search
for
 About Bioline  All Journals  Testimonials  Membership  News


African Crop Science Journal
African Crop Science Society
ISSN: 1021-9730 EISSN: 2072-6589
Vol. 6, Num. 3, 1998, pp. 215-225
African Crop Science Journal, Vol

African Crop Science Journal, Vol. 6. No. 3, pp. 215-225,

AN ADAPTATION BREEDING STRATEGY FOR WATER DEFICIT IN BEAN DEVELOPED WITH THE APPLICATION OF THE DSSAT3 DRYBEAN MODEL

C. S. Wortmann

CIAT, Pan Africa Bean Research Alliance, P.O. Box 6247, Kampala, Uganda

(Received 23 December, 1997; accepted 21 May, 1998)

Code Number:CS98024
Sizes of Files:
      Text: 76K
      Graphics: Line drawings and tables (gif) - 25K

ABSTRACT

Bean (Phaseolus vulgaris L.) productivity is much constrained by water deficits in many production areas in eastern and southern Africa. The DSSAT 3 Drybean model was used to analyze the effects of water deficits on bean using genetic coefficients of four cultivars including an early, determinate type and early, intermediate and late maturity indeterminate bush types. More than 2300 simulations were run using meteorological data from 19 locations. The late maturing ideotype gave highest yield in less stressful environments but its yield was the least stable. In more stressful environments, early maturing ideotypes had highest mean yield and their yield was most stable in all environments. Stress was most severe and frequent in the later stages of growth; stress affected yield most during pod formation and fill, less during the early reproductive stage and least during the vegetative stage. Elements of a possible strategy for improving adaptation to water deficits are discussed, and specific information for the application of such a strategy is given for three locations which appear suitable as primary screening sites.

Key Words: Africa, bean, crop growth simulation, DSSAT3 Drybean Model, water deficit

RÉSUMÉ

La productivité du haricot (Phaseolus vulgaris L.)est bien contrainte par les déficits hydriques dans beaucoup d'aires de production en Afrique de l'est et du sud. Le DSSAT3 drybean model a été utilisé pour analyser les effets de déficits hydriques sur le haricot en utilisant les coefficients génétiques de quatre cultivars comprenant les types précoces déterminés et les types nains indeterminés à maturité intermédiaire et tardive. Plus de 2300 simulations étaient performées en utilisant les données météorologiques de 13 stations écologiques. L'idiotype à maturité tardive a donné le plus haut rendement dans les environnements moins stressants mais son rendement a été le moins stable. Dans les environnements plus stressants, les idiotypes à maturité précoce ont eu le plus haut rendement et leur rendement a été le plus stable dans tous les environnements. Le stress était plus sévère et fréquent dans les derniers stades de croissance et a affecté plus le rendement pendant la période de formation et de remplissage des gousses; il était moins durant le stade précoce de reproduction et moins plus durant le stade végétatif. Les éléments d'une stratégie possible pour améliorer l'adaptation aux déficits hydriques sont analysés et l'information spécifique pour l'application d'une telle stratégie est donnée pour trois stations écologiques qui apparaissent convenables comme sites primaires pour le criblage.

Mots Clés: Afrique, haricot, simulation de la croissance de culture, DSSAT3 Drybean Model, Déficit hydrique

INTRODUCTION

Bean (Phaseolus vulgaris L.) crop performance in eastern and southern Africa is severely constrained by periodic water deficits in some production areas (Fig. 1). Frequency of occurrence of water deficits, severity of stress, timing of stress relative to plant age, and sensitivity of the plant at different stages of growth interact to determine yield loss associated with water deficits. These factors have not been quantified for bean production in Africa but the information is potentially useful for identifying some characteristics of bean ideotypes for stress environments and for devising a breeding strategy for drought tolerance.

Figure 1: Map of the relative importance of soil water deficits in bean production areas in sub-Saharan Africa.

Three classes of mechanisms of drought tolerance might be exploited for improving bean for water deficit conditions (White and Castillo, 1992). Drought escape might be achieved with early maturing genotypes or through plasticity such that drought stress accelerates the maturation process; the first is highly heritable but associated with lower yield potential than with later maturing genotypes. Drought tolerance with high plant water potential or drought avoidance might be achieved by plants conserving water or extracting a greater amount of soil water: studies involving use of carbon isotope discrimination (CID) (Ehleringer, 1988; White et al., 1990) to evaluate water use efficiency show large differences between genotypes, although CID is not always related to genotypic differences for yield under water deficit (White et al., 1994a); differences in root growth on deep soils are related to ability for water extraction (Sponchiado et al., 1989; White and Castillo, 1992); and drought yield has been correlated with a leaf thickness index (White and Izquerido, 1991). Drought tolerance with low water potential might be improved with increased desiccation tolerance or better maintenance of turgor (White and Izquierdo, 1991). Bean is especially sensitive to water deficit stress during the early podfill stage, less so during the vegetative stage and little during late podfill.

Narrow sense heritability for yield under water deficit can be high (White et al., 1994b) ranging from 0.09 to 0.75 for different environments. Realised gains in seed yield ranged from 0.4 - 7.4% and 2.9 - 15.7% at locations in Mexico and Colombia, respectively, due to selecting the top 20% from the F2 generation. However, performance under water deficit is related to environment; general combining abilities of parents tended to be higher in their environments of origin than in other environments. This may indicate the importance of good adaptation to drought tolerance.

Computer simulation models are able to capture nuances in weather and estimate their effects on crop growth giving a better understanding of the effects of soil water deficits. The DSSAT3 Drybean Crop Growth Model predicts dry matter growth, leaf area index, crop development and final seed weight and other yield components of bean. This is a function of daily weather data, characteristics of a one-dimensional multi-layered soil profile and crop management conditions. Genetic coefficients are required to simulate the differences in crop performance among varieties. Complete documentation of the DSSAT3 Drybean model is provided elsewhere (Hoogenboom et al., 1994; Tsuji et al., 1994). The Drybean Model has not been verified for prediction of bean yield in Africa where productivity is constrained by numerous abiotic and biotic stresses which the model does not consider. However, it appears to be particularly well suited for research on water deficits, assuming absence of other constraints, as it considers soil water and daily changes in weather (CIAT, 1992).

MATERIALS AND METHODS

Four genetic ideotypes were used in this study represented by the genetic coefficients previously estimated for four cultivars (Table 1; Tsuji et al., 1994). Seafarer is an early maturity, determinate type; Rabia de Gato, Kilyumukwe and Carioca are early, medium and late maturity types, respectively, of indeterminate bush growth habit.

Meteorological data were obtained from 19 locations. The locations were grouped into five climatic categories for eastern and southern Africa

i.e., I Sub-humid, medium altitude, low latitude; II Sub-humid, high altitude, low latitude; III Semi-arid, low latitude; IV. Sub-humid, medium latitude; and V. Semi-arid, medium latitude (Table 2).

TABLE 1. Characteristics of the genotypes used in the study and days required for emergence to first flower (S1), first flower to first seed (S2) and first seed to physiological maturity (S3) when grown near the equator at 1200 m asl

Genotype

Growth characteristics

S1

S2

S3

Seafarer

Early maturity, determinate

26

10

23

Rabia de Gato

Early maturity, indeterminate

26

9

21

Kilyumukwe

Medium maturity, indeterminate

35

9

21

Carioca

Late maturity, indeterminate

35

12

24

TABLE 2. Characteristics, years of simulations and planting dates (PD) for the locations of eastern and southern Africa used in analysing the effects of water deficits on bean yield

Location

Lat

Long.

Alt.

Years

PD1

PD2

PD3

PD4

Category I, sub-humid, medium altitude, low latitude (n= 580)

Ikulwe

0.6

33.7

1200

10

15/3

1/4

15/9

1/10

Kisumu

-0.1

34.7

1146

12

15/3

1/4

15/9

1/10

Kawanda

0.3

32.6

1195

7

15/3

1/4

15/9

1/10

Masaka

-0.3

31.7

1313

3

15/3

1/4

15/9

1/10

Lira

2.3

32.9

1085

10

15/7

1/8

Category Ii, sub-humid, high altitude, low latitude (n= 744)

Embu

0.5

37.4

1493

10

1/4

1514

1/9

15/9

Meru

-0.1

37.6

1554

12

15/3

1/4

15/10

1/11

M bale

1.0

37.1

1494

9

15/3

1/4

15/7

1/8

Kitale

1.0

35.0

1875

12

1/3

15/3

15/8

1/9

Bushenyi

-0.5

30.2

1616

6

15/3

1/4

15/9

1/10

Category Ill, semi-arid low latitude (n= 404)

Kilimanjaro Airport

-3.4

37.1

896

12

20/3

1/4

15/11

Katumani

-1.6

37.3

1646

9

1/11

15/11

15/3

30/3

Nakuru

-0.3

36.1

1871

12

15/3

1/4

1/10

15/10

Category IV, medium latitude (n= 520)

Awasa

7.0

38.5

1750

12

15/4

15/7

30/7

Harare

-17.9

31.1

1478

12

1/12

30/12

Rusape

-18.5

32.1

1430

7

15/1

15/12

30/12

Mbeya

-8.9

33.5

1704

8

X

X

15/2

1/3

Morogoro

-6.8'

37.6

526

6

1/3

15/3

15111

1/12

Category V, semi-arid medium latitude (n= 64)

Bulawayo

-20.0

28.6

1326

9

1/12

30/12

Lat., Long. and Alt. refer to the longitude, latitude and altitude of the locations

Locations of Category I fall in areas where soil water deficits are of moderate importance (Fig. 1). Category II overlaps areas where water deficits are rated as being of moderate or high importance. Deficits are of very high importance at the sites of Category III. Semi-arid Category V falls outside of the bean production area shown for Zimbabwe due to the low importance of bean around Bulawayo. Some sites in Category IV are in areas where water deficits are of low importance, but some common sowing dates do result in periodic stress. Early sowing dates which result in low probability of water deficits were excluded for Lira, Awasa and Mbeya; for these locations, the analyses were for typical mid-season sowing times whereby the bean crop takes advantage of the late rains but is more likely to encounter water deficits. Assuming that farmers generally sow only after enough rain has fallen to establish the crop, simulations were excluded if cumulative rainfall was less than 18 mm at one week after sowing. Actual daily rainfall data were used in all places. Daily maximum and minimum temperatures were available for most locations, but estimated monthly means were used for some locations in Category I and III. Estimated monthly mean values were used for solar radiation. WGEN option of Weatherman (Pikering et al., 1994) was used to estimate missing values. The profile description for a deep sandy clay loam soil of moderately low pH was used for all simulations; the soils and landforms in all major production areas are heterogeneous and it is recognised that the soil profile description used may have caused over- or under-estimation of the typical soil water availability for some locations. Simulations started two weeks before sowing with the water holding capacity of the soil at 20% of full. The nitrogen sub-routine of the model was switched off so nutrient supply to the plants was not limiting. Also, the effects of diseases and insect pests were not considered by the model although these can be important to a crop's water use efficiency and tolerance to water deficits. The simulations were done for sole crops of bean; intercropping effects were not considered due to limitations of the model.

The importance of stress at a location was determined for each of three growth stages as the sum of the products of frequency of occurrence and severity of stress. The stages of growth were emergence to first flower (S1), first flower to first seed (S2) and first seed to physiological maturity (S3). Severity was grouped as <0.10, 0.10 - 0.39, 0.40 - 0.69, 0.70 - 1.00, with 0.00 and 1.00 equal to no stress and plant death due to stress, respectively.

Optimal sowing dates of screening for adaptation to water deficits were determined for three locations representing Categories II, III and IV. The three locations have bean research programmes; proximity to the equator was considered to have temperatures and day lengths similar to those of the normal bean production seasons. Simulations across years were done for several sowing dates for each location and the optimal sowing date was considered to be the date that gave medium yield of 40% of the potential yield. Initial soil water was assumed to be 70% of field capacity as sowing normally commences once the early rains for the season have fallen, but before the profile has completely filled. Simulation was started 2 weeks before the sowing date. The same sandy clay loam profile as above was used.

RESULTS

TABLE 3. Stress levels at three stages of growth (S 1-3) for four bean ideotypes in five climatic categories of eastern and southern Africa as indicated by the sum of products of stress level and frequency

Climatic categorya

Seafarer

Rabia de Gato

Kilymukwe

Carioca

 

S1

S2

S3

S1

S2

S3

S1

S2

S3

S1

S2

S3

I

0.025

0.046

0.094

0.027

0.050

0,103

0.027

0.063

0,094

0.060

0.163

0.276

ll

0.053

0.071

0.196

0.058

0.063

0.172

0.051

0.112

0.184

0.052

0.121

0.278

Iii

0.046

0.166

0.289

0.073

0.136

0,241

0.114

0.171

0.323

0.100

0.192'

0.413

IV

0.034

0.089

0.155

0.031

0.091

0.155

0.040

0.101

0.157

0.041

0.105

0.220

V

0.050

0.169

0.337

0.050

0. 194

0.319

0.069

0.225

0.369

0.081

0.300

0.356

a| = Sub-humid, medium altitude, low latitude; II = Sub-humid, high altitude, low latitude; III = Semi-arid, low latitude; IV = Sub-humid, medium latitude; and V = Semi-arid,medium latitude

Water deficit stress was most important for Categories III and V (Fig.2; Table 3) with 48-49% mean yield loss and approximately 50% of the cases having a yield loss of 50% or more due to inadequate water. The model estimated yield loss to be 30-35% for Categories II and IV, but water is adequate in 50% of the years to achieve at least 75% of the yield potential. Category I was least affected by water deficits with a 15% mean yield reduction and with adequate water to achieve at least 90% of the potential yield in 50% of the cases.

The Seafarer, Rabia de Gato and Kilyumukwe ideotypes experienced similar levels of stress in Categories I, II and IV. Stress was greatest for late maturing Carioca in most cases. The results do not indicate any difference due to determinacy of growth habit.

The late maturing ideotype, Carioca, gave the highest mean simulated yield in Categories I and IV, but had little or no advantage in the other Categories (Table 4). Carioca also suffered the greatest yield reduction due to stress, and had the least stable yield. The intermediate maturity ideotype, Kilyumukwe, had a yield advantage over the earlier maturity ideotype in Category I only, but with less stability.

Drought escape is important in Category III and V, and early maturity cultivars have the most potential. Early maturity cultivars may be preferred for Category II as well; although mean yields are slightly less, yield is more stable with the short season ideotypes. In Category IV as well, small scale farmers may prefer to reduce risk, and sacrifice some yield potential, by using earlier maturing cultivars.

Early cessation of rainfall and depletion of soil water reserves resulted in the most severe and most frequent occurrence of stress during the seedfill stage (Table 4). Stress levels increased as the crop aged, and the seed fill stage of growth (S3) was most affected by water deficits due to a higher frequency and greater severity of stress than at other stages. Stress was more severe for the later maturing ideotypes for all stages of growth.

TABLE 4. Mean yields (and SE's) and yield reductions due to water deficits for four bean ideotypes in six climatic categories of eastern and southern Africa

Ideotype

Yield, kg ha-1

S.D. for yield

% yield reduction

Climatic category I, n = 580

Seafarer

2753

680

14.2

Rabia de Cato

2650

709

16.6

Kilymukwe

3020

936

18.2

Carioca

33 5 9

1039

20.6

Climatic category II, n = 744

Seafarer

2259

894

29.9

Rabia de Cato

2307

873

27.3

Kilymukwe

2235

1185

35.9

Carioca

2450

1334

37.8

Climatic category III, n = 408

Seafarer

1899

1079

42.6

Rabia de Cato

1925

1115

41.6

Kilymukwe

1778

1196

51.7

Carioca

1743

1323

58.0

Climatic category iV, n = 520

Seafarer

2494

967

28.2

Rabia de Cato

2469

1001

30.6

Kilymukwe

2594

1142

30.4

Carioca

2704

1292

35.0

Climatic category V, n = 64

Seafarer

1949

1153

43.2

Rabia de Cato

1828

1047

47.2

Kilymukwe

1921

1281

52.2

Carioca

1999

1385

53.0

Water deficit stress was negatively correlated with yield (Table 5). The relationship was strongest during the S2 and S3 stage, emphasising the need for more tolerance during the later stages of growth.

TABLE 5. Correlation coefficients for bean yield with water deficit stress at three stages of growth for four bean ideo.types in five climatic categories of eastem and southern Africa

Climatic category

Seafarer

Rabia de Gato

Kilymukwe

Carioca

 

S1

S2

S3

S1

S2

S3

S1

S2

S3

S1

S2

S3

I

-0.50

-0.70

-0.84

-0.60

-0.63

-0.76

-0.63

-0.78

-0.85

-0.53

-0.74

-0.84

II

-0.21

-0.69

-0.86

-0.28

-0.67

-0.80

-0.18

-0.74

-0.93

0.09

-0.76

-0.94

III

-0.55

-0.83

-0.87

-0.36

-0.79

-0.77

-0.52

-0.74

-0.86

-0.45

-0.71

-0.88

IV

-0.59

-0.79

-0.84

-0.62

-0.76

-0.82

-0.61

-0.74

-0.81

-0.54

-0.67

-0.78

V

-0.65

-0.53

-0.76

-0.72

-0.57

-0.88

-0.52

-0.83

-0.94

-0.44

-0.79

-0.87

a|= Sub-humid, medium altitude, low latitude; II = Sub-humid, high altitude, low latitude; III = Semi-arid, low latitude; IV = Sub-humid, medium latitude; and V = Semi-arid, medium latitude
b S1 = first flower to first seed; S2 = first seed to physiological maturity; and S3 = Severity was grouped as <0.10, 0.10 - 0.39, 0.40 - 0.69, 0.70 - 1.00, with 0.00 and 1.00 equal to no stress and plant death due to stress.

Three sites have been considered here as primary screening sites (Table 6). Optimal sowing dates for screening for water deficits for these potential sites were determined to be: Embu (Category II), 5 May for short season germplasm; Katumani (Category III), 15 April for short season germplasm; and Mbeya (Category IV), 10 March for short season and 5 March for long season germplasm. Mbeya appears to be the least satisfactory for screening as the cessation of rains is most variable resulting in much variation in stress during the screening period.

TABLE 6. Optimal sowing Oates for screening bean for tolerance to water deficits under natural rainfall conditions at four locations in eastern and southern Africa

Location

Embu

Katumani

Mbeya

Mbeya

Maturity group

Early

Early

Early

Late

Sowing date

5 May

15 Apr.

10 Mar.

5Mar.

Stress at first seed

0.18

0.13

.17

0.17

Stress during seed fill

0.54

0.47

0.55

0.53

Yield reduction (YR) (%)

60

60

61

62

Std. Dev. of YR1

12

17

30

26

1 = Std. Der. of YR: standard deviation of the reduction in yield across years

DISCUSSION

The DSSAT3 Drybean Model as a tool in bean breeding. The Drybean model has been informative in the study of water deficits for:

  • assessment of the effects of water deficits on bean productivity in different Categories,
  • assessment of the interaction effects of maturitytimes of cultivars and water deficits on yield and its stability,
  • determination of the stage of growth for which tolerance is most important,
  • evaluation of the potential of sites for screening under conditions of natural rainfall, and
  • determination of likely sowing dates for achieving desired stress conditions.

The model did not reveal any difference due to the use of indeterminate lines, while an advantage is expected during the S2 stage of growth as prolonged flowering increases the probability of pollination and seed set occurring when there is adequate rainfall.

For the more extreme categories, the results generally agree with the author's expectations. Early maturity for drought escape has been perceived as important in Category III. Much on-farm testing has been done in Category I locations of cultivars of varying time to maturity and farmers seldom express concern about later maturity. Similarly, in parts of Malawi which would be in Category IV, farmers seldom express concern about time to maturity. These observations suggest that either early or late maturity cultivars can be acceptable to farmers in Categories I and IV.

Simulated yields are much higher than yields typically obtained in the field where crops encounter numerous biotic and abiotic constraints to yield. These constraints are likely to affect the ability of plants to use soil water efficiently. Uptake may be inhibited by damaged root systems due to insect or disease damage. Defoliation due to disease or insect attack affects transpiration, as well as growth, and water uptake and use efficiency are likely to be affected. Nutrient inadequacies are expected to result in less root growth and less capacity for water uptake, and in less shoot growth with reduced transpiration. The cumulative effects may result in more or less stress depending on the distribution of rainfall during the growing season.

We opted to switch off the nitrogen sub-routine in order to consider water deficits alone. However, N deficiency is often a major constraint to crop production. Adequacy of N might result in much growth initially and earlier depletion of water reserves, but the root system should be more efficient in capture of deep water. The effects of N availability are expected to vary with rainfall distribution.

The model would be more useful for such ex ante evaluation of bean ideotypes if certain physiological processes could be adjusted by the user. It would have been of interest to adjust the root to shoot growth ratio, or even root architecture, and determine the simulated effects on performance with water deficits. The roles of leaf size, thickness, and orientation in performance with water deficits might be investigated with a more flexible model.

A regional bean breeding strategy for tolerance to water deficits. The results indicate that improved adaptation to conditions of water deficit stress is very important for Categories III and V, and of importance in Categories II and IV as well. Early maturing cultivars are likely to be more productive and more stable than late maturing cultivars in some environments.

A programme for improving adaptation to water deficits may include one or more of the following: selection of drought tolerant lines from existing cultivars and breeding lines; incorporation of drought tolerance in commercial cultivars; improvement of breeding parents for drought tolerance; and breeding for drought tolerance per se. Improved stress escape, and/or tolerance during the seedfill stage of growth, is most needed.

Stress levels vary much from season to season (Fig. 2). Therefore, breeders should evaluate materials either under field conditions with controlled water supply, or by planting towards the predicted end of the rainy periods to achieve good early growth with stress during the later stages of growth. Initially, priority might be given to identification of well-adapted cultivars that perform well in spite of water deficits, but eventually breeding for tolerance should be emphasised. Test materials might be of three types: materials known to perform well in spite of water deficits; agronomically superior materials from national and regional breeding programmes in Africa as these materials are likely to be otherwise well adapted and of preferred types; and populations bred for drought tolerance beginning evaluations with the F3 generation (White et al., 1994).

Figure 2: Cumulative frequency of yield production due to soil water deficits in six climatic categories of Eastern and Southern Africa.

A regional screening approach might be similar to that used for screening for tolerance to low soil fertility conditions (Wortmann et al., 1995). A large set of well-adapted entries (>100) of the desired maturity range would be screened at one or two primary screening site for two seasons. Then, the most promising materials (best 10 to 15% for drought yield) can be evaluated more widely. Environment effects, especially temperature, on the expression of drought tolerance are expected (White et al., 1994) and good selection of primary screening sites is crucial.

CONCLUSION

Water deficits are a major constraint to bean productivity in many production areas of eastern and southern Africa, reducing mean yields by 50% or more in some semi-arid areas. The later maturing ideotypes have the greatest yield potential in non-stress environments, but where water deficits are often severe, the early maturing ideotype gives higher mean yield. Yield was also more stable with the early maturing type in stress environments; this stability may be more attractive to risk-prone small scale farmers than a slight advantage in mean yield. The results indicate that drought escape through the use of early maturing varieties provides a valuable defence against unpredictable water deficits.

Achieving improved adaptation to water deficits is a major challenge and should be approached in a collaborative manner at a regional level. Preliminary screening of lines proven to be well adapted in the region for tolerance to stress at a few locations where stress can be well managed can be followed by multi-location testing of the most promising lines at drought prone sites throughout the region. This needs to be complimented by breeding to incorporate tolerance traits into preferred and agronomically superior cultivars.

ACKNOWLEDGEMENTS

The Director of Kawanda Agricultural Research Institute of NARO enabled this work by providing facilities to CIAT staff in Uganda. Dr. Philip Thornton provided consultative support and assisted with meteorological data sets used in the study. The financial assistance was provided by the Canadian International Development Agency, Swiss Development Co-operation and the United States Agency for International Development for this and other CIAT activities in Africa.

REFERENCES

CIAT (Centro International de Agricutura Tropical), 1992. Bean Program Annual Report, 1992. CIAT, pp: 120-124.

Ehleringer, J.R. 1988. Correlations between carbon isotope ratio, water-use efficiency and yield. In: Proceedings of the International Bean Drought Workshop, Working Document No. 41, 19-21 October, 1987, pp. 165-191.

Hoogenboom, G., White, J.W., Jones, J.W. and Boote, K.J. 1994. BEANGRO: a process-oriented dry bean model with a versatile user interface. Agronomy Journal 86:182-190.

Pickering, N.B., Hansen, J.W., Jones, J.W., Wells, C.M., Chan, V.K. and Godwin, D.C. 1994. Weatherman: a utility for managing and generating daily weather data. Agronomy Journal 86:332-337.

Richardson, C.W. and Wright, D.A. 1984. WGEN: A model for generating daily weather variables. U.S. Department of Agriculture, Agricultural Research Service, ARS-8.

Sponchiado, B.N., White, J.W., Castillo, J.A. and Jones, P.G. 1989. Root growth of four common bean cultivars in relation to drought tolerance in environments with contrasting soil types. Experimental Agriculture 25:249-257.

Tsuji, G.Y., Uehara, G. and Balas, S. (Eds.). 1994. DSSAT v3. University of Hawaii, Vol. 1-3.

White, J.W., Castillo, J.A. and Ehleringer, J. 1990. Associations between productivity, root growth and carbon isotope discrimination in Phaseolus vulgaris under water deficit. Australian Journal of Plant Physiology 17:189-198.

White, J.W. and Izquierdo, J. 1991. Physiology of yield potential and stress tolerance. IN van Schoonhoven, A. and O. Voysest (Eds.) Common Beans: Research for Crop Improvement, CABI and CIAT. pp. 339-359.

White, J.W. and Izquierdo, S.P. 1991. Breeding for adaptation to drought. In: Common Beans: Research for Crop Improvement. van Schoonhoven, A. and Voysest, O. (Eds.), pp. 501-560. CABI and CIAT.

White, J.W. and Castillo, J.A. 1992. Evaluation of diverse shoot genotypes on selected root genotypes of common bean under soil water deficits. Crop Science 32:762-765.

White, J.W., Castillo, J.A., Ehleringer, J.R., Garcia-C. J.A. and Singh, S.P. 1994a. Relations of carbon isotope discrimination and other physiological traits to yield in common bean (Phaseolus vulgaris) under rainfed conditions. Journal of Agricultural Science, Cambridge 122:275-284.

White, J.W., Ochoa, M. R., Ibarra, F. and Singh, S.P. 1994b. Inheritance of seed yield, maturity and seed weight of common bean (Phaseolus vulgaris) under semi-arid rainfed conditions. Journal of Agricultural Science, Cambridge 122:265-273.

Wortmann, C.S., Lunze, L., Ochwoh, V.A. and Lynch, J. 1995. Bean improvement for low fertility soils in Africa. African Crop Science Journal 3:469-477.

Copyright 1998, African Crop Science Society


The following images related to this document are available:

Line drawing images

[cs98024a.gif] [cs98024b.gif]
Home Faq Resources Email Bioline
© Bioline International, 1989 - 2024, Site last up-dated on 01-Sep-2022.
Site created and maintained by the Reference Center on Environmental Information, CRIA, Brazil
System hosted by the Google Cloud Platform, GCP, Brazil