A close, negative correlation was found between the mean prefectural yield of paddy rice and the mean daily air temperature during the Aug. -Sept. period which roughly corresponds to the "substantial yield production period". On the other hand, there was a positive, linear correlation observed between the yield and the mean daily amount of solar radiation during the same period among northern prefectures, although there was no correlation found between the two when southern prefectures were included. However, this was interpreted not as a proof to show the absence of the effect of solar radiation, but as a result of disturbance of its effect by the adverse effect of air temperature in warmar regions. Assuming, therefore, that the solar radiation effect to be always proportional, a parabolic function, f(t), which has a maximum at t=21.5°C, was derived to fit the temperature-yield relationship. Further, by combinating the solar radiation effect, s, with the former, a function, sf(t), has been obtained to express the "climatic productivity" of each prefecture. Quite a high correlation coefficient at a higher level than 0.1%, was shown to exist between the climatic productivity index thus calculated and the rice yield among all the prefectures. From these, it has been concluded that both the mean daily air temperature and the mean daily solar radiation during the Aug.-Sept. period play an important role in determining the local differences in the productivity of paddy rice in Japan. Some probable causes for this were discussed in relation to photosynthesis and dry matter production.