EVALUATION OF FUNGICIDES APPLIED TO SOIL TO CONTROL NATURALLY-OCCURRING TAKE-ALL USING A BALANCED-INCOMPLETE-BLOCK DESIGN AND VERY SMALL PLOTS

Six sterol biosynthesis-inhibiting fungicides representing several com binations of properties were applied to soil to control naturally-occu rring take-all (caused by Gaeumannomyces graminis var. tritici) in win ter wheat in field experiments in two successive years. The average ta ke-all severity category was never more than moderate in the different clay-loam and sandy loam sites used in each year. At each site in eac h year there were six treatments and an untreated control in an arrang ement based on a balanced-incomplete-block design for six treatments i n 10 blocks each with three treatments. Each block had three treated p lots and a control plot and was paired with the complementary block of three treatments (plus control) to form a complete replicate, of whic h there were 30 per site. Take-all assessments in June or July showed that after incorporation into the seed bed (at 2 kg ha-1 and sometimes at 1 kg ha-1) in autumn. two non-volatile, strongly lipophilic compou nds, nuarimol and triadimenol, with good intrinsic toxicity to the tak e-all fungus and slow rates of degradation, partially controlled take- all. However, another compound, flutriafol, with similar properties to nuarimol and triadimenol, controlled take-all less. Two slightly vola tile, strongly lipophilic compounds, flusilazole and penconazole, with good intrinsic activity, were less effective (at 2 kg ha-1). A volati le, less lipophilic compound, PP 969, with less intrinsic activity, al so partially controlled take-all, but only after application as a dren ch in the spring (2 kg ha-1). The most effective treatments were gener ally more effective the greater the level of disease (as indicated by assessments of disease in control plots), especially in spring assessm ents of disease. Although flutriafol did not perform as expected, it s till seems reasonable to conclude that the requirements for a soil-app lied fungicide to control take-all are likely to be: (i) good intrinsi c fungitoxicity, (ii) some mobility in soil water (i.e. not strongly l ipophilic), and (iii) season-long persistence.