Genetic Control of Aflatoxin Production in Maize

Publisher Website
Google Scholar
Abstract
Twenty-one possible F1 single crosses among seven inbred maize lines (Zea mays) were planted in 20 replicates of single-hill plots, and developing ears were inoculated with Aspergillus flavus by a pinboard technique. Inoculated kernels from each ear were assayed for aflatoxin by high-performance liquid chromatography. To evaluate the effect of the number of replicates on the standard error of the mean, we used the error mean square for 20 replicates and computed standard errors of the mean assuming 2, 4, 6, 8, 10, 12, 14, 16, 18, and 20 replicates. Relative efficiencies of changing r1 to r2 replicates were calculated. Comparison of various numbers of replicates versus standard error of the mean suggested that eight replicates would be the most efficient number for aflatoxin B1 field studies. Reductions in the standard errors of the mean were not appreciable with more than 10 replicates because relative efficiencies were changed little from 10 to 12 to 14 replicates, whereas standard errors were reduced when the number of replicates was changed from 6 to 8 to 10 replications. In the diallel analyses, genotypic differences were significant for both aflatoxin B1 and B2, as was the variance associated with general and specific combining ability effects. Specific combining ability sum of square estimates accounted for about 65% of the genotype sum of squares. Coefficients of variation were very high, 92 and 90% for aflatoxin B1 and B2, respectively. The estimates of general combining ability effects and rankings of aflatoxin levels from crosses and parental line means were in good agreement with a previous study conducted at the University of Missouri, even though only inoculated kernels were analyzed in this study versus whole ears in the previous study.