l on}? llllvlllfill'll.‘ {$.{OEI . ‘.1' I‘...- it uttl ..I‘rl o VII-‘ ‘DI’IIIHII vn Iy‘yhi‘r‘hloikl‘hic I It I -5 VI‘ t I't. l fl | 'I {Kill ‘0’ I." ' I‘d-r n. )IID'... 1 ll.‘ \dl'ltnlliili%i “IFIFI‘ llll.‘l!\l3l'l‘” , ‘ - .; IHAIM a. 1].... co .HIItllnfllu Ill-VJIHJV fit 1 : I‘D |M\I I iii-.“Il' tut u! ‘ .I 0.11) ».|| 1 0\ I (1 I ~ § Q it'll itl‘s‘t Ln] I II . iii ‘3 $11!“: )O‘vnlllduflluldo ‘ v 4. . . . J ‘ .Inlu .l. h|.n\§ollcla’}1nlul Eli \ In. III: .I) ‘3... 1, ‘Iuwfilt'i‘li «I. It ‘Gfllunfit t (.4 il‘a‘l‘ '7 .1‘4 tu.‘ . In I a t L o I x \ 4’. .‘. c. ”Jill!" AWWAs‘QKGoiQ In“ \H. . 1d . rt rflflkfi ‘ gnaw“: .l‘g...‘ agitéuulwfi 5,4113. - . - n r1 . h. ‘ .ch' ‘ l yr. .7 II. ,AI. . I ..4!A\.l .fla..l . I . . .. u: AI‘ W». . .IW ‘lr‘l ,- LIBRARY Michigan State University This is to certify that the dissertation entitled PRODUCTIVITY OF MALAWIAN LANDRACE DRY BEANS UNDER INTERCROPPING AND DROUGHT CONDITIONS presented by Alexander B. C. Mkandawire has been accepted towards fulfillment of the requirements for Ph.D. Crop and Soil Sciences degree in V/(,/./ — //_7‘9‘ ///!4_ ,42/777’é. Major professor /‘ __ , , , - Date A 4,"? '—/ 1-,1/lf L]. MS U it an Affirmative Action/Equal Opportunity Institution 0-12771 I f F \ MSU RETURNING MATERIALS: Place in book drop to remove this checkout from LIBRARIES —:__- your record. .FINES will be charged if book is returned after the date stamped below. 3 U '7 060 PRODUCTIVITY OF MALAWIAN LANDRACE DRY BEANS UNDER INTERCROPPING AND DROUGHT CONDITIONS By Alexander B. C. Mkandawire A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Science 1987 /§l35’5 1L;Q‘7 ABSTRACT PRODUCTIVITY OF MALAWIAN LANDRACE DRY BEANS UNDER INTERCROPPING AND DROUGHT CONDITIONS By Alexander B. C. Mkandawire Malawi is a country of great ecological diversity. The original bean (Phaseolus vulgaris L.) introductions. brought more than 300 years ago, have evolved with great morphological variation. The process of evolution is greatly influenced by two environmental parameters; namely: temperature and moisture. In Malawi, moisture seems to exert greater influence on net primary productivity than temperature. Studies were conducted to better understand the role of Malawian landrace bean components in bean crop productivity under conditions of limited soil moisture. In one set of experiments limited moisture conditions were imposed on the beans by intercropping them in the same row with a more aggressive species [maize (Zea mays L.)], which is a popular cropping system in Malawi. The other set of experiments involved controlled drought in field-simulated culture. The aim here was to observe any genotypic variability in both Phaseolus and Vigna species in water—use efficiency (WUE) under either well-watered or drought conditions. Intercropping beans and maize resulted in significantly lower soil moisture than when beans were grown in monoculture stands. This resulted in significant reductions in bean seed yields of some landraces under the intercropping system. However. there was no evidence for reduction in seed yield of other landraces grown under the same system. This indicates that some of these landraces are adapted to such intercropping stresses. Criteria of estimating relative drought tolerance substantiated this observation. Most of those landraces that were adapted showed higher leaf diffusive resistances. Combined over all experiments, under stress conditions, economic yield was significantly and positively (r=0.242) associated with leaf diffusive resistance but significantly and negatively (r=-O.472; r=-O.243) associated with leaf transpiration and leaf temperature, respectively. However, from data combined from one intercropping and one WUE experiment, no significant relationship was obtained between total biological yield and either leaf moisture retention capacity (LMRC) or specific leaf weight (SLW) under stress conditions. And there was no significant relationship between LMRC and SLW. Genetic variability exists among Phaseolus and Vigna species in their WUB and drought tolerance. Vigna species performed similarly to Phaseolus species agronomically although they were expected to show higher water-use efficiency and drought tolerance. In one of the two WUB experiments the Malawian landraces significantly increased their WUE under drought such that they yielded equally well under drought as under well-watered conditions. The Malawian landrace component 5 (Katolika) proved to be a drought tolerant genotype in 3 years. The genetic variability in the Malawian component landraces in drought tolerance suggests that the mixtures planted by farmers in Malawi are comprised of both higher yielding but probably susceptible and the lower yielding but drought tolerant components. This provides one explanation of why Malawian farmers grow bean mixtures. They may want to maximize seed yields during good years by planting landraces of higher yielding potential while at the same time minimizing yield losses, in the event of a drought, by including drought tolerant landrace components. DEDICATION To my parents, Benard and Tamala Mkandawire and my parents—in-law, Patson and Miriam Khonje. 111 ACKNOWLEDGMENTS I wish to express my sincere gratitude to Dr. M. Wayne Adams for serving as my major professor. He directed both my program of study and the research I conducted. I thank him for reading the manuscript and providing the necessary professional criticisms. I fell honored to have been one of his graduate students. I deeply appreciate the financial support from the Bean/Cowpea CRSP Project of which he is the U.S. Principal Investigator. I would also like to thank Drs. C. E. Cress, A. R. Putnam, and J. D. Kelly for serving on my advisory committee, providing me with wisdom through classroom instruction and reading the manuscript. I will always cherish the good moments my family enjoyed at the Cress’s (1984) and the Kelly’s (1986) homes for 'Thanksgiving’ dinners. More importantly, we enjoyed corresponding with parents, brothers, sisters, and a lot of friends back in Malawi. But God so loved my father-in-law that He called him before I completed my studies so that we could see each other again. May God bless them all. Lastly, but most important, I extend my love once again to my wife, Sheila Easter, our two daughters Maudie and Lusungu, and our son Alexander Jr., for making me realize the joy of going to graduate school while raising a family at the same time. May the Lord bless them too. iv TABLE OF CONTENTS Page LIST OF TABLES O O O O O O O 0 O O O O O 0 O 0 O 0 Vi LIST OF FIGURES 0 O O O O O O O O O O O O O O O O O x I NTRODUCT I ON C O O O O O O O O O O O O O O I O O O 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . 5 Definition of Drought . . . . . . . . . . . . 5 Measuring Plant Water Status. . . . . . . . . 6 Drought Resistance Mechanisms . . . . . . . . 11 Water-use Efficiency (WUE). . . . . . . . . . 3O Drought and Crop Yield. . . . . . . . . . . . 33 MATERIALS AND METHODS. . . . . . . . . . . . . . . 37 Intercropping Experiments . . . . . . . . . . 40 East Lansing 1984. . . . . . . . . . . . 40 Montcalm 1985. . . . . . . . . . . . . 44 Water-use Efficiency (WUE) Experiments. . . . 46 East Lansing 1985.. . . . . . . . . . . 46 East Lansing 1986. . . . . . . . . . . . 48 Statistical Analysis. . . . . . . . . . . . . 50 RESULTS 0 O O O O O O O 0 O O O 0 0 O O 0 O O O O O 5 l Microclimatology. . . . . . . . . . . . . . . 51 Intercropping Experiments . . . . . . . . . 56 Water-use Efficiency (WUE) Experiments. . . . 87 DISCUSSION 0 O O I O C O O O O O I O O O O O O O O 104 SUMMARY AND CONCLUSIONS. . . . . . . . . . . . . . 115 LITERATURE CITED 0 O O O O O O O O O O O O O O O O l 1 7 APPENDIX 0 O O O O O O O O O O O O O O O O O O O O l 27 Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Table 10. Table 11. Table 12. LIST OF TABLES Comparison of environmental parameters by their effect on net primary productivity across 19 locations in Malawi. . . . . . . Bean genotypes used in the water-use efficiency experiment (1986) . . . . . . . Leaf diffusive resistance of 18 Malawian landrace dry beans at 70 DAP (1984). . . . Leaf diffusive resistance of 14 Malawian landrace dry beans at 78 DAP (1984). . . . Leaf diffusive resistance of 12 Malawian landrace dry beans at 85 DAP (1984). . . . Leaf transpiration of 18 Malawian landrace dry beans at 70 DAP (1984) . . . . . . . . Leaf transpiration of 14 Malawian landrace dry beans at 78 DAP (1984) . . . . . . . . Leaf transpiration of 12 Malawian landrace dry beans at 85 DAP (1984) . . . . . . . . Leaf temperature of 18 Malawian landrace dry beans at 70 DAP (1984) . . . . . . . . Leaf temperature of 14 Malawian landrace dry beans at 78 DAP (1984) . . . . . . . . Leaf temperature of 12 Malawian landrace dry beans at 85 DAP (1984) . . . . . . . . Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 18 Malawian dry beans at 74 and 72 DAP, respectively (1984). . . . . . . . . . . . V1 Page 39 49 59 60 61 63 65 66 67 68 69 71 Table Table Table Table Table Table Table Table Table Table Table Table l3. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. (1985) (1985) Yield differential, geometric mean, and drought susceptibility index (8) index (8) of 16 Malawian landrace dry beans grown in monoculture and intercropping with maize (1984). Yield differential, geometric mean, and drought susceptibility index (8) of 22 dry bean genotypes grown in monoculture and intercropping with maize (1985) vii Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 18 Malawian dry beans grown in three different soil types (1984). Biological and economic yield of 16 Malawian dry bean genotypes. arithmetic mean, Leaf diffusive resistance, transpiration, dry bean genotypes (1985). leaf and leaf temperature of 22 Number of pods/plant and number of seeds/plant of 16 Malawian dry bean genotypes (1984) Other agronomic characteristics of 16 Malawian landrace dry beans (1984) arithmetic mean, Number of pods/plant and number of seeds/plant of 22 dry bean genotypes Yield stability analysis of 18 Malawian dry bean genotypes across 16 locations in Malawi Apparent harvest index (AHI) and loo-seed weight of 16 Malawian dry bean genotypes (1984) Economic yield at physiological maturity ,and leaf moisture retention capacity (LMRC) at 45 DAP of 22 dry bean genotypes Number of seeds/pod and IOO-seed weight of 22 dry bean genotypes (1985) Page 72 74 75 77 78 79 81 82 84 85 86 88 Table Table Table Table Table Table Table Table Table Table Table 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. Page Leaf elongation rates and stem elongation rates of nine dry bean and three cowpea genotypes (1985) . . . . . . . . . . . . . . 89 Leaf diffusive resistance, leaf transpiration, and leaf temperature of 9 dry bean and three cowpea genotypes at 40 DAP (1985) O O O O O O O O O O O O O O O O O 91 Biological yield, leaf weight ratio, and water-use efficiency (WUE) of nine dry bean and three cowpea genotypes (1985) . . . 92 Pod weight, number of pods/plant, and apparent harvest index (AHI) of 9 dry bean genotypes (1985) . . . . . . . . . . . . . . 94 Pod yield differential, arithmetic mean, geometric mean, and drought susceptibility index (S) of 9 dry bean genotypes grown under well-watered and drought conditions (1985) O O O O O O O I O O O O O O O O O O O 95 Stem and leaf weights of 9 dry bean and three cowpea genotypes (1985). . . . . . . . 96 Biological yield and water-use efficiency (WUE) of 9 dry bean and one cowpea genotypes (1986) . . . . . . . . . . . . . . 98 Biological yield differential, arithmetic mean, geometric mean, and drought susceptibility index (S) of 9 dry bean and one cowpea genotypes (1986). . . . . . . . . 99 Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 9 dry bean and one cowpea genotypes at 36 DAP (1986). O O O O O O O O O O O O O O O O 101 Leaf diffusive resistance, leaf transpiration, and leaf temperature of 9 dry bean and one cowpea genotypes at 47 DAP (1986) O O O O O O O O O O O O O O O O 102 Leaf area expansion of 9 dry bean and one cowpea genotypes (1986). . . . . . . . . 103 viii Page Table 36. Mean squares (ms) from analysis of variance for moisture tension of soil under monoculture and intercropped Malawian landrace dry beans (1984) . . . . 128 Table 37. Patterns of response by the landrace genotypes to drought in 1984 . . . . . . . 129 ix Figure Figure Figure Figure 1. 2. LIST OF FIGURES Meteorological characteristics Lansing, Michigan, 1984. . . . Meteorological characteristics Lansing, Michigan, 1985. . . . Meteorological characteristics Montcalm, Michigan, 1985 . . . Moisture tension of soil under monoculture and intercropped Malawian landrace dry beans (1984) . . . . . . . . . . . . at East at East at Page 52 54 55 57 INTRODUCTION Malawi comprises a narrow stretch of land west and south of Lake Malawi. It spans from about 7 to 18 degrees latitude south of the Equator and 33 to 36 degrees east of Greenwich. The country is characterized by extraordinary ecological and cultural diversity. Its area ranges from hot, humid lowlands in the Shire valley at below 100 m to cool, rugged peaks at over 2700 m above sea level. Among legumes in Malawi, dry beans (Phaseolus vulgaris L.) rank second only to peanuts (Arachia hypogaea L.) as evidenced by grain sales at ADMARC, an agricultural marketing firm. However, peanuts are a cash crop whereas beans are generally regarded as a food crop. Therefore, much larger quantities of beans, than can be documented, are produced. Beans originated in the Andean region (Peru and Bolivia) in South America and in Mesa-America (Mexico) (Kaplan, 1965). In general, the former center of origin comprises mostly the large-seeded types (mostly Types I and IV) and the latter center of origin comprises mostly the small-seeded types (mostly Types II and 111) although medium- to large-seeded pintos, bayos, and canarios are also found. Malawians generally prefer the large-seeded kidney bean types. Apparently in the 16th century the Spaniards and the Portuguese took beans, most probably from the South American (Andean) center of origin, to Europe and then to 2 Africa (Purseglove, 1968). Beans probably reached Malawi from the eastern coast of Africa (Mozambique) about'300 years ago though the influence of traders (Mughogho et al., 1972). Eastern Africa is now generally considered a center of cultivation. Probably as a consequence of Malawi’s ecological and cultural diversity, these original bean introductions have evolved with gross morphological variation (Adams, 1982) and are grown as such in mixtures by Malawian farmers. They fit Harlan’s (1975) description of ’landraces’ and will be referred to by this name. These landraces manifest a rich diversity in growth habits, seed sizes, colors, and shapes, among many other intrinsic and extrinsic characteristics. Archaic as this system of production of bean mixtures may seem, it confers important advantages under the tropical subsistence cropping systems. The different components of a single mixture may germinate at different rates thereby extending the seedling establishment period. Pure seed types would cause serious repercussions in the event of drought occurring during germination. The same differential in germination may stagger the maturation times. It is very important for a subsistence farmer, with labor constraints, to harvest the entire field by hand and at the right time for each component line. Bean mixtures may have other advantages such as in disease and insect pest resistance. Also, since these landraces have basically evolved over many generations, the environment has played a major role in 3 their evolutionary process. Two environmental variables exercise major control over crop species evolution; namely, moisture and temperature. Usually moisture varies more than temperature both within and among cropping seasons. And so moisture is widely regarded as exerting the largest influence on evolution, and also the maintenance of the landraces. One is led to speculate that the diverse bean mixtures contain some drought tolerant components which provide a buffering capacity during drought years to maintain the yield stability that is so much treasured by subsistence farmers. The objectives of this study were fourfold: a. To show that intercropping bean and maize (Zea mays L.) aggravates the drought beans encounter in the field in such cropping systems (and probably, to some extent, overcome by bean mixtures); b. To show variability in Malawian bean and other genotypes in response to the drought so caused (in a. above) or caused artificially under controlled conditions; c. To show variability in water-use efficiency (WUE) of different dry bean genotypes as a strategy used by some bean plants when grown under limited moisture conditions; and d. To show variability in simple characteristics that have been associated with drought resistance elsewhere. These characteristics are essential when 4 selecting for drought tolerant progeny among vast segregating populations in a breeding program where modern instruments are cumbersome, prohibitive in terms of cost, and time-consuming. LITERATURE REVIEW Definition of Drought 'Drought' is probably one of the most confusing, nebulous, and imprecise terms in agronomic plant science today. The confusion is portrayed through the perception of it by the different schools of thought. The Webster’s English Dictionary describes the term as derived from the old english term ’drugoth,’ which meant ’prolonged dry weather.’ Meteorologists often define drought as a ’period of lack of precipitation,’ usually with precipitation as a sole cause of drought. But lack of current precipitation is only one of the factors affecting the supply of moisture to crops. Initial amount of moisture contained in the soil at planting time is another source of moisture that needs to be incorporated in the definition. However, such a definition would still fall short in relation to crop productivity. Crop scientists generally regard a drought as a ’lack of both soil moisture and precipitation from planting up until physiological maturity of crops.’ Lack of both factors outside the crop season can not, therefore, qualify as a ’drought.’ It is still confusing, however, because during the same crop season a crop of one species in one field may experience drought whereas another crop of a different species in an adjacent field may not experience any drought. And drought is also 6 caused by a high demand for moisture by the atmosphere just as much as the shortage of moisture on the supply side (initial soil moisture and precipitation). As a result, crop physiologists define drought as ’the period of moisture deficit that adversely affects a specific crop’s growth and development or any of its related physiological processes.’ Menckel (1964) has defined drought resistant plants as ’those plants which, in the process of ontogenesis, are able to adapt to the effect of drought and which can normally grow, develop and reproduce under drought conditions because of a number of properties acquired in the process of evolution, under the influence of environmental conditions and natural selection.’ Suffice it to say that this same process can be directed carefully through crossing and. artificial selection in plant breeding procedures. Under such circumstances the span of time for obtaining drought resistant plants should be much abridged in comparison to evolution by natural selection. Measuring Plant Water Status Stomatal conductance Plants have to exist in an environment of continual dilemma of absorbing carbon dioxide for photosynthesis with consequential dehydration of tissue due to exposure to dry atmosphere. They have overcome this dilemma by evolving leaves with an epidermis composed of a relatively 7 impermeable cuticle and turgor-operated valves called ’stomata.’ The epidermis, therefore, not only reduces rates of carbon dioxide and water vapor exchange but it also provides a means of controlling the balance between water loss and dry matter production. Therefore, measuring the resistance to carbon dioxide and water vapor transfer between the atmosphere and the internal tissues of the leaf imposed by the stomata (stomatal resistance) is important in many studies of biomass production. If the limitation (resistance) offered by stomata is being compared with the flux of carbon dioxide or moisture, or even being correlated with some biological or environmental variable such as leaf water status, it is more meaningful and less prone to misinterpretation to express it as a conductance (reciprocal of resistance) rather than as a resistance (Burrows and Milthorpe, 1976). Fluxes are proportional to conductances but inversely proportional to resistances. But most instruments are calibrated against physical resistances. Therefore, measurements are made as resistances and then conductances are calculated. Stomatal conductance can be obtained by measuring the rate of gaseous loss by porometry. The mass-flow porometer measures the rate at which air is forced through leaves under pressure (Hsiao and Fischer, 1975). The diffusion porometer measures the rate at which water vapor diffuses out of leaves. Mass-flow porometers are simple, cheap, and usually do not involve electronic circuitry. They, however, 8 have the following disadvantages: 1. They are best used for comparative rather than absolute measurement because of errors; ii. Their use is restricted mainly to leaves with stomata on both surfaces; iii. Leakage of air from the apparatus can cause errors. Diffusion porometry, on the other hand, is based on measurement of the rate of water vapor loss from a leaf or portion of a leaf enclosed in a porometer chamber (Kanemasu, 1975). The rate of loss is determined from the rate of increase in humidity. Water loss is from both the stomata and the cuticle. It is generally understood that most of the loss occurs from the stomata, but the cuticular component becomes increasingly important as stomata close. Water Potential By definition, water potential (U) is the potential energy per unit mass of water with reference to pure water at zero potential. It is empirically calculated as: U = RT/Vw ln aw ......(l) where R = the gas constant, T = absolute temperature (K), Vw = partial molal volume of water, and aw = activity of water. As a result, water in most biological systems has less potential energy than pure water hence water potential is given in negative values. There are two main ways of measuring water potential, namely; i. The pressure chamber 9 technique (Scholander et al., 1965) and ii. Thermocouple hygrometry. The) pressure» chamber is simple, cheap, rugged, and ideally' suited for field studies (Ritchie’ and Hinckley, 1975). A leaf cut from the plant is placed in the chamber with the cut—end projecting through the hole in a rubber gromet assembly. The pressure applied to the leaf to return the water interface to where it was before detachment is equal and opposite to the tension in the xylem of the intact plant. The osmotic potential of xylem sap’ is usually extremely low (about 0.2 kPa) and as a result the hydrostatic pressure in the xylem is approximately equal to the water potential. The pressure chamber technique is now the most widely used method for characterizing plant water status. Plant Water Content Cell elongation is very sensitive to cell turgor. As a result plant water status strongly influences plant growth and dry matter production through its effect on leaf and root expansion and on photosynthesis. The amount of water in plants could be expressed in different ways (Slavik, 1974): 10 Water content = [fresh wt- dry wt/dry wtlx 100% ...(2) [fresh weight-dry weightlx100X Relative water content = ...(3) [turgid weight-dry weight) Water saturation deficit = [100-Relative water content]..(4) Turgid weight is obtained by floating leaves or leaf discs on water at the light-compensation point until constant weight is obtained. Determining water content requires relatively simple and cheap equipment and many samples can be taken as replicates or across a large number of treatments. However, the technique for determining relative water content requires considerable skill and precision to obtain accurate results. Leaf Water Retention Capacity This is a fairly new technique. Leaves are removed from the plants and weighed as soon as possible. They are then hung on a wire in a room kept at a constant room temperature for about 48h. After that they are weighed and oven-dried at 800 for 48h. The amount of water retained during air drying is obtained by subtraction and expressed on total water content as per cent. This method was found to show cultivar differences in wheat (Triticum aestivum L.); it was thought better than other methods such as ll stomatal conductance and leaf temperature (Clarke et al., 1982). It seems to be more consistent than other methods, and cultivar differences in the capacity to retain water have been indicated in beans (Tesha, 1984). Drought Resistance Mechanisms Plant mechanisms involved in adaptation to drought have recently been classified by three systems, namely; a. Drought escape, drought avoidance, and drought tolerance (Turner, 1979; Levitt, 1982); b. Phenological, morphological, and physiological mechanisms (Turner and Begg, 1981; Turner, 1982) and c. Adaptations 1. leading to the acquisition of the maximum amount of water; ii. to ensure the conservation and efficient use of water; and iii. protecting cells and tissues from damage and death during severe desiccation (Fitter and Hay, 1981). It should be emphasized, however, that although the systems of classification are different, they all aim at describing the same plant response to both increasing soil and plant moisture deficits. I will review the mechanisms enabling plants to adapt under drought conditions based on the second system of classification. 12 Phenological Mechanisms Phenology refers to the timing of major developmental events such as germination, flowering, senescence, and so forth. Phenological changes in timing of crop developmental changes that the plants go through from germination up to and including maturity play a role in plants subjected to drought. Two aspects of phenological mechanisms are; a. changes in phenological development and, b. developmental plasticity (Turner, 1982). A plant growing on an’increasing soil moisture deficit, without the probability of relief of drought stress, may use the strategy of going into the reproductive stage sooner and maturing earlier (Hall and Grantz, 1981). And actually, this particular strategy seems to be of major importance in some mediterranean zones characterized by terminal drought. Breeding for earliness under such conditions has been suggested (Turner, 1982; Quizenberry, 1982). The problem is that although such genotypes yield better under terminal drought they generally yield poorly when grown in zones with unpredictable droughts. This is because their capacity is limited by a shorter duration of crop growth whether adequate soil moisture is available or not. But effects of water deficits on phenological development are usually small. Mild droughts seem to hasten flowering and maturity in wheat (Turner, 1966) but not in sunflower (Heliamthus annuus L.) (Takami et al., 1981). It would seem that other variables of the environment such as high temperature during severe 13 droughts and photoperiods have a much more marked influence on phenological development (Turner, 1982). Muhammad (1983), working with 16 cultivars of beans, five day temperatures, five night temperatures, and a 12h daylength, found that the time to flowering decreased as the temperature was raised and vice versa. Morphological Mechanisms Water is absorbed from the soil by the roots. Increasing soil water deficits near the soil surface seem to induce compensatory extension of the roots to deeper, unexploited soil layers. The deep rooting systems characterize plants that are drought avoiders because they are able to exploit soil moisture in the lower soil profiles. This is more so in regions dependent on reserves of soil moisture where incomplete extraction of stored water may pose a limitation to high productivity (Jordan and Miller, 1980). Malawi lies in the arid region and is exposed to mainly unpredictable droughts. Usually the crops depend on current rainfall. Under such conditions, developing a deep rooting system may not necessarily be a good enough strategy since the crop virtually depends on current rainfall. However, some areas in Malawi experience a bimodal type of rainfall pattern. In such areas, e.g. Mulanje and Thyolo, beans are planted twice in a year. Usually, the first crop is intercropped with maize (H. Mloza-Banda, personal communication) and faces unpredictable l4 droughts. The second and primary crop is grown during the second rains in monocultures facing terminal droughts and depends largely on stored moisture during seed filling. Deep rooting systems would be beneficial under these conditions of terminal drought. However, under conditions of abundant soil moisture supply throughout the root zone, extensive root branching, as opposed to deep rooting systems, and eventual size of the root system assume utmost importance. Passioura (1972, 1976, 1982) has extensively studied the role of roots in the uptake of water and how water is translocated in the vascular systems of crop plants, especially wheat. He has worked in a mediterranean environment of Australia, where the wheat is largely grown on moisture reserves and experiences terminal drought. From his work two observations have been: a, the ability of a crop to produce a large rooting density at depth if deep subsoil water exists; and b, a longitudinal resistance to flow which may influence the rate at which water collected in the subsoil is transported to the shoots. He concluded that depending on the environment, it may be worthwhile to decrease the resistance, if there is evidence that roots are leaving available water in the soil at maturity. On the other hand, it may be worthwhile to increase the resistance if there is evidence that roots are leaving an inadequate supply of water in the soil at flowering. Modifying resistance to water flow is 15 accomplished by breeding for narrow or broad cross-sectional area of xylem vessels. Physiological Mechanisms Leaf expansion Leaf expansion is probably the first indicator of the plants’ response to drought stress (J. T. Ritchie, personal communication) and as a result irrigation managers have used it to decide when to irrigate their crops. This is because cell enlargement, which is responsible for growth apart from cell division, is the most sensitive process to droughts (Slatyer, 1967; Hsiao, 1973). In maize, Boyer (1970a) showed a rapid decline in leaf enlargement at leaf water potentials below -0.2 MPa, ceasing altogether at —0.7 to - 0.9 MPa. Photosynthesis begins to be affected at leaf water potentials below -0.8 MPa, i.e. only after cessation of leaf expansion. Cell enlargement occurs only when turgor pressure is positive, which can be as high as 0.6 to 0.8 MPa in sunflower (Boyer, 1970a). Usually, some species of crops will reduce their osmotic potentials to increase turgor pressure so that cell enlargement and leaf expansion can continue. This is called ’osmotic adjustment’ or 'osmoregulation’ and has been elaborated elsewhere in this review of literature. The relationship, however, between cell enlargement and turgor pressure has been described by 16 the following equation (Greacen and Oh, 1972; Green et al., 1971). G=B(P-Plin) sooosososs(5) where G is the growth rate, E is a coefficient for the gross extensibility of the cell walls, P is the pressure potential, and Pmin is the minimum pressure below which growth will not occur. Gross extensibility and threshold turgor are not constants and can change under water deficit so as to permit resumption of growth at reduced turgor. This adaptation to stress may occur as a result of an increase in extensibility and/or a decrease in the threshold turgor. However, if under severe water stress turgor pressure fell to zero, then no amount of adjustment of these two parameters would permit a resumption of growth. In situations where turgor pressure is approaching zero, the plant can only maintain growth through osmotic adjustment. Photosynthesis Stomata act as regulators for both carbon dioxide exchange and water loss in plants. Water deficits sufficient to close stomata should also depress photosynthesis through increased resistance to carbon dioxide uptake. And so the conventional explanation for reduced photosynthesis under drought conditions has been stomatal closure (Boyer, 1970a). The change in net 17 photosynthesis with leaf water potential follows that of stomatal conductance (Boyer, 1970b). And these changes are reflected in rates of transpiration measured concurrently. Troughton (1969) concluded that stomatal closure was the primary event in depression of photosynthesis under drought conditions. Using cotton (Gossypium hirsutum L.), he showed that leaf conductance decreased at relative water content of 80%, whereas mesophyll conductance was only reduced at 75%. However, as early as 1967, Nir and Poljakoff—Mayber reported that isolated chloroplasts from desiccated swiss chard (Beta vulgaris L.) showed less Hill activity and cyclic photophosphorylation. Photosynthesis of water stressed (below -1.1 to -1.2 MPa) sunflower leaves failed to respond to elevated carbon dioxide under high (or low) photosynthetically active radiation (PAR). But at low PAR where photochemical activity limits photosynthesis, a decline from -0.4 to -1.8 MPa resulted in a 70% drop in photosynthesis (Boyer, 1971b). The activity of isolated chloroplasts from stressed leaves shows that electron transport and photophosphorylation drop rapidly at leaf water potentials of less than -1.0 to -1.2 MPa, and it has been suggested that these activities limit photosynthesis below -1.1 MPa (Boyer, 1971b; Keck and Boyer, 1974). Since photosynthesis begins to decline at about -0.8 MPa, the contention is that stomatal resistance prevails as the primary factor at mild stresses, -0.8 to -1.1 MPa, but that chloroplast activities predominate at more severe stresses. 18 Recent evidence suggests that the reduction in photosynthesis is not as much as function of stomatal closure and/or light reactions as it is of adverse effects on the carbon fixation capacity. Johnson et al. (1985) determined the stomatal limitation to photosynthesis under drought in wheat. Their analysis of photosynthesis versus internal carbon dioxide concentration response curves indicated that, although stomata closed in response to stress, the stomatal limitation was less than 40% of the total limitation to photosynthesis. They suggested that internal leaf capacity for photosynthesis (’dark’ reactions) was largely responsible for the limitation to photosynthesis and that stomatal control may be in part mediated by the internal capacity of leaves to photosynthesize. Nitrogen Fixation Spent (1976) has performed experiments in this area and has critically reviewed the subject. The consensus of her review is that depending on the stage when it occurs, drought may depress nodulation or nitrogen fixation by the existing nodules. Finn and Brun (1980) have plotted specific nodule activity (SNA) as a function of nodule water potential, and for comparison the stomatal resistance of soybean (Glycine max (L.) Merr.) plants pretreated for 20h with polyethylene glycol (PEG)> As nodule water potential decreased from about -.5 to -1.5 MPa stomatal resistance increased from 8 to 38 sec/cm whereas SNA decreased from 36 19 to 15 umoles acetylene/g nodule fresh weight/h. A question arises as to whether nodule function is diminished because photosynthesis (source of ferredoxin, ATP, and NADPH for reduction of dinitrogen) is depressed or whether there is a direct effect of drought on the nitrogen fixation process. Using subterranean clover (Trifolium subterraneum L.) Dejong and Phillips (1982) found that acetylene reduction (AR) was depressed earlier and to a greater extent than carbon exchange rate (CER) when drought was imposed. Over three dry-down cycles CER was decreased less than 50% whereas there was a 90% inhibition of AR. Using soybean, Pankhurst and Sprent (1975b) determined that increasing the oxygen partial pressure reversed the water-stress induced depression in AR activity (by 75% at atmospheric levels of carbon dioxide) in moderately stressed (-0.75 MPa) nodules, Moreover, AR of nodule slices, nodule breis, and bacteroids from moderately stressed nodules was similar to turgid nodules. These results suggest that drought induces a barrier to oxygen diffusion into nodules, thereby reducing their activity by limiting energy production. With a scanning electron microscope Pankhurst and Sprent (1975a) further showed that soybean nodules have lenticels and that these collapse under drought. Oxygen limitation, therefore, seems to be the direct cause of reduced nitrogen fixation in moderately stressed nodules. With severe droughts other factors associated with tissue and enzyme alteration come into play. Recovery from such 20 stress involves regrowth of existing nodules in those species with a nodule meristem (most grain legumes), whereas species with non-meristematic nodules (spherical nodules) recover by growth of new nodules. Assimilate Remobilization In dry beans, much like other crop plants, reproductive development depends on vegetative organs as sources of carbohydrates, nitrogen, and other materials. During early stages of reproductive growth, allocation of newly assimilated carbon from the leaves is shifted from vegetative tissues to the developing reproductive sinks. As the reproductive demands become greater, and/or aggravated by severe soil moisture deficits, concurrent assimilation of carbohydrates is not sufficient and they must be remobilized from pools of nonstructural compounds accumulated in vegetative tissues during earlier growth stages. Adams et al. (1978) have indicated undetectable to large levels of starch in roots and stems of 23 dry bean cultivars. One may question the role of such large amounts of nonstructural carbohydrates in non-economic vegetative tissues. The case in point is that of attainment of high starch levels in some genotypes and not others. As it turns out, the root and stem may act as temporary storage organs to partially correct for an apparent imbalance. Sebasigari (1981) observed this concept among grain legumes grown at East Lansing, Michigan. Loss of reserve assimilates was 21 well correlated with grain growth. This concept closely associates with that of selecting for higher harvest indices especially under drought conditions (Sinclair et al., 1984). In late August of 1965 a killing frost occurred in Guelph, Canada. However, grain yield of maize doubled by early October (Daynard et al., 1969). This suggests that assimilates manufactured before the frost and stored in the maize stalks may have contributed to grain yield. Maize stalks accumulate carbohydrates, primarily as sucrose, glucose, and fructose (Loomis, 1935). In this crop, maximum amounts of soluble solids usually have accumulated 3-4 weeks after anthesis (Campbell and Hume, 1970) and rapid declines in stalk soluble solids often occur during grain filling (Daynard et al., 1969). Hume and Campbell (1972) found that most of the soluble solids that accumulated and disappeared were in internodes below the ear. When pollination and grain development were prevented, total soluble solids in stalks increased until the end of the growing season, indicating that the decline in stalk soluble solids during grain filling was caused primarily by translocation of metabolites from the stalk to the grain. Using wheat, during the droughts of 1975 and 1976 in England, Gallagher et al. (1976) assessed the importance of photosynthesis after anthesis and the translocation of materials assimilated before anthesis in supporting grain growth, using a simplistic method. They expressed the 22 increase in grain weight from anthesis to final harvest (de) by the following equation: dwg=th-dws ossssssss(6) where th is the increase in total crop weight, and dWs is the increase in weight of plant parts other than the grain between anthesis and harvest. The largest portion of dWs consists of the stem and for brevity this term is referred to as stem weight. In a determinate plant dWs/de is the fraction of final grain weight apparently derived from translocation of materials assimilated before anthesis. In 1975 and 1976, when production of dry matter after anthesis was smaller in relation to requirements for wheat grain growth, the contribution of translocated materials to grain yield was much greater. These measurements showed that pre- enthesis assimilate contribution to grain yield in wheat was 35% in non-stressed plants and 56% in stressed plants. Thorne (1966) indicated that in cereals the pre- anthesis assimilate contribution is small, being no more than 20% under non-stressed conditions. The 35% contribution indicated by Gallagher et al., (1976) probably was an overestimation. Using in situ labeling, with radioactive carbon, of the whole crop canopy at frequent intervals before and after anthesis, Bidinger at al. (1977) reported lower contributions, viz: 13% and 27% for wheat plants under non-stressed and drought-stressed conditions, 23 respectively. They also reported 12% and 17% assimilate contribution for barley (Hordeum vulgare L.) plants under similar respective conditions. McCaig and Clarke (1982) have shown that carbohydrates accumulate in wheat stems to a maximum (25% to 48% of total biomass) about the time of anthesis and then decline as maturity is approached. They reported significant differences in stem carbohydrate levels between wheat and oats (Avena sativa L.) and between cultivars in either species. A controlled method is required to reveal genotypes that can sustain translocation-based kernel growth in the absence of photosynthesis. Leaf removal after anthesis was found to affect kernel growth differentially among various genetic materials (Brenner, 1972). Blum et al. (1983) destroyed all the spring wheat’s important photosynthetic tissues, i.e. awns, glumes, leaf laminae and parts of the spike-peduncle, and leaf sheaths with a chemical desiccant. They grew four spring wheat cultivars (Lakhish, Cajeme, V591-51, and V582-58) under non-stress conditions in a mediterranean environment in Israel. They sprayed magnesium chlorate, a chemical desiccant, 14 days after anthesis. This induced earlier stem dry matter loss in two cultivars (Lakhish and V591-51, group I), and not in the other two (Cajeme and V582058, group II). The pattern of stem dry matter loss in the group II cultivars did not differ between sprayed and unsprayed treatments. Kernel growth rate was reduced by the treatment, to a relatively 24 lesser extent in group I than in group 11. They concluded, therefore, that cultivars of group I were better than cultivars of group II in sustaining kernel growth in the absence of photosynthesis and this difference was associated with a respective difference in stem reserves remobilization. From the breeders’ point of view, however, it should be emphasized that a highly productive cultivar should be capable of remobilizing reserve assimilates under both stress and non-stress conditions. A high yielding cultivar that does not remobilize reserves under non- stressed seasons may not be the best yielder under stressed conditions. Maintenance of Turgor As water deficits develop, water is removed from the cell and as a result, the solutes inside the plasmalemma get concentrated leading to reduction in water potential. Water moves from high to low water potential. Therefore, this acts as a driving force to replenish the cellular water content. In some species water stress not only concentrates solutes but induces solute accumulation, a process alluded to earlier called ’osmotic adjustment' or ’osmoregulation.’ This can be explained if solute accumulation does not follow the Boyle-van’t Hoff relationship which is expected if concentration of solute occurs with only loss of water. Using six potato (Solanum tuberosum L.) cultivars, Levy (1983a) reported genotypic variability in osmoregulation 25 capacity. Cultivar ’Alpha’ had the highest and ’Troubadour’ had the lowest osmotic adjustment. Concomitantly, Alpha had maintained the highest turgor potential under drought whereas Trobadour had the lowest values indicating a relationship between the degree of osmotic adjustment and maintenance of turgor. Alpha produced the highest tuber dry matter yield under drought whereas Troubadour produced the lowest yield under similar conditions (Levy, 1983b). Osmotic adjustment in a range of crop species has been observed in response to diurnal and seasonal changes in water status (Hsiao et al., 1976) leading to full or partial turgor maintenance. Turner and Jones (1980) reported that full or partial turgor maintenance during a drying cycle has been shown in the leaves of 14 species, in the root and hypocotyl of two more species, and in both the leaves and floral parts of wheat. Osmotic adjustment has also been shown in rice (Oryza sativa L.) (Cutler et al., 1980), sorghum (Sorghum vulgare L.) (Hsiao et al., 1976), sunflower (Takami et al., 1981) and maize (Turner, 1975). However, it has been shown not to occur in soybeans (Turner et al., 1978) and sugarcane (Saccharum Officinarum L.) (Ingram and Moore, 1985). Diurnal changes in osmotic adjustment help maintain physiological processes such as leaf expansion and photosynthesis that otherwise would be impaired by low turgor (Hsiao et al., 1976). They also hasten the unrolling of leaves wilted by low turgor (O’Toole and Moya, 1978). 26 Osmotic adjustment that occurs during a drying cycle, while rarely fully maintaining physiological functions, slow the rate of loss of physiological activity (Turner and Jones, 1980). Morgan (1980) has also reported genotypic variability in osmotic regulation among wheat genotypes. He crossed cultivars that showed osmotic adjustment with those that did not and selfed the progeny to the F4 generation. He showed that this character is heritable and can be selected for. After selfing, he came up with two populations that differed significantly in osmotic adjustment. These two populations were them grown under both drought and non-stress conditions. The group that was able to osmoregulate produced significantly higher grain yields than the other, possibly due to higher photosynthetic rates, lower respiration rates, and higher partitioning coefficients. However, negative associations between osmotic adjustment and productivity are expected because there is a biological cost in solute accumulation (Jefferies, 1981). Changes in Tolerance to Dehydration The ability of tissues to withstand dehydration becomes very important under conditions of severe and prolonged drought when other mechanisms have either failed or been exhausted. The cell membranes constitute a central site of activity for cellular metabolic processes. Their tolerance for dehydration stress is important (Bewley, 1979). It is 27 believed that dehydration tolerance may involve a complex of mechanisms at the enzyme and membrane levels that reduce protein denaturation and membrane degradation. Sullivan (1983) developed a fairly rapid method of assessing the rate of injury to cell membranes of sorghum by dehydration through the measurement in vitro of electrolyte leakage from water stressed leaf discs. This method seems applicable to other crop species (Blum and Eberccn, 1981) including bean plants (Tesha, 1984). Changes in dehydration tolerance occur with different rates of drying. Slow rates of drying enhance dehydration tolerance (Gaff, 1980), possibly due to the accumulation of solutes such as sugars and proline that protect proteins and aid in recovery. Free proline amino acid accumulation occurs in leaves of many plant species subjected to environmental stresses, including drought (Stewart and Hanson, 1980). However, that proline accumulation constitutes a drought-adaptive trait is rather controversial. Working with barley, Hanson (1980) concluded that proline accumulation actually is an indicator of drought susceptibility rather than drought resistance. He used two barley cultivars that differed in dehydration avoidance. The nondehydration-avoiding genotypes accumulated more proline when grown under drought in the field. Similar results were reported in cotton (Ferreira et al., 1979). However, when proline accumulation is assayed at a given water potential in all cultivars this characteristic seems 28 to be related to leaf survival rate during stress and recovery of growth upon rehydration (Singh et al., 1973; Blum and Eberccn, 1976). The possible role of proline may be to serve as an energy pool to be used upon recovery (Blum and Eberccn, 1976; Bogges et al., 1978) or as an osmotic agent (Gould and Measures, 1977). Proline is probably an effective cryoprotectant against freeze-induced dehydration of cells, possibly through its effect on membrane stabilization. Mycorrhizae ’Mycorrhiza’ is a term that refers to a mutualistic, symbiotic relationship formed between fungi (Greek mukes) and living roots (Greek rhiza) of higher plants (Miller et al., 1986). This type of association has been observed in most plant species except the Cruciferae and Chenopodiaceae. Mycorrhizae have been classified (Harley and Smith, 1983) through a description of 7 types, namely vesicular- arbuscular, ecto-, ectendo-, arbutoid, ericoid, monotropoid, and orchid. The vesicular-arbuscular mycorrhizae (VAM) comprise the major type observed in crop plants. They produce structures known as vesicles and arbuscules, as well as hyphae and spores. Arbuscules are intracellular, haustoria-like structures that develop by a repeated, dichotomous branching of hyphae. Vesicles are sac-like, usually terminal swellings at the tip of hyphae. The hyphae can be formed both within the root and outside it. The 29 transfer of mineral nutrients and other materials from the soil to the host plant is mediately by these hyphae. Safir et al. (1971, 1972) was the first group to report changes in the water relations of a plant when infected by a VAM fungus (Endogone mosseae L.). Soybean plants have been observed to suffer from drought stress in central Illinois under adequate soil moisture levels (Boyer et al., 1980). They contributed this to lack of enough root density to absorb enough water to meet atmospheric demand for moisture. However, Safir et al. (1971, 1972) showed that under well- watered conditions and low soil nutrition, mycorrhizal soybean plants had hydraulic conductivities 40% higher than nonmycorrhizal controls and showed no symptoms of drought stress. Addition of nutrients nullified these differences. It was suggested (Safir et al., 1972) that roots were the sites of the increased hydraulic conductivities of mycorrhizal plants based on comparisons of recovery rates from moderate water stress of whole plants with severed roots. Two possible explanations for increased conductivity exist. One is that it is possible that the mycorrhizal hyphae in the soil enable mycorrhizal plants to take up water. The other, which was discounted by Safir et al. (1972), is that mycorrhizal hyphae within the root could provide a low resistance pathway for water movement. Under limited water conditions in the short term, Safir et al. (1971, 1972) demonstrated that mycorrhizal soybean leaves recovered from moderate water stress faster than 30 nonmycorrhizal controls after soil rehydration. These differences were eliminated under high nutrient levels. Nelsen and Safir (1982) conducted a long term study by exposing onion (Allium cepa L.) plants (mycorrhizal and nonmycorrhizal) to 7 cycles of drought. The regime started 4 weeks after seeding and lasted 8 weeks. Mycorrhizal plants at 8 and 12 weeks, which were given low P levels, were more drought tolerant than nonmycorrhizal onions given high P levels. Drought tolerance of mycorrhizal plants was shown by greater fresh and dry weights and by higher tissue P concentrations. Nutrient analysis showed that the P concentrations of the nonmycorrhizal plants were at levels (0.1 to 0.3% dry weight) known to retard the growth of plants (Epstein, 1972). But P was not limiting in the mycorrhizal plants. It is not surprising that mycorrhizal plants are able to absorb P during drought considering their extensive fungal network in the soil. Water-use Efficiency Wittwer (1975) identified water as the second-most important factor, behind land area, to increasing food production. He argued that a high research priority should be given to improvement in the efficiency of water use by crops. And knowledge of the factors influencing water-use efficiency has continued to be an objective in many studies. As opposed to the sense of the word ’efficiency’ in the physical sciences, water-use efficiency (WUE) does not tend 31 toward unity if it is increased and can never reach it in a crop (J. L. Monteith, personal communication). However, the term is widely applied in crop studies where its place has been established. Sinclair et al. (1984) defined WUE as a ratio of biomass accumulation, expressed as carbon dioxide assimilation (A), total crop biomass (B), or crop grain yield (G), to water consumed, expressed as transpiration (T), evapotranspiration (ET), or total water input to the system (I). The time-scale for defining WUE can be instantaneous (i), daily (d), or seasonal (s). Water-use efficiency is written symbolically as a function of these three variables. For example, WUE (B, I, 8) refers to water-use efficiency expressed as total crop biomass (B) to total water input into the system (I) for a seasonal observation. It is calculated using the following formula: Total biomass production (3) WUE (B, I, 8) = ----------------------------- 000(7) Total water applied (kg) One approach to increasing the use of the total available water is the development of crops with better rooting systems. Greater rooting density and deeper rooting depths increase the surface area exposed to soil moisture leading to an increase in the amount of soil moisture absorbed by the plant (Taylor, 1983). Water reserves previously 32 unavailable to the transpiring plant could be made available with an improved root system. However, a deeper and more extensive rooting system may have disadvantages. A greater root biomass would result in lowered harvest index. Carbohydrates would be diverted to construction and maintenance of a larger root mass. In a prolonged drought, rapid and more efficient use of stored soil water may be deleterious. A less effective, slow growing root system may leave some water reserves that would be partially available for surviving a drought. Improvements in harvest index result directly in increased water-use efficiency based on seed yield. Substantial improvement in harvest index seems to be a desirable selection objective under conditions of limited moisture where drought stress, especially during reproductive growth, can lead to greatly reduced harvest indices. In Australia, Passioura (1972) describes manipulation of the water conducting system in plants for stress and nonstress environments. Conservation of soil moisture for later extraction is sought by narrowing the diameter of the vascular vessels by genetic selection. Shorter season cultivars that complete their life cycle and produce a high harvest index before the available water supply is exhausted is another approach. Cowpea (Vigna unguiculata (L.) Walp.) yields have been increased in a drought environment by developing lines with shorter growing seasons (Turk et al., 1980). 33 Drought and Crop Yield Crop growth and development can be divided into phases, namely; vegetative and reproductive growth periods and various stages within either of these periods (Fehr and Caveness, 1977). During the early part of the summer field- grown soybeans and maize are frequently under moisture stress which limits leaf expansion and stem growth during the day and may be sufficiently severe to limit photosynthesis during part of the day. Usually there is sufficient soil moisture to rehydrate the plants at night (Slatyer, 1967). As the season progresses and if there is limited rainfall the roots extract water from the surface and as they grow downward the depth of soil depletion gets deeper. During years when combined fall and spring rainfall does not recharge the moisture reserves droughts during July reduce canopy size. This is a conservative characteristic because if conditions are reversed the amount of leaf area attained will limit assimilate production for seed filling. Severe droughts will cause reduction in light interception and numbers of pod-bearing nodes. The most damaging drought in crops is that which occurs during the reproductive period and causes massive reductions in seed yields. In maize, a determinate species, a severe 4-day drought during some phases of the reproductive period is critical. From silking and a fortnight thereafter seems to be the most sensitive period of crop growth to drought (Claassen and Shaw, 1970). The number of kernels per ear is 34 the most drastically reduced yield component. Drought during this period causes barrenness in maize. In small cereal grains drought during the flower initiation and booting stages can reduce grain yields by about 25% and 65%, respectively (Hanson and Nelsen, 1980). Drought stress at different stages of indeterminate soybean development and its effect on seed yield and yield components was investigated by Shaw and Laing (1966). Indeterminate varieties have the potential to flower over longer periods of time and as a result may not be as sensitive to drought as determinate varieties. Short periods of drought during early flowering of soybean causes little reduction in seed yield; even though water stress caused flower abortion, the plant has time to generate a new flush of flowers after the drought. The number of pods per plant is the yield component most affected. And the growth stages most sensitive to drought are late pod development and mid-seed filling. At late pod development, Shaw and Laing (1966) observed pod abortion, poorer pod development (fewer seeds per pod), and reduced photosynthesis (reduced seed weight). During later stages of seed filling, although there was some effect on pods per plant and seeds per pod, the greatest influence was on seed weight. Determinate soybeans, however, behave similar to other determinate species, e.g. maize. 35 Therefore, drought resistance should ultimately be related to yield as a major economic consideration and as an integrator of the effects of plant drought stress in time and space. Yield under drought stress is a function of the genotype’s yield potential (Mederski and Jeffers, 1973; Fischer and Maurer, 1978). Yield under such conditions may not be a reliable selection criterion because the genetic component of variation relative to the environmental component of variation is usually low (Daday et al., 1973). If drought is uniform over the population and it is shown to be the major environmental factor then yield under drought is an important factor. The problem of dissociating the effect of drought resistance from that of potential yield was resolved by Fischer and Maurer (1978) using an index of susceptibility (S): Yd=Yp(l-SD) 00000000(8) where Yd is stress yield, Yp is potential yield under nonstress conditions, and D is the drought intensity. D is calculated as 1-(Xd/Xp), where Xd and Xp are the mean yields over all cultivars tested under stress and nonstress conditions, respectively, and ranges from 0 to 1. S is the drought susceptibility index. A cultivar with lower S than another has higher drought resistance. Although S should be independent of drought intensity, its exact value will depend on cultivars included in the calculation of D. Using 36 S, Samper (1984) was able to show significant differences among dry bean genotypes in drought resistance. Although yield differential has been used to show differences in drought resistance between snap bean cultivars (Bonanno and Mack, 1983) it is a poor method. Stability analysis (Finlay and Wilkinson, 1963; Eberhart and Russell, 1966) is very useful in defining drought resistance in terms of yield, provided that the major component of variation in the environmental index as used in the analysis can be attributed to the water regime (Nor and Cady, 1979; Blum, 1980). Laing and Fischer (1977) used the approach to show genotypic differences in drought resistance of wheat as shown by their yield under different environmental indices. Two cultivars, Mirian and 676, performed better than others at low soil moisture conditions. Their intercept (a) was higher than the other cultivars, Lakhish and Cajeme, which performed poorly. Therefore, differences in drought resistance among cultivars can be obtained by the intercept (a) of the regression line in a stability analysis. Potential yields under nonstress conditions are evaluated through their relative lepes (b). In their case, Laing and Fischer (1977) concluded that Mirian and 676 were both drought resistant but 676 had a higher yield potential than Mirian. MATERIALS AND METHODS Empirical equations are available that can be used to predict net primary productivity (NPP) at a particular location given environmental variables (Lieth, 1976). Each equation requires the use of but a single environmental variable. Such equations are given below: NPP = 30(1’e‘°'°°°“‘"] 0000000000‘9), where W = mean annual precipitation (mm); NNPP = 30/[l+e1-315'°-11’T] .......(10), where T = mean annual temperature (C); NPP = [(105XPET)-100]/100 000.0000‘11)’ where PET = potential annual evapotranspiration (mm). Such equations can be used to determine the most probable limiting environmental factor on crop productivity at any one location. Based upon Malawi Weather Tables (Malawi Government, 1982), three environmental variables were obtained from 19 different locations across that country. These variables were mean annual precipitation (mm), mean annual temperature (C) and PET (mm). PET was calculated as was done by Turc (1955) using the equation below: 37 38 PET = P(u)/[0.9+(P(u)/L)3]0'5 000000000000(12, where P = precipitation (mm), L = 300+25T+0.05T3, and T = mean annual air temperature (C). Calculations were done by the CALC procedure of the MSTAT (version 4) statistical package and NPP values were obtained and on an assumption of a growing season of 90 days and harvest index of 50% these values were adjusted to those in Table 1. This type of analysis helps identify the most limiting environmental variable on crop productivity. From the table, mean air temperature consistently gave the highest amount of yield and therefore it does not seem to be limiting. On the other hand, mean annual precipitation consistently gave the least yields at all locations except at Mimosa, Bvumbwe, Chichiri, Nkhata-Bay, Mzuzu, and Thyolo, where evapotranspiration was probably more limiting than precipitations. As a result, one would conclude that moisture is the most limiting environmental factor in crop productivity and is the factor that plays the greatest role in evolution of crop plants under Malawian conditions. Therefore, the studies conducted here were limited to moisture stress as a factor to which some Malawian bean components are adapted in order to maintain themselves and the mixtures in general. Malawi landrace bean genotypes were grown in three experiments at the Crop Science Field Laboratory in East Lansing, Michigan during 1984, 1985, and 1986. The 39 Table 1. Comparison of environmental parameters by their effect on net primary productivity across 19 locations in Malawi. Location Mean ann. Mean sun. A B C temp. precip. (C) (In) ------- kS/ha -------- Chitipa 21.0 1038.8 2588 1653 2150 Karoga 24.8 1164.8 2827 1788 2414 Mzimba 19.7 864.3 2491 1447 1670 Mzuzu 17.9 1218.3 2339 1842 1474 Nkhata-Bay 23.3 1694.6 2742 2249 1637 Chitedze 20.1 919.1 2522 1514 1661 Lilongwe 19.7 847.5 2491 1426 1775 Nkhota-kota 23.8 1630.8 2767 2202 2223 Salima 24.2 1281.2 2796 1903 2350 Chichiri 19.9 1122.1 2506 1744 1652 Dedza 17.7 905.2 2326 1498 1524 Makoka 20.5 1044.1 2556 1659 1627 Chileka 22.4 857.5 2685 1439 2485 Mangochi 24.3 823.6 2799 1396 2250 Mimosa 21.6 1724.5 2631 2271 1383 Bvumbwe 19.3 1158.9 2455 1782 1262 Thyolo 20.9 1273.1 2578 1896 1411 Ngabu 26.2 811.4 2902 1380 2167 Makhanga 25.5 764.5 2867 1318 2119 A = seed yields calculated based on mean annual temperature; B = seed yields calculated based on mean annual precipitation; C = seed yields calculated based on mean annual evapotranspiration. 4O experiment conducted in 1984 was a field maize and bean intercropping study. A similar experiment was carried out in 1985 at the Montcalm Potato Research Facility about 80 miles north-west of East Lansing. The experiments conducted in 1985 and 1986 at East Lansing were near and under the rain-out shelter, respectively, in a field-simulated crop culture to determine water-use efficiency of the bean genotypes under both well-watered and drought conditions. Intercropping Experiments East Lansing, 1984 By planting maize and beans in one row it was envisaged that the aggressive species, maize, would out-compete the bean for moisture, especially if the maize was sparsely populated to reduce shading and both crops were fertilized adequately. Eighteen selected pure bean lines (arranged as accessions 1 to 18), three of which were Type I’s (accessions 2,3 and 17) and the rest mostly Type IV’s, were grown in single-row plots in a split-plot design. The main plots consisted of two rows of a bean genotype. Sub-plots were cropping systems consisting of a single bean row in monoculture and another single bean row intercropped with maize in the same row. Each main plot was replicated three times. 41 Land Treatment and Crop Culture A site consisting of heavy, medium, and light soils was selected. Three replications were arranged so as to reflect the soil differences, with the first replicate on the light soil, the second replicate on the medium soil, and the third replicate on the heavy soil. The area was first plowed with a disc plow. On 12 June 1984 metolachlor and chloramheu were applied convert to at 2.36 kg/ha a.i. and 1.46 liters/ha, respectively, to control weeds. Adequate fertilizer Compound -10:20:20- was applied at 230 kg/ha on 13 June to remove interspecies competition for the major nutrients. Plating was done on 14 June. A main-plot consisted of two rows 3m long. The length of rows was dictated by availability of seed. To minimize competition through shading of the bean plants by maize, the inter-row spacing was In, conforming to the recommended space between ridges in Malawi. An early Michigan single cross maize hybrid (M422) was used in all intercropping sub-plots to provide competition for soil moisture. The maize was planted at an intra-row spacing of 1m and four seeds to a hole, again to conform to three plants of maize per hill after thinning two weeks after emergence. The beans were planted at an intra-row spacing of 20 cm in the same row with maize, a typical cropping system in Malawi, as one sub— plot, and in monoculture, at the same spacing, in the other row as a second sub-plot. 42 Soil Moisture Tension The availability of moisture was monitored using the ’Quick Draw’ soil moisture probe (Soil Moisture Equipment Corp., Santa Barbara, Ca) by determining moisture tension (kPa). Soil moisture tension was read at 15 cm away from the maize hill, between the hill and a bean plant in intercropping and anywhere inside 0.5 m borders in monocultures. Readings were recorded at 47, 68, and 82 days after planting (DAP). Using this method one can irrigate beans if soil moisture tension approaches or exceeds about 50 kPa; in other words, the sub-plot that exceeds 50 kPa has reached moderate drought stress levels. Porometry and other related characteristics Using a steady state porometer (LiCor Inc., Lincoln, NE) leaf diffusive resistance (s/cm), leaf transpiration [ug Hzo/sz/S), quantum [uE/cm'ls], relative humidity (%) near the leaf, and leaf temperature (C) were obtained from both sub-plots in the entire experiment at 70, 78, and 85 DAP. In this study readings were taken both just before dawn (pre-dawn) and in mid-afternoon. Diffusive resistance was the resistance offered by the leaf to the escape of moisture as opposed to that of carbon dioxide uptake. The two are different only due to differences in molecular weights of the two compounds. Leaf diffusive resistance was the combination of both the mesophyll and the stomatal resistances. 43 Leaf Moisture Retention Capacity (LMRC) At 72 DAP three young, fully expanded leaves were sampled from bean plants in each sub-plot. Their separate fresh weights (FW) were recorded. The leaves were then placed in shaded open air to dry for 24h. After this period the leaves were weighed again and these data are here-in- after called air-dried weight (ADW). At this point the leaves were oven-dried at 700 for 96h. These data are here- in-after called oven-dried weight (ODW). Leaf moisture retention capacity was calculated using the following formula: LMRC = [ADW-ODW] x lOO/[FW-ODW] ........(13) where ADW is air-dried weight (g); ODW is oven—dried weight (g); and FW is fresh weight (g). Specific Leaf Weight (SLW) At 74 DAP, five young, fully expanded leaves were sampled from all sub-plots. The leaves were dissected into leaflets which were then run on a portable area meter (model LI-3000, LiCor, Lincoln, NE) to determine their leaf area (LA). After determining leaf area the leaves were oven- dried at 70 C for 96h to obtain their oven-dried weights (ODW). Specific leaf weight (SLW) is a growth analysis parameter that was calculated as follows: 44 SLw=ODWILA 000000000(14) where ODW is leaf oven-dried weight (mg), and LA is leaf area [cm']. Yield and Yield Components At maturity three plants were sampled at random from each plot. The stem was cut at the base. At the field laboratory each plant was dissected into various plant parts, although much fewer leaves were retained on the stem at this stage. All plant parts were oven-dried at 70 C for 72h and the following data were obtained: stem weight, number of pods, seed number and seed weight. Harvest index was ’apparent’ due to lack of some leaves that had been shed in most bean genotypes but was calculated as follows: AHI = seed weight x 100/total biomass .......(15) Montcalm, 1985 A site consisting of a slight gradient in slope and a sandy soil (Montcalm sandy loam/ loamy sand) was selected in 1985. This was thought to provide rapid percolation of water, especially on the upper side, so as to emulate drought conditions. Dual (2.36 liters/ha) and Eptam (1.46 liters/ha) were applied on 10 June 1985 to control weeds. A mixed fertilizer -19:19:19— was applied at 341 kg/ha. Planting was done on 13 June and 14 June for bean and maize, 45 respectively. The same short season Michigan maize (M422) used in 1984 was planted in 1985. An experimental design similar to that used in 1984 was used in 1985. One difference was the length of rows which were extended to 5m. The following additional cultivars were included: Montcalm (from Michigan State University, U.S.A.), BAT 47, BAT 85, BAT 337, and BAT 798 (from C.I.A.T., Colombia) and Bayo Madero (from I.N.I.A., Mexico). Germination of the untreated materials (all except maize and Montcalm cultivars) was very poor and replanting was done on 30 June, 1985. As a result limited data were obtained from that nursery. Two young, fully expanded leaves were sampled at 53 DAP for water retention capacity data. Fresh weights (FW) were obtained and the leaves were put in a shaded space to air dry for 48h. Air dry weights (ADW) were obtained and the leaves were then oven dried at 70 C for 72h to obtain ODW. The formula (12) was used in calculating leaf water retention capacity. Porometry was done in a similar manner to that followed in 1984. 46 Water-use Efficiency Experiments 1985 WUE Experiment Crop Culture In 1985, five landrace dry bean genotypes (from Bunda College of Agriculture, University of Malawi), Montcalm (from Michigan State University, U.S.A.), BAT 47 and BAT 477 (from C.I.A.T., Colombia), Chiapas 7 (from I.N.I.A., Mexico) and three cowpea cultivars (from I.I.T.A., Nigeria) were grown in 200m diameter and about In high seam-sealed black plastic bags secured in place by wire mesh. Fifty-five kg of field soil was put in each bag. Five grams of a mixed fertilizer -19:19:19- were placed 100m deep. Seeds were pre-germinated for 48h and seedlings that were as uniform as possible were selected for planting. Two treatments were imposed and these were: a. adequate supply of water constituted well-watered plots and b. restricted supply of water constituted the drought plots. Well-watered treatments received 6 liters of water at planting, and additional water being supplied as necessary until a total of 9 liters of water had been applied by the stage of physiological maturity. The drought plots received only 2 liters at planting, and additional water being given fortnightly during the reproductive period until a total of 3 liters had been applied by physiological maturity. Planting was done on 30 June 1985, 24h after the initial and heavy dose of water was applied. Sevin was applied at 1.42 hg/ha on 28 July 1985 to control the mexican bean beetle. 47 Porometry and related characteristics A steady state porometer (LiCor, In., Lincoln, NE) was used at 28 DAP in 1985 to determine leaf diffusive resistance, leaf transpiration, quantum, relative humidity near the leaf, and leaf temperature. Again, diffusive resistance was the resistance imposed by the leaf to moisture escape as opposed to that against carbon dioxide uptake. Stem and leaf elongation rate The length of the stem from the base to the terminal bud was measured at 32 and 56 DAP as H1 and H2 in cm, respectively. The difference between H2 and H1 represented growth in 14 days and the rate was expressed on a daily basis. Leaf lengths were taken from small developing terminal leaflets at 32 DAP as L1. Leaf lengths were measured from the same leaflets at 35 DAP as L2. .The length difference between L2 and L1 in cm in 3 days was expressed as rate on a daily basis. In 1986 the equation cited by Gardner et al. (1985) was used to calculate leaf area from which leaf area expansion was derived. Pod weight and other characteristics Harvesting was done on 10 October 1985. Plants were cut at the base and taken indoors where they were dissected into various plant parts, i.e. stems, leaves, and pods. The number of pods was obtained. Plant parts were then placed 48 in the oven to dry for 72h at 70 C. Total biomass was the sum of all the dried plant parts. Leaf weight ratio (LWR) was calculated using the following formula: LWR = Leaf weight (g)/total biomass (g) ......(16) Water-use efficiency was calculated using formula (7). 1986 WUE Experiment Crop Culture In 1986 a similar experiment to that conducted in 1985 was carried out under the rain-out shelter. This time a different set of genotypes was used as given in Table 2 below. Planting was done on 10 June and all cultural practices were conducted as in the previous experiment. However, the imposition of treatments was slightly different from that in 1985. No large doses of water were applied at planting.) Instead all plots were well—watered up to the beginning of flowering. From then onwards well-watered plots were supplied with water every other week. Data were obtained as described in the previous WUE experiment. 49 Table 2. Bean genotypes used in the water-use efficiency experiment (1986). Entry Acc. # Origin Pedigree 1 186030 WH Tepary, Az WH Tepary, A2 2 186032 LEF—2-RB MX LEF-Z-RB MK 3 B76001 61065 MSU N2/BTS 4 186033 11-900-5-m MX II-900-5-M MK 5 N80068 802702 MITA/MSU N76001,61068, I81110,B-351 6 882008 80B001-53 MSU 49-242/TUS/2xBM /3/MID 7 N81017 77011-F3 MSU Kent/NZIIPIJ/ BUN 8 186040 Vita 3, IITA 9 186037 Acc.# 5, MALAWI 10 186039 Acc.# 6, MALAWI 50 Statistical Analysis All data obtained during the three years that these studies were carried out were analyzed using MSTAT version 4, a micro-computer statistical package developed, revised, and updated by the Department of Crop and Soil Science at Michigan State University, East Lansing, Michigan, U.S.A. RESULTS Microclimatology In 1984 planting was done on 14 June. There was no significant precipitation from this date until 26 days later on 10 and then 11 July when about 33 mm of precipitation were received (Figure 1). This was convenient because a week later the early genotypes (accessions 2, 6, 10 and 14) started flowering and the precipitation received at this time prevented large losses in seed yields especially through reductions in number of pods/plant. However, there was no significant precipitation again for another 23 days until 4 August when 20 mm and then 30 mm four days later were received. It was during this second drought period that soil moisture was first sampled (31 July) with the tensiometer. Another stress period of 21 days ensued up until August 30 when 7 mm were received. The second soil moisture sample was taken during this stress period (21 August). The third sample was taken 5 days after the small drizzle (4 September). Generally, the 1984 crop season was highly moisture stressed, receiving only 287 mm over the entire period of May, June, July, August, and September, distributed as 102, 4, 50, 65, and 66 mm, respectively. June had virtually no rainfall. On the other hand, pan evaporation totalled 810 mm over the same crop season. Therefore, although there might have been adequate moisture 51 52 1984 0 8mg °=Illn 0 so . ' - 2 30 P . . . . . O Q. 3 O O . O o . 'fi 0 0 . 5A 20 l. O O O " 3’ b o 00 .. 10-° 0 ° 0 o f .2 o 'C D 0 oo 10 - a .5 o ° on O O 8 8 .. a o D E) _ 8? 0 ° gbg’ 5H5 c o ° D o . m ‘3 o s D - C 4 5.. D - 6! 0" 9:1 ‘0 ‘ 4 SA - . 2.5 E 20 - I i ‘ E - . .L : ' i '+ d. 1 15 15 1 15 1 15 30 Dubs June July August Sspmmber Figure 1. Meteorological characteristics at East Lansing, Michigan, 1984. 53 reserves, drought stress was rampant merely due to poor distribution of precipitation and high atmospheric demand expressed here in terms of evaporation. In 1985, at East Lansing, the amount of water applied to beans was controlled. However, the distribution of precipitation during this crop season was better (Figure 2) than in the previous year. Amounts of precipitation obtained were 71, 57, 53, 96, and 88 mm for May, June, July, August, and September, respectively. This represented almost 100 mm over the previous season. However, atmospheric demand for moisture (858 mm) was comparable to the previous year. i The distribution of precipitation at Montcalm in 1985 was good and much better than that at East Lansing in 1984 (Figure 3). A total of 351 mm was obtained and distributed as follows: 40, 72, 67, 121, and 51 mm in May, June, July, August, and September, respectively. A pan evaporimeter was not installed as yet and so pan evaporation data were not obtained. 54 I985 8 :0 O O ('C) o o o o 0 o 0 Alt mature 8 O O O O O O O O 0 O 0 LE 6 EA 8% 0° ‘ gg 90 a0 ll)! 6 o 00 ‘ c 4 on 0° b a o- & u a D C M DEL D n 11 D 40 4 lie ‘ C m .4 .a II 4 c i l Ll _ 1 15 1 15 1 15 1 15 30 on. June July August September Figure 2. Meteorological characteristics at East Lansing. _ Michigan, 1985. 55 1985 O-Iss O'Im . .J 30 . 0 . 0 . L . 0 . . .0 . ' . . b. 0 .0 O 0 0 O . 1’ E>0 0 c o O O .3oh, * 1* -. E. it t * x 5‘20... * ti“? * t ~ “‘x '* * 3310' * * . v . 40*- -I —A -1 3:20 I , V d c a: :L l : . 1 15 1 15 1 15 1 15 .30 DO!” .wms Fwy thumb: Slmmmmst Figure 3. 1985. Meteorological characteristics at Montcalm, Michigan, 56 Intercropping Experiments 1984 Intercropping Experiment As was shown earlier (Figure 1), there was lack of precipitation from plating up to near flowering and then another period of lack of precipitation that extended into the seed-filling stage. The result of these stress periods was a general increase in soil moisture tension from 47 to 81 DAP (Figure 4). The increase meant that as water evaporated and/or was taken up by the plants the remaining water had to be taken up by a larger force than previously. This indicates that the amount of moisture in the soil was decreasing. This decrease was generally faster between 68 and 82 DAP than between 47 and 68 DAP, the mean increases being 24.1 kPa and 8.75 kPa, for the respective periods. This was probably due to a rather active reproductive period (seed-filling) when water is needed in the plant for translocation of assimilates from current photosynthesis and/or remobilization of assimilates from previous sinks. Moisture depletion was significantly greater in the soil under intercrops than under monocultures. At the first sampling (two weeks after a rain of 33 mm) the amounts of soil moisture were similar in soil under intercrops as those under monocultures. However, with subsequent lack of precipitation and with high atmospheric moisture demand, there was significantly less soil moisture under intercrops than under monocultures. This difference became even wider 5 '7 m l’ I 0 - Intercrop ' o - Monoctmurs o 8 00 Imohhws Iwmhm . (RP!) o 0 So“ l l 40 so so 100 Dari aflmr pumMng Moisture tension of soil under monoculture and Figure 4. intercropped Malawian landrace dry beans (1984). 58 (Figure 4) by 82 DAP. By this time, intercropped beans had reached a point where irrigation would have been necessary (at least 50 kPa) whereas those under monocultures had not reached that point as yet (Figure 4). This observation indicates that beans grown in intercropping systems with maize rely on reduced soil moisture levels and, should a drought occur, it would affect them before it would affect beans grown under monoculture systems. One could speculate that some Malawian landraces, selected naturally under intercropping systems, are expected to tolerate drought conditions better because the two stresses (intercropping and drought) seem to be confounded and not quite separable. Stated differently, drought seems to be a factor of the environment that some Malawian landraces of beans have had to overcome in order to have been selected naturally as they did and maintain themselves under the cropping systems prevalent in Malawi. In most of these systems, main crops out-compete the intercrops for water, light and nutrients. Leaf diffusive resistance was measured on three dates, namely; 70, 78, and 85 DAP in 1984. Both pre—dawn and mid- afternoon resistances were measured. Pre-dawn leaf diffusive resistances were similar in leaves of different genotypes and between cropping systems but were significantly lower than mid-afternoon diffusive resistances (Tables 3, 4, and 5). There were no differences among genotypes nor between cropping systems in mean leaf diffusive resistance at 70 DAP (Table 3). At 78 DAP there 59 Table 3. Leaf diffusive resistance of 18 Malawian landrace dry beans at 70 DAP (1984). Accession Leaf diffusive resistance number Pre-dawn Afternoon M I Mean M I Mean -------------------- s/cm—-------------------- 1 0.55 0.45 0.50 4.63 4.27 4.45 2 0.21 0.33 0.33 2.96 3.31 3.13 3 0.38 0.34 0.36 4.74 4.81 4.77 4 0.75 0.68 0.71 3.65 2.79 3.22 5 0.13 0.15 0.14 3.72 3.50 3.61 6 0.49 0.49 0.49 4.21 3.32 3.76 7 0.45 0.32 0.38 3.57 4.26 3.91 8 0.56 0.55 0.55 4.79 3.98 4.38 9 0.40 0.31 0.35 4.62 5.27 4.94 10 0.44 0.47 0.45 4.21 3.99 4.10 11 0.33 0.33 0.33 4.49 4.37 4.43 12 0.41 0.69 0.55 3.67 4.31 3.99 13 0.30 0.37 0.33 3.99 4.47 4.23 14 0.24 0.37 0.30 4.06 5.11 4.58 15 0.27 0.14 0.20 3.91 3.53 3.72 16 0.55 0.78 0.66 4.09 4.52 4.30 17 0.29 0.32 0.30 3.09 3.07 3.08 18 0.34 0.36 0.35 4.40 5.21 4.80 Mean 0.40 0.41 0.40 4.04 4.12 4.08 CV (%) 33.3 M = monoculture; H II intercrop. 6O Table 4. Leaf diffusive resistance of 14 Malawian landrace dry beans at 78 DAP (1984). Accession Leaf diffusive resistance number Pre-dawn Afternoon M I Mean M I Mean -------------------- s/cm--------------------- 1 0.50 0.43 0.46 3.92 6.533: 5.22 2 0.30 0.30 0.30 4.58 5.32 4.95 4 0.36 0.61 0.48 2.54 4.0733 3.30 5 0.36 0.41 0.35 4.16 4.42 4.29 7 0.39 0.56 0.47 3.20 4.2281 3.71 8 0.26 0.36 0.31 3.88 4.883 4.38 9 0.29 0.42 0.35 5.37 7.1933 6.28 10 0.16 0.38 0.27 4.00 5.6513 4.82 11 0.54 0.66 0.60 5.37 5.97 5.67 13 0.52 0.46 0.49 3.54 4.8333 4.18 15 0.25 0.41 0.33 5.87 6.05 5.96 16 0.56 0.59 0.57 4.12 5.18: 4.65 17 0.15 0.48 0.31 3.57 5.12:: 4.34 18 0.25 0.32 0.28 4.37 4.64 4.50 Mean 0.34 0.45 0.39 4.18 5.28 4.73 CV (%) 48.3 monoculture; intercrop; significant between cropping systems at P=0.05; 3* significant between cropping systems at P=0.01. 61 significant between cropping systems at P=0.05; ** significant between cropping systems at P=0.01. Table 5. Leaf diffusive resistance of 12 Malawian landrace dry beans at 85 DAP (1984). Accession Leaf diffusive resistance number Pre—dawn Afternoon M I Mean M I Mean -------------------- s/cm--------------------- 1 1.23 1.20 1.21 4.56 4.96 4.76 2 1.69 1.67 1.68 5.10 5.94 5.52 4 1.42 1.45 1.43 4.47 5.64: 5.05 7 1.57 1.41 1.49 3.82 5.41 4.61 8 1.07 1.02 1.04 4.63 5.06 4.84 9 0.99 1.03 1.01 4.66 4.68 4.67 11 1.17 1.97 1.57 4.75 7.528: 6.13 13 2.01 1.89 1.95 5.27 6.85:! 6.06 15 1.73 1.94 1.83 6.23 7.34 6.78 16 1.91 2.03 1.97 4.75 5.77: 5.26 17 1.61 1.67 1.64 4.68 5.83 5.25 18 1.98 2.13 2.05 3.96 5.22:: 4.59 Mean 1.53 1.62 1.57 4.74 5.85 5.29 CV (%) 50.8 = monoculture; = intercrop; 62 was a significant increase in mean leaf diffusive resistance in leaves on intercropped beans over those in monoculture (Table 4). Genotypes were not significantly different in leaf diffusive resistance. Accessions 1, 4, 7, 8, 9, 10, 13, 16, and 17 significantly increased their diffusive resistance in leaves under intercropping. This indicates that at 78 DAP these genotypes were already reacting to stress conditions under intercropping conditions by closing their stomata. As a result, these genotypes probably conserved more moisture than others. At 85 DAP there still was a significant increase in mean leaf diffusive resistance on leaves of intercropped beans over those in monoculture (Table 5). This came about because accessions 4, 7, 11, 13, 16, 17, and 18 significantly increased their leaf diffusive resistances under intercropping at 85 DAP. Some of these landraces (accessions 4, 7, 13, 16, and 17) had partially closed stomata at 78 DAP. Leaf diffusive resistance should be related to leaf transpiration rates because the partial closing of stomata under high leaf diffusive resistance causes less moisture to escape. The overall result is water conservation and loss of productivity, due to less photosynthesis, although much of it is regained in this case as opposed to plants that do not conserve water. At 70 DAP only 2 genotypes, accessions 7 and 13, were able to significantly reduce their rates of transpiration in the process of conserving water (Table 6). At 78 DAP all genotypes, except accession 11, significantly 63 Table 6. Leaf transpiration of 189 Malawian landrace dry beans at 70 DAP (1984). Accession Leaf transpiration number Pre-dawn Afternoon M I Mean M I Mean ---------------- ug HzO/cm3/s-------------—----- 1 1.22 1.11 1.16 1.40 1.11 1.26 2 1.00 0.91 0.95 1.67 1.54 1.60 3 1.10 1.01 1.06 1.43 1.16 1.30 4 1.15 1.08 1.11 1.60 1.61 1.60 5 1.12 1.07 1.10 1.57 1.46 1.52 6 1.12 1.06 1.09 1.40 1.43 1.41 7 1.13 1.13 1.13 1.51 1.0333 1.27 8 1.10 1.02 1.06 1.40 1.22 1.31 9 1.18 0.95 1.07 1.36 1.13 1.25 10 1.20 1.04 1.12 1.52 1.31 1.41 11 1.10 0.99 1.04 1.51 1.32 1.41 12 0.94 0.92 0.93 1.57 1.23 1.40 13 1.06 0.97 1.01 1.48 0.958! 1.21 14 1.04 1.12 1.08 1.44 1.23 1.34 15 0.91 0.92 0.91 1.50 1.51 1.51 16 0.93 0.86 0.90 1.47 1.34 1.41 17 1.03 0.99 1.01 1.63 1.39 1.51 18 0.88 0.79 0.83 1.45 1.27 1.36 Mean 1.07 0.99 1.03 1.50 1.29 1.39 CV (%) 33. M = monoculture; I = intercrop; it significant between cropping systems at P=0.01. n 11 it nc re So in ra ma th. Ch: thx 101 to ada} Plan 64 reduced their rates of transpiration under intercropping (Table 7) due to stomatal closure. Similarly, at 85 DAP some genotypes (accessions 1, 4, 7, 11, 16, and 18), most of which had significantly higher leaf diffusive resistance under intercropping, significantly reduced (Table 8) their rates of transpiration due to partial stomatal closure. Generally, the leaf temperature of leaves on plants under drought stress rises above that of leaves of nonstressed plants. On the contrary, in this study the mean leaf temperature of stressed plants (i.e. those under the intercropping system) was significantly lower than that of nonstressed plants (Tables 9, 10, and 11). Significant reductions were in accessions 3, 4, 5, 7, 8, 10, and 13. Some of these (4, 7, and 13) had both significant increases in leaf diffusive resistance and reductions in transpiration rates. Reductions in leaf temperature under intercropping may have been due to partial shading by the maize, although this was avoided by the sparse maize planting pattern, or to changes in leaf orientation under stress. At 78 DAP only three genotypes, accessions 2, 9, and 10, had significantly lower leaf temperatures (Table 10). However, at 85 DAP leaves from intercropped plants were at temperature similar to those of monoculture plants (Table 11). Specific leaf weight (SLW) has been shown to be an adaptive characteristic under evolutionary processes. Plants evolving in extremely low moisture environments (xerophytic plants) have vary large SLW as opposed to those 65 Table 7. Leaf transpiration of 14 Malawian landrace dry beans at 78 DAP (1984). Accession Leaf transpiration number Pre-dawn Afternoon M 1 Mean M I Mean ---------------- ug HzO/cm3/s------------------- 1 0.97 0.833 0.90 0.97 0.7133 0.84 2 0.83 0.76 0.79 0.98 0.76:: 0.87 4 0.86 0.91 0.88 1.21 0.80:: 1.00 5 0.82 0.81 0.81 0.97 0.68:: 0.82 7 0.89 0.81 0.85 1.05 0.80:¥ 0.92 8 0.89 0.77 0.83 0.96 0.78:: 0.87 9 0.86 0.85 0.85 0.89 0.7433 0.81 10 0.82 0.78 0.80 1.04 0.7834 0.91 11 1.04 1.00 1.02 0.90 0.79 0.84 13 0.82 0.74 0.78 1.00 0.6733 0.83 15 0.78 0.78 0.78 0.87 0.688! 0.77 16 0.90 0.83 0.86 0.97 0.8188 0.89 17 0.85 0.77 0.81 1.12 0.61:! 0.89 18 0.87 0.80 0.83 0.97 0.7533 0.86 Mean 0.87 0.82 0.84 0.99 0.74 0.86 CV (%) 13.2 M = monoculture; I I intercrop; significant between cropping systems at P=0.05; *3 significant between cropping systems at P=0.01. 66 Table 8. Leaf transpiration of 12 Malawian landrace dry beans at 85 DAP (1984). Accession Leaf transpiration number Pre-dawn Afternoon M I Mean M 1 Mean ---------------- ug H30/cm3/s------------------- 1 1.13 1.15 1.14 1.17 0.95! 1.06 2 1.01 1.03 1.02 1.12 0.95 1.03 4 1.12 1.06 1.09 1.16 0.9153 1.03 7 1.11 1.05 1.08 1.20 0.9333 1.06 8 1.12 1.12 1.12 1.11 1.11 1.11 9 1.11 1.01 1.06 1.15 1.08 1.11 11 1.13 1.23 1.18 1.16 0.96! 1.01 13 1.14 1.08 1.11 1.19 0.973 1.08 15 1.07 1.02 1.04 1.12 0.99 1.05 16 1.13 1.10 1.11 1.16 0.833! 0.99 17 1.07 1.00 1.03 1.15 0.99 1.07 18 1.18 1.06 1.12 1.19 1.00: 1.09 Mean 1.11 1.07 1.09 1.15 0.95 1.05 CV (%) 44.3 monoculture; intercrop; significant between cropping systems at P=0.05; 3* significant between cropping systems at P=0.01. 67 Table 9. Leaf temperature of 18 Malawian landrace dry beans at 70 DAP (1984). Accession Leaf temperature number Pre-dawn Afternoon M I Mean M 1 Mean .................... c -------------_-----_----- 1 15.5 15.8 15.6 21.2 21.4 21.3 2 14.8 15.0 14.9 22.4 21.9 22.1 3 15.1 15.4 15.2 22.1 21.3: 21.7 4 15.9 16.4 16.1 22.3 21.7: 22.0 5 14.8 14.9 14.8 22.4 21.6: 22.0 6 15.3 15.6 15.4 21.1 21.1 21.1 7 15.1 15.1 15.1 22.1 21.33 21.7 8 15.6 15.6 15.6 21.5 20.83 21.1 9 15.5 15.5 15.5 21.7 21.3 21.5 10 15.5 15.7 15.6 21.9 21.31 21.6 11 16.5 16.7 16.6 22.5 22.5 22.5 12 15.0 15.5 15.2 22.1 22.0 22.0 13 14.9 15.0 14.9 21.7 20.8t 21.2 14 15.3 15.3 15.3 22.0 21.5 21.7 15 14.9 14.9 14.9 22.1 21.6 21.8 16 15.7 15.9 15.8 22.1 22.0 22.0 17 14.8 14.9 14.8 22.1 21.5 21.8 18 14.6 14.8 14.7 22.4 22.0 22.2 Mean 15.3 15.4 15.3 22.0 21.5 21.7 CV (%) 9.7 monoculture; intercrop; significant between cropping systems at P=0.05; it significant between cropping systems at P=0.01. 68 Table 10. Leaf temperature of 14 Malawian landrace dry beans at 78 DAP (1984). Accession Leaf temperature number Pre-dawn Afternoon M I Mean M I Mean .................... C ------_----__----__--__-- 1 13.7 13.2 13.4 25.2 25.1 25.1 2 11.3 11.5 11.4 27.6 25.8: 26.7 4 12.8 13.1 12.9 26.5 26.0 26.2 5 11.0 11.2 11.1 25.6 25.2 25.4 7 12.0 12.1 12.0 26.0 25.3 25.6 8 12.5 12.7 12.6 25.6 25.8 25.7 9 12.4 12.4 12.4 27.2 23.0t 25.1 10 9.6 14.9! 12.2 27.9 25.98 26.9 11 ,14.1 14.4 14.2 28.4 27.1 27.7 13 11.7 11.8 11.7 26.4 25.4 25.9 15 11.2 11.4 11.3 27.4 26.7 27.0 16 13.6 13.2 13.4 27.2 26.3 26.7 17 11.1 11.5 11.3 27.5 26.1 26.7 18 10.8 10.7 10.7 26.4 25.6 26.0 Mean 12.0 12.4 12.2 26.8 25.6 26.2 CV (%) 19.4 M = monoculture; H II intercrop; 3 significant between cropping systems at P=0.05; 69 Table 11. Leaf temperature of 12 Malawian landrace dry beans at 85 DAP (1984). Acce s s i on Leaf temperature numbe r Pre-dawn Afternoon M 1 Mean M I Mean .................... C ---------_--_---------_-_ 1 17.2 17.0 17.0 19.9 19.7 19.8 2 17.1 16.9 17.0 19.9 19.8 19.8 4 17.6 17.3 17.4 19.8 19.5 19.6 7 17.2 17.1 17.1 19.9 19.9 19.9 8 16.7 16.7 16.7 19.8 19.9 19.8 9 16.6 16.7 16.6 20.0 19.9 19.9 11 17.9 18.84 18.3 20.5 20.2 20.3 13 18.4 18.1 18.2 20.9 20.2 20.8 15 17.2 17.2 17.2 20.5 20.1 20.3 16 18.4 18.0 18.2 20.1 19.8 19.9 17 17.2 17.2 17.2 19.8 19.5 19.6 18 19.3 18.8 19.0 20.5 20.4 20.4 Mean 17.5 17.5 17.5 20.1 19.9 20.0 CV (%) 29.2 M = monoculture; 1.: intercrop; 1 significant between cropping systems at P=0.05; 7O growing under fair to adequate moisture environments (mesophytic plants). Measured under both intercropped and monoculture conditions at 74 DAP, specific leaf weight tended to be lower, though insignificantly, in intercropped bean leaves. The eighteen genotypes tested differed significantly in SLW (Table 12). Accession 14 had the highest SLW. Leaf moisture retention capacity (LMRC) measured at 72 DAP did not differ between cropping systems but was significantly different among the malawian landrace genotypes (Table 12). Accessions 2, 15, and 17 had significantly more moisture retained in the leaves than other genotypes whereas 12 and 14 had significantly less moisture retained in their leaves. SLW was similar (Table 13) in leaves of beans grown on the different soil types (heavy, medium, and light). However, there was a significant decline in LMRC in leaves on plants growing on heavier soil. Rapid water percolation leads to faster moisture depletion in the light-textured soil and earlier drought as a result. These data show that LMRC is probably a better characteristic that plants use to adapt to soil moisture deficits than the SLW. One difference between SLW and.LMRC is that SLW becomes a permanent adaptation as opposed to LMRC. When the soil is rehydrated total leaf are willlmve been set and limited by leaf thickness as opposed to expansion. Therefore, LMRC is not as conservative a traHLas SLW and it responded more flexibly to the environment . 71 Table 12. Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 18 Malawian dry beans at 74 and 72 DAP, respectively (1984). Accession MM mmc number M 1 Mean M I Mean ------- mg/cm3------ ----------%------------ 1 3.33 3.31 3.32 cdef 28.5 30.9 29.7 bcd 2 3.61 3.88 3.74 bcdef 41.7 40.9 41.3 a 3 4.49 4.08 4.28 b 21.7 25.3 23.5 d 4 3.60 3.17 3.38 cdef 21.6 31.2 26.4 cd 5 3.67 3.33 3.50 bcdef 22.7 27.9 25.3 d 6 4.47 3.81 4.13 bc 24.8 34.0 29.4 bcd 7 2.97 2.84 2.91 f 27.8 30.9 29.3 bed 8 4.44 3.51: 3.98 bcd 22.8 25.3 24.0 d 9 3.30 2.73 3.02 ef 36.2 37.2 36.7 ab 10 3.44 3.37 3.41 cdef 38.5 31.8 35.1 abc 11 3.08 2.69 2.89 f 28.9 34.3 31.6 abcd 12 3.84 3.77 3.81 bcde 18.9 17.4 18.1 e 13 3.61 3.86 3.74 bcdef 34.6 35.9 35.3 abc 14 5.75 6.07 5.91 a 13.3 16.0 14.7 e 15 3.44 3.22 3.33 cdef 44.2 36.3 40.2 a 16 3.33 3.11 3.22 def 22.3 27.3 24.8 d 17 3.59 3.64 3.62 bcdef 42.9 39.2 41.1 a 18 3.35 3.09 3.22 def 34.2 38.7 36.4 ab Mean 3.74 3.53 3.63 29.2 31.1 30.1 CV (%) 14.5 32.2 M = monoculture ; I intercrop; isimuficant between cropping systems at P=0.05; mmnsfollowed by the same letter are not significantly (PfiLOH different. 72 Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 18 Malawian dry bean genotypes grown in three different soil types (1984). Tuitile: 1:3. Soil type SLW IMRC -- mg/cma -- -- % -- Light 3.55 33.7 a Medium 3.68 29.4 b Heavy 3.72 26.6 c Mean 3.63 29.9 CV (%) 14.5 32.2 73 There was a significant mean biomass yield reduction under intercropping as compared with monoculture. However, only four of the sixteen genotypes that were harvested yielded significantly lower biomass under intercropping than the monoculture cropping system (Table 14). Because there were no differences between cropping systems for apparent harvest index (AHI) (Table 17), economic yield production responded much the same as biomass production (Table 14). Assuming similar biomass production under either cropping system, some Malawian landraces of beans yield well under the intercropping system just as they do in monocultures. Four methods (Table 15) were used in assessing adaptability of the landraces to the intercropping systems and possible consequent drought stresses. Yield differential seemed to be a poorer method than the other three. For example, accession 9 had one of the smallest differentials but the other methods except S indicated that it was drought susceptible. The arithmetic mean identified accessions 4, 10, 7, 6, (2), 13, 5, and 17 to be better adapted to intercropping and drought stress, in that order, than other genotypes. Accessions 9, 1, 3, and 14 were least adapted to such conditions. The geometric mean identified 10, 4, 7, 6, 5, and 17 to be better adapted and 9, 1, and 3 to be least adapted. The drought susceptibility index (S) excludes the yielding potential of a genotype in assessing its drought resistance. In this case it identified 2, 4, 5, 3, 8, 12, and 17 to be better adapted and 13, 1, 16, 15, and 74 Table 14. Biological and economic yield of 16 Malawian dry bean genotypes (1984). Acc. Biological yield Economic yield no. M 1 Mean M I Mean ----------------------- g/plant ----—---------—-------- 1 119.9 66.04: 93.0 abcd 43.2 17.843 30.5 de 2 116.5 100.8 108.6 abc 38.3 50.4 44.3 abcd 3 64.1 50.1 57.1 d 37.3 31.6 34.4 cde 4 132.0 123.0 127.5 a 58.7 69.9 64.3 a 5 84.0 77.4 80.7 bcd 57.2 49.0 53.1 abcd 6 100.5 80.7 90.6 abcd 69.2 51.03 60.1 ab 7 123.0 96.9 109.9 ab 69.2 55.0 62.1 ab 8 99.4 75.6 87.5 abcd 50.1 42.1 46.1 abcd 9 79.6 98.4 89.0 abcd 16.2 12.3 14.3 e 10 115.7 90.4 103.1 abc 74.0 52.94 63.5 ab 12 70.0 55.7 62.9 cd 45.0 36.5 4.08 abcd 13 148.0 76.23: 112.1 ab 87.9 24.931 56.4 abc 14 80.5 45.8 63.2 cd 51.5 27.3tt 39.4 bcd 15 133.4 69.54! 101.5 abc 53.5 27.244 40.3 abcd 16 116.8 56.113 87.0 abcd 61.4 27.584 44.5 abcd 17 123.7 107.0 115.3 ab 58.0 47.1 52.6 abcd Mean 106.7 79.4 93.0 54.4 38.9 46.6 CV (%) 23.9 20.3 M = monoculture; 1 = intercrop; * significant between cropping systems at P=0.05; it significant between cropping systems at P=0.01. Means followed by the same letter are not significantly (P=0.01) different. 75 Table 15. Yield differential, arithmetic mean, geometric mean, and drought susceptibility index (8) of 16 Malawian landrace dry beans grown in monoculture and intercropping with maize (1984). Accession [Ym-Yi] Lxm+11| [meYiJUa S 2 number (1) (2) (3) (4) 1 25.4 30.5 27.7 2.06 2 -1201 4403 4309 -1011 3 5.7 34.4 34.3 0.54 4 -11.2 64.3 64.0 -0.67 _5 8.2 53.1 52.9 0.50 6 18.2 60.1 59.4 0.92 7 14.2 62.1 61.7 0.72 8 8.0 46.1 45.9 0.56 9 3.9 14.3 14.1 0.84 10 21.1 63.5 65.6 1.00 12 8.5 40.8 40.5 0.66 13 63.0 56.4 46.8 2.51 14 24.2 39.4 37.5 1.65 15 26.3 40.3 38.1 1.71 16 33.9 44.5 41.1 1.94 17 10.9 52.6 52.3 0.67 Ym = monoculture seed yield; Yi = intercrop seed yield; 1 = yield differential; 2 = arithmetic mean; 3 = geometric mean; and 4 = drought susceptibility index. 76 Table 16. Number of pods/plant and number of seeds/plant of 16 Malawian dry bean genotypes (1984). Acc. No. pods/plant No. seeds/plant no. M I Mean M I Mean ------------------------- no. -------------------—--—- 1 74.3 36.23! 55.2 a 234 101*: 167 abc 2 34.1 35.1 34.6 abc 108 120 114 3 25.1 24.9 25.0 c 103 94 98 4 56.1 41.9 49.0 ab 170 165 167 5 38.7 32.3 35.5 abc 220 185 202 a 6 35.4 31.8 33.6 abc 163 113 138 abc 7 38.7 37.3 38.0 abc 137 126 131 abc 8 38.7 32.2 35.4 abc 139 107 123 abc 9 22.4 16.8 19.6 c 41 28 34 d 10 28.2 27.9 28.0 bc 203 1333 168 abc 12 42.7 18.933 30.8 abc 207 155 181 ab 13 47.6 21.2tt 34.4 abc 261 7233 166 abc 14 29.8 18.3 24.0 c 112 79 95 cd 15 33.3 14.83* 24.0 c 118 60 89 cd 16 43.1 21.4! 32.2 bc 158 75* 116 be 17 37.9 26.9 32.4 bc 125 92 108 bcd Mean 39.1 27.4 33.2 156 107 131 CV (%) 31.5 28.8 M = monoculture; I = intercrop; * significant between cropping systems at P=0.05; 3* significant between cropping systems at P=0.01. Means followed by the same letter are not significantly (P=0.05 and P=0.01 for pods and seeds, respectively) different. ‘77 14 to be least adapted to intercropping and consequent drought conditions. The number of pods/plant is a characteristic used by those genotypes that were better adapted to intercropping to produce similar economic yields as in monoculture. There was a significant reduction in number of pods/plant (Table 16) in accessions 1, 12, 13, 15, and 16 under monoculture. It seems that the number of pods/plant largely controlled number of seeds/plant produced (Table 16). Genotypic variability existed for stem weight (Table 18). There was a significant reduction in stem weights of bean plants under intercropping as compared with those in monoculture in 1984. Shaded plants usually increase in height faster than unshaded plants so that they can be exposed to light. The less investment of assimilates into stem tissue leads to slender stems. Assuming a positive correlation between stem weight and amount of reserves, one expects that in the event of a drought occurring during seed-filling intercropped bean plants rely on significantly reduced amounts of reserves for remobilization to support seed growth. Most drought tolerant genotypes, identified by other criteria, did not have significantly smaller stem weights. Landrace 9 had higher stem weight under intercropping than under monoculture. This was probably due to the better climbing ability of this landrace which made it grow more than without support in monoculture. Table 17. 78 Apparent harvest index (AHI) and 100-seed weight of 16 Malawian dry bean genotypes (1984). Accession AHI 100-seed weight number M I Mean M I Mean --------- % ----------- ----- g/100 seeds ----- 1 37.9 27.2 32.5 e 18.5 17.7 18.1 c 2 33.4 50.9 42.1 cde 35.8 41.8 38.8 abc 3 58.1 63.0 60.5 ab 36.3 33.9 35.1 abc 4 44.1 57.4 50.7 abcd 34.3 44.2 39.2 abc 5 68.0 63.3 65.6 a 25.9 26.6 26.2 abc 6 70.7 63.3 67.0 a 42.3 45.4 43.8 ab 7 59.1 57.2 58.1 abc 50.2 43.6 46.9 a 8 50.9 58.1 54.5 abcd 36.1 59.5 47.8 a 9 20.4 12.7 16.5 f 41.9 45.4- 43.6 ab 10 63.9 59.0 61.4 ab 36.5 39.9 38.2 abc 12 _62.8 65.3 64.0 ab 22.0 23.5 22.7 bc 13 64.1 33.3 48.7 abcde 33.7 35.2 34.4 abc 14 63.8 60.4 62.1 ab 65.0 34.7 49.8 a 15 40.5 39.2 39.8 de 46.8 47.4 47.1 a 16 52.0 48.9 50.4 abcd 40.8 36.6 38.7 abc 17 47.0 44.6 45.8 bcde 46.9 51.0 48.9 a Mean 52.3 50.2 51.2 38.3 39.1 38.7 CV (%) 14.1 32.7 M = monoculture; I = intercrop; Means followed by the same letter are not significantly (P=0.01) different. 79 Table 18. Other agronomic characteristics of 16 Malawian landrace dry beans (1984). Accession No. non-seeded Stem weight pods/plant number M I Mean M I Mean ------- no. ------- ------ g/plant -------- 1 12.0 11.5 11.8 abc 60.6 39.74 50.1 ab 2 6.7 6.4 6.5 a-d 60.2 30.053 45.1 abc 3 3.7 6.1 4.9 bcd 14.7 9.4 12.1 e 4 4.0 4.5 4.3 cd 51.9 33.0 42.5 a-d 5 4.8 2.7 3.8 cd 16.9 18.1 17.5 de 6 7.3 6.0 6.7 a-d 15.5 14.4 15.0 e 7 4.0 3.5 3.8 cd 37.8 25.6 31.7 b-e 8 6.3 5.2 5.8 a-d 34.3 21.2 27.7 b-e 9 15.7 10.5 13.1 a 50.3 77.93! 64.1 a 10 8.7 8.8 8.8 a—d 20.9 21.4 21.2 cde 12 6.7 5.1 5.9 a-d 13.7 9.5 11.6 e 13 4.1 3.9 4.0 cd 33.2 37.1 35.1 b-e 14 12.6 12.7 12.6 ab 15.7 7.9 11.8 e 15 3.9 3.5 3.7 cd 56.1 29.2! 42.7 a-d 16 3.9 1.8 2.9 d 36.8 21.6 29.2 b-e 17 8.3 3.4 5.9 a—d 39.5 43.3 41.1 a-d Mean 7.0 6.0 6.5 34.9 37.5 31.2 CV (%) 58.4 37.7 M = monoculture; I intercrop; 3 significant between cropping systems at P=0.05; ** significant between cropping systems at P=0.01; Means followed by the same letter are not significantly (P=0.01) different. 80 1985 Intercropping Experiment Leaf diffusive resistances tended to be higher in some genotypes, though not significantly, in beans grown under the intercropping system than those grown in monocultures (Table 19). However, leaf diffusive resistance differed significantly among the dry bean genotypes. Accessions 2 and 5, and BAT 47 had significantly higher leaf diffusive resistances. Probably as a result of similar leaf diffusive resistance under the two cropping systems, leaf transpiration rates did not differ in beans grown under either intercropping or monoculture. But genotypes differed significantly in rates of leaf transpiration (Table 19). Accessions 2, 5, and 6, and BAT 47 had lowest leaf transpiration rates, thereby conserving the most water. There were no differences in leaf temperature between cropping systems and among the dry bean genotypes, probably due to lower levels of solar radiation. On the day these readings were taken there was a partly cloudy sky and a consequent decline in solar radiation. There was a significant decline in mean economic yield (Table 20) in beans under the intercropping system (20.9 g/plant) from those in monoculture (27.6 g/plant). The dry bean genotypes also differed significantly in economic yield. The highest yielders were Montcalm (an adapted cultivar), BAT 798, BAT 85, and accessions 10 and 7, in that order, although 6 and 5 were close behind these. The four methods of determining relative drought tolerance were again Table 19. Leaf diffusive resistance, 81 leaf transpiration, and leaf temperature of 22 dry bean genotypes (1985). Acc./ Leaf diffusive Leaf Leaf cv. resistance transpiration temperature M I Mean M I Mean M I Mean ------ s/cm --- ug H30/cm3/s ---- C ---- 1 0.48 0.60 0.54 14.5 12.63 13.6 cd 22.4 22.3 22.3 2 0.61 0.67 0.64 13.0 11.8 12.4 d 22.3 22.2 22.2 3 0.54 0.44 0.49 15.7 16.8 16.3 a-d 22.6 22.5 22.5 4 0.40 0.38 0.39 18.2 18.3 18.2 abc 22.6 22.7 22.6 5 0.66 0.56 0.61 13.2 14.0 13.6 cd 22.5 22.6 22.5 6 0.40 0.51 0.45 15.8 14.3 15.0 bcd 22.2 22.2 22.2 7 0.48 0.35 0.42 15.3 18.83 17.0 a-d 22.3 22.0 22.1 8 0.54 0.52 0.53 13.7 14.3 14.0 cd 22.2 22.1 22.1 9 0.46 0.50 0.51 15.9 14.7 15.3 bcd 22.4 22.6 22.5 10 0.64 0.41 0.53 13.7 16.83 15.3 bcd 22.3 22.2 22.2 12 0.57 0.54 0.55 15.4 16.5 15.9 bcd 23.1 23.1 23.1 13 0.38 0.57 0.48 18.8 15.2 17.0 a-d 22.2 22.2 22.2 14 0.40 0.40 0.40 18.7 17.1 18.2 abc 22.3 22.4 22.3 15 0.28 0.31 0.30 21.4 20.8 21.1 a 22.2 22.2 22.2 16 0.39 0.47 0.43 18.5 16.44 17.5 a-d 22.6 26.0 24.3 17 0.27 0.31 0.30 21.0 19.13 20.0 ab 22.3 22.2 22.3 BM 0.42 0.50 0.46 16.3 16.3 16.3 a-d 21.9 21.9 21.9 BAT337 0.40 0.33 0.36 16.2 17.9 17.1 a-d 22.0 21.8 21.9 BAT47 0.57 0.64 0.61 13.6 12.0 12.8 cd 22.2 22.1 22.1 BAT798 0.47 0.48 0.47 16.0 15.7 15.9 a-d 22.9 22.9 22.9 BAT85 0.44 0.38 0.41 16.3 17.9 17.1 a-d 22.2 22.1 22.2 MC 0.34 0.42 0.38 18.4 17.1 17.3 a-d 21.8 21.8 21.8 Mean 0.46 0.47 0.46 16.3 16.1 16.2 22.3 22.4 22.3 CV (%) 27.1 16.6 3.8 M=monoculture, I=intercrop; Bm=Bayo Madero; significant between cropping systems at P=0.05; Means followed by the same letter are not significantly (P=0.01) different. MC=Montcalm; * 82 Table 20. Economic yield at physiological maturity and leaf moisture retention capacity (LMRC) at 45 DAP of 22 dry bean genotypes (1985) Accession/ Economic yield LMRC Cultivar M I Mean M I Mean ----- g/plant ------- ------ % —---------- 1 21.6 9.033 15.3 de 21.1 52.9 47.0 ab 2 22.3 13.2tt 17.8 cde 42.1 36.1 39.1 abc 3 17.3 14.1 15.7 de 53.7 41.2 47.5 ab 4 15.8 15.9 15.8 de 43.7 48.1 44.2 ab 5 31.0 27.2 29.1 bcd 40.2 48.1 44.2 ab 6 34.5 25.448 29.9 bcd 50.2 50.0 50.1 ab 7 31.5 28.8 30.2 bcd 47.9 47.9 47.9 ab 8 32.3 24.534 28.4 bcd 41.1 38.1 39.6 abc 9 23.7 20.4 22.0 b-e 44.5 46.1 45.3 ab 10 32.2 29.5 30.9 be 45.7 45.0 45.4 ab 12 29.8 27.8 28.8 bcd 25.7 27.5 26.6 c 13 15.6 6.9: 11.2 e 48.2 40.3 44.3 abc 14 24.7 16.23 20.5 cde 34.8 34.4 34.6 bc 15 36.6 16.033 26.3 bcd 38.8 45.0 41.9 abc 16 15.7 6.93 11.3 e 43.6 41.2 42.4 abc 17 28.3 22.6 25.5 b-e 45.2 61.2 53.2 a Bayo Madero 15.7 14.3 15.0 de 40.4 50.6 45.5 ab BAT 337 28.8 16.5tt 22.7 b-e 47.5 49.1 48.3 ab BAT 47 8.1 21.85! 15.0 de 41.9 41.8 41.8 abc BAT 798 40.7 32.4: 36.5 b 52.9 51.3 52.1 ab BAT 85 35.0 27.7: 31.3 be 46.3 45.2 45.8 ab Montcalm 66.9 41.83! 54.4 a 54.2 52.0 53.1 a Mean 27.6 20.9 24.2 44.1 45.4 44.7 CV (%) 28.5 13.5 M = monoculture; 1 = intercrop; 4 significant between cropping systems at P=0.05; *4 significant between cropping systems at P=0.01; Means followed by the same letter are not significantly (P=0.05 and P=0.01 for yield and LMRC, respectively). 83 used during this year. Bat 47 was harvested before physiological maturity and so will not be used in comparisons. Some of the genotypes with smallest differences in yields between the two cropping systems were accession 4, Bayo Madero, and accessions 12, 10, and 7, in that order (Table 21). This method does not consider yielding ability of the genotypes. The arithmetic and geometric means do. These indicated that Montcalm, BAT 798, BAT 85, and accessions 10 and 7, were better adapted to intercropping conditions. The index of adaptability to stress conditions (8), which delineates adaptation apart from yielding potential, indicated that accessions 4, BAT 47, 12, and 7, Bayo Madero, accessions 5, 9, 17, and 3 were the most adapted, in that order, to intercropping systems. The number of pods/plant, which was previously shown to be a factor for adaptation, again differed significantly among the dry bean genotypes (Table 22). And the number of pods on plants under the intercropping system (15.5) was significantly reduced from that on plants in the monoculture system (20.2). As expected, the number of pods/plant seemed to influence the number of seeds/plant, the latter also being significantly different among the dry bean genotypes (Table 22). Accession 5 (Katolika) produced the most seeds/plant (Table 22) probably due to more seeds/pod and small seedsize (Table 23). As previously, 100-seed weight seemed to play no direct role in adaptation of dry bean genotypes to the intercropping stress (Table 23). Table 21. Yield differential, 84 arithmetic mean, geometric mean, and drought susceptibility index (8) of 22 dry bean genotypes grown in monoculture and intercropping with maize (1985). Accession] [Ym-Yi] IXB+Xi| [meYi]‘/3 S 2 Cultivar (1) (2) (3) (4) 1 12.6 15.3 13.9 0.027 2 9.1 17.8 17.1 0.019 3 3.2 15.7 15.6 0.009 4 -0.1 15.9 15.8 -3.271 5 3.8 29.1 29.0 0.006 6 9.1 29.9 29.6 0.012 7 2.7 30.1 30.1 0.004 8 7.8 28.4 28.1 0.011 9 3.3 22.0 22.0 0.006 10 2.7 30.8 30.8 0.004 12 2.0 28.8 28.8 0.003 13 8.7 11.2 10.4 0.026 14 8.5 20.4 20.0 0.016 15 20.6 26.3 24.2 0.026 16 8.8 11.3 10.4 0.026 17 5.7 25.4 25.3 0.009 Bayo Madero 1.4 15.0 15.0 0.004 BAT 337 12.3 22.7 21.8 0.020 BAT 47 -1307 1500 1303 -00079 BAT 798 8.3 36.5 36.3 0.010 BAT 85 7.3 31.3 31.1 0.010 Montcalm 25.1 54.3 52.9 0.017 Ym = monoculture seed yield; Yi = intercrop seed yield; 1 = yield differential; 2 = arithmetic mean; 3 = geometric mean; 4 = drought susceptibility index. 85 Table 22. Number of pods/plant and number of seeds/plant of 22 dry bean genotypes (1985). Acc./ No. pods/plant No. seeds/plant Cultivar M I Mean M I Mean --------------------- no. -----------—-------- 1 33.2 13.5 25.8 be 153 75 114 abc 2 14.9 9.5 12.2 def 57 34 46 de 3 11.9 8.7 10.3 cf 48 36 42 de 4 13.6 13.9 13.8 def 55 52 54 de 5 20.1 18.8 19.4 cdef157 145 151 a 6 20.4 15.1 17.8 cdef 81 47 64 de 7 23.1 18.0 20.6 bcde 78 68 73 cde 8 24.6 18.9 21.7 bcd 105 79 61 de 9 15.3 13.0 14.2 def 49 41 45 de 10 15.3 14.0 14.7 def 65 61 63 de 12 35.7 34.3 35.0 a 64 60 62 de 13 27.4 16.9 22.2 bcd 43 28 37 e 14 11.7 8.8 10.2 f 45 32 39 de 15 17.6 8.2 12.9 def 95 39 67 cde 16 12.9 5.4 9.2 f 56 24 40 de 17 16.7 11.7 14.2 def 69 50 61 de Bayo Madero 10.9 8.9 9.9 f 48 40 44 de BAT 337 15.7 9.0 12.3 def 46 25 36 e BAT 47 10.6 20.1 15.4 def 38 70 54 de BAT 798 44.4 28.3 36.4 a 158 164 161 a BAT 85 31.9 27. 29.8 ab 138 119 129 ab Montcalm 17.4 12.4 14.9 def 106 74 90 bcd Mean 20.2 25.5 17.8 80 62 71 CV (X) 29.2 30.8 M = monoculture; I = intercrop; B. Madero=Bayo Madero; 3 significant between cropping systems at P=0.05; ** significant between cropping systems at P=0.01; Means followed by the same letter are not significantly (P=0.01) different. 86 Table 23. Number of seeds/pod and 100-seed weight of 22 dry bean genotypes (1985). Acc./ No. seed/pod 100-seed weight cv. M I Mean M I Mean - ------ no. ---------- g/100 seeds ---— 1 4.54 4.09 4.32 def 14.2 12.0 13.1 k 2 3.86 3.60 3.73 efgh 39.0 38.5 38.8 defg 3 3.98 4.11 4.04 defg 36.4 39.5 38.0 efg 4 3.96 3.80 3.88 efgh 28.9 30.4 29.7 ghij 5 7.61 7.68 7.65 a 19.7 18.8 19.2 jk 6 3.98 3.13 3.55 gh 42.4 53.6 48.0 cde 7 3.44 3.77 3.61 fgh 40.4 42.6 41.5 def 8 4.26 4.15 4.20 defg 30.8 31.2 31.0 ghi 9 3.23 3.16 3.20 hi 48.1 49.6 48.9 cd 10 4.21 4.37 4.29 defg 49.4 48.3 48.9 cd 12 1.79 1.74 1.76 J 46.5 46.6 46.6 cde 13 1.66 1.64 1.65 3 34.2 24.8 29.5 ghij 14 3.89 3.55 3.72 efgh 54.6 51.4 53.0 bc 15 5.54 4.80 5.17 c 38.5 40.7 39.6 defg 16 4.32 4.34 4.33 def 28.4 29.5 29.0 ghij 17 4.09 4.32 4.20 defg 41.3 45.7 43.5 cdef Bayo Madero 4.42 4.52 4.47 de 32.7 35.8 34.2 fgh BAT 337 2.95 2.80 2.87 i 62.3 65.5 63.9 a BAT 47 3.84 3.39 3.61 fgh 21.2 31.3 26.2 hij BAT 798 3.87 5.70 4.78 cd 25.8 19.8 22.8 ijk BAT 85 4.30 4.29 4.30 defg 25.4 23.3 24.3 hij Montcalm 6.08 6.00 6.04 b 63.0 56.2 59.6 ab Mean 4.08 4.04 4.06 37.4 38.0 37.7 CV (%) 11.9 10.7 M = monoculture; I = intercrop; Means followed by the same letter are not significantly (P=0.01) different. 87 It is interesting to observe that yield stability analysis on data obtained by the Bean Cowpea CRSP from sixteen locations in Malawi on the same 18 accessions showed that accessions 2, 6, 7, 14, and 17 performed well under poor environments (Table 24). Accessions 3, 4, 5, 9, 10, 11, 12, 15, and 16 were stable across environments. Accessions 1, 8, 13, and 18 performed poorly under poor environments. These environments were not limited to differences in soil moisture and the beans were grown in monoculture systems. However, the genotypes identified in this study as more adaptable to intercropping and consequent drought stress were those that were either good under poor environments or were stable across environments in the stability analysis. WUE Experiments 1985 Experiment There was a tendency by all genotypes used in 1985 to incur a reduction, but not significantly, in leaf elongation rates under drought stress (Table 25). However, because there was a high coefficient of variation (45.2%), no statistically significant differences were observed between the drought and well-watered treatments. There were significant differences among the genotypes. Cowpea genotypes generally had faster leaf elongation rates than dry bean genotypes. On the other hand, there was a general significant reduction in stem elongation rates in five 88 Table 24. Yield stability analysis of the 18 Malawian dry bean genotypes across 16 locations in Malawi. Accession Slope number (b)! 1 1.49 2 0.51 3 0.99 4 1.01 5 1.17 6 0.83 7 0.80 8 1.20 9 0.94 10 1.04 11 1.01 12 0.99 13 1.30 14 0.70 15 0.91 16 1.09 17 0.65 18 - 1.35 4 Calculated from data obtained by Dr. E. Ayeh, Bean/Cowpea CRSP Project, Bunda College of Agriculture, University of Malawi. 89 Table 25. Leaf elongation and stem elongation rates of nine dry bean and three cowpea genotypes (1985). Accession/ Leaf elongation rate Steam elongation rate Cultivar WW DS Mean WW DS Mean ------ mm/day -—--- ------- cm/day -------- 4 2.02 0.98 1.50 cde 2.78 1.67 2.23 cde 5 2.02 1.24 1.63 cde 4.19 2.334 3.26 b 6 3.06 0.36 1.71 cde 3.44 1.604 2.94 bc 16 2.26 1.00 1.63 cde 4.34 1.5483 2.52 bcd 17 1.04 0.60 0.82 e 2.12 0.96 1.54 e Chiapas 7 1.62 0.66 1.14 de 3.50 2.068 2.78 bod BAT 477 2.74 0.52 1.63 cde 1.86 1.47 1.66 e BAT 47 2.66 1.96 2.31 abc 6.23 3.3243 4.77 a Montcalm 2.16 1.80 1.98 bcd 2.54 1.57 2.06 de Vita 4 3.00 2.56 2.78 ab 1.78 1.44 1.61 e Vita 5 3.16 3.16 3.16 a 1.57 1.47 1.52 e Vita 7 3.34 3.20 3.27 a 1.61 1.39 1.50 e Mean 2.42 1.50 1.96 3.00 1.73 2.36 CV (X) 45.2 27.2 WW = well-watered treatment; DS drought stressed treatment; 3 significant between treatments at P=0.05; *3 significant between treatments at P=0.01; Means followed by the same letter are not significantly (P=0.01) different. 90 genotypes: accessions, 5, 6, 16, and Chiapas 7 and BAT 47 (Table 25). Genotypes did not differ among themselves in leaf diffusive resistance, leaf transpiration, and leaf temperature (Table 26). However, there was a significant increase in leaf diffusive resistance and leaf temperature under drought conditions as compared with well-watered conditions. Also, there was a significant reduction in leaf transpiration rates under drought conditions. A significant reduction in total biological yield was observed in accessions 5 and 17, and BAT 47 and Vita 5 (Table 27). Means of total biological yield production among genotypes were significantly different. Chiapas 7 had the highest whereas accession 16 and BAT 477 had the lowest biological yield. Water-use efficiency (WUE), calculated on the basis of total biomass production, increased, but not significantly, under drought conditions from that under well-watered conditions (Table 27). Means of water-use efficiency were significantly different among genotypes. BAT 47 had the highest and accession 6 had the lowest water- use efficiency. Pod weight was significantly reduced in all genotypes, except Chiapas 7, under drought as compared with the well- watered conditions (Table 28). Chiapas 7 was still forming pods when the entire experiment was harvested. The mean number of pods/plant differed significantly between drought and well-watered conditions in all genotypes except 91 Table 26. Leaf diffusive resistance, leaf transpiration, and leaf temperature of 9 dry bean and three cowpea genotypes at 40 DAP (1985). Acc./ Leaf diffusive Leaf Leaf resistance transpiration temperature Cv. WW DS Mean WW DS Mean WW 08 Mean ------ s/cm --- ug H30/cm3/s ----- C ---- 4 0.55 1.65 1.10 28.7 16.4 22.5 31.0 31.8 31.4 5 0.78 2.09 1.43 23.3 11.4 17.4 31.4 32.3 31.9 6 0.63 1.39 1.01 26.2 17.2 21.7 30.9 31.4 31.1 16 0.66 1.25 0.96 25.8 16.3 21.0 31.2 32.1 31.7 17 0.56 1.11 0.84 28.3 18.4 23.3 30.6 31.2 30.9 C. 7 0.65 1.87 1.26 25.6 16.6 21.1 31.1 31.3 31.2 BAT 477 0.66 0.91 0.79 25.0 20.4 22.7 31.0 31.1 31.0 BAT 47 0.59 1.40 1.00 28.8 19.8 24.3 30.5 31.5 31.0 MC 0.65 2.04 1.35 22.6 15.0 18.8 30.2 32.7 31.4 Vita 4 0.52 1.80 1.16 27.2 14.4 18.8 29.8 32.2 31.0 Vita 5 0.57 1.11 0.84 23.5 15.7 19.6 29.7 31.8 30.7 Vita 7 0.78 1.86 1.32 20.5 15.4 18.0 30.2 31.3 30.8 Mean 0.64 1.54 1.09 25.5 16.4 20.9 30.6 31.7 31.2 CV (%) 45.1 28.0 4.57 WW = well—watered treatment; 08 = drought stressed treatment; * significant between treatments at P=0.05; 4* significant between treatments at P=0.01; 92 Table 27. Biological yield and water-use efficiency (WUE) of nine dry bean and three cowpea genotypes (1985). Accession/ Biological yield WUE Cultivar WW DS Mean WW DS Mean ------ g/plant ---- ------ g/kg water ----- 4 53.1 26.2 39.7 abc 5.90 8.76 7.33 abc 5 58.6 27.64 43.1 abc 6.51 9.21 7.86 ab 6 36.5 17.9 27.2 c 4.06 5.75 4.90 c 16 43.3 26.3 34.8 abc 4.81 8.78 6.80 abc 17 50.6 28.48 39.5 abc 5.62 5.16 5.39 bc Chiapas 62.3 43.3 52.8 a 6.93 6.87 6.90 abc BAT 477 38.0 17.4 27.7 c 4.19 5.59 4.89 c BAT 47 60.2 40.38 50.2 ab 6.69 11.87 9.28 a Montcalm 44.2 25.5 34.9 abc 4.92 6.99 5.95 bc Vita 4 36.5 25.5 31.0 bc 4.27 8.22 6.24 bc Vita 5 40.4 27.7 36.0 abc 4.93 9.22 7.07 abc Mean 47.3 27.2 37.3 5.26 7.79 6.52 CV (X) 56.0 55.7 WW = well-watered treatment; DS drought stressed treatment; 3 significant between treatments at P=0.05; Means followed by the same letter are not significantly (P=0.01) different. 93 accession 17 and Montcalm. BAT 47 produced the most pods/plant (Table 28). Significant increases in apparent harvest index (AHI) were observed only in accession 5, and Montcalm (Table 28). BAT 47 showed a decrease in AHI. However, the mean apparent harvest indices were not different among genotypes. Pod yield was used in methods of assessing drought resistance (Table 29). The yield differential proved to be a poor indicator of drought resistance because it generally did not agree with other indicators. For example, Chiapas 7 and BAT 47 had the least differentials. However, both the arithmetic mean and the geometric mean indicated that BAT 47 was the most drought resistant genotype among those that were used and Chiapas 7 was the most susceptible. At harvest, Chiapas 7 had not as yet reached physiological maturity. Accession 5 had the largest differential whereas the other methods indicated that it was drought resistant. According to the arithmetic and geometric means, cultivars BAT 47 and Montcalm, and accession 5 were the most drought tolerant genotypes. The drought susceptibility index (S) identified BAT 47 and Montcalm as the most drought resistant genotypes. 94 Table 28. Pod weight, number of pods/plant, and apparent harvest index (AHI) of nine dry bean genotypes (1985). Acc./ Pod weight No. pods/plant AHI cv WW DS Mean WW DS Mean WW DS Mean --- g/plant --—- ----- no. ------ ---- X ------- 4 25.6 10.734 18.2 a 19.5 11.33 15.4 31.3 34.3 32.8 5 34.2 13.23! 23.7 a 21.5 12.54 17.0 18.7 22.1! 20.4 6 26.2 10.33! 17.7 a 19.5 8.34 13.9 14.3 12.0 13.1 16 23.5 11.843 18.2 a 17.3 10.23 13.7 22.4 25.7 24.0 17 26.2 12.24! 19.2 a 12.0 8.0 10.0 24.0 27.5 25.7 C7 9.1 4.2 6.6 b 24.5 13.34 18.9 48.7 45.4 47.0 B477 26.0 10.54! 18.2 a 19.7 8.8! 14.2 13.9 12.5 13.2 B47 28.3 21.541 24.9 a 31.7 19.33 25.5' 38.0 34.1! 36.0 MC 29.1 15.23! 22.1 a 12.7 6.7 9.7 16.2 21.23 18.7 Mean 25.3 12.2 18.7 19.8 10.8 15.3 22.3 26.1 24.2 CV (X) 40.9 41.3 28.9 C7 = Chiapas 7; B477 = BAT 477; B47 = BAT 47; MC = Montcalm; WW = well-watered treatment; DS = drought stressed treatment; 4 significant difference between treatments at P=0.05; ** significant difference between treatments at P=0.01; Means followed by the same letter are not significantly (P=0.01) different. 95 Table 29. Pod yield differential, arithmetic mean, geometric mean, and drought susceptibility index (S) of 9 dry bean genotypes grown under well-watered and drought conditions (1985). Accession] [wa-Yds] waw+xg§1 [wadeslll3 S 2 Cultivar (1) (2) I (a) (4) 4 14.9 18.1 16.5 1.12 5 21.0 23.7 21.2 1.18 6 15.9 18.2 16.4 1.17 16 11.7 17.6 16.6 0.96 17 14.0 19.2 17.9 1.03 Chiapas 7 4.9 6.6 6.2 1.03 BAT 477 15.5 18.2 16.5 1.15 BAT 47 6.8 24.9 24.7 0.46 Montcalm 13.9 22.1 21.0 0.92 wa = pod yield under well-watered conditions; Yds = pod yield under drought-stressed conditions; 1 = pod yield differential; 2 = arithmetic mean; 3 = geometric mean; 4 = drought susceptibility index. 96 Table 30. Stem'and leaf weights of 9 dry bean and three cowpea genotypes (1985). Accession] Stem weight Leaf weight Cultivar WW DS Mean WW DS Mean -------------------- g/plant -----------------—-- 4 15.9 7.9 11.9 11.6 7.6 9.6 5 10.7 5.8 8.2 13.6 8.7 11.1 6 5.2 3.2 4.2 5.2 4.3 4.7 16 9.5 6.8 8.7 10.2 7.7 8.9 17 11.4 6.9 9.1 12.9 9.3 11.1 Chiapas 7 28.3 19.3 23.8 25.0 20.5 22.7 BAT 477 5.5 2.8 4.2 6.6 4.1 5.3 BAT 47 15.8 9.5 12.6 16.1 11.2 13.6 Montcalm 7.1 5.3 6.2 8.0 5.0 6.5 Vita 4 20.1 15.1 17.6 12.0 10.3 11.1 Vita 5 19.5 10.6 15.0 14.4 7.7 11.0 Vita 7 20.0 17.0 18.5 17.4 10.7 14.0 Mean 14.6 9.2 11 9 12.7 8.9 10 8 CV (%) 49.5 51.2 WW = well-watered treatment; DS = drought-stressed treatment; Means followed by the same letter are not significantly (P=0.01) different. ‘97 1986 WUE Experiment In 1986, a different set of genotypes was used (Table 2) in the water-use efficiency (WUE) experiment. Total biological yield production decreased significantly in all genotypes except breeding line B76001, Vita 3, and accession 5 (Katolika) (Table 31). Genotypic variability was observed for total biological yield production. 186032, 196033, N80068, and accession 6 had the highest mean biomass production. Under drought stress Malawian landraces 5 and 6 had the highest biomass production. Genotypic variability was also observed for water-use efficiency (WUE). Accession 6, B82008, 186033, and N80068 had the highest WUB. Mean WUE increased Significantly under drought over that under well- watered conditions. The significant genotype increases were observed only in accessions 5 and 6. The two Mexican lines also showed WUE over 5.0, although this was not significant statistically. The contention is that the Malawian landrace components probably had more rooting density that lead them to explore for moisture more efficiently than other genotypes, especially under drought conditions. But roots were not sampled in this study. Total biological yield was used in assessing relative drought resistance among genotypes using the various methods of numerical expression (Table 32). The yield differential and S identified accessions 5 and 6, B76001 and Vita 3 to be the most drought resistant genotypes. This contrasted with both the arithmetic and geometric means which identified 98 Table 31. Biological yield and water-use efficiency (WUE) of 9 dry bean and one cowpea genotypes (1986). Line/ Biological yield WUE Accession WW DS Mean WW DS Mean ------ g/plant ---- ------ g/kg water —---- 186030 41.7 21.143 31.4 b 3.27 3.52 3.49 0 186032 48.7 30.543 39.6 ab 4.06 5.08 4.57 ab B76001 40.7 28.7 34.7 b 3.39 4.79 4.09 be 186033 49.3 30.643 39.9 ab 4.12 5.09 4.60 ab N80068 49.8 28.544 39.1 ab 4.14 4.74 4.44 abc B82008 46.4 25.7tt 36.0 b 4.34 4.74 4.53 ab NB1017 44.8 27.033 35.9 b 3.78 4.47 4.12 bc Vita 3 39.1 28.1 33.6 b 3.26 4.68 3.91 bc 5 (Malawi) 35.9 32.6 34.2 b 2.99 5.20tt 4.10 bc 6 (Malawi) 52.3 38.5! 45.4 a 4.36 6.4243 5.39 a Mean 44.9 29.1 37.0 3.79 4.87 4.33 WW = well-watered treatment; DS 4 significant between treatments at P=0.05; ** significant between treatments at P=0.01; drought stressed treatment; Means followed by the same letter are not significantly (P=0.01) different. 99 Table 32. Biological yield differential, arithmetic mean, geometric mean, and drought susceptibility index (S) of 9 dry bean genotypes and one cowpea genotypes (1986). Line/ [wa-Yds] [Xmmidel [wadesll/2 S 2 Accession (1) (2) (3) (4) 186030 20.6 31.4 29.7 0.029 186032 18.2 39.6 38.5 0.022 B76001 12.0 34.7 34.2 0.017 186033 18.7 39.9 38.8 0.022 N80068 21.3 39.1 37.7 0.025 B82008 20.7 36.0 34.5 0.026 N81017 17.8 35.9 34.8 0.024 Vita 3 11.0 33.6 33.1 0.017 5 (Malawi) 3.3 34.2 34.2 0.005 6 (Malawi) 13.8 45.4 44.9 0.016 wa = potential; biological yield under well-watered conditions; Yds = biological yield under drought stress; 1 = biological yield differential; 2 = arithmetic mean biological yield; 3 = geometric mean biological yield; 4 = drought susceptibility index. 100 accession 6, and lines 186033, 186032, and N80068 as the most drought tolerant genotypes. The difference here was caused because the latter two methods incorporate yield potential and drought resistance whereas the former two methods remove the yield potential from the drought resistance attributes. Genotypic variability was observed for leaf moisture retention capacity (LMRC) (Table 33). White tepary, 186033, N81017, accession 5 (Katolika), and accession 6 had the highest LMRC. Line N81017 was shown to be the most drought tolerant line in a study conducted by my colleague (Samper, 1984), 186033, Vita 3, and accession 5 (Katolika) had significantly lower leaf transpiration rates (Table 34). There was a significant reduction in leaf area expansion in all genotypes under drought conditions (Table 35). Vita 3 had the fastest mean leaf area expansion rate. 101 Table 33. Specific leaf weight (SLW) and leaf moisture retention capacity (LMRC) of 9 dry bean and one cowpea genotypes at 36 DAP (1986). Line/ SLW LMRC Accession g/cm3 -- X -- 186030 2.46 43.8 a 186032 4.11 34.5 b B76001 3.65 39.3 b 186033 3.01 40.4 a N82008 3.67 34.8 b B82008 3.45 37.2 b N81017 3.49 43.8 a Vita 3 3.97 33.1 b 5 (Malawi) 3.10 44.4 a 6 (Malawi) 3.63 40.9 a Mean 3.51 39.9 CV (%) 29.9 17.8 Means followed by the same letter are not significantly (P=0.01) different. 102 Table 34. Leaf diffusive resistance, leaf transpiration, and leaf temperature of 9 dry bean and one cowpea genotypes at 47 DAP (1986). Line/ Leaf diffusive Leaf Leaf resistance transpiration temperature Acc. # WW DS Mean WW DS Mean WW DS Mean ---- s/cm ---- ug HzO/sz/S ----- C ----- 186030 2.21 4.16 3.18 9.00 7.72 8.36 26.6 28.543 27.5 186032 1.66 5.01 3.33 11.43 7.17 9.30 27.3 28.4: 27.9 B76001 2.07 3.72 2.90 9.95 8.4513 9.20 27.0 28.3 27.6 186033 2.86 6.44 4.65 8.99 4.19 6.59 27.9 28.8 28.3 N80068 1.30 4.54 3.02 12.19 6.36tt 9.28 27.9 29.83: 28.9 B82008 1.61 2.67 2.14 10.16 8.30 9.33 27.0 27.9 27.5 N81017 _ 1.48 6.79 4.13 12.58 5.3631 8.97 27.6 28.0 27.8 Vita 3 2.03 8.48 5.25 8.66 3.57 6.12 28.0 28.5 28.2 5 (Malawi) 3.36 5.93 4.64 8.71 4.45 6.58 28.0 28.0 28.0 6 (Malawi) 1.34 7.14 4.24 13.26 6.05:: 9.65 27.8 28.68 28.2 Mean 2.01 5.49 3.75 10.49 6.18 8.33 27.5 28.5 28.0 CV (%) 49.6 40.4 6.90 WW = well-watered treatment; DS drought-stressed treatment; * significant between treatments at P=0.05; 4* significant between treatments at P=0.01; 103 Table 35. Leaf area expansion of 9 dry bean and one cowpea genotypes (1986). Line/ Leaf area expansion Accession WW DS Mean - -------------- cm'lday -------------- ~-- 186030 5.08 1.63tt 3.35 186032 4.43 2.1543 3.29 b B76001 4.82 1.833! 3.32 b 186033 4.85 2.83:: 3.84 b N80068 5.00 1.7531 3.38 b B82008 5.42 2.4618 3.94 b N81017 4.91 1.933! 3.42 b Vita 3 7.18 3.50*¥ 5.34 a 5 (Malawi) 5.01 2.8643 3.93 b 6 (Malawi) 5.48 1.4814 3.48 b Mean 5.22 2.25 3.47 CV (X) 26.4 WW = well-watered treatment; DS drought-stressed treatment; *1 significant between treatments at P=0.01; Means followed by the same letter are not significantly (P=0.05) different. DISCUSSION It is important to understand causes of drought at any one location before a researcher proceeds into developing genotypes that are supposed to be drought tolerant at that particular location. Drought is caused by a number of factors that govern the supply of moisture (precipitation and initial soil moisture at planting) and the demand for moisture by the atmosphere (an interplay of air temperature, relative humidity, wind, and solar radiation) and the soil at the site (matric forces). When the cause of drought is known one can then proceed to strategically address the problem. For example, Boyer et al. (1980) have observed, through wilting symptoms, that soybeans growing on adequate supplies of moisture in central Illinois experience drought. They determined that this was due to failure by the root system to replenish, at the necessary rate, the moisture lost to the atmosphere. In such a case, one would perhaps employ methods dealing with the morphology of root systems to explore adequate amounts of moisture and also, perhaps, reduce hydraulic resistance to water conduction through the root system. Work done in this study aimed at identifying genotypes of dry beans that are tolerant to drought conditions, or characteristics that make them respond as such, under the Michigan environment. 104 105 This work was conducted during some of the most moisture stressed years in Michigan. As was shown in Figure 1, the month of June in 1984 had virtually no precipitation. This was during the vegetative phase of bean plant growth. Although there was precipitation during reproductive growth its distribution was poor; periods of three to four weeks of no precipitation occurred. One intercropping experiment was conducted during this particular year. Results indicated that beans growing in intercropping systems with maize suffered moderate levels of moisture stress (soil moisture about 50 kPa). Andow and Berkowitz (1985) indicated similar results using field bean and weed associations. Berkowitz (1985), however, reported similar moisture depletion in corn and sunflower monoculture systems as that when the two were intercropped. In their case the two species were of similar height and populations were grown in alternate rows and so inter-species competition was kept to a minimum. In the present study the planting patterns and populations used were those that are used by or recommended for the Malawian farmers. No special adjustments in plant populations were made. Berkowitz (1985) concluded that intercrops use the same amounts of water as monocultural stands. In this study intercrops used significantly more soil moisture than monocultural bean stands. Therefore, beans grown in intercropping seem more likely to encounter drought than those in monoculture systems. These landrace lines from Malawi have been naturally selected under these mixed 106 cropping systems for more than 300 years and some may have evolved in such a way as to contend with such competition for moisture under intercropping and have become more drought tolerant concomitantly. Two significant results here are: a, that one could identify drought tolerant bean genotypes though their performance under intercropping systems; and b, that when developing bean cultivars for Malawian farmers it is important to incorporate drought tolerance to assure high productivity under prevalent cropping systems. In the past two to three decades crop scientists have condemned intercropping as a primitive method of crop production. Farmers do not want to change because this system maximizes their labor and crop yield stability as well as minimizes risk. This study identifies drought as one of the most important factors that farmers have to deal with and against which natural selection probably has been operating over the last several decades. One of the mechanisms characteristic of some Malawian landraces exposed to drought under intercropping was stomatal resistance. The more stress tolerant landraces [accessions 4 (Nyauzembe), 7, and 17] significantly increased their leaf diffusive resistances under intercropping as compared with monoculture. The possible ways in which such stomatal behavior can enhance drought resistance in plants are: a, conservation of a limited supply of water so that some remains for later stages of the plant’s development; b, prevention of potentially damaging 10.7 drought; and c, maximization of total assimilation with the available water (Jones, 1980). However, although genotypic variation in stomatal conductance has been shown in soybean (Carlson et al., 1979), most later studies (Clarke and McCaig, 1982; Tesha, 1984) including these conducted here, have failed to detect genotypic variation for this characteristics. The approach, under conditions of unpredictable droughts, should be to breed for a genotype with low diffusive resistance (high stomatal conductance) under good conditions (to maximize photosynthesis) but under drought conditions such a genotype should efficiently close its stomata. Upon rehydration such stomata should rapidly reopen. The overall result of such an approach is an increase in water-use efficiency (WUE). Xerophytic plants have very thick leaves as one of the characteristics they have acquired in adaptation to high moisture deficit environments. Mesophytic plants seem to show similar increases in specific leaf weight (SLW) when grown under drought conditions (Turner, 1979). Turk and Hall (1980a) indicated increases in SLW of cowpea with increasing levels of drought. A similar response was observed with the bean genotypes grown under the rain-out shelter. However, beans grown in intercropping had significantly lower SLW as compared to those under monoculture due to expanded area and reduced weight per unit area under partially shaded conditions. The response observed under intercropping was only invoked due to the 108 partial shading and not as a result drought. This raises a problem in using SLW, under intercropping and drought, as a criterion for selection of drought resistant genotypes. The way SLW is calculated (equation 14) one does not necessarily know whether it is due to an increase in leaf thickness or accumulation of nutrients and starch. With the three different soil types, SLW showed no adaptation to soil moisture content accruing to differences in water-holding capacity of such soils. The problem with SLW is that if a plant increased SLW of its leaves when it is exposed to a drought during one of its growth stages the result would be an increase in SLW and a concomitant limitation of leaf area expansion. On rehydration SLW does not change as a result of which maximum leaf area is never attained. 1n cowpeas, seed yield was correlated with LAI at the end of flowering (Turk and Hall, 1980a). A more flexible characteristic than SLW is leaf moisture retention capacity (LMRC). This was significantly different among bean genotypes but no evidence of a change due to cropping systems was detected. Some of the genotypes shown to be drought tolerant in this study using other criteria (accessions 2 and 17) had significantly higher amounts of leaf moisture retained, whereas stress- susceptible ones, like accession 14, retained significantly less moisture. Beans grown under more drought (light soil type) retained significantly more moisture than those grown on less droughty (heavy soil types). These results show 109 that LMRC is a more reliable indicator of a genotype's capacity to adapt to increasing soil moisture deficits than SLW. LMRC does not seem to be as conservative a trait as SLW because it is expected not to respond permanently. There was no significant relationship between LMRC and SLW and between SLW and total biological yield production. There was also no significant relationship between LMRC and total biological yield production. This casts some doubt as the whether LMRC could be used as an alternative to the porometer although it is not as convenient. Genotypic variability was observed with LMRC but it was not detected with use of the porometer. There was significant genotypic variability in both biological and economic yield production under both monoculture and intercropping systems. Mean economic yield in 1984 was significantly reduced by 28.5% under intercropping from yields obtained under monoculture Comparable yield reduction in beans and soybeans have been reported (Mmbaga, 1980). See yield reductions in such cropping systems were explained as due mainly to partial shading. Interestingly, some genotypes did not yield differently under intercropping as compared with monoculture. This indicates that some dry bean genotypes are adapted to intercropping systems whereas others are not. Pooled over methods or criteria of quantifying drought tolerance in 1984, the data indicate that accessions 2, 4, 5, 6, and 17 were better adapted than others to 110 intercropping. Accessions 1 and 14 were the least adapted. Economic yield was significantly reduced (24.3%) under intercropping from the yields obtained under monoculture in 1985. Accessions 4, 5, 7, and 17, and Bayo Madero seemed better adapted to intercropping than the remaining entries in 1985. Under controlled conditions biological yield production declined significantly (42.6%) under drought conditions from biomass productivity observed under well-watered conditions. Turk and Hall (1980a), Bonanno and Mack (1983) and Jones et al. (1980) also reported reductions in biological yield in cowpea, snap bean, and perennial ryegrass (Lolium perenne L), respectively. In 1985, there was significant genotypic variability in biological yield production. Chiapas 7 had the highest mean biomass whereas accession 6 and BAT 477 had the least mean biomass production. Only accessions 5 and 17, and BAT 47 and Vita 5 significantly reduced their biomass productivity under drought. However, all genotypes produced significantly less (51.8%) economic yield under drought conditions and no genotypic variability was observed. BAT 47 and accession 5 (Katolika) were the most adapted genotypes to drought conditions in 1985. During 1986 there was again a significant decline in biological yield productivity (35.2%) under drought conditions from the potential productivity observed under well-watered conditions, except breeding line B76001, Vita 3, and accession 5, which showed no significant decline. Genotypic 111 variability existed in mean biological yield productivity. Accessions 5 and 6, lines B76001, Vita 3, 186033, 186032, and N80068 were more drought tolerant than other genotypes. Better adapted genotypes under intercropping did not reduce their number of pods/plant when grown under such a system as compared to their potential under monoculture. The less well adapted genotypes suffered significant reductions in this yield component. Mmbaga (1980) indicated similar results. It seems that the number of pods/plant largely controlled number of seeds plants produced. Under controlled conditions in 1985, there was a significant reduction in the number of pods/plant under drought from the number observed under well-watered conditions. In cowpea Turk et al. (1980) indicated that drought reduced seed yield due mainly to reductions in pods per square meter. There was also significant genotypic variation in the mean number of pods produced. BAT 47, a genotype that showed higher relative drought tolerance, produced the most pods/plant. Frederick et al. (1985) indicated that older soybean varieties were drought tolerant in terms of yield due to a smaller relative decrease in pod number/branch. However, using a different set of genotypes, Boyer et al. (1980) indicated consider yield improvement in newer cultivars when their yields were compared with those of older cultivars under a high evaporative demand environment. Rooting densities of two cultivars showing extremes of water deficits indicate that the cultivar with 112 the least water deficit had a greater root density than the cultivar with the largest deficit. The implication is that older cultivars have larger root densities than newer ones. If this is the case with large—seeded legumes crops in general, one would expect the Malawian landraces to have large rooting densities. Such a characteristic would be important especially under unpredicted droughts as compared with terminal droughts. The major portion of the environment in Malawi is such that crops have to rely on current rainfall and unpredicted droughts often occur. Large rooting densities would be most appropriate as opposed to deep rooting systems. This might explain why Malawian landraces had higher WUE in 1986 and lower WUE in 1985. In 1985 the large initial doses of water in WUE experiments simulated terminal drought whereas in 1986 unpredicted droughts were simulated. Under such latter conditions deep rooting genotypes would be at a disadvantage because they would be wasting biomass in developing deep roots when moisture is not available lower in the soil profile. This is probably why the tepary bean had lower WUE than other genotypes. Water-use efficiency (WUE) calculated on the basis of total biological yield production during both the vegetative and reproductive periods (equation 7) increased, but not significantly, under drought from that under well-watered conditions in 1985. In 1986 the increase was significant. This contradicts reductions in WUE of leaves indicated in 115 sorghum and soybeans (Rawson et al., 1978) with increasing soil water deficits. Turk and Hall (1980b) indicate that WUE (calculated on the basis of seed yield and evapotraspiration) increases when drought is imposed during the vegetative period and it decreases when drought is imposed during the reproductive period. It would seem that some Malawian landraces partially close their stomata to conserve water and maximize total assimilation with the available water, i.e. tend to have high WUE. This was indicated by an increase in leaf diffusive resistance and consequent decrease in leaf transpiration under stress. Drought tolerant genotypes were identified by some characteristics. They showed increased leaf diffusive resistance, reduction in leaf transpiration rates, and reduction in leaf temperature under stress conditions. LMRC and SLW produced mixed results. The drought tolerant genotypes had lower SLW in the 1984 intercropping experiment. Genotypic variability in SLW was not observed in the 1986 WUE experiment. Some of the drought tolerant genotypes, on the other hand, had higher LMRC than other drought tolerant and susceptible genotypes in both of the above experiments. However, in the 1985 intercropping experiment, the drought tolerant genotypes were among those with higher LMRC. Montcalm, a cultivar adapted to the area, produced the highest economic yield and also had higher LMRC than other entries. The Malawian landrace components 5 and 6, which increased WUE under drought conditions, were among 114 genotypes with significantly higher LMRC. In the 1984 intercropping experiment most of the drought tolerant genotypes also escape drought by either maturing early or by a coincidence of a relief from drought (precipitation) and either pod development or seed-filling (Table 37 in the Appendix). SUMMARY AND CONCLUSIONS Intercropping maize and beans resulted in significantly lower soil moisture indicating that the bean would suffer from drought stress earlier under such a cropping system than when beans are grown in monoculture systems. Since most of the beans are grown in intercropping in Malawi, such droughts comprise most of the environmental limitations to crop productivity. Genotypic variability was indicated in tolerance to stress caused by intercropping and/or drought in the selected genotypes of Phaseolus and Vigna species. This is probably a major explanation of why Malawian farmers use bean mixtures. They want to maximize seed yields during the good years by planting landraces of higher yielding potential while at the same time aiming at minimizing yield looses during drought years by including drought tolerant genotypes (a double-barrel strategy). Under simulated unpredictable droughts Malawian landrace lines had significantly higher water-use efficiency. However, under simulated terminal droughts the Malawian were comparable to other genotypes in water-use efficiency. 115 116 Leaf moisture retention capacity (LMRC) was significantly different among genotypes, although it was not significantly related to economic yield. Drought tolerance and yield potential are two different characteristics that are possibly differently inherited. However, the combination of the two should produce high yielding drought tolerant genotypes (within limits). There was no evidence for either LMRC of SLW being good indicators of drought tolerance. Intercropping maize and beans increased leaf diffusive resistance in beans and as a result decreased leaf transpiration. This indicates that some Malawian landraces overcome intercropping and consequent drought stresses by conserving moisture and probably utilizing it more efficiently. Both biological and economic yields were significantly reduced in susceptible genotypes to intercropping stress. There was no evidence for such reductions in the adapted landraces and cultivars. Under simulated terminal droughts (WUE, 1985) virtually all genotypes had significant reductions in biological yield. However, under simulated unpredictable drought (WUE, 1986) there was no evidence for such reductions in the drought tolerant genotypes. L I TERATURE C I TED LITERATURE CITED Adams, M.W.1982. Genetic, agronomic, and social-cultural analysis of diversity among bean landraces in Malawi. Bean/Cowpea CRSP Title X11 Project. Michigan State University, East Lansing. Adams, M.W., I.V. Wiersma, and J. Salazar. 1978. Differences in starch accumulation among dry bean cultivars. Crop Sci. 18:155-157. Andow, D.A. and A.R. Berkowitz. 1985. Competition for resources between field beans and weeds. Agron. Abstracts. 1985. ASA, CSSA, SSSA. Madison. Berkowitz, A.R. 1985. Water use in mixtures and monocultures for corn and sunflower. Agron Abstracts. 1985. ASA, CSSA, SSSA. Madison. Bewley, J.D. 1979. Physiological aspects of desiccation tolerance. Ann. Rev. Plant Physiol. 30:195-238. Bidinger, F., R.B. Musgrave, and R.A. Fischer. 1977. Contribution of stored pre-anthesis assimilate to grain yield in wheat and barley. Nature 270:431-433. Blum, A. 1982. Evidence for genetic variability in drought resistance and its implications for plant breeding. p.53-68. In IRRI. Drought resistance in crops, with emphasis on rice. Los Banos, Laguna, The Phillipines. Blum, A. and A. Ebercon. 1976. Genotypic responses in sorghum to drought stress. 111. Free proline accumulation and drought resistance. Crop Sci. 16:428- 431. Blum, A. and A. Ebercon. 1981. Desiccation and heat tolerance tests in wheat. Crop Sci. 21:43-47. Blum, A., H. Poiarkova, G. Golan, and J. Mayer. 1983. Chemical desiccation of wheat plants as a simulator of post-anthesis stress. I. Effects on translocation and kernel growth. Field Crop Res. 6:51-58. Bogges, S.F., D.E. Koeppe, and C.R. Stewart. 1978. Oxidation of proline by plant mitochondria. Plant Physiol. 62:22- 25. Bonanno, A.R. and H.J. Mack. 1983. Yield components and pod quality of snap beans grown under differential irrigation. J. Amer. Soc. Hort. Sci. 108:832-836. 117 118 Boyer, J.S. 1970a. Leaf enlargement and metabolic rates in corn, soybean, and sunflower at various leaf water potentials. Plant Physiol. 46:233-235. Boyer, J.S. 1970b. Differing sensitivity of photosynthesis to low leaf water potentials in corn and soybean. Plant Physiol. 46:236-239. Boyer, J.S. 1971a. Resistances to water transport in soybean, bean, and sunflower. Crop Sci. 11:403-407. Boyer, J.S. 1971b. Nonstomatal inhibition of photosynthesis in sunflower at low leaf water potentials and high light intensities. Plant Physiol. 48:532-536. Boyer, J.S., R.R. Johnson, and S.G. Saupe. 1980. Afternoon water deficits and grain yields in old and new soybean cultivars. Agron. J. 72:981-986. Bremner, P.M. 1972. Accumulation of dry matter and nitrogen in grains of different positions of the wheat ear as influenced by shading and defoliation. Aust. J. Biol. Sci. 25:657-668. Burrows, F.J. and P.L. Milthorpe. 1976. Stomatal conductance in the control of gas exchange. In T.T. Kozlowski. (ed.) Water deficits in plant growth. Academic Press, New York. Campbell, D.E. and D.J. Hume. 1970. Evaluation of a rapid technique for measuring soluble solids in corn stalks. Crop Sci. 10:625-626. Carlson, R.E., N.N. Momen, 0. Arjmand, and R.H. Shaw. 1979. Leaf conductance and leaf-water potential relationships for two soybean cultivars grown under controlled irrigation. Agron. J. 71:321-325. Claassen, M.M. and R.H. Shaw. 1970. Water deficit effects on corn. 11. Grain components. Agron. J. 62:652-655. Clarke, J.M. and T.N. McCaig. 1982. Evaluation of techniques for screening for drought resistance in wheat. Crop Sci. 22:503-506. Cutler, J.M., K.W. Shahan, and P.L. Steponkus. 1980. Influence of water deficits and osmotic adjustment on leaf elongation in rice. Crop Sci. 20:314-318. Daday, H., F.E. Biner, A. Grassia, and J.W. Peak. 1973. The effect of environment on heritability and predicted selection response in Medicago sativa. Heredity 31:293- 308. 119 Daynard, T.B., J.W. Tanner, and D.J. Hume. 1969. The contribution of stalk soluble carbohydrates to grain yield in corn. Crop Sci. 9:831-834. Dejong, T.M. and D.A. Phillips. 1982. Water stress effects on nitrogen assimilation and growth of Trifolium subterraneum L. using dinitrogen or ammonium nitrate. Plant Physiol. 69:416-420. Eberhart, S.A. and W.A. Russell. 1966. Stability parameters for comparing varieties. Crop Sci. 6:36-40. Epstein, E. 1972. Mineral nutrition of plants: Principles and perspectives. John Wiley and Sons, New York. Fehr, W.R. and C.E. Caveness. 1977. Special report 80. Co- operative Extension Service. Iowa State University, Ames. Ferreira, L.G.R., J.D. de Souza, and J.T. Prisco. 1979. Effects of water deficit on proline accumulation and growth of two cotton genotypes of different drought resistance. 2. Pflanzenphysiol. 93:189-199. Finlay, K.W. and G.N. Wilkinson. 1963. The analysis of adaptation in plant breeding programme. Aust. J. Agric. Res. 14:742-754. Finn, G.A. and W.A. Brun. 1980. Water stress effects on carbon dioxide assimilation, photosynthate partitioning, stomatal resistance, and nodule activity in soybean. Crop Sci. 20:431-434. Fischer, R.A. and R. Maurer. 1978. Drought resistance in spring wheat cultivars. 1. Grain yield responses. Aust. J. Agric. Res. 29:897-912. Fischer, R.A. and N.C. Turner. 1978. Plant productivity in the arid and semi-arid zones. Ann. Rev. Plant Physiol. 29:277-317. Fitter, A.H. and R.H.M. Hay. 1981. Environmental physiology of plants. Academic Press, New York. Frederick, J.R., J.T. Woolley, J.D. Hesketh, and D.B. Peters, 1985. Agronomic responses of old and new soybean varieties to soil moisture deficit. Agron. Abstracts 1985: 96. ASA, CSSA, SSSA. Madison. Gaff, D.E. 1980. Protoplasmic tolerance of extreme water stress. p.207-230. In N.C. Turner and P.J. Kramer (eds.) Adaptation of plants to water and high temperature stress. Wiley interscience, New York. 120 Gallagher, J.N., P.V. Biscoe, and B. Hunter. 1976. Effects of drought on grain growth. Nature 264:541-542. Gardner, F.P., R.B. Pearce, and R.L. Mitchell. 1985. Physiology of crop plants. Iowa State University Press, Ames. Gould, G.W. and J.C. Measures. 1977. Water relations in single cells. Phils. Trans. R., Soc. London, Ser. B, 278:151-166. Greacen, E.L. and J.S. Oh. 1972. Physics of root growth. Nature (London), New Biol. 235:24-25. Green, P.B., R.O. Erickson, and J. Buggy. 1971. Metabolic and physical control of elongation rate: in vivo studies in Nitella. Plant Physiol. 47:423-430. Hall, A.E. and D.A. Grantz. 1981. Drought resistance of cowpea improved by selecting for early appearance of mature pods. Crop Sci. 21:461-464. Hanson, A.D. 1980. Interpreting the metabolic responses of plants to water stress. HortScience 15:623-629. Hanson, A.d. and C.E. Nelsen. 1980. Water: adaptation of crops to drought-prone environments. In P.S. Carlson (ed.) The Biology of Crop Productivity. Academic Press, New York. Harlan, J.R. 1975. Geographic patterns of variation in some cultivated plants. J. of Heredity 66:184-191. Harley, J.L. and 8.3. Smith. 1983. Mycorrhizal symbiosis. Academic Press, New York. Henckel, P.A. 1964. Physiology of plants under drought. Ann. Rev. Plant Physiol. 15:363-386. Hsiao, T.C. 1973. Plant responses to water stress. Ann. Rev. Plant Physiol. 24:519-570. Hsiao, T.C., E. Acevedo, E. Fereres, and D.W. Henderson. 1976. Water stress, growth, and osmotic adjustment. Phils. Trans. R. Soc. London, Ser. B. 273:479-500. Hsiao, T.C. and R.A. Fischer. 1975. Mass flow porometers. In E.T. Kanemasu. (ed.) Measurement of stomatal aperture and diffusive resistance. p.5-11. Bull. 809, College of Agric. Res. Center, Washington State University, Pullman. 121 Hume, D.J. and D.K. Campbell. 1972. Accumulation and translocation of soluble solids in corn stalks. Can. J. Plant Sci. 52:363-368. Ingram, H.T. and P.H. Moore. 1985. Physiological responses of sugarcane to water deficit. Agron. Abstracts. 1985. ASA, CSSA, SSSA. Madison. Jefferies, R.L. 1981. Osmotic adjustment and the response of halophytic plants to salinity. BioScience 31:42-46. Johnson, R.C., D.W. Mornhinweg, H.T. Nguyen, D.M. Ferris, and J.J. Heitholt. 1965. Stomatal limitation to photosynthesis of wheat under water deficits. Agron. Abstracts. 1985. ASA, CSSA, SSSA. Madison. Jones, M.B., E.L. Leafe, and W.Stiles. 1980. Water stress in field-grown perennial ryegrass. 1. Its effect on growth, canopy photosynthesis, and transpiration. Ann. Applo B1010 96:87-1010 Jordan, W.R. and F.R. Miller. 1980. Genetic variability in sorghum root systems: implications for drought tolerance. p.383-399. In N.C. Turner and P.J. Kramer (eds.) Adaptation of plants to water and high temperature stress. Wiley Interscience, New York. Kanemasu, E.T. 1975. Measurement of stomatal aperture and diffusive resistance. Bull. 809, College of Agric. Res. Center, Washington State University, Pullman. Kaplan, L. 1965. Archeology and domestication in American Phaseolus beans. Economic Bot. 19.358-368. Keck, H.L. and J.S. Boyer. 1974. Chloroplast response to low leaf water potentials. Plant Physiol. 53:474-479. Laing, D.R. and R.A. Fischer. 1977. Adaptation of semidwarf wheat cultivars to rainfed conditions. Euphytica 26:129-139. Levitt, J. 1982. Responses of plants to environmental stresses. Academic Press, New York. Levy, D. 1983a. Varietal differences in the responses of potatoes to repeated short periods of water stress in hot climates. I. Turgor maintenance and stomatal behavior. Potato Research 26:303-313. Levy, D. 1983b. Varietal differences in the responses of potatoes to repeated short periods of water stress in hot climates. II. Tuber yield and dry matter accumulation and other tuber properties. Potato Research 26:315-321. 122 Lieth, H. 1976. The use of correlation models to predict primary productivity from precipitation or evapotranspiration. p.392-407. In O.L. Lange, L. Kappen, and D.E. Schulze. (eds.) Water and Plant Life: Problems and modern approaches. Springer-Verlag, New York. Loomis, W.R. 1935. The translocation of carbohydrates in maize. Iowa State 8. Sci. 9:509-520. Malawi Government. 1982. Climatological tables for Malawi. Meteorological Services, Chileka, Malawi. McCaig, T.N. and J.M. Clarke. 1982. Seasonal changes in nonstructural carbohydrate levels of wheat and oats grown in a semi-arid environment. Crop Sci. 22:963-970. Mederski, H.J. and D.L. Jeffers. 1973. Yield responses of soybean varieties grown at two soil moisture levels. Miller, J.C., S. Rajapakse, and R.H. Garber. 1986. Vesicular—arbuscular mycorrhizae in vegetable crops. HortScience 21:974-984. Morgan, J.M. 1977. Differences in osmoregulation between wheat genotypes. Nature 270:234-235. Morgan, J.M. 1980. Osmotic adjustment in the spikelets and leaves of wheat. J. Exp. Bot. 31:655-665. Mmbaga, M.B.T. 1980. Effect of associated culture on grain yield, percent protein, and percent oil of maize, dry beans, and soya beans. M.S. Thesis, Michigan State University, East Lansing. Mughogho, L.K., O.T. Edje, and U.W.U. Ayonoadu. 1972. Bean Improvement Program in Malawi. Ann. Report of Bean Improvement Co-operative. 15:69-71. Muhammad, A.F.H. 1983. The effects of temperature and daylength on days to flower and maturity in dry beans (Phaseolus vulgaris L.). Ph.D. Dissertation, Cornell University, Ithaca. Nelsen, C.E. and G.R. Safir. 1982. Increased drought tolerance of mycorrhizal onion plants caused by improved phosphorus nutrition. Planta 154:407-413. Nir, I. and A. Poljakoff-Mayber. 1967. Effect of water stress on the photochemical activity of chloroplats. Nature 213:418-419. 123 Nor, K.M. and F.B. Cady. 1979. Methodology for identifying wide adaptability in crops. Agron. J. 71:556-559. O’Toole, J.C. and T.B. Moya. 1978. Genotypic variation in maintenance of leaf water potential in rice. Crop Sci. 18:873-876. Pankhurst, C. and J. Sprent. 1975a. Surface features of root nodules. Protoplasma 85:85-98. Pankhurst, C. and J. Sprent. 1975b. Effects of water stress on respiratory and nitrogen-fixing activity of soybean root nodules. J. Exp. Bot. 26:287-304. Passioura, J.B. 1972. The effect of root geometry on the yield of wheat growing on stored water. Aust. J. Agric. Res. 23:745-752. Passioura, J.B. 1976. Physiology of grain yield in wheat growing on stored water. Aust. J. Plant Physiol. 3:559- 565. Passioura, J.B. 1982. The role of root system characteristics in the drought resistance of crop plants. p.71-82. In IIRI. 1982. Drought resistance in \ crops, with emphasis on rice. Los Banos, Laguna, The Phillipines. Purseglove, J.W. 1968. Tropical Crops: Dicotyledonous. Longmans, London. Quizenberry, J.E. 1982. Breeding for drought resistance and plant water use efficiency. p.193-212. In M.N. Christiansen and C.E. Lewis. (eds.) Breeding plants for less favorable environments. John Wiley and Sons, New York. Rawson, H.M., N.C. Turner, and J.E. Begg. 1978. Agronomic and physiological responses of soybean and sorghum crops to water deficits. 1V. Photosynthesis, transpiration, and water-use efficiency of leaves. Aust. J. Plant Physiol. 5:195-209. Ritchie, G.A. and T.M. Hinckley. 1975. The pressure chamber as an instrument for ecological research. Adv. Ecol. Safir, G.R., J.S. Boyer, and J.W. Gerdemann. 1971. Mycorrhizal enhancement of water transport in soybean. Science 172:581-583. Safir, G.R., J.S. Boyer, and J.W. Gerdemann. 1972. Nutrient status and mycorrhizal enhancement of water transport in soybean. Plant Physiol. 49:700-703. 124 Samper, C. 1984. Effects of water stress imposed at mid-pod filling upon yield and dry matter partitioning in dry beans (Phaseolus vulgaris L.) M.S. Thesis, Michigan State University, East Lansing. \ Scholander, P.F.; H.T. Hammel, E.D. Bradstreet, and E.A. Hemingsen. 1965. Sap pressure in vascular plants. Science 148:339-346. Sebasigari, K. 1981. Patterns of partitioning and remobilization of nonstructural carbohydrates in common bean and other selected grain legumes. M.S. Thesis, Michigan State University, East Lansing. Shaw, R.H. and D.R. Laing. 1966. Moisture and plant response. p.73-94. In W.H. Pierre, D. Kirkham, J. Pesek, and R.H. Shaw. (eds.) Plant environment and efficient water use. ASA, CSSA. Madison. Sinclair, T.R., C.B. Tanner, and J.M. Bennett. 1984. Water- use efficiency in crop production. BioScience 34:36-40. Slatyer, R.O. 1967. Plant-water relationships. Academic Press, London. Slavik, B. 1974. Methods of studying plant water relations. Chapman and Hall, London. Singh, T.N., L.G. Paleg, and D. Aspinall, 1973. Stress metabolism. III. Variations in response to water deficit in the barley plant. Aust. J. Biol. Sci. 26:65- 76. Sprent, J. 1976. Nitrogen fixation by legumes subjected to water and light stresses. p.406-420. In P.S. Nutman. (ed.) Symbiotic nitrogen fixation in plants. Cambridge University Press. Stewart, C.R. and A.D. Hanson. 1980. Proline accumulation as a metabolic response to water stress. p.173-190. In N. C. Turner and P.J. Kramer. (eds.) Adaptation of plants to water and high temperature stress. Wiley Interscience, New York. Sullivan, C.Y. 1983. Genetic variability in physiological mechanisms of drought resistance. Iowa State J. Res. 57:423-439. Takami, 3., N.C. Turner, and H.M. Rawson. 1981. Leaf expansion of four sunflower (Helianthus annuus L.) cultivars in relation to water deficits. 1. Patterns during plant development. Plant Cell Environ. 4. 125 Taylor, H.M. 1983. Managing root systems to increase efficient water use. p.87-114. In H.M. Taylor et al. (eds.) Limitations to efficient water use in crop production. ASA. Madison. Tesha, A. 1984. Drought resistance in some bean varieties (Phaseolus vulgaris) grown in Tanzania. Phaseolus Bean Newsletter for Eastern Africa. 2:22-23. Thorne, G.N. 1966. Physiological aspects of grain yield in cereals. p.88-105. In P.L. Milthorpe and J.D. Ivins. (eds.) The growth of cereals and grasses. Butterworths, London. Troughton, J.H. 1969. Plant water status and carbon dioxide exchange of cotton leaves. Aust. J. Biol. Sci. 22:289- 302. Turc, L. 1955. Le Bilan d'eau des sols. Ann. Agron. 6:5. Turner, N.C. 1966. Grain production and water use of wheat as affected by plant density, defoliation, and water status. Ph.D. Dissertation, University of Adelaide, Australia. Turner, N.C. 1975. Concurrent comparisons of stomatal behavior, water status, and evaporation of maize in soil at high or low water potential. Plant Physiol. 55:932-936. Turner, N.C. 1979. Drought resistance and adaptation to water deficits in crop plants. p.343-372. In H. Mussell and R.C. Staples. (eds.) Stress physiology in crop plants. Wiley Interscience, New York. Turner, N.C. 1982. The role of shoot characteristics in drought resistance of crop plants. p.115-134. In IRRI. Drought resistance in crops, with emphasis on rice. Los Banos, Laguna, The Phillipines. Turner, N.C. and J.E. Begg. 1981. Plant water relations and adaptation to stress. p.97-277. In J. Monteith and C. Webb. (eds.) Soil water and nitrogen in mediterranean- type environments. Martinus Nijhoff and Dr. W. Junk Publishers, Boston. Turner, N.C. and M.M. Jones. 1980. Turgor maintenance by osmotic adjustment: a review and evaluation. p.87-103. In N.C. Turner and P.J. Kramer. (eds.) Adaptation of plants to water and high temperature stress. Wiley Interscience, New York. 126 Turner, N.C., J.E. Begg, H.M. Rawson, S.D. English, and A.B. Hearn. 1978. Agronomic and physiological responses of soybean and sorghum crops to water deficits. III. Components of leaf water potantial, leaf conductance, photosynthesis, and adaptation to water deficits. Aust. J. Plant Physiol. 5:179-194. Turk, K.J. and A.B. Hall. 1980a. Drought adaptation of cowpeas. III. Influence of drought on plant growth and relations with seed yield. Agron. J. 72:428-433. Turk, K.J. and A.B. Hall. 1980b. Drought adaptation of cowpeas. IV. Influence of drought on water use and relations with growth and seed yield. Agron. J. 72:434- 439. Turk, K.J., A.E. Hall, and C.W. Asbell. 1980. Drought adaptation of cowpea. Agron J. 72:413-420. Wittwer, S.H. 1975. Food production: technology and resource base. Science 188:579-584. APPENDIX 127 Table 36. Mean squares (ms) from analysis of variance for moisture tension of soil under monoculture and intercropped Malawian landrace day beans (1984). Source of variation df ms Pr > F Rep (R) 2 3755.799 - Genotypes (G) 17 37.038 0.030 R x G 34 17.427 - Treatment (T) 1 3312.642 0.000 G x T 17 15.505 0.138 R x G x T 36 10.114 - Stage (S) 2 21208.892 0.000 G x S 34 21.382 0.452 T x S 2 323.429 0.000 G x T x S 34 6.684 ns R x G x T x S 144 21.017 - 128 Table 37. Patterns of response by the landrace genotypes to grought in 1984. Entry Stage when drought was critical 1 early pod development 2 escaped, early maturity 3 end seed-filling 4 escaped, synchronized‘ 5 escaped, synchronized 6 escaped, early maturity 7 escaped, synchronized 8 escaped, synchronized 9 problem related to fertility 10 late pod development 12 early pod development 13 early pod development 14 early pod development, seed-filling 15 early pod development 16 early pod development 17 escaped, early maturity ‘ ’escaped, the drought outside the development synchronized’ means that the genotype escaped largely from the fact that drought occurred important reproductive stages, namely; pod and seed filling. "mmmwmm