Prediction of eyespot severity on winter wheat or winter barley inoculated with W-type or R-type isolates of Pseudocercosporella herpotrichoides

In crops of winter wheat (1986-88) or winter barley (1987-88) inoculated with W-type or R-type isolates of Pseudocercosporella herpotrichoides and sown on different dates (1986) or at different seed rates (1987, 1988) eyespot epidemics developed in different ways. Methods of measuring eyespot incidence/severity during crop growth were compared for their ability to predict eyespot severity at grain filling. Regressions were calculated for eyespot severity score at GS 71 on earlier measurements, either at GS 30/31 (11 methods) or from GS 22 to GS 65 (3 methods). Based on measurements at GS 30/31, all the methods predicted eyespot severity at GS 71 well in plots of winter barley inoculated with W-type isolates (r, 0.83-0.97) but the accuracy of predictions in plots inoculated with R-type isolates was very variable (r, 0.09-0.71). Predictions for 1987 and 1988 were less accurate in wheat than in W-type plots of barley, but did not differ between W-type and R-type plots (r, 0.70-0.89). When the wheat data for 1986 were also included predictions were less accurate, especially in R-type plots (r, 0-0.59). Generally, it was easier to predict eyespot severity at GS 71 in W-type than in R-type plots, especially in barley and in wheat before GS 37/39. Predictions of eyespot severity at GS 71 based on measurements before GS 25 were inaccurate for both wheat and barley. After GS 25 the accuracy of the prediction was generally good in W-type plots and did not improve greatly except in wheat after GS 59. However, there was a steady improvement in the accuracy of the prediction in R-type plots of barley from GS 24 to GS 53. Assessments of eyespot incidence on stems predicted eyespot severity at GS 71 more accurately than assessments on leaf sheaths on wheat after GS 37/39, but were not as good on barley until GS 53.