This experiment was started at the same time of the experiment 3 with the same purpose and continued to fifth generations. In 1939, 34 samples of 7 rape varieties were recollected from the central and five local rape breeding centers where they settled during the past 10 years propagated by self pollinated seeds under paraffin paper bag or victoria lawn cage. Two to six individuals of each local strain representing earliest or latest flowering time were treated year after year. Throughout the progeny test such agronomic characteristics as date of flowering, plant height, number of first order branches, length of ear, size of seedling before transplanting, seed yield per plant, leaf color, waxiness and degree of anthocyan pigmentation were compared among local strains or intra-local lines. The same variation as found in wheat experiment was recognized in this case, that is, some varieties showed small variation among local strains or intra-local lines, however, others deviated greatly. Usually, early flowering type developed in the materials collected from southern regions, and late flowering line separated in the northern stations. From this experiment the author concluded that delivered rape varieties have been continuing to differentiate to new ecotypes reconstructing their genetic composition to adapt their new habitats. In conclusion, the author discussed the development of genecological studies in Japan and abroad or studies of varietal differentiation of main crop plant. Finally, he emphasized that in case of seed raising and propagation of crop plant variety, special attention should be given to the fact that the variety was not static but dynamic, usually has been reconstructing its genetic constitution and differentiating into new ecotype adapting to its new habitats.