Physiological attributes determining yield, both under drought and under irrigated conditions, of some advanced chickpea lines of recent origin were investigated over two seasons using a physiological model. Total shoot biomass, seed yield, and vegetative (Dv) and reproductive (Dr) durations were measured in field trials in peninsular India and the crop growth rates (C) and the rate of partitioning to seed (p) were estimated. The contribution of model parameters to variations in seed yield were determined by path analysis, and the relationships of the yield determinants with seed yield were obtained by regression techniques. The model was found to be suitable for chickpea, and when the parameters were fitted the model explained 98% of the variation. Irrigation enhanced Dr and C. While C was the major single yield determinant, the combination of C and p in non-irrigated environments explained most of the yield variation. Dv and Dr exhibited a negative relationship while C and p exhibited a positive relationship under drought stress and a negative relationship in the irrigated environment. There were indications of the existence of an optimum Dv for maximum C among the genotypes, suggesting the need to select for optimum duration genotypes. As high values for p and C in severe drought stress and Dr and C in the irrigated environments are advantageous for high yield, it is suggested that separate breeding strategies are needed for different soil water environments.