Field tests of bacteria and soil-applied fungicides as control agents for take-all in winter wheat

Putative biological and chemical treatments for controlling take-all were used in each of three consecutive years at two locations where winter wheat crops were grown in naturally-infested fields. The chemical treatments more often decreased take-all than the biological treatments, but no treatment consistently and significantly decreased take-all, nor did any cause a significant increase in yield. An isolate of Bacillus cereus var. mycoides and one of B. pumilis, applied as soil drenches in autumn or spring, or in the seed furrows, were usually ineffective. Of the few significant effects on disease, half were associated with increases and half with decreases, and most occurred in April and did not persist to late June. Two strains of Pseudomonas fluorescens applied to the seed were ineffective. The fungicide benomyl, applied as a drench in autumn and spring at 20 kg/ha was ineffective, while nuarimol, applied as a drench in autumn at 2 kg/ha was sometimes effective. Nuarimol incorporated into the seed bed at 2 kg/ha was the most effective treatment. In analyses using a functional relationship model for data from treated and untreated plots 12% of 176 data sets for biological treatments, 38% of 96 data sets for chemical treatments and 81% of 16 data sets for combined treatments showed increasing efficiency of the treatment with increasing disease intensity. These findings also demonstrate an additional advantage of the experimental design, namely that treatments are tested at different disease intensity levels within fields.