In order to investigate the variation of the quantitative characters and their correlations of 20varieties of upland rice, the rice was planted by the four replication randomized-block-design method and by the one-plant planting method in the experime...
In order to investigate the variation of the quantitative characters and their correlations of 20varieties of upland rice, the rice was planted by the four replication randomized-block-design method and by the one-plant planting method in the experimental farm of Jeonbug national university on the 20th of June, 1975, and was harvested on the 22th of Oct., 1975. The author investigated the correlations of the qualitative characters shown. By analyzing the relationship of these characters to the yield, it has become easy to discover the high yield varieties.
The results can be summarized as follows :
1. The factors that determined the yield of upland rice were not only the number of stems, the number of panicles, and the health of the rice plant but also the length of panicles and grain weight.
2. The longer the panicle was, the greater the number of grains.
3. There was a tendency of negative correlation between the number of stems or the number of panicles and grain weight. The more stems the variety had, the lighter its grain was.
4. It was found that there is little correlation between the number of grains and the yield in upland rice.
5. Katenwase, Kokukou, Nakabara, Housenjiyobouruchi, Seikokutaiou, ,and Rikutoukanto # 72 were found to be high-yield varieties.