Substantial reductions in winter wheat diseases caused by addition of straw but not manure to soil

Over a period of 3 years, five agronomically distinct crops of winter wheat were grown in plots in which straw (1 kg m −2 ), manure (4 kg m −2 ) or nothing were incorporated into the soil. Plant establishment and height, but not leaf area per tiller, were lower in straw‐treated plots. Fertilizer regimes differed between years. Soil and leaf nitrogen were recorded; there was no obvious link between N and any disease or soil amendment. The numbers of leaf layers scorable for disease were similar in all treatments. At the end of the season, plants from straw‐treated plots had consistently reduced septoria tritici blotch (caused by Mycosphaerella graminicola ), powdery mildew [caused by Erysiphe ( Blumeria ) graminis ], brown rust (caused by Puccinia recondita ) and foot rot (caused by Fusarium spp.). Early on, M. graminicola was worse in straw‐treated plots. In manure‐treated plots, P. recondita was reduced but effects on other diseases were inconsistent and slight. A fungicide, chlorothalonil, was applied in one crop; its effects did not interact with those of other treatments. Mycosphaerella graminicola was not suppressed by straw in outdoor pot experiments. Late in the season, straw‐treated plants had significantly higher leaf silica ( P < 0·01). In a glasshouse experiment, plants supplied with silicon had less E. graminis infection ( P < 0·001) and higher leaf silica, but effects on M. graminicola were inconsistent. A prior inoculation of M. graminicola primed plant defences against a subsequent attack of E. graminis , but only in the presence of adequate Si. It is postulated that straw acts in the field by increasing Si availability.