Strategies for fungicidal control of take-all

Abstract

Fluquinconazole (Jockey F) seed treatment was applied or not applied, in all possible sequences, in up to six consecutive crops of winter wheat in field experiments. Take-all, caused by the root-infecting fungus Gaeumannomyces graminis var. tritici, was controlled effectively. Grain yield was usually increased by treatment when the disease was moderate or severe in the non-treated crops. Control of the most severe take-all did not result in acceptably high yields or adequate grain quality. Treatment of a second wheat with little take-all did not benefit the subsequent crop. There was, however, a small residual benefit in a second wheat following a treated first wheat (tested once only). Non-treatment of a crop grown after a treated, diseased crop usually resulted in a marked increase in disease. Take-all was controlled by treatment of a crop grown after a treated, diseased crop but the extent of control and of increased yield was often less than that in a treated crop grown after a non-treated crop in the same crop sequence.

It is recommended that seed treatment should be applied to second or third wheat crops (at risk from damaging take-all) and that a break crop should follow the treated crop. Seed treatment should not normally be used in longer sequences of wheat or on take-all decline sites.

Application of fluquinconazole seed treatment to crops grown successively on the same site for up to 4 years did not adversely affect the natural community of saprophytic fungi on the roots. No resistance or decreased sensitivity to fluquinconazole was found in populations of the take-all fungus in these crops.

A DNA probe and restriction-digestion were used to characterise isolates of the fungus from crops that were treated or non-treated in every year as RFLP types, T1, T1c and T2. On this basis, population structure was found to change from year to year but was not affected significantly by fluquinconazole seed treatment. A hypothesis that the greatest proportion of T2 isolates anticipated increased take-all was tested but only partially validated, which precluded the development of these markers for a risk assessment system. T1 types, often more predominant later in crop sequences, were frequently more melanised, suggesting a mechanism for enhanced survival. Other sub-groups of the take-all fungus (A and B), identifiable by PCR after DNA extraction directly from root tissue, were found to correlate closely with the RFLP types. Because of a correlation with insensitivity in the take-all fungus to the take-all-specific fungicide, silthiofam, and selection by host-plant species, the PCR test has potential value in future research.