Fermentation Biochemistry Research, National Center for Agricultural Utilization Research, U.S. Department of Agriculture, Agricultural Research Service, Peoria, IL 61604
Go to article:
Accepted 7 November 1996.
Eighteen bacterial strains were individually assayed against Gibberella pulicaris (5 × 105 conidia per ml) by coinoculating antagonist and pathogen in wounds in cv. Russet Burbank potatoes. All antagonist concentrations (106, 107, and 108 CFU/ml) decreased disease (38 to 76% versus control, P < 0.05). When four strains were assayed at 11 concentrations (range 105 to 108 CFU/ml) against G. pulicaris, linear regression of the log-dose, log-response data was significant for all four strains (P < 0.001 to 0.01, R2 = 0.50 to 0.74). Challenging G. pulicaris with all possible antagonist pairings within 2 sets of 10 antagonist strains (5 × 105 CFU of each strain per ml) resulted in 16 of 90 pairs controlling disease better than predicted based on averaging the performance of the individual strains making up the pair (P < 0.10). Successful pairs reduced disease by ~70% versus controls, a level of control comparable to that obtained with 100 times the inoculum dose of a single antagonist strain. Neither strain genus nor soil of origin were useful in predicting successful antagonist pairs. Factors potentially influencing dose-response relationships and the effectiveness of antagonist pairs in controlling disease are discussed.
The American Phytopathological Society, 1997