Abstract
The interlaboratory Ames test variability of the Salmonella/microsome assay was studied by comparing 12 sets of results generated in the frame of the International Program for the Evaluation of Short‐Term Tests for Carcinogens (IPESTTC). The strategy for the simultaneous analysis of test performance similarities over the whole range of chemicals involved the use of multivariate data analysis methods. The various sets of Ames test data were contrasted both against each other and against a selection of other IPESTTC tests. These tests were chosen as representing a wide range of different patterns of response to the chemicals. This approach allowed us both to estimate the absolute extent of the interlaboratory variability of the Ames test, and to contrast its range of variability with the overall spread of test responses. Ten of the 12 laboratories showed a high degree of experimental reproducibility; two laboratories generated clearly differentiated results, probably related to differences in the protocol of metabolic activation. The analysis also indicated that assays such as Escherichia coli WP2 and chromosomal aberrations in Chinese hamster ovary cells generated sets of results within the variability range of Salmonella; in this sense they were not complementary to Salmonella.