Two assays based on the inhibition of 3H-thymidine incorporation into DNA were used to measure either the antimetabolic or the antiproliferative effects of anticancer drugs. A direct comparison of the two assays was made with cell suspensions obtained from 11 ovarian cancers and 22 malignant melanomas. Drugs with different effects on cell cycle phases were tested by both assays, for a total of 53 drug comparisons. When the sensitivity indices specific for each system was used, a significant association (p less than 0.01) was noted between the two assays. The agreement of both assays in defining in vitro sensitivity or resistance was 100% for ovarian cancer. For melanoma, 97% of samples resistant to the antimetabolic assay were also resistant to the antiproliferative assay; whereas, only 45% of samples sensitive to the antimetabolic assay were sensitive to the antiproliferative assay.