A proposed method for estimating loss in yield of wheat caused by take-all (Ophiobolus graminis SACC.)

During the seven year period (1943∼49), take-all disease of wheat and barley was quite severe, presumably predisposed by insufficient manuring during and after the war. The outbreak in 1948 was most severe and widespread. This study was undertaken to find out a method available for estimating the loss in yield, 20 to 30 days before harvest, of wheat infected by take-all. A naturally infected wheat field of 0.7 acre in area was divided into 20 plots, a row of 180cm length was chosen at random in each plot, and wheat plants on the row were sampled. The average culm lengths, the total numbers of culms, the total grain yields, and the weight of thousand grains of these 20 samples were recorded (Table 1). As shown in Fig. 2, it was found that there was a linear relation between the average culm length and the grain yield of each sample. The degrees of disease severity, as expressed by the decrease in culm length and grain yield, could be classified into three distinct groups, viz., none or light, moderate, and heavy. The indices of average grain yields of these groups were 100, 70, and 45, and those of the corresponding average culm lengths were 100, 90, and 60 respectively. This may be due to the fact that optimum infection usually occurs twice in a growth period of wheat, viz., in autumn and in early spring; the earlier the infection occurs, the heavier is the damage. No clear difference in grain yield was found betwee“none” and“light”groups. The possibility of estimating the loss in yield for the whole area of affected field by applying the regression line of grain yield. on culm length was discussed.