Abstract
Using the data of determining test of recommended variety of rice, the author tried to classify the plain part of baisin area in Yamagata prefecture into several groups from the yield of variety cline. Result of this investigation is following 1. The coefficient of rank correlation, the repeatability and the interaction of VP were applied to the yields data, however 10 places could not be classified clearly by those statistical methods. 2. Generally, this is due to the fact that the yield of variety (line) is very variable, by place and year, and the relative yield of variety is remarkably instability. 3. On the breeding area, it will be necessary to consider about both the point of yield and the ecological point.
