VARIABILITY AND HETEROSIS IN CO.WPEA 
(VIGNA UNGUICULATA (L ) WALP. ) 
ACCESSIONS FROM FOUR REGIONS OF GHANA
By
SAMUEL ODEI BENNETT-LARTEY
A thesis submitted in partial fulfilment of the 
requirements for a Doctor of Philosophy Degree in 
Science at the University of Ghana.
CROP SCIENCE DEPARTMENT 
FACULTY OF AGRICULTURE 
UNIVERSITY OF GHANA 
IjEGON
AUGUST, 1991
University of Ghana          http://ugspace.ug.edu.gh
“•<4329581
S T2> 10 5 <C 2 TL,
University of Ghana          http://ugspace.ug.edu.gh
DECLARATION
I do hereby declare that, except for references 
to works of other researchers which have been duly 
cited, this work is the result of my own original 
research and that this thesis either in whole or in 
part has not been presented for another degree 
elsewhere.
S.O. BENNETT-LARTEY 
(STUDENT)
DR. E.T. BLAY 
(SUPERVISOR)
University of Ghana          http://ugspace.ug.edu.gh
ACKNOWLEDGEMENTS 4  + y
I would like to thank Professor E.V. Doku for encouraging me 
to undertake this work, and Dr. E.T. Blay for her guidance. I 
owe a debt of gratitude to Professor E. Laing, of the Botany 
Department of the University of Ghana (UG), Dr. I. Ofori, of the 
Crop Science Department, (UG) and Dr. Osei Bonsu of the Cocoa 
Research Institute of Ghana (CRIG) for critically reading through 
the manuscript and making very useful suggestions at various stages 
of the work. I am also grateful to Mr. Joseph Adu and other 
staff of the Plant Genetic Resources Unit (PGRU) of the Crops 
Research Institute (CRI) at Bunso for their help in the field and 
laboratory work. I also thank Miss Agnes Asiedua Ankomah of CRI, 
Kumasi, for her help in the analysis of the results, Mr. Sammy 
Ofosu-Ampofo for taking the photograph of the cowpea accessions, 
Mr. A. A. -Brown for reproducing the figures appearing in the 
work, Mr. Emmanuel Ofori and Mr. Samuel A. Yeboah of CRI for 
typing the manuscript and Mrs Mercy Bennett-Lartey for her 
encouragement.
Finally, I would like to thank Mr. E.A . Addison, the 
Director of CRI for giving me permission to pursue this course 
and Dr. Quat Ng, the head of the Genetic Resources Unit (GRU), of 
the International Institute of Tropical Agriculture (IITA) for 
providing the funds for the collection of the cowpea germplasni 
used in the work.
- i i i -
University of Ghana          http://ugspace.ug.edu.gh
C O N T E N T S
Page
TITLE PAGE 
DECLARATION 
ACKNOWLEDGEMENTS 
CONTENTS 
LIST OF TABLES 
LIST OF FIGURES
ABSTRACT   x
1. INTRODUCTION   1
2. LITERATURE REVIEW ........................  3
2.1 Origin and distribution of cowpea
2.2 Economic importance of cowpea
2.3 Cowpea cultivation in Ghana ... 4
2.4 Constraints to cowpea production 5
2.5 Cowpea collecting explorations in Ghana 6
2.6 Preliminary characterization and evaluation 7
2.7 Further characterization and evaluation 9
2.8 Characterization and evaluation of cowpea 9
2.9 Variability in cowpea collections 10
2.10 Cowpea improvement   11
2.11 Heterosis and crop improvement ... 13
3. MATERIALS AND METHODS ................... 17
3.1 Exploration and collection of cowpea g'ermplasm 17
3.2 Seed multiplication, characterization and
preliminary evaluation ............. 17
3.2.1 Characterization of vegetative characters 20
University of Ghana          http://ugspace.ug.edu.gh
CONTENTS (CONTD.)
3.2.2 Characterization of inflorescence and
fruit characters ............. 21
3.3 Further characterization and evaluation 22
3.3.1 Vegetative characters ... 23
3.3.2 Inflorescence, fruit and seed characters 24
3.4 Heterosis studies .............  26
3.5 Statistical analysis  ....... 28
4. RESULTS .......................  29
4.1 Characterization and preliminary evaluation
data on vegetative characters 29
4.1.1 Growth h a b i t ............. 29
4.1.2 Twining tendency .......  29
4.1.3 Plant pigmentation .......  32
4.1.4 Other vegetative characters 32
4.2 Characterization and preliminary evaluation
data on flower and fruit characters 34
4.2.1 Flower colour   34
4.2.2 Raceme position .......  34
4.2.3 Pod attachment .......  34
4.2.4 Immature pod pigmentation 38
4.3 Further characterization and evaluation
data on vegetative characters 38
4.3.1 Hypocotyl length .......  38
4.3.2 Number of nodes on main stem 40
4.3.3 Number of branches per plant 40
4.3.4 Terminal leaflet length and width 42
Page
-  v -
University of Ghana          http://ugspace.ug.edu.gh
CONTENTS (CONTD.)
4.4 Further characterization and evaluation 
data on inflorescence, fruit and seed 
characters ............. ... 44
4.4.1 Days to flowering ... ... 44
4.4.2 Peduncle length ... ... 44
4.4.3 Peduncles per plant ... ... 45
4.4.4 Number of racemes per plant 47
4.4.5 Standard and calyx lobe length 48
4.4.6 Days to maturity ... ... 48
4.4.7 Pods per plant and pod length 48
4.4.8 Seeds per pod and 100-seed weight 51
4.4.9 Grain yield per plant ... 54
4.5 Heterosis estimates in selected crosses 55
5. DISCUSSION ......................... 58
5.1 Variability in qualitative characters in
cowpea germplasm . . . . . .  58
5.2 Variability in quantitative characters in
cowpea germplasm . . . . . .  62
5.3 Heterosis for maturity date, yield and
yield components . . . . . .  69
6. SUMMARY AND CONCLUSIONS ......... 7 3
7. REFERENCES .........................  7 6
APPENDIX .......................  86
Page
- v i -
University of Ghana          http://ugspace.ug.edu.gh
12
3
4
5
6
7
8
9
10
11
12
LIST OF TABLES
Some characteristics of parental 
accessions used in heterosis studies ...
Ranges, means and coefficients of variation 
for hypocotyl length (mm) of cowpea germplasm 
from four regions of G h a n a .......... ...
Ranges, means and coefficients of variation 
for number of nodes on the main stem of 
cowpea germplasm from four regions of Ghana
Ranges, means and coefficients of variation 
for number of branches per plant of cowpea 
germplasm from four regions of Ghana ...
Ranges, means and coefficients of variation 
for terminal leaflet length (cm) of cowpea 
germplasm from four regions of Ghana ...
Ranges, means and coefficients of variation 
for terminal leaflet width (cm) of cowpea 
germplasm from four regions of Ghana ...
Ranges, means and coefficients of variation 
for number of days to flowering of cowpea 
germplasm from four regions of Ghana ...
Ranges, means and coefficients of variation 
for peduncle length (cm) of cowpea germplasm 
from four regions of Ghana ... ...
Ranges, means and coefficients of variation 
for number of peduncles per plant of cowpea 
germplasm from four regions of Ghana . . .
Ranges, means and coefficients of variation 
for number of racemes per plant of cowpea 
germplasm from four regions of Ghana ...
Ranges, means and coefficients of variation 
for standard length (mm) of cowpea germplasm 
from four regions of Ghana .......  ...
Ranges, means and coefficients of variation 
for calyx lobe length (mm) of cowpea 
germplasm from four regions of Ghana ...
27
41
41
43 
4 3 
4 5 
46
46
47 
49 
49
Page
-  v  i i -
University of Ghana          http://ugspace.ug.edu.gh
LIST OF TABLES (CONTD. )
13 Ranges, means and coefficients of variation
for number of days to maturity of cowpea
germplasm from four regions of Ghana ... 50
14 Ranges, means and coefficients of variation 
for number of pods per plant of cowpea 
germplasm from four regions of Ghana ... 52
1 5 Ranges, means and coefficients of variation 
for pod length (cm) of cowpea germplasm 
from four regions of Ghana ............. 5 3
16 Ranges, means and coefficients of variation 
for number of seeds per pod of cowpea 
germplasm from four regions of Ghana ... 53
17 Ranges, means and coefficients of variation 
for 100-seed weight (g) of cowpea germplasm 
from four regions of Ghana ............. 54
18 Ranges, means and coefficients of variation 
for grain yield per plant (g) of cowpea 
germplasm from four regions of Ghana ... 56
19 Performance of parents and their FI 
hybrids, and midparent heterosis (%) in 
four crosses among some selected cowpea 
acce ss i o n s .......  ... ... ....... 5 7
University of Ghana          http://ugspace.ug.edu.gh
List of Figures
1
2
3
4
5
6
7
8
Figure
Cowpea germplasm collection sites in 
Ghana in 1987. ..................
Seed samples of cowpea germplasm
collected in four regions of Ghana in 1987
Growth habit of cowpea germplasm from
four regions of Ghana ............
Twining tendency of cowpea germplasm from 
four regions of Ghana ............
Plant pigmentation of cowpea germplasm 
from four regions of Ghana .......
Flower colour of cowpea germplasm from 
four regions of Ghana .......
Raceme position of cowpea germplasm from 
four regions of Ghana .......
Pod attachment of cowpea germplasm from 
four regions of Ghana ............
18
19
30
31 
33 
3 5
36
37
Page
9 Immature pod pigmentation of cowpea
germplasm from four regions of Ghana 39
University of Ghana          http://ugspace.ug.edu.gh
Abstract
Forty-five cowpea accessions, randomly selected from cowpea 
germplasm from four cowpea-growing regions of Ghana, namely, Upper 
East, Upper West, Northern and Eastern Regions, were characterized 
and evaluated at the Plant Genetic Resources Unit of the Crops 
Research Institute at Bunso with the objective to determine the 
range of variability in the samples with respect to vegetative, 
inflorescence and fruit characters. The variability in the samples 
was used to initiate hybridization among some selected accessions 
from the collection. Subsequently heterosis was determined for 
flowering and maturity dates and yield and yield components in four 
crosses among the selected accessions with the view to improving 
the yield.
Large variability was observed in most of the qualitative 
characters, particularly growth habit, twining tendency, plant and 
pod pigmentation, raceme position, pod attachment to peduncle and 
flower colour of the germplasm. The accessions differed
significantly in their days to flowering and maturity, peduncle 
length, number of peduncles per plant, number of pods per plant, 
pod length, number of seeds per plant, 100-se.ed weight and grain 
yield per plant. There were also significant between- aod within- 
region differences I'n these characters.
t Heterosis for pods per plant .and grain yield per plant in all 
the four crosses was positive and high. This was
particularly so for grain yield per plant for the cross between 
accessions 87/27 and 87/157 in which heterosis was about 130%.
University of Ghana          http://ugspace.ug.edu.gh
The possibility of using accessions 87/27 and 87/157 for future 
improvement programmes of the crop is discussed based on their 
performance in this work.
University of Ghana          http://ugspace.ug.edu.gh
Cowpea is one of the most important legumes in Ghana.
It has been assigned first priority for germplasm collection and 
conservation in West Africa by the International Board for Plant 
Genetic Resources (IBPGR, 1981) since West Africa is regarded as 
centre of domestication and diversity of the crop (Steele, 1976; 
Steele and Mehra, 1980).
Cowpea is a widespread and popular food legume in Ghana and 
it is imperative that the landraces of the crop be collected and 
conserved- Landraces of cowpea and their wild relatives are well 
adapted to their local environment and it may be necessary to use 
these adaptive traits in breeding programmes. These traits 
include resistance to pests and diseases, adaptation to poor 
soils and tolerance to drought and heat stresses.
In recent years, cowpea improvement has been intensified by 
research institutions with the result that improved varieties are 
being turned out quite rapidly. These improved varieties may 
tend to replace the well adapted local varieties. Furthermore 
the frequent bushfires in the cowpea-growing areas may destroy 
several cowpea varieties together with their wild relatives. 
Genetic erosion in the crop may also be caused by development 
activities like the construction of dams, roads and townships 
(Chang £i al-, 1972).
University of Ghana          http://ugspace.ug.edu.gh
_  —
The collection and conservation of germplasm are important 
to safeguard against the loss of the species due to genetic 
erosion and for use in cowpea improvement programmes. Collecting 
a large diversity of a crop provides a broad genetic base to 
satisfy the needs of future breeding programmes.
When germplasm has been assembled, it is necessary that it 
is characterized and evaluated using appropriate descriptors. In 
this respect, both qualitative and quantitative characters are 
evaluated. Such data are documented and supplied to plant
breeders together with the germplasm when requests for breeding 
materials are made.
Though several cowpea germplasm expeditions have been
carried out in Ghana, the collection is by no means complete. 
More expeditions are required to fill in gaps still existing in 
the collections. There is the need to ensure that majority of 
existing landraces in Ghana are present in the germplasm 
assembled at the genetic resources centres.
The objective of -this study was to assemble, characterize 
and evaluate the genetic diversity in a collection of cowpea 
genotypes from four cowpea-growing areas in Ghana and 
furthermore, to initiate hybridization among some selected
accessions from the collection and to estimate some genetic
parameters in the hybrid populations. The magnitudes of these 
genetic parameters will indicate the potential for making 
improvements in the populations.
University of Ghana          http://ugspace.ug.edu.gh
2. LITERATURE REVIEW
2.1 Origin and distribution of cowpea
Cowpea, Vigna unguiculata (L) Walp., is an important pulse 
crop in tropical Africa. It is reported to have originated from 
West Africa (Faris, 1963; Rachie and Rawal, 1976; Smithson, Redden 
and Rawal, 1980) and most probably in Nigeria, where wild and weedy 
species abound (Rawal, 1973; Rachie and Rawal, 1976; Steele, 1976).
Most of the world’s production is in Africa and major 
producing countries are Nigeria, Burkina Faso, Niger, Ghana, Kenya 
and Uganda (Dovlo e_t ajL . , 1976; Rachie and Rawal, 1976; Steele and 
Mehra, 1980; Rachie, 1985). Other producing countries outside 
Africa include, United States of America (USA), (Brazil, India, Sri 
Lanka, Burma and Bangladesh (Rachie, 1985). The crop is also 
cultivated in the Mediterranean region and the Carribean islands 
(IITA, 1983).
2 . 2 Economic importance of cowpea
Dry cowpea seeds have protein content rang ing from 18 to 29% 
(IITA, 1983). Hence, it is a cheap source of protein in the diet. 
An increase in the production of cowpea will therefore help to 
reduce the incidence of protein malnutrition in children, 
especially in the rural areas of developing countries.
-  3 -
University of Ghana          http://ugspace.ug.edu.gh
- 4 -
Cowpea is consumed in many different ways. The young
shoots, leaves and even roots are used as pot herb in most parts 
of Africa (IITA, 1982). The tender young leaves are eaten in the 
northern parts of Ghana (Asafo-Adjei, 1986) and elsewhere in 
Africa as vegetables (Dovlo , 1976; Grubben, 1977). In the
USA the canning and freezing of immature pods for food is very 
important (IITA, 1982; Purseglove, 1968). Cowpea may also be 
used as a fodder for livestock and as green manure (Dovlo ei &1. ,
1976).
Cowpea is most often eaten with cereals to enhance the 
protein value of the latter which are sometimes deficient in 
sulphur-containing amino acids (Dovlo grt &1-, 1976). The bulk of 
the seed is cooked and eaten in West Africa (Dovlo, 1976).
The cultivation of cowpea in Ghana is widespread. It is 
grown mainly in the savannah and the transitional zones. There 
is some amonunt of cultivation in the forest zone too. Indeed, 
it can be grown throughout the country (GGDP, 1990).
Cowpea cultivation in Ghana has been done mainly by peasant 
farmers who intercrop it with cassava, sorghum, millet and yams. 
The peasant farmers usually use local varieties which vary widely 
in various characters. Commercial cowpea farming involving large 
areas of sole cowpea is on a rapid increase in Ghana.
University of Ghana          http://ugspace.ug.edu.gh
- 5 -
2.4 Constraints to cowpea producti on
Several factors militate against the production of cowpea in
Ghana. Both biotic and abiotic factors contribute to this (IITA,
1982). Some of these factors are discussed below.
i. Planting materials - the planting materials used by-
peasant farmers are mostly locally adapted landraces which are
generally low-yielding (Rachie and Rawal, 1976; IITA, 1983).
ii. Climate - this is characterized by unfavourable
conditions like insufficient, poorly distributed or excessive 
rainfall, inadeqiiate insolation and extreme temperatures (Rachie, 
1985).
iii. Soils - which are usually poor in physical and
chemical properties and have low fertility due to deficiency in 
organic matter and often unfavourable microbiological conditions 
(Rachie, 1985).
iv. Pests - cowpea production is greatly hampered by 
several pests which attack various parts of the plant especially, 
thrips which attack flowers, pod-sucking pests like Haruca. 
storage bruchid and Striea. the plant parasite of the crop (Singh 
and -Jackal, 1985).
v. Djjseass -- several disease-causing organisms like virus, 
fungi, nematodes and bacteria cause a drastic reduction in the 
yield of cowpea (Ernechebe and Shoyinka, 1985).
University of Ghana          http://ugspace.ug.edu.gh
6vi. the storage weevil,
marnl ftt.\is causes high rates of post-harvest losses which begin in 
the field and continue in the shelled seeds in storage (Singh and 
Jackai, 1985).
which arise from the common practice of growing the crop in 
association with other crops, sub-optimal planting dates and 
little or no insecticidal and weed control methods lead to 
reduction in yield (Rachie, 1985).
2.5 Cowpea collecting exploration in Ghana
In Ghana the set-i?p mandated to engage in plant genetic 
resources activities is the Plant Genetic Resources Unit (PGRU) 
of the Crops Research Institute (CRI) based at Bunso in the 
Eastern Region. Systematic collection of cowpea germplasm in 
Ghana started in i976 (Adansi and Holloway, 1980; Bennett-Lartey, 
1988). The first collection of cowpea germplasm in Ghana was 
carried out in 1976 by a team of experts from the Genetic 
Resources Unit (GRU) of the International Institute of Tropical 
Agriculture fllTA) (GRU, 1976). The number of cultivated cowpea
v i i . Inadequate management practic.eig low plant densitie
Laiii) and wild cowpea species collected were 111
and 32 respectively fGRU, 1976). The wild V iena spec
collected included \Z. Y- racemc-sa . V . pubigera and V
University of Ghana          http://ugspace.ug.edu.gh
In 1977 a team from the PGRU, Bunso followed up the 1976 
collection. The team collected 48 accessions of cowpea from 
areas the 1976 team could not visit (Bennett-Lartey, 1986). In 
1978 a team from the GRU of IITA and the Faculty of Agriculture 
of the University of Ghana, Legon also collected 38 accessions 
from the Northern, Upper and Brong Ahafo Regions of Ghana (Adansi 
and Holloway, 1978; GRU, 1978; Perez, 1978).
Between October, 1982 and February, 1983, there was a multicrop 
germplasm collecting mission in Ghana by the PGRU. In this 
expedition 61 accessions of cowpea were collected from all 
regions of Ghana except the Greater Accra and Western Regions 
(Bennett-Lartey, 1983; Holloway, 1983). Between October and 
December 1987, 71 accessions of cultivated cowpea were collected 
in four cowpea-growing regions of the country, namely, Northern, 
Upper East, Upper West and Eastern Regions by PGRU.
One hundred and nineteen accessions of wild cowpea species were 
also collected in all regions of Ghana except the Greater Accra, 
Eastern and Western Regions between 1987-89 (Bennett-Lartey, 
1990; 1991).
Though there is a great diversity in the cowpea found in the 
savanna zones of Ghana, the diversity is not up to what is found 
in the savanna zones of Nigeria (GRU, 1976).
For collected plant genetic resources to be useful certain 
basic information has to be available on them. The provision of
-  7 -
University of Ghana          http://ugspace.ug.edu.gh
this basic information is considered the responsibility of the 
curator of the gene-bank. This work includes the basic 
morphological description of the accessions. This is usually 
done at the time of multiplication of the seeds or other planting 
materials using standardized crop descriptors (Howes, 1981).
Preliminary characterisation consists of recording 
characters which are highly heritable and which can be easily 
scored visually and will be exposed in most environments (Howes, 
1981; IBPGR, 1983). Some of these data can be collected in the 
field with the collection, otherwise they can be collected during 
multiplication of the germplasm.
Characterization data are useful because they can be used 
for identification purposes and they are a good guide for the 
future use of the material. Since such characters are heritable 
they will be reliable and constant. Examples of preliminary 
characterization data are flower colour, fruit size and fruit 
colour.
Preliminary evaluation consists of recording a limited 
number of additional agronomic traits thought desirable by a 
consensus of users of the particular crop (Howes, 1981). These 
characters are capable of visual assessment but not in all 
environment. Some of these descriptors may be recorded at the 
same time as characterisation. Examples of such data are 
reaction to diseases and pests, and resistance or tolerance to 
environmental stresses.
8
University of Ghana          http://ugspace.ug.edu.gh
2 . 7 Further characterization and evaluation
Beyond the basic descriptive work carried out on a 
collection are a range of other characters of interest to
breeders (Chapman, 1989). These include stress tolerance, 
disease and pest resistances, and quality characters. Evaluation 
for many of these traits is outside the ability or resources of
most curators. It is the responsibility of the users of a
collection and may require specialized laboratory or greenhouse 
work and the assistance of an expert familiar with the specific 
character and procedure for testing (Howes, 1981; Chapman, 1989).
2. 8 Characterization and evaluation of cowpea
Standardized characterization and evaluation data should be 
readily available on plant genetic resources. In this respect 
the International Board for Plant Genetic Resources (IBPGR) has 
produced standard descriptors for cowpea (IBPGR, 1983). The 
descriptors contain information necessary for recording passport, 
characterisation and preliminary evaluation data for cowpea 
germplasm.
Several workers have worked on the characterization and 
evaluation of cowpea. Faris (1963) collected both cultivated and 
wild cowpea from all over the world and he characterized them on 
the basis of morphological traits. Doku (1970) characterized 7 
local and 32 introduced cowpea cultivars on morphological 
characters including 100-seed weight, pod length, number of pods 
per plant, time from germination to maturity, weight of nodules
9
University of Ghana          http://ugspace.ug.edu.gh
per plant, seed yield per plant, number of seeds per pod and time 
from germination to flowering. Scientists at the Genetic 
Resources Unit of the International Institute of Tropical 
Agriculture have characterized the cowpea germplasm collected 
from all over the world and have produced a catalog of 
information on cowpea (Porter, 1973; Porter fii £l1-, 1974). A
similar catalog has been produced from the data on Botswana 
cowpea germplasm (deMooy, 1984). Cowpea germplasm collected in 
Ghana in 19S2 and 1983 have also been characterised and evaluated 
(Bennett-Lartey, 1984).
2.9 Variability in cowpea collections
Abifarin (1984) defines variability as the state or quality 
of being variable. In plant breeding, variability refers to the 
presence of genotypically different individuals.
Cultivated species are usually variable and cowpea is no 
exception (Ng and Marechal, 1985). Ng and Marechal (1985) 
attributed the variability in cultivated species to artificial 
selection under diverse environments. As a result of
characterization and evaluation carried out on cowpea germplasm, 
wide variability has been observed in several characters of the 
crop.
Dovlo (1976) reported the presence of many different grain 
colour in cowpea found in Ghanaian markets. Working on a world 
collection of cowpea at IITA, Porter _e£ a] . (1974) observed
variability in several characters including growth habit, twining
10
University of Ghana          http://ugspace.ug.edu.gh
tendency, days to flowering and maturity, eye colour and pattern 
and susceptibility to insect pests and diseases.
deMooy (1984; 1985) also found variability in several
characters in 108 accessions of Botswana cowpea germplasm. These 
characters included terminal leaflet length and width, peduncle
length, pod length, number of pods per plant and number of seeds
per pod. Amoatey (1987) working with cowpea accessions from Ghana 
and Nigeria found variability in seed size, seed colour, pod 
length, number of seeds per pod, 100-seed weight and grain yield. 
Doku (1970) observed variability in weight of nodules per plant in 
39 Ghanaian and exotic cowpea varieties he studied.
2.10 Cowpea improvement
Evaluated cowpea germplasm has proved very useful in crop
improvement programmes against known pests and diseases. The
cowpea B301 from Botswana was found to be resistant to the plant 
parasite Alectra vogelii in evaluations in Botswana. It was 
subsequently used in Striga trials in Burkina Faso, Mali, Niger and 
Nigeria and was found to be resistant. Similarly B301 which is 
a smooth-seeded variety of about 70 days maturity is being used in 
breeding programmes in Nigeria for its drought resistance 
characteristics (IITA, 1988a).
Several cowpea lines including TV 3236 with adaptability and 
high yields have been used to breed improved varieties like IT 845- 
2246-4 which has resistance against aphids, thrips and the storage 
weevil (IITA, 1988b). An IITA cowpea germplasm TVu 2027
11
University of Ghana          http://ugspace.ug.edu.gh
is known to be moderately resistant to the bruchid storage insect 
(Callosobruchus maculatus) (Singh and Jackai, 1985) and this was
the only accession out of 10,000 found to have resistance to this 
pest (Jackai and Singh, 1991). This accession has been the basis 
of the bruchid resistance in all improved cowpea varieties from 
IITA (Singh and Jackai, 1985). Research at the University of 
Durham revealed a biochemical factor which made cowpea meal toxic 
to the bruchid larvae and thereby conferring resistance in TVu 2027 
(Baker, 1978).
Desirable traits in cowpea germplasm have been used in the 
improvement of other crops by the use of genetic engineering 
techniques. A gene from the IITA-bred cowpea TVu 2027 which 
inhibits insect trypsin has been transferred into a tobacco plant 
by British plant scientists at the Plant Breeding Institute, 
Cambridge, U.K. (Anon, 1988). The gene encodes a protein that is 
a natural inhibitor of insect trypsin which helps the tobacco 
plant to resist insect pests.
Wild cowpea germplasm has also been found to possess high 
levels of resistance to insect pests. Sources of high levels of 
resistance to cowpea coreid bug, Clavigralla tomentosicollis and 
cowpea pod borer, Maruca testulalis ha' been identified from the 
IITA germplasm collections of Vigna vexillata (TVNu 72 and TVNu 73) 
(IITA, 1988b; Jack.i i and Singh, 1991). Though so far no successful 
cross has been possible between V. vex i1lata and V. unguiculata, 
there is the possibility that the barriers in the
12
University of Ghana          http://ugspace.ug.edu.gh
hybridization of wild and cultivated cowpea may be overcome using 
in-vitro methods (IITA, 1988b).
Lately, the major thrust within the cowpea research
programme has been to develop genotypes with high and stable
potential and favourable characteristics such as resistance to 
some diseases and insects (Ng, 1982), early (60 days) and medium 
(75-80 days) maturity, grain, vegetable and fodder types; and 
different seed sizes, coat texture and colours for various 
consumer preferences (IITA, 1988b).
In Ghana cowpea improvement has been intensified since 1979 
with the inception of the Ghana Grains Development Project (GGDP) 
(Asafo-Adjei, 1986). This increased research activity in co­
operation with IITA and the Canadian International Development
Agency (CIDA) has led to the release of several improved
varieties of cowpea (Wobil, 1986). Examples of the improved and 
high yielding varieties of cowpea recently released in Ghana are 
Soronko, Amantin, Asontem and Vallenga with medium (75-80 days) 
to very early (60-65 days) maturity (GGDP. 1990).
Heterosis is defined as the increased development or 
expression of one or more characters of hybrid plants above that 
of the parents (Abifarin, 1984). The characters could be vigour, 
growth, si.:je, yield, earliness and improved general fitness 
characteristics such as resistance to disease, lodging and 
drought. The character so expressed is usually more than the
13
University of Ghana          http://ugspace.ug.edu.gh
average of the two parents. Heterosis is also termed hybrid 
vigour.
Several theories have been put forward to explain the 
genetic basis of heterosis. According to Briggs and Knowles 
(1967), there are the over-dominance, dominance and epistasis 
theories. The overdominance or heterozygosity theory has resulted 
from the fact that theoretically heterozygotes for a single gene 
difference are superior to either of the two homozygotes.
This superiority is attributed to the metabolic advantage of 
the gene product produced by two different kinds of alleles rather 
than only one kind (Strickberger, 1985). In the dominance theory,
dominant alleles contribute equally to heterosis and the recessive 
alleles do not and therefore dominance is assumed to be complete 
and expressed in the hybrid. It has been postulated that 
epistasis, which includes complementary, inhibiting and duplicate 
gene interactions, may be a cause of heterosis even though no 
direct evidence to that effect has been cited (Briggs and Knowles, 
1967). According to Burton (1968), heterosis is
thought to have resulted from the combined action and interactions 
of allelic and non-allelic factors and is usually closely and 
positively correlated with heterozygosity.
Heterosis is usually measured with reference to either the 
midparent value or the higher parent value (Laosuwan and Atkins,
1977). Heterosis in which an expressed character of a hybrid is 
beyond the range of either parent is termed heterobe 11iosis
14
University of Ghana          http://ugspace.ug.edu.gh
(Abifarin, 1984). Both heterosis and heterobeltiosis can be 
computed from the following equations.
Heterosis (%) =( Fj - MP) / MP x 100
Heterobeltiosis (%) =( Fj - HP) / HP x 100
where F, , MP and HP are the means for the F, , midparent and higher 
parent respectively (Laosuwan and Atkins, 1977).
In practice the problems a hybrid breeder would have to 
resolve have been enumerated by Simmonds (1979) as follows:
1) isolating numerous inbred lines;
2) testing combinations of inbreds to identify excellent 
ones ;
3) devising short-cut methods of testing inbreds without 
resort to crossing everything by everything and
4 ) developing methods of making hybrid seed on 
commercial scale.
Weber et ad ■ (1970) also argued that for a highly self-
fertilizing species two requirements should be satisfied for
successful commercial production of F, hybrids. First, there1
must be heterosis for seed yield and, second an economical large- 
scale method of making hybrids must be found. Kheradnam ejt al ■ 
(1 975) concluded that the production of hybrid seed for cowx^ea 
could be commercially advantageous, if large-scale methods of
15
University of Ghana          http://ugspace.ug.edu.gh
emasculation and crossing were developed for this species. The 
discovery or development of male sterility would facilitate 
hybrid seed production in cowpea (Rachie and Rawal, 1976).
Several workers have observed heterosis for yield and yield 
components in such crops as cowpea (Kheradnam at al-, 1975), 
soybean (Paschal and Wilcox, 1975), safflower (Yazdi-Samadi 
.aJ*. , 1975), potato (Tarn and Tai, 1977) and Indian mustard
(Singh, 1973) .
16
University of Ghana          http://ugspace.ug.edu.gh
3 . MATERIALS AND METHODS
3.1 Exploration and collection of cowpea germplasm
A cowpea germplasm collecting mission was carried out in 
four cowpea-growing regions of Ghana, namely Northern, Upper East, 
Upper West and Eastern Regions. Collection started in October, 
1987 and ended in December, 1987. The collection sites of the 
germplasm are indicated in Fig.l. Collections were made in towns 
and villages (Appendix), from farmers’ fields, home gardens, farm 
stores and markets. All collections were landraces. Materials 
collected were pods and seeds. Seed samples of the collected 
accessions are shown in Fig.2. Passport data were collected with 
the germplasm (Appendix). A total of 71 accessions of cowpea
germplasm were collected in the four regions.
3.2 Seed multiplication, characterization and preliminary 
evaluation
Forty-five accessions of the cowpea germplasm were randomly 
selected for this study with representative accessions from each 
reg ion.
Seeds from the 45 accessions were sown in the field for seed 
multiplication, characterization and preliminary evaluation. The 
seeds were sown at a spacing of 90cm x 90cm using three seeds per 
hill on plots 4.5m by 1.8m.. The seedlings were thinned to one per 
stand one week after seedling emergence. There were two rows per
- 17 -
University of Ghana          http://ugspace.ug.edu.gh
FIG . 1 . COWPEA GERMPLASM COLLECTION SITES ( SHADED  REGIONS  )
IN GHANA IN 1987 .
University of Ghana          http://ugspace.ug.edu.gh
Im
87 /25
t t
87 H I
*
87 / 30 8 7 /3 2
u
8 7 /3  4 8 7 /3 5
« l
_B7 /_37____ * !87 /40 B 7 / 4 1 _ ^
*
_ £ Z ' « _
r r
87 /62
»
87/_49 8 7 /5 2
»
87 /55
»
87/56
«
87 / 58
t S !
1 87 /59
2 :
87/60
f t
87/61
a |
8 7 /6 4 87 /6  7
9
87 /73
«
87/83 8 7 / ^ ^
f t
87 /90
8
87 m
*
81 /  96
*
87/103
< i
87/106
n
87/132
S f e
w
87/133 □
#
87/135
I I
87 /136
&
87/137
8
87 /138
B  « J t
87/142 i 87/147 TOT7150
I I
87 /153 87 /156
*
87/157 87/163
Fig. 2 Seed samples of cowpea germplasm collected in 
four regions of Ghana in 1987.
University of Ghana          http://ugspace.ug.edu.gh
five plants per row. There was no replication because of the 
small number of seed samples.
Characterization and preliminary evaluation data collected 
at this stage were on vegetative, inflorescence and fruit 
characters based on the standard descriptors for cowpea (IBPGR,
1983). Mature pods were collected from all accessions for 
further studies.
3.2.1 _Ghay5.ct.er.izatiQn of yegetativ.s_G];mx:acters
i. Growth habit: - The accessions were classified
at six weeks after planting 
(WAP) into erect, semi-erect, 
intermediate, semi-prostrate 
or prostrate.
ii. Growth pattern: - The plants were classified as
determinate or indeterminate at 
six WAP.
iii. Twining tendency: - Classification was based on the
intensity of twining. It could 
be absent, slight or 
intermediate.
iv. Plant pigmentation: - Pigmentation of the vegetative
parts was recorded at six WAP. 
The options were no 
pigmentation (green), very 
slight, moderate or extensive 
pigmentat ion.
20
University of Ghana          http://ugspace.ug.edu.gh
21
v. Plant hairiness: The absence or presence of
hairs on stems, leaves and pods 
were recorded as glabrescent or 
hirsute respectively.
relation to the canopy was 
recorded when the peduncles 
were at full length. Racemes 
were either held above the 
canopy, in the upper canopy or 
throughout the canopy.
mature green stage. The
options were erect, 30-90 
degrees inclination or 
pendant.
3.2.2 Characterization of
fruit characters
i. Raceme position: The position of the flowers in
ii. Pod attachment
to penducle: This was evaluated at the
University of Ghana          http://ugspace.ug.edu.gh
22
iii. Immature pod 
pigmentat ion Pattern of pigmentation on 
green pods was recorded. The 
options were none, pigmented 
tip, pigmented sutures, 
pigmented tips and sutures, 
splashes of pigment and 
uniformly pigmented.
iv. Flower colour: Main flower colours 
identified were purple and 
white based on the Methuen 
Handbook of colour (Methuen, 
1981 ) .
3.3 Further characterization and evaluation
Harvested seeds from each of the 45 accessions were sown in 
the field for further characterization and evaluation in May, 1988. 
Seeds of each accession were planted at three seeds per hill in a 
randomized complete block design with three
replications. The seedlings were thinned to one per hill at one 
week after seedling emergence leaving ten plants per plot for each 
accession. Both vegetative and reproductive characters were 
stud ied.
University of Ghana          http://ugspace.ug.edu.gh
23
3.3.1 Vegetative characters
i. Hypocotyl length:
ii. Terminal leaflet 
length and width:
iii. Number of branches
iv. Number of nodes 
on main stem:
The hypocotyls of five 
seedlings per plot were 
randomly measured 14 days 
after planting (DAP) and the 
mean of the five taken.
The widths at the widest 
points and the lengths of the 
terminal leaflets of ten 
leaves per plant for five 
plants per plot were randomly 
measured at six WAP and the 
mean taken.
The mean number of branches 
per plant from a random 
sample of five plants per 
plot was recorded six WAP.
The mean number of nodes per 
plant from five randomly 
selected plants per plot was 
recorded six WAP.
University of Ghana          http://ugspace.ug.edu.gh
24
3.3.2 Inflorescence, fruit 
and seed characters 
i. Days to flowering:
ii. Standard length:
iii. Calyx lobe length:
Days to maturity:
Number of days from sowing to 
when 50% of the plants had 
flowered was recorded.
The mean of the lengths of 30 
freshly opened, randomly 
selected standard petals from 
five plants per plot was 
measured.
The mean of the lengths of 30 
randomly selected calyx lobes 
from five plants per plot was 
measured.
The number of days from
sowing to the stage when 50% 
of the plants had mature pods 
was recorded.
v. Number of racemes
per plant: - The mean number of racemes
from five randomly selected 
plants per plot was recorded.
University of Ghana          http://ugspace.ug.edu.gh
vi. Peduncle length:
vii. Number of peduncles 
per plant:
viii. Number of pods 
per plant:
ix. Pod length:
x. Number of seeds 
per pod:
xi. 100-seed weight:
The mean length of penducles 
from five randomly selected 
plants per plot was measured.
The mean number of peduncles 
per plant from a random 
sample of five plants per 
plot was recorded.
The mean number of mature 
pods from five randomly 
selected plants per plot 
was recorded.
The mean length of pods on 
five randomly selected 
plants per plot was measured.
The mean number of seeds from 
pods harvested from five 
randomly selected plants per 
plot was recorded.
The weight of a sample of 100 
seeds at 12% moisture content 
was measured.
25
University of Ghana          http://ugspace.ug.edu.gh
26
xii. Grain yield per
plant: - The mean grain yield from five
randomly selected plants per 
plot was recorded.
The same five randomly selected plants per plot were used 
for the above characters, besides the flowering and maturity 
dates.
3.4 Heterosis studies
Five accessions of cowpea possessing contrasting flowering
and maturity dates, number of pods per plant, p^ od length and
grain yield per plant, were selected, based on the results of 
previous studies, for heterosis. The objectives of the studies 
were:
1) to produce plants with combined early maturity and large 
number of seeds.
2) to combine large number of seeds per pod and large seeds,
3) to produce plants which have large number of seeds per plant 
combined with large seeds and
4) to produce plants with high seed yield.
The characteristics of the parental accessions are shown in 
Table 1. The selected accessions were planted in April, 1990 in 
single rows m  the field at different times to synchronise
flowering periods.
University of Ghana          http://ugspace.ug.edu.gh
The crosses made were as follows:
i. 87/56 x 87/94
li. 87/35 x 87/157
ill. 87/27 x 87/35
iv. 87/27 x 87/157
Table 1. Some characteristics of cowpea parental accessions used 
in heterosis studies.
27
Cowpea Days Racemes Days Pod Pods Seeds 100-seed Grain
acce- to per to len- per per wt.(g) yield
ssions flo- plant matu- gth plant pod per
wer- rity (cm) plant
(g)
87/27 40 56.0 62 17 . 1 32.0 16. 1 11.8 46.9
87/35 49 50.7 69 13. 2 4.7 8.1 15.2 17.9
87/56 39 63.7 57 13.7 28.3 11.3 9.6 29. 1
87/94 42 45.3 60 15. 1 23. 3 12. 3 10.7 28.2
87/157 47 49.7 66 15.6 44.7 15 . 6 9.3 36.9
Crosses and their reciprocals were made according to the 
method developed at IITA by Rachie, Rawal and Franckowiak (1975). 
Emasculation was done in the evening and crosses were made early 
in the morning. Mature pods were harvested as soon as they began 
to dry. Pods from identical crosses as well as those from 
parental plants were harvested separately.
The parents designated PI and P2 and their FI’s were planted 
in a randomised complete block design with three replications in
University of Ghana          http://ugspace.ug.edu.gh
August, 1990 for the heterosis studies. Five randomly selected 
plants per plot were used to evaluate the following yield 
components: days to flowering, days to maturity, pods per 
peduncle, pods per plant, pod length, seeds per pod, 100-seed 
weight and grain yield per plant. Backcrosses to both parents
(BC1 and BC2) were made from the remaining FI's and parents. 
Pods from each backcross were harvested separately and bulked. 
Heterosis was estimated by the formula of El-Hosary and Nawar 
(1984).
Heterosis (%) = (.Fi - MP)/ (MP) x 100 
where Fi and MP are the means of the FI and the midparent, 
respectively.
3.5 Statistical Analysis
Further characterisation and evaluation data recorded were 
subjected to analysis of variance. First the data on accessions 
from all regions were analysed as a whole. The data were further 
analysed on within- and between-region basis.
28
University of Ghana          http://ugspace.ug.edu.gh
-  29  -
4.1 Characterisation _ao£L£re.liminar-Y ..evaluation data on
Figure 3 shows the growth habit of cowpea germplasm from 
four regions of Ghana. The three main growth habits shown by the 
cowpea germplasm were intermediate, semi-prostrate and prostrate. 
Germplasm from the Upper East Region showed only the seini- 
prostrate habit whereas germplasm from the Upper West, Northern 
and Eastern Regions showed all the three types of growth habits. 
For each region the semi-prostrate type formed the majority of 
the accessions. This was followed by the prostrate and 
intermediate types, respectively.
4.1.2 Tv/in in e tendency
The twining tendency exhibited by the cowpea germplasm under 
study is shown in Fig. 4. Three categories of twining tendency 
were found; no twining, slight twining and intermediate twining. 
All accessions collected in the Upper East Region showed no 
twining. Accessions collected in the Upper West Region showed 
two categories of twining tendency; none and slight. Majority of 
the accessions showed slight twining. Most of the accessions 
from the Northern Region showed no twining. This was followed by 
slight and intermediate twining in that order. Accessions from
University of Ghana          http://ugspace.ug.edu.gh
NORTHERN REGION EASTERN REGION
Fig. 3 . Growth habit of Cowpea germplasm from four Regions
of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
FR
EQ
UE
NC
Y 
% 
FR
EQ
UE
NC
Y
100
UPPER EAST REGION UPPER WEST REGION
Fig. A . Twining tendency of cowpea germplasm from four Regions
of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
- 32 -
the Eastern Region showed a similar trend as those from the 
Northern Region. Generally majority of all accessions showed 
no twining and the least showed intermediate twining.
4.1.3 Plant Pigmentation
Figure 5 shows the frequencies of various plant pigmentation 
types in the germplasm collected. Five categories of plant 
pigmentation were observed. Accessions from the Upper East and 
Upper West Regions showed none, very slight or moderate 
pigmentation. In each of these regions, accessions with very 
slight pigmentation were in the majority. This was followed by 
moderate pigmentation and no pigmentation in that order. 
Accessions from the Northern Region were all pigmented and showed 
four categories of pigmentation, namely, very slight, moderate, 
intermediate and extensive. Similarly accessions from the 
Eastern Region were all pigmented with most accessions showing 
slight pigmentation. This was followed by moderate and
intermediate pigmentation in that order.
All the cowpea accessions studied showed indeterminate 
growth pattern and none, had hairs on stem, pods or leaves.
University of Ghana          http://ugspace.ug.edu.gh
FR
EQ
UE
NC
Y 
% 
FR
EQ
UE
NC
Y
1 0 0 r
^ 6 0 -
UPPER EAST REGION UPPER WEST REGION
100
80-
60-
40
20-
100r
NORTHERN REGION EASTERN REGION
F ig . 5 . Plant p igmentation of cowpea germplasm from four
Regions of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
34
4.2 Characterization and preliminary evaluation
data on flower and fruit characters
4.2.1
Figure 6 shows the distribution of flower colour of the 
cowpea germplasm . All the accessions from the Upper East Region
had white flowers. Most (81-88%) of the accessions from Upper
West, Northern and Eastern regions had purple flowers and the 
remainder of the accessions from these regions had white flowers.
The types of raceme positions exhibited by the cowpea
germplasm are shown in Fig. 7. Three types of raceme positions 
were observed; the racemes were mostly held above the canopy, 
they were held in the upper canopy or they were found throughout 
the canopy. The general trend shown by accessions from the four 
regions is that most of the racemes were mostly above the canopy. 
This was followed by the racemes being found in the upper canopy 
and then the racemes being throughout the canopy.
germplasm . Two conditions were observed: the condition in which 
the pod was inclined 30-90 degrees from erect and the condition 
in which the pod was pendant. Majority of accessions from each 
region had the pendant type of pod attachment.
OO
Figure 8 shows types of pod attachment in the cowpea
University of Ghana          http://ugspace.ug.edu.gh
FR
EQ
UE
NC
Y 
V. 
FR
EQ
UE
NC
Y
100r
80
60
40
20
100r
80
60
CJ
S  *0
CD 
Cf 
Uj 
Q; 
k. 20
NORTHERN REGION EASTERN REGION
Fig. 6 . Flower colour of cowpea germplasm  from  four
Regions of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
1 0 0 r 100
80
Sv
^  60
o
Lu
C2f
£  40
u.
20
KEY
■
Mostly about canopy 
In  upper canopy 
Throughout canopy
UPPER EAST REGION UPPER WEST REGION
100r 100r
NORTHERN REGION EASTERN REGION
Fig . 7 . Raceme position of cowpea germplasm from  four Regions
of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
FR
EQ
UE
NC
Y 
% 
FR
EQ
UE
NC
Y
UPPER EAST REGION UPPER WEST REGION
100r
NORTHERN REGION
o
5:
U j
OS
U j
or
u.
EASTERN REGION
Fig . 8 .  Pod a ttachm ent of cowpea germ plasm  from four
Regions of Ghana •
University of Ghana          http://ugspace.ug.edu.gh
4.2.4 Immature pod Pigmentation
Figure 9 shows the patterns of immature pod pigmentation 
observed in the cowpea germplasm- Six patterns of pigmentation 
were observed in the immature pods. About 17-42% of accessions 
from all regions showed no pigmentation in their immature pods. 
Immature pod pigmentation observed in accessions from the Upper- 
West Region was quite similar to that observed in accessions from 
the Upper East Region. Both did not have accessions with 
pigmented tips and those with pigmented tips and sutures. 
Accessions from the Northern Region had representation in ail six 
categories of pigmentation patterns.
38
4 .3.1
The hypocotyl lengths of the cowpea germplasm are shown in 
Table 2. The hypocotyl lengths of all the germplasm under study 
ranged from 14.3 to 27.3mm. Though there were no significant 
differences between the means of the hypocotyl length of the 
cowpea germplasm as a whole, and between—region, there were 
significant within-region differences in the hypocotyl lengths of 
the germplasm from the Northern and Eastern Regions but no 
significant differences in the hypocotyl lengths of those from 
the Upper East and Upper West Regions. The coefficient of
University of Ghana          http://ugspace.ug.edu.gh
FR
EQ
UE
NC
Y 
% 
FR
EQ
UE
NC
Y
100
80
60
o
U j
§  40
U j
Q:
U.
20
KEY
I
None
Pigmented t ips  
Pigmented sutures
Pigmented tips and sutures 
Splashes of pigment 
Uniform ly pigmented
UPPER EAST REGION UPPER WEST REGION
100
80
60
40
20
NORTHERN REGION 
Fig.  9 .
100r
80
0
1  40 
cy
U j
Cfc
U
20
0
EASTERN REGION
Im m ature pod pigmentation of cowpea germplasm from
four Regions of Ghana .
University of Ghana          http://ugspace.ug.edu.gh
variation for hypocotyl lengths of the germplasm from the Upper- 
East and Upper West Regions were high (>20%). This rnay be due to 
the fact that the number of accessions obtained from the two 
regions were low.
4.3.2 Number of nodes on main stem
The number of nodes on the main stem of cowpea germplasm
collected from four regions of Ghana is shown in Table 3. The
number of nodes on the main stem ranged from 11.3 to 13.7. There 
were large differences in the number of nodes on the main stem 
for the accessions from all four regions. However, differences 
both between and within regions were not significant.
4.3.3 Humber of branches per Plant
The number of branches per plant of the cowpea germplasm
collected from four regions of Ghana is shown in Table 4. The
number- of branches ranged between 7.3 and 10.0 with accessions 
from each region averaging about 8 to 9 branches per plant.
40
University of Ghana          http://ugspace.ug.edu.gh
Table 2. Ranges, means and coefficients of variation for
hypocotyl length (mm) of cowpea germplasm from four 
regions of Ghana.
- 41 -
Region No. of Range Mean1 Coef icient of
Accs . Variation (%)
Northern 16 14 . 3-24.0 19.8 14 . 7
Upper East 6 15 . 0-23.0 19.3 20 . 2
Upper West 8 17 . 7-25.0 20 . 7 22 . 2
Eastern 15 16.3-27. 3 21.0 15.3
Table 3. Ranges, means and coefficients of variation for the 
number of nodes on the main stem of cowpea germplasm 
from four regions of Ghana.
Reg ion No . of
Accs .
Range Meant Coefficient of
Variation ('
Northern 
Upper East 
Upper West 
Eastern
16
6
8
15
13.0-18.7
14.7-18.0 
11.3-17.0
12 .7- 17.0
15.7
15.8 
14 . 1 
14.7
17.9
12.9
25.8
22.9
+ Mean differences not significant (P>0.05)
University of Ghana          http://ugspace.ug.edu.gh
There were no significant between-region differences in the 
number of branches per plant. Similarly there were no
significant within-region differences in this character.
4.3.4 Terminal leaflet length and width
The terminal leaflet lengths of the cowpea germplasm are
shown in Table 5. The terminal length ranged from 6.8cm to 
12.8cm. Both the shortest and longest leaflet lengths were from 
the Northern Region collection. There were significant
differences among the accessions for the terminal leaflet length. 
There were also significant between-region differences in this 
character. There were significant within-region differences in 
the terminal leaflet length in the accessions from the Northern 
and Eastern Regions whereas there was none in those from the 
Upper East and Upper West Regions.
The terminal leaflet width ranged from 3.8cm to 8.6cm (Table 
6). Again the extremes are from the Northern Region collection. 
The 45 accessions showed significant differences with respect to 
their terminal leaflet width. There were however no significant 
differences between the regions in the terminal leaflet width of 
the cowpea germplasm. Significant differences were observed in 
the terminal leaflet width within each region except the Upper 
East Region.
- 42 -
University of Ghana          http://ugspace.ug.edu.gh
Table 4. Ranges, means and coefficients of variation for the
number of branches per plant of cowpea germplasm from 
four regions of Ghana.
-  4 3  -
Region No. of 
Accs .
Range Mean* Coefficient of 
Variation (%)
Northern 16 7.3-9 . 3 8 . 6 10 . 9
Upper East 6 7.7-9.3 8.4 12.2
Upper West 8 7.3-9-3 8.4 12.7
Eastern 1 5 7.6-10 . 0 8 . 7 9 . 7
+Mean differences not significant (P>0 .05)
Table 5 . Ranges, means and coefficients of variation for the 
terminal leaflet length (cm) of cowpea germplasm from 
four regions of Ghana.
Reg ion No. of 
Accs .
Range Mean' Coefficient of 
Variation (%)
Northern 16 6.8-12.8 8 . 6 10 . 9
Upper East 6 8.3-10.8 9 . 5 16.1
Upper West 8 7.6-9 . 8 8 . 7 14.1-
Eastern 1 5 7.8-11. 1 9 . 9 8 . 8
+Mean differences significant. LSD(5%) = 0.8
University of Ghana          http://ugspace.ug.edu.gh
4.4 Further characterization and evaluation data on
inflorescenceJ fruit and seed characters
4.4.1 Davs to flowering
The number of days to 50% flowering in the cowpea germplasm 
is shown in Table 7. The days to 50% flowering ranged from 31 
days in accessions from Northern and Upper West Regions to 72 
days in accessions from Upper East Region. There were
significant differences both between- and within-regions with 
respect to flowering date. Accessions from Northern and Upper 
West Regions averaged 38 days from sowing to flowering. The 
coefficients of variation were relatively small ranging from 3.5% 
to 7.2%.
The peduncle lengths of the cowpea germplasm collected from
four regions of Ghana are shown in Table 8. The peduncle lengths
ranged between 15.0cm and 41.3cm, indicating large differences in
the peduncle lengths of the germplasm. However there were no
significant differences between the regional means of peduncle 
lengths. There were significant within-region differences in the 
peduncle lengths of the germplasm from the Northern and Upper 
West Regions whilst there were small differences in those from 
the Upper East and Eastern Regions.
44
University of Ghana          http://ugspace.ug.edu.gh
- 45
4.4.3 Peduncles per plant
Table 9 shows the number of peduncles per plant of the 
cowpea germplasm studied. The number of peduncles per plant 
ranged between 10.0 and 14.3. There were significant differences 
between the accessions with respect to the number of peduncles 
per plant. However on regional basis, there were no significant 
differences in the number of peduncles per plant. Accessions 
from each region, with the exception of Upper East Region, showed 
no significant intra-regional differences.
Table S. Ranges, means and coefficients of variation for 
the terminal leaflet width (cm) of cowpea 
germplasm from four regions of Ghana.
Region Ho. of Range Mean+ Coefficient of
Accs. Variation (%)
Northern 16 3.8-8.6 5.1 14.0
Upper East 6 4.4-5.8 5. 2 16. 3
Upper West 8 4.4-6.8 5.7 11.2
Eastern 1 *r-1 4.2-7. 6 6.2 10. 5
+Mean differences not significant (P>0.05)
University of Ghana          http://ugspace.ug.edu.gh
the number of days to flowering of cowpea 
germplasm from four regions of Ghana
Region No. of Range Mean+ Coefficient of
Accs. variation (%)
Northern 16 31-51 38 3.5
Upper East 6 37-72 52 7.2
Upper West '7 31-54 38 6.6
Eastern 12 36-50 49 4.2
+Mean differences significant LSD (5%) = 5
Table S. Ranges, means and coefficients of variation for
peduncle length (cm) of cowpea germplasm from four
regions of Ghana.
Region No. of Range Meant Coefficient of
Accs. variation (%)
Northern 16 15.0-41.3 27 . 4 12.8
Upper East 4 28.1-34.1 30.6 18. 7
Upper West 7 15.8-35.2 29. 4 16.7
Eastern 12 27.5-38.9 32 . 4 14 . 6
+Mean differences not significant (P>0„05)
University of Ghana          http://ugspace.ug.edu.gh
Table 9. Ranges, means and coefficients of variation for 
the number of peduncles per plant of cowpea 
germplasm from four regions of Ghana.
- 47 -
Region No. of 
Accs.
Range Mean+ Coefficient of 
variation (%)
Northern 16 10.3-11.7 11.0 4.6
Upper East 4 10.0-12.3 11.3 2.1
Upper West 7 10.0-14.3 11.8 4.6
Eastern 12 10.3-12.0 11.3 5.1
+Mean differences not significant (P>0.05)
4.4.4 Number of racemes Per Plant
Table 10 shows the number of racemes per plant of the cowpea 
germplasm. Accessions from the Northern Region showed the
widest range of about 42 to 64 racemes per plant. There were
significant differences in the number of racemes produced by the 
different accessions. There were no between-region differences 
in the number of racemes per plant of the accessions. Similarly,
there were no significant differences between the accessions
within each region.
University of Ghana          http://ugspace.ug.edu.gh
4.4.5 Standard length and calvx lobe length
The standard lengths and calyx lobe lengths of cowpea 
germplasm are shown in Tables 11 and 12 respectively. No 
significant differences were observed in those two characters in 
the germplasm studied whether between-region or within-region.
4.4.6 Davs to maturity
The number of days to 50% maturity of cowpea germplasm is 
shown in Table 13. The earliest were accessions from Northern 
and Upper West Regions (57 days) and the latest were from Upper 
East and Upper West Regions (83 days). Significant differences 
were observed in the number of days to maturity of the cowpea 
germplasm.
There were also significant differences in the maturity 
period when the germplasm were considered on regional basis. 
With the exception of the Eastern Region, the accessions from the 
other regions showed significant differences within each region. 
The coefficients of variation were rather small being about 2% in 
the Upper 'West Region to about 6% in the Upper East Region.
- 48 -
The number of pods per plant of cowpea germplasm is shown in 
Table 14. The number of pods per plant ranged from 3.0 to 44.7. 
Significant differences were observed in the accessions as a 
whole.
University of Ghana          http://ugspace.ug.edu.gh
Table 10. Ranges, means and coefficients of variation for 
the number of racemes per plant of cowpea 
germplasm from four regions of Ghana.
-  49 -
Region No. of Range 
Accs.
Mean+ Coefficient of 
variation (%)
Northern 16 41.7-63.7 50.2 16. 6
Upper East 4 42.0-59-7 52.8 12. 1
Upper West 7 43.7-61.0 50. 1 18.2
Eastern 12 45.0-56.7 51.1 13.3
+Mean differences not significant (P>0.05).
Table 11. Ranges, means and coefficients of variation for
the standard length (mm) 
four regions of Ghana.
of cowpea germplasm from
Region No. of Range 
A ccs.
Mean+ Coefficient of 
variation (%)
Northern 16 15.0-20.0 18.0 10.2
Upper East 4 18.3-20.0 18.7 10.9
Upper West 7 16.7-20-0 18. 6 10.0
Eastern 12 16.7-20.0 17 . 2 9. 3
Hiean differences not significant (P>0.05)
University of Ghana          http://ugspace.ug.edu.gh
Table 12. Ranges, means and coefficients of variation for
- 50 -
the calyx 
from four
lobe length 
regions of
(mm) of 
Ghana.
cowpea germplasm
Region No. of 
Accs.
Range Mean+ Coefficient of 
variation (%)
Northern 16 3.7-5.0 4.6 9.0
Upper East 4 4.3-4.7 4.4 9.3
Upper West 7 4.3-5.0 4.5 9.6
Eastern 12 4.0-5.0 4.5 9.4
+Mean differences not significant (P>0.05)
Table 13. Ranges, means and coefficients of variation for
the number of days to maturity of cowpea
germplasm from four regions of Ghana.
Region No. of 
Accs.
Range Meant Coefficient of 
variation (%)
Northern 16 57-80 66 3.3
Upper East 4 68-83 76 5.7
Upper West 7 57-83 64 2.2
Eastern 12 64-72 68 4.9
+Mean differences significant LSD (5%) = 1.3
University of Ghana          http://ugspace.ug.edu.gh
and also between the regions with respect to the number of pods 
per plant. Within each region there were significant differences 
in the accessions except the Upper East Region. Upper East 
Region had the lowest range of 2.3 to 14.0 pods per plant. The 
coefficient of variability for each region was very high.
Pod length of the cowpea germplasm ranged from 8.3 to 17.5cm 
(Table 15). Significant differences were observed in the pod 
length of the germplasm. Similarly, there were significant 
between- and within-region differences in the germplasm.
4.4.8 Seeds per pod and. 100-seed weight
The number of seeds per pod of the coiv'pea germplasm is shown 
in Table 16. The number of seeds per pod of the cowpea germplasm 
ranged from 7.7 to 16.4, while the regional means ranged from 9.1 
to 11.5. There were significant differences in the number of 
seeds per pod of the accessions studied. Comparing the germplasm 
on regional basis, there were significant differences among the 
regions and also significant, differences in the accessions within 
each region.
The 100-seed weight of the cowpea germplasm ranged between 
S.7g and 21.4g (Table 17). There were significant differences 
between the accessions in 100-seed weight. There were also 
differences among the regions in 100-seed weight of the 
germplasm. With the exception of the Northern Region, there were 
significant differences in the accessions within each region with
- 51 -
University of Ghana          http://ugspace.ug.edu.gh
52 -
respect to 100-seed weight. Seed weight was fairly uniform 
within the regions except for the Upper East Region where it 
ranged from 100-seed weight of 7.7 to 21.4g. The coefficients of 
variation were relatively small.
Table 14. Ranges, means and coefficients of variation for 
the number of pods per plant, of cowpea germplasm
from four regions of Ghana.
Region N o . of 
Accs
Range Mean+ Coefficient of 
variation (%)
Northern 16 3.0-39.0 16.2 41.2
Upper East 4
0-st4tH1CO<N 6.3 58.1
Upper West 7 4.3-29.7 17 .0 41.8
Eastern 12 8.7-44.7 27 ? 29.9
+Mean differences significant. LSD (5%) = 4.1
University of Ghana          http://ugspace.ug.edu.gh
the pod length (cm) of cowpea germplasm from four
regions of Ghana.
Region No. of Range Mean+ Coefficient of
Accs. variation (%)
Northern 16 9.2-17.3 12.3 4.3
Upper East 4 9.6-15.2 12.7 4.8
Upper West 7 8.3-15.2 12.9 4.1
Eastern 12 12.4-17.5 14.3 5.5
+Mean differences significant LSD (5%) = 0.9
Table 16. Ranges, means and coefficients of variation for
the number of seeds per pod of cowpea germplasm
from four regions of Ghana.
Region Mo. of Range Mean+ Coefficient of
Accs. variation (%!
Northern 16 7.7-16.4 10.6 6.7
Upper East 4 8.0-10.3 9.1 6.2
Upper West 7 8.0-13.5 11.5 6.9
Eastern 12 8.7-15.6 10.9 8.0
-(•Mean differences significant. LSD (5%) = 0.8
University of Ghana          http://ugspace.ug.edu.gh
- 54 -
Table 17. Ranges, means and coefficients of variation for
100-seed weight (g) of cowpea germplasm from four 
regions of Ghana.
Region No. of 
Accs.
Range Meant Coefficient of 
variation (%)
Northern 16 8.7-11.9 10.0 3. 1
Upper East 4 7.7-21.4 14.9 2.6
Upper West 7 8.2-13.6 9 . 8 3.9
Eastern 12 6.7-14.0 9.4 4.5
+Mean 
4.4.9 (kali
differences 
a yield P6r i
significant. 
a] ant
LSD (5%) = 0.3
The grain yield per plant of the cowpea germplasm is shown 
in Table 18. The range of grain yield per plant was between 2.1 
and 46.9g. There were significant differences in the grain yield 
per plant of the cowpea accessions. The regions also differed 
significantly from one another with respect to their grain yield 
per plant. Except the Eastern Region, the accessions within 
each region showed differences in grain yield per plant. The 
coefficient of variation observed for each of the regions was 
very high
University of Ghana          http://ugspace.ug.edu.gh
- 55 -
4.5 Heterosis estimates in selected crosses
Table 19 shows the mean values for the parents, FI hybrids
as well as midparent heterosis (%) of yield and yield components 
of four crosses involving some selected cowpea accessions. There 
were significant differences (P>0.05) between the FI's and the
parental values of the characters studied. The FI's were the
same as the midparent for days to flowering and days to maturity 
in two crosses but earlier than the midparent in the other two.
The Fi's were intermediate between the parents in two of the 
crosses and higher than the midparental values in the other two 
for pods per peduncle, pod length and 100-seed weight. For
number of seeds per pod the FI's were intermediate between the
parents in three out of the four crosses and was higher than the
midparent in one cross. Midparent heterosis was very high and
positive in both number of pods per plant and grain yield per 
plant in all four crosses. This was particularly so for grain
yield per plant for the crosses 87/27 x 87/157 where heterosis 
was about 130%.
University of Ghana          http://ugspace.ug.edu.gh
-  5 6  -
Table 18. Ranges, means and coefficients of variation for 
grain yield per plant (g) of cowpea germplasm 
from four regions of Ghana.
Reg ion No . o f 
Accs .
Range Meant Coefficient of 
variat ion (%)
Northern 16 2.1-46 . 9 18.5 32 . 5
Upper East 4 3.2-18.8 8 . 2 5 3.5
Upper West 7 4 . 7-35 . 6 19 . 4 44 . 1
Eastern 12 18.5-36.8 2 7.5 29 . 6
+Mean differences significant. LSD (5%) = 12.0
University of Ghana          http://ugspace.ug.edu.gh
Table 19. P e r f o r m a n c e  o f  p a r e n t s  a n d  t h e i r  F1’s a n d  m i d p a r e n t  
h e t e r o s i s  ( % )  i n  f o u r  c r o s s e s  a m o n g  s o m e  s e l e c t e d  
c o w p e a  a c c e s s i o n s .
- 57 -
CROSS
Popn. 
and 
% Het'sis
Days 
to 
f1ower- 
i ng
Days 
to 
matu- 
r i ty
Pods 
per 
ped- 
uncl e
Pods 
per 
pi ant
Pod
1ength 
(cm)
Seeds 100- 
per seed 
pod weight
(g)
Grai n 
yield 
per 
pi ant
(g)
87/1 57 (P1 ) 47 66 2. 1 24.7 15.9 14.7 10.1 37 . 2
I 87/35 (P2) 50 68 1 .4 14.2 12.8 8 . 0 15.9 17.9
F1 49 66 1 .9 26 . 1 14.3 11.4 12.8 38. 1
% Heterosis 0 -2 . 0 8.5 34 . 1 -0 . 3 0 . 4 -1.5 38 . 2
87/27 (P3 ) 39 56 1 . 3 18.2 17.0 16.4 12.3 36 . 9
II 87/35 (P2 ) 49 70 1 . 5 15.1 13.5 8.2 15.6 19.0
F1 43 64 1 .8 29 . 1 17.3 10.5 16.6 51.2
% Heterosis -2 . 2 1 . 5 28.5 74 . 7 13.4 -14.6 18.9 83 . 1
87/56 (P4) 39 56 1 . 5 23 . 2 13.1 11.8 9 . 6 26 . 1
III 87/94 (P5) 43 66 1 . 5 23 . 1 15.3 13.5 10.8 33 . 6
F1 41 64 2.0 36 . 5 14.9 13.0 11.4 54 . 2
% Heterosis 0 4 . 9 33 . 3 57.6 4 . 9 2 . 7 11.7 81.2
87/27 (P3 ) 41 58 1 . 9 15.4 17.9 15.8 11.4 2 . 7
IV 87/1 57 (P1 ) 46 66 2 . 3 40 . 3 15.4 15.6 8 . 6 54.2
F1 42 61 2 . 2 50 . 8 19.1 16.7 10.9 93.3
% Heterosis -2.3 -1.6 4 . 7 82 . 4 14.7 6 . 3 9 . 0 129.8
University of Ghana          http://ugspace.ug.edu.gh
- 58 -
5. DISCUSSION
5.1 Variability In qualitative characters in cowpea germplasm
There was a wide range of variability in most of the 
qualitative characters of the cowpea germplasm, especially, in 
seed colour, plant, flower and pod pigmentation, growth habit, 
pod attachment, raceme, position and twining tendency. Ng and 
Mare'chal (1985) observed a large diversity in cowpea germplasm 
accessions from Ghana in collections at IITA.
The seed colours in the present collection ranged from 
white, red, brown to black with intermediate colours between 
these extremes, as well as mottled colours. Earlier collectors 
in the same regions recorded similar variability in seed coat 
colour in cowpea germplasm (Dovlo, 1976; Holloway,1983). Seed 
coat colour plays an important role in the consumer preference of 
cowpea in Ghana. White or cream is the most preferred colour for 
several dishes because it is quick-cooking and has a pleasant 
flavour, brown and mottled are preferred when rice or maize is 
cooked together with cowpea and red for stews (Dovlo, 1976).
There are indications that the white or cream cowpea types 
may be more susceptible to pests and diseases both as plants in 
the field and as seeds in storage than the other colour types. 
There is evidence for preferential selection by bruchids for 
seeds of a particular variety as food or site for oviposition and 
subsequent development based on factors like size, seed colour
University of Ghana          http://ugspace.ug.edu.gh
- 59
and seed coat texture (Osuji, 1976). Amoatey (1987) observed 
that bruchids preferred large and smooth seeds to small and 
wrinkled seeds for oviposition. He however did not observe any 
trend in the ovipositional behaviour of bruchids based on seed 
colour. Research efforts are being made to incorporate- desirable 
seed characteristics like white, creamy-white or light colours 
into improved strains to suit consumer preferences for seed 
colour (Rachie and Rawal, 1976).
Pigmentation on the stem and pods ranged from green to 
purple and the flowers were either white or purple with the 
latter being more predominant. Ezueh and Nwoffiah (1984) also 
observed white and purple flowers in cowpea accessions in Nigeria 
and the purple types were in the majority. Fery (1985)
recognised four principal flower colours in cowpea, namely, dark, 
pale, tinged and white and related these to the concentration of 
anthocyanin in the flower parts. The dark flowers contain a high 
concentration of anthocyanin in all of the principal flower 
parts, pale flowers contain small amounts in the wings, tinged 
flowers have a faint narrow band of pigmentation along the outer 
edge of the standard, and white flowers have no anthocyanin. 
Flower colour has been associated with seed-coat colour (Harland, 
1919). Generally, dark to pale flowers are characteristic of 
cultivars with pigmented seeds whereas white flowers are 
characteristic of cultivars with white or cream seed coats (Fery, 
1985 ) .
University of Ghana          http://ugspace.ug.edu.gh
-  60 -
Growth habit is a very important factor in the cropping 
system of cowpea in Ghana. The range of growth habits in cowpea 
germplasm is from the climbing types to the erect types. There 
are also intermediate, semi-prostrate and prostrate types. The 
present collection had the intermediate, serni-prostrate and 
prostrate types. Most landraces exhibit the prostrate and 
climbing types (Doku, 1970; Rachie and Rawal, 1976). Rachie and 
Rawal (1976) observed that the prostrate types are used by 
peasant farmers in mixed cropping as they can cover the ground 
quickly and thereby suppress weeds. Climbing types climb 
companion crops which act as live stake.
The intermediate, semi-prostrate and prostrate types of 
cowpea formed the majority of the accessions characterised at 
IITA by Porter si al. (1974). However, deMooy (1935) working on 
cowpea germplasrn from Botswana observed that majority of 
genotypes displayed the erect or semi-erect habits followed by 
the semi-prostrate. The production of cowpea in Botswana may be 
more recent than in Nigeria or Ghana hence the predominance of 
the erect and semi-erect types in that country. It is an. 
objective of recent cowpea improvement programmes to breed the 
erect and semi-erect types of cowpea for the mono-culture 
cultivation of the crop, and prostrate, creeping and climbing 
types for the mixed cropping system (Rachie and Rawal, 1976).
Steele and Mehra (1980) observed that landraces of cowpea
University of Ghana          http://ugspace.ug.edu.gh
- 61 -
are usually indeterminate and determinate types are rare. In 
characterization and preliminary evaluation work at IITA, almost 
all the accessions of cowpea studied showed the indeterminate 
pattern (Porter si fiJL-, 1974). It seems that over the years, 
peasant farmers have selected the indeterminate types to suit 
their farming systems and to ensure continuous production over a 
long period within the season. Similarly, the presence of hairs 
in cultivated cowpea accessions is a rare feature. Hairs are 
known to be present in some wild cowpea types and this is an 
important diagnostic feature between the cultivated and some wild 
species (Steele and Mehra, 1980).
Twining in landraces of cowpea is desirable to the peasant 
farmer who intercropped these types with cereals like millet and 
sorghum. Rachie and Rawal (1976) observed that twining types 
used by peasant farmers have lower performance than the erect and 
semi-erect types. Rachie and Rawal (1976) have proposed the 
breeding of erect and climbing strains for associated cropping.
Majority of the accessions collected had racemes mostly 
above the canopy. It is more desirable to have racemes above the 
canopy than in the upper canopy or throughout the canopy because 
that position facilitates harvesting of the mature pods. Porter 
et a I. (.1974) observed that accessions with racemes throughout
the canopy were the commonest among 4,000 accessions that they 
s tudied.
University of Ghana          http://ugspace.ug.edu.gh
Steele and Mehra (1980) observed that pods of cultivated 
cowpea are mostly pendant whereas those of wild cowpea are mostly 
erect, that is directed vertical to the peduncle. Porter al.
(1974) on the other hand observed that in most of the accessions 
they studied, the pods were inclined 30-90°.
5.2 Variability in quantitative characters in cowpea aermplasm
The cowpea germplasm collected from four cowpea-growing 
regions of Ghana revealed large variability in quantitative 
characters. Much variability was observed in the leaf size of 
the cowpea germplasm. Terminal leaflet length and width ranged 
from 6.8 to 12.8cm and 3.8 to 8.6cm respectively. This implies 
that there is a wide range of leaflet sizes from small to large. 
In a similar work at IITA, terminal leaflet lengths and widths 
had ranges of 7 to 20c;n and 4.5 to 14.5cm respectively (Porter slL 
al.. 1974) whilst in another work in Botswana these traits ranged 
from 5.4 to 14.5cm and 3.0 to 7.8cm respectively (deMooy, 1985). 
The accessions studied by Porter et a 1 . (1974) had the widest
ranges in both terminal leaflet length and width. This was 
followed by those studied by deMooy (1985) and those in the 
present study. This could be attributed to the number of 
accessions studied in each'" case: Porter si al. (1974) worked
with 4,000 accessions, deMooy (1985) worked with 108 accessions 
whilst in the present work there were 45 accessions. Another 
factor which could have influenced the variability of the 
accessions is the scope of the collections.. The collections in 
Porter fii &J.. (1974) were trom all over the world, those in
- 62 -
University of Ghana          http://ugspace.ug.edu.gh
deMooy (1985) were from throughout Botswana and those in this 
study were from four cowpea-growing regions of Ghana.
The leaves of cowpea play an important role in the 
utilization of the crop. Cowpea is grown throughout the tropics 
as leaf vegetable, pulse, fodder and as a cover crop (Steele and 
Mehra, 1980). The young leaves are used as spinach in the 
northern part of Ghana (Asafo-Adjei, 1986). It has been observed 
that if the tender green leaves are plucked before the 
reproduction phase begins, the plant continues to produce new 
leaves (IITA, 1982). This characteristic in cowpea makes it a 
useful spinach, fodder and cover crop (Purseglove, 1968; Steele 
and Mehra, 1980; Asafo-Adjei, 1986). The greater the size of the 
leaflets of cowpea, the more suitable they are for use as 
spinach, fodder and cover crop. The leaflets of cowpea are 
mostly ovate in shape but forms with narrow hastate leaves have 
been bred (IITA, 1977) to improve light penetration deeply into 
the leaf canopy and thereby improve yields (Steele and Mehra, 
1980) .
The cowpea gerrnplasm collected exhibited much variability in 
their flowering dates; the range was 31 to 72 days. There was 
variability both between and within regions. Though most 
accessions flowered, a few remained vegetative throughout the 
period of the experiment (March-July, 1988). The types which did
not flower were from the Upper West, Upper East and Eastern 
Regions. All the accessions from the Northern Region flowered.
- 63 -
University of Ghana          http://ugspace.ug.edu.gh
- 64 -
Amoatey (1987) working on 16 accessions from Ghana and
Nigeria observed days to flowering to range between 37 and 51
days while at IITA the range for over 4,000 accessions was from 
33 to 84 days (Porter al.. 1974). In Botswana, the range for 
108 accessions was 38 to 41 days (deMooy, 1985). It appears that 
the accessions that did not flower within the period of the 
experiment were probably late types, since farmers grow them also 
for their seed. Wien and Summerfield (1980) and Steele and Mehra 
(1980) observed that onset of flowering of local varieties of 
cowpea may be ascribed to photoperiod control which also depends 
on the latitude of origin of the germplasm and the variation in 
day and night temperature.
No significant differences were observed in both the
standard length and the calyx lobe length of the cowpea germplasm 
but significant differences were observed in the number of 
racemes per plant. The siaes of standard and calyx have been 
used to distinguish between Vlena species and subspecies 
(Verdcourt, 1970). Total yield will depend on the number of
racemes, the number of flowers per raceme and more importantly, 
the fertility or the ratio of pods to flowers. Many flowers and 
immature pods abort (Summerfield and Wien, 1980; Summerfield et 
iU^ ., 1985). This abortion in the flowers and immature pods has 
been attributed to warm or cool air, dry atmospheric conditions, 
water stress, abnormalities in pollen formation, poor pollen 
germination and arrested pollen tube elongation {Summerfield and
University of Ghana          http://ugspace.ug.edu.gh
Wien, 1980). Also implicated in the abortion of flowers is the 
abscission-causing insect, Megalurothrins (Summerfield et, al., 
1985 ) .
The variability in the number of days to maturity was closely 
related to the variability in the number of days to flowering. 
Accessions which flowered early also matured early. Doku (1970) 
further observed that for a good yield, flowering period must be 
short to enable most of the plant’s energy material to be diverted 
into seed and pod development and that the longer the pod and seed 
development take the better for seed yield.
The number of pods per plant averaged 2 to 45. Several 
factors, both biotic and abiotic could have influenced the 
variability observed in the number of pods per plant. Since the 
accessions were collected from diverse locations with varied 
ecological conditions, their transfer to Bunso, in the forest area 
might have affected the yield of some of them. Viral diseases were 
observed in some accessions especially those from the Upper East 
Region and this could also have affected the yield. Amoatev (1987) 
found in a study of local and exotic cowpeas that varieties which 
were very severely affected by virus had a very poor yield. 
Furthermore, landraces of cowpea naturally have low seed yield 
(Rachie and Rawal, 1976). In a characterization of cowpea 
germplasm from Botswana, a range of 1 to 99 was observed for the 
number of pods per plant and majority of the accessions had low 
production levels (deMooy, 1985).
-  6 5  -
University of Ghana          http://ugspace.ug.edu.gh
The accessions differed significantly in peduncle length and 
the number of peduncles per plant. The range of peduncle length 
was 15.0 to 41.3cm whilst that of number of peduncles per plant was 
10.0 to 14.3. There is normally one peduncle per node. However, 
there are some genetic stocks with multiple peduncles (Rachie and 
Rawal, 1976). The length of the peduncle determines the position 
of pod in relation to the canopy. There is a rare West African 
variety of cowpea called V i. g n a unguiculata text i 1 i s with long 
peduncles (<60cm) which are used for fibre (Ng and Marechal, 1985;
IITA, 1988c). The pods of this variety are reported to be small 
and upright with small, smooth seeds ( Ng and Marechal, 1985). Long 
peduncles appear to be a disadvantage to pod and seed production in 
cowpea. Dry matter, and its partitioning into useful products like 
pods and seeds is diverted into the long peduncles (Rachie and 
Rawal, 1976). There were therefore research efforts to breed for 
shortened peduncle (Rachie and Rawal, 1976). deMooy (1985) 
observed a range of 4 to 36cm for peduncle length in cowpea 
germplasm from Botswana.
Much variability was observed in pod length of the cowpea 
germplasm which ranged from 8.3 to 17.5cm. Pod length is a very 
important diagnostic feature between cultivated and wild cowpea 
species (Verdcourt, 1970). Steele and Mehra (1980) observed that 
pod length ranges from 4cm in some wild subspecies to 12
to 20cm in subsp. ungu iculata. and one metre in subsp. 
sesquipedalis■ In a world collection of cowpea germpiasm, Porter 
et ad. (1974) observed the range of 7.1 to 4 1.7cm for pod length.
-  6 6  -
University of Ghana          http://ugspace.ug.edu.gh
The cowpea germplasm exhibited variability in the number of 
seeds per pod. There was both between-region and within-region 
variability. The range was 8 to 16. Both the accessions with the 
lowest number of seeds per pod and those with the highest were 
collected from the Northern Region. Amoatey (1987) observed a 
positive correlation between number of seeds per pod and pod 
length. In this study, accessions which had the shortest pods also 
had the lowest number of seeds per pod. Similarly, accessions with 
the longest pods also had the highest number of seeds per pod. 
Research efforts are directed at producing long pods (Rachie and 
Rawal, 1976). Amoatey (1987) observed a range of 10 to 14 seeds 
per pod whilst deMoov (1985) observed a range of 2 to 16.
Significant differences were observed in the 100-seed weight 
of the cowpea germplasm. The range was 6.7 to 21. 4g. Accessions 
from the Eastern Region had the lowest 100-seed weight whilst the 
accessions which had the highest were from the Upper East Region. 
Steele and Mehra (1980 ) observed that the weight of 100 seeds 
varies from lg' in some wild forms to 34g in rare cultivars.
Amoatey (1987) had a range of 9.2 to 20.Og for 100-seed weight for
16 cowpea accessions he studied.
Much variability was found in the grain yield per plant of the 
cowpea accessions. The range was 2.1 to 46.9g per plant.
Both the accessions with the lowest and highest grain yield per
plant were from the Northern Region. Generally the region with
highest mean grain yield per plant was the Eastern Region whilst
-  6 7  -
University of Ghana          http://ugspace.ug.edu.gh
the Upper East had the lowest. Arnoatey (1987) had a range of
20.1 to 63.Og per plant for seed yield. The wide range of 
variability in the grain yield per plant of the accessions may be 
due to the reasons already advanced for the number of pods per 
plant.
A wide range of variability was observed in the 17 
quantitative characters studied as shown by their coefficients of 
variation. Very high coefficients of variation were recorded for 
number of pods per plant and grain yield per plant. High 
coefficients of variation were recorded for hypocotyl length, 
nodes on main stem, main branches per plant, terminal leaflet 
length and width, number of branches per plant, calyx lobe 
length, standard length, peduncle length, number of racemes per 
plant and number of seeds per pod. Characters which had low 
coefficients of variation were days to 50% flowering, peduncles 
per plant, days to 50% maturity, pod length, number of seeds per 
pod and 100-seed weight. Amoatey (1987) observed similar 
coefficients of variation as those in the present study. He 
found that coefficients of variation for grain yield per plant 
and number of pods per peduncle were high whereas those for 
number of peduncles per branch, days to 50% flowering, days to 
50% maturity, number of seeds per pod and pod length were low. 
Similarly, Opoku-Asiama (1978) observed that days to flowering 
and maturity were the least variable in a study of bambara 
groundnut accessions. Opoku-Asiama (1978) also observed that the 
observed variation in the agronomic characters associated with
- 68 -
University of Ghana          http://ugspace.ug.edu.gh
yield in bambara groundnut is largely under genetic control as 
evidenced by their high heritability estimates.
The current trend in cowpea research is to produce types 
which mature early (60 days) and are high yielding (Singh and 
Rachie, 1985). The cowpea-breeding programme at IITA has focused 
on breeding for preferences in different regions, resistance 
against insect pests and diseases and for extra-early and bushy- 
type varieties (Singh and Ntare, 1985). It is desirable to 
develop useful plant types with preferred qualities to suit 
farmers' specific situations (Rachie and Rawal, 1976).
An essential factor in breeding for heterosis is to breed 
hybrids which are more productive than the best available 
cultivars. Several cowpea accessions have been characterized in 
this study from which five accessions were selected for crosses 
for heterosis studies on yield and yield components based on some 
desirable traits they had. Accession 87/157 from the Eastern 
Region had a high number of pods per plant and high grain yield 
per plant; 87/35 from the Northern Region had large white seeds 
and high 100-seed weight; 87/27 from the Northern Region had 
desirable traits which include long pods, high number of seeds 
per pod and high grain yield per plant; accessions 87/94 from the 
Upper West Region and 87/56 from the Northern Region had high 
seed yield and early maturity respectively.
-  69 -
University of Ghana          http://ugspace.ug.edu.gh
-  70  -
Heterosis for yield is a very important economic 
consideration in breeding. In this study all the hybrids of the 
four crosses showed midparent heterosis for grain yield per 
plant. The heterosis for grain yield per plant ranged from 38.2 
to 129.8%, the highest being found in the cross 87/27 x 87/157. 
This should be expected since the parents in this cross were 
among accessions with the highest grain yield per plant.
Another trait in which the crosses had high heterosis was 
number of pods per plant. The range of heterosis in this trait 
was 34.1 to 82.4%. Cross 87/127 x 87/157 once again had the 
highest heterosis. Both accessions 87/27 and 87/157 normally 
produced high number of pods per plant therefore their hybrids 
showed significant heterosis for number of pods per plant.
All the four crosses showed positive midparent heterosis for 
number of pods per peduncle. The range for this trait was 4.7 to 
33.3%. The highest midparent heterosis for number of pods per 
peduncle occurred in the cross 87/56 x 87/94.
Three crosses out of four had positive midparent heterosis 
for pod length. The range for these crosses was 4.9 to 14.7%. 
The cross which showed the highest heterosis for pod length was 
once again 87/27 x 87/157. The parents in this cross were among 
the accessions with the longest pods. It is therefore not 
surprising that their hybrids had longer pods than the midparent.
University of Ghana          http://ugspace.ug.edu.gh
There were three crosses with positive midparent heterosis for 
number of seeds per pod. However, the values for the 
heterosis were all low. The range was 0.4 to 6.3%. The accessions 
87/27 and 87/157 had the highest number of seeds per pod and the 
cross between the two accessions fortunately also produced the 
highest midparent heterosis for the trait.
Three crosses had positive midparent heterosis for 100-seed 
weight whilst one had negative. The range for the positive 
heterosis was 9.0 to 18.9%. The cross with the highest heterosis 
for this trait was 87/27 x 87/35. Accession 87/35 had large seeds 
with high 100-seed weight and therefore influenced the 100-seed 
weight of the hybrid of the cross 87/27 x 87/35.
Heterosis was either negative, positive but low, or nil iri the 
four crosses for days to flowering and days to maturity. Two 
crosses had zero heterosis and two had negative heterosis for days 
to flowering. The implication is that the hybrids with zero 
heterosis flowered at the same time as their midparents whereas the 
hybrids with negative heterosis flowered earlier than their 
midparents. The two crosses in which the hybrids flowered earlier 
than the midparents were 87/27 x 87/35 and 87/27 x 87/157 and those 
in which the hybrids flowered at the same time as the midparents 
were 87/157 x 87/35 and 87/56 x 87/94. Similarly, two hybrids had 
negative midparent heterosis whilst two had positive heterosis for 
days to maturity. This means that the two crosses which had 
negative heterosis matured earlier than their midparents whilst 
those with positive heterosis matured later
- 71 -
University of Ghana          http://ugspace.ug.edu.gh
than their midparents. The two crosses in which the hybrids 
matured earlier than the midparents were 87/157 x 87/35 and 87/27 
x 87/157 whilst those in which the hybrids matured later were
87/27 x 87/35 and 87/56 x 87/94.
In this study, seed yield showed the highest rnidparent 
heterosis among other yield components. This shows the tendency 
for the hybrids to produce more seeds than the average of the 
parents or even than the higher parent. Similar observations 
were made by Yazdi-Samadi si al. (1975) for safflower, Singh
(1973) for Indian mustard, Paschal and Wilcox (1975) for soybean, 
Tarn and Tai (1977) for potato species and Kheradnam si al- 
(1975) for cowpea.
Kheradnam si al- (1975) therefore concluded that the 
commercial production of hybrid seed for cowpeas could be 
advantageous it large scale methods of emasculation and crossing 
were developed for this species. The discovery of male sterility
in cowpea at IITA permits rapid and efficient crossing by both
insects and by hand (Rachie si a±+, 1975) and this could be
exploited to facilitate hybrid seed production in the crop.
Spacing in this study (90cm x 90cm) was greater than the 
recommended spacing for commercial cowpea production in order to 
minimize competition within and between accessions and also to 
facilitate the recording of data for individual plants.
- 7 2  -
University of Ghana          http://ugspace.ug.edu.gh
6.1 Summary
Cowpea germplasm collected from four cowpea-growing regions 
of Ghana in 1987 were studied at Bunso to determine the amount of 
variability in vegetative and reproductive characters among 45 
randomly selected accessions.
The accessions exhibited variability in such qualitative 
characters as growth habit, twining tendency, plant, flower and 
pod pigmentation, raceme position and pod attachment to the 
peduncle. Variability was also exhibited in such quantitative 
characters as terminal leaflet length and width, days to 
flowering and maturity, peduncle length and number of peduncles 
per plant, number of racemes per plant, number of pods per plant, 
pod length, number of seeds per pod, 100-seed weight and grain 
yield per plant.
Based on the results obtained from the variability studies 
five cowpea accessions with contrasting or similar flowering and 
maturity dates, number of pods per plant, pod length and grain 
yield per plant, were selected for heterosis studies. Crosses 
were made between the selected accessions. The parents and their 
Fx hybrids were planted and evaluated over one season.
High and positive midparent heterosis was observed for pods 
per plant and grain yield per plant in all crosses. Heterosis 
for the grain yield for the cross 87/27 x 87/157 was about 130%.
University of Ghana          http://ugspace.ug.edu.gh
- 74 -
6 . 2 Conclusions
Both between- and within-region variability was observed in 
various qualitative and quantitative characters in the cowpea 
germplasm collected from the four regions of Ghana.
Majority of the accessions from all regions had the semi- 
prostrate growth habit, showed no twining, had slight plant 
pigmentation and had purple flowers.
All the accessions studied were indeterminate types. The 
racemes were mostly held above the canopy in most accessions from 
all regions.
In majority of accessions pod attachment to the peduncle was 
pendant.
Variability was observed in both terminal leaflet ’ength and 
width of the cowpea accessions. The accession ith both the 
longest and broadest terminal leaflet wa- from the Northern Region.
Accessions which flowereri and matured earliest were from the 
Northern and Upper Wes' Regions whereas those that flowered and 
matured la'est were from the Upper East Region.
Get^erally accessions from the Eastern Region had the longest 
pods, the highest number of pods per plant and the highest grain 
yield per plant. Accessions from the Upper West and Upper East 
Regions had the highest number of seeds per pod and the highest 
100-seed weight respectively.
University of Ghana          http://ugspace.ug.edu.gh
The cross between accessions 87/27 and 87/157 gave the 
highest midparent heterosis for number of pods per plant, po 
length, number of seeds per pod and grain yield per plant. It i 
recommended that these accessions could be used in futur 
improvement programmes for grain yield on the basis of thei 
performance in this work.
University of Ghana          http://ugspace.ug.edu.gh
-  76  -
7. REFERENCES
Abifarin, A.O. (1984).
Crop Improvement with Particular Reference to Ric-e. IITA, 
Ibadan. pp 178.
Adansi, M.A. and Holloway, H.L.O. (1978). Exploration for plant 
genetic resources in Ghana. Paper presented at the symposium 
on crop irnprovementat at the University of Ghana, Legon, 15 
December, 1978. 13pp Mimeo.
-----------------------------------(1980). The national programme
of Ghana. FAO/IBPGR Plant Genetic Resources Newsletter, 
40:2-5.
Amoatey, H.M. (1987). Genetic studies in some cowpea (Viana 
imguinn 1 at.fi (L) Waip) varieties in Ghana. M.Sc. Thesis, 
University of Ghana, Legon.
Anon, (1988). African Diversity. Newsletter of African Plant 
Genetic Resources Committee. Vol. 1 Addis Ababa.
Asafo-Adjei, B. (1986). Cowpea improvement program: Ghana
Grains Development Project. Tropical Grain Legume Bulletin 
32: 103-109.
Baker, A,.M.R. (1978). Protease inhibitors and the biochemical
basis of insect resistance in iLigrLS un.gu leu lata. Ph.D. 
Thesis. University of Durham, U.K.
University of Ghana          http://ugspace.ug.edu.gh
-  77 -
Bennett-Lartey, S.O. (1983). Plant Genetic Resources activities 
in Ghana. Paper presented at the 13th Biennial Conference of 
the Ghana Science Association at the University of Cape 
Coast, Ghana. 4-8 April, 1983. pp 12 Mimeo.
Bennett-Lartey, S.O. (1984). Characterization and preliminary
evaluation of germplasm collected by the Plant Introduction 
and Exploration (PIE), Bunso in 1982-1983. Technical Report 
84/2. Plant Genetic Resources Unit, Bunso, Ghana.
--------------------- (1986). Cowpea collection and conservation
in Ghana. Paper Presented at the 6th National Maize and 
Cowpea Workshop, 4-6 February, 1986, Kumasi, Ghana. pp 12. 
Mimeo.
----------------------  (1988). Genetic resources of some major
crops of Ghana. Paper presented at the African Workshop on 
Plant Genetic Resources, 17-20 october, 1988. IITA, Ibadan, 
Nigeria. pp 10. Mimeo.
---------------------- (1990). Crops Research Institute - IITA
collaborative collection of yam and cowpea germplasm in 
Ghana. Presented at the external programme review and 
external management review of IITA. 2 February, 1990. Accra, 
Ghana. pp 7. Mimeo.
--------------------- (1991). Plant Genetic Resources Unit
exploration in Ghana 1987-1990. Technical Report, Crops 
Research Institute, Ghana.
Briggs, F.N. and Knowles, P.F. (1967). Introduction t
University of Ghana          http://ugspace.ug.edu.gh
Breeding. Reinhold Publishing Corporation. pp 426.
Burton, G.W. (1968). Heterosis and heterozygosis in pearl millet 
forage producdtion. Crop Science. 8: 229-230.
Chang, T.T., Sharma, S.R., Adair, C.R. and Perez, A.T. (1972).
Manual for Field Collectors of Rice. IRRI, Manila. pp 17. 
Chapman, C. (1989). Principles of germplasm evaluation. In 
Scientific Management of Germplasm: characterization,
evaluation and enhancement (Eds. Stalker, H.T. and Chapman,
C . ) IBPGR, Rome. pp 55-64 . 
deMooy, B.E. (1984). Botswana cowpea germplasm catalogue Vol.1.
Min. of Agric., Dept, of Agric. Research, Sobele, Botswana. 
deMooy, B.E. (1985). Variability of different characterisation
in Botswana cowpea germplasm. Tropical Grain Legume Bulletin 
31:1-4.
Doku, E.V. (1970). Variability in local and exotic varieties of 
cowpea (Vigna unguicu1ata (L) Walp). in Ghana. Ghana Jnl. 
Agric. Sci. 3: 139-143.
Dovlo, F.E. (1976). Dietary uses of grain legumes in Ghana. 
Proceedings of the University of Ghana - Council for 
Scientific and Industrial Research Symposium on grain 
legumes. Legon, Ghana, 10-11 December, 1976.
Dov1o, F.E . , Williams, C . E . and Zoaka, L. (1976). CoKpaa. Home
preparation and use in West Africa IDRC, Ottawa.
El-Hosary, A.A. and Nawar, A.A. (1984). Gene effects in field 
beans. (Vicia faba L. ) II. Earliness and maturity. Egypt. 
Jnl. Genet. Cytol. 13: 109-119.
-  7 8  -
University of Ghana          http://ugspace.ug.edu.gh
Emechebe, A. and Shoyinka, S.A. (1985). Fungal and bacterial 
diseases of cowpeas in Africa. In Cowpea Research.
Production and Utilization (Eds. Singh, S.R. and Rachie, 
K.O.). John Wiley and Sons, Chichester. pp 173-192.
Ezueh, M.I. and Nwoffiah, G.N. (1984). Botanical observations 
on a local collection of vegetable cowpea cultivars in 
South-Eastern Nigeria. Tropical Grain Legume Bulletin 
29: 2-6.
Faris, D.G. (1963). Evidence for the West African origin of 
Vigna sinensis (L) Savi . Ph.D. Thesis University of 
California, California.
Fery, R.L. (1985). The genetics of cowpea: review of the world
literature. In Cowpea Research, Production and Utilization 
(Eds. Singh, S.R. and Rachie, K.O.). John Wiley and Sons, 
Chichester pp 25-62.
GRU, (1976). The plant collecting exploration of Ghana. Genetic 
Resources Unit Exploration 1976. IITA, Ibadan.
  (1978). Report of trip to Ghana, Togo and Benin. Genetic
Resources Unit Exploration 1978. IITA, Ibadan.
Grubben, G.J.H. (1977). Tropical vegetables and their genetic 
resources. (Eds. Tindal, H.D. and William, J.T.). IBPGR
pp 19 7.
GGDP (1990). Maize and cowpea production guide. Ghana/CIDA 
Grains Development Project, January, 1990.
Harland, S.C. (1919). Inheritance of certain characters in the 
cowpea (Vigna sinensis ) ■ Jnl. of Genetics 8: 101-132.
-  7 9  -
University of Ghana          http://ugspace.ug.edu.gh
Holloway, H.L.O. (1983). Germplasm collection in Ghana. Nov.
1982-Feb., 1983. AGPG: IBPGR/83/101.
Howes, C. (1981). Guidelines for developing descriptor lists.
FAO/IBPGR Plant Genetic Resources Newsletter, 45:26-32.
IBPGR (1981). Revised priorities among crops and regions.
IBPGR, Rome. pp 18.
  (1983). Descriptors for cowpea. IBPGR, Rome. pp.29.
IITA (1977). Annual Report. IITA, Ibadan, Nigeria.
  (1982). Cowpea production training manual. IITA Manual
Series No.11, Ibadan, Nigeria.
----- (1983). Cowpea germplasm collection, conservation and
utilization. A special proposal for funding to the 
Government of Italy. Submitted by IITA, Ibadan, Nigeria.
J anuary, 1983.
IITA (1988a). Cowpea improvement in semi-arid tropics of West 
Africa■ IDRC- Food legume Project. IITA, Ibadan.
  (1988b). IITA Annual Report and Research Highlights 1987/88.
Ibadan, Nigeria.
  (1988c). Genetic Resources for Tropical Agriculture. IITA,
Ibadan, Nigeria.
Jackai, L.E.N. and Singh, S.R. (1991). Evaluation of cowpea 
germplasm for insect pest resistance. In Crop Genetic 
Resources of Africa II. (Eds. Ng, N.Q., Perrino, P . , Attere, 
F. and Zedan, H.). Sayce Publishing, U.K. pp 85-93.
- 80 -
University of Ghana          http://ugspace.ug.edu.gh
Kheradnam, M. , Bassiri, A. and Niknejad, M. (1975). Heterosis, 
inbreeding depression and reciprocal effects for yield and 
some yield components in a cowpea cross. Crop Science 15: 
689-691.
Laosuwan, P. and Atkins, R.E. (1977). Estimates of combining 
ability and heterosis in converted exotic sorghum. Crop 
Science 17: 47-50.
Methuen (1981). Methuen Handbook of colours. Eyre Methuen, 
London. pp 252.
Ng., N.Q. (1982). Survey of plant genetic resources in Africa. 
IBPGR/IITA.
Ng, N.Q. and Marechal, R. (1985). Cowpea taxonomy, origin and
germplasm. In Cowpea Research, Producdtion and Utilisation
(Eds. Singh, S.R. and Rachie, K.O.). John Wiley and Sons, 
Chichester pp 13-21.
Opoku-Asiama, Y. (1978). Variation and yield in Bambara
groundnut. (Voandzftia subterranea Thouars. ) . M.Sc. Thesis 
University of Ghana, Ghana.
Osuji, F.N.C. (1976). A comparison of the susceptibility of
cowpea varieties to infestation by Callosobrachus maculatus. 
Entomol. Exp. Appl., 20: 209-217.
Paschal, E.H., II and Wilcox, J.R. (1975). Heterosis and
combining ability in exotic soybean germplasm. Crop Science. 
15: 344-349.
-  81  -
University of Ghana          http://ugspace.ug.edu.gh
Perea, A.T. (1979). The international genetic resources
conservation centre for major food crops based in Nigeria. 
Presented at the 14th Annual Conference of Agricultural 
Society of Nigeria, 2-7 July, 1978. Benin City, Nigeria.
Porter, W.M. (1973). Documentation and maintenance of the
germplasm collection of Viena ungnleu lata. Proceedings of 
the 1st IITA Grain Legume Improvement Workshop, 29 October - 
2 November, 1973, IITA, Ibadan, Nigeria. p 17.
Porter, W.M., Rachie, K.O. Rawal, K.M., Wien, H.C., William, R.J. 
and Luse, R.A. (1974). Cowpea Germplasm Catalog IITA, 
Ibadan, Nigeria. pp 209.
Purseglove, J.W. (1968). Tropical Crops. Dicotyledons.
Longmans, Singapore, pp 199-332.
Rachie, K.O. (1985). Introduction. In Cow p ea Re3earch,
Production and Utilization (Eds. Singh, S.R. and Rachie, K.O) 
John Wiley and Sons, Chichester.
Rachie, K.O. and Rawal, K.M. (1976). Integrated approaches to 
improving cowpeas, V lena. unsuiculata (L) Walp. Technical 
Bulletin 5, IITA, Ibadan, Nigeria.
Rachie, K.O., Rawal, K. and Franchkowiak, J.D. (1975). A rapid 
method of hand crossing cowpeas. Technical Bulletin 2.
IITA, Ibadan, Nigeria.
Rachie, K.O. , Rawal, K. , Franckowiak, J.D., and Akinpelu, M.A.
(1975). Two outcrossing mechanisms in cowpeas, Viana 
ungu.UEiiLa.tLa ( L ) Walp. Euph y t i c a 24: 15 9-163.
- 82 -
University of Ghana          http://ugspace.ug.edu.gh
Rawal, K.M. (1973). Systematic germplasm collection of grain 
legumes in West Africa. Proc. of the first IITA Grain 
Legumes Improvement Workshop, 29 October-2 November, 1973. 
IITA, Ibadan, Nigeria. p 15-16.
Simmonds, N.W. (1979). Principles of crop improvement.
Longman, London. pp 88-97.
Singh, S.P. (1973). Heterosis and combining ability estimates 
in Indian mustard, Brassica juncea (L) Czern. and cox.
Crop Science 13: 397-399.
Singh, B.B. and Ntare, B.R. (1985). Development of improved
cowpea varieties in Africa. In Cowpea Research. Production 
and Utilization (Eds. Singh, S.R, and Rachie, K.O.).
John Wiley and Sons. Chichester pp. 105-115.
Singh, S.R. and Jackai, L.E.N. (1985). Insect pests in cowpeas 
in Africa: their life cycle, economic importance and
potential for control. In Cowpea Research, Production and 
Utilization. (Eds. Singh, S.R. and Rachie, K.O.). John 
Wiley and Sons. Chichester.
Smithson, J.B., Redden, R. and Rawal, K.M. (1980). Methods of 
crop improvement and genetic resources in Vigna ung'uiculata 
In Advances in Legume Science. (Eds. Summerfield, R.J. and 
Bunting, A.H.). Royal Botanic Gardens, Kew. pp 445-457.
Steele, W.M. (1976). Cowpeas. In Evolution of crop plants.
(Ed. S immonds, N.W.). Longman, London. pp 183-185.
-  83 -
University of Ghana          http://ugspace.ug.edu.gh
Steele, W.M. and Mehra, K.L. (1980). Structure, evolution and 
adaptation to farming systems and environments in Vigna.
In Advances in Legume Science. (Eds. Summerfield, R.J. and 
Bunting, A.H.) Royal Botanic Gardens, Kew, pp 393-404.
Strickberger, M.W. (1985). Genetics■ 3rd Edition 
Macmillan, New York, pp 868.
Summerfield, R.J., Pate, J.S., Roberts, E.H., and-Wien, H.C. 
(1985). The physiology of cowpeas. In Cowpea Research, 
Production and Utilization (Eds. Singh, S.r. and Rachie, 
K.O.). John Wiley and Sons. Chichester. pp 65-101.
Summerfield, R.J., and Wien, H.C. (1980). Effects of
photoperiod and air temperature on growth and yield of 
economic legumes. In Advances in Legume Science (Eds. 
Summerfield, R.J. and Bunting, A.H.). Royal Botanic 
Garden, Kew, pp 17-36.
Tarn, T.R. and Tai, G.C.C. (1977). Heterosis and variation of 
yield components in FI hybrids between Group Tuberosum 
and Group Andigena potatoes. Crop Science 17: 517-521.
Verdcourt, B. (1970). Studies in the leguminosae -
Papi1ionoideae for the 'Flora of Tropical East Africa’:
IV Kew Bull. 24:543.
Weber, C.R., Empig, L.T. and Thorne, J.C. (1970). Heterosis 
performance and combining ability of two-way FI soybean 
hybrids. Crop Science 10: 159-160.
-  8 4 -
University of Ghana          http://ugspace.ug.edu.gh
Wien, H.C. and Summerfield, R.J. (1980). Adaptation of cowpea in 
West Africa: Effects of photoperiod and temperature
response in cultivars of diverse origin. In Advances in 
Legume Science (Eds. Summerfield, R.J. and Bunting, A.H.). 
Royal Botanic Gardens, Kew, pp 405-417.
Wobil, J. (1986). Cowpea production in Ghana. Tropical Grain 
Legume Bulletin 32: 101-102.
Yazdi-Samadi, B., Sarafi, A. and Zali, A.A. (1975). Heterosis 
and inbreeding estimates in safflower. Crop Science 15: 
81-83.
University of Ghana          http://ugspace.ug.edu.gh
- 86 -
APPENDIX
Passport data on cowpea germplasm col 1ected from 
four regions of Ghana
Col 1ec- Local 
tor’s No. name
Ethnic Name 
Group of
Vi n -  
age
Region Lat(N) Long(W) Col- Source
our of 
of sam- 
seeds pie
87/25 Sanji
87/27 
87/30 
87/32 
87/34 
87/35 
87/37 
87/40 Isagi
Tingbani Labaraga North 0912 0151 Red Farm
store
Dagomba Borterly 
Tingbani Tua
Sang
Dagbani Malzeri 
Kombani Tuwuwa
0925 0029 Mottled
0925 0027 Red
Whi te
0924 0016 Red
Whi te
0927 0001 Red
0944 0007 Dark
brown
87/41 Ituo
87/44 Sanji sambli Dagbani Shaipo
87/49 Sanji Dagomba Ziong
87/52 Tua
87/555 Sanji Dagbanle Limoh
87/56 
87/58
Whi te 
0958 0038 Light
0931 0030 Red
Whi te
0935 0039
Kokomba Duobunantor " 1024 0007
Red 
Mottled
Red Gar­
den
87/59 Nobega Kusar Manga Upper 1101 0015
East
Whi te
University of Ghana          http://ugspace.ug.edu.gh
_ 87 -
APPENDIX (CONTD)
Col lee- Local Ethnic Name Region Lat(N) Long(W) Col- 
tor’s Name Group of of
No. Vi 11 - seeds
age
87/60
87/61
87/62
87/64
87/67
Benga Kusar Bawku Upper 1103 0014
East
Zarebum- Busanga Manga 
nega
Zare
T i a Nankani Kandiga
1101 0015
Soone Kasena Navrongo
1050 0150
1054 0106
Whi te
Li ght 
brown
Whi te
87/73 Bonda Sisala Tumu Upper 1053 
West
0159 Black
87/81 Sona Buoti
87/83 Bibita- Dagarti Nandom 
kone
87/84 Napopog- 
be
87/90 Oapiala
87/94 Bengeh Wala
87/96 
87/103
87/106 Achibe Gonja
87/132 Asedua Akim
87/133
Sombo
Kampaha
Mengweh
Gel
Tuna
Akora 
Dar ko
1047 0206 White
1051 0042 Black
Light
brown
1015 0233 Mottled
1002 0225
0959 0220
0952 0230
North- 0929 0226
ern
East- 0622 0024 Cream
ern
Wh i te 
Mottled
source
of
sam­
ple
Field
Mar­
ket
Farm
store
University of Ghana          http://ugspace.ug.edu.gh
-  8 8  -
Collec- 
tor ’ s 
No.
Local
name
Ethni c 
Group
Name Region 
of
Vill­
age
Lat(N) Long(W ) Col- Source 
our of 
of sam- 
seeds pie
87/135 Asedua Akim Akora East- 
Darko ern
0622 0024 Red Farm 
store
87/136 ■■ Brown
87/137 ■■ " Red
87/138 ■■ „ Mott1ed
87/139 ■■ „ ■■ Whi te
87/142 Yor Krobo - ■■ Cream
87/147 Asedua Aki m Ahoma-
hamasu
0630 001 7 Red Mar­
ket
87/150 Adua
nsawa
Kwahu Bepong 0636 0043 "  "
87/153 Asak raka 0639 0041 Brown/ Farm 
Mottled store
87/156 „  .. Abene 0657 0057 Whi te
87/157 ■■ „ - Red
87/153 White Mar 
ket
University of Ghana          http://ugspace.ug.edu.gh