First, second, and sixth authors: Department of Plant Pathology, The Ohio State University, Ohio Agricultural Research and Development Center, Wooster 44691; third author: Department of Plant Pathology, University of Kentucky, Princeton 42445; fourth author: Department of Plant Pathology, North Dakota State University, Fargo; and fifth author: Plant Science Department, South Dakota State University, Brookings 57007
Go to article:
Accepted for publication 31 August 2006.
A meta-analysis of the effect of tebuconazole (e.g., Folicur 3.6F) on Fusarium head blight and deoxynivalenol (DON) content of wheat grain was performed using data collected from uniform fungicide trials (UFTs) conducted at multiple locations across U.S. wheat-growing regions. Response ratios (mean disease and DON levels from tebuconazole-treated plots, divided by mean disease and DON levels from untreated check plots) were calculated for each of 139 studies for tebuconazole effect on Fusarium head blight index (IND; field or plot-level disease severity, i.e., mean proportion of diseased spikelets per spike) and 101 studies for tebuconazole effect on DON contamination of harvested grain. A random-effects meta-analysis was performed on the log-transformed ratios, and the estimated mean log ratios were transformed to estimate the mean (expected) percent control for IND ( CIND ) and DON ( CDON). A mixed effects meta-analysis was then done to determine the effects of wheat type (spring versus winter wheat) and disease and DON levels in the controls on the log ratios. Tebuconazole was more effective at limiting IND than DON, with CIND and CDON values of 40.3 and 21.6%, respectively. The efficacy of tebuconazole as determined by the impact on both IND and DON was greater in spring wheat than in winter wheat (P < 0.01), with a 13.2% higher CIND and a 12.4% higher CDON in spring wheat than in winter wheat. In general, CIND and CDON were both at their lowest values (and not significantly different from 0) when mean IND and DON in the controls, respectively, were low (≤2% for IND and <1 ppm for DON). CIND was 25% higher in studies with mean IND between 2 and 15% than in studies with mean IND ≤ 2%, whereas CDON was 28.8% higher in studies with mean DON between 1 and 10 ppm than in studies with mean DON < 1 ppm. The between-study variance was significantly greater than 0 (P < 0.01), indicating considerable (unexplained) variability in percent control.
© 2007 The American Phytopathological Society