One of the most important concerns in the decontamination of aflatoxin-containing feed commodities is the safety of the products for food-producing animals and for human consumption of products derived from these animals. A new method, based on the use of florisil and C18 solid phase extraction columns, was developed for the preparation of extracts from decontaminated peanut meal, which allowed testing with in vitro genotoxicity assays without interference of the residual aflatoxin B1. Recovery of degradation products in the extracts was evaluated by the use of radiolabelled [14C]-aflatoxin B1 (AFB1) added to naturally-contaminated peanut meal (3.5 mg AFB1/kg). The meal was treated by a small-scale version of an industrial decontamination process based on ammoniation. Following decontamination, more than 90% of the label could not be extracted from the meal. AFB1 accounted for about 10% of the radiolabel present in the extractable fraction, indicating a total AFB1 reduction of more than 99%. Decontamination of the meal by a number of other small- and industrial-scale ammonia-based processes resulted in similar efficiencies. Application of the extraction procedure resulted in AFB1-rich and AFB1-poor fractions, the latter containing half of the extractable decontamination products but less than 1% of the residual AFB1. Testing in the Salmonella/microsome mutagenicity assay (TA 100, with S9-mix) of the original crude extracts and AFB1-rich fractions prepared from non-treated and decontaminated meal, showed the positive results expected from the AFB1 contents as determined by HPLC analysis. Analysis and testing of the AFB1-poor fractions showed that the various decontamination processes not only resulted in a successful degradation of AFB1 but also did not produce other potent mutagenic compounds. Slight positive results obtained with these extracts were similar for the untreated and treated meals and may be due to unknown compounds originally present in the meal. Results obtained with an unscheduled DNA synthesis (UDS) and Comet assay with rat hepatocytes supported this conclusion. Positive results obtained with the micronucleus assay, using immortalized mouse hepatocytes (GKB), did not clearly reflect the mycotoxin levels and require further examination. It is concluded that the newly developed extraction procedure yields highly reproducible fractions and hence is very suitable for examining the possible formation of less potent degradation products of aflatoxins in short-term genotoxicity tests.