mvensmr LIBRARIES N W W WW\\\\\\\\\\\\\\\ 3129 LIBRARY Michigan State University This is to certify that the thesis entitled CHARACTERISTICS RELATED TO YIELD OF DRY BEAN (PHASEOLUS VULGARIS L.) UNDER WATER STRESS CONDITIONS presented by Jorge Elizondo Barron has been accepted towards fulfillment of the requirements for M.S. . Plant Breeding & Genetics degree 1n Program in Department of Crop and Soil Science; /'>' c 0%;‘7W Major professor M. W. Adams [hue October 20, 1987 0-7539 MS U is an Affirmative Action/Equal Opportunity Institution i :22 PLACE N RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before ode due. DATE DUE DATE DUE DATE DUE MSU Is An Affirmetlve Action/Equal Opportunity Institution cumulus-m CHARACTERISTICS RELATED TO YIELD OF DRY BEAN (2E;§§QLQ§_!HL§ARI§_L,) UNDER WATER STRESS CONDITIONS BY Jorge llizondo Barron A THESIS Subnitted to Michigan State university in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Crop and Soil Sciences 1987 bosofib4 ACKNOWLEDGMENTS The author expresses his sincere appreciation to his major professor Dr. M. W. Adams for his guidance and support throughout every stage of the realization of this study. To the members of the Graduate Committee, Drs. J. D. Kelly, C. E. Cress, and I. E. Widders for their constructive criticisms of this manuscript. To my parents, my wife, my children, for their love and encouragement. Finally, I wish to acknowledge the financial support I received from the government of Mexico, who sponsored my study through I.N.I.F.A.P. Gratitude is also extended to the Department of Crop and Soil Sciences, who provided financial support in the form of an assistantship, during portion of my study at Michigan State University. ii TABLE OF CONTENTS Page LIST OF TABLES............' ....... . ......................... v LIST OF FIGURES. ...... ................. .............. ....vii INTRODUCTION. ..... . ........................................ 1 LITERATURE REVIEW ................ . ......................... 3 Drought definition............. ....................... 3 Plant response to water stress ....... ..... ............ 3 Morphological effects................... ............ 3 Physiological effects.............. ...... . .......... 5 a).-Leaf resistance and water potential ............ 5 b).-Photosynthesis..................... ............ 7 c).-Translocation and distribution of assimilates..9 Effects of water deficits on yield ................. 11 Mechanisms of drought resistance... ................ 13 a).-Drought escape........... ....... . ............ 13 b).-Drought avoidance.. ................. . ........ 14 c).-Drought tolerance.... .................... ....15 MATERIALS AND METHODS..... ....... ... ...................... 18 Indicators of water status and stomatal condition....23 Methods of analysis ...... .... ........................ 24 Principal Factor Analysis... ....................... 24 Stepwise Regression Analysis ....................... 27 RESULTS...... ..... . ................ . ...................... 29 Stomatal resistance and water potential .............. 29 Stepwise regression analysis. Water stress ........... 31 Stepwise regression analysis. Irrigated condition....32 Factor analysis. Water stress ........................ 34 Factor analysis. Irrigated condition ................. 36 Analysis of variance of the important traits ......... 39 Economic yield ..................................... 39 Biological yield ................................... 41 Days to beginning bloom ............................ 45 Days to maturity ................................... 46 Weight of stems .................................... 46 Weight of roots .................................... 48 iii Top/root ratio ........... ...... .................... 50 DISCUSSION......... ....................... ... ....... ......54 Interpretation of the results.......... .......... ....54 Stomatal resistance and water potential.. ......... ...54 Stepwise multiple regression analysis................57 Water stress. Important variables....... ........... 57 Irrigated condition. Important variables ........... 57 Factor analysis........ ..... .... ..... ..... ...... .....58 Water stress. Important variables....... ........ ...58 Irrigated condition. Important variables.. ......... 62 Final remarks............................. ........... 64 CONCLUSIONS.................... ..... ... ....... . ..... . ..... 66 LITERATURE CITED............ ..... ........ ............. ....68 APPENDIXOOOOOOOO 0000000000 00...... 000000000000 ... ......... 73 iv LIST OF TABLES Table Page 1. 10. ll. Names and descriptions of the 11 dry bean genotypes grown under water stress and irrigation. "outcalm' 1986.........OOOOOOOOOOOOO0.0.0.000........019 Measured and calculated characteristics and methods Of measurementOOOO......OOOOOOOOOO...OOOOOOOOOOOO 00000 22 -1 Stomatal resistance means (sec/cm ) of six dry bean genotypes under water stress and irrigation. Montcalm, 1986............................ ..... .. ..... 3O Xylem water potential means (-bars) of six dry bean genotypes under water stress and irrigation. Montcalm, 1986........................ ...... ..........3O Stepwise regression analysis of data under water stress with grain yield as dependent variable, when primary yield components were excluded from the set of independent variables......................... ..... 33 Stepwise regression analysis of data under irrigation with grain yield as dependent variable, when primary yield components were excluded from the set of independent variables ...................... 33 Factor analysis results: water stress condition ....... 35 Factor analysis results: irrigated condition .......... 38 Analysis of variance for grain yield, of 11 dry bean genotypes, under water stress and irrigation. Montcalm, 1986 ........................................ 4O 2 Economic yield (g/m ) of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ..... 4O 2 Biological yield (g/m ) at beginning bloom for 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 43 2 Biological yield (g/m ) at physiological maturity of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 43 Total dry matter production and grain mean yield 2 (g/m ) of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ........................ 44 Days to anthesis of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ........... 44 Days to maturity of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ........... 47 2 Weight of stems (g/m ) at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986..... ....................... . ........... 47 2 Weight of stems (g/m ) at physiological maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 49 2 Weight of roots (g/m ) at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ....... ........... ...................... 49 2 Weight of roots (g/m ) at physiological maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 51 Top/root ratio at anthesis of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ..... 51 Top/root ratio at physiological maturity of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ........................................ 53 Vi LIST OF FIGURES Figure Page 1. Summary of the rainfall and irrigation for the 1986 growing season at the Montcalm Research Farm .......... 21 vii LIST OF APPENDIX TABLES Table Page A. Analysis of variance for stomatal resistance of six dry bean genotypes grown under water stress and irrigation. Montcalm, 1986............................73 Analysis of variance for water potential of six dry bean genotypes grown under water stress and irrigation. Montcalm, 1986..................... ....... 73 Drought susceptibility index, yield differential, arithmetic mean, and geometric mean of 11 dry bean genotypes. Montcalm, 1986.............. ..... ..........74 2 Mean yield (g/m ) of the selected three top yielding dry bean genotypes, using four different selection criteria. Montcalm, 1986....................74 Analysis of variance for biological yield at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986..... ............ 75 Analysis of variance for biological yield at maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ................. 75 Analysis of variance for days to anthesis of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986........................... ............. 76 Analysis of variance for days to maturity of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986.. ...................................... 76 Analysis of variance for stem weight at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 77 Analysis of variance for stem weight at maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986 ............................ 77 viii Analysis of variance for root weight at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986............................78 Analysis of variance for root weight at maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986. ..... ....... ..... ..........78 Analysis of variance for top/root ratio at anthesis, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986.................79 Analysis of variance for top/root ratio at maturity, of 11 dry bean genotypes under water stress and irrigation. Montcalm, 1986........ ......... 79 Soil water content (8). at three stages of development of the dry bean plant under water stress and irrigation. Montcalm, 1986.................80 INTRODUCTION The evaluation of dry bean genotypes, using only seed yield as selection criterion, has helped to increase bean yields, but this kind of assessment is not going to show the way toward combinationg _£ gggired characters which can lead to higher seed yields. Seed yield alone, in the early generations of selection, is not always a reliable selection criterion, since the desired or best characterized genotypes generally can not be unmistakenly recognized. As a result, improvement of seed yield has been slow, because it has been based largely upon combinations of characters designed mainly by chance. There has been little breeding or direct selection for specific drought resistant characters, because the characters especially beneficial in stress environments have not been identified in dry beans. However, both physiological and morphological characteristics of the dry bean plant are believed to play major and inter-independent roles in determining seed yield under conditions of water stress. The purpose of the present investigation was to search for and identify patterns of physiological and morphological characteristics in a set of dry bean genotypes 2 which we could then relate to seed yield under water stress imposed at the most sensitive stages of plant development: flowering and seed filling (reported by studies made in dry beans (Robins and Domingo, 1956), snap beans (Miller and Gardner, 1972), soybeans (Sionit and Kramer, 1977), broad beans (E1 Nadi, 1969), cowpeas (Turk et. a1., 1980)). The main procedures used to analyze the data were Principal Factor Analysis, in which patterns of traits were equated to one or more principal factors, and Stepwise Regression Analysis, which introduces characters into a multiple regression equation in the order of each trait's contribution to yield as determined by total variance. LITERATURE REVIEW DROUGHT DEFINITION. Agriculturally, drought is defined as the condition that exists when there is insufficient water available to support vigorous crop growth (Van Bavel and Verlinden, 1956). Water deficit or water stress occurs whenever the loss of water by transpiration exceeds the rate of absorption. It is characterized in the crop by modifications of certain developmental and physiological processes which condition crop growth (Kramer, 1969). PLANT RESPONSE TO HATER STRESS. HORPHOLOGICAL EEEECTS. The growth and development of a plant depends on continuing cell division, on the progressive initiation of tissue and organ primordia and on the differentiation and expansion of cells until the characteristic form of the plant is realized (Slatyer, 1973). Cell division appears to be less sensitive to water deficits than cell enlargement. Evidence for this is given by the observation that cell number is frequently of the same general order in plants exposed to water stress, as 61:11 4 compared with irrigated ones, although cell size is greater in the latter (Petinov, 1965); and by the phenomena of more rapid growth on recovery (young tomato plant) from water stress, as compared with irrigated (Gates, 1955). The most obvious morphological change with the onset of drought is a reduction in leaf area, either through a reduction in leaf size, leaf senescence, or by the shedding or death of leaves (Hsiao and Acevedo, 1974). Studies with sunflower have shown that water stress during vegetative development reduces leaf size by about two thirds (Turner et. a1., 1978). Since yield in sunflower is positively correlated with leaf area at flowering (Rawson and Constable, 1980), an irreversible reduction in leaf size due to water stress during vegetative growth will also reduce yields, even if water stress is relieved during late vegetative and reproductive growth. Although total plant growth is reduced during water stress, root growth is generally favored relative to shoot growth as indicated by reported increases in the root to shoot ratio (Pearson, 1966). In fact, water stress leads to a reduction in the total growth of the plant. It influences the growth of different organs differently. There is normally an increase in the ratio of root to top growth, and a decrease in the proportion of the lateral roots to total root length. The ratio of leaf to stem is decreased. Therefore, during a period of soil water stress the growth of organs is influenced in order of decreasing severity: leaves) stems) roots. PHYSIOLOGICAL EFFECTS. a).-Leaf resistance and water potential. Water stress affects water transport through the plant, and hence the rate of water absorption and transpiration. As soil water content falls, the water potential gradient required to move water to the roots and into the leaves increase so that leaf water potential drops to levels which induce stomatal closure. This condition continues until leaf water potential rises and the stomata re-open. Stomata do not respond to changes in leaf water potential or relative water content until a critical threshold level of these parameters is reached. The stomata of field grown grapes, soybean and wheat, have been shown to close at -13 bars (Smart, 1974), -10 to -12 bars (Boyer, 1970), and -13 to -14 bars (Millar and Denmead, 1976), respectively. The threshold potential for stomatal closure and leaf water potential is lower for plants grown under field conditions than for those grown under controlled conditions. This was shown for cotton (Jordan and Ritchie, 1971), and confirmed for a number of other species in various situations (Begg and Turner, 1976), (Davies, 1977). Turner (1974) suggested two possible reasons for this occurrence: first, plants in the field are subject to much higher irradiances than are those in a growth chamber; and second, 6 they have an unrestricted soil volume into which the roots may grow. However, there is not a unique value of leaf water potential for stomatal closure in any particular species or cultivar (Turner, 1974). The leaf water potential for stomatal closure varies with position of the leaf in the canopy, plant age, and growth conditions. At zero water potential, the individual cells are fully turgid: as the water potential drops, cell turgor pressure drops, and when the cell water potential reaches the level at which cell turgor pressure is zero, the cell is quite flaccid (plasmolysis occurs). There is good evidence that, in those leaves which do exhibit complete and irreversible wilting, this occurs at the point when cell turgor pressure reaches zero (Slatyer, 1957). Turner (1974) suggested that because of the differences in osmotic potential and osmotic adjustment (under field conditions) between plant species, stomatal conductance may be better related to leaf turgor pressure than leaf water potential. Leaf turgor pressure is however, difficult to measure under field conditions. Osmotic adjustment within the leaf and particularly in the guard and subsidiary cells of plants undergoing water stress assist the maintenance of turgor by partially offsetting the decline in water potential and allowing partial stomatal opening (Begg and Turner, 1976). Pasture grasses characterized by C4 metabolism Show considerable 7 osmotic adjustment as compared with the small-seeded legume siratro (Macroptilium atropurpureum) which scarcely adjusts (Wilson et. a1., 1980). Partial stomatal opening in the grasses does allow some CO fixation and this is used to produce roots which are abIe to explore new soil and extract more water. This helps to explain the higher root/shoot ratio of stressed plants. This osmotic adjustment assists in the maintenance of turgor by a shift in carbon partitioning on a whole plant. The grass leaves adjust osmotically to water stress, apparently through accumulation of solutes. In contrast, in siratro, osmotic solutes do not appear to accumulate. Osmotic changes are relatively small and are derived from a decrease in the hydration of the leaf. Because its leaves are not drought tolerant (they die when their water potential is lower than -20 bars), it is disadvantageous for this species to make a large osmotic adjustment and keep its stomata open. Instead, under water stress, it adapts morphologically (reductions in absorbed radiation, and in evaporative surface) so that water loss is minimized and leaf water potential is maintained at relatively high values by comparison with grasses (Ludlow and Ibaraki, 1979). b).-Photosynthesis. It is now generally accepted that stomatal closure affecting conductance of CO through the stomata is the 2 8 primary cause of depressed photosynthesis under water limiting conditions. Slatyer (1969) stated that stomatal closure, may be the primary mechanism by which water stress leads to reduced net photosynthesis, by directly impeding CO supply and indirectly by increasing leaf temperature. 2 Consequently, a reduction in CO assimilation follows a pattern similar (in a general sinse) to transpiration reduction. This parallelism has been frequently observed (Boyer, 1970), (Brix, 1962). Throughton (1969), in a study with cotton leaves, provided evidence that the observed reduction in net photosynthesis with increasing water stress can be attributed to stomatal closure and changes in mesophyll resistance, until quite severe stress exists. Net photosynthesis and transpiration rates of dry bean under greenhouse conditions began to decrease simultaneously at about -5 bars leaf water potential, and were near zero at -9 to -10 bars (Moldau, 1973), (O'Toole et. a1., 1977). This suggests that stomatal closure was the principal causal factor in the reduction of net photosynthesis. Photosynthesis declines initially as a result of stomatal closure, but prolonged and severe water stress can lead to depression of chloroplast and enzyme activity and to non-stomatal effects on photosynthesis such as an increase in resistance for CO2 diffusion in the liquid phase from the mesophyll cell wall to the chloroplast. Non—stomatal effects on photosynthesis have been reported to be of considerable 9 magnitude in sunflower (Potter and Boyer, 1973), in certain varieties of maize, and in tobacco (Redshaw and Meidner, 1972). In some species under field conditions, the stage of development influences the rate of photosynthetic reduction; Ghorashy et. al. (1971) found a linear decrease in the rate of apparent photosynthesis in isolines of "Clark" soybeans as leaf water potential was lowered, although the decrease was greater during pod filling stage than during flowering. c).-Translocation and distribution of assimilates. High yields are associated with high photosynthetic rates and may result from currently photosynthetic activity or from greater demand for assimilates. The translocation (transport) of photosynthate involves three aspects: those affecting assimilate supply (source of assimilate), the transfer of assimilate through the phloem, and those concerned with the sink strength and capacity of assimilate (sink). A sink exists wherever the products of photosynthesis are utilized in the plant. According to Kramer (1969), Weibe and Wihrheim in 1962 reported that translocation of 14C labeled photosynthate out of sunflower leaves was reduced about one third as the water potential was decreased to -10 or -12 bars. There is ample evidence that the rate of translocation of assimilates out of the photosynthetic site is reduced considerably under water stress. This has been shown by 10 Wardlaw (1969) who studied wheat at the grain filling stage, and rye grass (Lolium temulentum L;) at the vegetative stage. In the case of wheat, the sink for assimilates (the grain) was less sensitive to water stress than the photosynthesis of the source leaf. A reduced source of assimilates and a reduction of loading into the vascular system were considered to be the most likely factors limiting translocation. On the other hand, in rye grass, the sink for assimilates (the expanding leaf) was more sensitive to water stress than photosynthesis of the source leaf. Here the inhibition of leaf growth was considered to be the major factor reducing translocation. Wardlaw concluded reduced translocation under water stress is the result of a reduction in the source strength due to a reduction in photosynthesis, or a reduction in the growth of the sink, and is not the result of any direct effect on the conducting system itself. Wardlaw (1967, 1969) also showed that the stage of development was important in the distribution of assimilates under water stress conditions. In rye grass, at the vegetative stage, labeled assimilates moved preferentially to young leaves, sheaths, and roots: whereas, in wheat at the grain filling stage, reduced leaf photosynthesis resulted in assimilates moving from the lower leaves, stems, roots and crown to the head. According to Wardlaw (1971), the velocity of assimilate movement in the vascular system, if influenced by the 11 resistance to volume flow, is little affected by water stress. The rate of assimilation and loading of the sieve elements are considered to be more sensitive to this stress and therefore, are the main factors limiting translocation. Yoshida (1972) cited several studies with limited nitrogen fertilization of rice in which up to 40% of the final grain yield was translocated from the stem. Passioura (1976) showed that the grain of severely water stressed wheat plants was filled largely (up to two thirds) from reserves produced before anthesis, rather than largely from current photosynthate, and only one third from current assimilation in the period after anthesis. EFFECTS OF HATER DEFICITS ON YIELD. Photosynthate deposited in grain can come from three major sources, current leaf photosynthesis, current photosynthesis from non leaf parts, and remobilization of assimilate deposited in other plant organs. How much each one of these factors contributes to final grain yield, is affected by species and environment. The stage of growth at which water stress occurs can exert important influences on the final yield of crop plants. For example, Robins and Domingo (1953), working with corn found that maximum yield was reduced by water stress at the tasseling stage, and Denmead and Shaw (1960), found a reduction of about 50% was due to water stress at the silking stage. For wheat, similar tendencies occur but 12 Watson (1952), and Asana and Saini (1958), have shown that continued active photosynthesis after fruit set is also a important determinant of final yield. There is general agreement that the flowering and pod development stages of legumes are the most sensitive phases affecting seed yield (El Nadi, 1975),(Sionit and Kramer, 1977). El Nadi (1969) found, in broad beans Vicia faba, that plants in the flowering phase proved to be more sensitive to water stress than plants in the vegetative period. Robins and Domingo (1956), working with dry beans, reported a reduction in the number of pods per plant if water stress occurred during vegetative growth. Reduction in the number of pods per plant and number of seeds per pod occurred if plants were stressed during flowering; and reduction of seed weight was observed when the stress was imposed during the pod filling stage. Millar and Gardner (1972) reported a reduction of 47% in dry matter production of field-grown snap beans when soil water potential decreased from -0.28 to -0.40 bars at flowering stage, and the leaf water potential was lower than -8 bars. Downey (1971) studied the timing of stress on grain yield and total weight in maize. He reported that the total, above—ground dry matter, at harvest, was reduced by 29%, though grain yield was unaffected when water stress was allowed for 20 days during male meiosis. But, when water stress was allowed during grain filling, the reduction in 13 total above ground dry matter was 30%, and the grain yield was reduced by 47%. Asana and Basu (1963) showed that, for wheat, reduced photosynthesis caused by water stress in the grain filling stage could be compensated for by enhanced translocation from the stem. In general, yield reductions are related to limitations of transpiration due to inadequate water supply, and to unfavorable effects on allocation, either from direct effects of lowered water potential to the plant, or indirectly via reduced assimilate production at key stages. MECHANISMS OF DROUGHT RESISTANCE. May and Milthorpe (1962) defined drought resistance as the ability of a crop variety, or species, to grow satisfactorily in areas subjected to periodic water deficits. There are three types of drought resistance which plants may exhibit: a).-Drought escape, b).-Drought avoidance, and c).-Drought tolerance. a).-Drought escape may be accomplished either by rapid phenological development or by developmental plasticity (Lawn, 1982). Success has been achieved by shortening the growth cycle of the crop so that the plants mature before soil water limits yield (May and Milthorpe, 1962). Developmental 14 plasticity may be achieved by tillering or progressive flowering associated with an indeterminate growth habit (Turk et. el., 1980). Developmental plasticity facilitates the matching of crop growth and development, to the constraints of the environment, especially by minimizing the occurrence of the critical reproductive phase during drought periods. b).-Drought avoidance is defined as the ability of a plant to maintain a favorable internal water status during periods of water stress. Drought avoidance includes mechanisms which maintain water uptake, and mechanisms which reduce water loss due to transpiration. Mechanisms enabling plants to maintain water uptake can take the form of a deep and dense root system, and increased hydraulic conductance as a result of low resistance to water movement within the plant (particularly within the root). These enable the plant to absorb water rapidly enough to keep up with extremely rapid water loss. Passioura (1972) showed experimentally that increasing the resistance of wheat root systems to water flow by reducing the number of seminal roots, conserved water during the vegetative phase resulting in greater yields than plants with a normal root system, when both were grown on stored water. Mechanisms enabling plants to avoid drought through water conservation include: 15 i).—Reduction in epidermal conductance by the control of stomatal aperture. Stomatal closure reduces transpiration and CO exchange (Boyer and McPherson, 1975). The succulents (pineagple) are an example of plants which are able to prevent dehydration. Szarek and Ting (1974) showed that the internal water potential of the cactus, Opuntia basilaris, did not fall below -16 bars even after four months without rain. ii).-Reduction in absorbed radiation and reduction in evaporative surface. This may be accomplished through leaf orienting the leaf parallel to the sun's rays, leaf rolling, and leaf hairs. Lawn (1982) states that paraheliotropic leaf movements act to reduce total light interception by the canopy. This implies a reduction in photosynthesis, though not necessarily depending on the crop. Drought avoidance mechanisms are often associated with reduced productivity, because mechanisms to reduce water loss are associated with reduced CO2 exchange. There is also a cost to the plant in developing a deep and dense root system to maintain water uptake. c).-Drought tolerance is defined as the ability of a plant to withstand low tissue water potential. Drought tolerance mechanisms include: i).—Maintenance of turgor pressure. This is crucial for cell expansion, growth, and for many of the associated 16 biochemical, physiological, and morphological processes. The two processes which contribute to the maintenance of turgor pressure are osmotic adjustment by solute accumulation, and tissue elasticity. As tissue water content and water potential decline, tissue with high elasticity have a greater ability to maintain turgor pressure than tissue of low elasticity. Osmotic adjustment, generally considered as the net increase in intracellular solutes, usually occurs in response to various environmental stresses. Osmotic adjustment within a leaf, and particularly within the guard and subsidiary cells of plants undergoing stress, assist in the maintenance of turgor by partially offsetting the decline in water potential, and allowing partial stomatal opening (Begg and Turner, 1976). As the leaf water potential declines, osmotic adjustment can maintain turgor, and turgor related processes, such as leaf growth, stomatal opening, and photosynthesis. According to Greacen and Oh (1972) some roots are able to adjust osmotically, maintaining a nearly constant turgor in root cells, through a buildup of solutes, regardless of the variation in soil water potentials. This suggests that root growth is not stopped and top/root ratio may be decreased under water stress conditions. This osmoregulation was considered by Hsiao and Acevedo (1974) as an advantage that may constitute an adaptive mechanism to avoid the effects of water stress by having a deeper root system. 17 ii).-Protop1asmic resistance (desiccation tolerance) is the ability of the protoplasm to survive a serious reduction in water content and being able to develop a regular metabolic functioning. The solute concentration forms the osmotic driving force for water entry into the cell. A high capacity for solute accumulation generates turgor maintenance and low water potential. The extreme case is shown by the resurrection plants, which include several grass species. Even after leaf water potential falls too low (high negative values), they have the ability to recover turgor and grow within a few hours of rewatering (Gaff and Hallam, 1974). MATERIALS AND METHODS The trial was carried out at the Montcalm Research Farm, located in Montcalm County, Michigan, during the growing season of 1986. Eleven strains of dry beans of diverse origin and plant characteristics were grown. The names and description of each entry are shown in Table 1. The experimental units were arranged as a two factor factorial in a split plot design with two replications. Water stress and irrigated condition were the whole plot factor, while genotypes were the split plot factor. The experimental units consisted of four row plots, 6m long with rows 50cm apart. A standard number of seeds (14/m) were sown on June 13. The experimental plots were planted in a deep sandy soil (Montcalm sandy loam). The center rows of each plot, eliminating plants at the row edge, were used for sample collection and final harvest. The two outer rows which acted as guard rows were discarded. All the plots were sampled at beginning bloom and at physiological maturity. Each subsample consisted of 1m row length which exhibited uniform competition. In order to preserve as much roots as possible, each plant sample was dug, and was separated into roots, stems, and leaves. This material was placed in an oven for drying, l8 19 TABLE 1. NAMES AND DESCRIPTIONS OF THE 11 DRY BEAN GENOTYPES GROWN UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. Seed Growth Source Drought Entry Genotype color habit* ** reaction 1 C20 Navy II MSU Susceptible 2 Black Magic Black II MSU Susceptible 3 N81017 Navy II MSU Tolerant 4 LEF-2-RB R.B.S.*** III MEX Tolerant 5 B76001 Black II MSU Tolerant 6 II-900-5-M R.B.S. III MEX Tolerant 7 BAT-85 Cream II CIAT Tolerant 8 A-195 R.B.**** I CIAT Tolerant 9 N80068 Navy II MSU Susceptible 10 B82008 Black II MSU Susceptible 11 San Juan Sel Pinto III CSU Tolerant * I = Determinate habit. II = Indeterminate upright short vine habit. III = Indeterminate long vine with postrate habit. ** MSU = Michigan State University. USA. MEX = Instituto Nacional de Investigaciones Forestales y Agropecuarias. MEXICO. CIAT = Centro Internacional de Agriculture Tropical. COLOMBIA. CSU = Colorado State University. USA. *** R.B.S. = Reddish brown striped. **** R.B. = Reddish brown. 20 after which, the weights were recorded. Water stress was created by covering the area between rows with plastic 15 days before anthesis, and by permanently terminating irrigation. Irrigated treatments were well—watered throughout the growing period. Rainfall was adequate throughout most of the 1986 growing season, except for a lack of enough rain during the flowering and seed filling stages. At those stages of plant development, the irrigated treatments were well-watered, with the supplimented of an overhead sprinkler system; 25 mm of water were applied, each time on August 3 and August 22, to meet the demands of the plant at the reproductive period. Rainfall and irrigation were recorded during the course of the experiment, and are summarized in Figure 1. Before covering the water stress area with plastic, the last significant rainfall was 27.1 mm on July 16; and the significant rainfall that terminated the drought, was a three days period of rain on September 10, 11, and 12, with 35.5, 222.5, and 29.4 mm, respectively. Twenty seven traits were taken or calculated for analysis. These traits and methods of measurement are listed in Table 2. Six of the eleven genotypes were selected for detailed sampling (porometer and pressure chamber measurements). This selection was based on previous information regarding water stress tolerance. ZOHH>OH$UpUH r‘r">'71 21 PLASTIC COVERING STRESS AREA END BLOOM HARVEST PLANTED 1 ANTHESIS 1 1 1 25. 201 10, o 222 222.5 1L 9 61.2 501i 4 +, 40 3o 20 10 1| ‘9 0 _ , I'lllll ' [1 f1 ‘ 1 15 1 15 1 15 1 15 JUNE JULY AUGUST SEPTEMBER FIGURE #1. SUMMARY OF THE RAINFALL AND IRRIGATION (mm) FOR THE 1986 GROWING SEASON AT THE MONTCALM RESEARCH FARM. 22 TABLE 2. MEASURED AND CALCULATED CHARACTERISTICS AND METHODS OF MEASUREMENT 1. ANTHES: Days to anthesis, when 50% of the plants had one open flower. 2. ENDFLO: Days to end bloom, 50% of the plants without flowers. '3. PHYMA: Days to physiological maturity, when one pod had a mature color on 50% of the plants. 4. HARVE: Days to harvesting. 5. STEANT: Dry weight of stems at anthesis, using 1m sample. 6. LEANT: Dry weight of leaves at anthesis, using 1m sample. 7. ROANT: Dry weight of roots at anthesis, using 1m sample. 8. IKIANT: IKI test in roots at anthesis. Using starch indicator solution (Adams et. a1., 1978), made by using 0.3g iodine and 1.5g potassium iodide in 100 ml water. Treating freshly cut root tissues with IKI and visually ranking the brilliant blue color development, the amount of starch was rated on a five-point scale, 1(least) to 5(most). 9. STEPHY: Dry weight of stems at maturity, using 1m sample. 10.LEPHY: Dry weight of leaves at maturity, using 1m sample. 11.ROPHY: Dry weight of roots at maturity, using 1m sample. 12.PODWAL: Dry weight of pod walls. 13.IKIPHY: IKI test in roots at maturity. l4.SEPHY: Number of seeds in 20 pods, at harvesting time. 15.GRAIN: Grain yield, using one 5m length row. 16.DUFLO: Duration of flowering; difference between anthesis and end bloom. 17.DUFIL: Duration of seed filling: difference between beginning seed filling and physiological maturity. 18.TORANT: Top/root ratio at anthesis; stems plus leaves divided by roots at beginning bloom. 19.TORPHY: Top/root ratio at maturity; stems plus leaves plus pods divided by roots at physiological maturity. 20.LESANT: Leaf/stem ratio at anthesis; leaves divided by stems at beginning bloom. 21.LESPHY: Leaf/stem ratio at maturity; leaves divided by stems at physiological maturity. 22.EFFI: Yield efficiency, grain yield in g/square m/day. 23.INDE: Harvest index, economic yield divided by biological yield. 24.BIANT: Biomass at anthesis; leaves plus stems plus roots at beginning bloom. 25.BIPHY: Biomass at maturity; leaves plus stems plus roots plus pods (including seeds) at physiological maturity. 26.WESEE: Weight of 100 seeds in grams. 27.LEGRO: Leaf growth rate in square cm, calculation by a=.624+.583(lxw); using length and width leaf. —_————-—-—-—-..—————_--——-——————-——————_————_————-—___.__-—————_-———_“ 23 INDICATORS OF WATER STATUS AND STOMATAL CONDITION. The most recognized parameters for studying plant water status and its dynamics in the soil-plant atmosphere continuum are, water potential and stomatal resistance. The value of water potential is numerically equal to the pressure that must be applied to the system in order to prevent the net movement of water molecules through a membrane from pure water into the system. Leaf water potential provides a useful indirect measure of water economy. Similarly leaf diffusive resistance, diffusion of water vapor from the substomatal cavities through the stomata, governed largely by stomatal aperture, is a major controlling factor in leaf photosynthesis. Since stomata are the ports of transport for both water vapor and CO , stomatal resistance is used to study plant responses 30 changes of water potential, and to infer the status of other plant parameters, such as transpiration rates, plant water status, and CO assimilation (Fischer et.al., 1977). In this siudy, water potential and stomatal resistance were taken at the end of bloom, on the 66th and 67th day after planting. Leaf water potential was estimated by xylem pressure potential, measured with the pressure chamber. It was taken in two days, on four plants, two plants per plot per day, between 11:00 AM and 3:00 PM, on completely developed trifoliate leaves from the top part of each plant. 24 This measurement was done immediately after the stomatal resistance reading had been recorded. The first hour was used to take measurements of plants under water stress on the first replication; the second hour for plants under irrigation on the first replication; the third hour for plants under stress on the second replication, and the fourth hour for plants under irrigation on the second replication. Leaf diffusive resistance was measured using a ventilated diffusion porometer. The readings were taken on the lower surface of the central leaflet of the same trifoliates used for leaf water potential measurement. These measurements were also taken between 11:00 AM and 3:00 PM, in order to avoid excessive cyclic variation, and to approximate the peak of resistance in all the genotypes used. METHODS OF ANALYSIS. Two methods of analysis were used to identify plant characters associated with seed yield under water stress conditions: Principal Factor Analysis (PFA), and Stepwise Regression Analysis (SRA). PRINCIPAL FACTOR ANALYSIS.- A branch of multivariate analysis, PFA is a statistical technique for reducing a large number of correlated variables to a small number of uncorrelated variables or factors (Veldman, 1967). The 25 correlated variables with which the calculation is begun, consist usually of measurements of observable traits. The uncorrelated variables are abstract hypothetical variables called factors (Burt and Banks, 1947). These factors are the basic influences in the data which can be used later, instead of the more numerous original variables, since they explain most of the variation in the data set (Catell, 1965). In the present study, versions of principal factor analysis, image analysis and varimax orthogonal rotation, have been used. Image analysis was employed as direct solution used to transform a correlation matrix into a factor matrix. Varimax orthogonal rotation was employed as a derived solution method, rotation of the common factor space, to accentuate the larger loadings in each factor and suppress the minor loading coefficients, facilitating the biological interpretation of each factor. In the principal factor analysis each observed variable is described as a linear regression on "n" factors (which result from the correlations among the variables) plus a unique factor (which represents the remaining variance of a particular variable not accounted for in the set of correlations). The computation starts from the image covariance matrix, which is a transformed correlation matrix. The correlation of each test (water stress and irrigated) with itself would give diagonal values of unity and these are 26 replaced by communalities of the variables. Communality is the amount of the variance of a variable accounted for by the factors taken together. Subsequently, the Jacobi method is applied to the image covariance matrix to extract the variances or portions of the total variation which are extracted by the principal components. Afterwards, the matrix of factor loadings is submitted to a varimax orthogonal rotation, to maximize the effect of the high loading variables, making easier the interpretation of the factors. The major variables in a factor would show high loading values. In the present study, only high-valued loading coefficients were considered, those displaying values of .8 and above. The high negative loadings were viewed as antifactor variables. An increase in the strength of the factor would tend to lead to a reduction in the level of the negative loaded variables. In this study, when the contribution of a factor to the total variance was less than 10%, the process stopped, allowing only the more important factors to be retained. The data set for the analysis consisted of 27 characteristics measured on 11 genotypes grown at one location under water stress and irrigated conditions, respectively. Analyses were performed on subsets of the data, namely, each condition separately. The factor vectors extracted in each set were compared on the basis of their factor loadings 27 and interpreted biologically. The computer program used was programmed on an IBM computer using the Proc Factor from SAS (Statistical Analysis System). STEPWISE REGRESSION ANALYSIS, computes a sequence of regression equations, at each step adding or deleting an independent variable in the order in which it contributes variance to the dependent variable until the regression equation is satisfactory. The order of insertion is determined by using the partial correlation coefficient as a measure of the importance of variables not yet in the equation (Draper and Smith, 1980). To begin, the SRA routine selects the independent variable most correlated with the dependent variable (seed yield) and find the regression equation. Afterwards, this variable is checked to know if is significant; if it is, the variable is not removed from the equation and the process continues examining the partial correlation coefficients of all the independent variables not in the equation, then the independent variable with the highest partial correlation coefficient is selected to be added into the equation. After that, the lower of these two partial F's is compared with an appropriate F percentage point, and the variable is retained or rejected. This procedure continues until none of the independent variables in the current equation can be removed, and the next best independent variable can not hold its place in the 28 equation, so at this point the routine terminates. In this study, both entry and exit tests were made at the level aC= 0.15. In summary, in this study Factor analysis was used to group plant characteristics into uncorrelated groups, which were constituted by characters that explained the same factor or biological concept; whereas, Stepwise Regression analysis was used to identify or corroborate the characters related to the yield factor disclosed by Factor Analysis. RESULTS STOMATAL RESISTANCE AND WATER POTENTIAL. As was mentioned earlier, porometer and pressure chamber measurements were taken only on six of the eleven genotypes studied, with prior knowledge that four of them, namely, LEF-2-RB, B76001, II-900-5-M, and BAT-85, are drought tolerant to some extent, whereas, the other two: N80068, and 882008, are drought susceptible. Stomatal resistance. The readings of this variable were taken on the 66th and 67th day after planting, i.e., during the reproductive stage. At this stage, there was a statistically non-significant difference between stressed -1 and irrigated treatments (8.15 sec/cm stress and 3.72 -1 sec/cm irrigated), and a significant difference between genotypes (Table A, in Appendix). In the stressed treatment (Table 3), the genotypes LEF- 2-RB, B76001 and II—900-5-M (11.66, 12.45, and 11.61 sec/cm-l, respectively) had the higher resistances, while lower resistances appeared for the genotypes BAT-85, N80068, and B82008 (6.18, 4.43, and 2.60 sec/cm—l, respectively). For the irrigated plants (Table 3), the genotype LEE-2- -1 RB (7.20 sec/cm ) had the highest resistance; while lower 30 -1 TABLE 3. STOMATAL RESISTANCE MEANS (sec/cm ) OF SIX DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. WATER TREATMENTS N s A Water stress 8.15 B Irrigated 3.72 GENOTYPES Signif. different at P é 0.05 --------- WATER TREATMENT A B 4.-LEF-2-RB 11.66 7.20 5.-B76001 12.45 3.90 6.-II-900-5—M 11.61 2.52 7.-BAT-85 6.18 4.09 9.-N80068 4.43 2.15 10.-B82008 2.60 2.51 . LSD at P é 0.1 = 10.71 ** LSD at P é 0.05 = 18.49 ***LSD at P é = TABLE 4. XYLEM WATER POTENTIAL MEANS (~bars) OF SIX DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. WATER TREATMENTS Significance at P é 0.1 A Water stress -8.91 B Irrigated -6.49 GENOTYPES Significance at P é 0.05 --------- WATER TREATMENT A B 4.-LEF-2-RB —8.10 -7.10 5.-B76001 -8.90 -5.55** 6.—II-900-5-M -10.80 —7.10** 7.-BAT—85 -7.70 -6.90 9.-N80068 —10.40 -6.50** 10.-B82008 -7.60 -5.80 .-—---—--—--————————--———————.——————---———————-—-—--————————-— * LSD at P é£0.1 = ** LSD at P = 0.05 *** LSD at P é 0.0 31 resistance appeared for the rest of the genotypes. Water potential. This variable was taken following the readings of stomatal resistance. There was a significant difference between stressed (-8.91 bars) and irrigated (-6.49 bars) treatments (P é 0.06), a significant difference between genotypes, and a significant interaction between genotype and water treatment (Table B, in Appendix). Genotypes II-900-5-M and N80068 showed the lowest values (-10.80, and -10.40 bars, respectively) in the stressed treatment (Table 4); while genotypes LEF-2-RB, B76001, BAT-85, and B82008 had higher values under stress (-8.10, -8.90, -7.70, and -7.6 bars, respectively). Under irrigation (Table 4) all genotypes showed higher water potential values, compared to the stressed ones with values ranging from -5.55 to ~7.10 bars. STEPUISE MULTIPLE REGRESSION ANALYSIS. WATER STRESS. According to the equation calculated by the multiple regression analysis for dependent variable grain yield, when the primary yield components, number of seeds (X14), yield efficiency (X22), harvest index (X23), and seed weight (X26), were excluded from variables going into the model, the best model selected included nine variables and gave a 2 R value of 1.0. Three variables accounted for almost 99% of variance. These variables were: biomass at maturity (X25), weight of roots at anthesis (X7), and weight of leaves at maturity 32 (X10); in decreasing order of importance according to the magnitude of the R2 value. These results are given in Table 5. Thus, according to the contribution in percentage to the total variance made by each variable, the character which made the largest single contribution to yield variance was biomass at maturity (X25), accounting for 91% of the total variation. STEPWISE MULTIPLE REGRESSION ANALYSIS. IRRIGATED CONDITION. According to the equation calculated by the multiple regression analysis for the dependent variable grain yield, when primary yield components (the same as those listed in the water stress condition) were excluded from variables going into the model, the best model selected included eigth variables and gave an R2 value of 0.999 (Table 6). Three variables accounted for 99% of the total variance. These variables were: biomass at maturity (X25), weight of leaves at maturity (X10), and top/root ratio at maturity (X19); in decreasing order of importance according to the magnitude of the R2 value. As in the water stressed treatment, the variable, biomass at maturity (X25), made the largest contribution to yield variance, accounting for 79% of the total variation. The positively loaded variable top/root ratio at maturity (X19), and the negatively loaded variable, weight of leaves at maturity (X10), both contributed 10% to the total 33 TABLE 5. STEPWISE REGRESSION ANALYSIS OF DATA UNDER WATER STRESS WITH GRAIN YIELD AS DEPENDENT VARIABLE. 2 VARIABLE PARTIAL REGR. COEFF. (b) F PARTIAL R INTERCEPT -3.697 BIPHY (X25) 0.696 486.42 0.9121 ROANT (X7) -1.741 28.27 0.0517 LEPHY (X10) -0.344 13.48 0.0238 SOURCE DF MEAN SQUARE F Regression 3 9168.78 186.56** Error 7 49.14 2 R = 0.9876 TABLE 6. STEPWISE REGRESSION ANALYSIS OF DATA UNDER IRRIGATION WITH GRAIN YIELD AS DEPENDENT VARIABLE. 2 VARIABLE PARTIAL REGR. COEFF. (b) F PARTIAL R INTERCEPT -83.863 BIPHY (X25) 0.777 802.81 0.7934 LEPHY (X10) -0.916 189.83 0.0974 TORPHY (X19) 1.246 98.34 0.1019 SOURCE DP MEAN SQUARE F Regression 3 9579.333 319.32** Error 7 29.998 2 -_---——————---——--————_——-——————_———————-————_———_-——~—~—-~— 34 variance. FACTOR ANALYSIS. WATER STRESS. As was mentioned earlier, we were able to offer biological interpretations on the basis of the high valued loading coefficients, of 0.8 and above, in both conditions. The results of the water stress data indicate that 99.67% of the total variation is common, that is, the common factors can account for that much of the variation, and the remaining 0.33% should be attributed to unique factors and errors. Four latent roots extracted 86.21% of the trace variation. The first two roots accounted for more than 66% before rotation and more than 50% after rotation. The varimax rotation results showed the third and fourth factors increasing in importance, from about 11 and 8% before rotation to about 16 and 11% after rotation, respectively; whereas, the first two factors retained their importance, from about 45 and 20% before rotation to about 30 and 20% after rotation, respectively (Table 7). There were five factors with roots greater than unity, and four factors with a contribution greater than 10% to the total variance. The first factor made the largest contribution to the variance and accounted for 45.47 and 30.79% of the total variation, before and after rotation, respectively. The variables with the highest loadings in the first factor were weight of roots at physiological maturity 35 TABLE 7. FACTOR ANALYSIS RESULTS: WATER STRESS CONDITION. TRAITS 1 2 3 4 ANTHES (X1) .8662 .3765 -.1046 .1416 ENDFLO (X2) .6928 .1254 -.4941 .1072 PHYMA (X3) .7618 .4648 - 2380 2021 HARVE (X4) .7618 4814 -.2484 .1804 STEANT (X5) -.2064 -.0647 .9360 -.1364 LEANT (X6) -.3813 .0701 .7881 .3440 ROANT (X7) .4405 .1909 .1167 -.2605 IKIANT (X8) -.5120 -.0303 .4206 - 0451 STEPHY (X9) .7476 .4457 -.1225 .1645 LEPHY (X10) -.4895 .6131 .2169 -.0282 ROPHY (X11) .8937 .2204 -.1479 .1342 PODWAL (X12) .4423 .7219 -.0453 -.1688 IKIPHY (X13) .3669 .3906 -.3411 -.1306 SEPHY (X14) -.3320 -.0921 -.0406 -.9151 GRAIN (X15) .2615 .9206 -.0152 .2378 DUFLO (X16) .3375 - 0933 - 6015 0468 DUFIL (X17) .6732 4695 - 2900 2165 TORANT (X18) -.5753 -.1749 .5482 .2416 TORPHY (X19) -.8532 .2339 .3368 .0535 LESANT (X20) -.0008 .2802 -.2714 .7704 LESPHY (X21) -.8472 .1742 .3197 -.0198 EFFI (X22) -.0526 .9643 .0936 .1735 INDE (X23) .1885 .4807 -.0700 .2644 BIANT (X24) -.1840 .0845 .9404 .0802 BIPHY (X25) .2622 .9369 .0205 .1630 WESEE (X26) .1291 .3046 .3501 .8450 LEGRO (X27) .7593 .0630 -.0297 .4805 % VARIANCE 30.79 19.89 16.01 11.40 CUM. % VAR. 30.79 50.68 66.69 78.09 36 (X11), and days to beginning bloom (X1). The variables with the highest negative loadings in this factor were leaf/stem ratio at physiological maturity (X21), and top/root ratio at physiological maturity (X19). This is essentially a vigor (strength shown in development) and development (growth to become into a more complete state) factor. The second factor accounted for 20.61 and 19.89% of the total variation, before and after rotation, respectively. This factor was highly associated with yield efficiency (X22), biomass at physiological maturity (X25), and grain yield (X15). Consequently, this factor was named yield, because it identified itself with known yield characters. The third factor accounted for 11.81 and 16.01% of the total variation, before and after rotation, respectively. This factor was highly associated with biomass at beginning bloom (X24), and weight of stems at beginning bloom (X5). Therefore, this factor was called biomass at anthesis. Factor four accounted for 8.32 and 11.40% of the total variation, before and after rotation, respectively. This factor was composed of yield components, represented on the positive axis by weight of 100 seeds (X26), and on the negative axis by number of seeds per pod (X14). This factor was named yield-component. FACTOR ANALYSIS. IRRIGATED CONDITION. The results of the irrigated condition indicate that 99.71% of the total variation is common. Four roots 37 extracted 82.82% of the trace. The first two roots accounted for more than 59% before rotation, and more than 50% after rotation (Table 8). There were five factors with roots greater than unity, and four factors with a contribution greater than 10% to the total variance. The first factor accounted for 42.65 and 34.60% of the total variation, before and after rotation, respectively. The variables with the highest loadings in this factor were days to harvesting (X4), days to physiological maturity (X3), weight of stems at physiological maturity (X9), duration of seed filling (X17), weight of roots at physiological maturity (X11), and days to end bloom (X2). This factor was thus named vigor and development. The second factor accounted for 16.54 and 15.73% of the total variation, before and after rotation, respectively. This factor was highly associated with grain yield (X15), biomass at physiological maturity (X25), and yield efficiency (X22). As in water stress condition, we have denoted this factor as yield. The third factor accounted for 14.04 and 13.24% of the total variation, before and after rotation respectively. This factor was highly associated with weight of seeds (X26), which is, as before, compensated by the negative loading for number of seeds per pod (X14). This factor was named yield component because it consisted of the same high loading variables as were under water stress condition. 38 TABLE 8. FACTOR ANALYSIS RESULTS: IRRIGATED CONDITION. TRAITS 1 2 3 4 ANTHES (X1) .7424 -.2596 .4428 -.1536 ENDFLO (X2) .8245 -.2637 .0730 -.2173 PHYMA (X3) .9491 .0212 .2517 -.1486 HARVE (X4) .9514 .0268 .2576 -.1438 STEANT (X5) - 2557 .0573 .1087 .9471 LEANT (X6) -.3837 .0468 .4453 .6006 ROANT (X7) .4374 -.0552 -.3691 .5604 IKIANT (X8) -.6079 .2448 -.1284 1265 STEPHY (X9) .9459 .0706 .1435 -.1182 LEPHY (X10) -.4941 .4849 .0906 .1368 ROPHY (X11) .8801 -.1824 .0864 - 0296 PODWAL (X12) -.0840 .7050 - 4261 - 0757 IKIPHY (X13) .2950 .1150 -.0212 -.1811 SEPHY (X14) -.3261 .1055 -.9198 .0193 GRAIN (X15) .0382 .9786 .0446 -.0026 DUFLO (X16) .6459 -.1933 -.1673 -.1952 DUFIL (X17) .9363 .1433 .1393 -.1301 TORANT (X18) -.6119 .1979 .4183 .1901 TORPHY (X19) -.7601 .3924 .1637 .1012 LESANT (X20) .2524 .0634 .4260 -.6l84 LESPHY (X21) -.7522 .1697 .0324 .0933 EFFI (X22) -.3320 .9233 -.0048 .0723 INDE (X23) .1242 .4713 -.0332 -.0647 BIANT (X24) -.1983 .0349 .1915 .9267 BIPHY (X25) .0254 .9523 .0669 .0352 WESEE (X26) .1255 .1334 .9373 .2062 LEGRO (X27) .5004 -.0661 .6814 .1751 % VARIANCE 34.60 15.73 13.24 11.88 CUM. % VAR. 34.60 50.33 63.57 75.45 5‘ 39 Factor four accounted for 9.59 and 11.88% of the total variation, before and after rotation, respectively. This factor was highly associated with weight of stems at beginning bloom (X5), and biomass at beginning bloom (X24). Thus, factor four was named biomass at anthesis, and it consisted of the same high loading variables as under the water stress condition. ANALYSIS OF VARIANCE OF THE IMPORTANT TRAITS. ECONOMIC YIELD. The analysis of variance revealed a significant genotypic effect, a significant water effect, and an interaction between both factors (the last one could be due to the unexpected behaviour of the genotype A-195). These results are presented in Table 9. Four of the eleven genotypes had a significant reduction of economic yield under water stress; Black Magic (33%), N81017 (23%), LEF-2-RB (25%), and San Juan Sel (46%). while only one cultivar, A-195 had a significant increase (34%), for this trait under stress. The other seven genotypes did not show any significant difference between treatments, even though there was reduced yield (Table 10). Economic yield was generally reduced under water stress. The yield reduction of the genotype A-l95 under irrigation when compared to the stress treatment, could be explained by considering that since this genotype exhibited the largest complete biological cycle (it needed eight more TABLE 9. GENOTYPES TABLE 10. 40 ANALYSIS OF VARIANCE FOR SEED YIELD OF 11 DRY BEAN UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. Replication Water treatment Error a Genotype Interaction Error b M.S 119.19 25129.46 ** 0.01 8063.94 ** 3295.73 * 1432.09 * Significant at P ** Significant at P C.V. = 15.59% ECONOMIC YIELD (fill ) OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. C20 Black Magic N81017 LEF-2-RB B76001 II-900-5-M BAT—85 A-195 N-80068 B82008 San Juan Sel ---—--———-—--—-———-———---—-—--———-—-———-———————————*—_—-———— * LSD at P ** LSD at *** LSD at P: P IRRIGATED STRESS 205.1 159.2 289.1 192.3** 276.1 . 212.2* 385.3 286.4** 223.8 198.2 286.1 262.7 201.5 174.3 233.6 314.1** 282.3 258.1 239.1 181.5 310.4 167.5*** 266.5 218 7 41 days to mature under irrigation than under stress), the seed already yielded by this genotype under irrigation, was too much exposed to the excess of moisture due to the wet conditions originated by heavy rainfalls during the end of the growing season. Thus, in this fashion the seed yield was severely reduced under irrigation.‘ Under water stress this seed damage did not occur because the plastic used to cover the stress area facilitated the drainage of the abundant rainfall. When selecting the three top yielding genotypes by the criteria of drought susceptibilty index, yield differential, arithmetic mean, and geometric mean; and comparing their yields under water stress and irrigated condition, the mean yields of lines selected by the drought susceptibility index, and yield differential were lower than the mean yields of lines selected on the basis of the arithmetic and geometric means. This suggests the use of either the arithmetic mean or the geometric mean, in order to avoid selecting low yielding genotypes under water stress conditions (Tables C and D, in Appendix). BIOLOGICAL YIELD. The differences between the irrigated and the stress treatments in respect to biomass production were striking. A significant genotype effect as well as a significant water effect were indicated by the analyses of variance of 42 biological yield at beginning bloom and at physiological maturity (Tables E and F, in Appendix). At beginning bloom, all the genotypes except LEF-2—RB suffered a reduction of biological yield under water stress, but only two of them had a significant reduction: N81017 (23%), and San Juan Sel (26%). The other nine genotypes did not show any significant difference between treatments (Table 11). At physiological maturity, all the genotypes except II- 900-5-M and A-195, incurred a reduction of biological yield under water stress, but only five of them had a significant reduction: Black Magic (30%), LEF-2—RB (22%), N80068 (22%), B82008 (26%), and San Juan Sel (43%). The other six genotypes did not show any significant difference between treatments (Table 12). The effects of the stress on the morphology and yield of these genotypes were significant, and were reflected in lower production of dry matter and grain. The weights of the whole plant at anthesis and at maturity and grain yield of the stressed plants were significantly reduced, when compared with the irrigated plants. The results (Table 13) indicate that the stressed beans produce lower amounts of total dry matter and seed yield when compared with the irrigated plants. The decrease in dry matter was 15.41% at anthesis, and 18% at maturity, and the percentage decrease in seed yield was 17.92%. Significant correlations of 0.925 under water stress, TABLE 11. 43 2 BIOLOGICAL YIELD (g/m ) AT BEGINNING BLOOM OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. ENTRY NO. IDENTIFICATION 1 C20 2 Black Magic 3 N81017 4 LEF-2-RB 5 B76001 6 II-900-5-M 7 BAT-85 8 A-195 9 N80068 10 B82008 11 San Juan Sel MEAN * LSD at P é<0.10 = 37.6 ** LSD at P =40.05 = 45.9 *** LSD at P = 0.01 = 66.0 C.V. = 15.34% TABLE 12. IRRIGATED STRESS 119 6 103 2 155 1 124.7 192 3 146.7* 154 2 158.3 172 7 155.3 136.1 116.6 171.3 153.3 175 2 142.4 108 9 98.7 172 2 139.8 202 7 149.9** 160 0 135 3 2 BIOLOGICAL YIELD (g/m ) AT PHYSIOLOGICAL MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. C20 Black Magic N81017 LEF-2-RB B76001 II-900-5-M BAT-85 A-195 N80068 882008 San Juan Sel HOKOCOQONU'IDUNH F‘H * LSD at P ** LSD at P = *** LSD at P C.V. = 14.69% IRRIGATED STRESS 407.9 351.1 532.1 371.0** 550.9 431.3 715.7 554.9** 472.6 461 2 501.4 530.1 451.4 355.4 529.6 578 6 616.6 477.6* 550.2 412.3* 597.4 337 7*** 538 7 441 9 44 TABLE 13. TOTAL DRY MATTER PRODUCTION AND GRAIN MEAN YIELD (g/square m) OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. MOISTURE REGIME BEG. BLOOM PHY. MAT. GRAIN YIELD Water stress 135.38 223.15 218.81 Irrigated 160.05 272.13 266.61 % decrease 15.41 18.00 17.92 * There was a significant difference between water treatments in the three traits. TABLE 14. DAYS TO ANTHESIS OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 45 46* 2 Black Magic 46 47* 3 N81017 44 45** 4 LEF-2-RB 43 44* 5 B76001 45 47*** 6 II-900-5-M 47 47 7 BAT-85 42 42 8 A-195 53 53 9 N80068 45 45 10 B82008 45 45 11 San Juan Sel 41 40* 4 MEAN 45 0 45 5 * LSD at P = 0.1 = 0.95 ** LSD at P é<0.05 = 1.17 *** LSD at P = 0.01 = 1.78 C.V. = 1.25% 45 and of 0.881 under irrigation, between biomass at physiological maturity and economic yield point out the importance of the biomass at physiological maturity in determining final seed yield. DAYS TO BEGINNING BLOOM. "A A significant genotype effect (P .05), and a significant water effect (P § .09) for days to beginning bloom were indicated by the analysis of variance (Table G, in Appendix). Five genotypes: C20, Black Magic, N81017, LEF-2-RB, and B76001, showed a significant increase for this trait under stress. The other six genotypes did not show any difference between water treatments, except San Juan Sel, which showed a significant reduction under the water stress condition (Table 14). A positive correlation of 0.49 between economic yield and days to anthesis, under stress, indicates that delayed flowering might represent greater accumulation of dry matter before the onset of stress, and it could act as a reserve pool of assimilates. If pre—anthesis assimilates are being utilized to fill the seeds, the genotypes which have a longer vegetative growth, those which start flowering later, would be expected to have a greater source of assimilates and, therefore, a greater potential for higher yield. Evidence for this is that, in general, longer season varieties yield more. 46 DAYS TO MATURITY. The analysis of variance revealed a significant water effect, a significant genotype effect, and an interaction between both factors (Table H, in Appendix). All the genotypes showed a significant decrease for this trait under stress, except the genotypes C20 and B82008 which did not show any difference (Table 15). In general, the results of this trait indicates that under water stress there was a tendency of the plant to accelerate the complete biological cycle to avoid in this way the lack of soil moisture, at the critical reproductive stage. WEIGHT OF STEMS. Significant genotype effect and water effect, were indicated by the analysis of variance at anthesis, as well as at maturity (Tables I and J, in Appendix). At beginning bloom, only two of the 11 genotypes, N81017 and San Juan Sel, had significant stem weight reductions, under water stress. The rest of the genotypes showed a tendency toward decreasing stem weight, except LEF— 2-RB which showed an increasing tendency, but the effect was not significant (Table 16). At maturity, all the genotypes showed a tendency toward decreasing stem weight under water stress, but only six genotypes: Black Magic, LEF-2-RB, II-900—5-M, N80068, 47 TABLE 15. DAYS TO MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 94 94 2 Black Magic 99 98** 3 N81017 102 100** 4 LEF-2-RB 94 92** 5 B76001 94 93** 6 II-900-5-M 104 102** 7 BAT-85 84 83** 8 A-195 112 104*** 9 N80068 103 102** 10 B82008 95 95 11 San Juan Sel 88 85*** 4 MEAN 97 95 * LSD at P =‘0.1 = 0.73 ** LSD at P =‘0.05 = 1.00 *** LSD at P = 0.01 - 2.41 C.V. = 0.39% 2 TABLE 16. WEIGHT OF STEMS (g/n ) AT ANTHESIS OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 39.9 31.3 2 Black Magic 52.8 47.6 3 N81017 64.4 46.9* 4 LEF-2-RB 51.8 54.4 5 B76001 63.2 51.3 6 II-900-5-M 42.6 33.7 7 BAT-85 52.5 49.7 8 A-195 54.2 42.1 9 N80068 30.1 29 9 10 B82008 61.6 50.1 11 San Juan Sel 74.1 50.0** 4 MEAN 53 3 44 2 * LSD at P =<0.1 = 15.3 ** LSD at P =<0.05 = 18.5 *** LSD at P = 0.01 = 25.3 C.V. = 19.09% 48 B82008 and San Juan Sel, exhibited significant stem weight reductions (Table 17). The genotypes Black Magic, LEF-2-RB, II-900-5-M, N80068, and B82008, did not show any significant difference between water treatments at anthesis, but showed significant reduction under water stress at physiological maturity. Genotypes BAT-85, and San Juan Sel decreased their stem weight from anthesis to maturity under both water treatments indicating that remobilization of stored assimilates had taken place. The increase in stem weight that most of the genotypes showed from anthesis to maturity under both water treatments, could be explained by considering that since most of the genotypes have an indeterminate growth habit (except A-195 which exhibits determinate growth), the plants kept growing after anthesis. This plant growth could have also been supported by some residual soil moisture still available to the plant, due to the fact that only 15 days before anthesis, the water stress area was covered by the plastic used to sled rain water. WEIGHT OF ROOTS. A significant genotype and water effects were indicated by the analysis of variance at anthesis (Table K, in Appendix); whereas, at maturity, only a significant genotype effect was manifest (Table L, in Appendix). At beginning bloom, three genotypes: LEF-2-RB, 49 2 TABLE 17. WEIGHT OF STEMS (g/m ) AT PHYSIOLOGICAL MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. -—-----————--------——-----—--_--_-----------b————_—-——--———- ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 64.9 47.0 2 Black Magic 71.8 48.6* 3 N81017 74.7 63.8 4 LEF-Z-RB 63.0 41.5* 5 B76001 65.6 61.6 6 II-900-5-M 96.9 74.9* 7 BAT-85 36.1 31.7 8 A-195 94.2 84.0 9 N80068 83.2 55.0** 10 B82008 77.3 55.3* 11 San Juan Sel 53.6 34.1* 4 MEAN 71 0 54 3 * LSD at P =<0.1 = 18.4 ** LSD at P =<0.05 = 23.5 *** LSD at P = 0.01 = 43.4 C.V. = 16.58% TABLE 18. WEIGHT OF ROOTS (g/m ) AT ANTHESIS OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 29.1 27.5 2 Black Magic 41.1 24.6* 3 N81017 54.1 39.6* 4 LEF-Z-RB 21.4 27.8 5 B76001 29.2 44.5* 6 II-900-5-M 30.3 32.1 7 BAT-85 25.3 21.5 8 A-195 32.1 29.4 9 N80068 30.5 26.8 10 B82008 38.3 32.2 11 San Juan Sel 29.4 20.5 < MEAN 32 8 29 6 * LSD at P =40.1 = 14.0 ** LSD at P =<0.05 = 17.0 *** LSD at P = 0.01 = 23.3 C.V. = 27.29% 50 B76001, and II-900-5-M, showed a tendency toward increasing root weight under water stress, but only in B76001 was this significant. These genotypes developed heavier root systems due, possibly, to deeper roots or to a large number of branch roots which allowed them to meet their water needs. The remaining genotypes showed an opposite tendency, but only in the Black Magic and N81017 genotypes was this decrease significant (Table 18). At physiological maturity one genotype, II-900-5-M had significant root weight reductions under water stress. Most of the rest of the genotypes (N81017 and BAT-85, showed a non-significant increase) showed a non-significant tendency to decrease root weight under stress (Table 19). TOP/ROOT RATIO. The statistical analysis showed significant differences between genotypes at anthesis (Table M, in Appendix); at maturity, there were significant differences between genotypes, and a significant genotype water treatment interaction, but no differences between water treatments (Table N, in Appendix). At anthesis, six genotypes: C20, LEF-2-RB, B76001, II- 900-5-M, A-195, and B82008, showed a tendency to decrease top/root ratio under water stress, but only in B76001 the decrease was significant. Three genotypes: N81017, N80068. and San Juan Sel, maintained almost the same ratio under both conditions, whereas, the rest of the genotypes showed a 51 2 TABLE 19. WEIGHT OF ROOTS (g/m ) AT PHYSIOLOGICAL MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 19.4 17.3 2 Black Magic 22.3 17.2 3 N81017 24.0 26.6 4 LEF-2-RB 12.4 10.6 5 B76001 20.4 18.8 6 II-900-5-M 22.3 16.3* 7 BAT-85 7.3 7.5 8 A-195 31.5 30.3 9 N80068 24.2 21.9 10 B82008 20.5 16.8 11 San Juan Sel 8.7 7.8 4 MEAN 19.3 17.3 * LSD at P =<0.1 = 5.2 ** LSD at P = 0.05 = 7 0 *** LSD at P é 0. 1 = 2 c.v. = 12.78% TABLE 20. TOP/ROOT RATIO AT ANTHESIS FOR 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. ENTRY NO. IDENTIFICATION IRRIGATED STRESS 1 C20 3.119 2.770 2 Black Magic 2.771 4 242 3 N81017 2.571 2 722 4 LEF-2-RB 6.263 5 045 5 B76001 5.057 2.885* 6 II-900—5-M 3.486 2.695 7 BAT-85 5.876 6 127 8 A-195 4.450 3 781 9 N80068 2.554 2 646 10 B82008 3.646 3 390 11 San Juan Sel 6.356 6 369 4 MEAN 4.195 3 879 * LSD at P =40.1 = 2.167 ** LSD at P =<0.05 = 3.109 *** LSD at P = 0.01 = 8.703 C.V. = 27.06% 52 non-significant tendency to increase top/root ratio under water stress conditions (Table 20). At physiological maturity, eight of the 11 genotypes showed a tendency to decrease the top/root ratio under water stress, but only in N81017, BAT-85, and San Juan Sel, was the decrease significant. Two of the remaining genotypes: B76001 and A-195, maintained almost the same ratio under both conditions; whereas, the genotype II-900-5-M showed a significant increment (Table 21). 53 TABLE 21. TOP/ROOT RATIO AT PHYSIOLOGICAL MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. 19.833 22.799 23.443 57.125 22.193 21.497 60.929 15.801 24.470 25.748 68.383 21.074 13.746* 51.710 23.461 31.983* 46.126** 17.992 20.453 23.543 41.968** * LSD at P E 0.1 ** LSD at P *** LSD at P = 12.67% C.V. IDENTIFICATION C20 Black Magic N81017 LEF-2-RB B76001 II-900-5—M BAT-85 A-195 N80068 B82008 San Juan Sel MEAN 9.681 é 0.05 = 14.751 0 é 1 = 47.959 32.929 28.303 DISCUSSION INTERPRETATION OF THE RESULTS. STOMATAL RESISTANCE AND WATER POTENTIAL. Analyzing the behavior of each genotype separately, we have the following results: With respect to stomatal resistance, the genotype LEF- 2-RB did not show differences between water treatments, this indicates that its stomata remained not completely opened under both conditions (drought and irrigation). However, under irrigation its stomatal resistance was the highest among all the genotypes, but it was lower when compared to the one shown by itself under water stress, indicating a greater stomatal opening under irrigation. This unusual plant behavior (partial stomatal opening under both water conditions), indicates that this genotype possesses the ability to make gradual stomatal closure, over a wide range of water potential, thus, this genotype could be suitable for large drought periods. Regarding water potential, when stomata are open, if the vapor pressure gradient is high, the water loss from the leaf is high too, this contributes to reduce the water potential. But if stomata close, that should conserve water and results in high water potential. 54 55 Thus, LEF-2-RB was able to exhibit the expected high water potential value, perhaps due to its ability to increase its stomatal resistance avoiding water loss during stress conditions. With respect to the genotypes B76001 and II-900-5-M, considering their stomatal resistance, they showed differences (although not statistically) between water treatments, this indicates that these genotypes possess stomata sensitive enough to permit opening when water is abundant, and closing under high evaporative demand. However, regarding water potential B76001 exhibited (although not statistically) higher values than II-900-5-M. This difference could be explained by considering other characters like root weight and top/root ratio. For example, the genotype B76001 exhibited under water stress, a significant increase of root weight at anthesis (Table 18), and a significant decrease of top/root ratio at anthesis (Table 20). On the other hand, the genotype II-900— 5-M exhibited under stress, a significant reduction of root weight at maturity (Table 19), and a significant increase of top/root ratio at maturity (Table 21), i.e., this genotype was not able to extend its root system into new areas of the soil that may have a higher potential of water. The stomatal resistance of the N80068 genotype did not show differences between water treatments, indicating that its stomata remained open under both water conditions. Therefore, this genotype exhibited the expected high water 56 potential value under irrigation, and low water potential under stress, indicating its inability to conserve water under conditions of high evaporative demand. Finally, the genotypes BAT-85 and B82008, were shown to possessinsensitive stomata (remained open under both water treatments); these genotypes, however, exhibited high water potential values. This behavior could be explained in BAT-85 by considering that under water stress this genotype exhibited a significant decrease of top/root ratio at maturity (Table 21), increasing in this fashion its root system. However, in the case of B82008 which was shown to possess low stomatal resistance and (although not significantly different) high water potential under stress conditions, and also a lack of capacity to develop compensatory characters such as root weight increases; we do not have evidence which permit a reasonable explanation. In general, the results obtained under both conditions indicated that water potential and stomatal resistance of the bean plants were sensitive to differential moisture regimes. The principal effects of irrigation were to create higher values for water potential, and lower values for stomatal resistance. Under water stress, the plants had lower values for water potential, and higher values for stomatal resistance, indicating a higher resistance to gas exchange which caused a reduction in plant growth and yield, and to transpiration which would affect leaf temperature. Summarizing this section on stomatal behavior, we can 57 conclude that plants already have stomata sensitive enough to permit normal transpiration when soil moisture is abundant. On the other hand, stomatal closure can conserve water under high evaporative demand, and also, if this stomatal behavior is combined with other plant features such as the ability to increase root growth and decrease top/root ratio, the result would be a better use of the limited soil moisture in environments with drought. STEPWISE MULTIPLE REGRESSION ANALYSIS. WATER STRESS. IMPORTANT VARIABLES. The positive loaded variable biomass at maturity which, according to the R2 value, explained 91% of the total variance was the most important variable. This is interpreted to mean that successful bean production under water stress depends on plants which are able to effectively exploit photosynthesis in order to render a greater part of their biological yield as grain. IRRIGATED CONDITION. IMPORTANT VARIABLES. According to the R2 values, the most important variables were: for the positive loading, biomass at maturity, and top/root ratio at maturity; and for the negative loading, weight of leaves at maturity. These variables explained 79%, 10%, and 10% of the total variation, respectively. These results are interpreted to mean that because of the availability of soil moisture 58 during the complete biological cycle, it is more important for the plant to develop a greater canopy than an extensive root system. Large values of top/root ratio represent a large photosynthetic area and therefore a substantial carbohydrate manufacturing site, which finally results in a heavy biomass at maturity. With respect to the negative value of weight of leaves at maturity, it indicates that there was an effect attributable to harvest index, but showing up in the form of the negative sign for leaf weight at maturity. Since leaf weight has long been established as important component of competitive ability (Hamblin and Donald, 1974), and since this characteristic will tend to increase the denominator (biological yield) of the harvest index, it follows that competitive ability and harvest index are likely to be negatively related. FACTOR ANALYSIS. WATER STRESS. IMPORTANT VARIABLES. The first factor, vigor and development, is so called because the development of the plant is reflected by the positive loading of days to anthesis (length of the vegetative cycle). Plant vigor is reflected by the negative loading of leaf/stem ratio at maturity, the positive loading of root weight at maturity, and by the negative loading of the top/root ratio at maturity. This indicates that because of the lack of soil moisture, it is important for the plant 59 to develop a heavy dense penetrating root system to absorb adequate water from a greater volume of soil, or to extract water within a particular volume of soil more efficiently. Although, the propensity of the root system to be highly modified by soil conditions, and the restrictions to productivity caused by the additional investment of photosynthate necessary to form and maintain a larger root system, are discouraging from relying only on this character as a criterion of selection, nevertheless, this variable and others must be taken into consideration. An explanation for the importance of the length of the vegetative cycle is that roots require assimilates from the shoot, while the shoot requires water and nutrients from the root system. A long vegetative cycle results in more total CO2 fixed during growth cycle, and this provides the required assimilates to grow a well developed root system, and therefore, a vigorous and well developed plant. During the vegetative phase, leaves and other green tissues are the original sources of assimilates, and at the same stage of development, roots, stems and leaves are competitive sinks for assimilate. The proportions of assimilate partitioned to these three organs can influence plant growth and productivity. The investment of assimilate into greater leaf area development results in greater light interception, however, the leaves also require water and nutrients, so a proportional investment in root growth is necessary. 60 Thus, extending the length of the vegetative cycle, and increasing root weight, would be the best way to strengthen vigor and development of the bean plant under water stress. However, increasing the length of the vegetative cycle without increasing the length of the complete biological cycle, would require a reduction of the reproductive cycle. This reduction would require that the plant increases the seed filling rate in order to maintain high grain production. This could be possible by selecting genotypes with the ability to produce heavy biomass and high yield efficiency. The second factor, which we called yield, identified itself with yield characters: yield efficiency, biomass at physiological maturity, and grain yield. Any factor affecting photosynthetic activity is likely to affect its total dry matter content (biomass), and within broad limits, grain production. Thus, high yielding genotypes are based on an effective exploitation of photosynthesis to translocate a great part of their heavy biomass at maturity, to grain. Increasing biomass at physiological maturity, would be the recommended way to increase grain yield under water stress. The combination of both characters requires the maintenance of a high harvest index (H.I. = Economic yield/Biological yield) if yield increase is to be achieved. The third factor, biomass at anthesis, appears important under water stress conditions, and it would be our 61 hypothesis that this importance rests in part, upon the stem serving as a temporary storage site for assimilates which may be translocated to pods and seeds during the seed filling stage. Other effects associated with this factor would, of course, be those associated with overall vigor at that stage, namely, greater leaf area for photosynthesizing, greater root size for more efficient moisture uptake, and even larger stems for the development of a more numerous array of reproductive units. During certain phases of development more assimilate is being produced than is being used in growth and development, and this excess, other than that portion lost in dark respiration or by root leakage, can be directed to storage sites. During later phases (fruiting), when current photosynthesis is not able to furnish the assimilate requirements of yield sinks, storage compounds can be remobilized and moved to active sites, such as seed development. Continued grain development under water stress, despite a drop in the photosynthetic activity of the leaves, suggests that there must follow a change in distribution of assimilates to overcome deficiencies from leaves normally supplying the grain. Most of the grain yield of severely stressed plants probably come from redistribution of existing (or storaged) plant reserves. Therefore, reduced photosynthesis caused by water stress could be compensated for by enhanced translocation 62 from the stem during the grain filling stage. The fourth factor, yield cemponent, is so called because its high loading variables indicate component compensation, in which the reduction in seed weight is associated with the increment of seed number. The components of seed yield (pods per plant, seeds per pod, and seed size) tend to be mutually compensatory, so that an advance in any one component tends to be accompanied by a negative change in another component. As a consequence, the pursuit of increased seed yield by selection for any individual yield component has usually proved to be disappointing. IRRIGATED CONDITION. IMPORTANT VARIABLES. The first factor is called vigor and development, as in water stress treatment. However, under irrigation the development of the plant is reflected by the positive loadings of days to physiological maturity, days to end bloom, and duration of seed filling, whereas plant vigor is reflected by the positive loading of weight of roots and stems at physiological maturity. This factor indicates that since there is sufficient soil moisture available, there are not water limitations to yield production, thus, the plant does not need to shorten its growth cycle. On the contrary, it seems to need to augment its reproductive and complete life cycle, to take advantage of the existing improved condition. The positive loading given by stem weight a: 63 physiological maturity suggests lack of remobilization of stored carbohydrates at this point. This possibly is due to the lack of a need to utilize the stored carbohydrates, because assimilate demand by the seeds is being satisfied by currently produced photosynthates. Only when photosynthesis is adversely affected would the seed filling process depend upon the stored assimilates and their reallocation. Factor two, yield, is composed of the same high positive loading variables that were seen in this factor under the water stress condition. The yield factor thus, remained invariant from water stress to irrigated condition. Factor three, the yield component factor, consists of the same high loading yield compensatory components, that were observed in factor four under water stress condition. Factor four expressed biomass at anthesis, which is mostly based upon the positive loading of weight of stems at beginning bloom. Factor four expressed the same concept as factor three under water stress conditions. Obviously, the biomass at anthesis factor, is more important under water stress than under the irrigated condition. Biomass at anthesis, was responsible for about 16 and 11% of the total variation after rotation, in water stress and irrigated condition, respectively. This indicates that biomass at anthesis has a lower impact on bean productivity under irrigated conditions. 64 FINAL REMARKS. Stomatal resistance and water potential measurements taken at the critical reproductive stage, presented evidence of how the bean plant might counteract the effects of water deficits. It demostrated that in a bean plant the combination of desired characters, such as sensitive stomata (allowing low hydraulic conductance), and the ability to decrease top/root ratio and increase root growth (allowing soil water availability) may promote the goal of achieving a more efficient water use by the plant. Likewise, the analyses used in this study provided useful information to determine the plant characters related to the production of dry bean. Stepwise regression analyses showed that biomass at maturity is the most important characteristic related to yield under water stress and irrigation. Factor analysis reduced the number of variables from 27 to nine under the two water conditions, and disclosed that there are three uncorrelated factors or patterns of production in dry beans. The three factors have been called vigor and development, yield, and biomass at anthesis. Thus, Factor analyses grouped traits that measured the same biological concept or function. It was found that the two water conditions did not differ much in the general concepts measured, and in the grouping of characters, except for the first factor which involved different variables 65 under each water condition. Likewise, the relationship from these factors to grain yield did not differ too much, except for the biomass at anthesis factor which is more important under stress than under irrigated condition. CONCLUSIONS We can conclude that better adapted bean cultivars, under water stress (imposed at the reproductive stage), can be developed by considering the following groups of plant characters: 1.- Vigor and deve10pment, accomplished by delayed flowering, a heavy and deeply penetrating root system, and a high root/top ratio. 2.- High yield, achieved by producing a heavy biomass at maturity. This combination results in a high harvest index. 3.- Heavy biomass at anthesis, attributable mainly to heavy stems. 4.- Sensitive stomata (able to close under drought), and the ability to increase root growth and decrease top/root ratio. Thus, our ideotype would be a high yielding cultivar able to make an efficient exploitation of photosynthesis to render a great. part of its heavy biomass at maturity as grain; this can be attained by having: a) a long vegetative cycle which allows for more prolonged photosynthetic activity providing the required assimilates to develop a heavy and deeply penetrating root system, b) a heavy biomass at anthesis, constituted mainly by a heavy stem that ensures enough carbohydrates to be used in the critical period of seed filling, C) sensitive stomata to induce water saving, 66 67 d) decreased top/root ratio and increased root growth to make soil water more available. This study has identified plant characteristics related with the production of dry beans under water stress conditions imposed at the reproductive stage of development. To make the desired combination of these characters, further research into the genetics of these identified traits would be appropriate, in order to choose the best breeding method, and design an efficient selection strategy to develop improved genotypes for water stress conditions. LITERATURE CITED LITERATURE CITED .-Adams, M.W., J. Wiersma, and J. Salazar. 1978. Differences in starch accumulation among dry bean cultivars. Crop Sci. 18:155-157. -Asana R.D., and R.N. Basu. 1963. Studies in physiological analysis of yield 6. Analysis of the effect of water stress on grain development in wheat. Indian J. Plant Physiol. 6:1-13. -Asana R.D., and A.D Saini. 1958. Studies on physiological analysis of yield IV. The influence of soil drought on grain development, photosynthesis surface and water content of wheat. Physiologia Plantarum, 11:666-674. .-Begg J.E., and N.C. Turner. 1976. Crop water deficits. Adv. Agron. 28:161-217. .-Boyer J.S. 1970. Leaf enlargment and metabolic rates in corn, soybean, and sunflower at various leaf water potentials. Plant Physiol. 46:233-235. -Boyer J.S., and H.G. McPherson. 1975. Physiology of water deficits in cereal crops. Advances in Agronomy. American Society of Agronomy (eds.). 27:1-22. .-Brix H. 1962. The effect of water stress on the rates of photosynthesis and respiration in tomato plant and loblolly pine seedlings. Physiologia Plantarum, 15:10- 20. .-Burt C., and C. Banks. 1947. A factor analysis of body measurements for british adult males. Ann. Eugen. 13:238-256. .-Catell R.B. 1965. Factor analysis: An introduction to essentials. I The purpose and underlying models. Biometrics, 21:190-210. 10.-Davies W.J. 1977. Stomatal response to water stress and light in plants grown in controlled environments and in the field. Crop Sci. 17:735-740. 11.-Denmead O.T., and R.H. Shaw. 1960. The effects of soil moisture stress at different stages of growth on the b8 69 development and yield of corn. Agron. J. 52:272-274. 12.-Downey L.A. 1971. Effect of gyspsum and drought stress on maize (Zea mays L.). I Growth, light absorption and yield. Agronomy J. 63:569-572. 13.-Draper N.R. and H. Smith. 1980. Applied regression analysis, second ed. pp., 307-311. 14.-El Nadi A.H. 1969. Water relations of beans. I Effects of water stress on growth and flowering. Expl. Agric. 5:195-207. 15.-El Nadi A.H. 1975. Water relations of beans. III Pod and seed yield of Haricot beans under different irrigation in the Sudan. Exp. Agr. 11:155-158. 16.-Fischer R.A., M. Sanchez, and J.R. Syme. 1977. Pressure chamber and air flow porometer for rapid field indication of water status and stomatal condition in wheat. Exp. Agr. 13:341-351. 17.-Gaff D.F., and N.D. Hallam. 1974. Resurrecting desiccated plants. In mechanisms of regulation of plant growth, R.L. Bieleski, A.R. Ferguson, M.M. Cresswell, eds. R. Soc. N.Z., Bull. 12:389-393. 18.-Gates C.T. 1955. The response of the young tomato plant to a brief period of water shortage. II The individual leaves. Austr. J. Biol. Sci. 8:215-230. 19.-Ghorashy S.R., J.W. Pendetlon, D.B. Peters, J.S. Boyer, and J.E. Beverlein. 1971. Internal water stress and apparent photosynthesis with soybeans differing in pubescence. Agron. J. 63:674-676. 20.-Greacen, E.L., and J.S. Oh. 1972. Physics of root growth. Nature (London) New Biol. 235:24-25. 21.-Hamblin J., and Donald C.M. 1974. The relationships between plant form, competitive ability and grain yield in a barley cross. Euphytica, 23:535-542. 22.-Hsiao T.C., and E. Acevedo. 1974. Plant response to water deficits, water use efficiency and drought resistance. Agric. Meteorol. 14:59-84. 23.-Jordan W.R., and J.T. Ritchie. 1971. Influence of soil water stress on evaporation, root absorption, and internal water status of cotton. Plant Physiol. 48:783- 788. 24.-Kramer P.J. 1969. Plant and soil water relations: a 70 modern synthesis. McGraw-Hill, New York, pp.,347-390. 25.-Lawn R.J. 1982. Response of four grain legumes to water stress in South Eastern Queensland. I Physiological response mechanisms. II Plant growth and soil water extractions patterns. III Dry matter production, yield and water use efficiency. Austr. J. Agric. Res. 33:481- 521. 26.-Ludlow M.M., and K. Ibaraki. 1979. Stomatal control of water loss in Siratro (Macroptilium atropurpureum (DC) Urb.), a tropical pasture legume. Ann. Bot. 43:639-647. 27.-May L.H., and F.L. Milthorpe. 1962. Drought resistance of crop plants. Field Crop Abstr. 15:171-179. 28.-Millar A.A., and W.R. Gardner. 1972. Effect of the soil and plant water potentials on the dry matter production of snap beans. Agronomy Journal, 64:559-562. 29.-Millar B.D., and O.T. Denmead. 1976. Water relations of wheat leaves in the field. Agron. J. 68:303-307. 30.-Moldau H. 1973. Effects of various water regimes on stomatal and mesophyll conductances of bean leaves. Photosynthetica, 7:1-7. 31.-O'Toole J.C., J.L. Ozbun, and D.H. Wallace. 1977. Photosynthetic response to water stress in Phaseolus vulgaris. Physiologia Plantarum, 40:111-114. 32.-Passioura J.B. 1972. The effect of root geometry on the yield of wheat growing on stored water. Austr. Journ. Agric. Res. 23:745-752. 33.-Passioura J.B. 1976. Physiology of grain yield in wheat growing on stored water. Austr. J. Plant Physiol. 3:559-565. 34.-Pearson R.W. 1966. In "Plant environment and efficient water use". (W.H. Pierre et al., eds.). Am. Soc. Agron. and Soil Sci. Soc. Am., pp., 45-126. 35.-Petinov N.S. 1965. The physiology of irrigated agricultural plants. Field Crop Abstr. 18:1-8. 36.-Potter J.R., and J.S. Boyer. 1973. Chloroplast response to lower leaf water potentials. Plant Physiol. 51:993- 997. 37.-Redshaw A.J., and H. Meidner. 1972. Effects of water stress on the resistance to uptake carbon dioxide in tobacco. J. Exp. Bot. 23:229-240. 71 38.-Rawson H.M., and G.A. Constable. 1980. Carbon production of sunflower cultivars in field and controlled environments. I Photosynthesis and transpiration of leaves, stems and heads. Aust. J. Plant Physiol. 7:555- 573. 39.-Robins J.S., and C.E. Domingo. 1953. Some effects of severe soil moisture deficits at specific growth stages of corn. Agron. J. 45:618-621. 40.- Robins J.S., and C.E. Domingo. 1956. Moisture deficits in relation to growth and development of dry beans. Agron. J. 48:67-70. 41.-Sionit N., and P.J. Kramer. 1977. Effect of water stress during different stages of growth of soybean. Agron. J. 69:274-278. 42.-Slatyer R.O. 1969. Physiological significance of internal water relations to crop yield. In: Physiological aspects of crop yield. Eastin et.al.(ed). pp., 53-88. 43.-Slatyer R.O. 1957. The significance of the permanent wilting percentage in studies of plant and soil water relations. Bot. Rev. 23:585-636. 44.-Slatyer R.O. 1973. Plant response to climatic factors. R.O. Slatyer (ed.) UNESCO, Paris, pp., 177-276. 45.-Smart R.E. 1974. Aspects of water relations of the grapevine (Vitig vinifera). Am. J. Enol. Vitic. 25:84- 91. 46.-Szarek S.R., and I.P. Ting. 1974. Seasonal patterns of acid metabolism and gas exchange in Opuntia basilaris. Plant Physiol. 54:76-81. 47.-Througton J.H. 1969. Plant water status and carbon dioxide exchange of cotton leaves. Austr. J. Biol. Sci. 22:289-302. 48.-Turk R.J., A.E. Hall, and C.W. Asbell. 1980. Drought adaptation of cowpea. I Influence of drought on seed yield. II Influence of drought on plant water status and relations with seed yield. Agronomy Journal, 72:413-439. 49.-Turner N.C. 1974. Stomatal response to light and water under field conditions. In mechanisms of regulation of plant growth, R.L. Bieleski, A.R. Ferguson, M.M. Cresswell, eds. R. Soc. N.Z., Bull. 12:423-432 50 51 52 53. 54 55. 56 57 58 72 .-Turner N.C., J.E. Begg, and M.L. Tonnet. 1978. Osmotic adjustment of sorghum and sunflower crops in response to water deficits and its influence on the water potential at which stomata close. Austr. J. Plant Physiol. 5:597-608. .-Van Bavel C.H.M., and P.J. Verlinden. 1956. Agricultural drought in North Carolina, N.C. Agric. Exp. St. Tech. Bull. 122. .-Veldman D.J. 1967. Fortran programming for the behavioral sciences. Holt, Rinehart, and Winston, Publ., New York, 406p. -Wardlaw I.P. 1967. The effect of water stress on translocation in relation to photosynthesis and growth. I Effect during grain development in wheat. Austr. J. Biol. Sci. 20:25-39. .-Wardlaw I.F. 1969. The effect of water stress on translocation in relation to photosynthesis and growth. II Effect during leaf development in Lolium temulentum L. Austr. J. Biol. Sci. 22:1-16. -Ward1aw I.F. 1971. The early stages of grain development in wheat: Response to water stress in single variety. Austr. J. Biol. Sci. 24:1047-1055. .-Watson D.J. 1952. The physiological basis of variation in yield. Adv. Agron. 4:101-145. .-Wilson J.R., M.M. Ludlow, M.J. Fisher, and E.D. Schulze. 1980. Adaptation to water stress of the leaf water relations of four tropical forage species. Austr. J. Plant Physiol. 7:207-220. .-Yoshida S. 1972. Physiological aspects of grain yield. Annu. Rev. Plant Physiol. 23:437-464. APPENDIX 73 TABLE A. ANALYSIS OF VARIANCE FOR STOMATAL RESISTANCE OF SIX DRY BEAN GENOTYPES GROWN UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M.S. Replication 1 7 809 Water treatment 1 117 572 Error a 1 20 925 Genotype 5 30 047 ** Interaction 5 13 543 Error b 10 8 414 ** Significant at P 0.05 TABLE B. ANALYSIS OF VARIANCE FOR WATER POTENTIAL OF SIX DRY BEAN GENOTYPES GROWN UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M.S. Replication 1 0.602 Water treatment 1 26.042 * Error a 1 0.282 Genotype 5 3.693 ** Interaction 5 5.973 *** Error b 10 1.069 * Significant at P < 0.06 ** Significant at P = 0.05 *** Significant at P 0.01 TABLE C. DROUGHT SUSCEPTIBILITY INDEX, ARITHMETIC MEAN. AND 74 GEOMETRIC GENOTYPES. MONTCALM, 1986. YIELD DIFFERENTIAL. MEAN OF 11 DRY BEAN 1.24 46 182 1.87 97 240 1 29 64 244 1.44 99 335 .64 26 211 .45 23 274 .76 28 188 -1.94** -80 274 .48 24 270 1.35 57 210 2.58 142 239 GENOTYPE 1.- C20 2.- Black Magic 3.- N81017 4.- LEF-Z-RB 5.- B76001 6.- II-900-5-M 7.- BAT-85 8.- A-195 .- N80068 10.-B82008 11.-San Juan Sel liYs/Ya *S: 14Xs/X = Drought susceptibility index. ** Negative value due to Stress yield > Control yield. TABLE D. MEAN YIELD YIELDING DRY BEAN GENOTYPES, 2 (all ) OF THE CRITERIA. MONTCALM, 1986. SELECTED THREE TOP USING FOUR DIFFERENT SELECTION Top three by drought suscep. index MEAN YIELD WATER STRESS Top three by yield differential Top three by arithmetic mean Top three by geometric mean MEAN YIELD GENOTYPES IRRIGATION A-195 267 3 II-900-5-M N80068 A-195 267.3 II-900-5-M N80068 LEF-Z-RB 301.6 II-900-5-M A-195 LEF-Z-RB 301.6 II-900-5-M A-195 ”———wm—_~w————._u—————-—————-e—e—...—..——..—._—.—_—~—~—u.__———a————...—-—~—~--——~——— 75 TABLE E. ANALYSIS OF VARIANCE FOR BIOLOGICAL YIELD AT ANTHESIS. OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M.S Replication 1 1093.010 Water treatment 1 6698.647 ** Error a 1 17.691 Genotype 10 2319.129 *** Interaction 10 262.230 Error b 20 513.201 ** Significant at P §<0.05 *** Significant at P = 0.01 TABLE F. ANALYSIS OF VARIANCE FOR BIOLOGICAL YIELD AT MATURITY. OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M S Replication 1 3391.901 Water treatment 1 103035.966 ** Error a 1 240.286 Genotype 10 20586.384 *** Interaction 10 8472.973 Error b 20 5187.691 ** Significant at P :20'05 *** Significant at P = 0.01 76 TABLE G. ANALYSIS OF VARIANCE FOR DAYS TO ANTHESIS OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M S Replication 1 0.023 Water treatment 1 1 114 * Error 3 1 0.023 Genotype 10 40.605 *** Interaction 10 0.714 Error b 20 0.323 * Significant at P g 2.09 *** Significant at P TABLE H. ANALYSIS OF VARIANCE FOR DAYS TO MATURITY OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M S Replication 1 0.091 Water treatment 1 52.364 ** Error a 1 0.091 Genotype 10 313.523 *** Interaction 10 11.414 *** Error b 20 0.141 ** Significant at P é 0.05 *** Significant at P = 0.01 77 TABLE I. ANALYSIS OF VARIANCE FOR STEM WEIGHT AT ANTHESIS, OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M S Replication 1 65.051 Water treatment 1 910.000 *** Error a 1 0.002 Genotype 10 404.163 *** Interaction 10 58.952 Error b 20 86.999 TABLE J. ANALYSIS OF VARIANCE FOR STEM WEIGHT AT MATURITY. OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. SOURCE DF M S Replication 1 151.702 Water treatment 1 3069.460 ** Error a 1 27.051 Genotype 10 1071.625 *** Interaction 10 64.833 Error b 20 108.083 —---*—--——-a—_*—-—t-—~-~—---‘--*.—. ** Significant at P <( *** Significant at P = 0.01 ll 0 O 01 78 TABLE R. ANALYSIS OF VARIANCE FOR ROOT WEIGHT AT ANTHESIS, OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. SOURCE DF M S Replication 1 130.927 Water treatment 1 107.266 ** Error a 1 0.247 Genotype 10 181.558 ** Interaction 10 80.906 Error b 20 72.786 TABLE L. ANALYSIS OF VARIANCE FOR ROOT WEIGHT AT MATURITY, OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. SOURCE DF M S Replication 1 0.001 Water treatment 1 43.204 Error a 1 6.568 Genotype 10 201.157 *** Interaction 10 5.632 Error b 20 5.531 *** Significant at P 79 TABLE M. ANALYSIS OF VARIANCE FOR TOP/ROOT RATIO AT ANTHESIS, OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. SOURCE DF M S Replication 1 0.454 Water treatment 1 1 093 Error a 1 1.273 Genotype 10 7.412 *** Interaction 10 0 857 Error b 20 1.194 TABLE N. ANALYSIS OF VARIANCE FOR TOP/ROOT RATIO AT MATURITY, OF 11 DRY BEAN GENOTYPES UNDER WATER STRESS AND IRRIGATION. MONTCALM. 1986. SOURCE DF M S Replication 1 8.836 Water treatment 1 205.457 Error a 1 26.540 Genotype 10 964.791 *** Interaction 10 91.540 *** Error b 20 15.074 80 TABLE O. SOIL WATER CONTENT (8). AT THREE STAGES OF DEVELOPMENT OF THE DRY BEAN PLANT UNDER WATER STRESS AND IRRIGATION. MONTCALM, 1986. DATE DEPTH (cm) IRRIGATION STRESS JULY 10 10 1.40 1.44 20 3.22 3.29 30 4.45 4 50 JULY 25 10 1.70 1 64 20 1.90 1 85 30 2 45 2 40 SEPTEMBER lst 10 1.76 1 50 20 3.45 3.10 30 3.95 3 22 SOIL MOISTURE. Two soil samples were taken in each replication at three stages of plant development: in July 10 (before the stress area was covered with plastic), in July 25 (at anthesis), and in September lst (at the end of bloom). The results (Table 0) indicate that in July 10, there was not difference between the irrigated and the water stress treatment. This result was expected, since the stress area was still not covered with plastic at that time. On July 25 there was a slight reduction in soil moisture under water stress condition; this reduction was more manifest in the soil samples taken on September lst.