The rice stem maggot, Chlorops oryzae, is one of the serious pests of rice in Japan. It has two generations annually in northern part but completes three generations in southern part of the country. In two-generation area the rice plant is infested only once from the middle of growing stage to heading, and in three-generation area twice in the seedling and heading stages. The symptoms of attack are the injured leaves and ears.<br>This investigation was designed, first of all, to know the differences in some biological habits of the two regional stocks of the rice stem maggot from two- and three-generation areas, rearing both stocks in comparison under the natural conditions at Ômagari which is in two-generation area. The results obtained in 1958 are summarized as follows:<br>1. Fly emergence from hibernating grasses peaked about 25 days earlier in the stock from three-generation area than that from two.<br>2. In the case of infesting normally cultured rice plants, larvae of the stock from two-generation area could not pupate without feeding on young ears, and pupation occurred after the heading period of the host plant which fell on late August. Therefore, the earlier the larvae fed into the plant, the longer their developmental period became. On the other hand, the phenomenon mentioned above was not observed in the larvae of three-generation area.<br>3. Within 3 months from June to August, stocks from two- and three-generation areas yielded one and two generations respectively on the same host plant. Furthermore, prior to entering into hibernation, an intermediate generation occurred in both stocks were artificially forced to feed on the rice plants cultured in autumn.<br>4. Injury of leaves caused by larval feeding was classified into four types according to the forms of feeding scars. Some marked differences were observed in the types of injury between two stocks.<br>5. In crosses between two regional stocks mating was frequently obtained regardless of their combinations, and produced fertile eggs. In F₁ generation, the larval developmental period resembled to that of the stock from two-generation area, but the symptoms of attack were rather different from the cross combinations.<br>6. Except the intermediate generation, the stock transferred from three-generation area held its original rythm of seasonal cycle noticed in the area of its origin, and seemingly has not been influenced enough by the environmental conditions of the new locality where it was reared. The differences between these two stocks especially in number of generations per year, could not be explained by climatic conditions such as total effective temperature. This may be understood more fully by the supposition of existing biological races and their inherent characteristics which will be maintained through some successive generations of years after transferring to new locality.