Randomly chosen, commercially available, market basket samples of on-the-shelf brands of peanut butter were routinely tested for the presence of aflatoxin from 1982 to 1989. Standard analytical procedures were employed which included high performance liquid chromatography and thin layer chromatography. During the eight years studied, 74% of the samples were positive for the presence of aflatoxin. Of those samples testing positive which excluded no detect samples or "zero values", 8.6% were found to be contaminated between 20 and 50 ppb, 3.7% were between 50 and 100 ppb, and 2.2% were above 100 ppb. The total for this evaluation approach was 14.5%. When all samples including nondetectable were considered, only 9.8% of 2510 samples exceeded the U.S. Food and Drug Administration action level for total aflatoxin. With the exception of 1986 and 1987, aflatoxin levels in commercially available peanut butter offered for sale to the public in this particular state have been well controlled. It is a monitoring effort such as this, conducted as checks on the final product on sale to the public, that provides the final line to protect public health from exposure to mycotoxins. Based on these regional data for an eight-year period reported herein, it is apparent that the findings could be used in conjunction with other regional and national studies to support a reduction of the 20 ppb U.S. action level.