Breeding for adaptation to drought and heat in cowpea

Europ. J. Agronomy 21 (2004) 447–454

Breeding for adaptation to drought and heat in cowpea

Anthony E. Hall∗

Department of Botany and Plant Sciences, University of California, Riverside, CA 92521, USA

Accepted 8 July 2004

Abstract

Accomplishments in breeding for adaptation to drought and heat are reviewed based upon work with the indeterminate grainlegume species cowpea. Plant traits and some crop management methods are examined that influence adaptation to rainfedproduction in the drought-prone, semiarid tropical Sahelian zone of Africa. Drought escape, drought resistance, delayed-leaf-senescence, and varietal intercrops are examined. In addition, adaptation to the heat that can detrimentally impact irrigatedproduction in the hot, subtropical arid zone of California is evaluated. Heat tolerance during reproductive development, electrolyteleakage, membrane thermostability, some aspects of crop management including date of sowing, and chilling tolerance duringemergence including the beneficial effects of a dehydrin protein are considered. Methods for breeding cowpeas with adaptationto drought and heat are described that have been effective.© 2004 Elsevier B.V. All rights reserved.

Keywords: Drought escape; Drought resistance; Dehydration avoidance; Delayed-leaf-senescence; Heat resistance; Heat tolerance; Chillingt

1

agWtaiait

pactne

tedtiveandalso

beenor

Sa-ely

1

olerance; Dehydrin protein; Electrolyte leakage; Membrane thermostability

. Introduction

I will review accomplishments in breeding fordaptation to drought and heat with the indeterminaterain legume species cowpea (Vigna unguiculataL.alp.). I mainly will discuss the work of a collabora-

ive research program involving scientists in the USAnd Africa. The history of our collaborative program

s described inHall et al. (2003). First I will considerdaptation to the droughts that can detrimentally

mpact rainfed production in the harsh, semiaridropical Sahelian zone of Africa. I will then examine

∗ Tel.: +1 951 236 1580; fax: +1 951 787 4437.E-mail address:[email protected].

adaptation to the heat that can detrimentally imirrigated production in the hot, subtropical arid zoof California. I will describe the breeding of adapcowpea cultivars for these zones including effecmethods for breeding for adaptation to droughtheat, and some aspects of crop management. Iwill describe some approaches that have not yetadopted by either crop improvement programsfarmers but appear to have some merit.

2. Breeding for adaptation to drought

Rainfed crops growing in the semiarid tropicalhelian zone of Africa can be subjected to extrem

161-0301/$ – see front matter © 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.eja.2004.07.005

448 A.E. Hall / Europ. J. Agronomy 21 (2004) 447–454

dry and hot conditions. Since 1968, droughts have oc-curred on many years in the drier part of this zonewhich were so severe that local landraces and mod-ern cultivars of drought hardy crops such as pearlmillet, sorghum and peanut failed to produce signifi-cant quantities of grain. Also, virtually all cowpea lan-draces that had evolved over 100’s of years in the Sa-hel did not produce significant quantities of grain inthe years and locations with the most severe droughtsdue to the likely climate change since 1968. I will de-scribe cultivars and crop management practices thathave been developed that have enabled cowpeas toproduce significant grain in most recent years in theSahel.

2.1. Crop phenology and adaptation to drought

Initially we attempted to select plants with optimalnumbers of days from sowing to first flowering andmaturity, and appropriate plant habit. However, rain-fall in the Sahel has been so variable and droughtshave been so extreme that this approach only was par-tially successful. I now consider it necessary to breedtwo types of cultivars and recommend that farmersgrow both types every year to enhance the chancesthat significant grain production will be achieved everyyear.

For about 30 years since 1968, there have been aseries of droughts in the Sahel that have resulted inthe growing season being considerably shorter than itw thefi cy-c se-l syn-c abita emn andt typeo‘ g3 atu-r estc ypeo out5 n ther raget read-i

spacing. ‘Ein El Gazal’ was evaluated in the Sahel asbreeding line 1-12-3 over many years (Hall and Patel,1985; Elawad and Hall, 2002). In 1981 at Louga, Sene-gal, ‘Ein El Gazal’ reached maturity 55 days from sow-ing and produced 1091 kg/ha of grain with a rainfall andtotal water supply of only 181 mm and hot conditionswith an evaporative demand of 6 mm/day. In contrast,a set of local landraces had only just begun to pro-duce pods at 55 days after sowing, and they producedvery little grain during this year at Louga. In 1983 atEl Obeid, Sudan with a rainfall of only 230 mm, ‘EinEl Gazal’ produced 500 kg/ha of grain while landraceswith later flowering and longer cycles from sowing tomaturity only produced 152 kg/ha. Early erect culti-vars, such as ‘Ein El Gazal’ and ‘Melakh’, have per-formed well when the rainfall season was short but dis-tinct due to their ability to escape late-season drought.Unfortunately, early erect cultivars have been damagedby mid-season droughts (Thiaw et al., 1993) due tothe extremely detrimental effects of drought on podset and pod-filling of erect synchronous-flowering va-rieties (Turk et al., 1980).

The second type of cultivar that we developed beginsflowering later than either ‘Ein El Gazal’ or ‘Melakh’and has a more spreading plant habit that provides itwith more sequential rather than synchronous flow-ering and a medium cycle from sowing to maturity.‘Mouride’ (Cisse et al., 1995) begins flowering in about38 days and reaches maturity 70 days from sowing.Cultivars that are more spreading than ‘Mouride’ andr n bee ouldb eens ene-g id-s asond tery ro-d de’h daryo redw nd‘P

r-f ings e tog rs,

as in the 50 years prior to 1968. Consequently,rst type of cultivar that we bred has a very shortle from sowing to maturity. This was achieved by

ecting plants that began flowering early and hadhronous flower production. They have an erect hnd produce their first floral buds on low main stodes and subsequent floral buds on the main stem

he first nodes on branches. We bred two of thisf cultivar. ‘Ein El Gazal’ (Elawad and Hall, 2002) and

Melakh’ (Cisse et al., 1997) are erect, begin flowerin0–35 days from sowing in the Sahel, and reach mity within 55–64 days from sowing with the shortycles occurring with late-season drought. This tf cultivar should be sown at close spacing of ab0 cm between rows and 25 cm between seeds iow. These erect plants exhibit less ground covehan traditional Sahelian landraces that have a spng habit and often have been sown at 100 cm× 100 cm

each maturity about 75 days from sowing also caffective in the Sahel. The spreading cultivars she sown in rows 50 cm apart with 50 cm betweeds within the row. In the Sahelian zone of Sal, ‘Mouride’ has exhibited greater resistance to meason drought but less ability to escape late-serought than ‘Melakh’. ‘Mouride’ also has greaield potential, presumably due to its longer repuctive period than the early erect cultivars. ‘Mourias produced about 3000 kg/ha at the wetter bounf the Sahel with about 450 mm of rainfall compaith grain yields of about 2400 kg/ha for ‘Melakh’ a

Ein El Gazal’ under well-watered conditions (Hall andatel, 1985).‘Mouride’ and ‘Melakh’ have complementary pe

ormance and it is important to consider the croppystem that would be most effective if farmers arrow both types of cultivars. Using other cultiva

A.E. Hall / Europ. J. Agronomy 21 (2004) 447–454 449

we have demonstrated that varietal intercrops can pro-duce more grain and hay, and be more stable underdry Sahelian conditions with infertile soil than any ofthe sole crops of cowpea that were tested (Thiaw etal., 1993). The varietal intercrops we tested consistedof alternating rows of early erect and medium-cyclespreading cowpea cultivars. Some mechanisms for thebeneficial effects of the varietal intercrops are as fol-lows. In years when a mid-season drought occurs, theearly erect cultivar becomes stunted and the medium-cycle spreading cultivar compensates by growing intothe space that is made available and produces muchof the grain and hay produced by the intercrop. Inyears when late-season drought occurs, the early erectcultivar produces abundant grain, while the medium-cycle spreading cultivar produces abundant hay but lit-tle grain. Farmers in the Sahel appreciate both the grainand the hay of cowpea, and are beginning to grow dif-ferent types of cultivars (Hall et al., 2003), but, the con-cept of varietal intercropping has not yet been extendedto them.

2.2. Drought resistance

‘Mouride’, ‘Melakh’ and ‘Ein El Gazal’ have sub-stantial resistance to vegetative-stage drought. ‘Cali-fornia Blackeye No. 5’ (‘CB5’) is one of the parents of‘Ein El Gazal’ (Elawad and Hall, 2002). In California,‘CB5’ has exhibited the ability to survive a vegetative-stage drought that would have killed most other annualc pro-d thatw ent( adb file,p top lantsu heri se-v thatc nutp hade

taged ciesh jectedt l inb n ei-

ther pearl millet or sorghum or peanut (Singh andMatsui, 2002). The authors hypothesized that the re-sistance mechanism may reside in the plant shoot.Under severe droughts, cowpea exhibits greater de-hydration avoidance and maintains leaf water poten-tials above−1.8 MPa, while pearl millet, sorghumand peanut can develop leaf water potentials as neg-ative as−4 to −9 MPa under these conditions.Petrieand Hall (1992)found that pearl millet developed morenegative leaf water potentials than cowpea, even atpredawn with plants of the two species growing in thesame pot.Petrie et al. (1992)hypothesized that thesespecies differences in leaf water potential may be dueto pearl millet having a root system that is less effectiveon a micro-scale in taking up water during soil drying,even though it has more roots than cowpea. They ar-gued that the numerous high-density clumps of rootsin pearl millet would develop internal dry soil layers,such that they would only take up water at the ends ofthe roots. In contrast, cowpea has a more uniform rootsystem and was hypothesized as using more of the rootsurface and thus being more effective in taking up wa-ter from drying soil than pearl millet. These hypotheseswere consistent with results from simulation modeling,but have not been adequately tested. Apparently, thereasons for the high level of vegetative-stage droughtresistance in some cowpea cultivars are controversialand largely unknown.

We probably incorporated resistance to vegetative-stage drought into ‘Ein El Gazal’, ‘Mouride’ and‘ ar-e asedo anyc ative-s beend lantss ins se-l ivev ndM ngl andm l ofv notbTa p-p nal

rop species, and to recover when re-watered anduce very high grain yields of about 4000 kg/haere similar to a weekly irrigated control treatm

Turk et al., 1980). The vegetative-stage drought heen imposed by sowing seed into a dry soil proroviding a small amount of water with sprinklersermit the seedlings to emerge, and then growing pnder hot sunny conditions for 43 days with no furt

rrigation or rain. In several years in the Sahel whenere vegetative-stage drought occurred, I observedowpea plants survived, while pearl millet and pealants that were growing in the same fields andmerged at about the same time had died.

Mechanisms for the resistance to vegetative-srought of cowpea compared with the other speave been studied. When these species were sub

o drought as seedlings growing in shallow soioxes, cowpeas survived 8–12 days longer tha

Melakh’ by chance, either through the choice of pnts, such as CB5, or through our selections bn yield trials in the Sahel. This indicates that mowpeas may have substantial resistance to vegettage drought. A simple screening technique haseveloped that uses visual observations of young pubjected to drought and recovery while growinghallow soil in boxes, which may be useful forecting cowpea genotypes with the ability to survegetative-stage drought (Singh et al., 1999; Singh aatsui, 2002). A large number of cowpea breedi

ines (190) were screened using the box methodost of them (148) exhibited substantial surviva

egetative-stage drought, even though they hadeen selected for this trait (Singh and Matsui, 2002).he research on improving drought resistance bySinghnd Matsui (2002)provides a promising avenue for aroaching this difficult problem; however, additio


information is needed on the extent of genetic variation,mechanisms and types of resistance to vegetative-stagedrought present in cowpea cultivars and accessions.

In general, cowpeas are very sensitive to droughtduring pod set and pod filling (Turk et al., 1980).A delayed-leaf-senescence (DLS) trait has been dis-covered in cowpea that conferred some resistance toreproductive-stage drought in erect cowpea cultivars.The DLS trait enabled them to recover after the droughtand produce a larger second flush of pods that com-pensated for the low yield by the first flush of pods(Gwathmey and Hall, 1992). The DLS trait appears tobe conferred by a single gene and may involve resis-tance to premature death caused byFusarium solanif. sp.phaseoli(Burke) Synd. & Hans., type A, whichprobably is widely distributed (Ismail et al., 2000). InSenegal, an early erect DLS cultivar began flowering inabout 35 days and produced about 2000 kg/ha of grainby 60 days, followed by a second flush of pods with thepotential to produce an additional 1000 kg/ha by 100days from sowing (Hall et al., 2003). Farm families inAfrica typically harvest cowpea pods by hand and donot uproot plants such that they can make multiple har-vests with individual crops. An early erect DLS cultivarmight be adapted to locations in the wetter boundaryof the Sahelian zone or the wetter Savanna zone wherethere is sufficient rainfall in most years to support acropping season of at least 100 days but a tendency formid-season droughts to occur.

2

im-p ght.I am-a sev-e e Sa-h‘ asesc risp rnem peaw eptcs rsh rec-

ommendation that farmers should grow both cultivars,since seasonal pest patterns often are not known at sow-ing time. Ideally one should breed cultivars that areadapted to drought and resistant to all of the majorpests and diseases that can occur in the target produc-tion zone, but this is not easy and will take several moreyears to accomplish for the Sahel.

2.4. Performance testing

Relative yield of potential new cultivars comparedwith current cultivars is an important measure of whathas been achieved by incorporating traits that improveadaptation to drought and resistance to pests and dis-eases. Relative yield provides a measure of the relativeefficiency of the cultivar in terms of the returns in grainper unit of land or other inputs when the new cultivarhas similar input requirements as the old cultivars. Thisoften is the case when cowpeas are grown in both theSahel and California, since they usually are not pro-vided with either fertilizer or manure.

Advanced cowpea breeding lines were evaluated inmultiple locations on experiment stations and farmersfields in the Sahelian zone of Senegal and Sudan overseveral years (Cisse et al., 1995, 1997; Elawad and Hall,2002). Performance testing of this type is critical forensuring that selected cultivars are adequately robust inrelation to the highly variable environments that theywill experience when grown on farmers fields under theharsh rainfed conditions that occur in semiarid zoness op-p n-c nota ual-i theg im-p cultf n thee

3

uinV ir-r m-m ub-j t in

.3. Pest and disease resistance

Farmers are not likely to adopt a new cultivar sly because it has improved adaptation to drou

n addition, farmers are concerned about the dge they readily see that is caused to cowpea byral pests and diseases that are prevalent in thelian zone (Hall et al., 1997). Both ‘Mouride’ and

Melakh’ have resistance to the seed-borne diseaused by bacterial blight (Xanthomonas campestvvignicolaBurkholder) Dye and cowpea aphid-boosaic virus. ‘Mouride’ also has resistance to coweevil (Callosobruchus maculatusFabricius) and tharastic weedStriga gesnerioidesWilld. Vatke. In con-

rast, ‘Melakh’ has resistance to cowpea aphid (Aphisraccivora Koch) and flower thrip (MegalurothripsjostedtiTrybom). The fact that these two cultivaave resistance to different pests also supports the

uch as the Sahel. On-farm trials also provided anortunity to obtain feedback from farm families coerning traits in the various lines they did and didppreciate, and especially their opinions on grain q

ty as perceived by both looking at and consumingrain in local dishes. Grain quality is an extremelyortant trait for buyers and consumers that is diffi

or breeders to assess and has a major impact oxtent of adoption by farmers.

. Breeding for adaptation to heat

The hot arid subtropical zone in the San Joaqalley of California where cowpea is grown underigation has a cool spring followed by a very hot suer. Cowpea often is sown in May and then is s

ected to hot weather during flowering and pod se


July and August (Hall and Frate, 1996). Studies havebeen conducted with cowpea plants subjected to highernight temperatures during flowering using enclosuresin field conditions (Nielsen and Hall, 1985a,b), andwith almost isogenic pairs of heat-resistant and heat-susceptible lines grown in field environments with con-trasting temperatures (Ismail and Hall, 1998). Thesestudies showed that increases in night temperaturecaused 4–14% decreases in both pod set and grain yieldfor each◦C above a threshold of 16◦C. The main mech-anism for these effects on cowpea is that high tem-peratures occurring in the late night during floweringcan cause pollen sterility and indehiscence of anthers(Hall, 1992, 1993). For cowpea, the heat-stress prob-lem mainly has been solved by breeding, however, cropmanagement also can be important in some cases.

3.1. Crop phenology and adaptation to heat

In principle, early sown cowpeas that flower earlycould escape the high night temperatures that occur inJuly and August in the San Joaquin Valley of Califor-nia. However, current cowpea cultivars cannot be sowntoo early in spring because they are sensitive to chill-ing and require a minimum soil temperature greaterthan 18◦C if they are to emerge adequately (Ismail etal., 1997). Also, the cool night temperatures of springcause cowpeas to develop slowly and the earliest flow-ering cowpea cultivars take about 60 days from sowingto first flowering in cool conditions. Some progress hasb nced d byI ge ecificd nu-c roms em-b rialsa -i eh cre-m in-h teine fect,w mic(

pres-e as-

say of a chip taken from a cotyledon in a manner thatdoes not harm the germination of the seed (Ismail etal., 1999). Several cowpea accessions were discoveredto have this dehydrin protein in their seed that could beused as parents (Ismail and Hall, 2002). Consequently,this trait can be readily bred into cultivars. Surprisingly,only three out of 61 US cowpea cultivars, which hadbeen developed in subtropical zones where soils can becool at sowing, contained the dehydrin protein (Ismailand Hall, 2002).

3.2. Heat tolerance during reproductivedevelopment

A method for breeding cowpeas with heat toleranceduring reproductive development has been developed(Hall, 1992, 1993) that was used to breed ‘CaliforniaBlackeye No. 27’ (‘CB27’) (Ehlers et al., 2000).‘CB27’ is both tolerant to heat during reproductivedevelopment and heat resistant in that it producesmore grain yield than other cowpea cultivars in hotfield environments (Ismail and Hall, 1998). Breedingfor heat tolerance, involved subjecting progeny to veryhigh night temperatures and long days in either fieldor glasshouse conditions, which only could be donein the summer, and selecting plants with the ability toabundantly produce flowers and set pods (Hall, 1992,1993). Long days must be used because under shortdays the detrimental effects of heat on reproductivedevelopment of cowpea are either much smaller or mayn ,p wersa enet sionoh rateb lve as lowa ti1 ods tion,f thato heren ong.

nced n de-v ance

een made in breeding cowpeas with chilling tolerauring emergence. An additive model was propose

smail et al. (1997), whereby chilling tolerance durinmergence is conferred by the presence of a spehydrin protein in the seed (positive single genelear effect) and the extent of electrolyte leakage feed (negative maternal effect) as a measure of mrane thermostability. Tests with backcross matend molecular genetic (Ismail et al., 1999) and inher

tance studies (Ismail and Hall, 2002) supported thypothesis that the dehydrin protein confers an inent of chilling tolerance under single nuclear geneeritance. They also showed that the dehydrin proffect is independent of the electrolyte-leakage efhich was shown to be nuclear rather than cytoplas

Ismail et al., 1999).Single seeds of cowpea can be screened for the

nce of the dehydrin protein using an immunoblot

ot occur (Ehlers and Hall, 1998). In the F2 generationlants were selected that abundantly produced flond set pods. This virtually fixed the recessive g

hat provides tolerance to heat-induced suppresf floral bud development (Hall, 1993). Tolerance toeat during pod set was more difficult to incorpoecause, even though it appeared to mainly invoingle dominant gene, the realized heritability wast about 0.26 (Marfo and Hall, 1992) and it is likely tha

nheritance also depends on some minor genes (Hall,993). Fixing a high level of heat tolerance during pet has taken several generations of family selecollowed by single-plant selection in these families,nly could be conducted in the summer in places wight temperatures were very hot and days were l

A more rapid method for breeding for heat tolerauring reproductive development recently has beeeloped based on the observation that heat toler


during pod set might be associated with slow electrolyteleakage from leaf discs subjected to heat stress and thusmembrane thermostability (Ismail and Hall, 1998).This latter study was conducted with sets of cowpealines that had differences in heat tolerance but differentgenetic backgrounds. Genetic selection experimentsnow have been conducted using a pair of these linesas parents that have shown there is a close relationshipbetween heat tolerance during pod set and slow elec-trolyte leakage from leaf discs subjected to heat stress(Thiaw and Hall, 2004). The protocol used consistedof taking leaf discs from recently expanded leaves andsubjecting them to 46◦C in aerated distilled water anddetermining the % electrolyte leakage into the water af-ter 6 h. Lines divergently selected for leaf-electrolyte-leakage under heat stress also differed with respect toheat tolerance during pod set with low leakage beingassociated with high pod set. Lines divergently selectedfor heat tolerance during pod set also differed in leaf-electrolyte-leakage with high pod-set lines also havinglow electrolyte leakage from leaf discs.

Realized heritability of leaf-electrolyte-leakageunder heat stress was 0.28–0.34 and the geneticcorrelation between this trait and pod set was−0.36.This gives an indirect realized heritability when usingslow electrolyte leakage to select for high pod set ofonly 0.10–0.12, which is smaller than the realizedheritability for direct selection for pod set of 0.26 ob-tained byMarfo and Hall (1992). However, genotypicdifferences in electrolyte leakage of leaf discs underh angeo sw atm werea hichi ctivew ern ofs

or-p vel-o irectsH ,s tiona withv thef g

in moderate-temperature glasshouses and short days,select single plants for low electrolyte leakage by leafdiscs under heat stress. During the second summer,grow the F5 generation in a field or glasshouse environ-ment with very high night temperatures and long days.Select families for low leaf-electrolyte-leakage underheat stress during the vegetative stage, and then selectindividual plants within these families that have abun-dant flower production and pod set. In the F6 and F7generations select single plants with slow electrolyteleakage of leaf discs under heat stress when growing inmoderate-temperature glasshouses during the fall andwinter when days are short. Finally, in the third sum-mer, begin performance testing selected F8 lines in hotlong-day commercial production environments.

3.3. Interactions between heat tolerance duringreproductive development and other traits

‘CB27’ and other genotypes we have bred withheat tolerance during reproductive development ex-hibit some dwarfing under hot and also moderate-temperature conditions (Ismail and Hall, 1998). Heat-tolerant genotypes were shorter due to shorter intern-odes, produced less vegetative shoot biomass and hadgreater harvest index (Ismail and Hall, 1998). Theheat-tolerant genotypes produce much more grain thanheat-sensitive genotypes under very hot conditions,i.e. they also are heat resistant. But, we were con-cerned that the dwarfing may reduce their value un-d xtento toy ivarsw tibleg s inda thatt eaterg ratet rrowr otem ad-v bys om-p ider od-e linesa less

eat stress were detected with plants grown in a rf environments (Thiaw, 2003). Genotypic differenceere most readily detected with plants grownoderate temperatures and long days, but theylso detected in hot conditions and short days w

ndicates that this selection technique may be effeith plants growing in both off-season and summurseries. This would permit more generationselection for heat tolerance in each year.

The following strategy was proposed for rapid incoration of heat tolerance during reproductive depment into cowpea that combines direct and indelection for heat tolerance during pod set (Thiaw andall, 2004). During the summer with the F2 generationelect single plants with abundant flower producnd pod set in a field or glasshouse environmentery high night temperatures and long days. Duringall and winter with the F3 and F4 generations growin

er moderate-temperature conditions, and the ef hot weather during flowering varies from yearear. Semi-dwarf heat-resistant genotypes and cultere compared with standard-height heat-suscepenotypes and cultivars under different row spacingifferent moderate-temperature environments (Ismailnd Hall, 2000). These and other studies indicated

he heat-resistant semi-dwarf lines can produce grrain yield than standard-height lines under mode

emperatures, providing they are grown under naow spacing (51 cm) with soil conditions that promoderate to vigorous early plant growth. The yield

antage was due to impaired production of podstandard-height, but not semi-dwarf lines when cetition for light was strong. In contrast, under wow spacing (102 cm), poor soil conditions and mrate temperatures, the semi-dwarf heat-resistantre not as competitive with weeds and can produce


grain than the standard-height lines, especially with in-effective weed management.

The heat-tolerance trait is associated with morerapid partitioning of carbohydrates to developing pods,under moderate or high temperatures, whereas the DLStrait is associated with the maintenance of greater car-bohydrate reserves in stem bases and probably roots(Gwathmey et al., 1992). Consequently, we testedwhether a negative interaction occurs between thesetraits (Ismail et al., 2000). We crossed a DLS heat-sensitive parent with a non-DLS heat-tolerant parentand bred four advanced populations with and with-out DLS and heat tolerance. These four populationswere evaluated in moderate-temperature field environ-ments where the soil organism (F. solani) for whichthe DLS trait confers resistance was present, and hotfield and glasshouse environments where the soil or-ganism was not present. We concluded from thesestudies that the DLS and heat-tolerance traits can beeffectively combined. Genotypes with both traits ex-hibited beneficial effects on grain yield in two cir-cumstances. The heat-tolerance trait enhanced pod setin hot environments, while the DLS trait enhancedplant survival whenF. solaniwas present in the soil,and there were only small detrimental interactive ef-fects due to the presence of both traits in the samegenotype.

Heat-tolerance during reproductive developmentmay depend on different membrane properties thanchilling tolerance during emergence. Consequently, itc iblet o bet bothc ler-a tionm icht sameg

A

DAS 08,U S-A ll.T f thea

References

Cisse, N., Ndiaye, M., Thiaw, S., Hall, A.E., 1995. Registration of‘Mouride’ cowpea. Crop Sci. 35, 1215–1216.

Cisse, N., Ndiaye, M., Thiaw, S., Hall, A.E., 1978. Registration of‘Melakh’ cowpea. Crop Sci. 37.

Ehlers, J.D., Hall, A.E., 1998. Heat tolerance of contrasting cowpealines in short and long days. Field Crops Res. 55, 11–21.

Ehlers, J.D., Hall, A.E., Patel, P.N., Roberts, P.A., Matthews, W.C.,2000. Registration of ‘California Blackeye 27’ cowpea. Crop Sci.40, 854–855.

Elawad, H.O.A., Hall, A.E., 2002. Registration of ‘Ein El Gazal’cowpea. Crop Sci. 42, 1745–1746.

Gwathmey, C.O., Hall, A.E., 1992. Adaptation to midseason droughtof cowpea genotypes with contrasting senescence traits. Crop Sci.32, 773–778.

Gwathmey, C.O., Hall, A.E., Madore, M.A., 1992. Pod removal ef-fects on cowpea genotypes contrasting in monocarpic senescencetraits. Crop Sci. 32, 1003–1009.

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 cow-pea, and comparison with other crops. In: Kuo, C.G. (Ed.), Pro-ceedings of the International Symposium on Adaptation of FoodCrops to Temperature and Water Stress, Taiwan, 13–18 August.Asian Vegetable Research and Development Center, Shanhua,pp. 271–284, Taiwan. Publ. No. 93-410.

Hall, A.E., Frate, C.A., 1996. Blackeye bean production in Califor-nia. Univ. Calif. Div. Agr. Sci. Publ. 21518, 23 pp.

Hall, A.E., Patel, P.N., 1985. Breeding for resistance to drought andheat. In: Singh, S.R., Rachie, K.O. (Eds.), Cowpea Research,Production, and Utilization. Wiley, New York, pp. 137–151.

Hall, A.E., Cisse, N., Thiaw, S., Elawad, H.O.A., Ehlers, J.D., Ismail,A.M., Fery, R.L., Roberts, P.A., Kitch, L.W., Murdock, L.L.,Boukar, O., Phillips, R.D., McWatters, K.H., 2003. Development

RSP.

H lant

I ef-i. 38,

I ow-p Sci.

I ithField

I ur-slow

I e-dling.

I af-dently

ould be hypothesized that it may not be posso combine these traits. This does not appear the case in that we have bred lines that havehilling tolerance during emergence and heat tonce during reproductive development. An interacay occur, however, that limits the extent to wh

hese tolerances can be further developed in theenotype.

cknowledgments

The research was partially supported by USpecific Cooperative Agreement No. 58-6659-8-0SDA NRICGP Award No. 98-35100-6129 and UID Grant No. DAN-G-SS-86-00008-00 to A.E. Hahe opinions and recommendations are those outhor and not necessarily those of USAID.

of cowpea cultivars and germplasm by the bean/cowpea CField Crops Res. 82, 103–134.

all, A.E., Singh, B.B., Ehlers, J.D., 1997. Cowpea breeding. PBreed. Rev. 15, 215–274.

smail, A.M., Hall, A.E., 1998. Positive and potential negativefects of heat-tolerance genes in cowpea lines. Crop Sc381–390.

smail, A.M., Hall, A.E., 2000. Semidwarf and standard-height cpea responses to row spacing in different environments. Cro40, 1618–1623.

smail, A.M., Hall, A.E., 2002. Variation in traits associated wchilling tolerance during emergence in cowpea germplasm.Crops Res. 77, 99–113.

smail, A.M., Hall, A.E., Close, T.J., 1997. Chilling tolerance ding emergence of cowpea associated with a dehydrin andelectrolyte leakage. Crop Sci. 37, 1270–1277.

smail, A.M., Hall, A.E., Close, T.J., 1999. Allelic variation of a dhydrin gene cosegregates with chilling tolerance during seeemergence. Proc. Natl. Acad. Sci. U.S.A. 96, 13566–13570

smail, A.M., Hall, A.E., Ehlers, J.D., 2000. Delayed-lesenescence and heat-tolerance traits mainly are indepenexpressed in cowpea. Crop Sci. 40, 1049–1055.


Marfo, K.O., Hall, A.E., 1992. Inheritance of heat tolerance duringpod set in cowpea. Crop Sci. 32, 912–918.

Nielsen, C.L., Hall, A.E., 1985a. Responses of cowpea (Vigna un-guiculata (L.) Walp.) in the field to high night air temperatureduring flowering. I. Thermal regimes of production regions andfield experimental system. Field Crops Res. 10, 167–179.

Nielsen, C.L., Hall, A.E., 1985b. Responses of cowpea (Vigna un-guiculata(L.) Walp.) in the field to high night air temperature dur-ing flowering. II. Plant responses. Field Crops Res. 10, 181–196.

Petrie, C.L., Hall, A.E., 1992. Water relations in cowpea and pearlmillet under soil water deficits. I. Contrasting leaf water relations.Aust. J. Plant Physiol. 19, 577–589.

Petrie, C.L., Kabala, Z.J., Hall, A.E., Simunek, J., 1992. Water trans-port in an unsaturated medium to roots with differing local ge-ometries. Soil Sci. Soc. Am. J. 56, 1686–1694.

Singh, B.B., Matsui, T., 2002. Cowpea varieties for drought toler-ance. In: Fatokun, C.A., Tarawali, S.A., Singh, B.B., Kormawa,P.M., Tamo, M. (Eds.), Challenges and Opportunities for Enhanc-

ing Sustainable Cowpea Production, World Cowpea ConferenceIII Proceedings, 4–8 September. International Institute of Tropi-cal Agriculture, Ibadan, Nigeria, pp. 287–300.

Singh, B.B., Mai-Kodomi, Y., Terao, T., 1999. A simple screeningmethod for drought tolerance in cowpea. Indian J. Genet. 59,211–220.

Thiaw, S., 2003. Association between slow leaf-electrolyte-leakageunder heat stress and heat tolerance during reproductive devel-opment in cowpea. Ph.D. Dissertation. University of California,Riverside, 100 pp.

Thiaw, S., Hall, A.E., 2004. Comparison of selection for either leaf-electrolyte-leakage or pod set in enhancing heat tolerance andgrain yield of cowpea. Field Crops res. 86, 239–253.

Thiaw, S., Hall, A.E., Parker, D.R., 1993. Varietal intercropping andthe yields and stability of cowpea production in semiarid Senegal.Field Crops Res. 33, 217–233.

Turk, K.J., Hall, A.E., Asbell, C.W., 1980. Drought adaptation ofcowpea. I. Influence of drought on yield. Agron. J. 72, 413–420.

Documents

Breeding for adaptation to drought and heat in cowpea