s" d ' .. ‘44 unm- sz‘ -. L n r'fwwix‘ 13 - iwmhuu "VFT' m. -m. I iWYréfiWi :- ~ “Nb SI am i". r r ‘3‘“! .%J .- .. .3 11-“ 7~ > I" . U , 11'7“?“ yaw: : . mm“ 4,‘ v] k ‘z‘ Rwy"- 5-... ‘3“; w! «m - - .03; 2:3: :4; . w‘ a ’- a. --u. ‘3?"3'777 Zn. . a ;¢:,!Ju._ ., fif’ I 1"; filo} V 3 I'm»: " :J . u '7'0 3 ' H n u~ -» llllllllllllllllllllllllllllllllllllIIHIIHIIHIIIIIIlHlllll m 1 3 1293 01581 4407 Z, 1/ m ./ LIBRARY " W Michigan State Unlverslty This is to certify that the dissertation entitled THE INHERITANCE 0F COOKING TIME AND CHEMICAL AND AGRONOMIC TRAITS IN SEEDS 0F DRY BEAN (Phaseolus vulgaris L.) presented by Frank Martin Edward Elia has been accepted towards fulfillment of the requirements for Ph. D. degreem Crop and Soil Sciences Date Sept. 6, 1996 MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 PLACE ll RETURN BOX to remove thb checkout iron your record. To AVOID F INES return on or More dete due. DATE DUE DATE DUE DATE DUE MSU Ie An N'flrmetlve Action/Equal Opportunlty lnetltmlon Wows-m THE INHERITANCE OF COOKING TIME AND CHEMICAL AND.AGRONOMIC TRAITS IN SEEDS OF DRY BEAN (Phaseolus vulgaris L.) BY Frank Martin Edward Elia A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 1996 ABSTRACT THE INHERITANCE OF COOKING TIME AND CHEMICAL AND AGRONOMIC TRAITS IN SEEDS OF DRY BEAN (Phaseolus vulgaris L.) BY Frank M. E. Elia Cooking time of dry bean (Rhaseolus vulgaris L.) influences consumer utilization of the crop in countries where beans are a staple in the diet and are cooked in open kettles and stoves. Since dry beans usually require a long cooking time to render the grains palatable, their cooking time determines household fuel needs. This study was undertaken to determine the inheritance and gain from selection for cooking time, water absorption, protein and tannin content, seed weight and maturity in dry bean. Fifteen genotypes from.the Andean gene pool and one form the Middle American were mated in a North Carolina design II scheme. The thirty two progenies from the mating and sixteen parents were grown during the 1993 short rain and the 1994 long rain growing seasons at the Crop Museum and the Morning Site farms in Morogoro, Tanzania. Estimation of genetic variances, heritabilities, and response to selection for the traits were made. Total genetic variance was partitioned into additive and non additive components. Highly significant Frank M. Elia differences were observed. among progeny for the traits studied. Genetic variance was mostly additive. The single seed descent procedure or pedigree breeding method may be an effective breeding strategy for improving the traits because of their high heritability. Although recurrent selection for general combining ability (RSGCA) or reciprocal recurrent selection (RRS) would lead to the accumulation of favorable alleles with additive effects in the population, the efficiency of recurrent selection would be low because of the slow response to selection. Beans representing the parents of the Design II mating scheme were stored for nine months to determine the effect of storage on the expression of cooking time, water absorption, and seed germination. Seed genmination is important to farmers because they store remnant seed from.one year's crop to plant the subsequent year. Results showed significant genotypic differences among the entries. The influence of accelerated aging’ on cooking time indicated that genotypes that maintained a short cooking time, and high water absorption and seed germination after extended storage could be identified rapidly for use as parents for population improvement. Note: This dissertation is presented as a series of two papers written in the style and format required by Crop Science and, the Jburnal of'the American Society fbr.Hbrticultural Science. This work is dedicated to my father, the late Martin Edward Elia. ACKNOWLEDGMENTS I wish to express my deepest gratitude to Dr. George L. Hosfield, my major professor, for his constant guidance, encouragement and valuable suggestions during the course of my work at Michigan State University. He has been a source of happiness and unlimited stimuli to my professional growth in my field through the many academic challenges he posed to me. His sense of humor and leadership qualities enabled me to tap his rich knowledge ranging from the basics in leadership to the specifics in academics. Special thanks go to members of my guidance committee Dr. James D. Kelly, Dr. Eunice F. Foster, and Dr. Mark A. Uebersax, Dr. Kelly has been helpful in suggestions and references relating to plant breeding. Dr. Foster has selflessly shared her experience in the MSTAT-C program that I used in the last two chapters. Dr. Uebersax provided useful suggestions and literature that were incorporated in the last two chapters of my dissertation. I would also like to thank Dr. Gill, for his assistance on some statistical work in the last two chapters. Dr. Harvey Dicks, a visiting professor from South Africa will always be remembered for the invaluable help he provided. me at the time I needed it most for the statistical analyses and estimation of components of variance using the GENSTAT software program. Gratitude is expressed to the African American Institute (AFGRAD III scholarship), and the Bean Cowpea CRSP for their financial support during my studies at MSU. Special thanks are extended to the Rockefeller Foundation that provided financial support which enabled me to conduct the research in Tanzania. I would also like to thank the University of Dar Es Salaam for granting me a study leave and retaining my teaching position there, while I pursued my Ph.D. studies at MSU. Gratitude is expressed to Sokoine University of Agriculture for providing me with research facilities that enabled the successful completion of this work. I am also grateful to MSU for believing in me and my research, and awarding me the prestigious Thoman fellowship to help me contribute to the alleviation of hunger and poverty in my country and other developing countries. Sincere thanks go to Crop and Soil Science graduate students with whom I had the privilege to work together and their constant assistance. Thanks also go to Ollie Trotter who spent long hours assisting in data entry and Karen Sussman and Mark Bitman for the final touches in my thesis. Appreciation and gratitude is expressed to all my friends who selflessly provided love, moral support, and encouragement throughout my study. Special thanks go to Dr. Brian Brunner, Dr. Imru Assefa, and Dr. Elias Sabry, Sandra Jones, and Zipporah Agheneza. Lastly but not least, I would like to thank and salute my mother Wintyapa, my wife Grace and my children Alfred, Emmanuel and Ester for their unwavering support, tolerance and encouragement during all my student years at Michigan State University. TABLE OF CONTENTS Page LIST OF TABLES . ...................................... x GENERAL INTRODUCTION . . . . . . . . . . . . . . . . 1 GENERAL LITERATURE REVIEW . . . . . . . . . . . . . . 6 Cookability of bean . . . . . . . . . . . . . . 6 Protein Content . . . . . . . . . . . . . . . 10 Tannin Content . . . . . . . . . . . . . . . 12 Maturity and Seed weight . . . . . . . . . . . 13 Beans Stored under adverse conditions . . . . 14 Accelerated.Aging . . . . . . . . . . . . . . 19 List of References . . . . . . . . . . . . . . 22 CHAPTER ONEzINHERITANCE OF COOKING TIME, WATER ABSORPTION, PROTEIN AND TANNIN CONTENT, MATURITY AND SEED WEIGHT IN DRY BEANS (Phaseolus vulgaris L.); THEIR RESPONSE TO SELECTION AND INTERRELATIONSHIP. . . . . . . . . . . . . . . 33 ABSTRACT . . . . . . . . . . . . . . . . 33 INTRODUCTION 0 O O O O O O O O O O O O I 35 MATERIALS AND METHODS . . . . . . . . . . 39 Genetic material . . . . . . . . . . 39 Procedures . . . . . . . . . . . . . 41 Trait evaluation . . . . . . . . . 43 Statistical Analysis . . . . . . . . 45 RESULTS AND DISCUSSION . . . . . . . . . 49 Breeding implications . . . . . . . 66 Use of correlated responses in breeding strategy. . . . . . . . . . . . .68 LIST OF REFERENCES . . . . . . . . . . . 70 CHAPTER TWO . . . . . . . . . . . . . . . . . . . . 75 THE EFFECT OF STORAGE TIME, TEMPERATURE AND RELATIVE HUMIDITY ON THE COOKABILITY AND SEED QUALITY TRAITS IN DRY BEAN (Rhaseolus vulgaris L.). . . . . . . . . 75 ABSTRACT . . . . . . . . . . . . . . . . 75 INTRODUCTION . . . . . . . . . . . . . . 77 MATERIALS AND METHODS . . . . . . . . . . 81 Genetic material . . . . . . . . . . 81 Procedures . . . . . . . . . . . . . 82 Character evaluation . . . . . . . . 86 Statistical Analysis . . . . . . . . 87 RESULTS AND DISCUSSION . . . . . . . . . 88 Breeding implications . . . . . . . 100 LIST OF REFERENCES . . . . . . . . . . . 102 LIST OF TABLES CHAPTER ONE 1. Description, Chemical Contents, Maturity, Seed Weight, Water Absorption, and Cooking time index of 16 Dry Bean .Accessions Used as Parents in a North Carolina Design II Mating Scheme and. Grown. at Two Locations in the Morogoro Region, Tanzania in the Short Rain (1993) and Long Rain (1994) Growing Seasons 0 O C O O O O O O O O O O O O O O O O O O O O 0 4O 2. Mean squares from the combined analyses of variance for maturity, protein, tannin, seed weight, water absorption and cooking time index for dry' bean progeny grown at two locations in the Morogoro region, Tanzania during the SR (1993) and LR (1994) growing season . . . . . . . . . . . . . . . . . . . . . . . . . 52 Estimates of variance components scaled to sum to 100 for maturity, protein, tannin, seed weight, water absorption and cooking time for dry bean progeny grown at two locations in the 'Morogoro Region, Tanzania during the SR (1993) and LR (1994) Growing Seasons 0 O I O O O O O O O O O O O O O O O O O O O O O 54 Estimates of additive and dominance variance, broad and narrow sense heritability and the degree of dominance for protein, tannin, maturity and seed weight for dry bean progeny grown at two locations in the Morogoro region, Tanzania during the SR(1993) O O O O O O O O O O O O O O O O O O O O O O O O 5.7 Means and ranges for maturity, protein, tannin, seed weight, and water absorption as related to Cookability for dry bean progeny grown at two locations in the Morogoro Region, Tanzania in the SR (1993) and LR(1994) growing season 0 O O O O O O O O O O O O O O O O O C O O O O O 60 Phenotypic correlation coefficients between protein, tannin, maturity, seed weight and cooking time index for dry bean progeny grown at two locations in the Morogoro region, Tanzania during the short rains (1993) and long rains(1994)growing season 0 O O O O O O O O O O O O O O O O O O O O O O O 64 Estimation of response to selection for maturity, protein, tannin, and seed weight, water absorption and cooking time for dry bean progeny grown at two locations in the Morogoro Region, Tanzania during the short rains (1993) and long rains (1994) growing season CHAPTER TWO 1. Production and consumption figures for major bean growing regions Mean squares from analyses of variance for eight dry bean entries for seed germination, water absorption and cooking time traits over nine months storage at varying temperature and relative humidity O O O O O O O O O O O O O O O O O O O O 0 Mean germination percent, water absorption, and cooking time of seed stored at 20°C, 25°C and 30°C and averaged over eight genotypes, two relative humidities and three storage times. Mean germination percent, water absorption, and cooking time of seed stored at two relative humidities and averaged over eight genotypes, three temperatures and three storage times. Mean germination percent, water absorption, and cooking time of seed stored at three storage times and averaged over eight genotypes, two relative humidities and three temperatures. Mean germination percent, water absorption, and cooking time of eight genotypes averaged over three storage times two relative humidities and three temperatures. Mean square analyses for cooking time trait of sixteen entries of dry bean seed that were accelerated aged from two locations grown during the SR (1993) and LR (1994) growing seasons at the Morogoro region, Tanzania. 80 89 91 92 93 94 97 8. Table of means for cooking time trait, of dry beans under three storage times and accelerated aging. . . . . . . . . . . . . . . . . . . . . . . . . 98 9. Table of means for cooking time of sixteen entries of dry bean seed that were accelerated aged from two locations grown during the SR (1993) and LR (1994) growing seasons at the Morogoro region, Tanzania. .APPENDIX: LIST OF TABLES 1. General soil characteristics at Morogoro at 0-12 cm, during the 1993-94 growing seasons 0 O O O O O O O O O O O O O O O O I O O O O O 104 2. Form of analysis of variance and mean square expectation for data of a design II model II experiment repeated over locations at Morogoro region, Tanzania in the 1993-94 growing season 0 O O O O O O O O O O O O O O O O O O O O 0 0 105 3. Protein content determination . . . . . . . . . . . . . . . . . . . . . . . 106 4. Tannin content determination O O O O O O O O O O 0 O O O O O O O O O O O O 106 5a. Analysis of variance for protein content SR Season 0 O O O O O O O O O O O O O O O O O O 0 O O O 110 5b. Estimated Components of Variance for protein content SR season 0 O O O O O O O O O O O O O O O I O O O O O O 110 6a. Analysis of variance for tannin content SR season 0 O O O O O O O O O O O O O O O O O O O O O O 111 6b. Estimated components of variance for tannin content SR season ‘. . . . . . . . . . . . . . . . . . . . . . . 111 7a. Analysis of variance for maturity SR season . . . . . . . . . . . . . . . . . . . . . . . 112 7b. Estimated Components of Variance for maturity SR season 0 O O O O O O O O O O O O O O O O O O O O O O 112 8a. Analysis of variance for loo-seed weight SR season . . . . . . . . . . . . . . . . . . . . . . . 113 8b. Estimated components of variance for loo-seed weight SR season 0 O O O O O O O O O O O O O O O O O O O I O O 113 9a. Analysis of variance for cooking time SR season . . . . . . . . . . . . . . . . . . . . . . . 114 9b. Estimated components of variance for cooking time SR season 0 O O O O O O O O O I O O O O O O O O O O O O 114 10a. Analysis of variance for water absorption SR season 0 O O O O O O O O O O O O O O O O O O O O O O 115 10b. Estimated components of variance for water absorption SR season 0 O O O O O O O O O O O O O O O O O O O O O O 115 11a. Analysis of variance for protein content LR season . . . . . . . . . . . . . . . . . . . . . . . 116 11b. Estimated components of variance for protein content LR season . . . . . . . . . . . . . . . . . . . . . . . 116 12a. Analysis of variance for tannin content LR season . . . . . . . . . . . . . . . . . . . . . . . 117 EN 12b. 13a. 13b. 14a. 14b. 15a. 15b. 16a. 16b. 17. 18. Estimated components of variance for tannin content LR season Analysis of variance for maturity LR season Estimated components of variance for maturity LR season Analysis of variance for loo-seed weight LR season Estimated components of variance for loo-seed weight LR season Analysis of variance for cooking time LR season Estimated components of variance for cooking time LR season Analysis of variance for water absorption LR season Estimated components of variance for water absorption LR season Analysis of variance table for percent seed germination of eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity Analysis of variance table for water absorption of eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity 117 118 118 119 119 120 120 121 121 122 123 19. 20. 21. 22. Analysis of variance table for cooking time of eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity Analysis of variance for cooking time of sixteen entries of accelerated aged dry bean seed grown in SR season Analysis of variance for cooking time trait of sixteen entries of accelerated aged dry bean seed grown in LR season Daily weathe data for 1994 growing season at Sokoine University of Agriculture (SUA) 124 125 126 127 GENERAL INTRODUCTION The availability and affordability of foods which supply an adequate amount of proteins and calories to human diets are of major concern in developing countries. In many poorer countries of Asia, Africa and Latin America, rice (Oryza sativa L.), maize (Zea mays L.), millet (Pennisetum americana), sorghum (Sorghum bicolor), banana (MUsa sapientum), cassava (Mbnihot esculenta Crantz), yams (Dioscorea cayanensis), sweet potatoes (Ipomea batatas L. Lamb) and cultivated potatoes (Solanum tuberosum) are dietary staples. Although these crops are low in protein, they can supply most of the daily energy requirements to consumers. Since meat is scarce and expensive in developing countries, consumers rely heavily on foods from plant sources for their source of protein, vitamins and minerals. Dry bean (Phaseolus vulgaris L.) is an important source of dietary protein and calories for persons in .Africa and Latin America. Although beans are low in sulfur amino acids, the bean protein deficiencies are supplemented by cereal protein which is rich in cysteine and methionine. Bean-cereal mixtures offer a balanced protein and energy I 2 diet for consumers. Although beans constitute a large part of the daily diet of consumers in.Africa and Latin America, the crop is under utilized as a staple food. One of the factors that detract from the utilization of beans by consumers in developing countries is the long time required to cook them to a desired softness for eating. In developing countries in.Africa and Latin America, firewood is the primary source of fuel for cooking beans. The prolonged cooking time required for beans often causes an excessive use of fuel wood which exacerbates deforestation in the countries. In Eastern Africa, fuel wood requirements are largely determined by how often beans are cooked and the cooking time required to render them. to desired softness for eating (Shellie-Dessert and Hosfield, 1990). Hence, some consumers due to frequent use of different genotypes are able to distinguish slow cooking genotypes from fast cooking ones in the market by such characteristics as seed size, color and seed coat appearance. They usually purchase beans known to cook more rapidly. In the Eastern and Central Africa countries of Rwanda, Tanzania, Malawi, Kenya and Uganda, the demand for fuel wood far exceeds the supply (Howe and Gulick, 1980; Bart, 1981); consequently, deforestation far surpasses any 3 afforestation programs (Sirven, 1981). The scarcity of firewood in bean consuming countries of Africa has made the reduction in resources to prepare beans for eating an important economic consideration (Shellie-Dessert and Hosfield, 1991). Consumers in rural areas in these countries avail themselves of a wide selection of unimproved local bean land races to meet their preferences. Although beans are available there is a cost in terms of human and physical resources because of the relatively long cooking time required by beans compared to most other plant foods. The development of bean cultivars with cooking times shorter than the cultivars currently grown for consumption could be a means to conserve fire wood within the region. In many studies involving cooking of dry beans the seeds are often soaked in water to improve the hydration characteristics of the seeds for uniform cooking. The amount of water absorbed by beans during soaking prior to cooking may be indicative of the amount of time required to render them to the desired softness for eating (Nordstrom and Sistrunk, 1977, Jackson and varriano-Marston, 1981, Castellanos et al., 1995 ). The storage of dry bean under tropical conditions of high temperature (>30°C) and humidity (>75% RH) typically results in deterioration of seed quality. Moreover, beans 4 become “hard-to-cook” and require a prolonged cooking time to render the seeds palatable, inactivate heat labile antinutrients, and permit the digestion of starch and protein (Jaffe, 1973). In addition to the increased energy required during preparation, hard-to-cook beans have an inferior nutritional quality and poor acceptance by consumers (Stanley and Aguilera, 1985). The most important nutritional factor in beans is protein. However, tannin can effectively cross-link with protein and other macromolecules through phenolic hydroxyl and other reactive groups characteristic of tannin molecules (Kakade, 1974). The cross-linking of tannin with protein has been shown to adversely affect the nutrition in animal studies (Kakade and Evans, 1966; Lindgren, 1975; Rannenkamp, 1977). Traits such as seed. weight and. maturity are of importance to consumers in East.Africa. Seed weight is a central yield component. Adams and Reicosky (1975) reported regression coefficient (b = 0.99) between seed weight and yield in dry bean but not for maturity. Moreover, maintaining a preferred seed weight is an important consumer preference characteristic. The presence of genetic variability is a major prerequisite for making genetic improvement for 5 quantitative traits in a crop. Genetic variability arises from differences in the assemblage of genes among parents. Selection in segregating generations following hybridization is the current practice to make genetic advance for metric traits in dry bean. However, the basis for selection of a trait depends on a knowledge of its inheritance and heritability. Bean cultivars that take a short time to cook to a palatable texture have been reported (CIAT, 1986; Shellie, 1986). The utilization of dry bean can be increased in diets of the consuming population in Eastern and Central .Africa through plant breeding. Breeding strategies should be directed to the development of cultivars with a relatively fast cooking time, a rapid water uptake during soaking, a low tannin content, protein content between 23- 268, 90 days maturity, and with seed weights between 30-40 9/100 seed. The objectives of this research were to: (1) determine the inheritance of bean cooking time, water absorption, protein and tannin content, loo-seed weight and maturity, (2) identify traits that are useful for indirect selection of cooking time, (3) measure the effect of accelerated aging on dry bean cooking time and (4) evaluate cooking time, and germination percentage of bean seed stored for nine months. GENERAL LITERATURE REVIEW The dry seeds of common bean (Phaseolus vulgaris L.) are an important staple for people in Central and South America and East and Central Africa where animal protein is too costly for the average income consumer. On a dry weight basis, beans have a high protein content, twice that of cereals (Goodhart and Shils, 1980). Beans are also relatively rich in lysine and threonine (Patel et al., 1980; Sahasrabudhe et al., 1981) although deficient in the sulfur amino acids. Beans proteins complement cereals which are deficient in lysine. Beans are an important source of minerals such as iron and calcium, and the water soluble vitamins, like niacin, thiamine, folic acid and riboflavin (Tobin and Carpenter, 1978; Fordham et al., 1975). Despite the fact that Latin America is the leading producer of dry beans (30%) of the world's 14 million tons/year (van Schoonhoven and Voysest, 1993), the mean per capita consumption is 13.3 kg/capita. Africa produces 10% of the world production but has a mean per capita consumption of 31.4 kg with some countries like Rwanda consuming up to 50.6 kg. Beans in East Africa provide up to 60% of the dietary' protein requirements to consuming populations. 7 Cookability of bean. The definition of ‘Cookability’ encompasses characteristics that have strong organoleptic appeal to consumers. The time it takes to cook individual grains to a palatable texture, cooked bean wholeness, color, splitting and clumping of beans, and the color and consistency of the broth comprise Cookability. Among the constraints limiting dry bean utilization is the prolonged cooking time of’ beans and the confounding effect of hardening in storage. Prolonged cooking increases firewood usage and places an economic burden on consumers in areas where firewood is scarce. The aspects of Cookability, the amount of softening and rate of softening are related to cellular breakdown in cell walls and the solubilization of starch. According to Silva et al., (1981), bean softening does not follow first order kinetic reactions. Sefa-Dedah and Stanley (1979) found that water absorption influenced cookability and that it was determined. by seed coat texture. Castellanos et al., 1995 reported a negative relationship between cooking time and water absorption and that ‘hard shell' rather than the ‘hard-to-cook' phenomenon is the main factor delaying the cooking time of newly harvested beans. However, Burr et al., (1968) and Molina et al., (1975) found no correlation between water uptake and Cookability. 8 Various methods have been employed for the assessment of cooking time. Some investigators (Mwandemele et al., 1984) counted the number of split seed coats after a specified cooking time of two hours. Edmister, (1990) applied pressure to cooked beans placed between the index finger and the thumb and related the force required to rupture the seed to a tactile evaluation. A cut-off point of 450 mean gram force based on a 100 bean sample was determined by squeezing individual cooked beans between the thumb and the index finger and comparing tactile sensations with the grams force registered by those beans using the puncture test cell. A number of instruments have been used to evaluate dry bean cooking time. The Instron (compression type), maturometer (penetration type) and the Mattson pin drop cooker have been used. The pin drop cooker method for assessing cooking time is the most preferred because it is economical, reliable and can evaluate 100 seeds simultaneously. Dry bean Cookability has been found to be related to seed storage. Observations (Burr et al., 1968) showed that some bean cultivars, when stored for periods of six months or longer become hard-to-cook..A number of factors have been suggested. for’ the hardening jphenomenon including lignification of the middle lamella, seed coat thickness 9 and the level of phytase, calcium and magnesium in the seed. Cooked. bean softening is associated with the breakdown of the middle lamella in the seed which allows separation of the cells. Jones and Boulter (1983 a,b) observed that the hard-to-cook phenomenon might be due to reduced water imbibition and reduced pectin solubility resulting in a reduced cell separation rate during cooking. These researchers suggested that the reduced water imbibition. and reduced pectin solubility act synergistically. Accompanying symptoms include solute leakage during soaking due to membrane breakdown, phytin catabolism.and pectin demethylation. Moscoso et al., (1984) investigating the hard-to-cook phenomenon in red kidney beans, found that the apparent rate constants of cooking decreased with increasing time storage. Jimenez et al., (1989) also observed that when some bean cultivars were stored under high temperature and high relative humidity for nine months, the time required for cooking increased three-fold. It has been observed that the cooking time of beans is under genetic and environmental control that often interact unpredictably (Shellie and Hosfield, 1991). Hard seed coats especially noticeable in small seeded cultivars (100 seed weights of 18-25 grams), prevent seeds from.imbibing water 10 during soaking leading to prolonged cooking times. It is interesting that the development of a hard seed coat in a cultivar determines the length of time it will take to cook the bean. The hard shell development has been shown to be under genetic and environmental control (Rolston , 1978; Stanley and Aguilera, 1985). Copeland, (1976) established that the heritability of the hard seed coat was high. However, this trait was controlled by a single recessive allele (Kyle and. Randall, 1963). Hard. seed. coat also develops under adverse storage conditions of temperature and relative humidity. Unlike beans with hard seed coats, hard-to-cook seeds imbibe water, but fail to soften adequately during cooking (Burr et al., 1968). Among the early information on the hardness of beans was that of Gloyer, (1928) in which he divided it into hard shell (the condition in which the seed fails to imbibe water) and the hard-to-cook condition (hardness produced by enzymatic action in which the cotyledons darken and harden and become hard to cook. Castellanos et al., 1995 reported that the hard shell plays a major role in extending the cooking time of beans with an initial moisture content of 90 9 Kg 4 or less. Protein content. The protein quality of a food is based on the relative contents of the amino acids. The limited 11 utilization of a protein is thus based on the essential amino acids present in the lowest amount. Bean proteins though rich in lysine, are limited in the sulfur containing amino acids, cysteine and methionine (Patel et al., 1980; Sahasrabudhe et al., 1981; Sathe et al., 1981). Bean proteins have low digestibility, between 48%-62% compared to meat proteins which range between 82%-86%. The presence of antinutritional factors, lectins (Jaffe, 1973; Barampama and Simard, 1994;) contribute to decreased protein utilization. Percent protein. has been found. to be negatively related to seed yield (Leleji et al., 1972; Kelly and Bliss, 1975;) thereby' making it difficult to improve protein content through breeding and selection (Bliss and Brown, 1983). Sullivan and Bliss, 1983 reported that it is possible to have an effective breeding strategy either through simultaneous selection for both protein and seed yield employing some form of a selection index, or using tandem selection first for high yield and then for high seed protein. Heritability estimates for seed protein percentage range from 0.10 to 0.85, (Porter, 1972; Leleji et al., 1972; Kelly and Bliss, 1975; Evans and Gridley, 1979;). In a study conducted by Emebiri, 1991, the inheritance of protein content in bean has been shown to be 12 both controlled by both additive and non additive gene effects. The broad sense heritability of protein content ranged from 0.7 and 0.8 Tannin content. It has been shown that condensed polyphenols in the seed coat (tannins) decrease protein digestibility (Elias et al., 1979) and that high tannin reduces nutritional value by binding with protein (Haslam, 1979; Griffiths and Moseley, 1980). Tannins in beans are located mainly in the testa of colored seeds (Ma and Bliss, 1978a ; Bressani and Elias, 1980; Deshpande et al.,1982). Tannins are polyphenolic compounds (high molecular weight 500 to 3,000) that are capable of precipitating proteins to form complexes that are not easy to digest (Griffiths and Moseley, 1980; Aw and Swanson, 1985; Elias et al., 1979; Reddy et al., 1985; Jansman, 1995). Apart from inactivating digestive enzymes and. increasing jprotein insolubility, tannins also lower feed efficiency which lead to a growth depression in experimental animals (Lindgren, 1975; Rannenkamp, 1977; Reddy et al., 1985). Tannins are categorized into hydrolyzable tannins which can be degraded to sugars and phenolic carboxylic acids when treated by acids or alkali, and condensed tannins which are not readily degraded by simple chemical treatments (Haslam, 1979). The tannin content of dry bean seed ranges from.0.0 13 to 2.0% (Sgarbieri and Garruti, 1986; Reddy et al., 1985). Quantitative variability for tannin has been reported (Ma and Bliss, 1978 a; Lyimo et al., 1992). These researchers showed that beans with colored seeds contained more tannin while tannins were not detected in beans with white seed coats. There was no strong relationship between tannin content of colored seeded genotypes and seed coat color. From a survey of literature on tannin (Allard, 1953; Picard, 1976; Rannenkamp, 1977; Ma, 1978 a,b; Marquardt et al., 1978; Crofts et al., 1980; Dalrymple et al., 1984;), it appears that seed coat color and tannin content are independently inherited. High broad sense heritabilities of 0.80 to 0.97 were obtained by Ma, 1978a when he analyzed populations resulting from crosses between parents that differed in seed coat color and tannin content. He also observed that low tannin was dominant to high tannin. However in a diallel analysis, Wassimi et al., found high tannin dominant to low tannin. Moreover significant estimates of general and specific combining ability were associated for tannin expression. .Maturity'and Seed weight. In East and central.Africa bean growing regions where short and long rain growing seasons occur, bean growers prefer growing early maturing, large seeded cultivars. The short rains occur between the end of 14 September to December while the long rains start in March and end in June. The short rain growing season exerts pressure on the farmers who have to plant and harvest within the period. Breeding for large seeded and early maturing genotypes is therefore important in the East and central Africa bean growing region to enable farmers to maximize the short rains. In their experiment to determine heritability and correlation of biomass, growth rate, harvest index and phenology to the yield in common bean, Scully et a1. , (1991) reported broad sense heritabilities of 0.96 and 0.87 for days to maturity and seed weight respectively. Cerna and Beaver, (1990) studying inheritance of early maturity in indeterminate dry beans reported a narrow sense heritability of 0.31 to 0.63. Singh et al. ,(1990) reported the estimated gain in selection values of < 3% and > 15% for maturity and seed weight respectively. Beans Stored under adverse conditione: It is recognized that storage environments can profoundly affect bean cookability. Prolonged storage of beans often leads to a long cooking time while freshly harvest beans require shorter cooking time. Jackson and varriano-Marston, 1981 found that freshly harvested beans cooked for 31 min while beans stored for one year took 45 min to cook. A loss of 15 nutritive value occurred with prolonged storage of beans (Nordstrom and Sistrunk, 1979). Beans stored at high relative humidity become increasingly darker and exhibit firmer textures. Beans stored at a relative humidity of 10% or lower maintained their cookability for up to two years (77??) while those stored at relative humidities above 13% showed a significant deterioration in flavor and textural characteristics within the six months (Morris and Wood, 1956). Storage at high moisture for four months resulted in longer cooking (Morris, 1963). Snyder, (1936) showed that in addition to storage time, temperature and humidity are factors that also lead to the development of hardness in beans. Bedford (1972) reported that beans stored at low moisture of 8-9%, and temperature of between 68-819? for four years maintained their cooking quality while those stored at high. moisture 15-18% showed a significant increase in cooking time. Unlike the pectase enzyme mechanism leading to the formation of calcium.pectate, the increase in seed toughness and hence the increase in cooking time has been suggested to be due to a non- enzymatic entropy increasing pectic chain entanglement that becomes dominant only at certain critical values of water 16 activity (Schwimmer, 1980). Significant changes in the bean proteins stored in high moisture levels (ll-14%) at 90%? were shown to be associated with the cookability of aged dry beans (Rockland, 1963). Cookability studies with sanilac, pinto and large lima beans stored.at 70° and 90°F, 6.5-14.4% moisture, for eleven months showed that those at 90° and high moisture required up to five hours to cook to eating soft after the first nine months of storage (Morris, 1963). When seeds with seed coats and those without seed coats were compared. in an experiment, 1Morris, (1963) concluded that seed coats were responsible for most of the increases in cookability in high moisture seeds due to an in increase in lipid acidity. Burr et al.,(1968) like Morris (1963) reported loss of cooking time in beans stored at high relative humidity and temperature in addition to seed darkening. Scanning electron microscopy studies (Rockland et al.,1973, Rockland and Jones, 1974 and Sefa- Dedah et al., 1979) showed that thermal degradation occurred when freshly harvested seeds were cooked. Observations indicated that thermal degradation of the middle lamella was more difficult in cooked seeds stored at high temperature and moisture than in cooked seeds that were freshly harvested. A.number of studies associated the development of hard 17 seed coat or the hard to cook. phenomena to loss of nutritive value. Sada, (1980) reported that dark red kidneys stored at 30°C and 80% RH for eight months, showed decreased contents of total soluble carbohydrates, stachyose and raffinose, tryptophan, lysine, histidine and methionine. Burr et al., (1968) reported that firmer seeds took longer to cook resulting in increased destruction of thiamine and a decrease in biologically available methionine and. cysteine. .Antunes and. Sgarbieri (1979) reported a decrease in the protein efficiency ratio values and protein digestibility. Molina et al., (1976) showed that short heat treatments(2.5 and 10 min. at 121°C and 10,20 and 30 min under steam (98W3) on black beans prior to storage at 25°C, 70% RH for nine months, reduced the development of seed hardness although germination capacity was affected (Molina et al., 1976). This research showed that lignified protein increased during seed storage. Molina and Bressani (1977) attributed similar increases to the migration of polyphenolic compounds from.the seed coat to the cotyledon. A study on the effect of different conditions on the development of seed hardness or the hard- to-cook condition in beans was conducted by Mejia (1979). This researcher used seeds with moisture contents of 9, 13, and 17% that were in equilibrium with relative humidities 18 of 40, 60, and 80%, and stored for six months at temperatures of 4, 20 and 36W2. Mejia (1979) observed a decreased tannin content with higher storage time and greater bean hardening at higher storage temperatures. Jackson and Varriano-Marston (1981) conducted accelerated storage tests which showed that black beans stored at 41°C and 100% RH for 7 to 14 days required longer times to cook. Decorticated beans of the same lot were observed to cook in about half the time. Bressani and Elias (1980) summarized the factors responsible for the development of hardness in legume seeds as influenced by : 1) cell structure changes; ii) condensation of tannins that may pose an obstacle to the free permeability of water; iii) reactions between tannins and.proteins resulting from adverse storage of the legume seeds; iv) reactions within the cotyledons of organic compounds and calcium and magnesium; v) reactions the cotyledons that are enzymatic. Srisuma et al., (1989) reported that storage of seeds at 20°C and 73% RH and 35°C and 80% RH induced changes of phenolic acid and promoted the development of the hard-to- cook beans. This work reported that large increases in the amount of hydroxycinnanic acid and increased hardening. Rozo et al., (1990), Hernandez-Unzon and Ortega-Delgado (1987), Shehata et al., (1984) and Hentges et al., (1991) 19 conducted storage studies on the hard-to-cook phenomenon in beans. Paredes-Lopez et al., (1989) reported an increase in the compact packaging of cotyledon cells and an increase in water activity but a decrease in water absorption capacity in stored beans. Mafuleka et al.,(1991) reported increased hardening and higher pectin methylesterase activity with storage. Uebersax (1972) observed that the influence of increased temperature storage became greater at higher relative humidity. Accelerated Aging. Accelerated aging as a test for seed quality was first developed by Delouche, (1965a) as a method of predicting the viability and germination of seed lots kept in storage. This test was later found to be important as a vigor test (Herrera, 1969; Haya, et al., 1978; Wilson et al.,, 1992;). The storage potential of seed lots could be predicted by Delouche (1965b) based on their germination responses after a short exposure to high temperature and. high relative humidity. By 'using the accelerated aging technique, Delouche(1965b) found a high correlation among seed lots of crimson clover (TrifOlium incarnatum L.) and tall fescue (Festuca arundinaceae ' Schreb). Pil (1967) found that accelerated aging was an effective test for evaluating the storability of alfalfa (Medicago sativa L.), corn (Zea mays L.), wheat (Triticum 20 aestivum L.) and cotton (Gossypium hirsutum L.). Mercado (1967) found the method to be effective for determining the rate and progress of deterioration of cotton seed during storage. Mercado, (1967) suggested using accelerated aging to supplement the standard germination test to evaluate field performance potential for seed lots. Unlike other researchers Herrera, (1969) reported the possibility that heritable factors may partially control the vigor and deterioration rate of wheat seed lots. He further pointed out that the accelerated aging technique can be used by plant breeders for the selection of breeding materials in a population improvement program. Other researchers Tippayaruk, (1975) while evaluating the usefulness of accelerated aging technique for selecting cotton seed for improved vigor in a segregating population, concluded that F3 cotton seeds that were accelerated aged, failed to show heritable differences in seed vigor in the F4 generation. Wilson et al., (1992) found a high correlation between accelerated aging test and final stand of shrunken-2 sweet corn seed, A similar response was also observed earlier by Baskin, (1970) in peanuts indicating that germinative responses are highly correlated with plant growth and development. Since storability is one aspect of vigor, accelerated aging tests are very effective in the 21 evaluation of seed vigor (Islam, 1976). LIST OF REFERENCES Adams, M. W. and D. A. Reicosky, 1975. Plant architecture and physiological efficiency in the field bean. Report to Rockfeller Foundation. Allard, R. W. 1953. Inheritance of some seed coat color and patterns in lima beans. Hilgardia 22: 167-177. Antunes, P. L. and V. C. Sgarbieri. 1979. Influence of time and conditions of storage on technological and nutritional properties of a dry bean (Phaseolus vulgaris L.) variety Rosinha GZ. J. Food Sci. 44: 1703-1706. Aw, T. L. and B. G. Swanson. 1985. Influence of tannin on Phaseolus vulgaris protein digestibility and quality. J. Food Sci. 50:67-71. Barampama, Z. and Simard, R. E. 1994. Oligosaccharides, antinutritional factors, and protein digestibility of dry bean as affected by processing. J. Food Sci.59(4):833-838. Bart, Fr.(1981). Le Paysan Rwandais et l'energie. In: M. De Lame(Eds). Societe, culture et Histoire du Rwanda. Encyclopedie Bibliographique 1963-1980/87. Koninklijk museum voor Midden-Africa. Musee royal de 1'Afrique centrale. Tervuren, Belgium, p. 178. Baskin, C. C. 1970. Relation of certain physiological properties of peanut (Arachis hypogea L.) seed to field performance and storability. Dissertation (Ph.D.), Mississippi State university, Mississippi State, MS. Bedford, C. L. 1972. Bean storage and processing. Proceedings of the International Dry Bean Symposium. .August 22-24, at. Michigan State University, East Lansing, MI. p.63 Bliss, F. A. and J. W. S. Brown. 1983. Breeding common bean for improved quantity and quality of seed protein. : In Janick, J. (ed).Plant Breeding Reviews 1:59-102.AVI Publishing Co., Westport, CN, USA. 22 23 Burr, H. K., S. Kon, and H. J. Morris. 1968. Cooking rates of dry beans as influenced by moisture content and temperature and time of storage. Food Tech. 22:336- 338. Bressani, R., 1975. Nutritional improvement of food legumes by breeding, edited by M, Milner, John Wiley and sons, New York , N. Y. Bressani, R. and L. G. Elias. 1980. Polyphenols in cereals and legumes, IN: J. H. Hulse (ed.)IDRC, Ottawa, Canada. p 61. Bressani, R., L. G. Elias, and J. E. Braham, 1982. Reduction. of digestibility' of legume proteins by tannins. J. Plant Foods 4:43-55. Burr, H. K., S. Ken; and H. J. Morris. 1968. Cooking rates of dry beans as influenced by moisture content and temperature and time of storage. Food Technol. 22:336- 338. Castellanos, J. 2., H. Guzman-Maldonado, J. A. Acosta- Gallegos and J. D. Kelly. 1995. Effects of hardshell character on cooking time of common beans grown in the semiarid highlands of Mexico. J. Sci. Food Agric. 69:437-443. Cerna, J. and J. S. Beaver 1990. Inheritance of early maturity of indeterminate dry beans. Annu. Rept. Coop. 33:10-13. CIAT (Centro Internacional de Agricultura Tropical). 1986. Bean program annual report. CIAT, Cali, Colombia. p. 248. Copeland, L.O. 1976. Principles of Seed Science and Technology. Burgess, Minneapolis, MN. USA. Crofts, H. J., L. E. Evans, and P. B. E. McVetty. 1980. Inheritance, characterization and selection of tannin -free faba beans (Vicia faba L.). Can. J. Plant Sci. 60(4): 1135-1140. 24 Dalrymple, E. J., B. P. Goplen, and R. E. Howarth. 1984. Inheritance of birdsfoot trefoil. Crop Sci. 24:920 -923. Delouche, J. C. 1965a. Development of methods for predicting the longevity of crop seed lots in storage. Seed World 97(6):4-6. Delouche, J. C. 1965b. An accelerated aging technique for predicting relative storability of crimson clover and tall fescue seed lots. Agr. Abst. (ASA 1965 Meeting). P.40. Delouche, J. C. and C. C. Baskin. 1973..Accelerating aging and techniques for predicting the relative storability of seed lots. Food Sci. Technol. 1:427-452. Deshpande, S. S., S. K. Sathe, D. K. Salunkhe, and D. P. Cornforth, 1982. Effects of dehulling on phytic acid, polyphenols, and enzyme inhibitors of dry beans (Phaseolus vulgaris L.). J. Food Sci. 47:1846. Dessert, K. C. 1986. Environmental influence study on cooking times and total protein content of ten (Phaseolus vulgaris L.) Varieties in Rwanda, Africa. Bean Improv. Coop. 29:123 Edmister, J. .A,, W. M. Breene, .A. Serugendo. 1990. Influence of temperature, water activity and time on cookability and color of a stored Rwandan dry bean (Phaseolus vulgaris L.). J. Stored Prod Res 26(3): 121-126. Elias, L. G., D. G. de Fernandez and R. Bressani. 1979. Possible effects of seed coat polyphenolics on the nutritional quality of bean protein. J. Food Sci. 44:524-527. Emebiri, L. C. 1991. Inheritance of protein content in seeds of selected crosses of cowpea (Vigna unguiculata). J. Sci. Food Agric. 54(1):1-7. Evans, A, M. and H. E. Gridley. 1979. Prospects for the improvement of protein and yield in food legumes. Curr. Adv. Plant Sci.32:l-17. 25 Fbrdham, J. R., C. E. Wells and L. H. Chen. 1975. Sprouting of seeds and nutrient composition of seeds and sprouts. J. Food Sci. 40:552 Gloyer, W. O. 1928. Hardshell of Beans: Its production and prevention under adverse storage Conditions. Proceedings of the Nineteenth-Twentieth Annual Meetings of the Association of official Seed Analysts of North America. p. 52- 55. Goodhart, R. S. and M. E. Shils. 1980. In Modern Nutrition in health and Disease, 6th edition,Philadelphia, PA. Griffiths, W. D. and G. Moseley. 1980. The effects of diets containing field beans of high or low polyphenolic content on the activity of digestive enzymes in the rats. J. Sci. Food Agric.31: 255-259. Haslam, E. 1979. vegetable Tannins. Rec. Adv. Phytochem. 12:475. Haya, G., Luria, L. W. Woodstock and M. Perl. 1978. The effect of accelerated aging of sorghum. seed on seedling vigor. J. Of Exper. Bot. 29(109):489-495. Hentges, D. L., C. M; Weavers, S. S. Nielsen. 1991. Changes of selected physical and chemical components in the development of the hard to cook bean defect. J. Food Sci. 56(2): 436-442. Hernandez-Unzon, H. Y. and M. L. Ortega-Delgado. 1987. Water absorption and cooking time in long stored common bean seeds (Phaseolus vulgaris L.). Annu. Rept. Bean Improv. Coop. 30:58-59. Herrera, C. V. 1969. Response of wheat seed to accelerated aging’ and. other ‘vigor tests for ‘vigor and deterioration Thesis (M.S.), Mississippi State University, Mississippi State, MS. Howe, J. W. And F. A. Gulick. 1980. Fuel and other renewable energies in Africa - a progress report on the problem and the response. Overseas Development Council, Washington D .C. p. 43. Islam, A" J. M. 1967. Comparison of methods of evaluating deterioration in rice seed. M. S. Thesis, Mississippi State University, Mississippi State, MS. 26 Jackson, M. G. and E. Varriano-Marston. 1981. Hard-to-cook phenomenon in beans: effects of accelerated storage on water absorption and cooking time. J. Food Sci. 46:799-803. Jaffe, W. G. 1973. Factors affecting the nutritive value of beans. In M. Milner (ed.). Nutritional improvement of food legumes by breeding. Protein Advisory Group of the United Nations System, New York. p.43-48. Jansman, A. J. M. 1995. Effects of hulls of faba beans (Vicia faba L.) with a low and high content of condensed tannins on the apparent ileal and fecal digestibility of nutrients and the excretion of endogenous protein in ileal digestion and feces of pigs. Amber. Soc. Anim. Sci. 73(1):115-127. Jimenez, J. A., D. P. Coyne and Saladin, F. 1989. Imbibition, germination and cooking time of seeds of dry beans (Phaseolus vulgaris) stored in different containers. J. Agric. Univ. P-R. 73(4):327-338. Jones, P. M. B. and D. Boutler. 1983a. The cause of reduced cooking rate in Phaseolus vulgaris following adverse storage conditions. J. Food Sci. 48:623-626. Jones, P. M. B. and D. Boutler. 1983b. The analysis of development of hard bean during storage of black beans (Phaseolus vulgaris L.) Qual. Plant Foods Hum. Nutr. 33:77-85. Kakade, M. L. 1974. Biochemical basis for the differences in plant protein utilization. J. Agric. Food Chem. 22:550 Kakade, M. L. and Evans, R. J. 1966. Growth inhibition of rats fed navy bean Phaseolus vulgaris L. J. Nutr. 90:191 Kelly, J. D. and F. A. Bliss. 1975. Heritability estimates of percentage seed protein and available methionine and correlations with yield in dry beans. Crop Sci. 15:753-757. Kyle, J. H. and T. E Randall. 1963. A new concept of the hard seed character in Phaseolus vulgaris L. and its use in breeding and inheritance studies. J. Amer. Soc. Hort. Sci. 83:461-475. 27 Lebedeff, G. A, 1947. Studies on the inheritance of hard seeds in beans. J. Agric. Res. 74:205-215. Leleji, O. I., M. H. Dickson., L. V. Crowder, and J. B. Bourke. 1972. Inheritance of crude protein and its correlation with seed yield in beans, Phaseolus vulgaris L. Crop Sci.12: 168-171. Lindgren, E. 1975. The nutritive value of peas and field beans. Swedish J. Agric. Res. 5: 159-161. Lyimo, Ms, J. .Mugula, and T. Elias. 1992. Nutritive composition of broth from selected bean varieties cooked for various periods. J. Sci. Food .Agric. 58(4):535-539. Ma, Y. and F. .A. Bliss. 1978a. Tannin content and inheritance in common bean. Crop Sci. 18:201-204. Ma, Y. and F. A. Bliss. 1978b. Seed proteins of common bean. Crop Sci. 18:431-437. Mafuleka, M. M., D. B. Ott, G. L. Hosfield and M. A. Uebersax. 1991. Dry'.bean (Phaseolus vulgaris L.) hardening and the consequences of pectin methylesterase activity in storage. J. Food Pross. Preserv. V01. 15(1):1-18. Marquardt, R. R., A, T. Ward, and L. E. Evans. 1978. Comparative properties of tannin-free and tannin containing cultivars of faba beans. Can. J. Anim. Sci. 58: 753-760. Meiners, C. R., N. L. Derise, H. C. Law, M. G. Crews, S. J. Ritchey and E. W. Murphy. 1976 . The content of nine mineral elements in the raw and cooked mature dry legumes. J. Agric. Food Chem. 24:1126. Mercado, A. T. 1967. Moisture Equilibrium and Quality Evaluation of Five Kinds of Seed Stored at Various Relative Humidities. Thesis (M.S.), Mississippi State University, Mississippi State, MS. Mejia, Elvira Gonzalez de. 1979. Effects of various conditions on general aspects of bean hardening. Final Report. Unu Fellow. Instituto De Nutrition De Centro America Y Panama. Guatemala. 28 Molina, M. R. and R. Bressani. 1977. Factores de almacenamiento y procesamiento. Taninos. In: valor nutricional de las leguminosas de grano y factores que afectan su produccion, disponibilidad y consumo. IV Congreso Latinoamericano De Nutricion. Caracas Venezuela. Arch. Lat. nutr. V01. XXVII, No. 2. Junio 1977. Suplemento 2:78-84 Molina, M. R., G. De La Fluente, R. Bresseni. 1975. Interrelationships between storage, soaking time, nutritive value and other characteristics of black beans (Phaseolus vulgaris). J. Food Sci. 40:587 iMorris, H. J. 1963. Cooking qualities of dry beans. In: 6th Annual Dry Bean Conference, Los Angels, January 1963. USDA-ARS. pp.11-23. Morris J. J. and E. Wood. 1956. Influence of moisture content on keeping quality of dry beans. Food Technol. 10:225-229. Moscoso, W., M. C. Bourne, and L. F. Hood. 1984. Relationship between the hard to cook phenomenon in red kidney beans and water absorption, puncture force, pectin, phytic acid, and. minerals. J. Food Sci. 49:1577-1583. Mwandemele, O. D., K. S. MCWhirter and C. Chesterman. 1984. Improving the quality of soybeans for human consumption: factors influencing the cookability of soybean seeds. J. Food Sci. Techn. 21:286-289. Nordstrom, C. and W. Sistrunk. 1977. Effect of bean soaking time, canning' media and storage time on quality attributes and nutritional value of canned beans. J. Food Sci. 42:795-798. Paredes-Lopez, O., E. C. Maza-Calvino, J. Gonzalez- Castaneda, 1989. Effect of the hardening phenomenon on some physico-chemical properties of common bean. Food Chem. 31(3): 225-236. Patel, K. M., C. L. Bedford and C. W. Youngs. 1980. Amino acid and mineral profile of air-classified navy bean flour fraction. Cereal Chem. 57:123-125 29 Picard, J. 1976. Apercu sur 1 heredite du caractere absence de tanins dans les graines de feverole (Vicia faba L.). Ann. Amelio. plantes. 26: 101-106. Pili, E. C. 1967. An accelerated aging technique for evaluating the storability of alfalfa, wheat, corn and cotton seed lots. Thesis (M.S.), Mississippi State University, Mississippi State, MS. Rannenkamp, R. R. 1977. The effect of tannins on nutrition quality of dry beans (P. vulgaris L.). Ph.D. Thesis. Purdue University. Reddy, N. R., M. D. Pierson, S. K. Sate; and D. K. Salunkhe. 1985. Dry Bean Tannins:.A review of nutritional implications. J. Amber. Oil Chem. Soc. 62:541-549. Rockland, L. B. 1963. Chemical and physical changes associated with processing of large lima. beans. Proceedings of the 6th Annual Dry Bean Conference, January 2-4, at Los Angeles, Ca. P. 9. Rockland, L. B. and F. T. Jones. 1974. Scanning electron mdcroscopy studies on dry beans. Effects of cooking on the cellular structure of cotyledons in rehydrated large lima beans. J. Food Sci. 39:342-346. Rockland, L. B., E. M. Zaragosa and D. M. Hahn. 1973. New information on the chemical, physical and biological properties of dry beans. In: Report of Bean Improvement Cooperative and.National Dry Bean Research .Association Conference, November 1973, Rochester, New York. Rolston, M. P. 1978. Water impermeable seed dormancy. Bot. Rev. 44:365-396. Rozo, C., M; C. Bourne, L. F. Hood,, P. J.‘Van-Soest, 1990. Effect of storage time, relative humidity and temperature on the cookability of whole red kidney beans and on the cell wall components of the cotyledons. Can. Inst. Food Sci. Technol. J. 23 (1):72-75. 30 Sada, G. 1980. Effects of different conditions of storage on germination, texture, nutritional quality and chemical composition of light red kidney beans (Phaseolus vulgaris). Ph.D. Thesis, Cornell University, Ithaca, New York. Sahasrabudhe, M. R., J. R. Quinn, D. Paton, C. G. Youngs and B. J. Skura. 1981. Chemical composition of white bean (Phaseolus vulgaris L.) and functional characteristics of air-classified protein and starch fractions. J. Food Sci. 46:1079-1088. Sathe, S. K.,‘V. Iyer and D. K. Salunkhe. 1981. Functional properties of the Great Northern bean (Phaseolus vulgaris L.) Proteins. Amino acid composition, in vitro digestibility, and application to cookies. J. Food Sci. 47:8-11. Schwimmer, S. 1980. Influence of water activity on enzyme reactivity and stability. Food Technology 34: 64-74 and 82. Scully, B. T.; Wallace, D. H. And Viands, D. R. 1991. The Heritability and Correlation of Biomass, Growth Rates, Harvest Index, and Phenology to the Yield of Common Beans. J. Am. Soc. Hort. Sci. 116(1):127-130. Sefa-Dedah, S. and D. W: Stanley. 1979. Textural implications of the microstructure of legumes. Food Technol. 53:77-83. Seidl, D., M. Jaffe and W. G. Jaffe. 1969. Digestibility and proteinase inhibitory action of a kidney bean globulin. J. Agric. Food Chem. 17:1318-1321. Sgarbieri, V. C. and Garruti, R. S. 1986. A review of some factors affecting the availability and the nutritional and technological quality of common dry beans, a dietary staple in Brazil. Can. Inst. Food Sci. Technol. J. 19:202. Shehata, A. M. E. T., A. S. Messallam,, A. A. El-Banna,, M. M. Youssef, and M. M. El-Rouby, 1984. The effects of storage under different conditions on cooking quality, viability and bruchid infestation of faba beans (Vicia faba L.). Trop. Stored Prod. Inf. Vol.1:9-18. 31 Shellie-Dessert, K. C. and G. L. Hosfield. 1990. Implications of genetic variability' for dry' bean cooking time and. novel cooking' methods for fuel conservation in Rwanda. Ecol. Food and Nutr.24:195- 211. Shellie-Dessert, K. C. and G. L. Hosfield. 1991. Genotype x environmental effects on food quality of common bean: resource-efficient testing procedures. J. Amer. Soc. Hort. Sci. Vol. 116 (4): 732-736. Singh, K. B., P. C. Williams and H. Nakkoul. 1990. Influence of growing season, location and planting time on some quality parameters of kabuli chick pea. J. Sci. Food.Agric. 53(4):429-441. Silva, C. A. B., R. P. Bates and J. C. Deng. 1981. Influence of pre-soaking on black bean cooking kinetics. J. Food Sci. 46(6): 1721-1725. Sirven, P. 1981. Le role des centres urbaines dans la deforestation de la campagne Rwandaise. In: Energie Dans les Communautes Rurales des Pays du Tiers Monde. P.243-254. Snyder, E. B. 1936. Some factors affecting the cooking quality of the pea and great northern types of dry beans. Nebraska Agric. Expt. Sta. Res. Bull. 85. Srisuma, N., Hammerschmidt, R., Uebersax,‘M.A~,Bennink, M. R. Ruengsakulrach, S., and Hosfield, G. L. 1989. Storage induced changes of phenolic acids and the development of hard to cook condition in dry beans (Rhaseolus vulgaris, var. Sea Farer). J. Food Sci. 54 (2):311-314. Stanley, D. W. and J. M. Aguilera. 1985. A review of textural defects in cooked reconstituted legumes -the influence of structure and composition. J.Food Biochem. 9:277-323. Sullivian, J. G. and F. A. Bliss. 1983. Expression of enhanced protein Content in inbred backcross lines of common bean. J; Amer. Soc. Hort. Sci. 108(5) :787-791. 32 Tippayaruk, S. 1975. Use of the accelerated aging technique in selecting for improved seedling vigor in cotton. Thesis (M.S.), Mississippi State University, Mississippi State, MS. Tobin, G. And K. J. Carpenter. 1978. The nutritive value of the dry bean (Phaseolus vulgaris L.): A literature review. Nutr. Abstr. Rev. Ser..A, 48: 919. Uebersax, M. A" 1972. Effects of storage and processing parameters on quality attributes of processed navy beans. MS Thesis, Mich. State Univ., East Lansing. van Schoonhoven, A. And 0. besest. 1993. Common Beans. Research for crop improvement. C. A. B. International in association with CIAT. Wilson, D. 0. Jr., J. C. Alleyene, B. Shafii, S. Krishna Mohan, 1992. Combining vigor test results for prediction of final stand of shrunken-2 sweet corn seed. Crop Sci. Vol. 32 (6):1496-1502. CHAPTER ONE INHERITANCE OF COOKING TIME, WATER ABSORPTION, PROTEIN AND TANNIN CONTENT, MATURITY AND SEED WEIGHT IN DRY BEANS (Phaseolus vulgaris L.): THEIR RESPONSE TO SELECTION AND INTERRELATIONSHIPS. ABSTRACT This study was conducted to determine the inheritance of cooking time, water absorption, protein and tannin content, maturity and seed weight in a population of dry bean representing the Andean gene pool. The response to selection for the traits and their interrelationships were also determined. Sixteen parents that differed morphologically, phenologically, and agronomically were crossed in a North Carolina design II mating scheme. General combining ability for males and females were highly' significant for' all traits. There ‘was a preponderance of additive genetic variance in the population, influencing trait expression. Because of the small gains expected from selection, the use of recurrent selection to improve the traits may be inefficient. It may be feasible to improve the traits under selection using single-seed descent or the pedigree method. Since beans are released commercially as pure lines, pedigree breeding should be considered for improving the traits under study. 33 34 Significant correlations were observed between cooking time and maturity, cooking time and tannin content, and cooking time and water absorption. Data indicate that the amount of water absorbed prior to cooking may be indicative of the amount of time required to render them soft for eating. Water uptake may provide a rapid and indirect method for screening genotypes for cooking time. INTRODUCTION Dry bean (Phaseolus vulgaris L.) is a major source of calories and protein for consuming populations in many .African and Latin American countries. Consumers generally eat beans with cereals in developing countries because the bean protein amino acids complement the cereal protein amino acids in cereal-legume diets (Bressani, 1975). .Although beans constitute a large part of the daily diet of low and middle income families in many developing countries, the crop is under-utilized as a staple food. One of the factors leading to the under-utilization of dry bean as a food is the prolonged time required to cook beans to a point where they can be digested and their nutrients assimilated (Deshpande, et al., 1982). In many developing countries, firewood is the primary source of fuel for cooking beans. However, the long cooking time required for beans compared to other foods has a cost in terms of human and physical resources. The cost is reflected in an excessive use of fuelwood which exacerbates deforestation. In the Eastern and Central African countries of Rwanda, Tanzania, Malawi, Kenya, and Uganda, the demand for fuelwood far exceeds the supply (Howe and Gullick, 1980; 35 36 Bart, 1981). Deforestation due to the excessive cutting of trees for fuel renders afforestation programs ineffective (Sirven, 1981). The scarcity of fuelwood in Eastern.Africa has made the reduction in resources to prepare beans for eating an important economic consideration (Shellie-Dessert and Hosfield, 1991). Bean varieties with fast cooking times may be a means to conserve fuelwood. Beans are generally soaked before cooking. Soaking improves the hydration characteristics of the seed so beans cook uniformly. Well soaked beans may also cook more rapidly than ones containing less water. A number of studies in dry bean have indicated an indirect relationship between cooking time and water absorption (Sefa-Dedah, 1979; Mera, 1982; Moscoso et al., 1984; Hernandez-Unzon and Ortega-Delgado, 1987; Paredes-Lopez et al., 1989;, Edmister et al., 1990, Castellanos et al., 1995). Since the degree of hydration of beans prior to cooking may be indicative of the amount of time required to cook than to a palatable texture, the amount of water absorbed by beans may provide an indirect method for screening genotypes for cooking time. Breeders routinely use indirect selection for improving traits when the procedureis more rapid and cost efficient for selecting a particular trait than the protocol necessary to select the trait per se. 37 In addition to cooking time and water absorption, protein and tannin content, maturity and loo-seed weight, are other traits which influence consumer utilization. Cooking time, water absorption, protein and tannin content do not fall into discrete phenotypic classes but show a continuum through a range of expression from a minimum to maximum for the respective trait. The phenotypic expressions of these traits are most often determined by measurement. The current practice to make genetic advances for traits of a metric nature in dry bean is selection insegregating generations following hybridization. However, the efficiency of selection for trait improvement depends (n: a knowledge of its genetic control and the degree to which the environment influences trait expression. Shellie- Dessert and Hosfield (1991) reported on genetic variability for cooking time of dry bean but there is limited published information on the inheritance of this trait. .An increase in the utilization of dry bean in the diets of consuming population of Tanzania and other major bean consuming countries of the region might be achieved by reducing the cooking time and tannin content and increasing the protein content through testing and selection. The hypothesis for the study was that cooking time in dry bean was highly heritable, and that significant genetic 38 variability existed in dry bean to develop acceptable cultivars that cook rapidly and conserve fuelwood. Specific objectives were to: (1) estimate additive and dominance variance for cooking time, water absorption, protein and tannin content, seed weight, and maturity in a genetic population of Andean dry bean, (2) estimate the heritability and degree of dominance for the traits and use this information for developing selection strategy, (3) determine the expected gain from selection for the traits and (4) calculate correlations among the traits and evaluate the feasibility of using other seed characteristics as a rapid and indirect method of screening for cooking time . MATERIALS AND METHODS Genetic.material. Sixteen genotypes of dry bean that differed morphologically, phenologically, and agronomically (Table 1) were used as parents in the study. Except for ‘Sierra', the parents were representative of the Andean gene pool of P. vulgaris and represented preferred grain types in Eastern and Central Africa. ‘Sierra' is a high yielding bean and representative of the Middle-American gene pool with the medium pinto market class seed size (Table 1). The genotypes were all adapted to the bean production areas of East and Central.Africa. Seeds of the entries are highly favored by consumers in the region because of their meat-like appearance and thick broth/soup quality after cooking. The appearance of the bean complements a stiffened porridge-like staple food prepared from corn called "ugali" in Kiswahili (East Africa), "nsima" in Malawi, "saza" in Zimbabwe, “bogobe” in Setswana, ”papa” in South Africa and "fufu" in Nigeria. 39 40 N.v o.~ H.v ~.m m.oH m.m va>o o.N o.om m.H m.mN m.om m.m Amo.ov 9mg ¢.mm m.mm> m.ov H.0mm v.vm~ w.v> coo: o.hm o.mom v.mv m.vwm o.mmm m.~> mfiflaou Hooofia m.mm 0.0mm m.mm o.mmm m.mHN m.mm muuofim m.mv m.mm> H.mm m.Hov m.~mm m.mp mm wz m.>v o.m>m m.vo m.>av m.mv~ m.o> Eamoucoz m.mm o.mam m.mv m.mmm o.mHN m.o> mauumo m.ooomn o.mv o.mvm b.vv o.mmv o.omm m.mh mNHH xmaw m.mm o.omoa m.>v m.mmm m.mmm m.om Hmmmow.H.m m.mv m.mmm m.wm o.m>v o.mom o.m> moon HmHu m.mm m.Hmm m.mv m.bam m.eom o.~> mEHme m.mv o.mmh h.vm m.>mm m.HmH o.~m mmm 029 m.~m m.mvm m.mv o.mmw o.mmw m.m> HNN Um: m.mm o.mHm m.mm o.mmm o.Ha~ m.mm mam soaaw» m.mv o.mmm e.>q m.mov o.mmm o.mp massesaaax m.Hv o.wm> m.om o.>nm o.omm m.Nm oucofimaq m.om o.va m.ov m.hom m.mvm m.o~ Susanaxmuflmz m.mm o.Hmoa >.mm m.vmm o.mvN o.m> AH um> :HE Tux o o Toooa 09 Tax o Imhmol Aooav osfiu coauouomom Davao: ucoucoo uaoudoo >ufluoumz ocwxoou nouns ooom canons nfiououm xuunm .memmom onfi3ouo Avmmavnfimu ocoa one Ammmav new“ uuonm on» ca waomncma .oowmou ouomouoz on» ad mnoHumooH 03» um oono one escrow oceans HH caflmmo mafiaonmo ruuoz m a“ mucouoo mo pom: mcofimmooom noon moo ma no mean mcaxooo one .nowuouomom noun: .unoaoz ooom .auausuma .mucouooo HmOHEmno you mumo oocfinEoo ocm .GOHuQHuomoa "H magma 41 Procedures. The 16 genotypes were intermated in the greenhouse in the fall of 1991 using a North Carolina Design II (Comstock and Robinson, 1948) mating scheme. Eight genotypes were randomly chosen as male parents and the remaining eight genotypes were used as female parents. The genotypes used as male parents were not crossed to each other nor were the genotypes used as females crossed to each other. In order to reduce the number of matings required, two sets were formed. Within each set, each male parent was crossed to four females resulting in a total of 16 crosses per set. Seeds of the 16 parents were increased, and the 32 F1 populations were advanced in January, 1992 in a nursery at the University of Puerto Rico substation near Isabela, Puerto Rico on a San Anton clay loam (fine-loam, mixed iso- hyperthemic cumulic haplustolls). Seed of the 48 entries (parents and F1 populations) were harvested in March 1992, bulked and returned to East Lansing. The 32 F2 populations were planted in a nursery at the Saginaw Valley Bean and Sugarbeet Research Farm near Saginaw, Michigan in a Misteguay silty clay [fine, illitic (calcareous),frigid typic Haplaquolls] . One hundred F2 plants from each cross were randomly selected, threshed, and the seed bulked. Because most of the parents were obtained from breeding 42 programs in Tanzania and other countries in the region of Eastern and Central Africa, the progeny were generally unadapted.in Michigan. Hence, the experiments on which data were taken were grown in 1993 and 1994 in Tanzania at the Crop Museum(530 masl) and.Morning Site(~1000 masl) research farms in the Morogoro Region. These farms were maintained by the Agricultural University of Sokoine. The soil characteristics at Morogoro are moderately acidic (Appendix Table 1). Rainfall, temperature and solar radiation are given in Appendix Table 22. The experiments were planted in a randomized complete block design with three replications at each location during the seasons in which the annual bimodal distribution of rain occurs. Because there was not sufficient hybrid seed to plant the experiment at both locations during the short(SR) and long(LR) rainy season, seed of the F3 was planted at the Crop Museum and Morning Site on September 30“, 1993 during the period of SR and seed produced from this crop (F3 generation) was planted at both locations in March loub 1994 during the LR period. Hence, generation and season effects were confounded. The confounding was not a consideration in the study' because the comparison of generations (i.e, F3 vs F,) was not made. Moreover, genetic variances can be estimated on populations with any level of 43 inbreeding. The field arrangement was based on the assignment of entries to the sets. Although the male groups were originally assigned to each set at random, the integrity of the entries within each set was maintained at each location in each growing season. The four females and four males used as parents of a particular set were included in the planting arrangement along with the 16 F3 or F, hybrid progeny produced by their intermating. The 48 entries (parents and progeny) making up the two sets were hand seeded into four row plots. Rows were 4 m long and spaced 0.5 m apart. The within row spacing was 0.1 m giving an approximate plant density of 160 plants per plot. On farm. practices for' herbicide, pesticide and fertilizer applications were followed. Mature plants were harvested by hand from the middle two rows of individual plots when the pods were dry enough to be threshed. W. Prior to harvesting the plots, physiological maturity was estimated on each plot as the number of days from.planting until about 90% of the pods in a plot changed from green to pale yellow or brown(no additional accumulation of dry matter). The freshly harvested seeds from each plot were sun dried for two 44 weeks. After drying the seed, they were stored in sealed plastic buckets and held at 20°C and 70% RH until evaluated (Shellie, 1990). Protein content was determined on 309 of raw beans that were ground to 40 um.particle size with a Udy Cyclone mill(Boulder, Colorado). Protein percentage was estimated on the bean flour by the micro-Kjedahl method (Association of Official Analytical Chemists, 1975). The nitrogen content of each sample was multiplied by 6.25 to calculate percent protein. The vanillin hydrochloric acid method of Burns (1963) was used to estimate tannin content. Since catechin was used as a standard in the tannin assay procedure, tannin content was estimated as catechin equivalent. However in this report, catechin equivalent will be referred as tannin. The percent water absorption of the entries was determined on triplicate samples of a known weight of 75 seeds from each plot. The seeds were soaked in distilled water for twelve hours at room temperature. The amount of water absorbed was taken as the difference in weight before and after soaking divided by the dry weight of the 75 seed sample. 45 Bean cooking time was estimated with a 25 seed pin- drop cooker (Jackson and Varriano-Marston, 1981). Cooking time was calculated as the elapsed time from initiation of cooking until 13 of the 25 pins of the instrument had dropped and penetrated seeds in the cooker. Data were taken on triplicate samples for each plot at each location. fila;istigal_flnalysis. Data were collected on both parents and progeny. Data on the parents were analyzed separately using analyses of variance (ANOVA) appropriate to a randomized complete block design RCBD. The progeny data were subjected to ANOVAS appropriate for a (RCBD) pertaining to the Design II mating scheme. The F-tests were straightforward for all sources of variation except for the male and female components (Table 3). An approximate F“ test, (Satterthwaite, 1946) was used to test the male and female effects. Mean squares from the ANOVA were used to estimate components of variance pertinent to a Design II mating scheme repeated over environments (Comstock and Robinson, 1948; Cockerham, 1956; Miller et al., 1959; Cockerham, 1963; Hallauer and.Miranda, 1988). Replications, locations and genotypes were considered as random effects in the mathematical model used to estimate variance components (Searle, 1971). The values for the variance 46 components were calculated using the GENSTAT 5 software program (1989) . In the Design 11 ANOVA, the variability due to crosses was broken-out as variation due to males within sets, variation due to females within sets, and variation due to male x female interaction within sets (Hallauer and Miranda, 1988) . Throughout the text, variation due to males within sets, females within sets, and males x females within sets will be referred to interchangeably as GCA males, GCA females, and SCA variation, respectively (Pixley and Frey, 1991) . The components of genetic variance were estimated as: ozIn = 02, -= cov HS = (1/4)ofi , and 03., = Cov FS - Cov HS. - Cov Hs, = (1/4)02,,. Where: 0’, ; 03,, ; 02.; oz, ; and 02,, are the additive and dominance variance and the variance due to GCA males, GCA females, and SCA, respectively. The Cov HS and Cov F8 are the covariances of half sib and full sib progeny. The degree of dominance governing the genes for trait expression was calculated by the formula: d =(202., /o‘.)“ where d = the degree of dominance. Heritability in the narrow sense was estimated from the components of variance values pertaining to a Design II mating scheme as: hzN -= oz, / ozph. 47 Where '02, is the additive genetic variance and c"-ph is the phenotypic variance. The approximate standard error of the narrow sense heritability estimate was calculated as: SE ($2,) $2?“ . Where 3% is the variance components due to progeny and $29,. SE (h2) = is the total phenotypic variance of progeny estimated from variance components (Dickerson, 1969). Response to selection was estimated from the formula: AGC = hZND. Where 41Gc is the genetic gain per cycle, and D is the selection differential. The selection differential is the difference between the mean of the genotypes selected from a population and the overall mean of the population from which they were selected. Since the selection differential can also be expressed as D = k c:ph where k is the selection differential expressed in standard units and oph is the square root of the phenotypic variance (Fehr, 1987), the equation for genetic gain per cycle can be modified by substituting for H2 and D as follows: 3 k oz, 02,. 02,... AG: = HZD = 48 Phenotypic correlations between the cooking time and maturity, protein, tannin, seed weight and water absorption were calculated using the GENSTAT 5 program. RESULTS AND DISCUSSION The means across locations and growing seasons for maturity, protein and tannin content, loo-seed weight, water absorption and cooking time index, of the 16 parents are given in Table 1. Seed weights of the parents ranged from 28.6 to 47.4 g 100’1 seeds. There was a range of 26.8 min in the cooking time between the slowest UAC 221 (52.3 min) and fastest VAR 11 (25.5) cooking parent. The range in water absorption for the 16 parents was 472.5 g kg”1 with CIAT-3005 and Var 11 imbibing the least and most water, respectively. Significant location mean squares were detected for the maturity, protein content and water absorption traits in both the SR and LR growing season and for only tannin content in the SR season. Significant variability between locations led to significant location (L) interactions for some traits. Significant GCA.males x L and GCA females x L mean squares were noted for maturity in the SR season; a significant GCA males x L interaction for protein content in the SR season and tannin content in the LR season. The GCA females x L mean square was significant for seed weight in the SR season. Significant variability between sets was observed for water absorption and cooking time in both SR 49 50 and LR seasons and maturity and seed weight in the SR season. Mean squares for GCA males and GCA females were highly significant for the six traits in both growing seasons (Table 2). A.design II mating scheme provides two independent estimates of GCA, one due to the female and one due to the male parents (Hallauer and Miranda, 1988). Except for the seed weight trait the mean squares for GCA females were larger than the GCA males. The differences between the two GCA reflected the variability for the trait among the parents per se (Pixley and Frey, 1991). The GCA male and GCA female variance components greatly overshadowed the SCA component indicating that the genetic variance influencing trait expression. was preponderantly additive. Although there was a difference in the level of inbreeding of the genotypes grown in the SR and LR seasons, the magnitude of variance components were in good agreement for most traits. The exceptions were the GCA male components for maturity and tannin content and the GCA female components for tannin and cooking time (Table 3). The GCA.female variance component for cooking time was about 8 times greater in the LR than the SR season. When the GCA.males and GCA.females variance components were combined there was good agreement between the SR and 51 LR growing seasons for most traits except for cooking time and tannin content. For cooking time the combined GCA males and GCA females component of variance was 3 times greater in the LR growing season than in the SR growing season accounting for 51% ( i.e. 15.7 + 35.0) and 17% (i.e. 12.3 + 4.4) of the total variance, respectively (Table 3). The combined variance components of these same sources of variation for tannin content in the SR growing season was twice that in the LR season accounting for 76 and 38% of the total variance, respectively. 52 H H o aehH H mm MA v o H H are «aHH mm m H x “my: arm «*vN ram arm H NH mH H ramm «aH ram *«H «N Mm wH Amvh x 2 iemvm «+OHMH aemmm eava «rem «iva mH «aHN rammMH armom aammm «cov «#NMH mm m h.AmvmmHmeh ««VHH *«NNM 4amvN «amH «aHN armm MA *er «*mHm aaHHN aiHm aewH *4MD Mm m E .AmvmmHmz H N H m H h mH m H o H o N mm m qurmvmomm m H NH H o m mH o H o o H 0 Km H AHVm «HmNH *OBFN owH hr m mHh MA ahHHH ammwN aHmH vm w ammb mm H m .mumm MHN «mom MON HHm *MFN *HFNH + MA mmm «mmm mm *va ammm *HomH + mm H A .GOHumOOH AOOH. oaHu coHuouomom brown: unoucoo unoucoo huHHSumz common coHumHum> uconoo nouns ooom oHccma cHouonm mnHzouo up no condom .aommmm seasons lemma. mo can .mmmflc mm on» mcHuoo MHcmNcma .cowmou ouooouoz onu cH mcoHumooH or» no czoum >nououo coon hue you 3335 93.... 023000 one coHuouoQO noun: £an9; ooom .chcmu .oHououo .auHuoumE How ouanum> mo momaHmco oocHnaoo can scum mouooom cmoz .N oHoma 53 .mnomeem mnHzouo nHeH onoH one phone n ma one mm + .Anw>nuomnmmu .mam>ma aufionnaaoun we use mm map on Damonennonm .4 .. r-lr-i NN ON r-lr-I HO OH HN HN H03 H‘O mH *eNH ma Mm ma Mm ma Mm Han Hence oNH uouum 3 f: 3.: 3: m .H x Hm; 1.».cou. "N manna 54 m.m v.e m.o o.o m.H >.N mg v.o H.v m.o m.H v.0 0.0 mm mH Ammo x z o.mm m.~m m.ov v.mm v.om m.vH ma v.v >.mm m.vv m.mw v.w~ H.vH am e Amen >.mH v.vH m.~w o.o H.NH H.m ma m.~H o.vH v.we N.m m.m m.n mm m Amvz o.o H.o o.o N.o m.o H.o mg m.o 0.0 0.0 H.o 0.0 «.0 mm m AHVHme m.o o.o m.o 0.0 0.0 o.o mg 0.0 o.o o.o o.o o.o o.o mm H Loom m.mm H.vH o.o o.o >.Nv m.mH ma o.mm H.NH o.o o.o m.mm H.5H mm H m m.> m.m e.m N.Hm >.Nv N.mm neg o.mH m.v ¢.m m.>H m.em >.Nm new H A AoOHV 08H» noHuouomoe unuHoz uneunoo uneunoo >uHunue2 nOmeem + noHueHue> oonooo Hopes oeom annea nHeuoum onwzouu no no mounom .maommmm Sansone lemma. me one immmH. mm web manage eHneunea .nOHmeu ouomouoz enu nH mnoHueooH ozu ue n30uu >neoouo neon xuo you eEHu monoou one nOHuouomne Hone: .unoHoz omen .anneu .nHeuouo .auHunueB you oOH on fine on oeHeom munenoonoo euneHue> mo moueEHumm .m mHneB 55 .mnomeem onHzouo nHeu onOH one phone n me one am” .moHeEom u m .neHeE n 2 .mnOHueOHHQeu u m .muom u w .mnoHueOoH u A + .mHo>HuueQmow .mHe>eH >uHHHQeoouo wH one wm may we uneOHanmHm «« .« H.0 N.0 m.NH m.NH v.0H ma 0.H m.N m.v >.N v.N mm 0NH Hounm 0.0 m.0 m.MH 0.0 0.0 mg 0.0 0.0 0.0 0.0 m.H mm mHahVAZVAmVH 0.0 H.0 0.0 0.0 m.N mg 0.0 0.0 0.0 0.0 H.N mm 0 H x Hmvh v0.0 5.0 m.m 0.0 b.m mg 0.0 0.0 0.0 m.v m.H mm 0 H x “my: i.e.aouc .m «Home 56 The narrow sense heritability (qu) for protein and tannin content and water absorption in the SR growing season were 0.71 and 0.77, respectively (Table 4), indicating that the environment had little influence on the expression of the traits. The H2N for maturity, loo-seed weight and cooking time in the SR growing season were also high (0.80, 0.98 and 0.97 respectively). In the LR growing season Hfl. for maturity’ was a low 0.56 while water absorption maintained the same HzN of 0.77 as in SR growing season. Protein and tannin content had H2N of 0.88 and 0.91 respectively in the LR growing season. Only loo-seed weight and cooking time traits maintained H2N >0.90 in both SR and LR growing seasons indicating that they were the most highly heritable in both seasons (Table 4). The results obtained in this study for Hfl.estimate for days to maturity (0.56) during LR compare favorably with those reported by Singh, (1990) (0.47) and Cerna and Beaver, (1990) who reported H2N of 0.31 - 0.63 for the trait. The results reported by Scully, et al.,(1991) who obtained a HQ estimate of 0.96 for days to maturity were much higher than those reported in this study. 57 Table 4. Estimates of additive and dominance variance, broad and narrow sense heritability and the degree of dominance for protein, tannin, maturity and seed weight for dry bean progeny grown at two locations in the Morogoro region, Tanzania during the SR(1993) and LR(1994) growing seasons. Character Growing 01,3 023* H‘MS Degree of season daninance Maturity SR 10.9 2.7 0.80 10.15 0.7 LR 4.3 3.8 0.56 $0.03 1.3 Protein SR 2.1 0.2 0.71 10.22 0.4 LR 3.2 0.4 0.88 10.04 0.2 Tannin SR 4.8 1.4 0.77 10.03 0.8 LR 26.2 2.7 0 91 10.11 0.0 loo-Seed SR 35.0 0.4 0.98 $0.04 0.2 weight LR 40.2 0.4 0.98 10.02 0.2 Water SR 48.9 14.5 0.77 10.05 0.8 absorption LR 49.6 15.2 0.77 10.04 0.8 Cooking SR 9.5 0.3 0.97 $0.10 0.2 time LR 18.1 2.7 0.90 10.02 ‘ 0.5 aflt additive genetic variance o‘pt dominance genetic variance H’MS narrow sense heritability with the standard errors SR and LR = short and long rain growing seasons. 58 The Hfl.estimate obtained in this study for seed weight was high (0.98) in both growing seasons, but Singh et al., (1989) reported a lower value of 0.61 for the trait. Except for maturity and tannin content in the LR growing season, all the six traits showed a degree of dominance lower than 1.0 but greater than zero (Table 4). The magnitude of degree of dominance indicated that the traits were governed by genes with partial dominance. For a given trait, no progeny had a higher or lower mean than the highest and the lowest parent. The degree of dominance for tannin content in the LR growing season was zero indicating no dominance. For maturity in the LR growing season the degree of dominance was 1.3 indicating overdominance. The magnitude of the degree of dominance in conjunction with gene frequency influences the magnitude of the genetic components of variance in a population (Falconer, 1981). Although knowledge of the degree of dominance is important to the .breeder, the dominance relationship of genes probably is of limited usefulness in P. vulgaris, a self-pollinated crop. Bean cultivars are released to commerce as pure lines. In a self-pollinated crop, the frequency of heterozygous individuals declines progressively with each generation of inbreeding, and the frequency of homozygous individuals increase, regardless of 59 the effectiveness of selection (Fehr, 1987). Dominance has its greatest utility in crops released as hybrid varieties. The means of the progeny grown in the SR and LR were essentially equivalent for cooking time(Table 5). However the range in cooking time was greater in the LR (16 min) versus the SR (10 min). The ranges in cooking time in the SR and LR while not of a great magnitude should be sufficient enough to reduce cooking time through selection and, thus lead to a conservation of fuel wood in improved lines. Shellie-Dessert and Hosfield (1991) showed an average fuel wood savings of 1.3 kg per cooking session associated with a 15-min reduction in bean cooking time. Consumers of beans in most bean producing East African countries build 14 fires per week for cooking and reheating the food (Shellie-Dessert and Hosfield, 1991). Given a 1.3 kg fuel wood savings per cooking session for beans, the annual fuel savings would be 1050 kg. Since the cooking time of dry beans is of critical importance to energy use and the overall preservation of fuel wood in Tanzania and other bean growing regions of Eastern Africa, even a small reduction of bean cooking time should lead to a greater bean consumption and thus improve the nutritional well- being of consumers. r 60 .mnomeem onHzouo nHeH onOH one uwonm u «A one mm + m H m m m m mg m m m N SH 0 mm 3:6 N HH H mm vH m Mn N mm H HH 0v v mm . Amo.ovnmn hlem 0.mhm|>.0>0 0vnmm mmvimmm mmmumow omihm ma mvimm 0.mmmlm.0v0 omuvm Hbvlmmm ovmubmH mmlmm mm monem mm mum vv mom mvm mp ma mm 0H0 we won emm we mm . Amvnemn nd: 79. m 0 0 700H on Tax m n>eo .oOHV oEHu noHuouonoe uanmz uneunoo unounou >uHuDuez H nomeom oonoou Houez oeom annea nHouowm 0nHzouw .nomeom mnHzouo .vmmvaA one .mmmH. mm onu nH eHneunee .noHueu ouooouoz ecu nH mnOHueooH 03» ue nzouo anemone neon auo How auHHHoexooo ou ooueHeu me noHuouomoe Hope: one £3363 oeom .anneu .nHeuouo SuHunuen pom mouneu one mneoz .m 933. 61 Phenotypic correlation coefficients between the cooking time and water absorption (based on 192 observations) were high and negative, (-0.87* and -0.78*, in the SR and LR growing seasons, respectively) and consistent across growing seasons. The results obtained in this study were in agreement with those reported by Castellanos et al., 1995 who obtained correlations on non scarified seed of between -0.69** and -0.81** for cooking time and water absorption after an 18 hour soak. However the magnitude of the correlations between cooking time and water absorption obtained in this study and that by Castellanos et al., 1995 differs from the -0.37* value reported by Shellie and Hosfield (1991) for the same traits, based on data from 270 observations. The negative correlations found in the present study indicated that slow cooking beans imbibed less water than fast cooking beans. The magnitude of the correlations between cooking time and water absorption in this study and that of Castellanos et al., 1995 suggested that the percent water absorption trait should be useful to predict or estimate cooking time in beans. Selection based on the water absorption of a breeding. line as an indirect estimation of its cooking time as opposed to measuring the cooking time per se can save valuable resources. 62 The phenotypic correlations between protein and tannin content and cooking time and protein content were negative and non-significant (Table 6) but the correlation between cooking’ time and tannin. was significant and. positive indicating that beans with low tannin content cook faster than beans with high tannin content. Table 7 shows the selection differential and the estimated gain in selection for days to maturity, protein, tannin, and for seed weight. The data indicate that several cycles of selection are needed in order to change protein content, seed weight and water absorption in beans and lower tannin content, days to maturity and cooking time to a point where practical benefits might be noticeable. The estimated. gain in selection for cooking time at 25% selection pressure was about two minutes per cycle while that for the water absorption trait was 60 g kg"1 per cycle on a weight basis (Table 7). In order to reduce cooking time in beans to a point where a practical benefit might accrue (e.g. 10 minutes in this study), several cycles of selection may be needed. If selection for reduced cooking time is based on the water absorption trait, five years of selection would increase water absorption by 301 g kg‘1 (about 300 mg per seed). A.300 mg increase in seed weight due to water absorption is almost double the seed weight of 63 the dry seed. Physical considerations of size impose an upper limit to the amount of water it can absorb and thus, the weight it can attain. .eueo mnomeem oanouo nHeu 0n0H ngv u oHonnnon .eueo unoneom unwkouu nHeH phone Humv I UHom Amo.o V my on ucaonencanm aconumHmunoo , I ipm.o- Ten.ou ,om.o- iam.ou mo.o- nonoonomnm neuez «m>.0n n 00.0 «vm.0 400.0 no.0: onHu monooo no.0- vo.o- . *eH.o mo.o *mv.o Deana: ooom eHm.0| ahm.0 ch.0| i +Nv.0 *vm.0l auHHnueZ «h0.0 35>.0 00.0 amv.0 I 00.0: unounou annee «0.0 «H.0: «no.0 amm.0| 00.0: n unounoo nHououm noHuouomoe eEHu unonz muHunuez uneunoo unounoo Hones oonooo oeom anneB nHououm uHeua .nomeem 0nH30u0 mnHew onOH one mnHeu uuonm on» onHuno eHneNnee .nOHmou ouomonoz enu nH mnOHueooH 03D ue nzouo anemone neon moo now oEHu oonooo one unoHez oeem .auHunuen .anneu .nHououo noe3uoo muneHoHumeoo nOHueHouuoo OHoxuonenm .0 oHoea 65 Table 7: Estimation of response to selection for maturity, protein, tannin, and seed weight, water absorption and cooking time for dry bean progeny grown at two locations in the Morogoro region, Tanzania during the short rains (1993) and long rains (1994) growing season. Trait x , t 52 . t H’.§ so he) Maturity (days) 74 75 0.56 0.3 0.2 Protein (9 kg”) 232 244 0.88 12.2 11.4 Tannin (mg 10049) 395 373 0.91 21.9 -19.9 loo-Seed weight(g) 42 44 0.98 1.7 1.7 Water absorption 744 823 0.77 78.7 60.1 (9 kg“) Cooking time (min) 41 39 0.90 2.0 -1.8 x ,t mean of the population x .4 mean of the individuals for selection at 25% selection Hfl,§ narrow sense heritability combined over seasons SD 1 selection differential LG 4 genetic gain in selection 66 Breeding implications. The significant genetic variation for the six traits found in this study indicated that they can be changed by selection. Recurrent selection (RS), single-seed descent (SSD) or pedigree methods may' be used in population improvement. However, RS takes two years per cycle; hence it would take about ten years to make any noticeable progress for most of the traits. Progress in population improvement depends on the level of heritability of the trait which is used to estimate genetic gain in selection. The amount of genetic gain realized for all traits except maturity were of sufficient magnitude for population improvement. Generally when narrow sense heritability is low for a trait(s) RS is an efficient method to improve a population since this procedure results in an increase in gene frequencies of the desirable trait(s). Since the traits evaluated in this study all had high narrow sense heritabilities, RS as a breeding strategy to improve them. would probably be inefficient and used as a last resort although in some cases this procedure may provide the only means for making genetic progress for some of the traits. When the traits in this study are considered, the breeder may wish to practice one cycle of recurrent selection to increase gene frequency 1” 67 of favorable alleles. Recurrent selection was successful in changing the growth habit and improving seed characteristic and several food quality traits in small-red market class dry bean germplasm (Hosfield et al., 1995). The SSD procedure is another selection strategy that has been successful in improving quantitative traits in self-pollinating crops (Brim, 1966). The major future of SSD is to make an initial selection (single seed) from.each plant in the F}. No selection is practiced until the F3 or 1% when a reasonably high degree of homozygosity is assured. The F2 derived F5 or F6 lines are then subjected to intense evaluation. Although SSD saves labor and time it might not be the most suitable selection procedure to improve food quality traits in bean exhibiting high heritabilities. The SSD method is most efficient for low heritability traits such as yield. In identifying the most suitable breeding approach one needs to consider efficiency. Since most of the traits in this study had a moderate to high heritability, pedigree or a combination thereof with SSD would be the most efficient breeding strategy because it allows the breeder to practice selection in early‘ generations without increasing the length of time for cultivar development (Fehr, 1987). The fact that the pedigree method is labor intensive and 68 requires considerable record keeping is no longer a tenable argument against the procedure since high speed electronic computers are used for keeping record. The breeder might select the best E} derived.F3 or F2 derived F.1ines and evaluate them in replicated trials. For highly heritable quality traits like water absorption and cooking time the pedigree procedure beginning with F2.3 or Fm lines might prove useful. If the breeder chooses a combination of inbreeding methods to improve the traits evaluated in this study, single seed selection should be started in the F3 for traits such as maturity, seed size. Further selection among the lines for traits such as yield could be done at the F6, when the individual lines are more homozygous. The selected F6 lines would be replicated, and grown in different locations and seasons or years. The advantage of using a combination pedigree-SSD or pedigree- SSD-RS breeding strategy to improve the traits evaluated in this study, is that when selection is effective inferior bean genotypes may be discarded long before homozygous lines are evaluated in costly replicated tests. 039 of'correlated.responses in breeding strategy. The significant and positive correlation (0.77*) obtained in this study between tannin content and cooking time indicated that beans low in tannin cooked faster than 69 beans with high tannin. The selection and intermating of fast cooking genotypes during the inbreeding process of cultivar development would lead to cultivars with reduced tannin content. The intermating of fast cooking genotypes with high protein content would be useful to improve the three traits simultaneously and lead to the development of cultivars with low tannin, and high protein content that are fast cooking. Although this strategy would lead to improved nutritional quality by increasing protein and decreasing tannin the breeder should proceed with caution and. be cognizant of seed coat color. Consumers have preference for particular colors of bean seed coats and seed coats with darker colors such as purple and deep red hues generally are high in tannin (Ma and Bliss, 1978a and 1978b). Whether or not the association between some seed coat colors and tannin is due to pleiotropy or tight linkage is unknown and requires study. LIST OF REFERENCES Association of Official Analytical Chemists 1975. Official methods of analysis. 12th ed. AOAC, Washington, D.C. Bart, Fr.(1981). Le Paysan Rwandais et l'energie. In: M; De Lame(Eds). Societe, culture et Histoire du Rwanda. Encyclopedie Bibliographique 1963-1980/87. Koninklijk museum voor Midden-Africa. Musee royal de l'Afrique centrale. Tervuren, Belgium, p. 178. Brim, A. c.1966. A modified pedigree method of selection in soybeans. Crop Sci. 20:507-510. Bressani, R. and L. G. Elias. 1980. Polyphenols in cereals and legumes. IN: J. H. Hulse (ed.) IDRC, Ottawa, Canada. p. 61. Bressani, R. 1975. In Nutritional improvement of food legumes by breeding, edited by M. Milner, John Wiley and sons. New York, N. Y. Bressani, R. 1993. Grain quality of common beans. Food Reviews Intern. 9(2):237-297. Burns, R. E. 1963. Method of tannin analysis for forage crop evaluation. Tech. Bull. N. S. 32, Georgia Agrc. Exp. Stn..Athens. Castellanos, J. 2., H. Guzman-Maldonado, J. A. Acosta- Gallegos and J. D. Kelly. 1995. Effects of hardshell character on cooking time of common beans grown in the semiarid highlands of Mexico. J. Sci. Food Agric. 69:437-443. Cockerham, C.C. 1956. Analysis of quantitative gene action. Brookhaven Symposia in Biology. 9:53-68. Cockerham, C. C. 1963. Estimation of genetic variances. In Genetics Statistics and plant breeding, W. D. Hanson and H. F. Robinson(eds.) Nat. Acad. Sci. Nat. Res. Coun.Pub. 982:53-94. Compton, W. A1, and R.E. Comstock. 1976. More on modified ear-to-row selection in corn. Crop Sci. 16:122. 70 71 Comstock, R. E. and H. F. Robinson. 1948. The components of genetic variation in population of biparental progenies and their use in estimating average degree of dominance. Biometrics 4:254-266. Deshpande, S. S., S. K. Sathe, D. K. Salunkhe, and D. P. Cornforth. 1982. Effects of dehulling on phytic acid, polyphenols, and enzyme inhibitors of dry beans (Phaseolus vulgaris L.). J. Food Sci. 47:1846. Dickerson, G. E. 1969. Techniques for research in quantitative animal genetics. In techniques and procedures in animal science research. Am. Soc. Anim, Sci. p. 36-79. Edmister, J. Am, W. M. Breene, As Serugendo, 1990. Influence of temperature, water activity and time on cookability and color of a stored Rwandan dry bean (Phaseolus vulgaris L.). J. Stored Prod Res 26(3): 121-126. Elias, L. G., D.G. de Fernandez and R. Bressani. 1979. Possible effects of seed coat polyphenolics on the nutritional quality of bean protein. J. Food Sci. 44:524-527. Falconer, D.S. 1981. Introduction to quantitative genetics. 340p. Longman Scientific & Technical Fehr, W. R. 1987. Principles of cultivar development. V01.1 Theory and technique. 536p..McGraw-Hill, Inc. GENSTAT 5. 1989. A.statistical software program from Rothamsted Experimental Station, Harpenden, Hertfordshire AL5 2J0. Hallauer, A, R. and J. B. Miranda. 1988. Quantitative genetics in maize breeding. 468p. Iowa State university'Press/Ames. Hernandez-Unzon, H. Y. and M. L. Ortega-Delgado. 1987. Water absorption and cooking time in long stored common bean seeds (Phaseolus vulgaris L.). Annu. Rept. Bean Improv. Coop. 30:58-59. 72 Hosfield, G. L., J. D. Kelly, M. J. Silbernagel, J. R. Stavely, M. W. Adams, M. A" Uebersax and G. V. Varner. 1995. Eight small-red dry bean germplasm lines with upright architecture, narrow profile, and short vine growth habit. HortSci. 30(7):l479-1482. Howe, J. W. And F. A. Gulick. 1980. Fuel and other renewable energies in Africa - a progress report on the problem and the response. Overseas Development Council, Washington D .C. p. 43. Jackson, G. M. and E. Varriano-Marston. 1981. Hard-to-cook phenomenon in beans: effect of accelerated storage on water absorption and cooking time. J. Food Sci. 46: 799. Jaffe, W.G. 1973. Factors affecting the nutritive value of beans. In.M; Milner (ed.). Nutritional improvement of food legumes by breeding. Protein Advisory Group of the United Nations System, New York. p.43-48. Kelly, J. D. and F. A. Bliss. 1975. Heritability Estimates of Percentage Seed Protein and Available Methionine and Correlations With Yield In Dry Beans. Crop Sci. 15:753-757. Ma, Y. and F. A. Bliss. 1978a. Tannin content and inheritance in common bean. Crop Sci. 18:201-204. Ma, Y. and F. Am Bliss. 1978b. Seed proteins of common bean. Crop Sci. 18:431-437. Miller, P. A., J. C. Williams and H. F. Robinson. 1959. Variety x environment interactions in cotton variety tests and their' implications on testing :methods. .Agron. J. 51:132-134. Mora, M. A. 1982. The influence of different temperatures and moisture contents on the cooking time of beans (Phaseolus vulgaris L.) Stored during 18 months [storage conditions]. Influencia de diferentes temperatura y contenidos de humedad sobre el tiempo de coccion de frijol (Phaseolus vulgaris L.) almacenado durante 18 meses. Agro. Costarric. 6(H):87-89. 73 Moscoso, W., M.C. Bourne and L. F. Hood. 1984. Relationship between the hard to cook phenomenon in red kidney beans and water absorption, puncture force, pectin, phytic acid, and minerals. J. Food Sci. 49:1577-1583. Nienhuis, J. and Singh, S. P. 1988. Genetics of seed yield and its components in common bean (Phaseolus vulgaris L.) of Middle American origin. II. Genetic variance, heritability and expected response from selection. Plant Breeding- Z-Pflanzenzucht. 101(2):155-163. Paredes-Lopez, O., E. C. Maza-Calvino, and J. Gonzalez- Castaneda. 1989. Effect of the hardening phenomenon on some physico-chemical preperties of common bean. Food Chem. 31(3): 225-236. Patel, K. M., C. L. Bedford and C. W. Youngs. 1980. Amino acid and mineral profile of air-classified navy bean flour fraction. Cereal Chem. 57:123-125. Pixley, K. V. and K. J. Frey. 1991. Combining ability for test weight and agronomic traits of Oat. Crop Sci. 31:1448-1451 Reddy, N. R., M. D. Pierson, S. K. Sate and D. k. Salunkhe. 1985. Dry bean tannins: a review of nutritional implications. J. Amber. Oil Chem. Soc. 62:541-549. Sahasrabudhe, M. R., J. R. Quinn, D. Paton, C. G. Youngs and B. J. Skura. 1981. Chemical composition of white bean (Phaseolus vulgaris L.) and functional characteristics of air-classified protein and starch fractions. J. Food Sci. 46:1079-1088. Sathe, S. K., V. Iyer and D. K. Salunkhe. 1981. Functional properties of the Great Northern bean (Phaseolus vulgaris L.) Proteins. Amino acid composition, in vitro digestibility, and application to cookies. J. Food Sci. 47:8-11. Satterthwaite, F. E. 1946. An approximate distribution of estimates of variance components. Biometrics. 2:110- 114. Scully, B. T., D. H. Wallace, and D. R. Viands. 1991. The heritability and correlation of biomass, growth rates, harvest index, and phenology to the yield of common beans. J. Am. Soc. Hort. Sci. 116(1):127-130. 74 Searle, S. R. 1971. Topics in variance component estimation. Biometrics 27:1-74. Sefa-Dedah, S. and D. W. Stanley. 1979. Textural implications of the microstructure of legumes. Food Technol. 53:77-83. Shellie-Dessert, K. C. and F. A. Bliss, 1990. Genetic improvement of food quality factors. In: Common Beans: Research for Crop Improvement. (eds.) .A. van Schoonhoven and O. Voysest. CIAT. Cali, Colombia. Shellie, K. C. 1990. Food quality and fuelwood conservation of selected common bean (Phaseolus vulgaris L.), cultivars and land races in Rwanda. Ph. D. Thesis. Michigan State University. East Lansing. Shellie-Dessert, K. C. and G.L. Hosfield. 1991. Genotype x environmental effects on food quality of common bean: resource-efficient testing procedures. J. Amer. Soc. Hort. Sci. 116(4):732-736. Singh, S. P., R. Lepiz,, J. A. Gutierrez, C. Urrea,, A. Molina, H. Teran. 1990. Yield testing of early generation populations of common bean. Crop Sci. 30(4):874 878. Sirven, P. 1981. Le role des centres urbains dans la deforestation de la campagne Rwandaise. In: Energie dans les communautes rurals des pays du tiers monde. p. 243-254. Wassimi, N. N., G. L. Hosfield, and M. A. Uebersax. 1988. Combining ability of tannin content and protein characteristics of raw and cooked dry beans. Crop Sci. 28:452-458. CHAPTER TWO THE EFFECT OF STORAGE TIME, TEMPERATURE AND RELATIVE HUMIDITY ON THE COOKABILITY AND SEED QUALITY TRAITS IN DRY BEAN (Phaseolus vulgaris L.). ABSTRACT This study was conducted to ascertain the effect of prolonged storage, high temperature and high humidity on seed. germination and. cooking time of dry' beans representative of the Andean gene pool of P. vulgaris. The study also investigated the effects that accelerated aging had on beans compared to those stored under unfavorable environmental conditions. Genotypic differences were observed for seed germination, water absorption, and cooking time. Highly significant differences were noted for temperature, humidity, and storage time effects. A genotype x storage interaction was noted for each of the three traits. A significant interaction was detected between temperature and relative humidity for cooking time indicating that cooking time for the eight genotypes was not linear across storage temperatures and humidities. A second order interaction of genotype x storage x temperature for water absorption trait was also detected. Seeds stored in high temperatures took longer to cook. When seeds were stored under the high humidity, they 75 76 absorbed less water and had a lower seed germination but took longer to cook than seeds stored at 40% RH. The mean germination percentage decreased significantly with increasing storage time. The results of this study indicate that seed water absorption is inversely related to the cooking time in dry beans. Experiment II dealt with the investigation of accelerated aging as a technique to mimic conditions associated with unfavorable storage of sixteen genotypes of dry bean. Significant differences existed for cooking time index in the entries grown in the SR and LR growing seasons. Significant differences were observed between the two seed conditions (non accelerated aged seed and accelerated aged seed for progeny grown during the SR and LR growing seasons. Differences between the soak and non-soak method for the cooking method of the progeny grown during the SR and LR growing seasons were also significant. Several interactions were detected. The accelerated aging genotypes took longer to cook than those that stored under adverse conditions for three months. The mean performance of seed stored for six months was 54.8 while that for accelerated aging was 53.7. In this study accelerated aging mimicked the 6 month storage. INTRODUCTION Dry bean (Phaseolus vulgaris L.) Is the leading food legume with an annual world production of 14 million metric tons. This production accounts for more than 30% of the total world food legume production of 49 nullion metric tons. Dry bean production accounts for more than 30% of the total world grain legume production of 49 million metric tons. Latin American countries produce 4.0 million metric tons (van Schoonhoven and Voysest, 1993), while Africa produces 1.4 million metric tons. Throughout Latin America and Africa, beans are a staple in the diets of consumers. Dry bean is a good source of calories and protein in human diets. Bean protein complements cereal protein in diets where little meat protein is available (Antunes and Sgarbieri, 1979). The mean dry bean consumption in the Central and Eastern Africa region is 31 kg. This consumption is about double that of the 13 kg/capita in Latin America (van Schoonhoven and Voysest, 1993). In Africa, the largest per capita bean consumption is in Rwanda ( > 50 kg per year) which is four times that in Tanzania. In developing countries beans are often inadequately stored; hence, considerable storage loss occurs. It has 77 78 long been known that dry beans differ in ability to imbibe water. Gloyer, (1921) described two conditions in which beans failed to hydrate after soaking. One kind of failure of water imbibition called “hardshell” was due to the impermeability of the seed coat to water. The second condition affecting water uptake in bean seeds referred to as sclerema (Gloyer, 1921) was due to the inability of the cotyledon to take up water and expand. Snyder (1936) confirmed the condition of sclerema and observed that bean cotyledons would not hydrate even though the seed coat was scarified or removed. The loss of cookability of beans stored under tropical environments in developing countries may be due to one or both phenomena. Castellanos et al., (1995) reported that “hard shell” is a major contributor to the long cooking time of beans grown in semiarid areas of Mexico. Considerable economic loss occurs in Latin America due to bean hardening during prolonged storage (Mejia, 1979). On the other hand, Jones and Boutler (1983); Burr et al., 1981; observed sclerema which rendered beans hard-to-cook and indigestible. Both “hard shell” and sclerema negatively affect bean seed quality. Seed quality of a food legume is viewed from two perspectives: 1) characteristics associated with seed 79 germination and seedling vigor and 2) characteristics that have a direct impact on human nutrition and those related to consumer acceptance and preparation for eating. In dry bean, most growers in developing countries save part of the current season's bean crop to plant in the subsequent year. Hence, seed germination is an important economic consideration. The ability of beans to imbibe water and cook in a reasonable length of time is important to the nutrition well-being of consuming populations in regions where beans are a staple food. Since seed quality in dry bean is often dependent on its post harvest history, an additional objective was to determine the effect of storage conditions on seed quality. Specific objectives were: 1) show that prolonged storage and high temperature, and high humidity alter seed germination and cookability of the grains and 2) ascertain whether accelerated aging of seeds can be used as a technique to mimic conditions associated with unfavorable storage conditions. 80 Table 1: Production and consumption figures for major bean growing regions. Region and Percent of Mean per capita country total production consumption(kg/year) Latin America Brazil 55 20.1 Mexico 23 12.6 Argentina 5 2.9 Paraguay 2 24.3 Nicaragua 1 23.8 Total production for (Latin America) 4 x 10‘tones 13.3 .Africa East and Central 59 Uganda 13 29.3 Kenya 9 21.0 Rwanda 13 50.6 Tanzania 11 . 12.0 Burundi 12 44.3 Total(East and Central Africa) 0.8 x 10‘tones 31.4 From: Shellie-Dessert and Bliss, 1993. MATERIALS AND METHODS Genetic material. This study was conducted in two experiments: mm It dealt with the investigation of the effect of prolonged storage, high temperature and high humidity on seed germination and cookability of eight genotypes of dry bean. The eight genotypes were representative of the Andean germplasm pool of P. vulgaris and named: Var 11, Nyirakizungu, Lyamungu, Kilyumukwe, Yellow eye, UAC 221, TMO 959, and Kalima. mm It investigated accelerated aging as a technique to mimic conditions associated with unfavorable storage. In this experiment sixteen genotypes of dry bean, were used. These included the same eight genotypes used in experiment I and CIAT 3005, P.’I. 605621, GLPX 1125, Jacob's cattle, Montcalm, NY 99, Sierra, and Diacol Calima. Except for ‘sierra', the additional eight genotypes were of Andean germplasm pool origin. ‘Sierra’ was a representative of the middle-American gene pool with the medium pinto market class seed size. The dry bean genotypes used as entries in the two experiments differed morphologically, phenologically, and agronomically. The entries were all adapted to the bean production areas of East and Central Africa. Seeds of each 81 82 entry are highly favored by consumers in the region because of their meat-like appearance and thick broth/soup quality after cooking. Procedures. The experiments on which data were taken were grown in 1993 and 1994 in Tanzania at the Crop Museum(530 masl) and Morning Site (~1000 masl) research farms in the Morogoro Region. Data for aging experiments were combined over locations and growing seasons. These farms used in this study were maintained by the Agricultural University of Sokoine. The soil characteristics at Morogoro are moderately acidic (Appendix Table 1) . Rainfall, temperature and solar radiation are given in Appendix Table 22. The experiments were planted in a randomized complete block design with three replications at each location during the seasons in which the annual bimodal distribution of rain occurs. The short (SR) rain growing season occurs between the end of September and the middle of December. This growing season is suitable for fast maturing bean genotypes as well as other quick maturing crops. Yield of slow maturing bean genotypes generally give low yield during this growing season. The long (LR) rain growing season is characterized by more rainfall, and for longer time (early March to June. Both early and late maturing 83 bean genotypes are grown during this season. The yield is much higher than in SR growing season. However farmers have to be cautious about the planting date since early planting of beans may result in poor yield due to excessive rainfall. The entries were hand seeded into four row plots. Rows were 4 m long and spaced 0.5 m apart. The within row spacing was 0.1 m giving an approximate plant density of 160 plants per plot. On farm practices for herbicide, pesticide and fertilizer applications were followed. Mature plants were harvested by hand from the middle two rows of individual plots when the pods were dry enough to be threshed. Wm. Freshly harvested seed of each entry were cleaned and size-graded. Beans selected for use in the study passed through a 0.675 cm. sieve but were retained on a 0.476 cm. sieve (Bourne, 1967). To prevent mold growth (Sefa-Dedah, 1979) the beans were treated with sorbic acid in absolute methanol (1:8 w/v) per each kg of beans (Monsanto, 1992). The methanol was applied with a thin layer chromatography sprayer using air as the propellant. Seeds were mixed adequately during the treatment. The treatment of beans with sorbic acid was repeated after four and a half months. 84 Beans were stored in desiccators at 43% and 73% relative humidity (RH). The 43% RH was maintained by using saturated solution of potassium bicarbonate while that of 73% RH was maintained by using a saturated solution of sodium chloride. As an additional safeguard to prevent mold growth, copper sulfate was added to water at a concentration of 1 gram per liter (American Society for Testing Materials, 1951) . The design of the experiment was a split plot in which the cultivar was the whole plot. Storage time, temperature, and relative humidity were sub-plots. The seed lots were stored at two levels of relative humidity for nine months in desiccators, in Gallenkamp temperature-controlled incubators set at three temperature levels i.e. at 20 °C 3: 2 °C; 25 °C :1: 2 °C and 30 °C i 2 °C under the following sets of conditions with three replicates per treatment: I) 20 °C, 40% RH; ii) 20 °C, 73% RH; iii)25 °C, 40% RH; iv) 25 °C, 73% RH; v) 30 °C, 40% RH; and vi) 30 °C, 73% RH. W. Accelerated aging is a technique which involves aging seeds for short periods of time, for example between 3 to 14 days depending on the crop species(Delouche and Baskins, 1973) . The seeds are aged at high temperatures between 41°C to 45°C and a relative humidity of 100%. During accelerated 85 aging, seeds deteriorate comparable to that of seed lots stored for several months under adverse conditions. The seed for accelerated aging study were obtained and pretreated for mold control. The seed lots for the experiment were divided into a control sample (freshly harvested but not aged) and seed sample that underwent accelerated aging. The control and aged seed were cooked with and without soaking. For the soaked sample, beans were soaked for 12 hours. The treatment comparisons of control vs aged seed was as follows: 1)fresh harvested seed, non-soak; 2)fresh harvested seed, soak; 3)accelerated aged, non-soak: 4) accelerated aged, soak The aging of seed was carried out at 42°C, at 100 RH for 72 hours. A single sample accelerated aging chamber was used to age the seed lots in this experiment. The outer chamber had an immersion type of heating element (Stultz Scientific Equipment Co, Springfield, IL) that was immersed approximately 5 cm deep in the water reservoir. Temperature was controlled by a Thermistemp temperature control unit. The model used (model 71-A) was capable of controlling temperature to CLIPC. On the side wall of the chamber, a mercury thermometer was inserted to 86 about 12.7 cm. into the outer chamber. The inner chambers or seed containers were constructed of plastic. The brass or bronze wire mesh baskets that were used to contain the seeds were locally made to sizes that could accommodate 210 seeds of each. of the entries that were used. in the experiment for each treatment. £ha;agter_eyalnatign. The actual moisture contents of the seed samples was not determined, but prior to analyses of water absorption and cookability the treatment samples were equilibrated for moisture to each other. This was accomplished by storing the samples for two weeks in sealed plastic buckets held at 20°C and 73% RH. Seeds were kept as per field replications. Seed were analyzed at three,six and nine months. After each storage period (i.e. 3, 6, and 9 months) germination of each genotype of each treatment was determined on a 400 seed sample that was allowed to germinate on moist filter paper. The amount of water absorbed by each genotype was calculated on triplicate samples of a known weight of 75 seeds from each plot. The seeds were soaked in distilled water for twelve hours. The amount of water absorbed was taken as the difference in weight before and after soaking divided by the difference in the weight of unsoaked seeds and expressed as a percent. 87 Cookability was determined as the time it took 25-seed sample of beans to cook to a point of softening where a weighted plunger would penetrate the seeds (Jackson and Varriano-Marston, 1981). Beans were considered cooked, when 13 of the 25 pins of the apparatus had penetrated through the seeds. Data were taken on triplicate samples for each plot. Statistical_flnalysi§..All data were subjected to analyses of variance (ANOVA) appropriate to a split plot design. The cultivar was the whole plot and the storage time, temperature and humidity were sub plots in the splits. Data analyses were conducted using the MSTATCR statistical software program (experimental model number 34). RESULTS AND DISCUSSION W Genotypic (G) differences among the eight entries were observed for seed germination, water absorption, and cooking time (Table 2 and Appendix Table 17, 18, 19). There were highly significant differences noted for effects of temperature (T), humidity (H), and storage time (S). The significant storage time effects led to several significant interactions. Significant interactions between genotypes and storage effects (G x S) indicated a lack of uniformity in strain response over storage times. There was a genotype x storage interaction noted for each of the three traits. A. significant temperature x humidity interaction was detected for cooking time indicating that cooking time responses of the eight genotypes was not linear across storage temperatures and humidities. A second order interaction of G x S x T for water absorption trait was detected. The significant second order interaction for water absorption demonstrated that water absorption for the eight genotypes was not linear across storage times and temperatures (Table 2). 88 89 .aHo>Huoemweu .mHo>eH .m.0 v onHe> e>HuHmoo HHenm + HDHHHaenoHQ eH nee we men on LeeUHeHemHm .. .i Hmv Heuoe v mH b ovN nouum N 0H 0 0N m x a x m x 0 N 0H H v m x a x m H 0H m 0H m x a x w evH H0 H N m x a N 0H 0 0H m x w x 0 0H 0H m N m x w v 0H m h m x 0 ranch «10mm ammw H HmVMHHUHSm.m m evN 0 mm a x m x u 0 H 0H 0 a x m 0 mN 5 0H 9 x w eemmOH ieeHoH eeeemH N LecwnsbeneQEma m mH 0 mm Hocuouum «awn. «comm Limo VH m x U eemNovH aeHNvm eammHHH N HmcenHu omeuoum 0 FN h «H Hevuouwm ..meoe eememNH .imee e iwcmomuoemo 0 HN m N noHueOHHoom enHu nOHuowomne nOHuenHfiuoo oonoou Hopes ooom m no mounom .auHoHann e>HueHeu one ounueneonou onH>ue> ue ooewoum enunoe ean ue>o muHeuu eEHu oonoou one noHuouomne nope: .noHuenHeueo omen mom moHuuno neon hwo 0 now ooneHue> mo momaHene noun mouenom neez "N eHnme 90 .An examination of the Table of means (Table 3)indicated that dry bean seed stored at the 20°C temperature had a significantly higher percent germination, and absorbed more water than seed stored at the 30°C temperature. Seeds stored in high temperatures also took longer to cook. The results of this study agree with. previous work (Jackson and Varriano-Marston, 1981; Castellanos et al., 1995) that seed water absorption is inversely related to the cooking time in dry beans. A range of 38 min in cooking time was observed among the genotypes across the different temperatures used to store the seed. The range in percent water absorption and seed germination across storage temperatures were 51% and 14% respectively (Table 3). When seeds were stored under the 73% RH regime, they absorbed less water, had a lower seed germination, and took longer to cook than seeds stored at 40% RH (Table 4). Table 5 shows the effect storage times averaged over temperature and humidities while Table 6 shows the mean performance for the three traits of the eight genotypes stored for nine months under varying temperature and relative humidity conditions. 91 Table 3: Mean germination percent, water absorption, and cooking time of seed stored at 20°C, 25°C and 30°C and averaged over eight genotypes, two relative humidities and three storage times. Storage Seed T Water t Cooking temperature germination absorption time (%) (%) (min) 30%? 81a 62a 57c 25°C 84b 65b 54b 20°C 87c 68c 51a Mean 84 65 54 Range (77-90) (33-85) (33-71) LSD (0.05) 0.8 1.1 0.6 C.V.(%) 3.2 6.0 3.8 t Means with the same letter in a column are not significantly different at the 0.05 probability level. 92 Table 4: Mean germination percent, water absorption, and cooking time of seed stored at two relative humidities and averaged over eight genotypes, three temperatures and three storage times. Relative Seed 1 Water t Cooking t humidity germination absorption time (%) (%) (%) (min) 73 83a 64a 55b 40 86b 66b 53a Mean 84 65 54 Range (78-89) (34-83) (34-68) LSD (0.05) 0.5 0.8 0.4 C.V.(%) 3.2 6.0 3.8 1 .Means with the same letter in the column are not significantly different at the 0.05 probability level. 93 Table 5: Mean germination percent, water absorption, and cooking time of seed stored at three storage times and averaged over eight genotypes, two relative humidities and three temperatures. Storage Seed t Water t Cooking t time germination absorption time (months) (%) (%) (min) 3 93a 73a 44c 6 85b 65b 55b 9 75c 57c 64a Mean 84 65 54 Range (69-96) (32-97) (28-79) LSD (0.05) 0.8 0.6 1.1 C.V.(%) 3.2 6.0 3.8 T Means with the same letter are not significantly different at the 0.05 probability level. 94 Table 6: Mean germination percent, water absorption, and cooking time of eight genotypes averaged over three storage times two relative humidities and three temperatures. Entry Seed Water Cooking germination absorption time (%) (g) (min) Kalima 87 81 44 Nyirakizungu 87 76 50 Var ll 86 82 35 Kilyumukwe 86 56 62 TMO 959 84 60 60 UAC 221 82 36 67 Lyamungu 85 82 61 56 Yellow eye 79 68 56 Mean 84 65 54 LSD(0.05) l 2 1 CV (%) 3 3 6 95 ExperimenLIL From.the analyses of variance, significant differences existed for the cooking time of the 16 genotypes (Table 7). The analysis of variance further indicated that the location effects significantly influenced the cooking time in the SR and LR growing seasons. Significant differences were observed between the two seed conditions (control vs accelerated aged seed for progeny grown during the SR and LR growing seasons. The fact that the genotypes differed in their response to accelerated aging implied the existence of significant genetic variability. The differences noted between soaking and non-soaking beans prior to cooking them indicated that soaking of the seeds improved their hydration characteristics. While Burr et al.,(1968) and Molina et al., (1975) found no correlation between water uptake and cooking time, Sefa-Dedah et al., (1979) found that water absorption influenced cooking time and that it was determined by seed coat texture. The results obtained in this study for soak and non-soak cooking method agree with those of Sefa-Dedah et al., (1979) and Castellanos, et al., 1995. Several interactions were detected. In the SR growing season all treatment interactions were significant. Table 8 indicates the mean performance of the bean genotypes that 96 underwent accelerated aging for 72 hours and control treatment that were stored at 20, 25, and 30°C and 40 and 70% RH for 3, 6, and 9 months. Accelerated aging increased the cooking time of each genotype compared to the three months storage period. On the other hand, aging had less of an effect on cooking time except for (UAC 221) than the 9 months storage effects. The average cooking time increase of the eight genotype stored for 9 months compared to the aging treatment was 12 minutes. There were mixed results for the beans stored for 6 months when compared with the aging treatment (Table 8). Table 9 shows that the effect of aging in the bean seed was greater in the SR growing season than in the LR growing season. This suggests that short rains have an adverse effect on the storage of the seed. This is important to the the consumers and farmers, as it concerns the shelf-life of the bean seed during storage. 97 Table 7: Mean square analyses for cookability of sixteen entries of dry bean seed that were accelerated aged from two locations grown during the SR (1993) and LR(1994) growing seasons at the Morogoro region, Tanzania. Mean squares Source df Short rain long rain Reps (R) 2 l 6 Location (L) 1 598* 858* Genotype (G) 15 2901* 2400* G x L 15 5 3 Seed cond. 1 20460* 13184* L x C 1 32* 44* G x C 15 74* 62* L x G x C 15 3 2 Soaking(S) 1 2954* 4874* L x S 1 71* 81* G x S 15 97* 59* L x G x S 15 4* 3 C x M 1 3744* 330* L x C x S 1 33* 126* G x C x S 15 55* 16* L x G x C x S 15 3 2 Error 254 2 2 Total 383 * significant at the 0.05 probability level 98 Table 8: Table of means for the cooking time trait of dry bean genotypes under three storage times under normal conditions and under accelerated aging conditions. Entry Cooking time (minutes) Storage time in months t .Accelerated 0* 3 6 9 aging Var 11 25 28 33 43 39 Kalima 28 36 46 51 43 Nyirakizungu 30 40 51 58 47 yellow eye 35 44 57 65 50 Lyamungo 85 42 50 61 70 53 TMO 959 42 47 60 72 55 Kilyumukwe 44 52 62 71 57 UAC 221 46 55 67 79 80 Mean 37 37 55 66 54 LSD(0.5) 1 1 1 1 1 C.V.(%) 3.2 3.2 3.2 3.2 2.9 1 non aged seed # control storage time ' 99 Table 9. Table of means for cooking time of sixteen entries that were accelerated aged during the SR (1993)and LR (1994) growing seasons from two locations at the Morogoro region,Tanzania. Cooking time (minutes) Entry Accelerated aged Freshly SR t LR t harvested Var 11 25 36 33 605621 27 39 36 Kalima 28 41 37 Nyirakizungu 30 44 39 Jacob's cattle 32 46 40 Sierra 32 48 42 Yellow eye 34 50 44 Diacol calima 30 53 45 Lyamungu 85 43 55 49 TMO 959 42 57 52 NY99 44 60 54 Kilyumukwe 44 62 56 Montcalm 45 64 58 GLPX 1125 45 67 60 CIAT-3005 45 68 62 UAC 221 46 71 65 Mean 37 54 48 LSD (0.05) 1 1 1 CV (%) 3.2 2.4 3.0 t SR = short rain growing season i LR = long rain growing season 100 Implications. Results of this study indicated that sufficient variability existed in the germplasm for germination and cookability (water absorption and cooking time) of seeds. Rapid water absorption is a favored food quality trait because genotypes that imbibed the most water during soaking i.e. cooked the fastest. This finding held true regardless of the storage conditions imposed. There was deterioration in the quality of dry beans stored at high temperature and high relative humidity. Measurements of cooking time indicated that beans were increasingly difficult to cook when stored at the higher temperature and relative humidity. The results from this study indicated that accelerated aging lengthened the cooking time in dry bean and even for beans stored for 3 months (Table 8). Accelerated aging should be useful to predict how well a bean will store under unfavorable environments. However, accelerated aging lost its utility for beans stored more than 6 months. The mean cooking time of the eight entries under accelerated aging was 54 minutes compared to 55 minutes and 66 minutes for beans stored six and nine months respectively. Kalima, Var 11, 605621 and Nyirakizungu out-performed others in maintaining higher percent germination, higher water absorption and fast cooking. The possession. of these 101 attributes make them superior genotypes for population improvement of shortening cooking time, increasing water absorption and seed germination. Var 11 consistently cooked fast across both SR and LR seasons (Table 9). The same was true for 605621, Kalima and Nyirakizungu although the magnitude between SR and LR was greater in the later genotypes. There was a more pronounced negative effect on cooking time when seed grown in the short rains was subjected to accelerated aging than for seed grown during the long rains season. LIST OF REFERENCES ASTM (American Society for Testing and Materials). 1951. Recommended practice for maintaining constant relative humidity by means of aqueous solutions. In: Selected .ASTMI standards for chemical engineering‘ Students. American Society for Testing and .Materials, Philadelphia. Antunes, P.L. and V.C. Sgarbieri. 1979. Influence of time and conditions of storage on technological and nutritional properties of a dry bean (Phaseolus vulgaris L.) variety Rosinha G2. J. Food Sci. 44: 1703-1706. Bourne, M.C. 1967. Size, Density, and Hardshell in Dry Beans. Food Technol. 21:17-20. Burr, H. K., S. Kon and H. J. Morris. 1968. Cooking rates of dry beans as influenced by moisture content and temperature and time of storage. Food Technol. 22:88- 90. Castellanos, J. 2., H. Guzman-Maldonado, J. A..Acosta- Gallegos and J. D. Kelly. 1995. Effects of hardshell character on cooking time of common beans grown in the semiarid highlands of Mexico. J. Sci. Food.Agric. 69:437-443. Delouche, J. C. and C. C. Baskin. 1973..Accelerated Aging Techniques for Predicting the Relative Storability of Seed Lots. Seed Sci. and Tech. 1:427-452. Jackson, M .G. and E. varriano-Marston. 1981. Hard-to-cook phenomenon in beans: effects of accelerated storage on water‘ absorption and cooking time. J. Food. Sci. 46:799—803. Jenes, P. M. B. and D. Boutler. 1983. The cause of reduced cooking rate in (Phaseolus vulgaris L.) Following adverse storage conditions. J. Food Sci. 48:623-649. Mejia, Elvira Gonzalez de. 1979. Effects of various conditions on general aspects of bean hardening. Final Report. Unu Fellow. Instituto De Nutrition De Centro America Y Panama. Guatemala. 102 103 Molina, M. R. , de la Fuente, G., and Bressani, R. 1975. Interrelationships between storage, soaking time, cooking time, nutritive value and other characteristics of the black bean (Phaseolus vulgaris L.). J. Food Sci. 40: 587. Monsanto. 1992. Sorbic acid and potssium sorbate. Mbnsanto industrial chemicals co, St. Louis, Missouri 63166. Sefa-Dedah, S., D.W. Stanley and P.W.Voisey. 1979. Effect of storage time and conditions on the hard to cook defect in cowpeas (Vigna unguiculata). J. Food Sci. 44:790-796. Shellie-Dessert, K. and F. A. Bliss. 1993. Genetic improvement of food quality. In: Common Beans. Research for crop improvement. A. van Schoonhoven and O. VOysest (eds). C.A.B. International in association with CIAT. van Schoonhoven, A. and O. Voysest. 1993. Common Beans. Research for crop improvement. C. A. B. International in association with CIAT. APPENDIX 104 Table 1: General soil characteristics at Morogoro at 0- 12 cm. During the 1993-94 growing seasons. Characteristic Determination Value method Soil pH:H20 pH meter 4.60 Soil texture % Clay(loam) Hydrometer 20 % Silt 20 % Sand 60 Textural class: sandy loam % Total Nitrogen Kjedahl 0.10 (semi micro) % Organic carbon 0.83 Phosphorus (ppm) 0.04 % Soil moisture Volumetric method 13.09 Bulk density Density (cm3/cm3) method 1.18 Exchangeable cations Ca (me/1009'1 soil) Saturation‘ 2.20 method K " " 1.33 Mg " " 1.31 Ad. " " 1.03 ECEC " " 5.87 % Al " ” Saturation 17.54 method Drainage class moderate Permeability moderate Color reddish brown 105 one meHeE mo noHuoeuounH men on e no eoneHue> .xHe>Huooomeu .mnoHueuoH one meHeEeu No «euneHueer uouuo ooHooo .1. No A .meHenem one .moHeE .mnoHueoHHQeu .mnoHueooH anqu moon .mnoHueOOH mo Honenn enu ou Heuew b one .2 .m .m .H e HlunumH . Heuoa «0 Hz HHuunvaHuwva uouue ooHoom 3on + No «2 3-: 2-3 3.57. E. :0 63 neon... + 3.3 + No m2 3an Size .H x Am: aunt. + 3.9. + so an 3-: 3-15m H 3. GE when + 340... + .6 ma :tt 315m 6; x 2 Launch + Recon + 349.5 + 3ND“ + No 02 Sign him; #0on + swoon + Rebuu + acaou + No >2 HHLEm 2;sz LHuucmH Anoimcmomm HHImVHHIHv H x m Hum m .mumm HIH H .noHueooH # mowenom neon oeuoooxm onenom neez no .mounom .nomeom 0nHzouu «mummmH onu nH eHneNnea .noHoem ouououoz ue mnOHueooH Hobo oeueeoem uneEHuooxm HH Heooz HH noHnoo e no eueo won noHueuoooxm onenom new: one eoneHue> no mHmmHenm no Euom "N eHoea 106 Table 3: Protein Determination. Prior to estimating protein content the raw dry bean seed were processed to bean flour. The seed were ground to 40 u with a Udy-Cyclone mill. The flour was then analyzed by the micro- Kjedahl method of nitrogen determination. The nitrogen content of each sample was multiplied by 6.25 to obtain the total protein. Thus % protein = % nitrogen x 6.25. Table 4: Tannin Content Determination. Tannin content was determined according to vanillin hydrochloric method of Burns (1971) as modified by Telek (1983). These stages are: I. Preparation and Extraction of the Samples of the Dry Bean. 1- A.0.Zg sample of ground bean seed flour (40p) was weighed in a 100 ml sample bottle. 2- An acidic methanol solution was then made (V/V/V) by mixing absolute methanol (80 ml): distilled water (19.5 ml): concentrated hydrochloric acid (0.5 m1). 3- 35 ml of the acidic 80% methanol solution were added to the ground bean flour sample. 4- The treated samples were then put in a shaker bath maintained at 70°C for 30 minutes. 107 5- The extract was the decanted over a porcelain crucible lined with glass microfiber filter (GF/D whatman, 2.5cm) into a 100 ml volumetric flask. 6- The residual from the filter was decanted two additional times. All the extracts were combined together and volume made up with 80% acidic methanol solution. 7- 5 ml of the extract were pipetted into a 25 ml volumetric flask and brought to volume with a 30% sulfuric acid solution. 8- From this 25 m1 volume, 3 ml sample were pipetted into each of three 10 ml volumetric flasks. 9- 3 ml of a 0.5% vanillin solution were added to two of the 10 ml sulfuric acid solution flasks. 10- To the third 10 ml flask only sulfuric acid was added. 11- The 3 flasks were allowed to stand for 20 minutes and then absorbance of each flask read at 500 nm. 12- Two vannilin blanks were then prepared by pipetting 3 ml of 0.5% solution into a 10 ml volumetric flask and bringing up to volume with a 30% sulfuric acid solution. 108 II. Preparation of the Catechin Standard. 1- A.0.05 9 sample of catechin was weighed and dissolved in 2 ml absolute methanol in a 50 ml volumetric flask and brought up to volume with distilled water. 2- A.5 ml sample of this catechin solution was pipetted into 200 ml volumetric flask and brought to volume with a 30% sulfuric acid solution. 3- From this 200 ml solution 3 ml sample were pipetted into 10 ml volumetric flasks in duplicate. To each of the flasks 3 ml of a 0.5% vanillin solution were added. The 0.5% vanillin and catechin solutions were prepared fresh each day prior to pipetting the ground bean seed flour. III. Reading the Absorbance l- The spectrometer was set to zero with a vannilin blank by putting the blank in both sample and reference cuvette. 2- The catechin standard is read at 500 nm against a vannilin blank which is left in the reference cell. 3- The Sample blank is placed in both the reference 109 and sample cell and read. The sample cuvette is rinsed and the actual sample poured into the cuvette and read against the sample blank. IV. Determining the Catechin Equivalent A. Day Factor Day Factor = (wt. Of catechin/ O.D of catechin) x (dilution factor of sample ldilution factor of catechin) x 100 B. % catechin equivalent = (O.D. of sample/ wt. of sample) x Day Factor. 110 Table 5a: Analysis of variance for protein SR season Source of variation d.f. s.s. m.s. v.r. site 1 339.5 339.5 2036.0 sets 1 5.8 5.8 34.6 site.sets 1 1.1 1.1 6.6 site.sets.reps 8 1.2 0.1 0.9 sets.males 6 98.8 16.5 98.8 sets.females 6 240.4 40.1 240.3 sets.males.females 18 11.0 0.6 3.7 site.sets.males 6 21.2 3.5 21.2 site.sets.females 6 1.3 0.2 1.3 site.sets.males.females 18 6.0 0.3 2.0 Residual 120 20.0 0.2 Total 191 746.2 Table 5b: Estimated Components of Variance and standard errors (s.e.)for protein SR season s.e. site 3.525 5.037 sets 0.0000167 0.8679 site.sets 0.0000167 0.1127 site.sets.reps 0.0000167 0.005389 sets.males 0.5275 0.4055 sets.females 1.649 0.9667 sets.males.females 0.04657 0.03860 site.sets.males 0.2672 0.1705 site.sets.females 0.0000167 0.01841 site.sets.males.females 0.05486 0.03750 *units* 0.1667 0.02152 111 Table 6a: Analysis of variance for tannin content SR season Source of variation d.f. s.s. m.s. v.r. site 1 324.4 324.4 376.8 sets 1 94.1 94.1 109.2 site.sets l 0.4 0.4 0.4 site.sets.reps 8 9.0 1.1 1.3 sets.males 6 188.0 31.3 36.4 sets.females 6 1957.2 326.2 378.9 sets.males.females 18 54.3 3.0 3.5 site.sets.males 6 3.9 0.6 0.8 site.sets.females 6 2.9 0.5 0.6 site.sets.males.females 18 15.4 0.9 1.0 Residual 120 103.3 0.9 Total 191 2752.9 Table 6b: Estimated Components of Variance and standard errors (s.e.) for tannin content SR season s.e. site 3.376 4.791 sets 0.0000861 5.595 site.sets 0.0000861 0.03889 site.sets.reps 0.01672 0.03594 sets.males 1.189 0.7605 sets.females 13.48 7.857 sets.males.females 0.3600 0.1745 site.sets.males 0.0000861 0.04789 site.sets.females 0.0000861 0.04789 site.sets.males.females 0.0000861 0.1026 *units* 0.8610 0.1112 112 Table 7a: Analysis of variance for 90% maturity for SR season Source of variation d.f. s.s. m.s. v.r. site 1 1800.8 1800.8 2041.0 sets 1 768.0 768.0 870.5 site.sets 1 0.2 0.2 0.2 site.sets.reps 8 14.1 1.8 2.0 sets.males 6 451.3 75.2 85.3 sets.females 6 790.3 131.7 149.3 sets.males.females 18 37.7 2.1 2.4 site.sets.males 6 65.1 10.8 12.3 site.sets.females 6 71.1 11.8 13.4 site.sets.males.females 18 51.9 2.9 3.3 Residual 120 105.9 0.9 Total 191 4156.3 Table 7b: Estimated Components of Variance and standard errors (s.e.) for 90% maturity SR season s.e. site 18.76 26.83 sets 6.071 11.67 site.sets 0.0000882 0.6400 site.sets.reps 0.05521 0.05563 sets.males 2.715 1.848 sets.females 5.028 3.201 sets.males.females 0.0000882 0.2267 site.sets.males 0.6632 0.5278 site.sets.females 0.7465 0.5754 site.sets.males.females 0.6677 0.3228 *units* 0.8823 0.1139 113 Table 8a: Analysis of variance for lOO-seed weight SR season Source of variation d.f. s.s. m.s. v.r. site 1 98.2 98.2 222.9 sets 1 130.8 130.8 297.1 site.sets 1 0.01 0.01 0.02 site.sets.reps 8 1.4 0.2 0.4 sets.males 6 1267.6 211.3 479.7 sets.females 6 1218.0 203.0 460.9 sets.males.females 18 23.3 1.3 2.9 site.sets.males 6 3.4 0.6 1.3 site.sets.females 6 11.7 2.0 4.4 site.sets.males.females 18 11.6 0.6 1.5 Residual 120 52.9 0.4 Total 191 2819.1 Table 8b: Estimated Components of Variance and standard errors (s.e.) for loo-seed weight SR season s.e. site 1.023 1.475 sets 0.0000440 6.335 site.sets 0.0000440 0.06307 site.sets.reps 0.0000440 0.01424 sets.males 8.752 5.084 sets.females 8.350 4.884 sets.males.females 0.1084 0.08048 site.sets.males 0.0000440 0.03591 site.sets.females 0.1089 0.09567 site.sets.males.females 0.06854 0.07425 *units* 0.4405 0.05686 114 Table 9a: Analysis of variance for cooking time SR season Source of variation d.f. s.s. m.s. v.r. site 1 277.9 277.9 144.0 sets 1 1116.5 1116.5 578.4 site.sets 1 0.1 0.1 0.1 site.sets.reps 8 27.7 3.5 1.8 sets.males 6 370.4 61.7 32.0 sets.females 6 127.2 21.2 11.0 sets.males.females 18 22.7 1.3 0.7 site.sets.males 6 25.9 4.3 2.2 site.sets.females 6 2.7 0.4 0.2 site.sets.males.females 18 15.2 0.8 0.4 Residual 120 231.6 1.9 Total 191 2218.0 Table 9b: Estimated Components of Variance and standard errors (s.e.) for cooking time season s.e. site 2.894 4.196 sets 10.82 16.55 site.sets 0.0001930 0.2195 site.sets.reps 0.09583 0.1094 sets.males 2.375 1.517 sets.females 0.8472 0.5493 sets.males.females 0.06944 0.1689 site.sets.males 0.2894 0.2654 site.sets.females 0.0001930 0.1074 site.sets.males.females 0.0001930 0.2301 *units* 1.930 0.2492 115 Table 10a: Analysis of variance for water absorption SR season Source of variation d.f. s.s. m.s. v.r. site 1 382.5 382.5 277.6 sets 1 2662.6 2662.6 1932.1 site.sets 1 0.6 0.6 0.5 site.sets.reps 8 11.3 1.4 1.0 sets.males 6 1892.0 315.3 228.8 sets.females 6 8148.9 1358.2 985.5 sets.males.females 18 405.8 22.5 16.4 site.sets.males 6 1.8 0.3 0.2 site.sets.females 6 5.2 0.8 0.6 site.sets.males.females 18 14.6 0.8 0.6 Residual 120 165.4 1.4 Total 191 13690.9 Table 10b: Estimated Components of Variance and standard errors (s.e.) for water absorption SR season s.e. site 3.978 5.651 sets 10.54 40.13 site.sets 0.004851 0.05852 site.sets.reps 0.002083 0.04549 sets.males 12.22 7.619 sets.females 55.65 32.69 sets.males.females 3.622 1.286 site.sets.males 0.0001378 0.07665 site.sets.females 0.005015 0.07911 site.sets.males.females 0.0001378 0.1643 *units* 1.378 0.1779 Table 11a: Analysis of season 116 variance for protein content LR Source of variation site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females Residual Total 0. H O‘O‘CDO‘O‘CDHHH 18 120 191 .f. UH ON QDUQUUNUDOQU N (D OSLOH UdmdmmmUlI-‘Hmb UlN OOHOHOOHOQU .C.. ooquwmoiHr-aoab 328.9 H 00 GM OHOHODH O O... WUICDOSQQDHOS Table 11b: Estimated Components of Variance and standard errors (s.e.) for protein content LR season site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females *units* 2.847 0.0000831 0.0000831 0.01899 0.8087 2.027 0.08858 0.0000831 0.04182 0.0000831 0.8311 Seee 4.050 1.086 0.05362 0.03609 0.5006 1.216 0.08869 0.04623 0.06817 0.09905 0.1073 117 Table 12a: Analysis of variance for tannin content LR season Source of variation d.f. s.s. m.s. v.r. site 1 510.9 510.9 231.5 sets 1 76.8 76.8 34.8 site.sets 1 1.1 1.1 0.5 site.sets.reps 8 21.3 2.7 1.2 sets.males 6 89.6 14.9 6.8 sets.females 6 981.3 163.5 74.1 sets.males.females 18 152.1 8.4 3.8 site.sets.males 6 102.2 17.0 7.7 site.sets.females 6 43.2 7.2 3.3 site.sets.males.females 18 163.7 9.1 4.1 Residual 120 264.8 2.2 Total 191 2406.9 Table 12b: Estimated Components of Variance and standard errors (s.e.) for tannin content LR-season s.e. site 5.311 7.773 sets 0.0002207 2.775 site.sets 0.0002207 0.5700 site.sets.reps 0.02845 0.08508 sets.males 0.0002207 0.6067 sets.females 6.541 4.005 sets.males.females 0.0002207 0.7146 site.sets.males 0.6621 0.8578 site.sets.females 0.0002207 0.5053 site.sets.males.females 2.296 1.015 *units* 2.207 0.2849 118 Table 13a: Analysis of variance for 90% maturity LR season Source of variation d.f. s.s. m.s. v.r. site 1 1271.0 1271.0 200.4 sets 1 713.0 713.0 112.4 site.sets 1 7.5 7.5 1.2 site.sets.reps 8 53.4 6.7 1.1 sets.males 6 319.1 53.2 8.4 sets.females 6 865.9 144.3 22.8 sets.males.females 18 212.6 11.8 1.9 site.sets.males 6 129.6 21.6 3.4 site.sets.females 6 93.9 15.6 2.5 site.sets.males.females 18 110.6 6.1 1.0 Residual 120 761.3 6.3 Total 191 4538.0 Table 13b: Estimated Components of Variance and standard errors (s.e.) for 90% maturity LR season s.e. site 13.16 19.09 sets 5.739 10.91 site.sets 0.0006344 0.9913 site.sets.reps 0.02083 0.2148 sets.males 1.080 1.400 sets.females 5.125 3.502 sets.males.females 0.9444 0.7547 site.sets.males 1.288 1.064 site.sets.females 0.7917 0.7825 site.sets.males.females 0.0006344 0.7561 *units* 6.344 0.8190 119 Table 14a: Analysis of variance for loo-seed weight LR season Source of variation d.f. s.s. m.s. v.r. site 1 203.0 203.0 138.0 sets 1 139.7 139.7 95.0 site.sets 1 12.0 12.0 8.2 site.sets.reps 8 8.1 1.0 0.7 sets.males 6 1472.6 245.4 166.9 sets.females 6 1395.3 232.5 158.1 sets.males.females 18 42.3 2.3 1.6 site.sets.males 6 22.3 3.7 2.5 site.sets.females 6 11.3 1.9 1.3 site.sets.males.females 18 30.7 1.7 1.2 Residual 120 176.5 1.5 Total 191 3513.8 Table 14b: Estimated Components of Variance and standard errors (s.e.) for loo-seed weight LR season s.e. site 1.989 3.002 sets 0.0001471 7.420 site.sets 0.1783 0.3711 site.sets.reps 0.0001471 0.04755 sets.males 10.05 5.905 sets.females 9.584 5.594 sets.males.females 0.1072 0.1613 site.sets.males 0.1675 0.1850 site.sets.females 0.01528 0.1025 site.sets.males.females 0.07843 0.1999 *units* 1.471 0.1899 Table 15a: Analysis of season 120 variance for cooking time LR Source of variation site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females Residual Total 0.. H memmml-‘l-‘H 18 120 191 .f. s.s. 212.5 1281.0 5.3 7.9 682.0 1487.0 86.2 6.1 10.1 13.7 190.8 3983.0 m.s. v.r. 212.5 133.7 1281.0 806.1 5.3 3.4 1.0 0.6 113.7 71.5 247.8 155.9 4.8 3.0 1.0 0.6 1.7 1.1 0.8 0.5 1.6 Table 15b: Estimated Components of Variance and standard errors (s.e.) for cooking time LR season site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females *units* 2.158 9.596 0.08346 0.0001590 4.527 10.09 0.6713 0.02083 0.07639 0.0001590 1.590 3.153 18.97 0.2038 0.05138 2.756 5.983 0.3244 0.09894 0.1284 0.1895 0.2052 121 Table 16a: Analysis of variance for water absorption LR season O. H) Source of variation site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females 18 Residual 120 Total 191 H mmmmmmHl-‘H s.s. 2.755E+02 2.776E+03 O.521E+00 1.692E+01 1.931E+03 7.880E+03 4.238E+02 6.458E+00 1.875E+00 1.229E+01 1.011E+02 1.342E+04 m.s. 2.755E+02 2.776E+03 0.521E+00 2.115E+00 3.218E+02 1.313E+03 2.354E+01 1.076E+00 0.313E+00 0.683E+00 0.842E+00 Vere 327.08 3294.93 0.62 2.51 382.05 1559.02 27.95 1.28 0.37 0.81 Table 16b: Estimated Components of Variance and standard errors (s.e.) for water absorption LR season site sets site.sets site.sets.reps sets.males sets.females sets.males.females site.sets.males site.sets.females site.sets.males.females *units* 2.865 12.13 0.0000842 0.07951 12.41 53.75 3.810 0.03279 0.0000842 0.0000842 0.8424 s.e. 4.088 41.72 0.07958 0.06643 7.753 31.61 1.318 0.06391 0.04685 0.1004 0.1087 122 Table 17: Analysis of variance table for percent seed eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity germination of Sum of Mean F Source df squares square value Replication 2 9.6 4.8 0.7 Genotype (G) 7 3142.1 448.9 67.0 Error 14 93.8 6.7 Storage time(S) 2 22318.6 11159.3 1212.2 G x S 14 880.4 62.9 6.8 Error 32 294.6 9.2 Temperature (T) 2 2751.4 1375.7 192.4 G x T 14 94.7 6.83 0.9 S x T 4 61.3 15.3 2.1 G x S x T 28 259.9 9.3 1.3 R. Humidity (H) 1 665.2 665.1 93.0 G x H 7 32.5 4.6 0.6 S x H 2 17.9 9.0 1.3 G x S x H 14 76.5 5.5 0.8 T x H 2 1.6 0.8 0.1 G x T x H 14 73.5 5.3 0.7 S x T x H 4 3.7 0.9 0.1 G x S x T x H 28 123.2 4.4 0.6 Error 240 1716.0 7.2 Total 431 32616.4 Coefficient of Variation: 3.18% 123 Table 18: Analysis of variance table for water absorption of eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity Sum of Mean F Source df squares square value Replication 2 42.8 21.4 0.8 Genotype (G) 7 87978.7 12568.4 473.6 Error 14 371.5 26.5 Storage time (S) 2 18842.0 9421.0 740.5 G x S 14 3781.1 270.1 21.2 Error 32 407.1 12.7 Temperature (T) 2 2034.1 1017.1 66.2 G x T 14 327.2 23.4 1.5 S x T 4 3.2 0.8 0.1 G x S x T 28 658.6 23.5 1.5 R. Humidity (H) 1 736.3 736.3 47.9 G x H 7 72.7 10.4 0.7 S x H 2 32.2 16.1 1.0 G x S x H 14 220.4 15.7 1.0 T x H 2 0.04 0.02 0.001 G x T x H 14 245.2 17.5 1.1 S x T x G 4 39.9 10.0 0.6 G x S x T x H 28 379.5 13.6 0.9 Error 240 3689.2 15.4 Total 431 119861.9 Coefficient of Variation: 6.03% 124 Table 19: Analysis of variance table for cooking time of eight cultivars of dry bean seed stored for nine months at varying temperature and relative humidity index Sum of Mean F Source df squares square value Replication 2 11.6 5.8 0.7 Genotype (G) 7 42544.5 6077.8 712.7 Error 14 119.4 8.5 Storage time (S) 2 28055.5 4027.8 3028.5 G x S 14 1062.9 75.9 16.4 Error 32 148.2 4.6 Temperature (T) 2 2198.8 1099.4 259.3 G x T 14 82.6 5.9 1.4 S x T 4 23.7 5.9 1.4 G x S x T 28 125.7 4.5 1.1 R. Humidity (H) 1 705.3 705.3 66.4 G x H 7 26.6 3.8 0.9 S x H 2 19.3 9.6 2.3 G x S x H 14 32.7 2.3 0.6 T x H 2 27.5 13.8 3.2 G x T x H 14 18.6 1.3 0.3 S x T x H 4 8.8 2.2 0.5 G x S x T x H 28 56.2 2.0 0.5 Error 240 1017.4 4.2 Total 431 “762853 Coefficient of Variation: 3.81% 125 Table 20: Analysis of variance for cooking time of 16 entries of accelerated aged SR dry bean seed grown in SR growing season Sum of Mean F Source df squares square value Replication 2 2.69 1.34 0.80 Location (L) 1 597.50 597.50 355.72 Genotype (G) 15 43522.21 2901.48 1727.37 G x L 15 67.79 4.52 2.69 Seed cond. 1 20460.44 20460.44 12180.95 L x C 1 32.09 32.09 19.10 G x C 15 1114.35 74.29 44.23 L x G x C 15 38.71 2.58 1.54 Soaking(S) 1 2953.71 2953.71 1758.47 L x S 1 70.90 70.90 42.21 G x S 15 1454.41 96.961 57.72 L x G x S 15 67.23 4.48 2.67 C x S 1 3743.75 3743.75 2228.81 L x C x S 1 33.25 33.25 19.81 G x C x S 15 817.87 54.51 32.46 L x G x C x 15 50.37 3.36 2.00 Error 254 426.65 1.68 Total 383 75454.00 Coefficient of Variation: 2.41% 126 Table 21: Analysis of variance for cooking time of 16 entries of accelerated aged dry bean seed grown in LR growing season Sum of Mean F Source df squares square value Replication 2 12.56 6.28 3.13 Location (L) 1 858.01 858.01 428.36 Genotype (G) 15 35993.66 2399.52 1197.97 G x L 15 38.49 2.57 1.28 Seed cond. 1 13183.59 13183.59 6581.81 L x C 1 44.01 44.01 21.97 G x C 15 928.41 61.89 30.90 L x G x C 15 27.49 1.83 0.91 Soaking(S) 1 4873.50 4873.50 2433.06 L x S 1 80.67 80.67 40.27 G x S 15 885.67 59.04 29.48 L x G x S 15 50.00 3.33 1.66 C x S 1 330.04 330.04 164.77 L x C x S 1 126.04 126.04 62.93 G x C x S 15 242.13 16.14 8.06 L x G x C x 15 26.63 1.76 0.87 Error 254 508.77 2.00 Total 383 58210.00 Coefficient of Variation: 2.94% 127 Table 22: Daily Weather Data for 1993/94 Growing Season at Sokoine University of Agriculture (SUA). Day of Temperature %: Solar Rainfall Year Month Day Maxim. Minimn (radiation (mm) (MJ.m4) 94060 March 01 31.0 22.0 16.6 1.2 94061 March 02 31.5 20.5 18.4 0.0 94062 March 03 30.5 21.5 15.8 1.2 94063 March 04 31.2 19.5 21.8 18.0 94064 March 05 30.5 21.2 17.1 4.2 94065 March. 06 29.7 20.8 17.5 6.2 94066 March 07 31.0 22.0 18.5 0.0 94067 March 08 31.7 21.6 20.2 0.0 94068 March 09 31.5 21.3 22.2 0.0 94069 March 10 31.5 19.8 17.3 0.0 94070 March 11 32.0 20.0 22.7 0.0 94071 March 12 32.4 20.9 17.1 0.0 94072 March 13 32.0 20.0 17.6 0.0 94073 March 14 30.7 19.5 13.4 0.0 94574 March 15 31.1 20.6 17.4 7.4 94065 March 16 32.5 20.7 19.2 0.0 94076 March 17 28.5 21.0 7.2 1.5 94077 March 18 30.7 19.5 7.2 7.2 94078 March 19 29.6 20.0 18.1 6.4 94079 March 20 30.0 19.0 15.0 0.0 94080 March 21 30.5 20.5 14.6 0.6‘ 94081 March 22 31.0 19.5 15.3 1.7 94082 March 23 31.6 20.2 14.6 2.2 94083 March 24 30.4 20.0 14.3 0.0 94084 March 25 31.5 19.5 16.3 2.1 94085 March 26 30.3 21.9 14.3 12.0 94086 March 27 30.5 18.4 16.4 6.4 94087 March 28 31.6 20.0 21.3 0.0 94088 March 29 32.0 21.5 20.3 0.0 94089 March 30 31.7 21.5 16.2 0.3 94090 March 31 30.7 22.5 15.6 0.0 Mean values 29.0 21.1 17.3 2.2 128 Table 22 (cont.) Day of Temperature %: Solar Rainfall Year Month Day Maxim. Minimn radiation (mm) (MJ m’z) 94091 April 01 30.5 21.4 19.5 0.0 94092 April 02 32.0 17.1 16.3 0.0 94093 April 03 32.0 18.8 22.0 0.0 94094 April 04 30.0 19.9 12.5 0.0 94095 April 05 28.0 20.5 11.8 1.3 94096 April 06 28.0 20.3 10.6 31.9 94097 April 07 26.4 20.3 9.7 11.1 94098 April 08 29.5 18.1 13.2 6.9 94199 April 09 29.0 20.0 14.7 24.4 94100 April 10 27.5 20.5 10.9 24.2 94101 April 11 27.1 20.0 10.6 4.1 94102 April 12 29.4 19.8 15.0 15.0 94103 April 13 29.8 18.0 17.9 1.0 94104 April 14 30.1 19.2 15.7 0.0 94105 April 15 30.5 20.8 20.6 7.1 94106 April 16 30.5 20.4 12.7 2.6 94107 April 17 30.5 19.5 20.1 0.0 94108 April 18 33.3 20.0 17.8 0.0 94119 April 19 31.5 18.3 17.5 5.9 94110 April 20 28.5 20.2 12.5 7.2 94111 April 21 26.5 20.2 7.8 16.0 94112 April 22 31.0 20.6 19.1 14.0 94113 April 23 30.0 19.4 15.6 1.2 94114 April 24 27.0 20.3 9.0 26.0 94115 April 25 29.0 19.6 18.1 26.5 94116 April 26 29.0 18.8 11.7 35.0 94117 April 27 27.6 19.5 14.4 5.9 94118 April 28 28.4 17.0 14.5 0.0 94119 April 29 26.5 17.5 11.0 13.4 94120 April 30 23.5 19.6 3.7 33.0 Mean values 29.0 19.5 14.2 10.6 129 Table 22 (cont.) 94121 May 01 25.2 19.2 9.2 0.5 94122 May 02 24.5 19.8 6.8 8.0 94123 May 03 27.5 17.5 13.7 17.2 94124 May 04 28.5 19.5 14.4 16.1 94125 May 05 30.3 20.5 13.3 0.6 94126 May 06 30.7 19.8 18.1 0.6 94127 May 07 29.5 19.4 19.7 0.8 94128 May 08 27.0 19.5 10.6 0.0 94129 May’ 09 26.5 19.6 11.6 0.0 94130 May 10 27.3 18.8 11.1 0.0 94131 May 11 25.5 20.0 7.5 15.0 94132 May 12 25.5 19.0 6.7 8.4 94133 May 13 29.2 19.8 11.8 9.6 94134 May' 14 27.5 19.4 11.0 0.0 94135 May' 15 28.4 19.8 14.8 18.5 94136 May 16 29.5 19.5 14.1 0.0 94137 May' 17 28.9 18.9 14.5 0.8 94138 May’ 18 29.2 19.0 16.0 1.1 94139 May 19 25.6 18.8 9.8 3.7 94140 May 20 28.6 19.0 16.2 6.9 94141 May 21 29.5 19.1 16.2 1.9 94142 May 22 27.8 18.7 14.3 0.5 94143 May 23 29.5 17.6 19.5 0.0 94144 May’ 24 28.6 17.1 18.8 0.0 94145 May 25 29.0 17.5 16.5 0.5 94146 May’ 26 26.0 19.0 11.2 0.0 94147 May 27 26.5 16.0 11.4 0.0 94148 May 28 26.0 15.1 11.1 0.0 94149 May 29 26.0 16.2 11.0 0.0 94150 May 30 25.0 16.0 6.4 1.9 94151 May 31 26.8 17.5 11.1 0.0 Mean values 27.6 18.6 12.9 5.9 130 Table 22 (cont.) 94152 June 01 26.8 13.2 18.6 0.0 94153 June 02 28.0 12.6 17.0 0.0 94154 June 03 28.5 14.5 19.0 0.0 94155 June 04 28.6 13.9 19.4 0.0 94156 June 05 29.0 15.7 17.1 0.0 94157 June 06 29.5 14.8 21.2 0.0 94158 June 07 29.3 14.1 20.7 0.0 94159 June 08 28.5 15.4 15.5 0.0 94460 June 09 27.5 15.0 14.3 0.0 94161 June 10 27.2 14.0 15.9 0.0 94162 June 11 27.5 12.4 19.5 0.0 94163 June 12 26.5 11.2 18.9 1.2 94164 June 13 28.3 11.0 17.1 0.0 94165 June 14 27.7 13.9 16.2 0.0 94166 June 15 26.2 16.3 11.5 0.0 94167 June 16 27.0 14.4 15.2 0.0 94168 June 17 28.5 13.0 18.7 0.0 94169 June 18 27.0 13.2 15.4 0.0 94170 June 19 25.0 13.8 12.6 0.0 94171 June 20 26.6 14.1 15.0 0.0 94172 June 21 25.5 13.7 13.2 0.2 94173 June 22 23.0 16.1 5.8 6.6 94174 June 23 28.5 16.2 17.1 0.0 94175 June 24 28.8 16.3 17.5 0.0 94176 June 25 28.5 14.2 16.9 0.0 94177 June 26 28.6 13.2 14.0 0.2 94178 June 27 29.0 17.0 17.5 0.0 94179 June 28 27.8 17.0 14.3 0.0 94180 June 29 27.0 16.6 12.9 0.0 94181 June 30 28.5 12.5 27.2 0.0 Mean values 27.6 14.3 16.4 0.3 131 Table 22 (cont.) 94182 July 01 29.0 13.9 17.8 4.7 94183 July 02 24.4 18.2 6.1 0.0 94184 July 03 27.4 17.1 12.9 0.2 94185 July’ 04 27.5 16.1 13.8 13.6 94186 July 05 28.0 18.0 16.5 1.1 94187 July 06 27.0 16.5 12.9 17.2 94188 July’ 07 25.5 17.4 10.6 0.0 94189 July 08 27.4 14.5 17.1 0.0 94190 July 09 28.0 14.5 17.4 0.0 94191 July 10 29.3 13.5 18.7 0.0 94192 July 11 29.2 15.1 16.9 0.0 94493 July 12 28.1 15.0 19.0 0.0 94194 July 13 25.0 16.1 10.8 0.0 94195 July' 14 27.0 11.4 19.1 0.0 94196 July’ 15 26.5 10.2 19.0 0.0 94197 July 16 28.4 11.1 21.0 0.0 94198 July' 17 27.0 10.8 16.2 0.0 94199 July’ 18 27.5 14.0 16.2 0.0 94200 July 19 27.6 13.7 18.3 0.0 94201 July 20 28.0 16.1 12.7 0.0 94202 July 21 27.0 17.5 15.9 0.0 94203 July 22 27.9 16.0 13.4 0.0 94204 July’ 23 29.0 17.4 10.3 0.0 94205 July 24 24.9 19.2 16.1 0.0 94206 July' 25 28.7 14.8 16.1 0.0 94207 July 26 30.0 15.5 18.6 0.0 94208 July 27 30.0 16.2 19.3 0.0 94209 July 28 27.6 15.0 18.7 0.0 94210 July' 29 27.0 16.0 18.2 0.0 94211 July 30 27.7 14.6 16.2 0.0 94212 July 31 26.2 16.8 7.0 0.0 Mean values 27.5 15.2 16.6 1.2 132 Table 22 (cont.) 94213 August 01 27.5 16.0 18.4 0.0 94214 August 02 26.5 15.1 11.0 8.4 94215 August 03 24.6 16.1 8.6 0.5 94216 August 04 25.9 6.6 12.5 0.0 94217 August 05 28.0 13.1 19.3 0.0 94218 August 06 28.0 14.6 22.9 0.0 94219 August 07 27.0 15.4 14.2 0.0 94220 August 08 27.5 15.5 18.1 0.0 94221 August 09 29.0 17.0 17.9 0.0 94222 August 10 30.3 14.1 18.1 0.0 94223 August 11 28.2 5.0 17.7 0.0 94224 August 12 28.5 17.4 14.0 0.0 94225 August 13 30.0 16.0 19.3 0.0 94226 August 14 29.0 14.7 19.9 0.0 94227 August 15 28.3 15.5 13.4 0.0 94228 August 16 28.2 16.8‘ 16.8 0.0 94229 August 17 30.0 16.0 17.1 0.0 94230 August 18 29.0 14.8 17.8 0.0 94231 August 19 30.5 17.6 16.9 0.5 94232 August 20 24.5 17.2 6.3 0.5 94233 August 21 27.5 15.6 16.2 0.0 94234 August 22 28.6 14.1 19.9 0.0 94435 August 23 28.8 15.2 19.5 4.7 94236 August 24 27.5 16.8 15.7 0.0 94237 August 25 28.6 15.1 16.2 2.7 94238 August 26 28.5 17.4 12.4 1.0 94239 August 27 28.2 16.6 16.0 0.4 94240 August 28 28.6 15.8 16.2 0.0 94241 August 29 29.0 16.1 18.1 0.0 94242 August 30 29.5 15.2 18.5 0.0 Mean values 28.1 15.8 16.0 0.6 "ITlll'fll'jlllMW