Development of cowpea cultivars and germplasmby the Bean/Cowpea CRSP

Anthony E. Halla,*, Ndiaga Cisseb, Samba Thiawb, Hassan O.A. Elawadc,Jeffrey D. Ehlersa, Abdelbagi M. Ismaild, Richard L. Ferye, Philip A. Robertsf,

Laurie W. Kitchg, Larry L. Murdockh, Ousmane Boukari,R.D. Phillipsj, K.H. McWattersj

aDepartment of Botany and Plant Sciences, University of California, Riverside, CA 92521, USAbCentre National de la Recherche Agronomique, B.P. 53, Bambey, Senegal

cAgricultural Research Corporation, El Obeid Research Station, P.O. Box 429, El Obeid, SudandCrop, Soil and Water Sciences Division, International Rice Research Institute, DAPO 7777, Metro Manila, Philippines

eUS Vegetable Laboratory, ARS, USDA, 2875 Savannah Highway, Charleston, SC 29414, USAfDepartment of Nematology, University of California, Riverside, CA 92521, USA

gFAO Sub-Regional Office for Southern and Eastern Africa, P.O. Box 3730, Harare, ZimbabwehDepartment of Entomology, Purdue University, West Lafayette, IN 47907, USA

iInstitut de la Recherche Agronomique pour le Developpement, B.P. 33, Maroua, CameroonjDepartment of Food Science and Technology, University of Georgia, Griffin, GA 30223, USA

Abstract

This paper reviews accomplishments in cowpea cultivar and germplasm development by the Bean/Cowpea Collaborative

Research Support Program (CRSP) which was funded by the United States Agency for International Development for a period of

about 20 years. Drought-adapted, pest and disease resistant cultivars ‘Mouride’, ‘Melakh’ and ‘Ein El Gazal’ were developed for

rainfed production in the tropical Sahelian zone of Africa. Cultivars ‘CRSP Niebe’ and ‘Lori Niebe’ which have seed and pod

resistance to cowpea weevil and some disease resistance were developed for rainfed production in the tropical Savanna zone of

West Africa. Cultivar ‘California Blackeye No. 27’ was developed for irrigated production in subtropical California, USA and is

a semidwarf with heat tolerance and broad-based resistance to root-knot nematodes and Fusarium wilt. Various cultivars with

persistent-green seed color including ‘Bettergreen’ and ‘Charleston Greenpack’ were bred for use in the food freezing industry

in the USA. Germplasms were developed with unique traits including: snap-type pods, green manure/cover crop capabilities,

heat tolerance during reproductive development, chilling tolerance during emergence, delayed leaf senescence as a mechanism

of adaptation to mid-season drought and high grain yields, differences in stable carbon isotope discrimination, harvest index,

rooting and plant water- and nutrient-relations traits, broad-based resistance to root-knot nematodes and Fusarium wilt, and

resistance to flower thrips, cowpea aphid, lygus bug and cowpea weevil, and various quality traits including all-white and sweet

grain. These germplasms provide a valuable resource for breeding additional cowpea cultivars for Africa and the USA.

# 2003 Published by Elsevier Science B.V.

Keywords: Cowpea breeding; Vigna unguiculata (L.) Walp.; Blackeye beans; Blackeye peas; Southern peas; Snap peas; Green manure

cowpeas; Cover crop cowpeas; Niebe

Field Crops Research 82 (2003) 103–134

* Corresponding author. Tel.: þ1-909-236-1580; fax: þ1-909-787-4437.

E-mail address: anthony.hall@ucr.edu (A.E. Hall).

0378-4290/03/$ – see front matter # 2003 Published by Elsevier Science B.V.

doi:10.1016/S0378-4290(03)00033-9

1. Introduction

Improved cultivars provide a powerful stimulus for

rural development enhancing the output of agricultural

systems without requiring substantial additional

inputs, thereby enhancing the productivity, efficiency,

profitability and sustainability of farming systems.

This is particularly true of grain legumes, such as

cowpea (Vigna unguiculata L. Walp.), that have sub-

stantial ability to fix atmospheric nitrogen and often

do not require additional fertilizer to achieve higher

grain yields. For the cowpea production systems in

Africa and the United States that are discussed in this

chapter, typically no nitrogenous fertilizer is applied.

Where soils are infertile and fertilizers are available in

only limited quantities, first priority should be given to

applying the fertilizers to the staple cereals grown in

rotation in the cropping system. In most cases, cereals

are much more responsive to nitrogen fertilizer than is

cowpea.

Resistance to pests is particularly important because

it can enhance productivity, product quality and envir-

onmental conditions as well as decrease input costs

and enhance profits. Reductions in applications of

pesticides is an important goal for cowpea research

programs in Africa and the United States. New culti-

vars also can facilitate the extension of improved

farming systems. It can be much easier to extend

improved crop management, storage and processing

methods if they are recommended as part of a package

that includes a new cultivar, especially if the cultivar is

very attractive to farmers and consumers.

Cowpea is known under several names in the United

States including: blackeye beans, blackeye peas, and

southern peas. In Francophone countries of Africa the

name niebe often is used. Local names for cowpea are

seub and niao in Senegal, wake in Nigeria, and luba

hilu in the Sudan.

The United States Agency for International Deve-

lopment (USAID) provided funding for the Bean/

Cowpea Collaborative Research Support Program

(CRSP) since 1980. This program developed colla-

borative cowpea breeding projects involving several

research programs in Africa and the United States.

A brief history is provided to show the evolution of

these projects and acknowledge the people, organiza-

tions and other funding agencies that made major

contributions.

Collaborative efforts to breed cowpea cultivars for

Africa began prior to the initiation of the Bean/Cow-

pea CRSP. A few years earlier, Dr. A.E. Hall of the

University of California Riverside (UCR) worked in

West Africa as part of a USAID-funded institutional

development program at UCR. During a visit to

Senegal in 1976, Dr. Hall developed a plan for breed-

ing cowpea adapted to the Sahelian zone. About the

same time, he also initiated a program to breed cowpea

for irrigated conditions in California which has

received annual funding support from the cowpea

farmers of California through the Blackeye Varietal

Council of the California Dry Bean Advisory Board.

The collaboration between UCR and the Institut Sene-

galais de Recherches Agricoles (ISRA) that was

initiated in 1976 had a common goal: to develop

cowpea cultivars for target production zones in

Senegal and California. This collaboration has con-

tinued under CRSP funding until the present time and

Dr. Ndiaga Cisse has been the principle ISRA cowpea

breeder.

In 1983 the Agricultural Research Corporation

(ARC) of the Sudan began evaluating cowpea breed-

ing lines developed by UCR. The following year, Dr.

Hall worked in the Sudan with funding from the CRSP

and the USAID-funded Western Sudan Agricultural

Research Project. During this period, he developed a

plan for breeding improved cowpea cultivars for

rainfed production in the Sahelian zone of Sudan.

Dr. Hassan Elawad who had studied with Dr. Hall

returned to ARC, Sudan in 1984 and evaluated cowpea

lines developed by the UCR/ISRA CRSP project and

other organizations until the present time. Research in

the Sudan was funded by the Food and Agricultural

Organization of the United Nations, and the Govern-

ment of the Sudan.

From 1987 to 1990, a scientist from the Savanna

Agricultural Research Institute (SARI) in northern

Ghana, K.O. Marfo, conducted dissertation studies

with Dr. Hall at UCR funded by the German Founda-

tion for International Development. He also

began collaborating with the UCR cowpea breeding

program by making crosses between cultivars from

Ghana and UCR breeding lines. During the 1990s,

Dr. K.O. Marfo used these materials to breed

cowpea cultivars for northern Ghana. Unfortunately,

Dr. K.O. Marfo died in an airplane crash in Ghana

on 5 June 2000 and only a summary is provided of

104 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

his cowpea breeding accomplishments. Dr. Richard

Fery, USDA, Charleston provided advice to CRSP

cowpea breeding programs in Ghana and bred cowpea

cultivars for the southeastern USA, mainly using

USDA funding.

A CRSP cowpea breeding project was initiated in

Cameroon by the Institut de la Recherche Agronomi-

que pour le Developpement (IRAD). During the

1980s, Dr. Moffi E. Ta’Ama worked with IRAD

personnel in evaluating local landraces and importing

breeding lines for use as cultivars in Cameroon.

Hybridizations were initiated in Cameroon in 1990

by Dr. L.W. Kitch of Purdue University who subse-

quently collaborated with plant breeder Ousmane

Boukar of IRAD in breeding improved cultivars. In

later years the IRAD/Purdue project collaborated with

Dr. J.D. Ehlers who had joined the UCR CRSP project

as a cowpea breeder.

Key personnel working with Dr. Hall on breeding

cowpea for irrigated production in California and

rainfed production in Africa include plant pathologist

Dr. P.N. Patel from 1982 to 1992, and plant breeder Dr.

J.D. Ehlers from 1992 through the present time. All of

the various cowpea breeding projects have benefited

from collaboration with the International Institute

of Tropical Agriculture (IITA) and especially with

Dr. B.B. Singh who developed many important breed-

ing lines while at the Ibadan headquarters and the sub-

station at Kano in Nigeria.

2. Cowpea cultivars

In Africa there is considerable cowpea production

in two zones described by some scientists as the

Sahelian Savanna and Sudanian Savanna. The term

Sudanian Savanna can be confusing, however, because

this zone occurs in several African countries. The

more concise terms Sahelian and Savanna are used

in this paper to describe these zones. Development of

cowpea cultivars for rainfed production in the tropical

Sahelian and Savanna zones of Africa, and irrigated

and rainfed production in subtropical zones of the

United States will be considered separately because

these zones have different constraints. In general,

cowpea cultivars tend to have narrow adaptation with

cultivars developed for one zone usually not being

very effective in other zones.

2.1. Cultivars for the tropical Sahelian zone

of Africa

The Sahelian zone represents a tropical semiarid

transition between the Sahara Desert to the north and

the wetter tropical Savanna zone to the south. Changes

in climate that have occurred in this zone during this

century make it difficult to provide a precise descrip-

tion of the location of the Sahel. An approximate

definition is that it stretches across Africa from central

and northern Senegal and southern Mauritania in the

west to central Sudan in the east passing through

central Mali, northern Burkina Faso, southern Niger

and central Chad.

Crops are sown at the beginning of the rainy season

in about July in this zone. Since 1970, the rainy season

often has been short with the annual rainfall only

partially supporting a crop-growing season of about

2 months in most years. For example, average annual

rainfall at Louga, Senegal from 1970 through 1998

was only 267 mm. Drought can be intense because due

to the long dry season there is little available moisture

in the soil from the previous year. Also, the evapora-

tive demand is high during the growing season due to

hot conditions. Evaporative demand can be quantified

by potential crop water use and is about 6 mm per day

during the cropping season at Louga (refer to Fig. 10.5

in Hall, 2001 and Dancette and Hall, 1979). The most

productive cowpea landraces available in Senegal in

the 1970s, such as ‘58-57’, began flowering about 45

days from sowing and took about 80 days from sowing

to maturity. Hydrologic balance estimates predicted

this landrace required 460 mm of water to achieve

maximum grain yields (Hall and Patel, 1987). The

most productive landraces available in other Sahelian

countries had longer cycle lengths from sowing to

maturity (about 90–100 days) and required much more

than 460 mm of water to produce maximum grain

yields. Yet in the 34 years from 1968 to 2001 there

have been 25 years with less than 344 mm rainfall at

Louga. Thus, low rainfall together with limited moist-

ure in the soil at the beginning of the season and high

evaporative demands have resulted in traditional cow-

pea landraces experiencing extreme droughts in most

years from 1968 through 2001. The most productive

cowpea landraces have produced little grain during

the many dry years in the Sahel, and cultivars of

pearl millet (Pennisetum glaucum L. R. Br.), sorghum

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 105

(Sorghum bicolor L. Moench) and peanut (Arachis

hypogaea L.) have been even less productive.

Presumably the cowpea, pearl millet and sorghum

landraces were adapted to the wetter climate prevail-

ing during earlier years in the Sahel. Average annual

rainfall at Louga for the 50 years prior to 1968 was

442 mm compared with the 276 mm received for the

30 years after 1968. Droughts occurred throughout the

Sahelian zone of Africa between 1968 and the mid-

1990s with a severity that was similar to those that

occurred at Louga (Fig. 10.9 in Hall, 2001). Since

1968, the Sahel experienced what may have been the

longest and most severe series of droughts, with

respect to effects on agriculture, that has ever been

recorded.

An ideotype of a cowpea cultivar that would be

adapted to the harsh conditions of the Sahel was

proposed by Dr. Hall in 1976 based upon hydrologic

budget analysis and studies of cowpea responses to

drought in California (Turk et al., 1980). This cultivar

would begin flowering within about 30 days from

sowing, which requires that it be erect and develop

flowers and pods low on the main stem and on early

branch nodes. This type of extra-early cultivar was

predicted to produce a grain yield of about 1000 kg/ha

within 60 days from sowing while using only 200 mm

of water for plants sown at relatively close spacings. In

1976, no cultivar of cowpea had been shown to flower

this early and produce significant quantities of mature

grain within 60 days from sowing, and there were no

other crops with the ability to produce significant food

in this harsh environment. Dr. Hall also predicted that

the Sahelian zone might have some advantages for

cowpea production in that there are fewer insect pests

and diseases in this zone compared with the wetter

Savanna zones to the south where much cowpea was

being grown.

This analysis of the development of cowpea culti-

vars for the Sahelian zone by the CRSP will consider

breeding conducted at UCR, agronomic research in

the Sudan and the breeding of ‘Mouride’ and ‘Melakh’

in Senegal. In the Sudan, Dr. Hassan Elawad evaluated

a total of about 150 breeding lines developed by UCR

and IITA, and accessions from Senegal and elsewhere

over a period of 15 years. Four of these lines were

released as cultivars for use in the central western

Sudan in 2000 by the National Release Committee.

The cultivars were ‘Dahab El Goz’, which was bred by

IITA as the line ‘IT84S 2163’, ‘Gamar Dorein’ and

‘Haydoob’, which are landraces ‘58-57’ and ‘58-185’

from ISRA, Senegal, respectively, and ‘Ein El Gazal’,

which involved a greater contribution by the CRSP in

breeding and is discussed below.

2.1.1. ‘Ein El Gazal’

‘Ein El Gazal’ was bred for use as a dry grain

cultivar in the Sahelian zone (Elawad and Hall, 2002).

It was derived from a cross between a California

cultivar, ‘CB5’ (Mackie, 1946), and a breeding line

from Senegal, ‘Bambey 23’ (Sene and N’Diaye, 1974)

which was made at UCR in 1977. The parents used are

similar in that they are erect, flowered earlier than

other available cowpeas, and have large cream-

colored seed with a rough seed coat and a black

eye. This type of grain is preferred by many cowpea

consumers in the United States and Africa. The par-

ents differed in that ‘CB5’ was bred for irrigated

cropping in subtropical California and has substantial

yield potential; whereas ‘Bambey 23’ was bred and

selected under rainfed conditions in the tropical Cen-

tral Peanut Basin of Senegal. The main objective in

crossing two early flowering lines with different

genetic backgrounds was to achieve transgressive

segregation for even earlier flowering.

Subtropical California is a more effective location

for selecting day-neutral genotypes with early flower-

ing than tropical Senegal. The period from sowing to

flowering is much longer in the subtropics due to the

cooler night-time temperatures compared with the

tropics (Hall, 2001, 2003) and genotypes with differ-

ences in earliness are more easily detected. An F2-

derived F4 population of 585 families and the parents

were grown in a well-irrigated field at Riverside, CA in

1978. Twenty-three lines were selected as single

plants from families that began flowering a few days

earlier and had the same type of large high quality

grain as the parents. Small differences in days to

flowering and appearance of mature pods had been

shown to have large effects on grain yield under

terminal drought (Hall and Grantz, 1981). The

selected lines were erect with synchronous flowering

and abundant pod production. Further line selections

were made out of the 23 lines based upon grain yield

in replicated trials at Riverside, CA under well-

irrigated conditions in 1979, and well-irrigated and

dry stored soil-moisture conditions in 1980 and 1981.

106 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

The selected lines were then tested and further line

selections were made under rainfed conditions with

variable levels of drought in the Sahel at Bambey,

Senegal in 1980, 1981 and 1982, and at Louga,

Senegal in 1981 and 1982. For example, in 1982 at

Louga with only 181 mm of rainfall, UCR line ‘1-12-

3’, which subsequently was released as ‘Ein El Gazal’

in the Sudan, produced 1091 kg/ha of dry grain in only

55 days from sowing (Hall and Patel, 1985). In con-

trast, some Sahelian landraces in an adjacent experi-

ment had only just begun flowering at this time (Hall,

1984) and produced virtually no grain due to terminal

drought. ‘Ein El Gazal’ is extra-early in Senegal with

50% of the plants producing their first flowers about 34

days after sowing and plants reaching physiological

maturity in about 60 days under well-watered condi-

tions and a few days sooner with terminal drought. In

addition, ‘Ein El Gazal’ also has substantial yield

potential and at Bambey, Senegal in 1982 with

452 mm of rain produced 2406 kg/ha of grain (Hall

and Patel, 1985; seasonal rainfall and evaporative

demand at Bambey are described in Fig. 10.8 in Hall,

2001).

The UCR lines were not released for use by farmers

in Senegal due to a special set of circumstances. With

famine impending following extreme droughts in the

period 1982 through 1984, several hundred tons of

cowpea seed were needed in Senegal and only a few

hundred kilograms of seed of the breeding lines was

available. One of the parents used in the breeding,

‘CB5’, had performed reasonably well in dry years in

the Sahelian zone (Hall and Patel, 1985), so in 1985,

650 t of ‘CB5’ seed were imported from California

and sown in Senegal in a project funded by the

European Economic Community and USAID (Bingen

et al., 1988). This project was extremely effective for

famine relief in that the production of ‘CB5’ increased

national cowpea production in Senegal three- to four-

fold in 1985 and 1986 (Bingen et al., 1988). However,

‘CB5’ was not sufficiently robust for long-term use

as a cultivar in Senegal, due to its susceptibility

to the seed-borne disease bacterial blight (caused by

Xanthomonas campestris pv vignicola (Burkholder)

Dye), the insect pest cowpea aphid (Aphis craccivora

Koch) and various wet and dry pod rots. ‘CB5’ had

been bred for irrigated production environments in

California that typically do not experience rain and

pod rots during the cropping season.

A subset of the UCR lines from ‘CB5’�‘Bambey 23’ that produced high average grain yields

in Senegal, including ‘Ein El Gazal’ was sent to the

Sudan. The first yield trial was conducted in the

Sahelian zone at El Obeid Experiment Station, Sudan

in 1983, which was a dry year with only 230 mm of

rain. In this trial, ‘Ein El Gazal’ produced 500 kg/ha of

grain while two local landraces, ‘Garn Elkabish’ and

‘Gambaru’, only produced 135 and 169 kg/ha, respec-

tively (Hall and Patel, 1985). From 1985 through

1993, ‘Ein El Gazal’ was evaluated in seven annual

trials with no insecticide applications on the experi-

ment station at El Obeid with an average annual

rainfall of 285 mm. ‘Ein El Gazal’ produced an aver-

age grain yield of 596 kg/ha and had greater yield

stability than a local landrace that produced an average

grain yield of only 215 kg/ha (Elawad and Hall, 2002).

In these trials, ‘Ein El Gazal’ had an average indivi-

dual seed weight of 186 mg and the seed is cream with

a black eye and a rough seed coat (Elawad and Hall,

2002). ‘Ein El Gazal’ also was evaluated in a total of

60 on-farm trials conducted over 5 years in three

locations with no insecticide applications in northern

Kordofan, Sudan. In these on-farm trials ‘Ein El

Gazal’ produced an average grain yield of 363 kg/ha

compared with the average yield of a local landrace,

‘Baladi’, of only 85 kg/ha (Elawad and Hall, 2002).

By 2001, about 500,000 farmers in the Sahelian zone

of the Sudan were growing ‘Ein El Gazal’ and the

Arabian Sudanese Seed Company began multiplying

large quantities of seed of ‘Ein El Gazal’, ‘Dahab El

Goz’, ‘Gamar Dorien’ and Haydoob’ in response to

a special request for them by the farmers’ union in

Kordofan, Sudan.

2.1.2. ‘Mouride’

‘Mouride’ was bred by ISRA in Senegal (Cisse

et al., 1995) for use as a dry grain cultivar in the

Sahelian zone. It was derived from a cross between

‘58-57’, and a line developed by IITA, ‘IT81D-1137’.

The ‘58-57’ is a selection from a landrace that origi-

nated from Podor in the drier part of the Sahelian zone

(Sene et al., 1971). Line ‘58-57’ was being widely

used as a cultivar throughout northern Senegal when

the cross was made and has resistance to the seed-

borne disease bacterial blight. Breeding line ‘IT81D-

1137’ flowers earlier than ‘58-57’ and has resistance to

cowpea aphid-borne mosaic virus (CABMV), which

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 107

causes a seed-borne disease, and partial resistance to

cowpea storage weevil (Callosobruchus maculatus

Fabricius). Resistance to seed-borne diseases is parti-

cularly important in the many parts of Africa where

farmers are unable to purchase pathogen-free seed.

Cowpea weevil is present in all cowpea producing

areas and is particularly damaging when farmers do

not use appropriate harvesting and storage procedures

(chapter by L.L. Murdock et al.).

F3 seeds from individually harvested F2 plants

were screened for resistance to cowpea weevil. Single

plants were selected from F4 families that had no

symptoms of mosaic virus or bacterial blight under

natural field conditions in Senegal. Selection for

resistance to cowpea weevil was conducted again

during the F5 generation. The presence of resistance

to bacterial blight and some resistance to CABMV

was confirmed using artificial inoculation in the F10

and F11 generations. Research at ISRA has shown that

it can be difficult to obtain totally effective resistance

to CABMV because there are several distinct races

of this virus in Senegal and several resistance genes

are needed (Mbaye Ndiaye, unpublished).

Beginning with the F6 generation, ‘Mouride’ was

performance tested in the central and northern Peanut

Basin of Senegal under the designation ‘IS86-275’.

Tests were conducted on four experiment-station sites

per year with optimal management from 1986 through

1991, and on about 35 on-farm sites per year with little

application of pesticides from 1989 through 1991.

During on-farm tests, it was discovered that ‘Mour-

ide’ has some resistance to the parasitic weed Striga

gesnerioides (Willd.) Vatke. The partial resistance was

confirmed by nursery trials over several years in fields

heavily infested with Striga. It was shown that the

Striga resistance of ‘Mouride’ probably came from

parent ‘IT81D-1137’ and is not as effective against the

Striga biotypes in Senegal as that of cowpea line

‘B301’ (Moctar Wade, unpublished).

In the multilocation yield trials in Senegal, ‘Mouride’

has consistently produced 18% more grain but 17%

less hay than landrace ‘58-57’ and much greater

average grain yield than other traditional cultivars.

‘Mouride’ is semierect and indeterminate, begins

flowering about 37 days after sowing and reaches

physiological maturity under well-watered conditions

about 68 days after sowing. It is the most robust of the

cultivars developed for the Sahel by the CRSP. With

adequate rainfall, ‘Mouride’ has achieved grain yields

as high as 3000 kg/ha at Bambey, Senegal, which is

higher than has been achieved by either ‘Ein El Gazal’

or another important cultivar developed by the CRSP,

‘Melakh’. ‘Mouride’ also has greater resistance to

midseason drought but less resistance to terminal

droughts than ‘Melakh’, which has earlier flowering

and a shorter cycle length.

Seed of ‘Mouride’ is cream with a brown eye and an

average individual seed weight of 160 mg for plants

grown in Senegal. The grain are considered smaller

than is desirable by some consumers, but the local

landrace, ‘58-57’, has even smaller seed (average

individual seed weight of 120 mg), and yet has been

widely used as a cultivar in northern Senegal (Sene

et al., 1971; Cisse et al., 1995) and has been released as

a cultivar in the Sudan. The Institut de Technologie

Alimentaire (ITA) in Dakar, Senegal evaluated the

grain quality of ‘Mouride’ (ITA 1990 Report, unpub-

lished). Protein content of ‘Mouride’ based on total

nitrogen measurements was 23% on a dry weight basis

compared with 22% for ‘CB5’ and 25% for ‘Bambey

21’. Cooking time of ‘Mouride’ was 44 min compared

with 25 min for ‘CB5’ and 41 min for ‘Bambey 21’. In

appearance during cooking, ‘Mouride’ rated poorly

due to the presence of a high percentage of split grain,

whereas ‘Bambey 21’ rated the highest due to it having

firm texture and no split grain, while ‘CB5’ received a

good intermediate rating. In sensory evaluations, con-

sumers rated ‘Bambey 21’ as being the highest with

‘Mouride’ being the lowest due to perceived problems

with texture and color, whereas ‘CB5’ was given a

good intermediate ranking. The overall conclusion

was that ‘Mouride’ had acceptable but marginal grain

quality that was not as good as that of either ‘CB5’ or

‘Bambey 21’. It should be noted, however, that after

the early 1990s neither ‘CB5’ nor ‘Bambey 21’ were

widely grown in Senegal due to various weaknesses

such as their susceptibility to several diseases.

‘Mouride’ was released by ISRA in Senegal in 1992

(Cisse et al., 1995). In the summer of 1993, seed of

‘Mouride’ was distributed to about 1000 farmers in

300 villages in northern Senegal by World Vision

International (WVI). Average on-farm grain yield

by sole crops of ‘Mouride’ was estimated to be about

1000 kg/ha by WVI. ‘Mouride’ also was released for

use in Guinea-Bissau in 1993. In a 5-year InterCRSP

project that ended in February 2001, WVI evaluated

108 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

and extended ‘Mouride’ in Ghana, Niger and Chad. In

Niger, ‘Mouride’ produced high grain yields and was

popular, with farmers ‘stealing’ seed from experimen-

tal plots prior to the planned harvest and requesting

large-scale multiplication of seed. ‘Mouride’ has also

produced high grain yields in trials in the Sudan

conducted by the ARC since 1997.

2.1.3. ‘Melakh’

‘Melakh’ was bred by ISRA in Senegal (Cisse et al.,

1997) for use as a dual-purpose dry grain/fresh south-

ern pea cultivar in the Sahelian zone. It was derived

from a cross between ISRA breeding line, ‘IS86-292’,

and IITA breeding line, ‘IT83S-742-13’. Line ‘IS86-

292’ is from the same cross that produced ‘Mouride’

(Cisse et al., 1995) and has high yield potential and

resistance to bacterial blight and CABMV. Line

‘IT83S-742-13’ flowers early and has resistance to

cowpea aphid (A. craccivora Koch). In some years,

cowpea aphid can be a major pest of cowpea in the

Sahelian zone, killing young plants, severely stunting

older plants and damaging pods and grain.

Single F2 plants having no infestations of cowpea

aphid and no symptoms of either mosaic virus or

bacterial blight were selected from a field nursery.

In the F4 generation, artificial infestation and inocula-

tion were used in screening for resistance to cowpea

aphid and CABMV. Seeds from a single F6 plant were

bulked and introduced in performance trials in the

central and northern Peanut Basin of Senegal under

the designation, ‘B89-504’. These tests were con-

ducted on four experiment-station sites per year with

optimal management from 1989 through 1992, and on

35 on-farm sites per year with little application of

pesticides in 1991 and 1992.

Segregation for resistance to bacterial blight was

observed, and single resistant plants were selected,

self-pollinated and screened for resistance to bacterial

blight in the F9 and F10 generations using artificial

inoculation. During the F11 generation, seedlings of

bacterial blight-resistant lines were screened for resis-

tance to cowpea aphid using artificial infestation.

Seeds of resistant lines were bulked to form the final

version of ‘Melakh’. Subsequent field tests indicated

‘Melakh’ also has partial resistance to flower thrips

which can be a moderate problem in the Sahelian zone

and is the major constraint for cowpea production in

the Savanna zone.

‘Melakh’ is an early cowpea cultivar flowering

about 35 days from sowing and reaching physiological

maturity about 64 days from sowing under well-

watered conditions in Senegal. It is semierect and

indeterminate with strong resistance to terminal

drought but can be damaged by midseason drought.

When ‘Melakh’ is grown in Senegal, its seeds have

moderate size with average individual seed weight of

190 mg and are white with a brown eye. In the multi-

location yield trials in Senegal, ‘Melakh’ produced

30% more grain and forage than other early cowpea

cultivars, such as ‘CB5’ and ‘Bambey 21’.

In the summer of 1993, seeds of ‘Melakh’ were pre-

released and distributed to about 1000 farmers in 300

villages in northern Senegal by WVI. Average on-farm

grain yields by sole crops of ‘Melakh’ were estimated

to be about 1000 kg/ha by WVI and farmers rated

‘Melakh’ as being more desirable than ‘Mouride’.

‘Melakh’ was released by ISRA in Senegal in 1996

(Cisse et al., 1997). In subsequent years, WVI and the

Belgian NGO, Aquadev extended ‘Melakh’ in north-

ern Senegal. More recently, WVI organized farmer-

based cowpea seed production cooperatives affiliated

with the national union of personnel involved in seed

production in Senegal (UNIS). An Austrian NGO,

EWA, has been working with farmers in Senegal to

increase production of ‘Melakh’, build cowpea storage

facilities, and promote cowpea export to neighboring

countries. A factory in Mauritania has begun using

flour from grain of ‘Melakh’ to make biscuits. In

addition to its use as a dry grain, ‘Melakh’ has become

popular in Senegal as a source of fresh southern peas

during the typical period of hunger between mid-

August to mid-September, which is prior to the harvest

of the staple cereals, pearl millet and sorghum. As with

‘Mouride’, WVI evaluated and extended ‘Melakh’ in

Ghana, Niger and Chad. ‘Melakh’ produced high

yields in Ghana and Niger and was even more popular

than ‘Mouride’. ‘Melakh’ also has produced high

grain yields in trials conducted in the Sudan by

ARC since 1997 and is being considered for release

in this country.

‘Mouride’ and ‘Melakh’ are complementary with

‘Mouride’ having higher yield potential and stronger

resistance to mid-season drought, and ‘Melakh’ hav-

ing greater resistance to terminal drought. Each culti-

var has resistance to different major pests. Since the

types of drought and attacks by pests can vary from

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 109

year to year, and among fields, farmers are encouraged

to grow both cultivars as a means for increasing the

stability of their farming systems. The cultivars can be

grown as either sole crops or varietal intercrops.

Varietal intercrops with alternating rows of early erect

and later flowering spreading cowpea cultivars were

shown to produce more grain and hay and be more

stable under dry Sahelian conditions than any of the

sole crops that were tested (Thiaw et al., 1993). The

varietal intercrops or sole crops of cowpeas should be

grown in annual rotation with pearl millet and peanut

since ISRA studies have shown that rotations can

provide certain advantages. Pearl millet crops can

suppress the levels of sclerotia of the fungus Macro-

phomina phaseolina (Tassi) Goid. which causes the

ashy stem blight disease of cowpea (Mbaye Ndiaye,

unpublished). This disease can devastate cowpea and

other control methods are not available. Germinating

seeds of some cowpea cultivars, but not other culti-

vars, can stimulate suicidal germination of seeds of

Striga hermonthica Benth. (Moctar Wade, unpub-

lished). This Striga species is a major parasitic weed

of pearl millet and sorghum for which few control

methods are available. Cowpea cultivars and cultural

practices used in Senegal have been described by

Cisse et al. (1996) and Hall et al. (1996).

2.2. Cultivars for the tropical Savanna zone of Africa

In the tropical Savanna zone of Africa there are

several insect pests that cause major damage to cow-

peas, including flower thrips (Megalurothrips sjostedti

Trybom), pod sucking bugs (Clavigralla spp., Ano-

plocnemis curvipes Fabricius, Nezara viridula L.,

Riptortus dentipes Fabricius) and pod borer (Maruca

testulalis Geyer) and many organisms causing dis-

eases that are not prevalent in the Sahelian zone (Hall

et al., 1997a). The IRAD Cameroon CRSP cowpea

breeding project evaluated some 400 local Cameroo-

nian landraces and 300 imported cowpea breeding

lines mainly from IITA, Nigeria that had resistance

to some of these diseases. Following these evaluations,

the project promoted farmer use of a local landrace,

‘VYA’. This is a spreading type cowpea with a med-

ium cycle length taking 85–90 days from sowing to

maturity, and producing white grain of moderate size

when grown in Cameroon (average individual seed

weight of 150 mg). IRAD also released two IITA

breeding lines in 1985: ‘IT81D-985’, under the name

‘BR1’, and ‘IT81D-994’ under the name ‘BR2’. These

cultivars have a shorter cycle length than ‘VYA’

reaching maturity in about 70 days from sowing. They

have partial seed resistance to cowpea weevil, non-

dehiscent tough pods and a moderate level of pod

resistance to cowpea weevil. Their grain color is

similar to ‘VYA’. ‘BR1’, ‘BR2’ and ‘VYA’ are sus-

ceptible to races of CABMV and biotypes of Striga

occurring in Cameroon. However, ‘VYA’ is more

resistant to CABMV than either ‘BR1’ or ‘BR2’. This

seed-borne disease and the parasitic weed Striga are

major constraints to cowpea production in Cameroon.

A hybridization breeding program was initiated in

1990 that developed cultivars ‘CRSP Niebe’ and ‘Lori

Niebe’. Selection of advanced lines by this breeding

program was guided by a group of expert farmers

(Kitch et al., 1998). Farmer participation in the breed-

ing program was intended to enhance farmer adoption

of cultivars. It also produced an unexpected benefit,

the discovery of an important new trait that can

influence consumer acceptance of cowpea grain (see

Section 3.17 on cowpeas with ‘sweet’ grain).

2.2.1. Development of cultivar ‘CRSP’ Niebe’

‘CRSP Niebe’ was bred by IRAD, Cameroon for

use as a dual-purpose dry grain/hay cultivar in the

Savanna zone. It was derived from a cross between

‘VYA’ and ‘BR1’. The objectives were to combine the

seed and pod resistance to cowpea weevil of ‘BR1’

with the enhanced resistance to CABMVof ‘VYA’ and

also to achieve greater grain yields. Selections were

made in Maroua, Cameroon that had some seed and

pod resistance to cowpea weevil, non-dehiscent tough

pods and absence of foliar mosaic virus symptoms.

When grown in Cameroon, ‘CRSP Niebe’ produces

grain with an average individual seed weight of

176 mg and a white rough seed coat. It has a spreading

habit and reaches maturity in about 85 days from

sowing.

‘CRSP Niebe’ was evaluated as breeding line

‘C93W 2-38’ in multilocation station trials in Camer-

oon from 1994 to 1996. The trials were conducted on

experiment stations in two locations in the Savanna

zone of the Extreme North Province which has aver-

age annual rainfall of 700–900 mm, and in two loca-

tions in the wetter Savanna zone of the North Province

which has average annual rainfall of 900–1200 mm.

110 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

The trials were conducted under a moderate input

regime consisting of two insecticide sprays to protect

the plants against pests at flowering and at early

podding, two manual cultivations to remove weeds

and no fertilizer. Grain yields of ‘CRSP Niebe’ for

1994–1996 averaged 881 kg/ha compared with

537 kg/ha for the check cultivar ‘BR1’. Fodder yields

of ‘CRSP Niebe’ for 1995 and 1996 averaged 2195 kg/

ha compared with 2489 kg/ha for ‘BR1’. On-farm

trials were conducted at one site in 1996, five sites

in 1997 and about 50 sites in 1998 and 1999. ‘CRSP

Niebe’ received high rankings in cooking and taste

tests in Cameroon. Planting density trials were con-

ducted on experiment stations to determine optimal

planting densities. ‘CRSP Niebe’ was released by

IRAD and the Cameroon Ministry of Agriculture in

September 1999.

2.2.2. Development of cultivar ‘Lori Niebe’

‘Lori Niebe’ was bred by IRAD, Cameroon for use

as a dual-purpose grain/hay cultivar in the Savanna

zone. It was derived from a single F4 plant selection

made in Maroua, Cameroon in 1993 from a cross

between IITA breeding lines ‘IT86D-364’ and

‘IT81D-1138’. Both lines possess partial seed resis-

tance to cowpea weevil and ‘IT86D-364’ had exhib-

ited resistance to CABMV in Cameroon. Selections

were made in Maroua, Cameroon that had some seed

and pod resistance to cowpea weevil, non-dehiscent

pods, absence of foliar mosaic virus and bacterial

blight symptoms, and moderate sized grain (average

individual seed weight of 194 mg for plants grown in

Cameroon) with a rough white seed coat and a dark

brown eye. ‘Lori Niebe’ has a spreading plant habit

and reaches maturity about 82 days from sowing.

‘Lori Niebe’ was evaluated as breeding line ‘C93W

24-130’ in the same multilocation station trials and on-

farm trials in Cameroon that were used to select

‘CRSP Niebe’. Grain yields of ‘Lori Niebe’ in the

multilocation trials averaged 994 kg/ha compared

with 537 kg/ha for the check cultivar ‘BR1’. Fodder

yields of ‘Lori Niebe’ for 1995 and 1996 averaged

1918 kg/ha compared with 2489 kg/ha for ‘BR1’.

Grain yields in on-farm trials averaged about

800 kg/ha. ‘Lori Niebe’ received high rankings in

cooking and taste tests in Cameroon and was released

by IRAD and the Cameroon Ministry of Agriculture in

September 1999.

2.3. Cultivars for subtropical zones in

the United States

The San Joaquin Valley of California and the high

plains of Texas are the major regions of dry grain

cowpea production in the United States. Grains are

marketed as dry blackeye beans to as many as 30

countries. When provided with complete irrigation the

San Joaquin Valley of California can be one of the

most productive areas for cowpea in the world due to

its long sunny days and cool night temperatures that

result in relatively long cycles from sowing to flower-

ing and maturity (Hall, 1999, 2003). Farm grain yields

of up to 7 t/ha have been achieved in California with

crops that went from sowing to harvest in about 145

days (Hall, 2003). Cowpea cultivars used in Califor-

nia, and methods for growing them under irrigation

have been described by Hall and Frate (1996). Cow-

peas also are grown in the southeastern United States

under rainfed conditions mainly for use as southern

peas for canned and frozen products, and for home use

as fresh southern peas.

2.3.1. ‘California Blackeye 27’

‘California Blackeye 27’ (CB27) was bred for use

as a dry grain type of cultivar in the San Joaquin Valley

of California (Ehlers et al., 2000a). It resulted from a

complex series of crosses with some selections

between the crosses. In 1981 ‘Prima’ and ‘TVu4552’

were crossed to combine different components of

tolerance to high night temperature during reproduc-

tive development that were present in these West

African accessions (Hall, 1992, 1993). Progeny were

then crossed to California cultivar ‘CB5’ (Mackie,

1946), breeding line ‘7977’ from the University of

California, Davis and then ‘CB5’ with some selections

between the crosses. ‘CB5’ was used to provide adap-

tation to irrigated production in California, large cream

seed with a black eye and the Rk resistance to root-knot

nematodes. Line ‘7977’ was used as a parent to provide

resistance to Fusarium wilt (Fusarium oxysporum f.

sp. tracheiphilum (E.F. Smith) W.C. Snyder & H.N.

Hansen) and it also has large cream seed with a black

eye and adaptation to irrigated production in Califor-

nia. Progenies from this series of crosses were selected

for reproductive-stage heat tolerance as determined by

ability to produce flowers and set pods under extremely

hot field conditions during the summer at the UC

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 111

Desert Research and Extension Center, near El Centro,

CA (temperatures for this location are provided in

Fig. 10.2 in Hall, 2001). Selection for agronomic traits

and grain yield and quality was conducted at locations

with intermediate temperatures, the UC Riverside

Experiment Station and the UC Kearney Agricultural

Research and Extension Center in the San Joaquin

Valley, CA (temperatures for these locations are shown

in Figs. 5.2 and 10.4 in Hall, 2001). A heat-tolerant line

was produced ‘1393’ that is well adapted to the San

Joaquin Valley, CA. Unfortunately, a small percentage

of the grain of ‘1393’ had brown eyes (<0.1%) rather

than black eye markings, which was not acceptable to

the market.

Another promising breeding line had been devel-

oped, ‘336’, by selecting for agronomic traits among

progenies from a cross between two California culti-

vars, ‘CB5’ and ‘CB3’, with ‘CB3’ providing a type of

resistance to Fusarium wilt that was different from that

in line ‘7977’ (Rigert and Foster, 1987). In 1986, lines

‘1393’ and ‘336’ were crossed and progenies were

screened for reproductive-stage heat tolerance at

two UC Experiment stations: as before near El Centro

in the Imperial Valley, CA, and also near Thermal in

the Coachella Valley, CA. The progenies also were

subjected to multilocation performance testing at

locations with intermediate temperatures in the San

Joaquin Valley, CA and ‘CB27’ was selected.

‘CB27’ was evaluated in performance trials and

various screening trials under the designation ‘H8-8-

27’ from 1994 through 1998 (Ehlers et al., 2000a). In

addition, six pairs of lines with and without heat

tolerance during reproductive development, including

‘H8-8-27’, were evaluated in eight field environments

with contrasting temperatures. ‘H8-8-27’ and the

other heat-tolerant lines were shown to have higher

grain yields than the heat-susceptible lines, including

‘CB5’, when hot weather occurred at flowering (Ismail

and Hall, 1998).

Various forms of root-knot nematode species Meloi-

dogyne incognita (Kofoid and White) Chitwood and

Meloidogyne javanica (Treub) Chitwood occur in

California and many other countries that can cause

substantial damage to cowpea (Roberts et al., 1997).

Some current California cultivars (‘CB46’, ‘CB88’

and ‘CB5’ are described in Hall and Frate, 1996) carry

the nematode resistance gene Rk that confers strong

resistance to common strains of M. incognita. ‘CB27’

was shown to carry gene Rk and an additional reces-

sive non-allelic gene from ‘TVu4552’, designated rk3,

that act together in an additive fashion to provide

protection against both Rk-avirulent and Rk-virulent

forms of M. incognita and M. javanica (Ehlers et al.,

2000c). Reproduction of the nematodes and root gal-

ling on ‘CB27’ caused by the Rk-virulent forms were

about half those observed on ‘CB46’ and ‘CB88’

(Roberts et al., 1997).

Various races of F. oxysporum f. sp. tracheiphilum

occur in California and many other countries and can

cause the Fusarium wilt disease of cowpea which can

result in substantial reductions in grain yield (Patel,

1985). Some current California cultivars, ‘CB46’ and

‘CB88’, have resistance to Fusarium wilt incited by

Race 3, the predominant race in California, but they do

not have resistance to Race 4, which recently has been

identified in different parts of the main production

zone in the San Joaquin Valley (Ehlers et al., 2000a).

‘CB27’ was shown to have resistance to both Race 3

and Race 4 of Fusarium wilt with the resistance likely

coming from ‘CB3’ but with some contribution being

possible from line ‘7977’.

‘CB27’ has yielded more than ‘CB46’ and ‘CB88’

in fields in California where both Race 4 of Fusarium

wilt and gene Rk-virulent M. incognita were present.

In two on-farm trials in soils where these diseases and

pests were present, ‘CB27’ had an average grain yield

of 2547 kg/ha, while ‘CB46’ yielded 2089 kg/ha, and

‘CB88’ yielded only 1215 kg/ha (Ehlers et al., 2000a).

‘CB27’ and the major California cultivar ‘CB46’

had similar average grain yield in fields where soil-

borne diseases and pests were not detected. In 16 yield

trials at several sites in the San Joaquin Valley ‘CB27’

and ‘CB46’ had average grain yields of 4350 and

4360 kg/ha, respectively (Ehlers et al., 2000a). ‘CB27’

had higher yields in trials conducted under hot con-

ditions, whereas ‘CB46’ had higher yields in trials

conducted with wide rows.

‘CB27’ and ‘CB46’ are the first semidwarf cowpea

cultivars and produce shorter main stems and less

vegetative shoot biomass than traditional erect inde-

terminate cultivars, such as ‘CB5’ (Ismail and Hall,

2000). However, ‘CB27’ is more dwarfed than ‘CB46’

with a shorter main stem, shorter internodes and less

vegetative shoot biomass. At narrow row spacing

(51 cm apart) and soil conditions that promoted mod-

erate to vigorous early plant growth, ‘CB27’ produced

112 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

greater grain yield than ‘CB46’ and standard-height

lines, such as ‘CB5’, in California. The yield advantage

of ‘CB27’ was due to less impairment of reproduction

when competition for light was strong compared with

standard-height lines (Ismail and Hall, 2000). The

extreme semidwarf habit of ‘CB27’ confers disadvan-

tages, with respect to grain yield and competition with

weeds, when grown in wide row spacing (e.g. 102 cm

apart) and where soil conditions do not permit at least

moderate early vegetative growth.

Selection was conducted for visual aspects of grain

quality. The heat-tolerant parental line ‘TVu4552’ can

exhibit an undesirable brown discoloration of the seed

coat that is more pronounced under higher night

temperatures (Nielsen and Hall, 1985b). This undesir-

able trait is simply inherited and is not linked to heat

tolerance during floral bud development (Patel and

Hall, 1988). Consequently, lines with the brown dis-

coloration trait can be removed by selection under

warm to hot conditions while retaining lines with heat

tolerance. The important IITA line ‘TVx 3236’, which

has been used as a cultivar and in many breeding

programs due to its partial resistance to flower thrips,

also has exhibited undesirable brown discoloration of

the seed coat when grown in Senegal. Selection alsowas

conducted to eliminate lines with seed-coat cracking

and seeds that were too small. ‘CB27’ has a bright white

seed coat that is more attractive to some consumers

than the cream seed coat of ‘CB46’, and typical ‘black-

eye bean’ appearance. In California, average individual

seed weight of ‘CB27’ was moderately large, 224 mg,

compared with 217 mg for ‘CB46’ (Ehlers et al.,

2000a), and about 240 mg for the large-seeded cultivar

‘CB5’ (Hall and Frate, 1996).

Selection also was conducted to eliminate lines that

leak pigments when soaked in boiling water (Hall

et al., 1997a). Two of the parents used in breeding,

‘Prima’ and ‘7977’, leak a purple pigment from the

black eye when soaked in boiling water, which would

be an undesirable feature for many consumers.

However, the black eye of ‘CB27’ does not leak

pigments during boiling. Canning tests conducted

by a commercial company and the CRSP (G.L. Hos-

field, unpublished) indicated that ‘CB27’ has excellent

canning quality. ‘CB27’ has been granted cultivar

protection under the US Plant Variety Protection

Act and Title V of the Federal Seed Act (Certificate

No. 200000183).

2.3.2. Development of cultivars for the USA

with persistent-green seed color

Development of horticultural-type cowpea cultivars

with persistent-green seed color has been of much

interest to US food processors, especially those

involved in the freezing industry. Grain of persis-

tent-green cultivars can be harvested at the dry-seed

stage of maturity without losing the green color people

associate with fresh southern peas. This color reten-

tion is important because mechanical harvesting of

dry grain is cheaper than mechanical harvesting of

moist grain. Chambliss (1974) proposed that the

green testa gene (gt) conditions a green seed coat

that persists in the dry seed, and results in a processed

product with improved consumer appeal. Although a

cultivar homozygous for the gt gene was released

(Chambliss, 1979), the green testa trait was not

accepted by the processing industry. A green cotyle-

don mutant was discovered in the cultivar ‘Carolina

Cream’, and used to develop the green cotyledon

cultivar ‘Bettergreen’ (Fery et al., 1993). The new

trait is similar to the green cotyledon trait reported

in lima bean by Magruder and Wester (1941) that

revolutionized the lima bean industry. The new cow-

pea cultivar ‘Bettergreen’ was quickly accepted by

the freezing industry.

Fery and Dukes (1994) studied inheritance of the

green cotyledon trait and the genetic relationship

between the green cotyledon and green testa traits.

They concluded that the green cotyledon trait in cow-

pea is conditioned by a single recessive gene, symbo-

lized gc, and that this gene is neither allelic to nor

linked with the gt gene. Both genes together condition

a green seed color that is richer and more uniform than

the green color exhibited by seeds of genotypes homo-

zygous for the gc gene alone. However, the decision

was made to first convert standard pinkeye-type,

blackeye-type, and cream-type germplasm to persis-

tent-green seed color using the gc gene alone because

it was thought that the use of both gc and gt genes

might result in the development of a persistent-green

seed color that would be too dark for ready acceptance

by industry. These efforts resulted in the release of

the pinkeye-type cultivars ‘Charleston Greenpack’

and ‘Petite-N-Green’ (Fery, 1998, 1999). ‘Charleston

Greenpack’ was quickly accepted by the freezing

industry in part because it forms an attractive product

when blended with grain of the widely grown pinkeye

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 113

cultivar ‘Coronet’. Other cowpea cultivars with the

green cotyledon phenotype that were bred are the

cream-type cultivar ‘Green Pixie’, the blackeye-type

cultivar ‘Green Dixie Blackeye’ and the nematode-

resistant, cream-type cultivar ‘Charleston Nemagreen’

(Fery, 2000, 2002). ‘Charleston Greenpack’ and

‘Petite-N-Green’ have been granted cultivar protec-

tion under the US Plant Variety Protection Act and

Title V of the Federal Seed Act (Certificate Nos.

9700286 and 9900074, respectively).

Various horticultural types of cowpea cultivars are

being developed that have both the gc and gt genes.

‘DoubleGreen Delight’, a cream-type cultivar, was

released in 2001 by R.L. Fery. A major attribute of

the new cultivar is the persistence of the green color

in unharvested dry seeds that enables growers of

‘DoubleGreen Delight’ to have a harvest window of

long duration during which they still can obtain a

high-color product. Additionally, the dry seeds have a

richer and more uniform green color than dry seeds of

other green-seeded cowpea cultivars available to US

growers prior to 2002.

Cowpea cultivars with persistent-green seed phe-

notypes have the potential for wide-spread application

in the US freezing and dry-pack cowpea industries.

Cowpea breeding programs at the US. Vegetable

Laboratory in Charleston, SC, UCR and Texas

A&M University, College Station are breeding cow-

pea cultivars with persistent-green seed phenotypes

suitable for use in all of the cowpea producing areas of

the United States.

3. Cowpea germplasm

3.1. Promising breeding lines

Breeding lines are described that are likely to be

released as cultivars in the near future.

3.1.1. Line ‘ISRA-819’ which is adapted to

the Sahelian zone of Africa

‘ISRA-819’ was bred by Ndiaga Cisse, ISRA,

Senegal for use as a dry grain cultivar in the Sahelian

zone and has similar grain yield, morphology and

phenology as ‘Melakh’ but a different grain type.

‘ISRA-819’ has seed that are brown and large (average

individual seed weight of 230 mg), whereas ‘Melakh’

has seed that are white with a brown eye and of

moderate size (190 mg). While rough white grain is

a very popular grain type, many cowpea consumers in

Africa do prefer brown grain, and large grain are

widely appreciated (chapter by A.S. Langyintuo

et al.). ‘ISRA-819’ was derived from a cross made

in 1989 between ‘Melakh’ and a landrace widely used

in northern Senegal, ‘Ndiaga Aw’, which has smooth

brown grain having a seed weight of 180 mg. Selec-

tions were made to obtain an erect plant with the same

earliness and resistance to cowpea aphid, bacterial

blight and CABMV as ‘Melakh’ and large brown

grain. Unfortunately, ‘ISRA-819’ is more sensitive

to flower thrips than is ‘Melakh’. Consideration is

being given to releasing ‘ISRA-819’ for use in north-

ern Senegal where flower thrips are not a major pest

problem.

3.1.2. Lines ‘Sul 518-2’ and ‘ITP 148-1’ which

are adapted to the Savanna zone of Africa

‘Sul 518-2’ and ‘ITP 148-1’ were bred for use as dry

grain cultivars in the Savanna zone in a collaborative

program between SARI, Ghana and UCR. In 1987, Dr.

Marfo crossed a set of Ghanaian cultivars with two

heat-tolerant blackeye breeding lines bred by UCR.

Line ‘518-2’ was derived from ðTVu 4552 � CB5Þ�CB5 and ‘148-1’ was derived from ½ðPrima�TVu 4552Þ � CB5� � 7977. In the summer of 1988,

Dr. Marfo screened 12 F2 populations for heat toler-

ance at the UC Desert Research and Extension Center

near El Centro, CA. He selected single plants that had

heat tolerance during reproductive development and

produced many flowers and pods under very hot

conditions. In subsequent generations, he screened

the selections for their adaptation to the Savanna zone

in northern Ghana. He selected two stable lines that

flower early and have short cycles, and are moderately

resistant to cowpea aphid, bacterial blight and Striga:

‘Sul 518-2’ came from ‘Sumbrisogla’ � ‘518-2’, and

‘ITP-148-1’ came from ‘IT81D-1137’ � ‘148-1’.

‘Sumbrisogla’ is a traditional landrace from northern

Ghana, whereas ‘IT81D-1137’ had been bred by IITA

and released in Ghana. Lines ‘518-2’ and ‘ITP-148-1’

have produced relatively high grain yields compared

with other cultivars and local landraces in experi-

ment station and on-farm trials and have been recom-

mended for release as new cultivars in northern

Ghana.

114 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

3.2. Snap-type cowpeas

Opportunity is present for greater use of cowpea in

Africa and the USA as fresh immature pods (snap

types). Snap-type cowpeas should have succulent

tender pods with good taste. Some food processors

in the USA have traditionally included a portion of

snaps in their processed cowpea products. In early

years, immature pods of southern pea types of cowpea

cultivars were used for this purpose but they are not

very succulent and a cowpea cultivar was developed,

‘Snapea’, that has attractive, long, low-fiber pods

(Lorz and Halsey, 1964). ‘Snapea’ was the major

cultivar used by processors for the snap component

until the 1990s.

Cowpea lines bred for edible pod production in

India were evaluated, and five germplasm lines were

released that are adapted to subtropical California

(Patel and Hall, 1986). These lines are erect bush

types with early, synchronous flowering and some

heat tolerance. In a yield trial at the UC Riverside

Experiment Station during the summer of 1984, fresh

immature pod yields of these lines over a 45-day

harvesting season were 25–28 t/ha compared with

only 13 t/ha for ‘Snapea’. In the same trial a snap

bean (Phaseolus vulgaris L.) cultivar ‘Contender’

gave a green-pod yield of only 7 t/ha due to the hot

weather. Clearly, these snap-type cowpea germplasms

have the potential for use in areas and seasons that are

too hot for snap beans.

In1994, snap-typecultivar ‘Bettersnap’ was released,

andquickly replaced ‘Snapea’ (Fery andDukes, 1995a).

These USDA researchers then initiated work to incor-

porate the superior yield potential of the Asian cowpea

germplasm released by Patel and Hall (1986) into

a genetic background adapted to the southeastern

United States. A snap-type cultivar derived from a cross

between ‘Bettersnap’ and line ‘UCR-204’ likely will be

released by the USDA in 2002.

3.3. Green manure/cover crop cowpeas

Organic vegetable producers in the USA have dif-

ficulty providing sufficient nitrogen to their crops, and

controlling weeds and nematode pests. A nematode-

resistant cowpea cultivar grown in rotation could be

used to enhance the fertility of the soil and suppress

weeds and nematode pests. An ideotype was devel-

oped at UCR for a green manure/cover crop cowpea

cultivar for use in the continental United States. The

cultivar would have extreme sensitivity to photoperiod

(group 3 in the system of Ehlers and Hall, 1996).

When sown in late spring or early summer in the

continental United States, the cultivar would not begin

flowering until late September. Plants that flower late

should fix more nitrogen and produce more biomass

than early flowering cultivars. The shoots would be

incorporated into the soil in September. A vigorous

shoot type would be needed that suppresses weeds

under low planting densities. The cultivar should have

resistance to root-knot nematodes that is effective

against a broad range of nematode species and bio-

types (refer to Section 3.10). Seed of a photoperiod-

sensitive cowpea of this type only could be produced

in warm areas with short days, such as during the fall

in the Coachella Valley in California with sowing in

mid-August. The decreasing photoperiod in late Sep-

tember would cause initiation of flowers, and this

location stays warm permitting the crop to continue

to grow and produce seeds that would be harvested in

early December. Seeds of this cultivar should be small,

smooth and round to permit efficient production and

harvesting of large numbers of seed. Cultivars with

small round seeds could be threshed with minimum

damage to the seed using modified grain combines. In

contrast, large-seeded cowpeas require specialized

bean threshers if they are to be harvested with minimal

damage to the seed (Hall and Frate, 1996) and these

threshers are not widely available. The seed costs of

the ideotype would be low due to the low cost of

harvesting with modified grain combines and the fact

that few kilograms of seed would be required per

hectare due to the relatively large number of seeds

per kilogram.

An old cowpea cultivar, ‘Iron Clay’, is being grown

in the US as a green manure/cover crop in rotation

with vegetable crops. This cultivar has vigorous vege-

tative growth, late flowering when grown in the sum-

mer, and the Rk form of resistance to root-knot

nematodes. Unfortunately, the pods tend to open pre-

maturely under low-humidity conditions, such that

seed yields can be low when grown in the low-eleva-

tion deserts of California in the fall. Also, ‘Iron Clay’

has a tendency to produce seed lots with a high

proportion of hard seed when grown in California

(Ismail and Hall, 2002). Accessions in the UCR

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 115

germplasm collection have been screened to deter-

mine their suitability for use as green manure/cover

crop cultivars. ‘Speckled Purple Hull’, an heirloom

cultivar from the southeastern US, and ‘UCR 1340’, a

landrace from India, appear superior to ‘Iron Clay’

with respect to their fall seed production and seed

quality, and have similar summer biomass production

and nematode resistance as ‘Iron Clay’. Also, attempts

are being made to breed a cultivar with broader-based

resistance to root-knot nematodes and all of the other

desirable green manure/cover-crop traits. ‘IT89KD-

288’, a breeding line developed by IITA for grain

production in Africa, was shown by studies at UCR to

have the broader-based Rk2 resistance to root-knot

nematodes and all of the desirable green manure/

cover-crop traits except that it has large seed.

‘UCR’ 779, a landrace from Botswana, has all of

the desirable cover-crop traits and resistance to the

California biotype of the cowpea aphid but it is

susceptible to root-knot nematodes. Advanced breed-

ing lines have been developed by UCR from crosses

made between ‘IT89KD-288’ and ‘UCR 779’ or ‘Iron

Clay’ which combine several of the traits that can

enhance the value of a cowpea green manure/cover

crop cultivar for use in the continental United States.

3.4. Heat tolerance during reproductive

development

Cowpeas can be very sensitive to high night tem-

peratures during reproductive development (Hall,

1992, 1993). An experimental system was developed

in which cowpea plants could be subjected to different

higher night temperatures in the field, but similar day

temperatures, by enclosing plots at night (Nielsen and

Hall, 1985a). Using this system it was shown that pod

set and grain yield can be substantially decreased by

increases in night-time temperature during early flow-

ering (Nielsen and Hall, 1985b; Hall, 1993). In Cali-

fornia, heat-sensitive cowpea cultivars, such as ‘CB5’,

were shown to exhibit 4–14% decreases in pod set and

grain yield for each 8C increase in daily minimum

night temperatures above 16 8C (Nielsen and Hall,

1985b; Ismail and Hall, 1998). Low pod set was

associated with damage to anthers by high night

temperature occurring 9–7 days before anthesis which

resulted in male sterility (Ahmed et al., 1992). Two

weeks or more of consecutive or interrupted hot nights

during the first 4 weeks after germination also caused

complete suppression of floral bud development such

that plants do not produce any flowers (Ahmed and

Hall, 1993). The suppression of floral buds caused by

high night temperatures only occurred in long-day

conditions (Patel and Hall, 1990). Most cowpea acces-

sions either produced no flowers under hot long-day

conditions or if they produced flowers they did not set

any pods (Patel and Hall, 1990; Ehlers and Hall,

1996).

Breeding provides a solution to the problem of heat-

induced suppression of flower production and reduc-

tions in pod set (Hall, 1992, 1993). Genetic lines were

bred that have a recessive gene that confers heat

tolerance during early floral development permitting

the plant to produce flowers when subjected to heat

(Hall, 1993), and a dominant gene and some minor

genes that enhance the ability of the plant to set pods

during hot weather (Marfo and Hall, 1992; Hall, 1992,

1993). Realized and narrow-sense heritabilities of heat

tolerance during pod set were low at about 0.26 (Marfo

and Hall, 1992). An approach used to incorporate this

set of heat-tolerance genes involves subjecting F2

plants to high night temperatures and long days in

very hot field or greenhouse conditions and selecting

plants that produce flowers and set many pods per

peduncle (Hall, 1992, 1993, 2003). F2 selection fixes

the ability to produce flowers in most but not all lines;

however, ability to set many pods requires several

generations of family selection. Six heat-tolerant lines

were bred, including ‘CB27’, and were shown to have

greater grain yield in fields with hot weather in

California than six heat-sensitive lines with similar

genetic backgrounds, including cultivar, ‘CB5’

(Ismail and Hall, 1998).

In greenhouse conditions these heat-tolerant lines

had high grain yield under hot conditions in either the

long days typical of subtropical zones or the short days

typical of tropical zones (Ehlers and Hall, 1998).

However, grain yield of heat-sensitive lines was not

reduced as much by high night temperature under

short days as it was under long days (Ehlers and Hall,

1998). Under short-day tropical conditions in northern

Ghana and Senegal with very high night temperatures,

no differences in grain yield were observed between

the six heat-tolerant and six heat-sensitive lines but

grain yields were low for all lines (Hall et al., 2003).

Note that night temperatures usually are much hotter

116 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

in tropical locations than subtropical locations (com-

pare Bambey, Senegal (Fig. 10.8) and Louga, Senegal

(Fig. 10.5) with Riverside, CA (Fig. 10.4) and Bakers-

field, CA (Fig. 10.3) in Hall, 2001). These results may

be explained by two possibilities: (1) the set of heat-

tolerance genes is of no adaptive value under hot short-

day conditions due to a photoperiod interaction or (2)

to be effective the set of heat-tolerance genes needs to

be combined with other genes conferring adaptation to

tropical target production zones. For example, genes

for resistance to wet and dry pod rots are needed

because the UCR lines are very susceptible to these

diseases which are prevalent during the rainy season in

Ghana and Senegal. Some evidence indicates the sec-

ond possibility is more likely. In developing the heat-

tolerant lines used in the experiments in California and

West Africa, accessions ‘TVu4552’ and ‘Prima’ were

used as sources of heat tolerance because they are very

heat tolerant under hot long-day conditions (Warrag

and Hall, 1983; Patel and Hall, 1990; Ehlers and Hall,

1998). In growth chamber studies with high night

temperature and short days, ‘TVu4552’ had greater

pod set (57%) than heat-sensitive lines ‘CB5’ (21%)

and ‘7964’ (15%) (Mutters et al., 1989). Craufurd et al.

(1998) reported that ‘Prima’ had greater grain yield

under hot short-day conditions in growth chambers due

to its better ability to maintain peduncle and flower

production and pod set than ‘IT84S-2246’, which is

sensitive to heat under long days (Patel and Hall, 1990).

Apparently, incorporation of heat-tolerance genes that

enhance pod set will only increase grain yield if the

plants are sufficiently adapted to the environment that

the photosynthetic source is not limiting and the pods

become well filled. Also, embryo abortion must not be

excessive.

Under hot conditions embryo abortion can occur

with no seeds being produced at the end of the pods

such that the pods are ‘pinched’ (Hall, 2003). Two

lines appear to have heat-tolerance with respect to

embryo abortion, ‘B89-600’ from Senegal, and

‘TN88-63’ from Niger (Ehlers and Hall, 1998). Breed-

ing for this trait may be difficult because stresses such

as drought (Turk et al., 1980), and plant age can cause

pod pinching. Also, due to negative correlations, it is

necessary to select plants that have both well-filled

pods and many pods. Comprehensive descriptions of

cowpea breeding for heat tolerance may be found in

Hall (1992, 1993) and in the section on heat stress at

http://www.plantstress.com. A comparison between

cowpea and common bean (P. vulgaris L.) responses

to heat is presented in Hall (2003).

3.5. Chilling tolerance during emergence

Chilling temperatures can reduce seedling emer-

gence of warm-season annuals. For cowpea, soil

temperatures below 20 8C result in slow germination

(El-Kholy et al., 1997) and reduced maximal emer-

gence (Ismail et al., 1997). Incomplete plant emer-

gence can be a problem for cowpea farmers in

subtropical regions, such as the continental United

States, where the crop usually is sown in the spring

when the soil is still cool. Farmers sow cowpea early

because this practice can increase grain yield, provid-

ing crop emergence is adequate. Early sowing

increases the potential length of the growing season.

In California, this has allowed some cowpea crops to

accumulate two flushes of pods prior to the onset of

cool weather or rain at the end of the growing season

and produce very large grain yields (Gwathmey et al.,

1992a). Early sowing also can enable the crop to

escape high temperatures during the stages of flower-

ing that are very sensitive to heat (Hall, 1992; Ismail

and Hall, 1998). However, if crop emergence is sub-

stantially reduced by chilling, the farmer must decide

between two poor alternatives: either accept a reduc-

tion in grain yield due to an inadequate plant stand or

pay the increased costs associated with replanting the

crop at a date that is later than optimal.

Breeding for chilling tolerance during emergence

can solve this problem, and selection efficiency can be

increased if traits associated with chilling tolerance

are identified. Ismail et al. (1997) observed that a pair

of genetically similar cowpea lines (‘1393-2-1’ and

‘1393-2-11’) had differences in emergence under

chilling field conditions. They also demonstrated that

line ‘1393-2-11’ had greater emergence than line

‘1393-2-1’ in growth chambers with constant tem-

peratures of 15 8C. Based upon trait associations in

these lines and F1 hybrids, Ismail et al. (1997)

hypothesized that the chilling tolerance during emer-

gence of line ‘1393-2-11’ was conferred by two

independent and additive traits: the presence of a

specific dehydrin protein, and membrane integrity

associated with slow electrolyte leakage from seed

under chilling conditions. The dehydrin protein was

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 117

purified and partially characterized (Ismail et al.,

1999a). Studies with lines that were more isogenic

than the original pair showed that the presence of the

specific dehydrin protein was associated with an

increase in chilling tolerance during emergence, and

that the effect was not associated with differences in

electrolyte leakage (Ismail et al., 1999b). Allelic

differences in the structural gene encoding the dehy-

drin protein were described by Ismail et al. (1999b)

and had the same position on a cowpea genetic-

linkage map as the dehydrin protein presence/absence

trait (Menendez et al., 1997; Ouedraogo et al., 2002).

Genetic studies confirmed that the presence of the

dehydrin protein in seed is consistent with the action

of a single dominant gene (Ismail and Hall, 2002).

Few US cultivars contain significant quantities of

the dehydrin protein in their seed (Ismail and Hall,

2002); yet cowpeas grown in the continental United

States often are subjected to cool soil at sowing. This

suggests that cultivars used in the United States might

be improved by incorporating this specific dehydrin

into their seed. A non-destructive immunoblot assay of

individual seeds for the presence of this dehydrin

protein can be used in backcross breeding to effi-

ciently incorporate this trait (Ismail et al., 1999b).

UCR has bred lines that have both the dehydrin protein

to confer some chilling tolerance during emergence,

and the set of genes required to confer heat tolerance

during reproductive development (Ehlers et al.,

2000b). The possibility that chilling tolerance in cow-

pea is conferred by two independent traits (Ismail

et al., 1997) is important. It indicates that combining

the gene for presence of the specific dehydrin

protein with a gene(s) conferring membrane integrity

under chilling conditions may result in very strong

tolerance to chilling during emergence. Many cowpea

accessions have been screened and some with low

electrolyte leakage under chilling conditions, indicat-

ing membrane thermostabilitity, have been detected

(Ismail and Hall, 2002).

3.6. Delayed-leaf-senescence

Early erect cowpeas can be extremely sensitive

to mid-season drought (Thiaw et al., 1993). Lines

were discovered that have a delayed-leaf-senescence

(DLS) trait which enhanced adaptation to mid-season

drought in California by permitting plants to recover

after the drought and produce a second set of pods that

compensated for the loss in yield by the first flush

of pods (Gwathmey and Hall, 1992). The DLS trait

also enhanced yield potential by enabling plants to

more consistently produce two sets of pods under

long-season conditions (Gwathmey et al., 1992a).

The DLS trait appears to be conferred by a single

gene and may involve resistance to premature death

of cowpea caused by Fusarium solani (Mart) Sacc. f.

sp. phaseoli (Burke) Synd. & Hans., type A (Ismail

et al., 2000). In Senegal, plants with the DLS trait

remained alive after producing one flush of pods and

retained the ability to produce a second flush of pods

whereas control cultivars died (Hall et al., 1997b). An

ideotype was proposed to combine ability to withstand

mid-season and terminal droughts that consisted of

combining early flowering and an indeterminate plant

habit with the DLS trait. Studies in Senegal indicated a

plant of this type could begin flowering in about 35

days and produce 2 t/ha by 60 days followed by a

second flush of pods with the potential to produce an

additional 1 t/ha by 100 days from sowing. Also, an

early flowering cultivar with DLS would produce

more hay and may fix more atmospheric nitrogen than

an early flowering cultivar that does not have DLS.

DLS involves preferential partitioning of carbohy-

drate to stem bases (Gwathmey et al., 1992b) and

presumably roots, whereas earliness or heat tolerance

during reproductive development involves preferential

partitioning of carbohydrate to pods. Consequently, it

was hypothesized that these traits might interact in a

negative manner. This hypothesis was tested by breed-

ing sets of lines with and without DLS, and with and

without heat tolerance. The four sets of lines were

evaluated in California in soils with and without the

organism causing premature death of cowpea, and in

hot and moderate temperature field conditions (Ismail

et al., 2000). It was shown that the DLS and heat-

tolerance traits can be effectively combined and that

they would have beneficial effects on grain yield in

specific circumstances with only small detrimental

interactive effects.

In California, the syndrome involved in premature

death of cowpea has increased in some fields where

cowpea has been grown several times, even with

alternate year rotation to other crop species. Fields

of this type can be used to select for DLS. Plants are

selected that have not senesced after producing a first

118 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

flush of pods and still have substantial green leaf area.

It is necessary, however, to select plants that have both

substantial green leaf area and many mature pods to

avoid selecting a type of DLS with little agronomic

value. For example, plants in breeding nurseries that

have very low pod loads due to male sterility or some

other factor typically stay green for a long period but do

not necessarily have the DLS trait that is agronomically

useful. Family selection for DLS is more effective than

single-plant selection because plants at the end of row

sections tend to have more green leaf area, after

accumulating the first flush of pods, than plants grow-

ing under more competitive conditions.

3.7. Genotypic differences in stable carbon

isotope discrimination

Stable carbon isotope composition of leaves reflects

the extent of their discrimination (D) against carbon

dioxide with the heavier isotope of carbon (13CO2)

during diffusion and photosynthetic fixation. For cow-

pea, it was shown that D is negatively correlated with

intrinsic, time-integrated water-use efficiency, where

water-use efficiency (W) is the ratio of plant biomass

production to transpiration (Ismail and Hall, 1992).

Possible use of selection for D to improve adaptation

to drought has been considered (Hall et al., 1994a).

Genotypic differences in D were discovered in cowpea

with similar rankings in both well-watered and dry

environments (Hall et al., 1990). Realized heritabil-

ities were low indicating selection for D may only be

effective with families in advanced generations

(Menendez and Hall, 1995, 1996). Genetic correla-

tions were observed indicating that selection for low Dmight also result in late flowering (Menendez and

Hall, 1995), and a low harvest index where HI is grain

yield/total shoot biomass (Menendez and Hall, 1996).

Genetic selection studies with cowpea gave positive

correlations between grain yield and D under both

well-watered and water-limited field conditions (Hall

et al., 1997b; Condon and Hall, 1997). For well-

watered environments, these results may be explained

if high D is associated with greater stomatal conduc-

tance which in turn results in greater photosynthesis,

due to diffusion effects, which results in greater

biomass production and thus greater grain yield.

The results are somewhat contrary to expectations

for water-limited environments in that they indicate

low W might be adaptive which appears counter intui-

tive. These results may be explained, however, by the

hypothesis that the genotypes with greater grain yields

had deeper roots that obtained more soil water causing

them to have more open stomata and thus greater D. In

this case the greater photosynthesis would have had to

overcome the inefficiency inherent in a lower W if the

plants were to produce more biomass. In the genetic

selection studies the plants with higher D and greater

grain yield under water-limited conditions did produce

more total shoot biomass (Hall et al., 1997b; Condon

and Hall, 1997).

Comparisons of cowpea genotypes across different

subtropical and tropical climatic zones in California,

Texas and Senegal indicated a tendency for adaptation

to be associated with high D (Hall et al., 1994b). Also,

some genotypes with relatively high D and high grain

yield in one environment, had relatively low D and low

grain yield in a radically different environment. This

indicates that different sets of genes may be needed to

confer high D and high grain yield in radically dif-

ferent environments. An alternative hypothesis is that

the best lines under water-limited conditions had high

yields due to their earliness and high HI, and that

their high D was due to positive genetic correlations

with these traits. If this hypothesis is valid, an addi-

tional increment of grain yield may now be gained in

water-limited environments but not in well-watered

environments by selecting for low D.

For well-watered environments the genetic selec-

tion experiments and theoretical analyses support the

argument that selection for high D might enhance

grain yield (Hall et al., 1997b; Condon and Hall,

1997). Consequently, selection for high D may be

effective for enhancing yield potential, but it still is

not clear for water-limited environments whether one

should select for low D or high D.

3.8. Genotypic differences in harvest index

From studies with contrasting genotypes at extre-

mely high and moderate plant densities, Kwapata and

Hall (1990) hypothesized that it may be possible to

enhance grain yield of cowpea in favorable environ-

ments by selecting for high harvest index in early

generations at wide spacing, and then growing the

crop at high plant densities. ‘CB27’ fits the ideotype

sought by this approach in that it has a very high

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 119

harvest index and is very productive at high plant

densities (Ismail and Hall, 2000).

Global atmospheric carbon dioxide concentration is

increasing and temperatures may be increasing, espe-

cially at night (Hall and Ziska). An indication of the

traits needed to improve cowpea adaptation to these

expected changes in climate was obtained by compar-

ing cowpea genotypes with differences in heat toler-

ance growing under normal or elevated atmospheric

carbon dioxide and moderate or high night tempera-

tures (Ahmed et al., 1993). Heat-tolerant genotypes

had enhanced pod set and higher HI and were more

responsive to elevated carbon dioxide under both hot

and moderate night temperatures than heat-sensitive

genotypes. Based on these results it was hypothesized

that for cases where elevated carbon dioxide increases

the photosynthetic source, genotypes are needed with

a greater HI to maintain an optimum balance between

this source and the reproductive sink (Hall and Ziska,

2000). Selection for higher HI may be particularly

effective for those production zones where the repro-

ductive sink is more severely stressed than is the

photosynthetic source capacity, such as in some hot

environments in California (Ismail and Hall, 1998).

When determining HI in a breeding program it was

shown that it may be effective with plants grown at

wide spacing (Kwapata and Hall, 1990), which means

that it may be effective with single plants and in earlier

generations than can be used for effectively determin-

ing grain yield. Also, HI, defined as the ratio of grain

yield to total shoot biomass, may be effectively deter-

mined at final harvest after plants have senesced with

no need to collect leaves (Kwapata and Hall, 1990).

Selecting lines for high HI in this way would, however,

tend to select early erect lines and eliminate lines that

have the DLS trait.

3.9. Genotypic differences in rooting and plant

water- and nutrient-relations traits

Rooting traits are important in dry infertile soils,

such as occur in the Sahelian zone, but screening for

them can be difficult. A technique was developed that

can quantify differences in rate and thus extent of

rooting by many cowpea lines under field conditions

(Robertson et al., 1985). This technique consists of

growing cowpea lines on soil moisture stored deep

in the soil profile. An herbicide is placed in a narrow

band deep in the soil in-between pairs of rows of plants

having the same genotype. Plants are scored daily for

leaf herbicide symptoms as an indicator of the time

taken for roots to grow and reach the herbicide. When

a set of cowpea accessions was screened using this

technique, some genotypes differed in the number of

days taken to exhibit foliage herbicide symptoms. In a

separate experiment, genotypes exhibiting herbicide

symptoms earlier were shown to extract more water

from deep in the soil, indicating they had more rapid

root development (Robertson et al., 1985). A diverse

set of 32 cowpea accessions was screened with this

technique at Riverside, CA and shown to differ sig-

nificantly in number of days to first herbicide symp-

toms (Hall and Patel, 1985). Only prospective parental

lines and stable breeding lines can be screened using

this method because the herbicide kills the plants. This

technique has been used to screen peanut genotypes

for differences in rooting in Senegal (Khalfaoui and

Havard, 1993). Genotypic differences in days to first

appearance of herbicide symptoms were detected.

Peanut genotypes that exhibited symptoms earlier,

and presumably had faster rates of root development,

had longer cycle lengths from sowing to maturity

(Khalfaoui and Havard, 1993) and the same associa-

tion probably occurs in cowpea.

Drought-induced osmotic adjustment has been pro-

posed as a mechanism that may enable cultivars to

develop deeper root systems under water-limited con-

ditions. A diverse set of 100 cowpea accessions was

screened under extreme soil drought in field condi-

tions and found to exhibit only small differences in

leaf solute potential ranging from �1.1 to �1.3 MPa

(Shackel and Hall, 1983). Cowpea has exhibited very

little drought-induced osmotic adjustment (0–0.4 MPa

in Shackel and Hall, 1983) compared with peanut

(0.3–1.6 MPa in Turner et al., 2000).

Performance of cowpea in infertile soils is influ-

enced by symbiotic associations with rhizobia, asso-

ciations with fungi resulting in mycorrhizal roots and

root architecture. Measurement of ureides in xylem

sap was shown to be an effective technique for quan-

tifying recent nitrogen fixation in cowpea under

field conditions (Elowad et al., 1987; Elowad and

Hall, 1987). Cowpeas were shown to have substantial

nitrogen fixation in several field conditions where this

was measured in both California (biological N fixation

of about 200 kg/ha) and Senegal. Cowpeas with

120 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

mycorrhizal roots were shown to be more effective in

uptake of phosphate, zinc and copper from deficient

soils than cowpeas with non-mycorrhizal roots (de

Faria, 1984; Kwapata and Hall, 1985). It is likely,

however, that cowpea grown in most parts of the world

already have mycorrhizal roots so there may not be

much merit in trying to modify this trait. Root hairs are

thought to be important in uptake of phosphate from

infertile soils. A growth pouch technique (Omwega

et al., 1988) was shown to be effective for screening

cowpea for differences in extent of root hairs. Using

the growth pouch technique, it was shown that the new

cultivar ‘Melakh’ and a widely grown line developed

by IITA, ‘IT82E-18’, had more frequent and longer

root hairs than other accessions that were screened

(W.C. Matthews and J.D. Ehlers, unpublished).

3.10. Broad-based resistance to root-knot and

other plant-parasitic nematodes

Cowpea resistance to root-knot nematodes (Meloi-

dogyne spp.) was one of the first cases of resistance to

nematodes described in the scientific literature with

Webber and Orton (1902) reporting that the cultivar

‘Iron’ exhibited low levels of root-galling in field plots

infested with root-knot nematodes. Since that time,

cowpea cultivars with nematode resistance have been

developed by several breeding programs. Most of

these resistant cultivars have the same resistance gene,

designated Rk by Fery and Dukes (1980). Fery and

Dukes (1980) determined through genetic studies with

resistant cultivar ‘Mississippi Silver’ that the resis-

tance is inherited as a single dominant gene, and that

gene Rk was effective against populations of three

species of root-knot nematode, M. incognita, M. java-

nica, and M. hapla. The Rk gene and genes conferring

resistance to Fusarium wilt and various viral diseases

were found to map to different locations on a genetic-

linkage map of cowpea (Ouedraogo et al., 2002). The

Rk resistance is highly effective in blocking both

nematode reproduction on roots and nematode-

induced root-galling for nematode populations that

are avirulent (Swanson and Van Gundy, 1984; Roberts

et al., 1995). The resistance enables cowpea plants

to exhibit normal growth and yield in nematode-

infested fields. The blocking of nematode reproduc-

tion also can enhance the growth of subsequent crops,

because it suppresses or reduces nematode population

densities in the soil. This is an attribute that makes

resistant cowpeas attractive as green manure/cover

crops (Section 3.3). Root-knot nematode resistance

also can reduce the extent of the Fusarium wilt disease

in cowpea by limiting points of entry for the pathogen

(Roberts et al., 1995).

In the 1950s, M. javanica populations from Califor-

nia were found that attacked blackeye cultivars with the

Rk gene, including ‘CB5’ (Thomason and McKinney,

1960). More recently a significant number of cowpea

fields have been found in California that contain

resistance-breaking populations of M. javanica and

M. incognita that are virulent on cultivars such as

‘CB46’’ that have the Rk gene (Roberts et al., 1995).

Some of these fields have been cropped frequently with

cowpea cultivars carrying the Rk gene, apparently

resulting in selection for Rk-virulence. These virulent

forms induce a stronger susceptible reaction on plants

with the Rk gene than they do on plants that do not

carry this gene.

The need for a broader-based type of root-knot

nematode resistance in cowpea prompted efforts

to identify novel resistance traits in cowpea. Earlier,

Fery and Dukes (1982) had identified a recessive allele

at the Rk locus that conferred an intermediate level

of resistance, indicating that other resistance factors

existed in cowpea. Subsequently additional superior

resistance was identified in three germplasm lines

(‘US-566’, ‘US-567’ and ‘US-568’) by Fery and

Dukes (1995b) and this resistance may be linked to

the Rk locus (Fery et al., 1994). Another form of

resistance associated with the Rk locus, designated

Rk2, was reported by Roberts et al. (1996) that

confers a very high level of resistance to a broad

range of Meloidogyne spp. isolates, including those

M. javanica and M. incognita populations that are

virulent on gene Rk. The Rk2 resistance was identified

in line ‘IT84S-2049’ (Table 1), a breeding line devel-

oped at IITA, and is inherited as a single dominant

gene. Allelism tests indicated that Rk2 is another

dominant allele at the Rk locus, or is very tightly

linked to gene Rk, within 0.17 map units. The Rk2

resistance could be effective in all nematode-infested

sites in California cowpea production areas and most

other cowpea production areas worldwide.

An additional source of resistance to M. javanica

and M. incognita populations that are able to develop

on cowpea cultivars with gene Rk has been discovered.

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 121

A recessive gene, designated rk3, that is independent

of the Rk locus, and when combined with gene Rk,

provides high levels of broad-based resistance through

an additive gene action was described by Ehlers et al.

(2000c). The rk3 resistance was identified in a heat-

tolerant breeding line, and accession ‘TVu 4552’ in its

pedigree was identified as the probable donor parent of

the recessive gene (Ehlers et al., 2000c). On its own,

rk3 confers only a moderate level of nematode resis-

tance, weaker than that conferred by Rk. The broad-

based form of resistance conferred by the Rk plus rk3

gene combination (Table 1) is present in the recently

released cultivar ‘CB27’ (Ehlers et al., 2000a). Pyr-

amiding resistance genes may provide a more durable

resistance than has been seen with gene Rk. Progress is

being made at UCR to combine both Rk2 and rk3 into

blackeye backgrounds to produce alternative forms of

broad-based durable resistance (Ehlers et al., 2001;

Roberts et al., 2001). For example, line ‘00-11-1430’

has the Rk2 resistance in a California blackeye genetic

background (Table 1). Resistance of several cowpea

cultivars and breeding lines developed by CRSP

research and mentioned elsewhere in this chapter is

described in Table 1. Note that all of the CRSP

cultivars developed for Cameroon and Senegal and

the advanced breeding lines developed by the CRSP

for Ghana are susceptible to root-knot nematodes.

In 2001, high levels of another type of plant-para-

sitic nematode, Scutellonema cavenessi Sher, were

detected in soil in several fields in the main cowpea

production zone of Senegal where cowpea growth

often is stunted (Mame Birame and Ndiaga Cisse,

unpublished). The ISRA scientists also observed that

multiplication of this nematode was substantial on

eight widely used cultivars and breeding lines but

very low on two old cultivars. Further studies should

be conducted in West Africa to determine the extent to

which plant parasitic nematodes are reducing cowpea

production, and whether this problem can be solved

by breeding nematode-resistant cowpea cultivars.

Table 1

Cowpea resistance to M. javanica and M. incognita determined as extent of nematode egg mass production using growth pouches (Roberts

et al., 2001)a

Cultivar or lineb Origin and

adaptation

Type of cultivar Egg masses

per plant

Resistance classification for

M. javanica and M. incognita

IT84S-2049 Nigeria Dry grain 51 Very strong Rk2 resistance

00-11-1430 California Dry grain 71 Very strong Rk2 resistance

UCR 193 California Snap pod 86 Very strong

India Snap pod

IT89KD-288 Nigeria Grain/hay 90 Very strong

California Cover-crop

Cameroon 7-29 Cameroon Grain/hay 94 Very strongc

CB27b California Dry grain 146 Strong Rk þ rk3 resistance

UCR 97-15-33 California All-white grain 215 Moderate

C93W-125B Cameroon Sweet grain 226–283d Moderate to susceptiblec

CB46b California Dry grain 229 Moderate Rk resistance

UCR 779 California Cover-crop 260 Susceptible

ITP 148-1 Ghana Dry grain 262 Susceptiblec

Sul 518-2 Ghana Dry grain 291 Susceptiblec

Vyab Cameroon Grain/hay 292 Susceptiblec

CRSP Niebeb Cameroon Grain/hay 300 Susceptiblec

B22xVal Ghana Dry grain 307 Susceptiblec

Melakhb Senegal Fresh/dry grain 334 Susceptible

CB3b California Dry grain 359 Susceptible check

Lori Niebeb Cameroon Grain/hay 370 Susceptiblec

Mourideb Senegal Dry grain 404 Susceptible

a Only M. javanica data are presented.b These entries are cultivars; the others are advanced breeding lines.c Response to M. incognita has not been determined at UCR for these cultivars and lines.d Range indicates variation observed within subline mean values.

122 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

A cowpea cultivar that reduces multiplication of

S. cavenessi might enhance the overall cropping

system because all major crop species grown in the

Peanut Basin of Senegal can be damaged by this plant

parasitic nematode.

3.11. Resistance to flower thrips

Flower thrips are the major insect pest of cowpea in

the Savanna zone and a moderate insect pest in the

Sahelian zone of Africa (Hall et al., 1997a). In Senegal,

‘Melakh’, accession ‘58-77’ and breeding line ‘ISRA-

2065’ have shown partial resistance to flower thrips

in trials comparing grain yield with and without

insecticide applications. The resistance was slightly

greater than that of ‘TVx 3236’. In addition, accessions

‘Bun 22’ and ‘Sanzibili’ from Ghana have shown some

resistance to flower thrips in northern Ghana (Abdulai

Baba Salifu, unpublished) and Senegal in terms of

fewer flower thrips being present in unsprayed condi-

tions than were present on either ‘TVx 3236’ or its

progenitor ‘TVu 1509’.

3.12. Resistance to cowpea aphid

Cowpea aphid is a moderately severe pest of cowpea

in the Sahelian and Savanna zones of Africa and the

USA. Seedling resistance to cowpea aphid was identi-

fied in cowpea accessions by IITA and other organiza-

tions 20–30 years ago. Aphid-resistant lines from IITA

include ‘TVu 86’, ‘TVu 801’, and ‘TVu 3000’ (Inter-

national Institute of Tropical Agriculture, 1984). Since

that time, these sources of resistance have been used

extensively in cultivar development at IITA (Singh

et al., 1997) and in several national cowpea breeding

programs in Africa, including in the development of

‘Melakh’ for the Sahelian zone (Section 2.1.3). Other

sources of cowpea aphid resistance include those dis-

covered in Kenya (Pathak, 1988) and India (Chari et al.,

1976). Resistances present in the African and Indian

lines have not been effective against those biotypes of

cowpea aphid occurring in California that have been

tested (Abdalla, 1992; Martyn, 1991).

There has been extensive screening to identify

resistance to California biotypes of cowpea aphid.

‘IT84S-2049’, mentioned in Section 3.10 as a source

of broad-based root-knot nematode resistance, was

observed to have moderate resistance to cowpea

aphid in greenhouse seedling tests (Abdalla, 1992).

However, the extent of the resistance has varied in

several subsequent seedling screening tests. This in-

consistency of results could have been due to different

aphid biotypes being used in the tests or effects of

differences in experimental methods, such as age of

the seedling at infestation or greenhouse environmental

conditions.

More recently, ‘UCR 5038’ (‘TVu 8381’), a sesqui-

pedialis type of cowpea exhibited resistance to cowpea

aphid with adult plants in field conditions and in

greenhouse seedling tests (Ehlers et al., 1996). This

resistance is being transferred into blackeye-type

cowpeas adapted to California. Lines from second

backcross materials have shown resistance in the field.

‘UCR 5038’ and the backcross lines derived from

‘UCR 5038’ do become infested under field condi-

tions, but the extent of the infestation is limited and

after several days the plants resume growth, while

susceptible cowpeas are killed (Ehlers et al., 2000b).

The resistance of ‘UCR 5038’ may be an inducible

form of adult-plant resistance that is different from

the more typical seedling-stage resistance that has

been used by breeding programs in Africa.

In addition, ‘UCR 779’, a landrace from Botswana,

and IITA breeding lines ‘IT93K-503-1’ and ‘IT93K-

2046’ were identified as having effective resistance to

cowpea aphid in field conditions in California (Ehlers

et al., 1997, 2000b). The resistance was similar to, but

even more effective (i.e. quicker regrowth) than the

resistance observed in ‘UCR 5038’ and its backcross

derivatives. Selected breeding lines developed from

crosses between cowpea aphid susceptible blackeye

cultivars and resistant line ‘IT93K-2046’ have shown

strong resistance in the field similar to that observed

for ‘IT93K-2046’. Crosses have been made to transfer

the cowpea aphid resistance of ‘IT93K-503-1’ and

‘UCR 779’ to green manure/cover crop and dry grain

types of cowpeas adapted to California. It should be

noted that early season applications of insecticides

to control cowpea aphids can destroy populations of

beneficial insects, resulting in severe outbreaks of

other insect pest species later in the season. The

resistance available in cowpea may be strong enough

to eliminate the need for using insecticides to control

cowpea aphid, thus reducing the need for additional

insecticide applications later in the season to control

other insect pests.

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 123

3.13. Resistance to lygus bug

Lygus bug (Lygus hesperus Knight) is the major

insect pest of cowpea in California. It is a mobile,

polyphagous pest that can devastate cowpea and other

crops, such as cotton. It is a sucking insect that feeds by

inserting its stylet into plant tissues, and secreting

digestive enzymes into the feeding site. These enzymes

cause necrotic lesions and abscission of developing

floral buds severely reducing flower and pod produc-

tion and grain yield. Lygus feeding on developing seeds

within pods cause brown necrotic lesions to develop on

the grain drastically reducing their market value.

Substantial effort has been devoted to searching for

resistance to lygus bug but progress has been moder-

ate. Thousands of accessions, cowpea breeding lines,

and lines developed from crosses between wild and

cultivated cowpeas have been evaluated for resistance

to this pest. Several screening methods have been

employed to try to detect strong resistance to this

pest, including caged plant techniques in the field and

greenhouse, seedling screening techniques, and field

screening in ‘hot-spot’ locations. Several lines have

been identified that exhibit less yield loss and grain

damage due to lygus feeding than the standard cultivar

‘CB46’. For example, breeding line ‘96-11-27’, which

was developed from a cross between wild � cultivated

cowpeas, had lygus caused yield losses of 14 and 13%,

while ‘CB46’ had losses of 27 and 29% in 1999 and

2000, respectively (Ehlers et al., 2000b). Also, sig-

nificant differences among lines for percentage of seed

damaged by lygus have been observed. ‘IT95M-43’, a

wild � cultivated line developed at IITA, had only 7%

damaged seed, while ‘CB46’ had 22% of its seeds

damaged by lygus. Line ‘96-11-27’ also had fewer

seed damaged by lygus (6 and 12%), than ‘CB46’

(14 and 22%), for trials conducted in 1999 and 2000,

respectively (Ehlers et al., 2000b). Similar results

were observed in 2001 (Ehlers et al., 2001). However,

as of early 2002, only a moderate level of resistance to

lygus-induced damage to flower buds and grains had

been discovered.

3.14. Resistance to cowpea weevil

Cowpea weevil is a major storage pest of cowpea in

all places where cowpeas are grown. Pod and seed

resistance to cowpea weevil have been bred into

cowpea cultivars ‘CRSP Niebe’ and ‘Lori Niebe’.

Pod resistance was associated with tough pod walls

and a tendency for the pods to remain intact during

storage. Partial seed resistance was bred into these

cultivars and ‘Mouride’ using resistance from cowpea

accession ‘TVu 2027’. Selection for resistance mainly

involved bioassays in which adult weevils were placed

on a seed lot in a Petri dish for about 24 h. About 5

days after infestation the number of weevil eggs on the

seed was determined. About 30 days after infestation,

the number of holes where adult weevils had emerged

from the seeds was determined permitting calculation

of the percentage of eggs producing emerged adults.

The percentage adult emergence can be much less

for resistant seedlots (about 20%) than susceptible

seedlots (more than 60%).

An improved method for detecting resistance to

cowpea weevil has been developed. Larval bruchids

living within cowpea seeds generate ultrasonic signals

when they strike hard internal tissues of the seed as

they feed (Shade et al., 1990). These signals are

exceedingly weak, but a biomonitor was developed

that detects and records them (Shade et al., 1989). The

Purdue biomonitor revealed dramatic differences in

feeding behavior when C. maculatus larvae feed in

seeds of resistant cowpea line ‘TVu 2027’ compared

with susceptible ‘CB5’ seeds (Fig. 1 in chapter by

Murdock et al.). Larval feeding rates were similar for

both genotypes during the first few days of larval life,

but decreased markedly during the second instar in

seeds of ‘TVu 2027’. Time required to reach the third

instar was prolonged in ‘TVu 2027’. In resistant seed,

larvae often lingered on for long periods in the second

or third instar, feeding only occasionally and spora-

dically, with some individuals surviving as long as 4

months but not completing their development.

The procedure was improved by developing a multi-

channel-biomonitor that allowed many seeds to be

studied simultaneously and software essential to

acquiring, managing, and analyzing large amounts

of data. This biomonitor system can screen seed of

many cowpea genotypes per unit time because it takes

only a few minutes of monitoring to determine with

95% confidence whether a seed is resistant or suscep-

tible to bruchids (chapter by Murdock et al.). Several

seeds from each of the many thousands of accessions

in the IITA cowpea germplasm collection could prob-

ably be screened in a year using a single Purdue

124 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

biomonitor operated by a single technician. This

screening probably should be done because to-date

only a single source of cowpea weevil resistance,

namely that of ‘TVu 2027’(Singh and Jackai, 1985),

has been detected and this resistance only is partially

effective. Other bruchid-resistant accessions, ‘TVu

11952’ and ‘TVu 11953’, appear to have the same

resistance genes as ‘TVu 2027’ (Kitch, 1987). The

discovery of an ecotype of C. maculatus from Nigeria

that is capable of overcoming the resistance in ‘TVu

2027’ (Shade et al., 1999) shows how risky it is for plant

breeders to rely on this single source of resistance. In

earlier years, when bruchid resistance in ‘TVu 2027’

was discovered, seed of a large number of cowpea

accessions (about 12,000) in the IITA collection was

screened (Singh and Jackai, 1985). But re-screening of

these accessions is justified because a batch method was

used during the original screening. Batch screening,

because it takes an average of the results with about 40

infested seeds, could easily have resulted in individual

highly resistant seeds being overlooked. Accessions in

cowpea germplasm collections do exhibit some genetic

variability because of the occasional outcrossing that

occurs in cowpea and other reasons. The biomonitor

screens individual seed in a non-destructive manner and

thus is very effective for searching for rare forms of

resistance. The ability of the biomonitor to rapidly and

non-destructively screen large numbers of seeds also

make it a useful tool for cowpea breeders who wish to

incorporate bruchid resistance into their cultivars.

3.15. Grain hardness, size and density

When grain of most cowpea cultivars are placed in

water at room temperature they rapidly imbibe water

and after a few hours have swollen substantially. Grain

characterized as being ‘hard’ does not imbibe appreci-

able water for at least 24 h. Hard grain is undesirable

to consumers using dry grain because it can take

longer to cook. Also, hard grain is undesirable for

commercial US food freezing companies because

processing operations can occur so rapidly that hard

seed do not have sufficient time to imbibe water. Two

of the major pinkeye-type cultivars traditionally pro-

cessed by US freezing companies, ‘Pinkeye Purple

Hull-BVR’ and ‘Coronet’, have produced hard seed

under some production conditions resulting in some

hard seed being included in the frozen pack retailed to

American consumers, which is undesirable. Hard

grain also is undesirable from a seed standpoint in

that it results in delayed emergence and non-uniform

plant stands. Screening of grain produced in a field

nursery at Riverside, CA showed that the frequency of

hard grain was high among 140 landraces from the

Mediterranean (48%) and 61 US cultivars (21%) but

low among 59 tropical accessions from Africa (3%)

(Ismail and Hall, 2002). The nursery experienced hot,

low humidity conditions during grain maturation, and

this probably enhanced the expression of the hard-

grain trait.

Many cowpea consumers prefer grains larger than

200 mg but this trait is difficult to breed for because

most cowpea accessions, breeding lines and even

cultivars have individual seed weights less than

200 mg. This is a greater problem in the tropics where

the higher night temperatures accelerate seed devel-

opment causing the grain to be smaller than in sub-

tropical zones (Nielsen and Hall, 1985b; Hall, 2003).

For example, the average individual seed weight of

‘CB5’ varies from 240 mg under cooler conditions in

California to 190 mg in Senegal. There is a tendency

for large grain to be more fragile and become damaged

during threshing and cleaning. Grain with a smaller air

gap between the cotyledons would have higher density

and may exhibit less damage during threshing and

cleaning. A simple rapid method is available for

measuring grain density (Wessel-Beaver et al., 1984).

Genotypic differences in grain density were detected

using this method that were highly heritable and con-

sistent with control by two loci (Robertson, 1985).

California cultivar ‘CB46’ has a higher grain density

than other cultivars that have been studied. For exam-

ple, in a yield trial at the UC Kearney Research and

Extension Center, CA in 1993, genotypic differences in

grain densities were observed: 1.05 g/cm3 for ‘CB46’,

1.03 g/cm3 for ‘CB27’, 1.00 g/cm3 for ‘CB88’ and

0.99 g/cm3 for ‘CB5’. These small differences in seed

density were significant (the LSD0.05 was 0.02) and may

be important because ‘CB46’ has exhibited more resis-

tance to the mechanical forces that can occur during

mechanical harvesting than either ‘CB88’ or ‘CB5’.

3.16. All-white grain

Widespread consumption of convenience foods

containing significant amounts of cowpea would

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 125

substantially increase the demand for and value of

cowpea grain. Flours and batters produced from cowpea

grain have unique properties such as gelation, cohesion,

and the ability to hold air when whipped; these proper-

ties are useful in the manufacture of value-added foods

(Prinyawiwatkul et al., 1996). The adhesive properties

of cowpea grain can hold other ingredients together

without the need to add eggs. US food processors are

considering developing several new value-added foods

based on cowpea. Possible products include ‘bean

chips’ (Kerr et al., 2001) and soy-free, egg-free vege-

tarian ‘burgers’. A significant market may be present for

some of these products because some people have an

allergic reaction to foods containing soybean proteins.

The fat and water binding, and heat-induced gelation

properties of cowpea flour also are beneficial for some

meat-based applications and cowpea flour is being

considered for use as a meat extender (Prinyawiwatkul

et al., 1993, 1997).

In West Africa and Brazil, cowpea already is pro-

cessed into an array of traditional foods, such as

‘akara’, that are delicious yet virtually unknown out-

side these regions. Akara is a fried ball of whipped

(aerated) cowpea paste that usually is seasoned with

chopped fresh peppers (hot or mild) and other spices

or seasonings. Traditionally akara paste is prepared by

soaking cowpeas followed by manually removing seed

coats and then milling the wet grain. Dry milling of

whole grains would be much more efficient than the

traditional process, which requires much labor, and

dry flour can be stored longer than can wet paste. Dry

milling would make possible the development of

ready-to-cook cowpea flour mixes for producing akara

or for use in other products such as bean chips, or as

protein-rich supplements for breads or other cereal

products. For most processed products that use cow-

pea flour milled from whole grain, all-white grain may

be more desirable than pigmented grain.

‘Bambey 21’ was developed by ISRA, Senegal

(Sene and N’Diaye, 1974) and is different from most

other cowpeas in that it lacks pigments in the seed and

has all-white grain. Other cowpeas that have white

seed coats also have a pigmented ring around the

hilum that produces a colored ‘eye’. The ‘colorless’

grain trait in ‘Bambey 21’ is controlled by a single

recessive gene (J.D. Ehlers, unpublished). All-white

grain lines that are similar to California blackeye

cultivars in their agronomic type and yield potential

have been developed from crosses between ‘Bambey

21’ and several California blackeye lines. Six lines

were yield tested in two trials for the first time in 2001

and had reasonable yields that were only slightly less

than those of ‘CB46’ and ‘CB27’ (Ehlers et al., 2001).

Grain of line ‘97-15-33’ was found to be highly suited

for the production of akara (Section 4 and McWatters

et al., 2001).

3.17. ‘Sweet’ grain

Farmers who participated in the IRAD, Cameroon

cowpea breeding program noted that a specific line

bred by the program, ‘C93W-24-125B’, tasted ‘sweet’

or ‘good’ (Kitch et al., 1998). Line ‘C93W-24-125B’

was shown to have a sucrose content of about 6%

compared to about 2% for typical cowpea cultivars

(Kitch et al., 2003 and Table 2). For over 3 years,

Cameroonian farmers consistently chose this line as

being one of their favorites. Triangular taste panel

tests conducted at Purdue also showed that US tasters,

who were generally unfamiliar with cowpeas, chose

this sweet cowpea more than 80% of the time. Infor-

mal taste tests conducted at UCR indicated that the

‘Hopping John’ dish made using ‘C93W-24-125B’

had a more pleasant ‘nutty’ taste and firmer texture

than the same dish made with California blackeye-

type cultivars, but that ‘C93W-24-125B’ did not have

a distinct sweet flavor like a fresh garden pea (Pisum

sativum L.).

Another sweet cowpea breeding line, ‘KVx 61-1’,

was recently developed by the national breeding pro-

gram in Burkina Faso (Issa Drabo, unpublished) which

had an average sucrose content of 5% with seed to

seed variation ranging from 3 to 8% (L.L. Murdock,

unpublished). A Senegalese sensory panel preferred

the taste of ISRA breeding line ‘283 N’ over three

cultivars, ‘CB5’, ‘Bambey 21’ and ‘Mouride’ because

of the ‘sweet’ taste of ‘283 N’ (ITA 1990 Report,

unpublished). Apparently, the ‘sweet’ trait could be a

valuable flavor trait for enhancing the desirability of

cowpea cultivars for traditional cowpea consumers

and consumers to whom cowpea is less familiar.

Sub-lines of ‘C93W-24-125B’ have been developed

from seed of individual plants that have greater

sucrose content in their seed than the average for

the population. The sub-lines have been crossed with

‘CB27’, which has a greater than normal sucrose

126 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

Table 2

Grain quality traits of cowpea cultivars and linesa

Cultivar or

lineb

Origin and

adaptation

Seedc

(mg)

Ashd (% dry

weight basis)

Proteine (% dry

weight basis)

Sucrosef (% dry

weight basis)

Raffinosef (% dry

weight basis)

Stachyosef (% dry

weight basis)

Thiaming

(mg kg�1)

Riboflaving

(mg kg�1)

Niacing

(mg kg�1)

Mouride Senegal 199 3.36 23.0 1.48 0.65 3.76 1.09 2.58 32.7

Melakh Senegal 243 3.35 23.4 1.46 0.57 3.44 0.90 2.37 25.2

Bambey 21 Senegal 267 3.53 24.7 2.23 0.60 4.34 1.02 2.34 31.6

Ndiambour Senegal 190 3.51 29.1 1.77 0.77 4.13 1.03 2.54 32.4

CRSP Niebe Cameroon 176 4.12 23.1 2.27 0.69 4.34 0.90 2.75 21.2

Lori Niebe Cameroon 194 3.94 21.8 1.73 0.55 3.97 1.20 2.49 20.8

CB5 California 270 3.40 23.7 2.02 0.59 4.11 1.18 2.57 22.9

CB46 California 239 3.94 25.6 2.75 0.77 5.29 1.32 2.67 23.5

CB27 California 223 3.87 23.3 3.11 0.85 6.24 1.31 2.47 24.8

UCR 24b California 274 3.31 23.7 2.22 0.54 3.68 0.90 2.68 22.4

97-15-33b California 198 3.94 25.2 2.16 0.68 4.54 1.22 2.60 23.3

Charleston

Greenpack

Southern US 185 3.80 28.3 2.02 0.74 4.12 1.04 2.72 24.8

a Mean values from three replicates (R.D. Phillips, unpublished).b UCR 24 is an advanced breeding line with delayed leaf senescence, 97-15-33 is an advanced breeding line from UCR with all-white grain.c Mean individual seed weight.d Ash content determined according to the American Association of Cereal Chemists Approved Methods (AACC 1983, 8th ed., pp. 8–30).e Protein content determined from nitrogen content multiplied by 6.25 and using AACC combustion method 46-30.f Sugar contents determined by High Performance Liquid Chromatography with refractive index selection and extraction with chloroform/methanol.g These water soluble vitamins were assayed by microbiological techniques after a combined acid and enzyme extraction.

A.E

.H

all

eta

l./Field

Cro

ps

Resea

rch8

2(2

00

3)

10

3–

13

41

27

content (3.1% in Table 2). The objective is to try and

develop cultivars with greater sucrose content than is

present in ‘C93W-24-125B’ that would confer a dis-

tinct sweet taste to the grain.

The discovery of the sweet trait opens up the

possibility of developing new products that are very

attractive to consumers and thus larger markets for

cowpeas. One possibility is the development of sweet

versions of many current cultivars. Another possibility

is the development of a new ‘green-sweet’ market

class for use fresh or in frozen vegetable blends.

4. Nutritional and processing quality ofcultivars and breeding lines

Nutritional quality of cowpea grain is particularly

important because of its use as a food by many poor

people who do not have access to a broadly based diet

(chapter by R.D. Phillips et al.). The nutritional ana-

lyses presented are preliminary in that grain nutri-

tional quality can be influenced by both genotype and

environment. However, grain for the 12 cultivars and

breeding lines from Senegal, California and the south-

ern United States whose nutritional and processing

qualities are described in Tables 2–4 was produced in

the same summer field nursery at Riverside, CA.

Proximate analysis of 12 cultivars and lines showed

that ash content, which reflects the mineral content

of the grain, varied between 3.3 and 4.1% (Table 2).

Fat content is low in cowpea grain and varied from 1.4

to 2.7% for 100 advanced breeding lines from IITA

(Nielsen et al., 1993). This variation in fat content

may have some impact on processing quality of the

grain. Crude protein content varied among cultivars

from 22% for ‘Lori Niebe’ to 29% for ‘Ndiambour’

(Table 2). Seven of the cultivars and breeding lines had

protein contents between 22 and 24%—‘Mouride’,

‘Melakh’, ‘CRSP Niebe’, ‘Lori Niebe’, ‘CB5’,

‘CB27’ and ‘UCR24’. Grain of a set of 100 advanced

breeding lines from IITA that were analyzed had

protein contents ranging from 23 to 32% (Nielsen

et al., 1993). Thus some cowpea cultivars have protein

contents that may be less than is optimal, and genetic

variability exists for crude protein content that might

be exploited in improving grain quality. A major role

of dietary protein is to provide essential amino acids.

Only small variation was observed in amino acid

profiles of grain from the 12 cultivars and breeding

lines (Table 3).

Grain content of sucrose and oligosaccharides is

important from two standpoints. The oligosaccharides

raffinose and stachyose are considered to be a major

cause of flatulence when cowpeas are consumed by

Table 3

Amino acid profiles of cowpea cultivars and linesa

Cultivaror lineb Percentage of total recovered amino acidsc

Asp Ser Glu Gly His Arg Thr Ala Pro Cys Tyr Val Met Lys Ile Leu Phe

Mouride 13.9 6.6 19.7 4.8 3.6 7.0 3.9 5.1 4.7 1.05 3.4 3.8 0.56 6.9 3.4 7.3 4.2

Melakh 13.2 6.3 20.1 4.8 3.5 7.2 4.0 5.0 4.8 1.01 3.7 4.1 0.68 6.7 3.4 7.2 4.3

Bambey 21 14.4 6.2 21.7 4.5 2.9 7.7 4.1 5.0 4.6 0.96 3.4 4.0 0.48 7.3 3.2 6.8 3.6

Ndiambour 14.0 5.6 20.1 4.5 3.6 9.3 3.9 4.8 4.3 1.04 3.4 3.9 0.45 7.0 3.3 6.9 3.8

CRSP Niebe 13.9 6.3 20.9 4.5 3.4 7.3 4.0 4.9 4.6 0.99 3.4 4.1 0.60 7.0 3.4 7.0 3.7

Lori Niebe 13.6 5.7 20.6 4.4 3.3 6.8 2.9 5.1 4.8 1.09 3.6 4.3 0.64 7.0 3.6 7.4 4.4

CB5 14.0 6.0 20.5 4.7 3.4 8.0 3.9 4.9 4.5 0.88 3.3 4.0 0.51 7.5 3.3 6.9 3.6

CB46 13.4 5.9 20.6 4.6 3.6 8.4 3.9 5.1 4.4 0.88 2.9 4.1 0.61 6.7 3.4 7.2 4.2

CB27 14.3 6.1 20.2 4.6 3.4 7.5 4.0 4.9 4.6 0.90 3.5 4.3 0.52 7.1 3.4 7.0 3.7

UCR 24b 13.9 6.2 20.8 4.7 3.2 7.9 3.8 4.8 4.5 0.92 3.4 4.0 0.50 7.4 3.3 6.9 3.7

97-15-33b 13.6 6.0 20.0 4.5 3.1 8.3 4.1 5.0 4.7 1.05 3.4 4.4 0.46 7.4 3.3 7.0 3.8

Charleston Gp 14.2 5.6 19.1 4.5 4.1 9.6 3.8 4.9 4.5 0.91 3.4 4.1 0.60 6.6 3.3 6.9 3.9

a Mean values from three replicates (R.D. Phillips, unpublished).b UCR 24 is an advanced breeding line with delayed leaf senescence, 97-15-33 is an advanced breeding line from UCR with all-white

grain.c Amino acid profiles were determined from acid hydrolysates using chromatographic separation and detection of fluorescent derivatives of

the essential amino acids. Acid hydrolysis totally destroys tryptophan so values are not included for that essential amino acid.

128 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

humans and thus high concentrations may be unde-

sirable. Conversely, high sucrose content may enhance

the flavor of cowpea grain. Most of the 12 cultivars

and breeding lines had sucrose concentrations of

1–2%, whereas the new California cultivar ‘CB27’

had a higher sucrose concentration of 3.1% (Table 2).

‘CB27’ also had the highest level of raffinose plus

stachyose that was 77% greater than that of ‘Melakh’

which had the lowest concentration of these oligosac-

charides.

Cowpea grains contain useful levels of some

B-vitamins. Genetic variation was apparent for niacin

with three cultivars from Senegal having the highest

values averaging 32 mg kg�1 and two cultivars

from Cameroon having the lowest values averaging

21 mg kg�1 (Table 2).

Some consumers prefer cooked grain with a firm

texture. Texture of cooked grains was determined

for the cultivars and lines listed in Table 2 at the

University of Georgia (R.D. Phillips, unpublished).

After 90 min of cooking, grains of ‘Mouride’,

‘Ndiambour’, ‘CRSP Niebe’ and ‘Lori Niebe’ had a

firmer texture; whereas grains of ‘Bambey 21’, ‘CB5’,

‘UCR 24’, ‘CB46’ and ‘CB27’ had a softer texture.

In contrast a taste panel conducted in Dakar, Senegal

found that cooked grain of ‘Bambey 21’ had excellent

firm texture, whereas grain of ‘Mouride’ had soft

texture and grain of ‘CB5’ had a good intermediate

firmness (ITA 1990 Report, unpublished). These con-

trasting results may be explained by the fact that

firmness can depend on cooking time with the optimal

cooking time varying among cultivars.

Paste- and akara-making characteristics of the

grain from the 12 cultivars and lines were determined

(Table 4). Capacity of proteins to entrap air and make a

foam structure is responsible for the desirable spongy,

elastic texture of the cooked akara balls. Specific

gravity of the paste after whipping provides an indica-

tion of the amount of air that has been incorporated.

Paste from ‘Lori Niebe’ had the lowest specific gravity

after whipping and produced the lightest balls. In

contrast, ‘Melakh’ had the highest specific gravity

after whipping and produced the heaviest balls.

Viscosity after whipping controls the flow behavior

of cowpea paste. If the paste is too thick or too thin, it

does not dispense effectively when dropped by spoon-

ful portions into hot oil and does not form attractively

shaped round balls. ‘Melakh’, ‘CB27’ and ‘UCR 24’

Table 4

Cowpea paste and akara ball characteristicsa

Cultivar

or line

Specific gravity Viscosity Ease of paste

dispensing

(1–3)b

Ease of ball

formation

(1–3)b

Shape

of ball

(1–2)c

Appearance of

ball surface

(1–2)d

Balls per batch

of 50 g dry

seed (number)

Weight

per ball

(g)Beforee Afterf Beforee

(Pa s)

Afterf

(Pa s)

Mouride 1.02 0.77 36 27 2 2 2 1 5 17.7

Melakh 1.03 0.82 27 22 2 2 1 2 5 18.2

Bambey 21 1.01 0.70 46 29 1 3 2 2 5 16.8

Ndiambour 1.01 0.75 39 29 2 2 2 2 5 17.8

CRSP Niebe 1.00 0.71 49 31 2 3 2 2 5 17.7

Lori Niebe 1.00 0.68 48 30 2 3 2 2 6 15.4

CB5 1.02 0.76 43 28 2 3 2 2 5 17.1

CB46 1.02 0.71 45 27 2 3 2 2 5 16.9

CB27 1.02 0.77 30 22 2 2 1 2 5 18.0

UCR 24 1.03 0.80 30 23 2 2 1 2 5 17.6

97-15-33 1.01 0.70 52 30 2 2 2 2 5 17.0

Charleston Gp 1.03 0.77 41 27 2 2 2 2 5 18.0

LSD0.05 0.03 0.03 4 3

a Mean values from three replicates (K.H. McWatters, unpublished).b 1: difficult, 2: easy, 3: very easy.c 1: distorted shape, 2: uniform round shape.d 1: rough surface, 2: smooth surface.e Before whipping.f After whipping.

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 129

had low viscosities after whipping and produced balls

with a distorted shape. The advanced UCR breeding

line with all-white grain, ‘97-15-33’, the cultivar from

Senegal with all-white grain, ‘Bambey 21’, the major

California cultivar, ‘CB46’, and the new cultivars from

Cameroon, ‘CRSP Niebe’ and ‘Lori Niebe’ all had

desirable paste characteristics and produced round,

uniformly shaped akara balls of good texture. All of

the cultivars and lines were considered as having

adequate paste characteristics for making akara.

Analyses of the nutritional and processing quality of

the cowpea cultivars developed by the CRSP were

interpreted as indicating their quality is adequate and

similar to those of older varieties that have been in use

for many years. Sufficient variability was observed in

some important parameters, however, to justify testing

all advanced lines for these parameters prior to their

release as cultivars. Parameters deserving attention

include protein and sucrose contents. This then leads

to the question of whether selection could be imposed

for these traits in early stages of cowpea breeding.

Implementing this selection on a practical scale

requires rapid assay techniques for protein and sucrose

contents. A rapid assay for sucrose content recently

has been developed by the CRSP project at Purdue

University (L.L. Murdok, unpublished) and rapid

assays for protein content have been developed for

other grain crops.

5. Cowpea germplasm collections

Germplasm collections provide a foundation for

breeding programs. When the Bean/Cowpea CRSP

began in 1980, the USDA cowpea germplasm collec-

tion had about 2000 accessions. The UCR CRSP

project in collaboration with USDA, IITA and African

national programs has conducted a cowpea germplasm

introduction and multiplication project at Riverside.

This work has resulted in a more than three-fold

increase in the number of cowpea germplasm acces-

sions in the USDA collection, which had about 7000

accessions in 2001, and a major cowpea germplasm

collection at UCR, which had about 5500 accessions in

2001. The UCR germplasm collection has been useful

to colleagues in Africa. For example, in earlier years

ISRA provided germplasm to UCR for the USDA

collection. In recent years the germplasm store of ISRA

broke down and several accessions were lost at that

facility. However, UCR was able to provide seed of

these ‘lost’ accessions to ISRA. Additional security has

been achieved because many of the cowpea accessions

that were processed by UCR have been placed in the

very cold, long-term storage facility of USDA at Fort

Collins, CO. Germplasm in the USDA and UCR

collections is available to scientists throughout the

world. The CRSP project at UCR has provided thou-

sands of seed lots to collaborating scientists and there

has been an active exchange of cowpea germplasm

among the various CRSP projects. The exchange and

enhancement of cowpea germplasm represent major

accomplishments of the CRSP because other than the

CRSP, USDA and IITA there were few major programs

in the world that were involved in the enhancement of

cowpea germplasm collections during the period from

1980 through 2001.

6. Conclusions

Considerable progress has been made by the colla-

borative African and US cowpea breeding projects

supported by the Bean/Cowpea CRSP but much

remains to be done. Further progress can be made in

breeding cowpea cultivars with phenologies and

plant habits that are suited to specific target production

zones and crop product utilizations. For example, an

improved spreading type with intermediate cycle

length is needed for dual-purpose grain and hay pro-

duction in the Sahelian zone. California could benefit

from a cultivar that begins flowering later and on higher

main stem nodes than current cultivars that would

produce a large single flush with a potential grain yield

of about 5000 kg/ha in a cycle length of 120 days from

sowing to harvest. Cultivars should be developed with

grain types that meet consumer needs more completely,

such as by incorporating the improved taste of the

sweet cowpeas and/or the improved appearance of the

persistent-green cowpeas. Also, all-white grain type

cultivars should be developed for Africa and the United

States to facilitate flour production and the use of

cowpea in various value-added foods, such as akara.

The major opportunity and challenge, however, is to

breed to incorporate resistances to all of the major

pests and diseases occurring in specific target produc-

tion zones, such that the cowpea cultivars can be

130 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

grown without using pesticides. For the San Joaquin

Valley of California resistances to lygus bug and

cowpea aphid must be combined with broad-based

resistances to root-knot nematodes and Fusarium

wilt. For the Sahelian zone of Africa, resistances to

Striga, cowpea aphid, flower thrips, hairy caterpillar

(Amsacta moloneyi Druce), bacterial blight and

CABMV are needed. For the Savanna zones, many

pest resistances are needed including those described

for the Sahelian zone plus resistances to additional

insect pests, such as pod bugs and pod borer, and

additional diseases. Genetic engineering can contri-

bute to this effort by providing unique genes that

confer resistance to certain insect pests. However,

many of the resistance genes will have to be incorpo-

rated by classical breeding procedures. DNA-marker-

assisted selection schemes should be developed to

enhance the efficiency of the classical breeding pro-

cedures (chapters by Kelly et al. and A.E. Hall). A

comprehensive DNA-marker linkage map has been

constructed on which several major diseases and pests

have been placed (Ouedraogo et al., 2002) that can

facilitate the development of DNA-marker selection

schemes. A sustainable, environmentally benign

‘Green Revolution’ would occur in Africa if cowpea

cultivars could be developed that are very productive

when grown without using pesticides. These cultivars

also would enhance soil fertility and other soil traits

that benefit cereals and other non-leguminous crops

grown in rotation with them, thereby enhancing the

sustainability of the whole farming system.

References

Abdalla, M.E., 1992. Resistance in cowpea (Vigna unguiculata [L.]

Walp.) to a California biotype of the cowpea aphid (Aphis

craccivora Koch), inheritance and mechanisms. Ph.D. Dis-

sertation. University of California, Riverside, p. 110.

Ahmed, F.E., Hall, A.E., 1993. Heat injury during early floral bud

development in cowpea. Crop Sci. 33, 764–767.

Ahmed, F.E., Hall, A.E., DeMason, D.A., 1992. Heat injury during

floral development in cowpea (Vigna unguiculata, Fabaceae).

Am. J. Bot. 79, 784–791.

Ahmed, F.E., Hall, A.E., Madore, M.A., 1993. Interactive effects of

high temperature and elevated carbon dioxide concentration on

cowpea (Vigna unguiculata (L.) Walp.). Plant Cell Environ. 16,

835–842.

Bingen, R.J., Hall, A.E., Ndoye, M., 1988. California cowpeas and

food policy in Senegal. World Dev. 16, 857–865.

Chambliss, O.L., 1974. Green seedcoat: a mutant in southernpea of

value to the processing industry. HortScience 9, 126.

Chambliss, O.L., 1979. ‘Freezegreen’ southernpea. HortScience 14,

193.

Chari, M.S., Patel, G.J., Patel, P.N., Raj, S., 1976. Evaluation of

lines for resistance to Aphiscraccivora Koch. Gujurat Agric.

Univ. Res. J. 1, 130–132.

Cisse, N., Ndiaye, M., Thiaw, S., Hall, A.E., 1995. Registration of

‘Mouride’ cowpea. Crop Sci. 35, 1215–1216.

Cisse, N., Thiaw, S., Ndiaye, M., Hall, A.E., 1996. Guide de

Production de Niebe, vol. 6, No. 2. Institut Senegalais de

Recherches Agricoles, Dakar, Senegal, 22 pp.

Cisse, N., Ndiaye, M., Thiaw, S., Hall, A.E., 1997. Registration of

‘Melakh’ cowpea. Crop Sci. 37, 1978.

Condon, A.G., Hall, A.E., 1997. Adaptation to diverse environ-

ments: variation in water-use efficiency within crop species. In:

Jackson, L.E. (Ed.), Ecology in Agriculture, Physiological

Ecology Series. Academic Press, San Diego, CA, pp. 79–116.

Craufurd, P.Q., Bojang, M., Wheeler, T.R., Summerfield, R.J.,

1998. Heat tolerance in cowpea: effect of timing and duration

of heat stress. Ann. Appl. Biol. 133, 257–267.

Dancette, C., Hall, A.E., 1979. Agroclimatology applied to water

management in the Sudanian and Sahelian zones of Africa. In:

Hall, A.E., Cannell, G.H., Lawton, H.W. (Eds.), Agriculture in

Semi-Arid Environments. Ecological Studies 34. Springer,

Berlin, pp. 98–118.

Ehlers, J.D., Hall, A.E., 1996. Genotypic classification of cowpea

based on responses to heat and photoperiod. Crop Sci. 36,

673–679.

Ehlers, J.D., Hall, A.E., 1998. Heat tolerance of contrasting cowpea

lines in short and long days. Field Crops Res. 55, 11–21.

Ehlers, J.D., Hall, A.E., Roberts, P.A., Matthews, W.C., Ismail,

A.M., Sanden, B.L., Frate, C.A., Eckard, A.N., 1996. Blackeye

varietal improvement, 1996 progress report. In: University of

California Dry Bean Research, 1996 Progress Report. Califor-

nia Dry Bean Advisory Board, Dinuba, CA, pp. 51–64.

Ehlers, J.D., Hall, A.E., Sanden, B.L., Roberts, P.A., Matthews,

W.C., Frate, C.A., Ismail, A., Eckard, A.N., 1997. Blackeye

varietal improvement, 1997 progress report. In: University of

California Dry Bean Research, 1997 Progress Report. Califor-

nia Dry Bean Advisory Board, Dinuba, CA, pp. 44–57.

Ehlers, J.D., Hall, A.E., Patel, P.N., Roberts, P.A., Matthews, W.C.,

2000a. Registration of ‘California Blackeye 27’ cowpea. Crop

Sci. 40, 854–855.

Ehlers, J.D., Hall, A.E., Roberts, P.A., Matthews, W.C., Sanden,

B.L., 2000b. Blackeye varietal improvement, 2000 progress

report. In: University of California Dry Bean Research, 2000

Progress Report. California Dry Bean Advisory Board, Dinuba,

CA, pp. 48–63.

Ehlers, J.D., Matthews Jr., W.C., Hall, A.E., Roberts, P.A., 2000c.

Inheritance of a broad-based form of root-knot nematode

resistance in cowpea. Crop Sci. 40, 611–618.

Ehlers, J.D., Hall, A.E., Roberts, P.A., Matthews, W.C., Sanden,

B.L., Frate, C.A., 2001. Blackeye varietal improvement, 2001

progress report. In: University of California Dry Bean

Research, 2001 Progress Report. California Dry Bean Advisory

Board, Dinuba, CA, pp. 23–37.

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 131

Elawad, H.O.A., Hall, A.E., 2002. Registration of ‘Ein El Gazal’

cowpea. Crop Sci. 42, 1745–1746.

El-Kholy, A.S., Hall, A.E., Mohsen, A.A., 1997. Heat and chilling

tolerance during germination and heat tolerance during flower-

ing are not associated in cowpea. Crop Sci. 37, 456–463.

Elowad, H.O.A., Hall, A.E., 1987. Influences of early and late

nitrogen fertilization on yield and nitrogen fixation of cowpea

under well-watered and dry field conditions. Field Crops Res.

15, 229–244.

Elowad, H.O.A., Hall, A.E., Jarrell, W.M., 1987. Comparisons of

ureide and acetylene reduction methods for estimating

biological nitrogen fixation by glasshouse-grown cowpea. Field

Crops Res. 15, 215–227.

de Faria, R.M.., 1984. Mycorrhizal influences on cowpea (Vigna

unguiculata [L.] Walp.) under different levels of soil phos-

phorus and drought. Ph.D. Dissertation. University of Califor-

nia, Riverside, p. 141.

Fery, R.L., 1998. ‘Charleston Greenpack’, a pinkeye-type south-

ernpea with a green cotyledon phenotype. HortScience 33,

907–908.

Fery, R.L., 1999. ‘Petite-N-Green’, a small-seeded, full season,

green cotyledon, pinkeye-type southernpea. HortScience 34,

938–939.

Fery, R.L., 2000. ‘Green Pixie’, a small-seeded, green cotyledon,

cream-type southernpea. HortScience 35, 954–955.

Fery, R.L., 2002. ‘Green Dixie Blackeye’, a green cotyledon,

blackeye-type southernpea. HortScience 37, 233–234.

Fery, R.L., Dukes, P.D., 1980. Inheritance of root-knot nematode

resistance in cowpea (Vigna unguiculata (L.) Walp.). J. Am.

Soc. Hort. Sci. 105, 671–674.

Fery, R.L., Dukes, P.D., 1982. Inheritance and assessment of a

second root-knot resistance factor in southernpea (Vigna

unguiculata (L.) Walp.). HortScience 17, 152 (abstract).

Fery, R.L., Dukes, P.D., 1994. Genetic analysis of the green

cotyledon trait in southernpea [Vigna unguiculata (L.) Walp.].

J. Am. Soc. Hort. Sci. 119, 1054–1056.

Fery, R.L., Dukes, P.D., 1995a. ‘Bettersnap’ southernpea.

HortScience 30, 1318–1319.

Fery, R.L., Dukes, P.D., 1995b. Registration of US-566, US-567

and US-568 root-knot nematode resistant cowpea germplasm

lines. Crop Sci. 35, 1722.

Fery, R.L., Dukes, P.D., Maguire, F.P., 1993. ‘Bettergreen’

southernpea. HortScience 28, 856.

Fery, R.L., Dukes, P.D., Thies, J.D., 1994. Characteristics of new

sources of resistance in cowpea to the southern root knot

nematode. HortScience 29, 678–679.

Gwathmey, C.O., Hall, A.E., 1992. Adaptation to midseason

drought of cowpea genotypes with contrasting senescence

traits. Crop Sci. 32, 773–778.

Gwathmey, C.O., Hall, A.E., Madore, M.A., 1992a. Adaptive

attributes of cowpea genotypes with delayed monocarpic leaf

senescence. Crop Sci. 32, 765–772.

Gwathmey, C.O., Hall, A.E., Madore, M.A., 1992b. Pod removal

effects on cowpea genotypes contrasting in monocarpic

senescence traits. Crop Sci. 32, 1003–1009.

Hall, A.E., 1984. Developing cowpea varieties with improved yield

under conditions of extreme drought and heat. In: Ghaderi, A.

(Ed.), Research Highlights, vol. 1, No. 1. Bean/Cowpea CRSP,

Michigan State University, p. 4.

Hall, A.E., 1992. Breeding for heat tolerance. Plant Breed. Rev. 10,

129–168.

Hall, A.E., 1993. Physiology and breeding for heat tolerance in

cowpea, and comparison with other crops. In: Kuo, C.G. (Ed.),

Proceedings of the International Symposium on Adaptation of

Food Crops to Temperature and Water Stress, Taiwan, August

13–18, 1992. Publ. No. 93-410. Asian Vegetable Research and

Development Center, Shanhua, Taiwan, pp. 271–284.

Hall, A.E., 1999. Cowpea. In: Smith, D.L., Hamel, C. (Eds.), Crop

Yield Physiology Processes. Springer, Berlin, pp. 355–373.

Hall, A.E., 2001. Crop Responses to Environment. CRC Press,

Boca Raton, FL, p. 232.

Hall, A.E., 2003. Comparative ecophysiology of cowpea, common

bean and peanut. In: Nguyen, H.T., Blum, A. (Eds.), Physiology

and Biotechnology Integration for Plant Breeding. Marcel

Dekker, New York.

Hall, A.E., Frate, C.A., 1996. Blackeye Bean Production in

California. Univ. Calif. Div. Agr. Sci. Publ. 21518, 23 pp.

Hall, A.E., Grantz, D.A., 1981. Drought resistance of cowpea

improved by selecting for early appearance of mature pods.

Crop Sci. 21, 461–464.

Hall, A.E., Patel, P.N., 1985. Breeding for resistance to drought and

heat. In: Singh, S.R., Rachie, K.O. (Eds.), Cowpea Research,

Production, and Utilization. Wiley, New York, pp. 137–151.

Hall, A.E., Patel, P.N., 1987. Cowpea improvement for semiarid

regions of sub-Saharan Africa. In: Menyonga, J.M., Bezuneh,

T., Youdeowei, A. (Eds.), Proceedings of the International

Drought Symposium on Food Grain Production in Semiarid

Regions of Sub-Saharan Africa, SAFGRAD, Nairobi,

Kenya, May 1986. SAFGRAD, Ouagadougou, Burkina Faso,

pp. 279–290.

Hall, A.E., Ziska, L.H., 2000. Crop breeding strategies for the 21st

century. In: Reddy, K.R., Hodges, H.F. (Eds.), Climate Change

and Global Crop Productivity. CAB International, Oxon,

pp. 407–423.

Hall, A.E., Mutters, R.G., Hubick, K.T., Farquhar, G.D., 1990.

Genotype differences in carbon isotope discrimination by

cowpea under wet and dry field conditions. Crop Sci. 30,

300–305.

Hall, A.E., Richards, R.A., Condon, A.G., Wright, G.C., Farquhar,

G.D., 1994a. Carbon isotope discrimination and plant breeding.

Plant Breed. Rev. 12, 81–113.

Hall, A.E., Thiaw, S., Krieg, D.R., 1994b. Consistency of genotypic

ranking for carbon isotope discrimination by cowpea grown in

tropical and subtropical zones. Field Crops Res. 36, 125–131.

Hall, A.E., Cisse, N., Thiaw, S., Ndiaye, M., Faye, M.D., Balde,

M., 1996. Cowpea Production Bulletin for Senegal. Univ. Calif.

Special Publ., 11 pp.

Hall, A.E., Singh, B.B., Ehlers, J.D., 1997a. Cowpea breeding.

Plant Breed. Rev. 15, 215–274.

Hall, A.E., Thiaw, S., Ismail, A.M., Ehlers, J.D., 1997b. Water-use

efficiency and drought adaptation of cowpea. In: Singh,

B.B., Mohan Raj, D.R., Dashiell, K.E., Jackai, L.E.N.

(Eds.), Advances in Cowpea Research. IITA, Ibadan, Nigeria,

pp. 87–98.

132 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134

Hall, A.E., Ismail, A.M., Ehlers, J.D., Marfo, K.O., Cisse, N.,

Thiaw, S., Close, T.J., 2003. Breeding cowpea for tolerance to

temperature extremes and adaptation to drought. In: Proceed-

ings of the World Cowpea Research Conference III, September

4–7, 2000. IITA, Ibadan, Nigeria.

International Institute of Tropical Agriculture, 1984. Research

Highlights for 1983. IITA, Ibadan, Nigeria.

Ismail, A.M., Hall, A.E., 1992. Correlation between water-use

efficiency and carbon isotope discrimination in diverse cowpea

genotypes and isogenic lines. Crop Sci. 32, 7–12.

Ismail, A.M., Hall, A.E., 1998. Positive and potential negative

effects of heat-tolerance genes in cowpea lines. Crop Sci. 38,

381–390.

Ismail, A.M., Hall, A.E., 2000. Semidwarf and standard-height

cowpea responses to row spacing in different environments.

Crop Sci. 40, 1618–1623.

Ismail, A.M., Hall, A.E., 2002. Variation in traits associated with

chilling tolerance during emergence in cowpea germplasm.

Field Crops Res. 77, 99–113.

Ismail, A.M., Hall, A.E., Close, T.J., 1997. Chilling tolerance

during emergence of cowpea associated with a dehydrin and

slow electrolyte leakage. Crop Sci. 37, 1270–1277.

Ismail, A.M., Hall, A.E., Close, T.J., 1999a. Purification and partial

characterization of a dehydrin involved in chilling tolerance

during seedling emergence of cowpea. Plant Physiol. 120, 237–

244.

Ismail, A.M., Hall, A.E., Close, T.J., 1999b. Allelic variation of a

dehydrin gene cosegregates with chilling tolerance during seed-

ling emergence. Proc. Natl. Acad. Sci. USA 96, 13566–13570.

Ismail, A.M., Hall, A.E., Ehlers, J.D., 2000. Delayed-leaf-

senescence and heat-tolerance traits mainly are independently

expressed in cowpea. Crop Sci. 40, 1049–1055.

Kerr, W.L., Ward, C.D., McWatters, K.H., Resurreccion, A.V.,

2001. Milling and particle size of cowpea flour and snack chip

quality. Food Res. Int. 34, 39–45.

Khalfaoui, J.-L.B., Havard, M., 1993. Screening peanut cultivars in

the field for root growth: a test by herbicide injection in the soil.

Field Crops Res. 32, 173–179.

Kitch, L.W., 1987. Relationship of bruchid (Callosobruchus

maculatus) resistance genes in three cowpea cultivars. Ph.D.

Thesis. Purdue University.

Kitch, L.W., Boukar, O., Endondo, C., Murdock, L.L., 1998.

Farmer acceptability criteria in breeding cowpea. Exp. Agric.

34, 475–486.

Kitch, L.W., Boukar, O., Ehlers, J.D., Shade, R., Ntoukam, G.,

Murdock, L.L., 2003. Registration of ‘‘C93W-24-125B’’ cow-

pea germplasm. Crop Sci., in press.

Kwapata, M.B., Hall, A.E., 1985. Effects of moisture regime and

phosphorus on mycorrhizal infection, nutrient uptake, and

growth of cowpeas (Vigna unguiculata (L.) Walp.). Field Crops

Res. 12, 241–250.

Kwapata, M.B., Hall, A.E., 1990. Determinants of cowpea (Vigna

unguiculata) seed yield at extremely high plant density. Field

Crops Res. 24, 23–32.

Lorz, A.P., Halsey, L.H., 1964. Snapea, a new cream type southern

pea variety for snap pod use, Univ. Florida Agr. Expt. Sta. Cir.

S-160.

Mackie, W.W., 1946. Blackeye Beans in California. University of

California, Agricultural Experiment Station, Bulletin 696, p. 56.

Magruder, R., Wester, R.E., 1941. Green cotyledon, a new

character in the mature lima bean (Phaseolus lunatus L.).

Proc. Am. Soc. Hort. Sci. 38, 581–584.

Marfo, K.O., Hall, A.E., 1992. Inheritance of heat tolerance during

pod set in cowpea. Crop Sci. 32, 912–918.

Martyn, K.P., 1991. Genetic performance and behavioral variation

between clones of Aphis craccivora Koch, with special

references to resistance in Vigna unguiculata [L.] Walp.

(cowpea). Ph.D. Dissertation. University of London.

McWatters, K.H., Hung, C.-Y.T., Hung, Y.-C., Chinnan, M.S.,

Phillips, R.D., 2001. Akara making characteristics of five

US varieties of cowpeas Vigna unguiculata. J. Food Qual. 24,

53–65.

Menendez, C.M., Hall, A.E., 1995. Heritability of carbon isotope

discrimination and correlations with earliness in cowpea. Crop

Sci. 35, 673–678.

Menendez, C.M., Hall, A.E., 1996. Heritability of carbon isotope

discrimination and correlations with harvest index in cowpea.

Crop Sci. 36, 233–238.

Menendez, C.M., Hall, A.E., Gepts, P., 1997. A genetic linkage

map of cowpea (Vigna unguiculata) developed from a cross

between two inbred, domesticated lines. Theor. Appl. Genet.

95, 1210–1217.

Mutters, R.G., Hall, A.E., Patel, P.N., 1989. Photoperiod and light

quality effects on cowpea floral development at high tempera-

tures. Crop Sci. 29, 1501–1505.

Nielsen, C.L., Hall, A.E., 1985a. Responses of cowpea (Vigna

unguiculata (L.) Walp.) in the field to high night air

temperature during flowering. I. Thermal regimes of production

regions and field experimental system. Field Crops Res. 10,

167–179.

Nielsen, C.L., Hall, A.E., 1985b. Responses of cowpea (Vigna

unguiculata (L.) Walp.) in the field to high night air

temperature during flowering. II. Plant responses. Field Crops

Res. 10, 181–196.

Nielsen, S.S., Brandt, W.E., Singh, B.B., 1993. Genetic variability

for nutritional composition and cooking time of improved

cowpea lines. Crop Sci. 33, 469–472.

Omwega, C.O., Thomason, I.J., Roberts, P.A., 1988. A non-

destructive technique for screening bean germplasm for

resistance to Meloidogyne incognita. Plant Dis. 72, 970–972.

Ouedraogo, J.T., Gowda, B.S., Jean, M., Close, T.J., Ehlers, J.D.,

Hall, A.E., Gillaspie, A.G., Roberts, P.A., Ismail, A.M.,

Bruening, G., Gepts, P., Timko, M.P., Belzile, F.J., 2002. An

improved genetic linkage map for cowpea (Vigna unguiculata

L.) combining AFLP, RFLP, RAPD, biochemical markers and

biological resistance traits. Genome 45, 175–188.

Patel, P.N., 1985. Fungal, bacterial and viral diseases of cowpea in

the USA. In: Singh, S.R., Rachie, K.O. (Eds.), Cowpea

Research, Production, and Utilization. Wiley, New York,

pp. 205–213.

Patel, P.N., Hall, A.E., 1986. Registration of snap-cowpea

germplasms. Crop Sci. 26, 207.

Patel, P.N., Hall, A.E., 1988. Inheritance of heat-induced brown

discoloration in seed coats of cowpea. Crop Sci. 28, 929–932.

A.E. Hall et al. / Field Crops Research 82 (2003) 103–134 133

Patel, P.N., Hall, A.E., 1990. Genotypic variation and classification

of cowpea for reproductive responses to high temperature under

long photoperiods. Crop Sci. 30, 614–621.

Pathak, R.S., 1988. Genetics of resistance to aphid in cowpea. Crop

Sci. 28, 474–476.

Prinyawiwatkul, W., Beuchat, L.R., Resurreccion, A.V.A., 1993.

Optimization of sensory qualities of an extruded snack based

on cornstarch and peanut flour. Lebensm.-Wiss. Technol. 26,

393–399.

Prinyawiwatkul, W., McWatters, K.H., Beuchat, L.R., Phillips,

R.D., 1996. Cowpea flour: a potential ingredient in food

products. CRC Crit. Rev. Food Sci. Nutr. 36, 413–436.

Prinyawiwatkul, W., McWatters, K.H., Beuchat, L.R., Phillips,

R.D., 1997. Physicochemical and sensory properties of chicken

nuggets extended with fermented cowpea and peanut flours. J.

Agric. Food Chem. 45, 1891–1899.

Rigert, K.S., Foster, K.W., 1987. Inheritance of resistance to two

races of Fusarium wilt in three cowpea cultivars. Crop Sci. 27,

220–224.

Roberts, P.A., Frate, C.A., Matthews, W.C., Osterli, P.P., 1995.

Interactions of virulent Meloidogyne incognita and Fusarium

Wilt on resistant cowpea genotypes. Phytopathology 85, 1288–

1295.

Roberts, P.A., Matthews, W.C., Ehlers, J.D., 1996. New resistance

to virulent root-knot nematodes linked to the Rk locus in

cowpea. Crop Sci. 36, 889–894.

Roberts, P.A., Ehlers, J.D., Hall, A.E., Matthews, W.C., 1997.

Characterization of new resistance to root-knot nematodes in

cowpea. In: Singh, B.B., Mohan Raj, D.R., Dashiell, K.E.,

Jackai, L.E.N. (Eds.), Advances in Cowpea Research. IITA,

Ibadan, Nigeria, pp. 207–214.

Roberts, P.A., Matthews, W.C., Hall, A.E., Ehlers, J.D., Temple,

S.R., Helms, D., 2001. Blackeye bean tolerance and resistance

to root-knot nematodes, 2001 progress report. In: University of

California Dry Bean Research, 2001 Progress Report. Califor-

nia Dry Bean Advisory Board, Dinuba, CA, pp. 38–42.

Robertson, B.M., 1985. Studies of rooting pattern, earliness, and

seed size and density in Vigna unguiculata [L.] Walp. Ph.D.

Dissertation. University of California, Riverside, p. 217.

Robertson, B.M., Hall, A.E., Foster, K.W., 1985. A field technique

for screening for genotypic differences in root growth. Crop

Sci. 25, 1084–1090.

Sene, D., N’Diaye, S.M., 1974. L’amelioration du niebe (Vigna

unguiculata) au CNRA de Bambey: de 1959 a 1973 resultats

obtenus entre 1970 et 1973. Agron. Trop. 29 (8), 772–801.

Sene, D., Laurent, P., N’Diaye, S.M., 1971. Les varietes de niebe

actuellement conseillees au Senegal. Les Cahiers d’Agriculture

Practique des Pays Chaude, No. 2, p. 18.

Shackel, K.A., Hall, A.E., 1983. Comparison of water relations and

osmotic adjustment in sorghum and cowpea under field

conditions. Aust. J. Plant Physiol. 10, 423–435.

Shade, R.E., Furgason, E.S., Murdock, L.L., 1989. Ultrasonic

insect detector. US Patent No. 4,809,554.

Shade, R.E., Furgason, E.S., Murdock, L.L., 1990. Detection of

hidden insect infestations by feeding-generated ultrasonic

signals. Am. Entomol. 36, 231–234.

Shade, R.E., Murdock, L.L., Kitch, L.W., 1999. Interactions

between cowpea weevil (Coleoptera: Bruchidae) populations

and Vigna (Leguminosae) species. J. Econ. Entomol. 92,

740–744.

Singh, S.R., Jackai, L.E.N., 1985. Insect pests of cowpea in Africa.

In: Singh, S.R., Rachie, K.O. (Eds.), Cowpea Research,

Production, and Utilization. Wiley, New York, pp. 217–231.

Singh, B.B., Chambliss, O.L., Sharma, B., 1997. Recent advances

in cowpea breeding. In: Singh, B.B., Mohan Raj, D.R.,

Dashiell, K.E., Jackai, L.E.N. (Eds.), Advances in Cowpea

Research. IITA, Ibadan, Nigeria, pp. 30–49.

Swanson, T.A., Van Gundy, S.D., 1984. Cowpea resistance to root-

knot caused by Meloidogyne incognita and M. javanica. Plant

Dis. 68, 961–964.

Thiaw, S., Hall, A.E., Parker, D.R., 1993. Varietal intercropping

and the yields and stability of cowpea production in semiarid

Senegal. Field Crops Res. 33, 217–233.

Thomason, I.J., McKinney, H.E., 1960. Reaction of cowpeas, Vigna

sinensis, to root-knot nematodes, Meloidogyne spp. Plant Dis.

Rep. 44, 51–53.

Turk, K.J., Hall, A.E., Asbell, C.W., 1980. Drought adaptation of

cowpea. I. Influence of drought on seed yield. Agron. J. 72,

413–420.

Turner, N.C., Wright, G.C., Siddique, K.H.M., 2000. Adaptation of

grain legumes (pulses) to water-limited environments. Adv.

Agron. 71, 193–231.

Warrag, M.O.A., Hall, A.E., 1983. Reproductive responses of

cowpea to heat stress: genotypic differences in tolerance to heat

at flowering. Crop Sci. 23, 1088–1092.

Webber, H.J., Orton, W.A., 1902. Some diseases of cowpea. II. A

cowpea resistant to root-knot (Heterodera radicicola). USDA

Bureau of Plant Industry, Bulletin No. 17, pp. 23–38.

Wessel-Beaver, L., Beck, R.H., Lambert, R.J., 1984. Rapid method

for measuring kernel density. Agron. J. 76, 307–309.

134 A.E. Hall et al. / Field Crops Research 82 (2003) 103–134