.3
7.3.1
Results
Nodule biomass at 50% flowering
During the long rains, nodule biomass ranged from 3 to 334 mg plant" at Kavuthu
while highest nodule biomass of 28 mg plant" was recorded at Ndunguni (Table 7.2).
Generally
treatment
significant
(p<0.05)
addition
enhanced
nodule
formation
increases especially at Kavuthu.
in most varieties
However,
with
the highest nodule
biomass of 334 mg plant" at Kavuthu and 28 mg plant" at Ndunguni were recorded in
M 14 in a manure +TSP treatment.
During the short rains, highest nodule biomass of 300 rng plant"
was recorded at
Kavuthu, whi le the highest nodule biomass at Ndunguni was 833 mg plant" (Table
7.3). The highest nodule biomass of 300 mg plant" recorded at Kavuthu was recorded
in Kang'au
in a manure+TSP
treatment, while at Ndunguni
highest biomass of 833
mg plant" was again recorded in Kang'au but in a P (TSP) treatment. In both sites,
treatment addition enhanced
most varieties.
nodule biomass with significant
(p<0.05)
increases in
74
Table 7.2
Nodule
Site
Variety/Treatment
El
E6
biomass
Kavuthu
Control
23a
100a
c
(mg/plant)
at 50% flowering
Manure
14a
49b
SC
E7
3
SSb
22C
M7
c
SOb
12
M8
98c
74C
MIO
SOd
182b
M14
Kang'au
97a
IS0a
34c
CP21
74
47
Treatment means
0.069
P-Value Variety
P-value Treatment
0.377
P-Value
treatment*Variety
O.SIS
44
SED Variety
27
SED Treatment
SED
variety*treatment
82
Same letter on the same row by site indicates
replicates.
8r
TSP
39a
55b
93a
67b
109a
212a
113c
104a
67b
during the long rains
Manure+TSP
13a
32b
9S
Ndunguni
Control
3a
Manure
Oa
a
SIb
3
87a
l2c
14Sb
334a
24b
2Sc
80
Ib
Ib
7a
lC
TSP
Manure+TSP
2a
2a
Oa
Oa
4a
la
la
Oa
2a
1b
6a
Sa
1b
Sa
3b
2b
c
28a
16
a
7b
3b
S
63
Ob
Ob
b
20
8b
1be
4b
Oc
3
3
9
6
0.034
0.388
0.318
4
2
7
that treatments
are not significantly
different
at (p<O.OS). Data values are means of 3
75
Table 7.3
Nodule biomass (mg/plant) at 50% flowering during the short rains
Site
Kavuthu
Ndunguni
Variety/Treatment
Manure
TSP
Manure
Control
TSP
Manure+TSP
Control
Manure+TSP
33c
133a
6r5
675
El
100a
67a6
335
33c
a
a
a
a
b
a
Ob
OC
133
100
E6
133
100
33
100
a
b
a
b
Ob
Ob
Ob
Ob
133
67a
67
E7
33
a
a
Ob
Oa
Oa
Ob
Ob
M7
67
33
a
a
b
a
b
Oc
OC
Ob
M8
100
133
33a
100
33
iooM10
200a
100b
200a
267a
100b
67d
133b
100b
roe=
M14
133ab
33c
133a
33c
167a
33c
Kang'au
133c
233b
833a
133c
233b
300a
267b
33d
c
b
c
b
c
bc
100a
3673
33
67
CP21
167
67
33
100
Treatment means
41
78
178
105
118
104
70
85
P-Value Variety
0.021
0.054
P-value Treatment
0.185
0.620
P-Value
0.416
treatment*Variety
0.894
SED Variety
42
92
SED Treatment
28
73
SED
variety*treatment
85
212
Same letter on the same row by site indicates that treatments are not significantly different at (p<0.05). Data values are means of 3
replicates.
6r
76
7.3.2
Shoot biomass production at crop maturity
At Kavuthu biomass production ranged from 1219 to 3301 kg/ha compared to 131 to
1608 kg/ha at Ndunguni
(Table 7.4). In addition, the highest shoot biomass of 3301
kg/ha at Kavuthu was recorded
in Kang'au
in a manure+TSP
treatment,
while at
Ndunguni highest biomass of 1608 kg/ha was recorded for M7 in a TSP treatment.
Generally, addition of treatments enhanced biomass accumulation
the long rains and treatment
addition
resulted
in both sites during
to significant
(p<0.05)
biomass
increases in most varieties
During the short rams, a generally
compared
to that recorded
higher biomass was recorded
during the long rains. At Kavuthu
in the trial sites
biomass
produced
ranged from 3090 to 7283 kg/ha compared to a range of 4834 to 12725 kg/ha at
Ndunguni
(Table 7.5). However,
recorded in Kang'au
kg/ha at Ndunguni
highest biomass of 7283 kg/ha at Kavuthu
in a manure+TSP
treatment while highest biomass of 12725
was recorded in M8 in a control treatment.
short rains, treatment application
was
enhanced biomass production
Further, during the
in both Kavuthu and
Ndunguni with significant (p<0.05) increases in most varieties.
7.3.3
Grain yield
During the long rains, grain was only harvested at Kavuthu (Table 7.6) and ranged
between 586 to 2235 kg/ha and highest yield of 2235 kg/ha was recorded in variety
El in a manure treatment.
As in biomass production,
grain yield was generally
enhanced by treatment application with significant (p<0.05) increases in most cow pea
varieties.
77
Table 7.4
Shoot dry weight (kg/ha) at crop maturity during the long rains
Site
Variety/
Treatment
El
E6
E7
M7
M8
MIO
M14
Kang'au
CP2l
Treatment means
P-Value Variety
Kavuthu
P-Value Treatment
P-Value Variety*
Treatment
0.291
0.987
0.677
0.45
Control
1663cb
2294b
1569b
1424bc
1794ab
1630c
1369c
2036c
1846a
1736
0.386
Ndunguni
Manure
2630a
1835c
1863a
1513b
1821 ab
2141 b
1944b
1341 d
2505a
1954
TSP
1869b
2766a
1933a
1219c
1869a
1760c
2435a
2485b
1249c
1953
M+TSP
1444c
2530ab
1388b
2396a
1566b
2977a
2024b
3301 a
2521a
2238
Control
722a
325c
344c
367c
1544a
711c
472c
1147a
614b
694
0.358
Manure
4835
800b
269c
702b
525b
827b
1211 a
c
555
925a
700
TSP
180c
772b
689b
1608a
131 c
1130a
455c
a
1180
c
197
705
Manure + TSP
392b
1127a
986a
664b
161 c
811 b
752b
b
962
950a
756
SED variety
209
336
SED treatment
191
257.6
SED variety *
treatment
749
538.4
Same letter on the same row by site indicates that treatments are not significantly different at (p<0.05). Data values are means of
3 replicates.
78
Table 7.5
Site
Shoot dry weight (Kglha) at crop maturity during the short rains
Kavuthu
Ndunguni
Variety/Treatment
Manure
Control
TSP
M+TSP
C
c
a
3671
El
3829
5989
55596
E6
4615c
4373c
5042b
5542a
c
a
a
3224
E7
5478
5148
4215b
b
a
a
M7
3468
4854
4737
4576a
c
a
4851 b
5426a
M8
4296
5817
a
MI0
4332b
4951
4743ab
5026a
ab
c
bc
M14
5253
4762
5115
6150a
c
5501 b
Kang'au
5528b
3090
7283a
CP21
4296b
4454b
3290c
5642a
Treatment means
4891
4398
4667
5491
P-value variety
0.375
P-value treatment
0.042
P-value
variety*Treatment
0.582
SED variety
479
SED Treatment
414
SED
Variety*Treatment
1176
Same letter on the same row by site indicates that treatments are not
means of 3 replicates.
11-
Control Manure TSP
5764b
7719a
61756
b
b
5559
6064
7447a
a
b
6650
7139a
5723
ab
c
7914
6370
7336b
a
b
12725
8685
8530b
6695c
1231
7072c
a
a
8677
8985
5323b
a
b
10748
6253
5017c
7119b
10835a
7108b
7880
8208
6794
0.343
0.438
r
Manure
TSP
5631b
7369a
7455a
8533a
4834c
8144b
9088a
11048a
6242b
7593
+
0.297
1116
893
2574
significantly different at (p<0.05). Data values are
79
During the short rains, grain yield ranged from 586 to 2524 kg/ha at Kavuthu and
from 344 to 2368 kg/ha at Ndunguni (Table 7.7). In addition, at Kavuthu, highest
grain yield of 2524 kg/ha was recorded in E I in manure+ TSP treatment whi le at
Ndunguni
same variety
had highest grain yield of 2368 kg/ha in a control
treatment. In addition, treatment addition generally enhanced grain yield in most
varieties with significant (p<0.05) increases in both study sites.
Table 7.6
Grain yield at Kavuthu during the long rains
Manure
Variety
Control
Manure
TSP
TSP
c
a
b
c
El
716
2235
1352
586
a
c
b
I747b
E6
2060
1013
1674
b
a
841 c
E7
1075b
1094
1552
a
a
891 a
M7
844
702a
750
c
b
a
M8
855
1I02
1397
747c
1427a
1272ab
MIO
1322a
1105b
c
a
741 c
M14
594
1910
1416b
Kang'au
1344b
2127a
922c
1280b
CP21
1408b
164r
1455ab
1183c
Treatment
1213
means
1152
1186
1380
P-value variety
0.545
P-value
treatment
0.671
P-value
Variety*treatme
nt
0.392
SED variety
363.7
SED treatment
198.5
SED
variety*treatme
nt
631
Same letter on the same row indicates that treatments are not significantly
different at (p<0.05). Data values are means of 3 replicates.
+
80
Table 7.7
Grain yield (Kg/ha) during the short rains
Site
VarietylTreatment
El
E6
E7
M7
M8
MI0
M14
Kang'au
CP21
Kavuthu
Control
914d
1100d
586c
1258d
1069c
1433a
1491 c
1558c
1619b
Ndunguni
Manure
2169b
1202c
1438b
1760c
1508b
1041 b
822d
1813b
1358c
TSP
1685c
2010a
1483b
1958b
2163a
1055b
1758b
1816b
1122d
M+TSP
2524a
1735b
1724a
2380a
1477b
1150b
2235a
2452a
1849a
Control
2368a
1413b
922c
1266b
1083c
1113b
127r
344b
1452a
Manure
969d
783c
1488b
1572a
1599a
1647a
866b
536b
1508a
TSP
1827b
1261b
1763a
1449a
1369b
1472a
991 b
397b
1597a
Manure+
TSP
1619c
1624a
977c
1444ab
722d
1166a
964b
875a
1438a
1249
1497
Treatment means
1203
1225
1457
1672
1347
1219
0.114
P-value variety
0.119
P-value treatment
0.852
0.006
P-value
0.73
variety*Treatment
0.767
329.3
SED variety
249.8
178.4
SED Treatment
202
SED
Variety*Treatment
581.2
568.6
Same letter on the same row per site indicates that treatments are not significantly different at (p<0.05). Data values are means of 3
repl icates.
- -
---- --- - ..
-- •...... ..... .-.-,-.
--~
81
7.4
7.4.1
Discussion
Nodule biomass
Lower nodule biomass recorded at Ndunguni compared to that recorded at Kavuthu
was most likely due to low total rainfall received at Ndunguni (194 mm) compared to
a higher rainfall (233 mm) received at Kavuthu during the long rains. In addition, it
was noted that most of the rainfall recorded at Ndunguni
fell during the last two
weeks of April, which were also the first two weeks after cowpea was planted and
about three weeks before cowpea reached 50% flowering. Therefore
samples
were being taken for nodule assessment
when cowpea
at 50% flowering,
the soil was
already dry (had limited moisture) and therefore very few nodules were recovered and
in some plants no nodules were recovered from cowpea plants at Ndunguni. The few
nodules recovered at Ndunguni
under limited soil moisture could have been caused
,
•
by water stress on the cowpea plants that most probably reduced nodule formation.
Sprent
(1971)
suggested
that
water
stress
affect
formation
and
longevity
of
leguminous root nodules.
In addition to rainfall differences, treatment differences in nodule biomass were noted
in both study sites although most treatment differences were noted at Kavuthu than at
Ndunguni
treatment
were
probably
indicating
effectiveness.
recorded
that high rainfall
At Kavuthu,
in TSP
manure+ TSP (80 mg/plant)
treatments
probably
amounts
at Kavuthu
enhanced
highest mean nodule biomass (95 mg/plant)
compared
to
manure
(74
mg/plant)
and
indicating that the cowpea plants used TSP
more efficiently for nodule formation than they did use other nutrient forms. Miao et
al., (2007)
studying
in the effect of P supply on nodule
formation
in soybean
I
82
observed that nodule mass was increased by P supply through
increased
nodule
development and functioning.
Results of nodule biomass
application
did not always
during the long rams further indicated
that treatment
increase nodule biomass and that cow pea response to
treatment application varied with site and variety. For example, during the long rains,
nodule
biomass
of M14
at Kavuthu
was increased
by all treatments
while at
Ndunguni nodule biomass of the same variety was increased by only two treatments,
manure and manure+TSP
treatments. In the same season, nodule biomass of Kang'au
was not affected by treatment application at Kavuthu but was increased by addition of
TSP at Ndunguni.
This observation
agreed with Audi et al., (1996) who suggested
that cowpea does not always respond to nutrient inputs. During the short rains, a
relatively higher nodule biomass was recorded at Ndunguni than at Kavuthu although
more total seasonal
Ndunguni
rainfall was still recorded at Kavuthu (397 mm) compared to
(386 mm). The differences
observed in nodule biomass in the two sites
during that season could have been caused by site differences that probably included
soil chemical characteristics
(Table 7.1).
During the short rains, the highest mean nodule biomass (178 mg/plant) was recorded
in TSP treatment
at Ndunguni
nutrient for nodule formation.
further indicating that TSP was the most effective
However,
as opposed to the observations
made on
treatment effect on nodule biomass during the long rains, where treatment effect were
common at Kavuthu,
probably
indicating
nodule biomass in both sites indicated clear treatment
the importance
of rainfall in treatment
effectiveness.
effect
Also as
83
observed in the long rains, not all treatment additions enhanced nodule biomass and
that variety response to treatment addition varied with site. For example, during the
short rains, there was no effect of treatment addition in nodule biomass of M 14 at
Kavuthu while at Ndunguni nodule biomass of the same variety was increased by two
treatments,
manure and manure+ TSP, as in the first season. Besides site variation,
treatment
effect on nodule
response to treatments.
biornass also indicated
Observations
seasonal
variation
of variety
made on nodule biomass could be compared to
the studies done by Jemo et al., (2006), who, when studying the effect of P in low
phosphorus
soils of southern
treatments,
Cameroon, recorded highest nodule biomass in TSP
and obtained nodule biomass that ranged from 234 to 676 mg/plant that
was within the nodule biomass range obtained in this study.
7.4.2
Shoot biomass
As observed
significantly
biomass
However,
with
nodule
biomass
during
the long
rains,
shoot
biomass
was
higher at Kavuthu than at Ndunguni and as with the nodule biomass, low
at Ndunguni
was attributed
to low total rainfall
received
in the site.
as opposed to highest overall mean nodule biomass at Kavuthu that was
observed in TSP treatment during the long rains, highest shoot biomass was recorded
in manure+T
P treatment
in both sites. This observation
was in agreement
with
Ojiem et al., (2004), who suggested that combining organic and inorganic P results in
synergistic
effects,
particularly
in drier moisture
stressed
growing
seasons.
In
addition, treatment effect for same varieties on shoot growth did not always match
treatment effect on nodule biomass indicating that treatments
that enhanced nodule
84
biomass did not necessarily enhance shoot biomass and probably that response of the
shoots to nutrients was different from that of nodules.
As observed
in nodule biomass, during the short rains, significantly
higher shoot
biomass was recorded at Ndunguni compared to Kavuthu and this observation
attributed to site characteristics
as opposed
production
to the observations
Ndunguni
other than rainfall differences in the sites. In addition,
made during the long rains where shoot biomass
in both sites was in overall enhanced by addition of manure+TSP,
only at Kavuthu
was
where the treatment
in overall enhanced
it was
shoot biomass while at
highest shoot biomass in overall was recorded in manure treatment. The
later observation
could have been caused by mineralization
(Campo et al., 1998;
Cornejo et al., 1994) of the high carbon content in manure and soil (Table 7.1) under
high rainfall at Ndunguni. These observations confirmed the need to combine organic
and inorganic inputs during drier seasons in both sites and in all seasons at Kavuthu.
It also indicated
inorganic
that during high rainfall seasons there might be no need to use
inputs at Ndunguni
observed that majority
application
to enhance cowpea biomass production.
of the varieties at Ndunguni
It was also
did not respond to treatment
probably indicating that there was a release of nutrients to the soil from
carbon mineralization
at the site. Loomis and Connor (1992) suggested that organic
matter could increase soil microbial biomass that can lead to an increase nutrient
availability to a growing crop.
85
7.4.3
Grain yield
As observed in other parameters
(nodule and shoot biomass) treatment application
generally enhanced grain yield at Kavuthu during the long rains. However, in overall
highest mean grain yield was recorded in the TSP treatment
most efficient nutrient
in grain filling. In addition,
increase grain yield of about 50% of the varieties
responded to treatment application.
indicating that P was
treatment
indicating
application
did not
that not all varieties
Most grain yields recorded during the long rain
season was within cowpea yield ranges of 1653 to 1827 kg/ha recorded by Kihanda et
al., (2004) in the semi-arid eastern Kenya.
As observed during the long rains, highest grain yield overall was recorded in TSP
treatment
in both sites probably
indicating that P applied
as TSP was the most
efficient nutrient in grain filling in both seasons. This observation
while inorganic inputs did not enhance biomass accumulation
also indicated that
at Ndunguni, inorganic
inputs were needed for grain filling. Varieties responded differently to nutrient inputs
depending
on the site indicating
treatments.
Most grain yield values recorded during the short rains ranged 791 to
by Jemo et al., (2006) in low phosphorus
1596 kg/ha obtained
Cameroon.
In addition,
yields of improved
1996).
the effect of site differences
on response
to
soils of southern
most cow pea varieties had grain yields within the potential
cow pea varieties (1200 to 1800 kg/ha) in Kenya (Audi et al.,
86
CHAPTER EIGHT
8
NITROGEN FIXATION, POPULATION AND DIVERSITY OF
COWPEA RHIZOBIA UNDER ISFM
8.1
Introduction
There is a trend towards soil fertility loss in many soils in the semi-arid zones of
Africa and planting legumes that can be used for both soil fertility and food may
arrest this trend. Grain legumes have been important crops and have been used for
centuries as food (van Kessel and Hartley, 2000). They supply proteins for animals
and mankind and cover about 11% of the world's arable lands (FAO, 1993). Grain
legumes
are also unique
in that they acquire atmospheric
symbioses with bacteria collectively
(N) through
called rhizobia in a process commonly termed
biological nitrogen fixation (Stevenson
and van Kessel,
fixation (BNF) is an important
aspect of sustainable
food production
crop productivity
and long-term
nitrogen
1997). Biological
nitrogen
and environmentally
friendly
(van Kessel and Hartley, 2000).
However, grain legumes contribute N to the soil only when the total quantity of N
fixed symbiotically
is larger than the amount of N removed at harvest in seed or in
crop residue (Evans et al., 1989; Giller et al., 1994; Stevenson and van Kessel, 1997).
Several methods
have been used to estimate BNF and the most commonly
used
methods are:
(1) Nitrogen difference method also called N balance method. This method is based
on difference in total nitrogen between a legume and a non-fixing reference plant. It
is based on the assumption that both legume and the non-fixing reference plant must
absorb the same amount of soi I
87
(2) 15N-isotope dilution either with enriched 15N-fertilizers (Fried et al., 1983) or
through changes at the natural 15N abundance level (Shearer and Kohl, 1986). Both
methods are used to estimate nitrogen fixed over a season. In both methods, it is
assumed that the rates of mineral N uptake are same for both the N-fixing legume and
the non-fixing
reference
plant. However,
differences
in seasonal
N accumulation
patterns of legume and reference crop under field conditions, the concurred decline in
%15N of available
soil N pool and differences
in root distribution
can lead to
erroneous estimates of N fixation. In addition, use of different non-fixing reference
plants can also lead to variable estimates of N fixation (Danso et al., 1993; Witty,
1983).
(3) Xylem-solute
method,
where N is estimated
by measuring
changes
in ureid
content in the xylem sap (Herridge et al., 1990). This method is limited to ureid
producing
legumes such as cowpea (Vigna unguiculata
L.) and can only be used to
estimate N fixed over a short period of time.
(4) Acetylene
observation
Reduction
Assay
(A RA) (Hardy
et al., 1968)
is based on the
that N2-fixing enzyme, nitrogenase, catalyses the reduction of acetylene
(C2H2) to ethylene (C4H4)
ARA method
(HC=CH + 2H+2e- ~
is that it only provides
activities under prevailing
H2C=CH2). The disadvantage
an instantaneous
measure
of
of nitrogenase
assay conditions and lacks linearity in the rate of C2H2
reduction during the assay.
Rhizobia by, definition,
are bacteria that establish symbioses
with legumes forming
root or stem nodules on the host and fixing atmospheric
nitrogen (N2) (Bala and
Giller, 2001). Exploitation
of the low-input tropical
of BNF to improve productivity
88
cropping
systems
populations.
demands
in
part
characterization
of
indigenous
rhizobial
Minimum criteria for describing new rhizobial species include cultural,
morphological,
physiological
al., 1991). Antibiotic
symbiotic traits and phylogenetic
characters (Graham et
resistance (Mueller et al., 1988), colony morphology
(Zhang et
al., 1991), stress tolerance to external temperatures, salinity and acidity (Keyser at al.,
1979) and nucleic
acids (Johnson,
1984) have been used as important
criteria to
distinguish rhizobia.
Cowpea
rhizobia
were first classified
rhizobia
nodulating
promiscuous
"Cowpea cross-inoculation
in heterogeneous
tropical and sub-tropical
group
of slow-growing
legume
species called
group" (Allen and Alien, 1981) but were later transferred
to the genus Bradyrhizobia that consisted of three species, B. japonicum
(groups 1
and la), B. elkanii and B. liaoningese (Hollis et al., 1981; Jordan, 1982; Kuykendall
and Saxena, 1991; Xu et al., 1995). However, Willems et al., (2001), using DNADNA
hybridization
Brydyrhizobium
and
consists
16S-23S
rDNA
of four highly
IGS
related
sequence
analysis
genospecies
showed
(B. japonicum,
that
B.
liaoningese), and at least three other genospecies one of which is B. elkanii).
"Cowpea miscellany"
well characterized
promiscuous
rhizobia indigenous to African soil environments
(Mpepereki
nodulating
have not been
et al., 1996). These bacteria are often described as
a wide range of legumes
but with
poor effectiveness
(Singleton et al., 1992). In addition, bacteria from cow pea nodules are said to contain
both slow- and fast-growing
rhizobia strains (Giller, 2001).
89
8.2
8.2.1
Estimation
Materials
and Methods
of biological nitrogen fixation (BNF)
Plant samples were taken from two on-farm trial sites, Kavuthu and Ndunguni in the
semi-arid Makueni District where nine cowpea varieties had been planted in the long
rains. Treatments
in the trails included a control (no amendment used) (Tl), manure
at 2.5 tlha (T2),
15 kg/ha P applied as TSP (P20S) (T3) and a combination
manure+TSP
of
(T4) at the singly applied rates. Cowpea varieties tested included a very
good nodulator in the trial sites (MI4), a local variety (Kang'au),
a medium nodulator
(E6) and a poor nodulator (M8) (full variety descriptions have been given in chapter
6). Plants used for BNF estimates were sampled at 50% flowering along with nine
common non-N-fixing
plants consisted
reference plants, consisting of shrubs and herbs. The reference
of Cordia sinensis,
Trichodesma zeylanicum,
Zea mays, Solanum
incanum, Melia volkensii, Ipomea sp. (4 species). The amount of nitrogen fixed was
assessed using the
ISN
natural abundance method. For
ISN
analysis, dry and ground
plant samples were sent to the UC Davis Stable Isotopes Facility, California, USA,
where they were analysed using a continuous flow Isotope Ratio Mass Spectrometer
(IRMS). An average
olsN value for the 9 non-fixing
plants (5.63) was used as a
reference value and %N fixed was calculated as follows:
100[(Non-fixing-Legume
sample)/(Non-fixing-l00%
fixation value)]
(SyiJa et al., 2002)
A olsN value of 0 was assumed for plants fixing 100% of their nitrogen. Data were
analysed using two-way ANOV A.
90
8.2.2
Rhizobia enumeration
Soils were collected
and assessment
of nodule and shoot biomass
at the top 20 cm from Kavuthu on-farm trial during the long
rains, at planting and during harvesting. In the trial, nine cow pea varieties had been
planted but only two of the test varieties (M14 and Kangau)
were used in this study
because they were the best nodulators among the nine varieties used in the on-farm
trials. During the planting time, soils were collected as described
in chapter 7 but
during harvesting time soi Is were collected by treatment and variety. Treatments used
at planting were a control, animal manure at 2.5 t/ha, P as TSP at 15kg/ha and a
combination
of animal manure and TSP at the singly applied rates (Coded T I, T2, T3
and T4, respectively).
The collected soil samples were stored at 4°C immediately after
collection. Rhizobia enumeration was done using the Most Probable Number Method
(MPN), as described
M14 and Kang'au,
by (Vincent,
1970). Clean seeds of the test cow pea varieties,
were placed in a flask and shaken for 3 minutes in 95% ethanol.
After ethanol was drained, the seeds were further shaken in 3% sodium hypochlorite
(NaOCI) for 3 minutes and rinsed in 10 changes of sterile water. The seeds were then
soaked for 2 hours in sterile warm water and then transferred to a sterile beaker and
incubated at 28°C in an oven for 48 hours. Pregerminated
transferred
into sterilized
Leonard jar assembly
(CaCh2H20,
KH2P04,
ZnS04.7H20,
CuS04.7H20,
1985). An uninoculated
Iron-citrate,
containing
MgS04.7H20,
CoS04.7H20,
seedlings were aseptically
N-free nutrient
K2S04, MnS04.H20,
NaMo02.2HO)
(Somasegaran
and a +N nitrogen (0.75g/L KN03)
media
H3B03,
and Hoben,
controls were included.
After 4 days, the root systems of the plants were inoculated with I ml of a 1:9 (Soil:
water) soil suspension
diluted serially to give dilutions of
s',
y2,
s',
y4, y5 and y6.
All jars were labeled to include variety, soil treatment, dilution, replication and date
91
of planting. To reduce evaporation,
the top surfaces of the jars were covered with
sterile gravel (ballast). The jars were placed in a green house whose floor and walls
had been cleaned using tap water. The plants were destructively
weeks. Data collected at harvesting
harvested
after 4
included number of positive jars (nodulated)
in
each dilution as well as nodule and shoot biomass of the harvested cow pea plants.
Rhizobia estimates were done using MPNES Programme
(Bennet et af., 1990). To
determine shoot and nodule biomass cow pea plants were destructively
harvested after
4 weeks and nodules carefully plucked from the root systems while the shoots were
cut at the shoot base. Nodule and shoot biomass were determined
as described
in
chapter 7.
8.2.3
Determination
8.2.3.1
Characterization
of rhizobia diversity
of rhizobia using cultures
Fresh nodules obtained from the MPN experiment (above) were surface sterilized by
immersion
in 95% ethanol and finally in 3% sodium hypochlorite
(NaOCI) for 3-4
minutes. The nodules were rinsed in six changes of sterile water and then each nodule
crushed in a drop of sterile water. A suspension
from each crushed
nodule was
streaked on Yeast extract mannitol agar (YMA), (lOg mannitol, 0.5 g K2HP04,
MgS04,
0.1 g NaCI,
CuS04.7H20,
0.5 g, and K2S04, MnS04.H20,
CoS04.7H20,
NaMo02.2H20
H3B03,
and iron citrate)
0.2 g
ZnS04.7H20,
yeast extracts
and
topped to 1000 ml with distilled water, adjusted to pH 6.8 using IM NaOH and IM
HCI, 15 g agar was added and the media was sterilized by autoclaving
at 121°C and
103 kPa for
(non-rhizobia
contaminants
15 minutes)
containing
25 mgl'
Congo
red
dye
absorb Congo red). The plates were incubated at 28°C until visible
92
bacterial growth could be seen on the streak line. A total of 376 rhizobia isolates were
cultured.
The isolates
purification.
on a fresh Congo
However only 151 cultures were used to determine
in the cultures
production
were then sub-cultured
using
colony
shape and plasticity.
characteristics
such
red media for
strain types present
as colony
colours,
mucous
To determine growth rate, 70 cultures out of the
purified 151 cultures were selected and streaked on YMA without Congo red. The
rate emergence
and growth of the rhizobia colonies to maximum
line helped to characterize
the isolates into fast or slow-growing
size on the streak
types. To determine
whether isolates were neutral, acidic or alkaline, the pure isolates were streaked on
YMA containing
bromo-thymol
blue (25 ppm), a pH indicator, and colour changes
from green (neutral) to blue (alkaline) or yellow (acid) helped to categorize the 70
isolates (Appendix 4) into neutral, alkaline or acidic culture types.
8.2.3.2
i)
Determination
of rhizobia diversity by direct polymerase chain
reaction (peR) amplification of 16s rRNA gene from rhizobia culture
cells
rRNA amplification
Eighteen strains of rhizobia were analysed using PCR-Restriction
polymorphism
(PCR-RFLP)
of 16S-23S rRNA intragenic
spacer region (IGS). For
RFLP analysis, primers used were, a forward primer Yl (5'-TGG
AAC GCT GGC GGC-3')
corresponding
(5'-TAC CTT GTT ACG ACT TCA
1482-1507 in the Escherichia
fragment length
CTC AGA ACG
to positions 20-43 and a reverse primer Y3
ccc
CAGTC-3')
corresponding
to positions
coli 16s rRNA gene sequence (Odee et al., 2002; Cruz
et al., 2001). 25 1-11 of PCR reaction product contained 50 ng pure DNA from fully-
KENYATTA UNiVERSITY LIBRARY
93
grown cultures, one dried bead (Ready-to-go
PCR beads (GE Healthcare illustra TM),
50 mM KCI and 1.5 mM MgCI2), 1.5 III Y I, 1.5 III Y2 and 22 III PCR grade water. 10
III of the PCR reaction
programmable
product were used for amplification.
Thermal
a
Controller (PTC-IOO™ MJ Research Inc., Watertown, MA)
was used as described
denaturation
For amplification,
here: Initial denaturation
at 93 QC for 2 min; 35 cycles of
(45 sat 93 QC) annealing (45 sat 62 QC), extension (2 minutes at 72 QC)
and final extension
at 72 QC for 5 minutes. Amplified DNA products were separated
by horizontal gel electrophoresis
of 2 f.ll aliquots of PCR product (Odee et al., 2002)
in 1% agarose (Invitrogen ™ Life Technologies)
(107.8 g tris-base,
in TBE (Tris-borate
EDTA) buffer
7.44 g EDTA and 55 g boric acid) for 2'/2 hours at 80 V and
stained with (1 ug/ml) ethidium bromide for 30 minutes. Stained DNA profiles were
photographed
under
UV with Kodak Molecular
tOEastman Kodak Company,
ii)
Imaging
Software
Version
4.0,
1998-2005.
rRNA restriction
Aliquots
of 6 III from the PCR reaction
endonucleses
product
MspI in 0.3 III enzyme (in a restriction
PCR reaction product,
Serum Albumin
were digested
with restriction
product that consisted of 6 III
I III enzyme buffer, 0.3 III restriction enzyme, 0.1 III Bovine
and 2.6 III water giving a total volume of 10 Ill) for 2 hours. The
restricted rRNA was analysed by horizontal electrophoresis
in 2.5% (w/v) agarose gel
in TBE buffer, stained and photographed as described above. To determine the size of
base pairs of the restricted rRNA, the pattern bands obtained after restriction were
compared
bromide.
to the base pairs of the standard,
1.5% agarose
stained with ethodium
94
8.3
8.3.1
Results
Biological nitrogen fixation
Large and significant
differences
(p<0.05) were found between amounts of nitrogen
fixed at the two sites with plants at Kavuthu fixing 46-53% of their nitrogen while
only one treatment (Kang'au
treated with TSP, 815N=4.96)
at site 2. The local variety (Kang'au)
of its nitrogen (815N=2.28),
(815N=I.58).
nitrogen
showed fixation of 12% N
supplied with Triple super phosphate fixed 60%
while Ml4 treated with manure+TSP
However,
differences
between
varieties
fixed 72% of its
and
differences
between treatments were not significant (Figures 8.1-8.3).
56
54
.-----------------------------~===,
DTI
DT2
~ 52
'2f.
'--' 50
"2x 48
t;:: 46
Z
44
bJT3
DT4
42
40
Treatment
Figure 8.1
Amount of nitrogen fixed (%) at Kavuthu during the long rains
Treatments (TI-T4); I) Control, 2) Manure at 2.5 tlha, 3) P at 15kglha and 4)
manure+ P at the singly applied rates.
95
12
10
8
~
6
4
,[====;------
-
-
DS1T1
DS1T2
GS1T3
DS1T4
DS2T1
DS2T2
DS2T3
DS2T4
~'------'
2
Site and treatment
Figure 8.2
N in sites 1 and 2 during the long rains
J5
S1 and S2 represents site 1 (Kavuthu) and 2 (Ndunguni), respectively. Treatments,
T1-T4 means; 1) Control, 2) Manure at 2.5 t/ha, 3) P at 15kg/ha and 4) manure + P at
the singly applied rates.
la
8
6
4
OLS1
OM14S1
OSlM8
OS2L
J S2M14
DS2E6
.1
-,
T
;.l:
..
.
T
.L
T
2
I
T
..I..
.,...
~
I
,.,
--
...........
..
.~
zI,'
o
Variety by site
Figure 8.3
15N values for site by variety during the long rains
SI and S2 represents site 1 and 2, respectively, while L, M14 and E6 represents
cowpea varieties Kang'au, M4 and E6, respectively.
96
8.3.2
Rhizobia
populations
in the soil, and nodule and shoot biomass
Rhizobia counts ranged between 4.89x 102 and 2.0x 104 cells/gram of soil (Table 8.1).
With an exception
of rhizobia counts in manure treatment
in M 14 and in manure
+TSP treatment in local (L) variety, Kang'au, generally rhizobia counts were lower at
planting than during harvesting.
Further, the highest rhizobia count in variety M 14
was recorded in TSP and manure+TSP treatments, while in Kang'au
highest rhizobia
counts were recorded in manure and TSP treatments.
Nodule biomass ranged from 0.9 to 3.4 mg with highest nodule biomass of 3.4 mg
being recorded in a control treatment ofKang'au
(Figure 8.4). Generally a low nodule
biomass was recorded in cowpea plants grown in soils collected at the start of the
season and treatment
application
overall a higher and significant
did not improve nodule biomass.
In addition,
(p<0.05) nodule biomass was recorded
in
in Kang'au
compared to M14.
Shoot biomass ranged from 29 to 52 mg/plant, with highest biomass production being
recorded in M 14 in a manure treatment (Figure 8.5). There was no difference in shoot
biomass between cowpea plants grown in the soils collected at the start of the season
and the control treatments.
In addition, treatment application
enhanced shoot growth in manure and manure+ TSP treatments.
significantly
(p<O.O1)
97
Table 8.1
Rhizobia estimates (rhizobia cells/gram of soil) in soils collected at the
beginning and at the end of long rain season
Variety
Time of soil collection
Treatment
Cells/gram soil
M14
Planting
0
1.074x 103
M14
Harvest
1
1.625xl04
MI4
Harvest
2
4.89x102
MI4
Harvest
3
2.0x104
M14
Harvest
4
2.0x104
Kang'au
Planting
0
6.98x102
Kang' au
Harvest
1
1.625x 104
Kang'au
Harvest
2
2.0x104
Kang' au
Harvest
3
2.0x 104
Kang'au
Harvest
4
8.75x103
TO- Soils collected before ISFM experiment was started; Tl- Soils collected from the
control treatment; T2- Soils collected from 2.5 t/ha manure amended soils; T3- Soils
collected from 15 kg/ha P amended soils and T4- Soils collected from manure+P
amendments. M14 and Kang'au are cowpea test varieties.
4
3.5
C
'"P-
3
M
S 2.5
VI
VI
'"E
.~
oD
~
2
I.5
='
-0
0
Z
0.5
0
TO
TI
T2
T3
T4
Treatment
Figure 8.4
Nodule biomass of nodules recovered from MPN experiment
TO- Soils collected before ISFM experiment
control treatment; T2- Soils collected from
collected from 15 kg/ha P amended soils
amendments. MI4 and Kang'au are cowpea
was started; Tl- Soils collected from the
2.5 t/ha manure amended soils; T3- Soils
and T4- Soils collected from manure+P
test varieties.
98
60
--~
El Kang'au
-z: 50
c
OMl4
C<:!
0..
00 40
E
'--'
Vl
~ 30
E
o
..D 20
o
o
..c
if)
10
TO
T2
Tl
T3
T4
Treatment
Figure 8.5
Shoot biomass of cow pea plants harvested from MPN experiment
TO- Soils collected before ISFM experiment was started; Tl- Soils collected from the
control treatment; T2- Soils collected from 2.5 t/ha manure amended soils; T3- Soils
collected from 15 kg/ha P amended soils and T4- Soils collected from manure+P
amendments. M14 and Kang'au are cow pea test varieties.
8.3.3
Rhizobia characterization
Using colony characteristics
of rhizobia isolates, 9 groups of rhizobia were identified
(Table 8.2). Most of the strains fell under strain group 3, while only two strains fell
under group 9. In addition, assessment of 70 isolates for growth rate revealed that
97% (68) of the isolates were fast growing
while only 3% isolates
were slow
growing. Slow growing isolates had dry colonies while most of the fast growing ones
had wet colonies.
With an exception of strain group 9, all other strain groups were
fast growing. Further, analysis of 18 using PCR-RFLP of 16S rRNA intragenic spacer
region revealed
four rhizobia groups represented
by four patterns (Figure 8.6). The
patterns had base pairs (bp) that ranged from 271 to 827 (Table 8.3). Variety and
treatment did not appear to affect restriction patterns (Table 8.4).
99
Table 8.2
Rhizobia isolate characterization
Isolate Colony characteristic
group using 376 cultures
Round, small, flat,
clear with brown
centres
Tiny, sticky and
2
mucoid
Large, round,
translucent, white,
3
gummy
Tiny, dry, flat, sticky
4
brown
5
Large, pink, dome
Small, round, dry,
6
flat, orange
Large, brown,
7
mucoid, spreading
Large, watery,
spreading, clear with
8
suspensions
Flat, orange, shiny,
9
dry
using cultures
Colour
pH
Number of
cultures
tested for
growth rate
and acid
Growt characteristi
h rate cs
Yellow
Acid
Fast
6
33
Green
Neutral
Fast
11
11
Yellow
Acid
Fast
18
59
Yellow
Yellow
Acid
Acid
Fast
Fast
8
8
8
21
Yellow
Acid
Fast
3
3
Yellow
Acid
Fast
9
9
Yellow
Acid
Fast
5
5
Blue
Alkaline Slow
2
2
M
Number of
cultures
selected for
detai led
studies
bp
1,353
1,078
872
Figure 8.6
Four PCR-amplified
16S-23S rRNA IGS patterns obtained
restriction of rhizobia strains with MpsI restriction endonuclease
M is 1.5 agarose gel stained with ethidium bromide and bp is base pairs.
after
100
Table 8.3
Fragment sizes in base pairs (bp) obtained after rhizobia restriction
rRNA IGS patterns
Fragment size (bp) of amplified rRNA
IGS digested with MspI
271 383 603 684
II
271 684711
III
271 408657
IV
Table 8.4
271 408603
827
Strain group, rRNA IGS pattern, cowpea varieties and treatments used
in the PCR-RFLP analysis of 16S-23S rDNA IGS
Pattern number
1
2
3
4
5
6
7
8
9
la
11
12
13
14
I5
16
17
18
Strain group
8
8
8
8
9
9
7
7
7
7
7
2
2
1
2
2
2
2
rDNA
group
I
I
II
II
III
III
III
III
III
I
I
I
II
II
I
I
IV
IV
TO- Soils collected before ISFM experiment
control treatment; T2- Soils collected from
collected from 15 kg/ha P amended soils
amendments. M14 and Kang'au are cowpea
IGS
Variety
Treatment
M14
Kang'au
Kang'a
Kang'au
M14
Kang'au
Kang'au
Kang'au
MI4
Kang'au
Kang'au
M14
Kang'au
Kang'au
Kang'au
MI4
MI4
MI4
TO
T3
T3
Tl
T4
T2
Tl
T2
TO
T2
T2
T4
T2
T3
T3
TI
T4
T4
was started; Tl- Soils collected from the
2.5 t/ha manure amended soils; T3- Soils
and T4- Soils collected from manure+P
test varieties.
101
8.4
8.4.1
Discussion
Biological nitrogen fixation
Biological nitrogen fixation was recorded in the wetter site (Kavuthu), where it varied
from 46 to 53%. Nitrogen
fixation at the drier site (Ndunguni)
was most probably
limited by low and unevenly distributed rainfall (chapter 7) that could have caused
water stress to the cowpea plants to the extent that nitrogen fixation was inhibited.
Sprent (1971) suggested
leguminous
root
that water stress reduces the formation
nodules
thereby
reducing
nitrogen
fixation
and longevity of
in nodules.
This
suggestion agrees with the observation that very low nodule biomass was recorded at
the drier site (Ndunguni) during the long rains (Chapter 7) compared to the wetter site
(Kavuthu) and hence low or no nitrogen fixation was recorded in the cowpea plants
recovered from Ndunguni.
At Kavuthu, nitrogen fixation was higher in manure and
TSP than in manure+ TSP treatment.
manure and TSP in the manure+TSP
This was probably
because presence of both
treatment could have resulted to high levels ofP
in the soil that could have decreased
N fixation. Miao et al., (2007) reported a
decrease in N fixation with increasing P supply. Nitrogen fixation values obtained in
this study were comparable
to ranges of 46-59%,
36-54%,
35-50%
and 36-51 %
obtained by Gathumbi et al., (2002) in western Kenya in Sesbania sesban, Calliandra
callothyrsus,
Macroptilium
(groundnut),
respectively,
shrubs to provide
atropurpureta
hypogaea (Siratro) and Arachis hypogaea
indicating that cowpea can be used to replace leguminous
fixed N for crops. Cowpea will have added advantage
over
leguminous shrubs because of its multiple uses. In addition to edible grain production,
and its leaves that can be used locally as a green vegetable,
generate income for households.
it can also be sold to
102
8.4.2
Rhizobia
Rhizobia
counts
harvesting
time.
February/March
populations
in the soil
were generally
This
lower at the beginning
observation
was most probably
of the season than at
due to a dry spell of
that could have lowered rhizobia populations
to the levels observed
at the beginning of the long rains, in the soils collected at cowpea planting time. This
observation
was in agreement with the observation made by Mapfumo et al., (2000),
where low rhizobia counts were recorded in soil samples collected following a dry
spell. Seasonal
documented
increase
by Mulongoy
in rhizobia
populations
and Ayanaba
in the cowpea
(1985) while studying
trials was also
seasonal
rhizobia
populations
in three locations in West Africa. In the study, they noted that rhizobia
populations
were higher
in wetter than in the drier sites. Rhizobia
populations
observed in this study were generally lower than the rhizobia populations observed by
Maingi et al., (2006) at Kiboko research station about 40 km east of Kavuthu trial
using cowpea, cultivar Ken Kunde 1, which ranged from 2.59x104 to 1.89xl05.
differences
in rhizobia counts could have been due to moisture differences
The
between
the two sites given that the Kavuthu trial was rain fed while all crops at Kiboko
Research Station are usually cultivated using irrigation. In addition, the differences
could have also been due to the differences in cowpea cultivars used in both studies.
High rhizobia populations
growth hormones
multiplication
are important in cultivated soils because rhizobia produce
which can stimulate plant growth and compounds
of pathogenic bacteria (Williams and Bartholomew,
which inhibit
2008).
Soils collected at harvesting had relatively higher nodule biomass compared to those
collected at planting confirming the observation that more rhizobia populations were
103
present in the soil at harvesting compared to the start of the season. This indicated
that the high rhizobia populations
present in the soil at harvesting enhanced nodule
formation when soils collected at harvesting time were inoculated to cow pea plants.
Shoot
rhizobia
biomass
was enhanced
population
sizes
by treatment
indicating
application
that shoot
that had also enhanced
growth
responded
to treatment
application than did nodule formation in this study.
8.4.3
Rhizobia characterization
Most fast growing cultures formed wet colonies while slow growing colonies formed
dry colony types. These observations
were most probably the characteristic
colony
types of both slow and fast-growing colonies of cowpea. For example, Mpepereki et
al., (1997) reported similar observations in cowpea rhizobia at Zimbabwe. However,
they observed similar proportions of both fast-slow growing rhizobia (49% and 51 %,
respectively)
as opposed to this study where 97% of the isolates were fast growing.
In this study indigenous
studied
after trapping
rhizobia populations
rhizobia
obtained
from an ISFM trial were
strains using two cowpea
varieties.
PCR-RFLP
analysis of 16S-23S IGS region groups genetically related strains (Jensen et al., 1993;
Vinuesa et al., 1998). Therefore the 18 strains represented
four genotypes. Wade et
al., (2003), studying on diversity of indigenous rhizobia associated with three cowpea
cultivars, also observed four genotypes in the rhizobia isolated from the cultivars.
104
CHAPTER NINE
9.
CONCLUSIONS
In this chapter conclusions
AND RECOMMENDATIONS
drawn from this study and recommendations
for further
research are summarized.
9.1
1)
Conclusions
Farmers
considered
in Makueni
District
were aware of declining
it as a problem to crop production.
soil fertility
and
This was shown by use of animal
manure by fanners to enhance soil fertility in their farms.
2)
Most farmers in Makueni District experienced
food deficit for about eight
months in a year and short rains were most reliable for crop production
in the study
sites.
3)
Grain legumes occupied substantial
proportions
of cultivated
land with the
cultivated area occupied by the grain legumes increasing with decrease in cultivated
farm size.
4)
Local cowpea variety, Kang'au, had very high potential for nodule formation,
biomass production
and grain yield and performed better than all improved cowpea
varieties,
to the belief that local varieties are low yielding.
contrary,
Low yields
observed by farmers may be probably due to poor management of the local varieties.
5)
Cowpea nodulation, biomass production
and grain yields were enhanced by
addition of inputs in form of manure, TSP or manure+ TSP and that manure+ TSP
treatment
was commonly
effective during the drier season, as well as during the
wetter season at Kavuthu. This observation
sites to enhance cowpea production.
implied need to add inputs in the study
105
6)
Grain
yield
was
enhanced
by addition
of
phosphorus
implying
that
phosphorus was limiting cowpea grain production in the study sites.
7)
Treatment
addition
enhanced
nitrogen
fixation
in cowpea
and improved
indigenous soil rhizobia populations.
9.2
1)
Recommendations
There is need to train farmers on the importance of soil fertility improvement
in the study sites, especially
in the management
and use of locally available farm
inputs such as animal manure, compost and crop residues. Farmers should also be
sensitized
on the use of inorganic
inputs, especially
phosphorus
(P), which can
significantly be improved in the soils by addition ofP fertilizers.
2)
Improved cowpea varieties, especially those used for on-farm trials in chapter
seven should be included in the farming systems of Makueni District and cultivated
along with the locally available
varieties to enhance
cowpea diversity
and boost
cowpea yields in the district.
3)
Soil fertility improvement
management
nitrogen
(ISFM)
fixation,
should
growth
interventions such as use of integrated soil fertility
be considered
to maximize
and in grain legume
cowpea
production
Techniques such as spot application of inputs and microdozing
to enhance effective use of inputs in the study sites.
performance
in the study
in
sites.
should be introduced
106
LITERA TURE CITED
Abeyasekera,
S., J.M. Ritchie and Lawson-McDowall.
scores
to determine
Malawi. Experimental
Alien,
O.N.
and
E.K.
characteristics,
farmers'
preferences
2004. Combining
for bean varieties
ranks and
in Southern
Agriculture. 38:97-107.
Alien.
1981.
The
uses and nodulation.
leguminoseae:
University
a source
of Madison,
book
of
Wisconsin,
U.S.A.
Amanullah,
P. and M. Hatam. 2000. Yield potential of cow pea germplasm. Pakistan
Journal of Biological Sciences. 3(5): 858-861.
Audi, P.O., e.K.
research,
Ngugi, F.P. Wandera and R. Ndengwa.
extension
Recommendations
KARI-Katumani,
and
development
1996. Dryland farming
project.
Dryland
for smallholder farmers. Proceedings
farming:
of a workshop held in
Machakos, Kenya. Pp. 47.
Ayaga, G. and P.e. Brookes.
2005. Increasing fertilizer phosphorus
use efficiency
through organic matter cycling for enhanced maize production
in Kenya. Pp.
345-362.
In: Mugendi,
P.N. Macharia,
D.N., G. Kironchi,
M. Mburu,
J.G. Mureithi
P.T. Gicheru,
and F. Maina
e.K.K.
Gachene,
(eds.). Capacity
building for land resource management to meet the challenges of food security
in Africa. Proceedings
of the 21 st annual conference.
Soil Science Society of
East Africa, Nairobi, Kenya.
Bala, A. and K.E. Giller. 2001. Symbiotic specificity of tropical tree rhizobia for host
legumes. New Phytologist.
Bationo, A., BJ. Ndunguru,
149: 195-207.
B.R. N'tare, B.e. Christianson
1991. Fertilizer management
strategies for legume-based
and A.V. Mokwunye.
cropping systems in
107
West Africa semiarid tropics. Pp 213-225. In: Johansen, C, K.K. Lee and K.L.
Sahrawat
(eds.).
Phosphorus
tropics. Proceedings
nutrition
of grain
legumes
of the international' workshop
in the semi-arid
held at International Crop
Research Institute for the Semi-Arid Tropics (lCRISAT), Patancheru, India.
Bekunda, M.A., A. Bationo, and H. Ssali. 1997. Soil fertility management
A review
of selected
research trials. Replenishing
in Africa:
soil fertility in Africa.
SSSA special publication. 51 :63-79.
Bennett, J.E., P.L. Woomer and R.S. Yost. 1990. Users manual for MPNES mostprobable-number
enumeration
system.
Version
1.0. NiITAL
Project
and
1990. Agriculture
and
University of Hawaii, Paia.
Bockman,
c.,
O. Kaarstad,
O.H.
Lie and 1. Richards.
fertilizers. Agriculture group, Olso. Pp. 245 .
Bressani,
R. 1985. Nutritive
value of cowpea.
Cowpea
research,
production
and
utilization. In: S.R. Singh and K.O Rachie (eds.). John Wiley and sons Ltd.
Bronson,
J. 1995. Conducting
resource management
Buresh,
R.J., P.c.
capital
Publication.
in natural
from Vanuatu. Clark University, Worcester, U.S.A.
Smithson
in Africa.
PRA in The South Pacific: Experiences
and D.T. Hellums.
Replenishing
1997. Building
Soil Fertility
in Africa.
soil phosphorus
SSSA
Special
51:111-149.
Campo, J., V.J. Jaramillo, J.M. Maas. 1998. Pulses of soil phosphorus availability in a
Mexican
Oecologia.
tropical
dry forest: Effects of seasonality
and levels of wetting.
115: 167-172.
Chambers, R. 1983. Rural development: Putting the last first, Longman scientific and
technical. Essex, U.K.
108
Chaturverdi,
G.S., P.K. Aggarwal
and S.K. Simba.
determinate
and indeterminate
cowpeas
Agricultural
Sciences. 94: 137-144.
1980. Growth
in dryland
and yield of
agriculture.
Journal of
Chien, S.H. and R.G. Menon. 1995. Factors affecting the agronomic effectiveness of
phosphate rock for direct application. Fertilizer Research. 41 :227-234.
Chui, J.N and M. Nadar. 1984. Evaluation of effects of Rhizobium phaseoli strains on
nodulation,
dry matter and grain yield of two bean (Phaseolus
vulgaris)
varieties. East African Agricultural and Forestry Journal. 44: 109-112.
Conway, G. and J. McCracken.
development:
Development,
1988. Training notes for agroecosystem
Ethiopia.
International
Institute
for
analysis for
Environment
and
London.
Cornejo, F.H., A. Varela and J. Wright. 1994. Tropical forest litter decomposition
under seasonal drought: nutrient release, fungi and bacteria. Oikos. 70: 183190.
Crow,
V.L., B.D.W.
Jarvis and R.M. Greenwood.
homologies among acid-producing
of Systematic Bacteriology.
1981. Deoxyribonucleic
strains of Rhizobium. International
16S ribosomal
isolated
from banana
DNA
Merril). Applied Environmental
problems
characterization
in the use of
Microbiology.
and F. Zapata.
ISN
(Ananas
bacteria
comosus
(L.)
67(5):2375-2379.
1993. Misconceptions
soil enrichment
fixation. Plant and Soil. 152:25-52.
and F.O. Pedrosa.
of nitrogen-fixing
(Musa spp.) and pineapple
Danso, S.K.A., G. Hardarson
Journal
31: I 52- 172.
Cruz, L.M., E.M. Souza, O.B. Weber, J.1. Baldan, J. Dobereiner
200l.
acid
techniques
and practical
for estimating
N2
109
Date, R.A. and J. Halliday.
of rhizobia.
1987. Collection, isolation, cultivation and maintenance
Pp 1-29. In: Elkan, G.!. (ed.).
Symbiotic
nitrogen
fixation
technology. Mercel Dekker, Inc. New York. Pp. 440.
Davis, D.W., .E.A. Oelke, E.S. Oplinger, J.D. Dol, C.V. Hanson and D.H. Putman.
1991. Cowpea. Alternative field crops Manual. Pp. 1-9.
de Lajudie, P., F. Laurent-Fulele,
M. Neyra,
A. Willems, B. Pot, D. Dewettinck, G. Maestrojuan,
M.D. Coli ins, B. Dreyfus,
K. Kersters
and M. Gillis.
1994.
Polyphasic taxonomy of rhizobia: emendation of the genus Sinorhizobium
description
of Sinorhizobium
Sinrhizobium
teranga
Bacteriology.
44:715-733.
Dennison,
E.B.
sp.
melitoti comb, nov., S. saheli sp. Nov., and
novo
International
1961. The value of farmyard
manure
northern Nigeria. Empire Journal of Experimental
Journal
of
in maintaining
Systematic
fertility
extracts
using
leguminosarum
randomly
Journal of Microbiology.
amplified
biovar
trifolii
polymorphic
111
Agriculture. 29:330-336.
Dye, M., L.Sk0t, L.R. Mytton, S.P. Harrison, J.J. Dooley and A. Cresswell.
study of Rhizobium
and
populations
DNA profiles.
1995. A
from soil
Canadian
41:336-344.
Eardy, B.D., l.P.W. Young and R.K. Selander.
1992. Phylogenetic
Rhizobium
sp. strain Or 191, a symbiont
Phaseoulus
vulgaris, based on partial sequences
genes. Applied Environmental
Microbiology.
of both Medicago
position
saliva
of
and
of the 16S rRNA and nijH
58: 1809-1815.
Evans, J., G.E. Oconnor, G.L. Turner, D.R. Coventry, N. Fettell, J. Mahoney, E.L.
Armstrong
and D. Walsgott. 1989. Nitrogen fixation and its value to soil N
110
Increases in Lupin, field pea and other legumes in south-eastern
Australia.
Australian Journal of Agricultural Research. 40:791-805.
Fairhurst, T., R. Lefroy, E. Mutert and N. Batjes. 1999. The importance,
and causes of phosphorus
tropics. Agroforestry
FAOSTA T. 2000.
Site
Organization
distribution
deficiency as a constraint to crop production in the
Forum. 9(4): 2-8.
Internethttp/www.fao.org/statists.
Food
and
of United Nations (FAO). 1993. Year production
Agriculture
book. Volume
46, FAO, Rome.
Food and Agriculture
Organization
of United Nations (FAO). 1993. Year production
book. Volume 46, FAO, Rome.
Fried, M., S.K.A. Danso and F. Zapata. 1983. The methodology
fixation by non-legumes
as inferred from field experiments
Canadian Journal of Microbiology.
ofN
with legumes.
29:1053-1062.
Gathumbi, S.M., G. Cadisch and K.E. Giller. 2002.
for assessing N2-fixation
of measurement
N Natural abundance
15
as a tool
of herbaceous, shrub and tree legumes in improved
fallows. Soil Biology and Biochemistry. 34:1059-1071.
Gibberd, V. 1995. Yield response offield crops to animal manure in semi-arid Kenya.
Tropical Science. 35 :418-426.
Gibson, A.H., B.L. Dreyfus and Y.R. Dormmergues.
tropics. P. 328. In: Dormmergues,
of tropical
soils
and
plant
1982. Nitrogen
fixation in the
Y.R. and H.G. Diem (eds.). Microbiology
productivity.
Martinus
NijhofflDr.
W. Junk
Publishers, The Hague.
Giller, K.E. 2001. Nitrogen fixation in tropical cropping systems, 2nd Edition. CABl
publishing, Wellingford,
U.K. Pp. 423.
111
Giller, K.E., J.F. McDonagh and G. Cadisch. 1994. Can biological nitrogen fixation
sustain
Rimmer
agriculture
in the tropics? Pp. 173-191.
(eds). Soil and sustainable
International,
In: Syers, J.K. and D.L.
land management
in the tropics. CAB
Wellingford, U.K.
Giller, K.E., G. Cadisch, C Halitosis, and E. Adams. 1997. Building soil nitrogen in
Africa. Replenishing
soil fertility in Africa. SSSA Special Publican. 51: 151-
192.
Graham, P.H., MJ. Sadowsky, H.H. Keyser, Y.M. Barnet, R.S. Bradley, J.E. Cooper,
DJ. Deley, B.D.W. Jarvis, E.B. Roslycky, B.W. Stijdom and J.P.W. young.
1991. Proposed minimal standards for description
of root- and stem-nodulating
Bacteriology.
of new genera and species
bacteria. International
Journal
of Systematic
41:582-587.
Hardy, R.W.F., R.D. Holsten, E.K. Jackson and R.e. Burns. 1968. The acetyleneethylene
assay
for N2-fixation:
laboratory
and
field
evaluations.
Plant
Cresswell.
1992.
Physiology. 43:1185-1204.
Harrison,
S.P.,
L.R.
Characterization
polymorphism
Mytton,
of
L.
Sket,
Rhizobium
using random
M.
Dye,
isolates
primers.
by
Canadian
and
C.
amplification
Journal
of
DNA
of Microbiology.
38: 1009-1 0 15.
Hartley, K.T. 1937. An explanation
Nigeria. Experimental
Herridge,
of the effect of farmyard
manure in Northern
Agriculture. 19:224 -263.
D.F., F.J. Bergerson
and M.B. Peoples.
1990. Measurement
of nitrogen
fixation by soybean in the field using the ureid and natural 15N abundance
methods. Plant Physiology. 90:708-716.
112
Higgins, C.F., G. Ferro-Luzzi
1982. A novel
Ames, W.M. Barnes, J.M. Clement and M. Hofnung.
intercistronic
regulatory
element
of prokaryotic
operons.
hybridization
studies of
Nature. 289:760-762.
Hollis., A.B. W.E. Kloos and G.H. Elkan. 1981. DNA:DNA
Rhizobium
Microbiology.
Huber,
1.
and
japonicuni
and
Rhizobiaceae.
Journal
of
General
123 :215-222.
S.
fingerprinting,
related
Selenska-Pobell.
genome
1994.
Pulsed-field
size estimation
gel
electrophoresis-
and rrn loci number
of Rhizobium
galegae. Journal of Applied Bacteriology. 77:528-533.
Hulton, C.S.J., D.F. Higgins and P.M. Sharp. 1991. ERIC sequences:
of repetitive
typhimurium
elements
and other enterobacteria.
Ikera, S.T. 2006. Combining
improvement
in the genomes
of Escherichia
a novel family
coli, Salmonella
Molecular Microbiology.
Tithonia diversifolia
5:825-834.
and Mijingu phosphate
rock for
of P availability and maize grain yields on a Chromic Acrisol in
Morogoro Tanzania. Nutrient Cycling in Agroecosystems.
76(2-3): 249-260.
Ikombo, B. 1984. Effects of farmyard manure and fertilizers on maize in semi-arid
areas of eastern
Kenya.
East. African
Agricultural
and Forestry
Journal.
44:266-274.
Imungi, J.KJ.
and N.N. Porter. 1983. Nutrient content of raw and cooked cowpea
leaves. Food science. 48: 1252-1254.
Jama, B., C.A. Palm, RJ. Buresh, A. Niang, C. Gachengo,
Amadalo.
2000. Tithonia
improvement
diversifoloa
G. Nzinguheba
as a green manure
in western Kenya: A review. Agroforestry
and B.
for soil fertility
Systems. 49:201-221.
113
Jama, B., R.A. Swinkles and R.J. Buresh. 1997. Agronomic and economic evaluation
of organic and inorganic sources of phosphorus in Western Kenya. Agronomy
Journal. 89:597-604.
Janssen, B.H. 1994. Integrated nutrient management:
the use of organic and mineral
fertilizers. Pp. 89-105. In: Van Reuler, H. and W.H. Prins (eds.). The role of
plant nutrients for sustainable
Wageningen,
Jemo, M., R.e.
food crop production
in Sub-Saharan
The Netherlands.
Abaidoo, e. Nolte and W.J. Horst. 2006. Genotypic
phosphorus
Africa.
variation
uptake and dinitrogen fixation in cowpea on low-phosphorus
for
soils
of southern Cameroon. Journal of Plant Nutrition and Soil Science. 169:816825.
Jensen, M.A., J.A. Webster and N. Straus. 1993. Rapid identification
the basis of Polymerase
polymorphism.
Chain Reaction-Amplified
Applied Environmental
and J.J. Holt (eds). Bergeys's
rhibosomal
Microbiology.
Johnson, J.L. 1984. Nucleic acid in bacterial classification.
of bacteria on
DNA spacer
59:945-954.
Pp. 8-11. In: Kreig, N.R.
manual of systematic bacteriology.
Volume
1.
Williams and Wilkins, Baltimore, USA.
Johnson, J.L. 1991. DNA reassociation
and M. Goodfellow
experiments.
Pp. 21-44. In: Stackebrandt,
(eds.). Nucleic acid techniques
E.
in bacterial systematics.
John Willey & Sons Ltd, England.
Jordan,
D.e.
1982.
Bradyrhizobium
from leguminous
32:136-139.
Transfer
of
Rhizobium
japonicum
genus novo a genus of slow-growing
plants. International
Buchanan
1080
to
root nodule bacteria
Journal of Systematic
Bacteriology.
114
Kajalainen,
S.
and
K.
Lindstrorn.
polymorphismanalysis
1989.
Restriction
fragment
length
of Rhizobium ga/egae strains. Journal of Bacteriology.
171 :5561-5566.
Kamprath,
EJ.
1991. Appropriate
measurements
of phosphorus
tropics.
(eds.).
Phosphorus
International
Crop
(ICRISAT), Patancheru,
in the
c.,
K.K Lee and K.L.
nutrition of grain legumes
in the semi-arid
soils of the semi-arid tropics. Pp. 23-32. In: Johansen,
Sahrawat
availability
Research
Institute
for
Semi-arid
Tropics
India.
Keyser, H.H., D.N. Munns and J.S. Hohenberg.
1979. Acid tolerance of rhizobia in
culture and symbiosis with cowpea. Soil Science Society of American Journal.
43:719-722.
Khan, T.N. 1973. A new approach to the breeding of pigeon peas (Cajanus cajan
Millsp): Formation of composites. Euphytiea. 22: 373-377.
Kihanda, F.M., G.P.Warren
and S.S. Atwal. 2004. The influence
of goat manure
application on crop yield and soil nitrate variation in semi-arid eastern Kenya.
Pp.163-186.
In: Bationo, A. (ed.). Managing
fertility in sub-Saharan
nutrient cycles to sustain soil
Africa. Tropical Soil Biology and Fertility Institute of
CIA T, Nairobi, Kenya.
Kimani, S.K. and J.K. Lekasi. 2004. Managing
manures through their production
cycles to enhance their usefulness: A review. Pp. 187-197. In: Bationo, A.
(ed.). Managing nutrient cycles to sustain soil fertility in sub-Saharan Africa.
Tropical Soil Biology and Fertility Institute of Cl AT, Nairobi, Kenya.
115
Kuykendall,
L.D.
and
sinnorhizobium
B. Saxena.
1991. Genetic
diversity
in Bradyrhizobium
Jordan 1982 and a proposal for Bradyrhizobium
L.M. elkanii
sp. novo Canadian Journal of Microbiology. 38:501-505.
KWAP Project. 1993. Extension
methods and techniques:
A manual for training of
field extension workers. ETC-Kenya. Nairobi, Kenya. Pp. 80.
Laguerre, G., M.P. Fernandez,
heterogeneity
V. Edel, P.
among
French
vulgaris L. International
ormand and N. Amarger.
Rhizobium
strains
Journal of Systemic Bacteriology.
Laguerre, G., M.R. AlIard, F. Revoy and N. Amarger.
rhizobia
by restriction
isolated
fragment
1993. Genomic
from
Phaseolus
43:761-767.
1994. Rapid identification
length polymorphism
amplified 16S rRNA genes. Applied Environmental
analysis
Microbiology.
of
of PCR60:56-63.
Laguerre, G., P. Mavingui, M.R. Allard, M.P. Charnay, P. Louvrier, S.l. Mazurier, L.
Rigottier-Gois
and N. Amarger.
fingerprinting
and PCR-restriction
chromosomal
leguminosarum
Microbiology.
and
symbiotic
and
its
1996. Typing
by PCR DNA
fragment length polymorphism
gene
different
regions:
Application
biovars.
Applied
to
analysis of
Rhizobium
Environmental
62:2029-2036.
Lara, M., J.L. Ortega, B.E. Olgluin and F. Sanchez.
nitrogen
of rhizobia
metabolism
in Phaseolus
1988. Nodule expression
vulgaris root nodules.
Bothe, H., F.J. de Bruij and W.E. Newton (eds.). Nitrogen
Pp. 617-622.
and
In.
fixation: hundred
years after. Gustav Fischer, Stuttgart.
Loomis, R.S. and P.J. Connor. 1992. Crop Ecology. Productivity
and management
agricultural systems. Cambridge University Press, Cambridge.
in
116
Ludwig,
W., R. RosseIl6-Mora,
R. Aznar, S. Klugbauer,
Beimfohr,
E. Brockmann,
Klugbauer,
N. Springer, D. Lane, R. Nietupsky,
Schleifer.
1995.
proteobacteria
G. Kirchihof,
Comparative
sequence
S. Dorn,
M. Bachleitner,
M. Weizenegger
analysis
system. Appl ied Microbiology.
of
23S
and K.H.
rR A
from
18: 164-188.
Macharia, P.N. 2003. Integrating fanners' and scientists'
soil mapping and management:
S. Spring, K. Reetz, C.
knowledge
in participatory
A case study from semi-arid eastern Kenya.
East African Agricultural and Forestry Journal. 69:39-47.
Maingi, J.M., N.M. Gitonga, C.A. Shisanya, B. Hornetz
Population
levels
of
indigenous
Bradyrhizobia
and G.M. Muluvi. 2006.
nodulating
soybean in two soils of the semi-arid and semi-humid
Journal of Agriculture
and Rural Development
promiscuous
agroecological
zones.
in the Tropics and Subtropics.
107(2):149-159.
Mapfumo, P., S. Mpepereki and P. Mafongoya. 2000. Pigeon pea rhizobia prevalence
and crop response to inoculation in Zimbabwean
Experimental
Marschner,
smallholder-managed
soils.
Agriculture. 36:423-434.
H. 1986. Mineral Nutrition of Higher Plants. Academic
Press, London.
Pp. 674.
Martin, B., O. Humbert, M. Camara, E. Guenzi, J. Walker, T. Mitchell, P. Andrew,
M. Prudhomme,
G. Alloing, R. Hakenbeck, D.A. Marrison, G.J. Boulnois and
J.P. Claverys. 1992. A highly conserved repeated DNA element located in the
chromosome
of streptococcus. Nucleic Acids Research. 20:3479-3483.
117
Mason, PJ. and J.G. Williams.
1985. Hybridization
in the analysis of recombinant
DNA. Pp. 113-137. In: Hames, B.D. and SJ. Higgins (eds.). Nucleic acid
hybridization:
a practical approach. ILR Press Ltd, Oxford, England. Pp. 245.
Masumba, B.A.N. 1984. Physiological
measurement
for ranking bean cultivars with
respect to relative drought and heat resistance. East African Agricultural
and
Forestry Journal. 44:290-296.
Mathuva, M.N., M.R. Rao, P.e. Smithson and R. Coe. 1996. Improving
mays)
yields
in semiarid
highlands
of Kenya:
agroforestry
maize (Zea
or inorganic
fertilizers? Field Crops Research. 55:57-72.
McCown,
R.I., Keating
sustainable
B.A., Probert
M.E. and Jones R.K.
crop production
in semiarid
Africa.
Qiao,
and M. An. 2007.
1992. Strategies
Outlines
of
of Agriculture.
21(1):21-23.
Miao,
S.J.,
Y.F.
development
X.Z. Han
formation
in soybeans (Glycine max L.) in response to phosphorus
in solution culture. Podosphere.
Mortimore,
Nodule
and
supply
17(1 ):36-43.
M.J., B.B. Singh, F. Harris and S.F. Blade. 1997. Cowpea in traditional
cropping
systems. Pp. 99-113. In: Singh, B.B., D.R. Mohan, K.E. Dashiell
and L.E.N. Jackai (eds.). Advances
International
in cowpea research.
Institute of Tropical Agriculture
Co-publication
of
(UTA) and Japan International
Research Centre for Agricultural Sciences (JIRICAS). IITA, Ibadan, Nigeria.
Mpepereki,
S., A.G. Wollum
specificity
and F. Makonese.
1996. Diversity
of cowpea rhizobia indigenous to Zimbabwean
186:167-171.
in symbiotic
soils. Plant Soil.
118
Mpepereki, S., F. Makonese
of indigenous
and A.G. Wollum. 1997. Physiological
rhizobia nodulating
characterization
Vigna unguiculata in Zimbabwean
soils.
Symbiosis. 22:275-292.
Mueller, J.G., H.D. Skipper, E.R. Shipe and L.W. Grimes. 1988. Intrinsic antibiotic
resistance
in Bradyrhizobium japonicum. Soil Biology
and Biochemistry
20(66):879-882.
Muleba, N. and. Ezumah
H.C. 1995. Optimizing
cultural practices for cowpea in
Africa. Pp 289-295. In Singh, S.R. and K.O. Rochie (eds.). Cow pea research,
production and utilization. John Wiley and Sons Ltd, Chichester, Uk.
Muli, M.B. and H.M. Saha. 2000. Participatory
evaluation
of cowpea cultivars for
adaptation and yield performance in coastal Kenya. Pp. 267-272. In: Mureithi.
J.G., G.K.K. Gachene, F.N. Muyekho, M. Onyango, L. Mose and O. Magenya
(eds.). Participatory
technology
development
for soil management
by small
holders in Kenya. Kenya Agricultural Research Institute, Nairobi, Kenya.
Mulongoy, K. and A. Ayanaba.
1985. Dynamics of the population
and soybean
rhizobia
Microbiology
and Biotechnology.
sizes of cowpea
at three locations in West Africa. World Journal of
2(2):301-308.
Muruli, B.I., R.S. Pathak, D.M. Mukunya, A.K. Karel, S.O. Keya and H. Ssali. 1980.
Cowpea research in Kenya. Tropical grain legume Bulletin. 19: 13-16.
Muthamia,
J.G.N. and F.K. Kanampiu.
1996. On-farm
cowpea
evaluation
In the
marginal areas of eastern Kenya. Pp. 677-685. In: Fungoh, P.O. and G.C.O.
Mbadi (eds.). Focus on Agricultural Research for Sustainable Development
Changing
Economic
Environment.
Conference, Nairobi, Kenya.
Proceedings
in
of the 5th KARI Scientific
119
Nadar, H.M. 1984. Intercropping
rainfall conditions
and intercrop component
interaction under varying
in eastern Kenya. I: Maize/bean
intercrop.
East African
Agricultural and Forestry Journal. 44:166-175.
Nakawuka,
C.K. and E. Adipala.
component
interactions
1999. A path coefficient
analysis
of some yield
in cowpea. African Crop Science Journal. 7(4):327-
331.
Nelson,
D.e.
and R.H.
Robichaux.
1997. Identifying
associated with yield under intercropping
plant
implications
architectural
traits
of genotype-cropping
system interactions. Plant Breeding. 16: 163-170.
Ng, N.Q. and R. Marechal.
1985. Cowpea taxonomy, origin and germplasm.
Pp. 11-
21. In: Singh S.R. and K.O. Rachie. (eds.). Cowpea research, production and
utilization. John Wiley and & Sons, New York.
Nour,
S., J.e.
Cleyet-Marel,
Genotyphic
D. Beck, A Affosse
and phenotyphic
and M.P.
diversity of Rhizobium
S., J.C. Cleyet-Marel,
heterogeneity
description
of strains
of Rhizobium
Systemic Bacteriology.
Nziguheba
P. Normand
G, R Merckx,
nodulating
40:345-354.
and M.P. Fernandez.
chickpea
mediterraneum
1994.
isolated from chickpea
(Cicer arietinum L.). Canadian Journal of Microbiology.
Nour,
Fernandez.
(Cicer
1995. Genomic
arietinum
sp. novo International
L.) and
Journal of
45:640-648.
e.A.
Palm and P.K. Mutuo.
2004. Combined
use of
Tithonia diversifolia and inorganic fertilizer for improved maize production in
a phosphorus
deficient soil in western Kenya. Pp. 329-346. In: Bationo, A.
(ed.). Managing nutrient cycles to sustain soil fertility in sub-Saharan
Tropical Soil Biology and Fertility Institute of Cl AT, Nairobi, Kenya.
Africa.
120
Ochman, H. and A.c.
substitution
Wilson.
1987. Evolution in bacteria: evidence
rate in cellular genomes. Journal of Molecular
for universal
Evolution. 26:74-
86.
Odee, D.W., K Haukka, S.G. Mclnroy, J.I Sprent, J.M. Sutherland and LP.W. Young.
2002. Genetic and symbiotic characterization
and herbaceous
of rhizobia
isolated from tree
legumes grown in soils from ecologically
diverse sites in
Kenya. Soil Biology and Biochemistry. 34:801-811.
Ogbonnaya,
c.r, B.
Selection
Sarr, C. Brou, O. Diouf, N.N. Diop and H. Roy-Macauley.
of cow pea genotypes
in hydroponics,
2003.
pots and field for drought
tolerance. Crop Science. 43:1114-1120.
Ojiem,
J.O.,
C.A.
combining
Palm,
and M.A.
organic and inorganic phosphorus
a humic-Nitosol
Managing
E.A. Okwuosa
in western
nutrient
Kenya.
Mudeheri.
2004. Effect of
sources on maize grain yield in
Pp. 347-358.
cycles to sustain soil fertility
In: Bationo,
A. (ed.).
in Sub-Saharan
Africa.
Tropical Soil Biology and Fertility Institute of CIAT, Nairobi, Kenya.
Okalebo, J.R., C.A. Palm, J.K. Lekasi, S.M. Nandwa, C.O. Othieno, M. Waigwa and
K.W. Ndung'u.
2004. Use of organic and inorganic
resources
to increase
maize yields in some Kenyan infertile soils: A five-year experience. Pp. 359372. In: Bationo, A. (ed.). Managing nutrient cycles to sustain soil fertility in
sub-Saharan
Africa. Tropical
Soil Biology and Fertility
Institute of CIAT,
Nairobi, Kenya.
Okalebo, J .R., K.W. Gathua and P. Woomer. 2002. Laboratory
plant analysis:
A working
Nairobi, Kenya. Pp. 128.
manual. The second edition.
methods of soil and
SACRED Africa,
121
Okoko,
E.N. and S. Makwaro.
2000. Evaluation
of the effect of compost
and
inorganic fertilizer on maize yield in Nyamira District South West Kenya. Pp.
23-29. In: Mureithi,
J.G., C.K.K. Gachene, F.N. Muyekho, M. Onyango, L.
Mose and O. Magenya (eds.). Participatory
management
by smallholders
conference
technology
in Kenya. Proceedings
of the soil management
development
for soil
of the 2nd scientific
and legume research network projects.
KARI, Nairobi, Kenya.
Olufajo, 0.0. and B.B. Singh. 2000. Advances in cowpea cropping systems research.
In: Fatokun, CA, S.A. Tarawali,
Pp. 267-277.
B.B. Singh, P.M. Kormawa
and M. Tamo (eds.). Challenges and opportunities
for enhancing sustainable
cowpea production. lIT A, Ibadan Nigeria.
Onyango, R.M.A., TJ. Mwangi, W.W. Kiinya, M.K. Kamidi and M.W. Wanyonyi.
2000. Evaluation
production
for smallholder
maize
in North Rift Kenya. Pp. 1-12. In: Mureithi, J.G., c.K.K. Gachene,
F.N. Muyekho,
technology
of organic and inorganic fertilizers
M. Onyango, L. Mose and O. Magenya (eds.). Participatory
development
Proceedings
for soil management
of the 2nd scientific conference
by smallholders
in Kenya.
of the soil management
and
legume research network projects. KARI, Nairobi, Kenya.
Owolade, O.F., M.O. Okande, B.S. Alabi and J.A. Adediran. 2006. Phosphorus level
affects brown blotch disease development and yield of cowpea. World Journal
of Agricultural
Palm, C.A.
Sciences. 2(1): 105-108.
1995. Contribution
intercropped
of agroforestry
trees to nutrient
plants. Agroforestry Systems. 30: 105-124.
requirements
111
122
Peffetti, D., C. Scotti, S. Gnocchi, S. Fancelli and M. Bazzicalupo.
1996. Genetic
diversity of an Italian Rhizobium meliloti population from different Medicago
saliva
varieties. Applied Environmental Microbiology. 62:2279-2285.
Pilbeam, C.A., M. Wood and P.G. Mugane. 1995. Nitrogen use in maize-grain
legume cropping systems in semi-arid Kenya. Biology of Fertilizers and Soils.
20:57-62.
Polhill, R.M. and L.J.G. Van der Maesen. 1985. Taxonomy of grain legumes. Pp. 336. In: Summerfield,
R.J. and E.H. Roberts (eds.). Grain Legume Crops.
ColI ins, London.
Poshiwa, X., R. Nyoka, N. Chikumba, T. Gatsi and 1. Chakoma. 2007. A baseline
socio-economic
evaluation of an European Union funded project on use of
micro-symbionts
in agroforestry systems in Zimbabwe. Journal of Sustainable
Development in Africa. 8(1): 186-196.
Powell,
J.M.
1986. Manure for cropping: A case study from central Nigeria.
Experimental Agriculture. 22: 15-24.
PRA Programme
Egerton University.
rural appraisal practitioners.
1995. PRA field handbook for participatory
1995. The PRA Programme Egerton University,
Njoro, Kenya. Pp. 84.
Probert,
M.E.,
J.R. Okalebo
smallholders'
and R.K. Jones.
1995. The use of manure
fams in semiarid eastern Kenya. Experimmental
on
Agriculture.
31:371-381.
Remanandan,
P., A. Shakoor and E.C.K. Ngugi. 1982. Pigeon peas germplasm
mission in Kenya. Genetic Resources Progress Report (ICRISAT). 48: 1-40.
123
Republic
of Kenya.
2001. The 1999 population
and housing
census. In: Central
Bureau of Statistics and Ministry of Finance and Planning. Volume 1. Pp. 415.
Rowe, A.G.
1980. Alternative
crops 3: Grain
legumes.
Zimbabwe
Agricultural
Journal. 79: 221-227
Saha, R.e.
and M.B. Muli. 2000. Effect of combining
green manure
legumes,
farmyard manure and inorganic nitrogen on yield of maize in coastal Kenya.
Pp.l03-113.
In: Mureithi, J.G., e.K.K. Gachene, F.N. Muyekho, M. Onyango,
L. Mose and O. Magenya (eds.). Participatory
soil management
conference
technology
by smallholders in Kenya. Proceedings
of the soil management
development
for
of the 2nd scientific
and legume research network projects.
KARI, Nairobi, Kenya.
Saidou, A.K., R.e.
Variability
Abaidoo,
B.B. Singh, E.N.O. Iwuafor and N. Sanginga.
of cowpea
breeding
lines to Iow phosphorus
tolerance
2006.
and
response to external application of phosphorus. Pp. 413-422. In: Bationo A, B.
Waswa, J Kihara and 1. Kimetu (eds.) Advances
management
in sub-Saharan
Africa: Challenges
in integrated soil fertility
and opportunities.
Springer,
Dordrecht, The Netherlands.
Sakala, WD., G. Cadisch, K.E. Giller. 2000. Interactions
and pigeon pea and mineral nitrogen fertilizers
nitrogen mineralization.
between residues of maize
during decomposition
Soil Biology and Biochemistry.
and
32:679-688.
Sanchez, P.A., K.D. Shepherd, MJ. Soul, F.M. Place, RJ. Buresh and A-M. N. Izac.
1997. Soil fertility replenishment
capital. Replenishing
46.
in Africa: An investment in natural resource
Soil Fertility in Africa. SSSA Special Publication.
51: 1-
124
Sande, S., c.K.P.O.
Ogol and P.L. Woomer. 2004. Soil food webs as influenced by
soil fertility management
(ed.)
Improving
empowering
practices. P. 55. In: Bationo, J Kimetu and J Kihara.
human
farmers
welfare
and
to combat
environmental
soil fertility
conservation
degradation.
by
Tropical
Soil
Biology and Fertility Institute of Cl AT, Nairobi, Kenya.
Selenska-Pobell,
S., L. Gigova and N. Petrova. 1995. Strain-specific
fingerprinting
of
Rhizobium galegae generated by PCR with arbitrary and repetitive primers.
Journal of Applied Bacteriology. 79:425-431.
Shakoor, A., E.C.K. Ngugi, M.S. Muthoka and J. W. Karnau. 1984a. Improvement
cowpea
varieties
for dry areas in Kenya. East African
Agricultural
of
and
Forestry Journal. 44: 312-317.
Shakoor, A., W.K. Rono and E.C.K. Ngugi. 1984b. Performance
determinate
varieties
of green gram in semi-arid
of early maturing
areas.
Eastern
African
Agricultural and Forestry Journal. 44:324-326.
Sharma, R.C., P.M. Govindakrishnan,
different
needs
green manures
of potato-wheat
R.P. Singh and H.c. Sharma. 1988. Effect of
and fodder crops on crop yields and phosphorus
systems
in the
north-western
plains
of
India.
Experimental Agriculture. 24:79-103.
Sharples, G.J. and R.G. L1oyd. 1990. A novel repeated DNA sequence located in the
intergenic
regions
of bacterial
chromosomes.
Nucleic
Acids
Research.
18:6503-6508.
Shearer, G. and D.H. Kohl. 1986. N2-fixation
natural
in field settings: estimations
N abundance. Australian Journal of Plant Physiology.
J5
based on
13:699-756.
125
Sheldrake, A.R. and A. Narayana.
1979. Growth development
pigeon peas tCajanus cajan). Journal of Agricultural
Shetty, S.V.R., B.R. Ntare, A. Bationo and C. Renard.
mixed cropping
productivity
and nutrient uptake in
Sciences. 92: 513-526.
1995. Millet and cowpea in
systems of the Sahel: A review of strategies
and sustainability.
Pp. 293-303. In: Powel, J.M., S. Fernandez-
Rivera and T.O. Williams (eds.).
in mixed farming
for increased
Livestock and sustainable
systems of Sub-Saharan
Africa.
nutrient cycling
International
Livestock
Centre for Africa (ILCA), Addis Ababa, Ethiopia.
Siderius
W. and F.M.
Muchena.
1977. Soils and environmental
agricultural research stations in Kenya. Miscellaneous
conditions
in
paper No. M5. Ministry
of Agricultural University.
Sikora, S., S. Redzepovic,
Bradyrhizobium
I. Pejic and V. Kozumplik.
japonicum
field population revealed by RAPD fingerprinting.
Journal of Applied Microbiology.
Singh, B.B., O.L. Chamblis
1997. Genetic diversity of
82:527-531.
and B. Sharma.
1997. Recent
advances
in cowpea
breading. Pp. 30-49. In: Singh, B.B., D.R. Mohan, K.E. Dashiell. and L.E.N
Jackai (eds.). Advances
in cow pea research. IITAIJIRCAS,
Ibadan, Nigeria.
Ibadan, Nigeria.
Singleton, P.W., B.B. Bohlool and P.L. Nakao. 1992. Legume rhizobia inoculation in
the tropics:
Myths and realities. Myths and science of soils in the tropics.
ASA, SSSA. Special publication. 29: 135-155.
Smaling,
E.M.A.
management
1993.
An agroecological
with special
University Wageningen,
reference
framework
to Kenya,
The Netherlands.
for
PhD.
integrating
thesis.
nutrient
Agricultural
126
Smaling, E.M.A. and A.R. Braun. 1996. Soil fertility research in Sub-Saharan Africa:
A new dimension,
new challenges communication
in soil science and plant.
Analysis. 27:365-386.
Smaling, E.M.A., S.M. Nandwa and B.H. Janssen. 1997. Soil fertility in Africa is at
stake. Replenishing
soil fertility in Africa. SSSA special publication.
51 :47-
61.
Sobral, B.W.S., RJ. Honeycut and A.G. Atherly. 1991. The genomes of the family
Rhizobiaceae:
size, stability
Journal of Bacteriology.
Somasegaran,
and rarely cutting
restriction
endonucleases.
173 :704-709.
B. and HJ. Hoben. 1985. Methods in legume-Rhizobium
technology.
University of Hawaii NiIT AL * Project and MIRCEN*. Pp.367.
Sprent, J.1. 1971. Effects of water stress on nitrogen fixation in root nodules. Plant
and soil. 35(1):225-228.
Stern, MJ.,
G. Ferro-Luzzi
Ames, N.H. Smith, E.C. Robinson
and C.F. Higgins.
1984. Repetitive extragenic palindromic sequences: A major component of the
bacterial genome. Cell. 37:1015-1026.
Stevenson,
F.C. and C. Van Kessel. 1997. Nitrogen
contribution
of pea residue in
hummocky terrain. Soil Science Society of American Journal. 61 :494-503.
Stoetzer, H.M. and B.H. Waite. 1984. Evaluation of bean diseases in grain legume
project
trials, Katumani
Research
Station.
East African
Agricultural
and
Forestry Journal. 44:333-354.
Stoorvogel,
J.J., E.M.A. Smaling and B.H. Hansen.
balances
in Africa
Research. 35:227-235.
at different
scales:
1993. Calculating soil nitrogen
IJ supra-national
scale.
Fertility
127
Subarao, G.Y., C. Renard, A. Bationo, N. van-Duivenbooden
Alternative technologies
Pp. 121-132.
and C. Bielders.
1999.
for Sahelian crop production systems in West Africa.
In: Faroda,
advances in management
A.S., N.L. Joshi and S. Kattju
(eds.).
Recent
of arid ecosystems. Arid zone research Association
of India, Jodhpur, India.
Swift, MJ., P.D. Seward, P.G.H. Frost, J.N. Quresh and F.N. Muchena.
term experiments
1994. Long-
in Africa: Developing a database for suitable landscape use
under global change. Pp. 229-251. In: Legh, R.A. and A.E. Johnson (eds.).
Longterm
experiments
International
in
agricultural
G. Gueye,
A.T. Ba and B. Dreyfus.
biological nitrogen fixation by pterocarpus
park in Senegal
Tarawali,
ecological
sciences.
CAB
Wellingford, England.
Sylla, S.N., 1. Ndoye,
Biotechnology.
and
using
15N
2002.
Estimate
of
luvens in semi arid natural forest
natural abundance
method. African
Journal
of
1(2):50-56.
S.A., B.B. Singh, M. Peters and S.F. Blade. 1997. Cowpea
fodder. Pp 313-325.
haulms as
In: Singh, B.B., D.R. Mohan Raj, K.E. Dashiell
L.E.N Jackai (eds.). Advances
in cowpea research. lIT AIJIRCAS,
and
Ibadan,
Nigeria.
Tesfaye, M., DJ. Peterson and F.B. Brian Holl. 1997. Comparison
of partial 23S
rDNA sequences from Rhizobium species. Canadian Journal of Microbiology.
43:526-533.
The MacmiJlan Atlas. 1997. The Macmillan primary school atlas. Macmillan
(Publishers) Ltd. Nairobi, Kenya. Pp. 57.
Kenya
128
Tisdale,
S. and W.
elson.
Macmillan Publishing,
1975. Soil fertility
and fertilizers.
Third
New York.
Urtz, B.E. and G.H. Elkan. 1996. Genetic diversity among Bradyrhizobium
that effectively
Microbiology.
edition.
nodulate
peanut (Arachis hypogaea).
Canadian
isolates
Journal of
42:1121-1130.
Van Ek, G.A., J. Henriet, S.F. Blade and B.B. Singh. 1997. Quantitative
assessment
of traditional cropping systems in the Sudan Savanna of Northern Nigeria. 11.
Management
and productivity
Agricultural Research.
of major cropping systems. Samaru Journal of
14:47-60.
Van Kessel, C. and C. Hartley. 2000. Agricultural management
of grain legumes: has
it led to an increase in nitrogen fixation? Field Crops Research. 65: 165-181.
Van Rheenen, H.A., O.E. Hasselbach and S.G.S. Muigai. 1981. The effect of growing
bean together
with maize on the incidence
of beans diseases
and pests.
Netherlands Journal of plant pathology. 87:193-199.
Vandamme,
P., B. Pot, M. Gillis, P. de Vos, K. Kersters
Polyphasic
Microbiologyl
Vanlauwe,
taxonomy,
a consensus
approach
to
and J. Swings.
bacterial
1996.
systematics.
Reviews. 60:407-438.
B., K. Aihou, P. Houngnandan,
J.Diels, N. Sanginga
and R. Merckx.
2001. Nitrogen management
in 'adequate' input maize-based
agriculture in the
derived savanna benchmark
zone of Benin Republic. Plant and Soil. 228:61-
71.
Versalovic,
J., M. Schneider,
fingerprinting
F.J. de Bruijn and J.R. Lupski.
of bacteria using repetitive sequences-based
1994. Genomic
polymerase
reaction. Methods in Molecular and Cellular Biology. 51 :25-40.
chain
129
Vincent,
J.M.
1970.
International
A
manual
Biological
for practical
Programme
study
Handbook
of
root-nodule
bacteria.
No. 15. Blackwell,
Oxford.
Pp. 164.
Vinuesa, P., J.L.W.
Rademaker,
characterization
FJ de Bruijin and D. Werner.
of Bradyrhizobium
strains nodulating endemic wood legumes
of the canary islands by PCR-Restriction
genes
encoding
16S rRNA (l6S
Spacers. Repetitive
partial
16S
Length Polymorphism
rDNA)
Extragenic Palindromic
rDNA
sequencing.
1998. Genotypic
and
16S-23S
rDNA
analysis of
Intergenic
PCR genomic fingerprinting
Applied
Environmental
and
Microbiology.
64:2096-2104.
Vos, P., R. Hogers, M. Beeker, M. Reijans, T. Van de Lee, M. Hornes, A. Frijters, J.
Pot, J. Peleman, M. Kuiper and M. Zabeau. 1995. AFLP: a technique for DNA
fingerprinting.
Nucleic Acids Research. 23:4407-4414.
Wade, T.K., 1. Ndoye, S. Braconnier, B. Sarr, P. de Lajudie and M. Neyra. 2003.
Diversity of indigenous bradyrhizobia associated with three cowpea cultivars
(Vigna unguiculata L.) grown under limited and favourable water conditions
in Senegal (West Africa). African Journal of Biotechnology.
Waite, B.H., A. Shakoor,
W. Songa. and E.C. Ngugi.
2(10: 13-22.
1984. Principal
diseases of
pigeon peas, cowpeas, chickpeas and green gram in the semi-arid areas in
Kenya. East African Agricultural and Forestry Journal. 44: 364-375.
Wanjekeche, E., TJ. Mwangi, P. Powon and J. Khaemba. 2000. Farmer participation
in the evaluation
of grain legumes in north-western
Kenya. Pp. 13-21. In:
Mureithi, J.G., C.K.K. Gachene, F.N. Muyekho, M. Onyango, L. Mose and O.
Magenya
(eds.). Participatory
technology
development
for soil management
130
by smallholders in Kenya. Proceedings of the 2nd scientific conference of the
soil management
and legume research network projects.
KARI, Nairobi,
Kenya.
Wedlock, D.N. And B.D.W. Jarvis. 1986. DNA homologies
fredii,
rhizobia
Bradyrhizobium
that
nodulate
Galega
sp.,
and
between Rhizobium
other
Rhizobium
species. International Journal of Systematic
and
Bacteriology.
35:550-558.
Wild, A.1988. Russell's
Soil Conditions and Plant Growth. Eleventh edition. John
Wiley and Sons, New York.
Willems, A., R. Coopman and G. Gillis. 2001. Comparison of sequence analysis of
16S-23S DNA spacer regions, AFLP analysis and DNA:DNA hybridization in
Bradyrhizobium.
International
Journal
of
Systematic
Evolutionary
Microbiology. 51 :623-632.
Williams, R.D. and P.W. Bartholomew. 2008. Effect of cowpea and pea inocula on
cool season grass. Allelopathy Journal. 21(2):381-388.
Witty, J.F. 1983. Estimating of N2-fixation in the field using 15N labeled fertilizer:
some problems and solutions. Soil Biology and Biochemistry.
Xu, L.M., C. Ge, Z. Cui, J. Li and H. Fan. 1995. Bradyrhizobium
15:631-640.
Liaoningensis sp.
novo Isolated from the root nodules of soybean. International
Journal of
Systematic Bacteriology. 45: 706-711.
Young, lP.W.
1996. Phylogeny and taxonomy of rhizobia. Plant and Soil. 186:45-52.
Zhang, X., R. Harpen, M. Karsists and K. Lindstrom. 1991. Diversity of Rhizobium
bacteria isolated from the root nodules of leguminous
Journal of Systematic Bacteriology. 41: 104-113.
trees. International
131
APPENDICES
Appendix 1
Maps ofYikivumbu, Ndunguni, Kavuthu and Matiku Sublocations drawn by farmers during farmer meetings
VOLOLO
SUB. LOCATION
,
\
I
1
I
I
\
\
\
\
,
"-
-"
,
,
I
I
"" " ,
--- " , " \
", \
r
/
I
" " +6
I
I
I
I
I
/
:\p...~\
~\)p...
'-l\~
I
I
~f>.0t-/\
•
I
+L>
Assistant
Chief's Office
I
I
I
I
I
~
--------....----.-__
I
/...................
/
NGOMANO
VILLAGE
/
\\ @
\
'.•... ~
/
tSsl.
OD
:::.:::
>
1'--.
z
::)
W
>
I
0
I
z
(9
<C
-.:~.•..
,"'"
""'\
j
;;
""":......,;::"
\'....
\
-- .-""
KYENI
\ VILL.
SR I
\
~
\
@
LEGEND:
-SChOOI
SS -Seed store
~
-Market
»
-Shrin£>s
CD - CaUl£> dip
H
-Hospital
+ -
Figure 1
-
-
__
Village boundaries
Boundaries
__________
- River (Stream)
(Dtspensary)
Church
Map ofYikivumbu
Sub-location
132
".
'
S.Army
Church
..'"
K
Muthyaamui
I
,.SOWETO
MARKET
Dam
,!
/
....
MAKUTANO
• MARKET
~~.::..- ~:luia
•
':~'r~
J.- V':
Nyumu
':\
.•
Mullu
\- •.•
1;?
~, //'
<y
V
Killi
.".\"'?/
":'"
-'.....
-.\,
Nzioka
Z\,\"
:\(
~
<\\',\~\""
~
\'\-
~
N7..rmgi
•..••
•
Ale
Kuvuva
\~
\{
»>:
.' ~.~::~_\~;::::::.:;"-
+NI
••;-;-
.•...::"',/
/"
••
-"
,~
.y/'~<':, •
.,Y/. \~\
.r,:
.e>
7'
/
Ass, Chiefs
Office
:'
:
Donald
\\
(/1--0
./
••.
Ii 3
\ , go
'C-'-'MUlhusi
Muteti
\
\7.
Mutuku
-:>: :)1l
~
£,MilOlU
\\~.
S:
("'
Wambua
Ngule
""<::",.
",,- '-,:.
.&.
~:
'-
., River
:::::=== Road
____
Figure 2
Dam
•
Outside boundary
Map ofNdunguni
Nyumu
<:".............::..
-...
~,-_...
Sub-location
"
1\
\
". "\
~
Town/Market
!
:
i
I,
y
:.:.\\
I
!I
0
\\
v
I
•• \..
..\-:
•••••••••••....• ~ .• ,
••••
----._~....-.~-•.•.-.,._...
'-"-_ ... _ .._.-.---_ .
Homestead
ji
,I
1'-
~ \),-
~" •. A Ngule
......
..
.. "
-.v
".
,
~
~.
\'\.
\~~.
\;'~ •••.
I
~\KinYili
•••.
CATHOLIC
CHURCH
Ndonye
KEY:
Muteli!
Ngu,"~
....~?~_--:;,
;
: L..-;r';: ..--'
1\
.-<;:.-::y i
1
NDUNGUNI
\\\±..
!I!
1:'<
I'
-.\
Nyamai",
i
I i !!l
.:\
e
\
,:
i
If
I
.\
: \
r,
l.-
I!.
: ~
\'-'\\
'7
,:
:\
\'\.
~
Mwau
-. ~
••.•·.NDUNGUNi\ '\
SCHOOL
\,',
.o.~·
l'
I:
..-..
• :
\·r_.·;:-~OhfiKitenge
',-'P
.'
;JUNCTION
,'~IKWAMBITI
-:> .. I
Mukula ••...
:;:::~~:>/-.-".\
\\~~ KATlVANI
....·.TOWN
i
~~'
.~>-. '!:-;::~~
"V/
i~ •••••
133
\
I
,I:
lJ'~-~~!£J11
~~rHu
hI [ ------_
-, O(
l
_
MARKO-
" (I,
!If Kavuthu
(j i 1+ Cdtholic
..J 0 j
+ Redeemtld
/- J...
J
1~
U)
"11
t~
\
avu u
_, PI!, School
'tAlC,
\ Kavuthu
I
I
+ABC
I U\nhi
~/-------~
j-TK
Church
51
GNCA
K AV UT H U
I
-----
0 KioSKS
"
, Salvation army
+ Church
A~C\t Catholic
Ngo~d Church
Church I T U M BUN /
•
IN G 0 M A N 0 I V ILL
I
I
I
9lnne, I
~
~\\ll'
AGE
Liani
Shrine
MASUE
SI LOCATION
+AIC
Tumbuni
I
IVILLAGEI
VILLAGE
--
+
Catha/!: Church
\
~~ I
I
W
E
5
MUIU
V
ILL A 'G E
,MUllSYA
I\ VILLAGE
I
I
Figure 3
Map ofKavuthu
MULU
\
Sub-location
LEGEND
::== Roads
-
S/Location
boundary
•• - -
Vil~ge boundary
134
IVISUNI
VILLAGE
N
Church
\j
School
Shrine
- ~
Roads
Dam
-------
Riv~r
•
®
OD
o
Figure 4
LEG END
+
-
SlLocation boundary
-
Market
Map ofMatiku
Sub-location
Village boundary
~
135
Appendix 2
Questionnaire
Questionnaire No. D
DD
MM
YY
ITIJ
Date of interview:
Name of respondent
District
Code
Division:
codeD
Location:
1
Sub Location:
__________
Household code:
code
1
---'1
--'
D
code D
--'
__________
Village:
1
D
code D
D
SECTION A: FAMILY INFORMATION
Table A.I. Personal information for members of the household currently resident: Fill where
appropriate
ID
Name
Age
(Group)
1-::;20yrs
2- 21-30
3- 31-40
4-41-50
5- 51-60
6-~61 yrs
1
2
3
Marital status
Sex
1- Single
2- Married
3- Widowed
4- Divorced
5- Separated
Male-I
Female-2
4
5
Relationship
to HIH head
1- Head
2- wife/Husband
3- Son/daughter
4- mother/father
5-other
l-None
2- Primary
3-Secondary
4-University
5-0ther
7
6
I
2
3
Total resident HIH members
Highest
education
Level
.
Main occupation
I-Farming!
Livestock keeping
2·Business
3-Employed
4- Other
5-None
8
136
I (A) What is the general condition of housing in the HH? Specify condition for up to five
houses:
Table A 2 Housing conditions within the Household
Wall
House No
1- Earth mud
2- Bricks/Stones
3- Iron sheets
4- Wood slabs
5- Maize stovers
6- Withies
I
1. Main House
2
Floor
Roof
I Earth/mud
2- Cement
3- Withies
I-Thatch
2-lron sheets
3-Tiles
4-0ther
3
4
2
3
4
5
6
SECTION B: FARM INFORMATION
Farm owned and managed by household
1'--__
I(B) Which year did you settle in this farm?
2(B) How many fields do you operate?
Field
ID
I
Total
area
(Acres)
2
Cultivated
area
(Acres)
Area under
Legumes
(Acres)
3
4
Tenure System
I-surveyed
2-Lease
3-unsurveyed
4-Communal
5- Other
Ownershi
p
I-Own
2-Rent
Rent In
Land cost:
Rent out
land income
5
6
7
8
Table B.l. Please provide fields information as specified in the table below for the current
season
3(B) Besides the legumes you grow on your fields, would you like to try others? 1- Yes
2-N°D
4(B) Who makes decisions concerning types of crops grown in Iyour farm?
0
l-Household head
3- Household head and son
5- Wife of household head
0
0
2- Household head and wife
4- Son of household head
6- Other (specify)
c=J
0
137
Table B.2. Information
fields
Field
ID
1
on the crops grown in the fields and production
Crops grown in
field
1- Beans
2- Pigeon peas
3- Cow peas
4-Green grams
Legume
location
where
terraced
1- Inside
2- On top
3- On slopes
4 Away
5- Other
2
3
Table B.3: Please provide information
Unit
Of
yield
of the
Total
produced
I
5
4
on soil fertility, Soil Erosion Problems and Control
Field
[D
Soil fertility
Status
Fertile
Moderately
fertile
Poor
Farm inputs
I-Manures
2-Fertilizers
3-Crop
residues
4-Compost
5-None
6-0ther(Specify)
Soil Erosion
Problem
Yes
No
Soil
Conservatio
n structures
present
Yes
No
Conservation structures
types present
Bench terrace
Bench terrace with grass
Bench terrace with trees
Bench terrace with grass
and trees
Check dam
Cut off drain
Other (specify)
I
2
3
4
5
6
5(B) In case you use fertilizer
when required?
I-Yes
D
2-No
In your farm,
IS
fertilizer available
In local market
D
6(B)Incase you have soil erosion problems and you do not have terraces, give reasons why you don't
have them?
7(B) Where do you get your manure?
I-own
2-buy
3-NeighbourD
D
D
Other (specify)
_
138
D
D
8(B)Would you like to try other fertility improvement methods? I-Yes
2-No
9(B) If you were trained on best fertilizer management practices, would you use fertilizer
in your farm?
l-yeSO
2-noO
SECTION C: WEEDS, PESTS AND DISEASES
T a bl e ID - W ee d s, pests an d diiseases
Field
Weeds in your farm
ID
l-Kithangai
2-Ikoka
3-Munyeeli (Mukuutu)
4-Munzee
5-Mbiu (kiviu)
6-Kikatu
7-Ukuku
§.-Untunga
9-mbete (kavete)
IO-Ikongo
II-Song'e
12-Lamuyu
I3-0ther (specify)
1
2
I-hoe/panga weeding
2-plough-weeding
3-use of herbicide
4-burning
5-carrying out offarm
6-other (specify)
Pests problems
I-Squirrels
2-birds
3-monkeys
4-scania "Osarna"
5-weevils
6-Wiu
7-Umuu
8-0ther (specify)
3
4
Weed control in your
farm
Disease
problems
I-mbaa
2-rust
3-other
(specify)
5
SECTION D: LIVESTOCK KEEPING
D
D
1(0) Do you keep livestock? I-Yes
2-.No
2(0) Which of the following livestock do you ke~
l-Cattle
2-goat
3-sheep L-J
5-donkey
6-rabbit
7-other (specify)
3(0) How do you keep your livestock
l-free-range
2-tethering
3-paddocking
(Specify)
_
c::::=::J
c=::J
c::::=::J
C=::J
C=::J
4-poultry
_
C=::J
0
C=::J
c:=J 4-other
SECTION E: FOOD SECURITY
1(E) Do you have food shortage during the dry seasons?
1-Yes IL-__
2(E) What is your ~ource
of food during dry seasons
l-reserve
4-neighbours
(own)
U
0
2-purchaseD
5-0ther (specify)
--'
2-No
3-ReliefD
_
139
Appendix 3
Experimental
design (ISFM trial)
On-farm trial design with 4 treatments and 9 cow pea varieties in three blocks
Block 1
Treatment/varieties
3
I
2
4
I
2
3
4
5
6
7
8
9
5
2
I
4
7
6
8
3
9
2
7
4
I
6
5
3
8
9
I
5
2
3
6
4
9
8
7
4
I
2
3
8
7
6
9
5
Block 2
1
2
3
4
5
6
7
8
9
)
4
2
3
8
5
7
6
9
)
2
)
7
6
8
5
3
2
4
9
4
3
8
9
5
4
7
6
)
2
3
1
9
2
3
6
4
8
7
5
Treatment/varieties
Block 3
1
2
3
4
5
6
7
8
9
3
9
5
1
4
3
6
2
7
8
4
5
4
7
8
1
6
9
2
3
2
1
3
8
5
2
4
6
9
7
1
9
6
3
2
4
1
8
7
5
Treatment/varieties
140
Appendix
a)
4
Rhizobia
cultures
isolated from cow pea plant nodules
Isolates used to group rhizobia strains
Strain
group
I
Number of
cultures
.,.,
11
2
3
11
59
4
5
8
21
6
7
8
3
9
5
2
b)
Culture codes
3,6,1 1,12,15,16,19,20,3 1,32,69,70,2 I 1,212,2 I 5,2 I 6,2 I 9,240,247,
248,25 I ,252,261 ,284,286,299,303,304,343,345,346,374
7,8,152, 154,184,186, I 88, I 89,1 90,233,279,347
5,15,16,36,40,109,1
10,1 13,127, I 28, 129, 130,1 3 1,145, I 51,153,165
,167,168,171, I 74, I 79, 181, I 82, 183,187,223,225,226,228,234,239,
262,263,27 I ,272,275,276,277,278,280,283,289,290,3
I 1,312,323,
324,333,334,335,336,342,343,348,354
39,111,112,133,166,207,208,227
22,24,25,26,27,29,30,38,107,108,
114,220,285,293,329,330,353,3
54,365,366,373
23,132,146
20 I ,202,204,231 ,232,245,246,34 I ,369
218,291,292,300,370
180,224
Isolates used to determine rhizobia acidity and growth rate
Strain
group
I
II
III
IV
V
VI
VII
VIII
IX
Number of
cultures
6
11
18
Culture codes
3,6,212,211,247,346
7,8,152,154,186,188,189,190,233,279,347
5,l5,84, 110, 130, 131,151,153,174,179,181,182,228,27
2,280,290,333,348
39, 111,112,133,166,207,208,227
22,24, 25,38,107,285,330,354
23,132
201,202,204,231,232,245,246,341,369
218,291,292,300,370
180,224
8
8
3
9
5
2
Meaning of culture codes
Numbers 1-50: M14 treatment one; Numbers 51-88: M14 treatment two;
Numbers 145-192: M14 treatment four; Numbers 193-220: Kang'au treatment one;
Numbers 221-264: Kang'au treatment two; Numbers 265-304: Kang'au treatment
three; Numbers 305-344: Kang'au treatment four; Numbers 345-360: Kang'au April
(soils collected before planting); Numbers 361-376: M14 April (soils collected before
planting).