Crop scientifical studies on the yield-forecast of lowland rice. : Preliminary report : III. Curves of tillering relating to earliness of lowland rice varieties and the forecast of number of panicles.

摘要

By examining the curves of tillering in rice obtained from the data of experiments in nineteen years on climate-adaptability of rice in Konosu Agricultural Experiment Farm, differences among the curves of early, intermediate and late varieties and the relations between panicle number and curves of tillering were studied. And the results were summarized as follows. (1) The authors discovered the fact that there had been no differences among early, intermediate and late maturing varieties so far as the date of the highest point of the curve is concerned. When the plants were cultivated under similar conditions and transplanted on the same day, the highest point of the tillering curve of each of all the varieties fell on the same date, irrespective of varieties and cultivating methods. (2) In the case of normal transplanting condition, it was pointed out that the beginning of young panicle formation always occurred before that highest point of the curve in early varieties, almost at the same time in intermediate varieties and always after that in late varieties. (3) The period from the date of beginning of tiller-vanishing to that of heading was short in early varieties and long in late ones. As a result of it, it was well understood that, according to the average in thirteen years, an early variety decreased 12.6 %, an intermadiate one 41.1 % and a late one 70.8 % of all vanished tillers by their heading time. (4) The percentages of panicle numbers to maximum tiller numbers, as an average of thirteen years, were 79∼72 % in early varieties, 71∼64 % in intermediate ones and 66∼62 % in late ones. As a whole, it seemed that the earlier a variety matured, the higher the percentage of the successful panicles became. (5) Regarding the finishing period of successful tillering in thirteen years, there was a nine-day interval, at least, between earliest and latest year. So that, it was proved that it had been very difficult to forecast panicle numbers by means of the tiller numbers in that period. (6) According to the result of investigation of the correlation between panicle numbers and tiller numbers of every point of tillering curves it was proved that the methods of forecasting panicle number by means of tiller numbers had been unavailable until the period of the maximum tillering or the young panicle formation. Moreover, as correlation coefficients between panicle numbers and tillering numbers in these periods ranged at about 0.88∼0.62, forecasting the number of panicle may not be done satisfactorily even in these periods.