Surveys of powdery mildew in wheat and an estimate of national yield losses

Abstract

Three hundred and one wheat crops in Canterbury were sampled to assess the incidence of powdery mildew. In 1948, 181 crops were surveyed and the disease was recorded by collecting samples of the leaf below the flag leaf, between the Feekes growth stages 9 and 10.2. The 1972 survey covered 120 crops, and recordings were taken by sampling flag leaves at around Feekes growth stage 10. Varietal differences in mildew incidence were substantial. The average incidence in 1948 was 6.4% on ‘Hilgendorf’, 0.7% on ‘Cross 7’, 0.2% on ‘Fife Tuscan’, and 0.1% on ‘Dreadnought’; and in 1972 3.3% on ‘Hilgendorf 61’, 0.5% on ‘Aotea’, 0.3% on ‘Kopara’, and 0.1% on ‘Arawa’. Glasshouse trials indicated that both mildew infection and yield reduction varied between varieties. A formula for yield loss derived from the estimation of leaf area affected by mildew at just one time during crop growth is not likely to be reliable. The possibility of analysing mildew infection recorded at the base of the plant and on the middle and the upper leaves at about Feekes growth stage 10 will be investigated to see whether a more reliable estimate of yield loss can be made for individual varieties. Aerial photographs of spray trials in wheat crops did not show any differences in mildew intensity, although the effect of soil-type differences on crop growth was readily discernible. Application of fungicides for the control of powdery mildew is likely to be economic only if restricted to the moderate and highly susceptible varieties in the more fertile areas. Also, crops should be sprayed well before the Feekes growth stage 10. The national loss in yield of wheat crops from powdery mildew infection is estimated at 1.54%.