Abstract
Field studies were conducted with the split plot design of 20 treatment with a combination of 4 levels of 3, 5, 7 and 9 plants per hill and 5 levels of 60, 80, 100, 120 and 140 hills per 3.3$m^2$ on non-, low- and high-salty areas. Rice variety, Kusabue was grown under jthe standard fertilization and cultivating. Investigation was made on the productive structure of plant population, leaf-area index, light intensity curve by stratum of crop population at the panicle differentiation stage. The competition density effect on the photosynthetic capacity was low as the salt concentration became higher. This seemed to suggest the possiblity of an increased yielding capacity by closer planting in the salty areas. The effect of an increased number of hills per unit area was greater than that of an increased unmber of plants per hill due to the total leaf area and space distribution of the actie assimilation parts of rice plants. The number of ppanicle per unit area in the salty areas were increased when the number of hill per 3.3$m^2$ increased over an increased number of plants per hill, and the panicle weight was reduced by close planting in the non-lalty area, while it was not reduced so much in the salty areas. The number of grains per panicle was significantly decreased by close planting in the salty areas as in the non-salty area, and ratio of matured grain was not decreased even by close planting in the salty areas, while it was significantly decreased by close planting in the non-salty area. An increase in the rice yield was possible by close planting and greatly related to leaf area index in the salty areas but not in the non-salty area. Increasing the number of hills per unit area showed greater effect on the increase of the rice yield than an increased number of plants per hill in the salty areas. Relationships between the growth characteristics and the rice population affected by plant spacing mode for maximum production were discussed.