.35.k§.§.n .mu . .1. .. . ..5. . ...........H¢._... . 3.”... .1... 57.15.55. (a \ o. 7!. :1 .57.: .L .35.... 5., s, .- 5 , $fiw...t .4. t. y . a . .3 .l 5 My .2... p... ., . .I 1.1.. 25:... 55%... 5.5 4.. . 5...! J3. Vamiunw. .. ~ . .. 4%.. . . - “u a . €n...¢..5..s.......umu.. v .3513. its}. n......n..,nnh................. .. . A» 3&3?" 1.... (1.“: 5.. - a: 2. 12...... 3...- .. ., .3. .. ....5..5... uh. {I}! 'l‘)5\03-l"¢l ruin t.v¥l. 5.1....3: :3: i; 5 0.1 {.14. ,u ) . ‘1... o: w I... Inl. 1.4 x c. .19.". 1. t: . 2.1. (4‘1 .‘ 1. >21. is... 1.951.} I, .1— . 1.1... i i!!! o .. .1! “Attila... fun..- I... i... - c. .33.». 5.1. 4...... ....-z.:.... I. . .3532}. 2:: 3 7.515 55...}? 2 3 2.5... f. . ”£53358 2 Tm” LIBRARY Michigan State University This is to certify that the thesis entitled IDENTIFICATION OF DROUGHT RESISTANCE IN LARGE SEEDED COMMON BEAN GENOTYPES presented by ESTEBAN FALCONI has been accepted towards fulfillment of the requirements for the M. S. degree in PLANT BREEDING AND GENETICS /%aQ/@/ Maj r Professor’ 3 Signature Jt/ / 5 5160 5/ Date MSU is an Affirmative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 2/05 c:/ClRC/DateDue.indd-p.15 IDENTIFICATION OF DROUGHT RESISTANCE IN LARGE SEEDED COMMON BEAN GENOTYPES By Esteban F alconi A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Plant Breeding and Genetics Program — Department of Crop and Soil Sciences 2005 ABSTRACT IDENTIFICATION OF DROUGHT RESISTANCE IN LARGE SEEDED COMMON BEAN GENOTYPES By Esteban F alconi The bean production areas in Ecuador are subject to intermittent drought and the available irrigation systems do not always provide the minimum water requirements of the bean crop resulting in yield and economic losses. Selection for drought tolerance in common bean (Phaseolus vulgaris L.) should be considered as the most practical strategy to help stabilize bean production. The objectives of this study were to: i) evaluate l6 bean genotypes and an inbred backcross line (IBL) population for drought resistance under field conditions in Michigan and Ecuador, ii) compare bean root systems in the greenhouse to identify root traits associated with superior performance under drought stress in the field. Five genotypes in the IBL population were selected based on high geometric mean (GM) yield in the field, yield under stress, seed weight, and seed quality. The selected genotypes will be further evaluated in Ecuador. Genotypes showing drought resistance and commercial traits were also selected as parents to develop new IBL populations to evaluate under Ecuadorian conditions. Low correlations were observed between GM yield and root traits measured in lm-long PVC tubes in the greenhouse. The tube methodology did permit the identification of genetic differences in root traits among genotypes grown under stress conditions. Dedication: In memory of my father, Angel F alconi iii ACKNOWLEDGMENTS I want to express my gratitude to Dr. James D. Kelly, my major advisor, for the help, understanding, and knowledge that I received from him during my stay at Michigan State University. Special thanks to my guidance committee: to Dr. Widders for the time and direction he dedicated to my project, and to Dr. Snapp and Dr. Hancock for their valuables suggestions during my research. My admiration, respect, and appreciation for all of them. Thanks to the dry bean lab, Halima E. Awale, Veronica Vallejo, Karolyn Terpstra, Ann Armenia, Evan Wright, Shitaye Moges, Norm Blakely, and Brian Long. Thanks for all the assistance, support, and good times. I wish to thank the Legume Program in Ecuador (PRONALEG) with all its members, Angel Murillo, Mary lBatallas , Nelson Mazon, Cristian Subia, Paola Estrella, Juan Ocafia, Marco Rivera, and specially to lug. Eduardo Peralta and José Pinzon for their friendship and help. Finally, I wish to thank my beloved family, my father Angel, my mother Yolanda, my brothers Santiago and Pablo, my aunt Maria Augusta, and my cousins Soledad, Johanna, and Juan José. Thank you for all the support, understanding, encouragement, andlove. iv TABLE OF CONTENTS LIST OF TABLES .................................................................................... vi LIST OF FIGURES ................................................................................. xii KEY OF ABBREVIATIONS ..................................................................... xiv INTRODUCTION .................................................................................... 1 CHAPTER 1 INTRODUCTION ......................................................................... 1 8 MATERIALS AND METHODS ........ i .................................................. 21 Plant Material and Population Development .................................. 21 Montcalm, MI 2004, 2005 ....................................................... 24 Tumbaco, Ecuador 2005 ......................................................... 26 Statistical analyses ................................................................ 27 RESULTS ................................................................................... 28 DISCUSSION .......................... ' ..................................................... 50 CONCLUSIONS ........................................................................... 53 REFERENCES ............................................................................. 55 CHAPTER 2 INTRODUCTION ......................................................................... 59 MATERIALS AND METHODS ......................................................... 61 Plant Material ...................................................................... 61 Experiment management ........................................................ 61 Irrigation management ............................................................ 63 Variables recorded ................................................................ 64 Statistical analyses ................................................................ 64 RESULTS ................................................................................... 66 DISCUSSION .............................................................................. 94 CONCLUSIONS ......................................................................... 104 REFERENCES ............................................................................ 106 APPENDIX A: DATA TABLES FROM DROUGHT EXPERIMENTS CONDUCTED IN THE FIELD IN MICHIGAN AND ECUADOR .......................................... 102 APPENDIX B: DATA TABLES FROM DROUGHT EXPERIMENTS CONDUCTED IN THE GREENHOUSE IN MICHIGAN ..................................................... 109 LIST OF TABLES Table 1. Characteristics of 30 Genotypes evaluated for drought in Michigan, US. and Tumbaco, Ecuador. 2004-2005 ........................................................ Table 2. Genotypes from the standard test evaluated in Tumbaco, Ecuador. 2005 ............................................................................................. Table 3. Geometric mean yield, yield under stress conditions, yield under non- stress conditions, and stress susceptibility index of the IBL population evaluated in Michigan, US. 2004 ........................................................................ Table 4. Analysis of variance for the IBLs experiment for yield, 100 seed weight (g), and harvest index under non-stress and stress conditions of the IBLs evaluated in Michigan, US. 2004 ............................................................ Table 5. Yield, biomass, 100 seeds weight, harvest index (HI), number of pods per plant, and number of seeds per pod across treatments of the IBL population evaluated in Montcalm, US. 2004 ........................................................... Table 6. Geometric mean yield, yield under stress conditions, and yield under non-stress conditions of the IBL population evaluated in Michigan, US. 2005 ...... Table 7. Yield, biomass, harvest Index (HI), common bacterial blight reaction, and desirability score across treatments of the IBL population evaluated in Michigan, US. 2005 ........................................................................... Table 8. Analysis of variance for yield (kg/ha) for 30 genotypes grown under stress and non-stress conditions over two years in Montcalm, MI 2004, 2005 ....... Table 9. Geometric mean yield, yield under non-stress conditions, and yield under stress conditions of the IBL population evaluated in Tumbaco, Ecuador. 2005. Table 10. Yield, biomass, harvest index (HI), 100 seeds weight, number of lateral roots per plant, number of pods per plant, and number of seeds per pod across water treatments of the IBL population evaluated in Tumbaco, Ecuador. 2005. Table 11. Geometric mean yield, yield under stress conditions, yield under non- stress conditions, and drought susceptibility index of the 16 genotypes in SGT in Tumbaco, Ecuador. 2005 ..................................................................... Table 12. Yield, biomass, harvest index (HI), 100 seeds weight, plant height, stem thickness, number of lateral roots, number of pods per plant, and number of seeds per plant across treatments of 16 genotypes at the SGT in Tumbaco, Ecuador. 2005 ............................................................................................. vi 22 24 30 32 34 36 38 39 41 44 46 48 Table 13. Characteristics of the common bean genotypes in experiments IBLl and IBL2 used to conduct the root study. Michigan State University. 2004 ............... Table 14. Characteristics of the common bean genotypes in SGT experiment used to conduct the root study. Michigan State University. 2004 ............................. Table 15. Amount of water/solution applied, date of the experiment, mean temperature, and relative humidity registered in the root study. Michigan State University. MI 2004 ........................................................................... Table 16. Root length and percentage of the total root length in categories A (0.0 — 0.5mm), B (0.5 — 1.0 mm), C (1.0 — 2.0 mm), and D (> 2.0 mm) under water stress and non stress conditions in two different depths of the PVC tube and in the whole PVC-tube in SGT experiment. Michigan State University. MI 2004 .......... Table 17. Total root length (TRL), length of the taproot (LTR), number of the lateral roots (No of LR), projected root area (PA), root volume (RV), root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 genotypes in STG experiment grown in greenhouse in Michigan State University, MI 2004 ............................ ‘ .............................................................. Table 18. Root length and Root density at two different depths of the PVC-tube in IBLl experiment ............................................................................... Table 19. Total root length, root length above 0.3 m of the PVC-tube, root length below 0.3 m of the PVC-tube, percentage of root distribution per category, percentage of root distribution per category above 0.3 m, and percentage of root distribution per category below 0.3 m under stress and non-stress conditions for 16 genotypes in IBLl experiment grown in greenhouse in Michigan State University, MI 2004 ........................................................................... Table 20. Total root length (TRL), length of the taproot (LTR), number of the lateral roots (No of LR), root weight (RW), surface area (SA), projected area (PA), and root volume (RV) for 16 genotypes in IBLl experiment in Michigan State University, MI 2004 .................................................................... Table 21. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 genotypes in IBLl experiment in Michigan State University, MI 2004 ....................................................... Table 22. Root length and percentage of root length of 16 genotypes in four root categories A (0.0 — 0.5mm), B (0.5 — 1.0 mm), C (1.0 — 2.0 mm), and D (> 2.0 mm) under stress and non-stress conditions in two different regions of the PVC- tubes in IBL2 experiment in Michigan State University, MI 2004 ..................... vii 62 63 64 67 76 81 83 86 86 91 Table 23. Total root length (TRL), length of the taproot (LTR), number of the lateral roots (No of LR) , root weight (RW), surface area (SA), projected area (PA), and root volume (RV) for 16 genotypes in IBL2 experiment in Michigan State University, MI 2004 .................................................................... Table 24. Root length in categories A (RL (A)), B (RL (B)), C (RL(C)), and D (RL (D)) for 16 genotypes in IBL2 experiment in Michigan State University, M1. 2004 ............................................................................................. Table 25. Percentage of reduction of four root categories (A-D) of SGT experiment under stress conditions related with the same category under non- stress conditions. MI 2004 ................................................................... Table 26. Yield, biomass, harvest index (H1), 100 seed weight, number of pods per plot, number of seeds per pod, and stem thickness of the 26 IBLs evaluated under non-stress conditions in Michigan, US. 2004 ...................................... Table 27. Yield, biomass, harvest index (HI), flowering, maturity, common bacterial blight reaction (CBB), and desirability score D8 of 26 IBL evaluated under stress and non-stress conditions in Michigan, US. 2004 ......................... Table 28. Yield, biomass, harvest index (H1), 100 seed weight, number of pods per plot, number of seeds per pod under, and number of lateral roots under non- stress and stress conditions of the IBLs evaluated in Tumbaco, Ecuador. 2005 ............................................................................................. Table 29. Stem thickness, plant canopy height, days to harvest, and root rot score in a scale 1 to 9 under non-stress and stress conditions of the IBLs evaluated in Tumbaco, Ecuador. 2005 ..................................................................... Table 30. Yield, biomass, harvest index (H1), 100 seed weight, number of pods per plot under non-stress and stress conditions of the SGT evaluated in Tumbaco, Ecuador. 2005 ................................................................................. Table 31. Stem thickness, number of seeds per pod, number of pods per plant, plant height, days to harvest, and root rot score under non-stress and stress conditions of the SGT evaluated in Tumbaco, Ecuador. 2005 .......................... Table 32. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes above 0.3 m of the PVC-tube in STG experiment in Michigan State University, MI 2004 ................ viii 93 93 94 109 110 111 112 113 114 116 Table 33. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), root length in category A (RL (A)), root length in category B (RL (B)), root length in category C (RL(C)), and root length in category D (RL (D)) for16 bean genotypes below 0.3 m of the PVC-tube in STG experiment in Michigan State University, MI 2004 ....................................................... Table 34. Total root length (TRL), length of the taproot (LTR), number of the lateral roots (No of LR), projected area (PA), root volume (RV) for 16 bean genotypes under two water treatments in STG experiment in Michigan State University, MI 2004 ........................................................................... Table 35. Root length in category A (RL (A)), root length in category B (RL (B)), root length in category C (RL(C)), and root length in category D (RL (D)) for 16 bean genotypes under two water treatments in STG experiment in Michigan State University, MI 2004 ........................................................................... Table 36. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), and root diameter average (DA) for 16 bean genotypes above 0.3 m of the PVC-tube under two water treatments in STG experiment in Michigan State University, MI 2004 ........................................................................... Table 37. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 genotypes above 0.3 m of the PVC-tube under two water treatments in STG experiment in Michigan State University, MI 2004 ............................................................................................. Table 38. Total root length (TRL), surface area (SA), projected area (PA), and root volume (RV) for 16 bean genotypes below 0.3 m of the PVC-tube under two water treatments in STG experiment in Michigan State University, MI 2004 ........ Table 39. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes at below 0.3 m of the PVC-tube under two water treatments in STG experiment in Michigan State University, MI 2004 ........................................................................... Table 40. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), root diameter average (DA), root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes above 0.3 m of the PVC-tube in IBLl experiment in Michigan State University, MI 2004 Table 41. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), root diameter average (DA), root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes below 0.3 m of the PVC-tube in IBLl experiment in Michigan State University, MI 2004 ........................................................................... ix 116 117 117 118 119 119 120 121 122 Table 42. Length of the taproot (LTR), number of the lateral roots (No of LR) , root weight (RW), total root length (TRL), projected area (PA), root volume (RV) for 16 bean genotypes under two water treatments in IBLl experiment in Michigan State University, MI 2004 ....................................................... Table 43. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes under two water treatments in IBLl experiment in Michigan State University, MI 2004 ............... Table 44. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), and root diameter average (DA) for 16 bean genotypes above 0.3 m of the PVC-tube under two water treatments in IBLl experiment in Michigan State University, MI 2004 .................................................................... Table 45. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes above 0.3 m of the PVC-tube under two water treatments in IBL] experiment in Michigan State University, MI 2004 ........................................................................... Table 46. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), and diameter average (DA) for 16 genotypes below 0.3 m of the PVC-tube under two water treatments in IBLl experiment in Michigan State University, MI 2004 ........................................................................... Table 47. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes below 0.3 m of the PVC-tube under two water treatments in IBLl experiment in Michigan State University, MI 2004 ........................................................................... Table 48. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), diameter average (DA) for 16 bean genotypes above 0.3 m of the PVC-tube in IBL2 experiment in Michigan State University, MI 2004 ................ Table 49. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes above 0.3 m of the PVC-tube in IBL2 experiment in Michigan State University, MI 2004 ............... Table 50. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), and root diameter average (DA) for 16 bean genotypes below 0.3 m of the PVC-tube in IBL2 experiment in Michigan State University, MI 2004 ....... Table 51. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), category D (RL (D)) for 16 bean genotypes below 0.3 m of the PVC- tube in IBL2 experiment in Michigan State University, MI 2004 ...................... 123 124 125 126 127 128 129 129 130 130 Table 52. Length of the taproot (LTR), number of the lateral roots (No of LR) , total root length (TRL), projected area (PA), root volume (RV) for genotypes under two water treatments in IBL2 experiment in Michigan State University, MI 2004 ............................................................................................. Table 53. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes under two water treatments in IBL2 experiment in Michigan State University, MI 2004 ............... Table 54. Total root length (TRL), surface area (SA), projected area (PA), root volume (RV), and root diameter average (DA) for 16 bean genotypes above 0.3 m of the PVC-tube under two water treatments in IBL2 experiment in Michigan State University, MI 2004 .................................................................... Table 55. Root length in categories A (RL (A)), B (RL (B)), C (RL(C)), and D (RL (D)) for 16 bean genotypes above 0.3 m of the PVC-tube under two water treatments in IBL2 experiment in Michigan State University, MI 2004 ............... Table 56. Total root length (TRL), surface area (SA), projected area (PA),. root volume (RV), and diameter average (DA) for 16 bean genotypes below 0.3 m of the PVC-tube under two water treatments in IBL2 experiment in Michigan State University, MI 2004 ........................................................................... Table 57. Root length in category A (RL (A)), category B (RL (B)), category C (RL(C)), and category D (RL (D)) for 16 bean genotypes below 0.3 m of the PVC-tube under two water treatments in IBL2 experiment in Michigan State University, MI 2004 ........................................................................... xi 131 132 132 133 I33 134 LIST OF FIGURES Figure 1. Annual rainfall distribution in Chota’s and Mira valley. Meteorological Station of Salinas, Imbabura, Ecuador. (1982 — 2002) .................................... Figure 2. Total root length of the 16 bean genotypes in the SGT experiment ......... Figure 3. Root length in four root categories under two water treatments in SGT experiment ....................................................................................... Figure 4. Percentage of the total root length in four root categories under two water treatments in SGT experiment ................................................................ Figure 5. Root length of the 16 bean genotypes in category C (1.0 — 2.0 mm), under stress and non-stress conditions in the SGT experiment .................................. Figure 6. Taproot length of the 16 bean genotypes under two water treatments in SGT experiment ................................................................................. Figure 7. Number of lateral roots of the genotypes in SGT ............................... Figure 8. Projected root area (cm2) of the 16 bean genotypes in SGT experiment. . .. Figure 9. Projected root area (cm2) of the 16 bean genotypes below 0.3 m of the PVC-tube in SGT experiment ................................................................. Figure 10. Projected root area (cm2) of the 16 bean genotypes in SGT experiment above 0.3 m of the PVC-tube .................................................................. Figure 11. Root volume (cm3) of the 16 bean genotypes in SGT ........................ Figure 12. Total root length average of two treatments of 16 bean genotypes in IBLl experiment in Michigan State University ............................................. Figure 13. Total root length of 16 bean genotypes under stress and non-stress conditions in IBLl experiment in Michigan State University ............................ Figure 14. Root length of 16 bean genotypes in category A in IBLl experiment in Michigan State University ..................................................................... Figure 15. Root length of 16 bean genotypes in category B in IBLl experiment in Michigan State University ..................................................................... Figure 16. Root length of 16 bean genotypes in category C in IBLl experiment in Michigan State University ..................................................................... xii 20 68 68 69 70 71 72 73 73 74 75 77 78 78 79 79 Figure 17. Root length of 16 bean genotypes in category D in IBLl experiment in Michigan State University ..................................................................... Figure 18. Root length of IBLl experiment per category under stress and non-stress conditions ........................................................................................ Figure 19. Percentage of root length per root category in IBLl experiment ............ Figure 20. Percentage of root length per root category in IBL2 experiment in Michigan State University ..................................................................... Figure 21. Root length of IBL2 experiment per category under stress and non-stress conditions in Michigan State University ..................................................... Figure 22. Root length of genotypes in IBL2 experiment in four different root categories under two water treatments in Michigan State University ................... Figure 23. Temperature and Relative humidity in the greenhouse from Watchdog 3529 in STG experiment ....................................................................... Figure 24. Temperature and Relative humidity in the greenhouse from Watchdog 3529 in IBLl experiment ...................................................................... Figure 25. Temperature and Relative humidity in the greenhouse from Watchdog 3529 in IBL2 experiment ...................................................................... xiii 80 82 82 89 90 90 100 100 AM ANOVA CBB CIAT CV dap DII DS GM HI IBL INIAP LSD MAS masl NSL PRONALEG QTL , RCBD RIL SSD SSI Xp Xs Yp Ys KEY OF ABBREVIATIONS Arithmethic Mean Analysis of Variance Common Bacterial Blight International Center for Tropical Agriculture Coeficient of Variance Days after planting Drought Intensity Index Desirability Score Geometric Mean Harvest Index Inbred Backcross Line National Institute of Agricultural Research - Ecuador Least Significant Difference Marker Assisted Selection Meters Above Sea Level Negro San Luis National Legume Program of INIAP-Ecuador Quantitative Trait Loci ' Pearson Correlation Coefficient Randomized Complete Block Design Recombinant Inbred Line Single Seed Descent Stress Susceptibility Index Mean yield under non-stress conditions Mean yield under moistures stress conditions Yield under non-stress conditions Yield under moisture stress conditions xiv INTRODUCTION From an agricultural perspective, drought is a condition in which water supply is insufficient to meet the needs of the crop. As a result of the reduction in soil moisture, plants suffer water stress and yield is reduced (Subbarao et al., 1995). Two different kinds of drought, intermittent and terminal, can be distinguished. Intermittent drought is due to climatic patterns of sporadic rainfall that cause intervals of drought and can occur at any time during the growing season (Schneider et al., 1997). A similar effect occurs when farmers have the option of irrigation, but water supply is limited. In contrast, terminal drought occurs when plants suffer lack of water during the later stages of development, mainly during reproductive growth (Frahm et al., 2004; Acosta-Gallegos and Kohashi-Shibata, 1989). An example of terminal drought occurs in Central America when common bean is planted toward the end of the first rainy season and water supply is insufficient to support yield. Lack of water interferes especially with the normal metabolism of plants during flowering and pod-fill, since these are the stages when water deficits cause the greatest yield reduction (Halterlein, 1983; Thung and Rao, 1999). According to Singh (1992) the water requirements for a bean crop is at least 400 mm. In semi-arid production areas that lack adequate amounts of water (< 400 mm) (Thung and Rao, 1999), and have sandy soils with low organic matter content, water holding capacity is limited, so bean yields are reduced further (Acosta and Adams, 1991). Furthermore, drought is intensified by other factors such as high temperatures, presence of root pathogens, and low soil fertility (Singh and White, 1988). Plant adaptation strategies to drought. Drought adaptation is the ability of plants to thrive and produce more biomass and seed, compared with non-adapted plants growing in the same water-limited environment (Hall, 1993). This adaptation is the result of an evolutionary process that allows specific genotypes to survive in areas with low precipitation. Common bean and plants in general have developed different adaptation strategies in response to the environmental conditions where they grow. These strategies allow plants to be more competitive in terms of water use, utilization of light and/or nutrients, and successful in terms of producing progeny under adverse conditions. In the specific case of environments with water deficits, plants have developed different strategies to perform better than other cultivars or different species. These strategies, according to Ludlow (1989), are recognized as escape, avoidance, and tolerance mechanisms. Escape strategy. Plants exhibiting this strategy are able to complete their life cycle in shorter periods of time by taking advantage of the available water. With seasonally limited amounts of water, plants maximize water use efficiency and complete seed production early. Ludlow (1989) describes the important characteristics of plants that base their survival on escape as rapid germination after rainfall, fast growth, early flowering and seed production before the water supply is exhausted. Some annual crops also use this strategy. One example in common bean is the cultivar “Bola 60”, which has a shorter vegetative and reproductive cycle than other common bean cultivars planted in the Andean region. However, this strategy is usually associated with low yields in most cultivars because the shorter vegetative growth period prevents maximum seed production. Subbarao et al. (1995) reported that earliness reduces the potential yield by reducing dry matter before flowering and the number of sites for post-anthesis seed filling. Reduction of potential yield is the penalty for individual early-maturing plants. However, an increase in plant density can be considered as an approach to compensate for the limited productivity of individual plants. Certainly, more plants will be competing for reduced amounts of water, but the efficiency of crop utilization of water may be improved, since less soil surface is available for evaporation. This strategy could be applied to certain crops and deserves further examination. Ludlow (1989) also suggested that some crops, such as pearl millet, show developmental plasticity in dry areas. Plasticity, as defined by Acosta-Gallegos and White (1991), is an adaptative feature to highly variable rainfall at the beginning and during the rainy season in semi-arid regions. Adapted crops have the ability to flower and produce seed after short periods of rain. Acosta—Gallegos and White (1995) identified common bean genotypes such as ‘Pinto Villa’ from the Mexican highlands that showed phenological plasticity. Avoidance strategy. Plants that use avoidance strategies maximize water uptake or minimize water loss by different mechanisms (Acosta-Gallegos et al., 1996). Traits associated with avoidance include the development of deep tap roots, stomatal regulation, compact canopies, small leaves, paraheliotropic leaf movements, and thick cuticles. One of the most important water stress avoidance mechanism is deep rooting (White and Castillo, 1989). This allows plants to reach water at depth. This feature has been widely studied and there is evidence that differences exist among common bean genotypes for this trait (Yabba and Foster, 1997). Incorporating this important characteristic into commercial common bean cultivars through breeding appears promising, however, the energy that the plant is investing in the developing deep roots could negatively affect yield. Hence, breeders should consider developing root architecture ideotypes able to reach deeper soil layers without losing the nutrients in the top soil and without wasting energy that should go to seed production. Stomatal regulation is another feature used by plants to maximize water use efficiency. Once the stomata are open, transpiration occurs and photosynthesis takes place. As a result of photosynthesis, plants accumulate biomass. The risk is that plants can experience excessive water loss during the day and suffer water stress. To prevent the consequences of the stress, plants should maintain internal plant water status above critical threshold levels (Subbarao et al., 1995). Selecting genotypes able to control transpiration in an efficient manner would be useful, provided that adequate variability exists for this trait in common bean germplasm. Aguirre et a1. (2002) reported that bean genotypes with larger stomatal index in the abaxial than in the adaxial surface exhibited less gas exchange compared with the variety Bayo Madero, which had the same stomatal index on both leaf surfaces. This study suggests that transpiration could be regulated when fewer stomata are present in the abaxial region. Transpiration could be regulated without reducing yield through biochemical control as well. Itai and Bimbaum (1991) reported that plant hormones production under stress conditions are related with the increase in stomata resistance. Aguirre (1999) studied stomatal response to stress in common bean using a split-root system. This experiment demonstrated that signals originating in the roots under water stress controlled the stomatal aperture in the leaves. The system appears useful for conducting studies designed to select more efficient common bean genotypes with optimum regulation of stomatal opening. Only one cultivar, Cacahuate 72, was used in this study and more bean genotypes would need to be studied to investigate if genetic differences exist among genotypes. Although White and Castillo (1989) reported that root characteristics of common bean are the major factors responsible for plant response to drought, shoot traits have also been described that confer avoidance capabilities when water is limiting. These characteristics include leaf pubescence, small leaves, a thick cuticle layer, number of stomata, and paraheliotropic movements of the leaves (Aguirre-Medina et al., 2002; Berg and Hsiao, 1986; and Ludlow, 1989). Characteristics such as leaf size, shape and thickness deserve further study, but may be linked to seed size. Large genetic variability in leaves can be found among the common bean germplasm. Characteristics such as leaf pubescence and thick cuticle may not affect yield to the same degree as other traits that demand larger amounts of energy. Tolerance. This strategy refers to the ability of the plant to maintain metabolic activity under low water availability. Kohashi-Shibata et a1. (2002) demonstrated that the cultivar Pinto Villa, identified as drought resistant, is less affected by water stress, and that leaves of Pinto Villa continue to grow faster than those of cultivar Bayo Madero under the same stress conditions. Certain compounds are responsible for maintenance of cell membrane integrity and cell proteins during stress periods. These compounds include the amino acids proline, betaine and glycine (Showalter, 1993). Osmotic adjustment is an additional characteristic that allows certain genotypes to retain water and thus avoid dehydration. When water availability is low, osmotic adjustment aids in the maintenance of turgor by producing organic compounds in the cell that facilitate the uptake of water through differences in osmotic potential. The plant is able to continue normal functions such as carbon acquisition through open stomata, and root growth under stress. Eventually, if conditions allow, the roots will reach deeper soil layers where more moisture is available. According to Ludlow (1989), no particular yield reductions have been identified in plants with tolerance, however, this strategy does not necessarily allow more carbon fixation than the avoidance strategy. Breeding for drought resistance.- Important drought resistance sources of common bean have been identified in races Durango and Mesoamerica germplasm (Singh, 1995). Unfortunately, single crosses with large-seeded Andean races would result in genotypes with phenotypes lacking commercial characteristics. The Ecuadorian market demands large seeds that differ from the majority of Mesoamerican seed classes. The Inbred backcross has been suggested as a breeding method to recover commercial plant and seed characteristics in crosses between Mesoamerican and Andean beans (Beaver, 1999). Bliss (1993) proposed “The Inbred Backcross Line Method of Breeding” to develop populations with genotypes possessing genes from promising donors in a background of well adapted germplasm. The method seems practical to develop new drought resistance bean cultivars for Ecuador, because the materials obtained, after two or more backcrosses, are phenotypically similar to the recurrent parent while maintaining sufficient variability for the trait being improved since no selection is applied until obtain homozygous lines in the population. Interracial populations should result in progress toward obtaining new cultivars with high levels of drought resistance in Andean plant and seed phenotypes. Evaluation in the field.- Several studies have been conducted to select for drought resistance in common bean (Frahm et al., 2004, Schneider et al., 1997b, and Singh, 1995). Direct selection in the stress environment characterized by low rainfall is the most common method of selection for drought resistant genotypes. This method consists of selection of the highest yielding genotypes subjected to water stress. The method is considered highly accurate since performance is measured in the stress environment, however, this method exhibits some limitations. One difficulty is the large numbers of accessions that need to be evaluated in the field for drought resistance. If the parameter for selection is yield, the plots must be large enough to get accurate information. In addition, drought periods can be erratic in some locations, so entire tests could be lost when dry periods do not occur. This problem slows breeding progress since the differences in yield of the genotypes evaluated under stress conditions could be masked by seasonal and location interactions (van Ginkel et al., 1998). Ramirez-Vallejo and Kelly (1998) conducted a drought study over two years (1988 and 1990). In 1988 the authors observed significant differences between genotypes in the traits studied, whereas in 1990 no significant differences were observed among treatments. One difficulty in detecting genotypic differences result from variable soil type and moisture conditions that that may affect the results. Moreover, since yield is the selection criteria, yield under stress might be also a function of other cultural management practices that may be interacting with drought stress. For these reasons, spatial and temporal limitations are problems that bean breeders face in direct screening for drought tolerance. In the field, the most effective method to select drought resistant common bean genotypes is evaluating for yield under stress and non stress conditions. These trials provide information on both potential yield and yield under stress. To estimate the intensity of the stress in individual experiments, the drought intensity index can be calculated (DII) (Ramirez-Vallejo and Kelly, 1998). DII= (l-Xs/Xp), where X5 is the experiment mean under stress conditions, and Xp is the experiment mean under non- stress conditions. Additionally, the D11 can be used to compare the stress imposed between two or more experiments conducted in different years or locations. Using data on yield under non-stress and water stress, Ramirez-Vallejo and Kelly (1998) suggested that the most effective selection for drought resistance is based first on those genotypes with high geometric mean (GM) yield followed by the selection of the genotypes with high-yielding individuals with low to moderate levels of drought susceptibility index. Schneider et al. (1997) used a similar breeding strategy, selecting first on high a GM yield followed by selection on yield in stress environment. Geometric mean yield is calculated as the square root of the product of yield under stress and non—stress. (GM= (prYs)'/2), where, Yp is the potential yield and Y5 is the yield under stress conditions of each genotype. Geometric mean yield identifies genotypes with high yield under stress and non-stress conditions without the influence of extreme values since the result is normalized through the use of the square root. The use of the GM yield requires two treatments, one under normal conditions with supplemental irrigation (potential yield) and other under stress conditions. Other selection criteria have been proposed such as the selection based on the mean yield [Yx = (Yp + Ys)/2], or the drought susceptibility index [DSI = [(1 — (Ys/Yp)]/DII] (Fisher and Maurer, 1978). However, Fernandez (1993) showed that mean yield favored the genotypes with high yield potential under non-stress and D81 failed to differentiate drought tolerant genotypes with high and low yield potential. Indirect evaluations.- Field evaluations under local conditions in production areas are the most accurate source of information on genotypic performance. However, the unpredictable nature of the rainfall has forced breeders to consider alternative ways to evaluate genotypes and, under controlled environments, select for genotypes showing drought resistance. These approaches attempt to identify traits that could confer the drought resistance in plants and reduce the impact of drought on plant performance (Ludlow and Muchow, 1990, van Ginkel et al., 1998). One plant organ with the greatest influence on drought resistance in common bean is the root system. Root parameters, which are highly correlated with drought resistance (White and Castillo, 1989), have been the subject of many studies. Yabba and Foster (1997) used growth pouches in a growth chamber to study root traits in eight common bean genotypes with different performance under drought stress. In this study, the drought resistant genotype BAT 477 produced less number of lateral roots compared with the other genotypes under study, but a deeper taproot helped to avoid the effect of water stress. The conclusion was supported by the findings of Gregory (1994), who showed that BAT 477 has a deeper taproot than susceptible genotypes under stress conditions. Using the pouch methodology, F rahm et a1. (2003) evaluated a population of 81 recombinant inbred lines (RILs), but did not find significant differences between drought resistant and susceptible RILs. Probably, the pouch method did not allow sufficient time for root growth to develop significant differences between genotypes. An indirect method to study roots and the effects under both stress and non-stress conditions in alfalfa (Medicago sativa L.), was developed by Pennypacker et a1. (1990). Plants were grown in 0.9 m tall by 0.2 m diameter containers. Using tensiometers to monitor and control water availability, the authors were able to impose a gradual drought stress and at the same time collect information on different plant grth parameters. Selection for deep taproots under artificial conditions appeared to be a promising approach to identify drought resistant genotypes. Caution should be exercised when root parameters. Kramer and Boyer (1995) documented how environmental factors (soil texture, structure, aeration, temperature, competition, and mineral content) can affect root growth. Roots are highly plastic, and morphology can change significantly depending on the substrate and the container used. Such factors could explain the frequent observance low correlations between field and greenhouse results. Researchers must recognize these factors and develop screening systems that are highly correlated with field performance of common beans. Ogbonnaya et a1. (2003) used a hydroponics method to select cowpea genotypes for drought resistance. The authors found significant correlations between yield and biomass production, shoot-root ratio, and root growth under well-watered conditions. However, under water stress conditions there were no correlations between these variables. Apparently, the hydroponic low oxygen environment altered normal root growth morphology and distribution. The use of molecular markers is another indirect option that could be applied in the selection of drought resistant common bean genotypes. Schneider et al. ( 1997b) found correlations between molecular markers and drought resistance based on high geometric means, for two RILs in Michigan, however, there were no significant correlation between the geometric mean and the markers were not detected in the same population tested in Mexico. Frahm et a1. (2003) screened the same markers in a black-seeded bean RIL 10 population but obtained low correlation with yield under stress. A probable reason for low correlation could be that the mechanisms of drought resistance linked to the marker are only effective in specific environments. Molecular markers that explain a large portion of the phenotypic variation that confers drought resistance in a broad range of semi-arid environments will become a powerful selection tool. The effectiveness of marker assisted selection (MAS) is inversely proportional to the heritability of the trait under consideration (Paterson et al., 1990). Therefore, MAS could facilitate the selection of traits such drought resistance with low to moderate levels of heritability (h2 = 0.19 to 0.59; Scheider et al., 1997b; h2 = 0.09 to 0.80; Acosta-Gallegos et al., 1996). Babu et al. (2003) observed a QTL in rice which explained a high percentage of the variability in yield under stress conditions. This region of the genome was associated with root traits related to drought resistance in rice. Identification of QTLs is a challenge since the QTLs may be affected by the environment (Collard et al., 2005) since complex analyses and effort are required to identify them (Staub et al., 1996). Once QTL are identified the efficiency of MAS can be greater than through direct selection. To date, there are no new drought resistant cultivars developed using molecular techniques, and more research is needed to enable breeders to effectively utilize molecular approaches. Tuberosa et a1. (2002) is optimistic when he stated that QTL analysis that detects morphological and physiological traits related with the adaptation to drought conditions can be integrated in plant breeding programs to develop improved cultivars. More traits indirectly correlated to drought resistance should be used to select putative resistant genotypes. Ramirez-Vallejo and Kelly (1998) studied yield 11 components, biomass and partitioning traits under stress and non stress conditions. Harvest index (HI) measurement was considered due to the high correlation with yield, but was thought to be no more useful than yield data since H1 is a product of yield influenced by the environment. The authors also found a significant correlation between stem diameter and biomass traits (r=0.707**). Data were consistent being useful to select putative drought resistant genotypes in future screening tests. The stem diameter can be used without difficulty since it is non destructive and easy to collect in the field. More research is needed to confirm the correlation between stern diameter and drought resistance to generalize the statement for different bean classes. In greenhouses, several methods have been evaluated and developed. Such methods requiring labor intensive evaluations have not become part of routine breeding programs due to the low correlation with high yield under stress conditions in the field. A practical method, able to overcome the problems observed under field evaluations due to the uncontrolled environmental factors, would help improve efficiency of bean breeders in developing new cultivars with drought resistance. 12 CONCLUSIONS From the bean breeding perspective, the selection of drought resistant genotypes should be based on performance under stress and non-stress environments, instead of the selection of genotypes showing tolerance per se. Seed production must be the fundamental component to consider. Breeders could focus on yield trials exclusively, or indirectly selecting traits that confer drought resistance followed by field selection. High and stable yields are the ultimate goal. Additionally, breeders who concentrate on selecting traits that confer drought resistance should recognize that the trait must be effective for the specific environment where the new cultivars are to be grown. The selection of the parents plays an important role, since breeders must take advantage of the opportunities offered by genotypes adapted for dry environments. Utilizing such germplasm will increase the genetic base of Andean beans by introducing novel traits and genes to enhance the current cultivars or to develop new genotypes with superior performance. Such strategies represent a challenge due to the genetic barriers, genetic distance, and lack of commercial traits, but it may provide opportunities. Finally, bean breeders should emphasize work with indirect screening methods and molecular markers. Even though, progress to date has been limited in improving drought resistance, breeders must recognize that new techniques and greater knowledge will be generated and could have potential value in future improvement programs for drought resistance in common bean. l3 REFERENCES Acosta-Gallegos, J. A., and J. Kohashi-Shibata. 1989. Effect of water stress on growth and yieldin indeterminate dry-bean (Phaseolus vulgaris L.). Field Crop Research. 20: 81-93. Acosta-Gallegos, J. A., and M. W. Adams. 1991. Plant traits and yield stability of dry bean (Phaseolus vulgaris) cultivars under drought stress. Journal of Agricultural Science, Cambridge. 1 17:213-219. Acosta-Gallegos, J. A., and J. W. White. 1995. Phenological plasticity as an adaptation by common bean to rainfed environments. Crop Science. 35: 199-204. Acosta-Gallegos, J. R. Rosales-Sema, and J. W. White. 1996. Logros, problemas y posibilidades del mejoramiento genetico del frijol Mesoamericano y de las Ecoregiones semiaridaes de America Latina. In: Singh, S and O. Voysest. Taller de mejoramiento de frijol para el siglo XXI. Cali Colombia. October 23 — November 1. p. 405-414. Aguirre-Medina, J. F., J. Kohashi-Shibata, C. Trejo-Lopez, and J. Acosta-Gallegos. I999. Respuesta fisiologica del frijol a la sequia, en un sistema de raiz dividida. Agronomia Mesoamericana. 10(1):31-36. Aguirre-Medina, J. F., J. A. Acosta-Gallegos, L. M. Ruiz-Posadas, J. Kohashi-Shibata, and C. Trejo-Lopez. 2002. Diferencias morfologicas de la epidermis de la hoja de frijol comt'rn y se relacion con la tolerancia a sequia. Agricultura Técnica en México. 28(1):53-64. Babu, R. Ch., B. D. Nguyen, V. Chamarerk, P. Shanmugasundaram, P. Chezhian, P. Jeyaprakash, S. K. Ganesh, A. Palchamy, S. Sadasivam, L. J. Wade, and H. T. Nguyen. 2003. Genetic Analysis of drought resistance in rice by molecular markers: Association between secondary traits and field performance. Crop Science. 43: 1457- 1469. Beaver, J. S. 1999. Improvement of large-seeded race Nueva Granada. 275-288. In: Singh S. P. Common bean improvement in the twenty-first century. Kluwer Academic Publisher. Dordrecht. The Netherlands. Berg, V. and T. Hsiao. 1986.Solar tracking : light avoidance induced by water stress in leaves of kidney bean seedlings in the field. Crop Science. 26:980-986. Bliss, F. A. 1993. Breeding common bean for improved biological nitrogen fixation. Plant Soil. 152271-79. l4 Collard, B. C., M. Z. Jahufer, J. B. Brouwer, and E. C. Pang. (2005). An introduction to markers, quantitative trait loci (QTL) mapping, and marker-assisted selection for crop improvement: The basic concepts. Euphytica. 142: 169-196. Fernandez, G.C.J. 1993. Effective selection criteria fos assessing plant stress tolerance. In: Kuo, G. Adaptation of food crops to temperature and water stress. Proceedings of an international symposium. Taiwan, 13 — 18 August 1992. AVRDC. p 257 — 270. Fischer, RA, and R. Maurer. 1978. Drought resistance in spring wheat cultivars. 1. Grain yield responses. Australian Journal of Agricultural Research. 29: 897-912. Frahm, M. A. 2002. Effects of terminal drought stress on black beans. Thesis for the degree of Master in Science. Michigan State University. 121p Frahm, M. A., E. F. Foster, and J. D. Kelly. 2003. Indirect screening techniques for drought resistance in dry bean. Annual Report of Bean Improvement Cooperative. 46:87-88. Frahm, M. A., J .C. Rosas, N. Mayek-Perez, E, Lopez-Salinas, J. A. Acosta-Gallegos, and J. D. Kelly. 2004. Breeding beans for resistance to terminal drought in lowland tropics. Euphytica 136:223-232. Gregory, P. J. 1994. Root growth and activity. In: Boote, K. J., J. M. Bennett, T. R. Sinclair, and G. M. Paulsen. ASA International Symposium. p 65 — 93. Hall, A. E. 1993. Is dehydration tolerance relevant to genotypic differences in leaf senescence and crop adaptation to dry environments? In: T. J. Closeand E. A. Plants responses to cellular dehydration during environmental stresses. Rockville, Maryland. The American Society of Plant Pathologists. p 1-10. Harterlein, A. J. 1983. Bean. In: I. D. Teare and M. M. Peet. Crop and Water Relations. New York. Wiley Publishers. p. 157-185. Itai, C., and H. Bimbaum. 1991. Synthesis of plant growth regulators by roots. P 163- 177. In: Waisel, 1., A. Eshel, U. Kafl