Some factors affecting the growth and yield of winter wheat grown as a third cereal with much or negligible take-all

Winter wheat cv. Avalon was sown in autumn 1981, 1982 and 1983 on a clay loam soil following two cereal crops. Multifactorial experiments tested the effects of combinations of the following eight factors, each at two levels: rotation, sowing date, timing of nitrogen, amount of nitrogen, growth regulator, pesticide, spring fungicide and summer fungicide. The best 16-plot mean grain yields in 1982–4 were respectively 8·7, 10·2 and 11·1 t/ha. Rotation had the largest effect on grain yield. Wheat following barley was severely infected with take-all and yielded, on average over 3 years, 2·2 t/ha less than wheat following oats. Take-all was more severe on wheat sown in mid-September than in mid-October; its effects on yield were lessened by early timing of N in 1982. Take-all decreased growth and N uptake mainly after anthesis, and also number of ears and dry weight per grain. Sowing in mid-September compared with mid-October decreased yield of wheat after barley by an average of 0·8 t/ha because take-all was more severe. Early sowing had negligible effects on grain yield of wheat after oats, but increased straw dry weight by 1·1 t/ha. Spring fungioide increased yield by an average of 0·3 t/ha. Effects were larger after barley than after oats, associated with a greater incidence of eyespot after barley. Summer fungioide increased yield by an average of 0·3 t/ha. Foliar diseases were slight in all 3 years. Fusarium ear blight and sharp eyespot were prevalent in 1982 and were not well controlled by the fungioide treatments. Fungicide temporarily decreased the incidence of some components of the mioroflora on the ears. Pesticide increased grain yield of wheat after oats only in 1984, when aphids on the ears were numerous. Aphids were present on early-sown plots in all three autumns but there was little barley yellow dwarf virus infection even without pesticide. Pesticide always decreased the number of nematodes after harvest to fewer than present before sowing. Populations never approached levels expected to affect yield. Early N application (main application early March) resulted in a larger grain yield in 1982 than N applied a month later. In 1983 and 1984 grain yield and N uptake by the grain were greater with the late application, especially when wheat was sown early. The soil contained more mineral N in the autumn of 1982 and 1983 than in 1981. Straw weight was always greater with early than with late application. Increasing the amount of N applied from 163 to 223 kg/ha increased N uptake by 40 kg/ha and grain yield by 0·5 t/ha after oats and by 0·6 t/ha after barley. N uptake in grain plus straw by the best yielding crops ranged from 205 kg/ha in 1982 to 246 kg/ha in 1984. Chlormequat applied at the start of stem extension shortened the stems at maturity by 2 cm each year. In 1984 it inoreased yield of early-sown wheat by 0·3 t/ha and also decreased lodging, which did not occur in the first 2 years.