Five tall chickpea genotypes, four of Russian and one of Indian origin, were mated to nine dwarf genotypes in line × tester design. None of the tall genotypes used as testers showed consistently good combining ability for all the characters studied. However, P 1387 among the Uses was a good combiner for all the characters. Majority of the best specific cross combinations for different characters emanated from crosses involving parents with high × low or low × high gca effects. Nonadditive effects were predominant for most characters, although appreciable additive effects were noticed for pod number, seed weight, and biological yield. It is suggested that for chickpea improvement, a breeding method that can combine both fixable and nonfixable types of gene action would be more suitable.
Gene action, chickpea, combining ability
