A method of testing postnatal carcinogenicity in rats was elaborated. The substances to be tested were administered from the 1st day after birth to the 5th, 10th and 20th days of age. The subsequent supply of a substance in diet from weaning (28th day of age) up to the end of the first year of survival was an optimal combination. A total of 12 substances were tested by this method. For all substances it was possible to provide a comparison with the results of long-term studies. For nine substances carcinogenicity was demonstrated in both tests, out of them in eight cases in identical organ systems. In three cases no carcinogenicity was ascertained in the postnatal study, in two compounds identical negative results were obtained by the two approaches. In one case (the cytostatic TS-160) development of sarcomas was found at the site of subcutaneous administration in the long-term study, and this effect was not observed in the postnatal study. Results identical for the two methods (carcinogenic or noncarcinogenic) were achieved in 11 substances, i.e. 91.7%. In addition, in 10 compounds, for which carcinogenicity was demonstrated either in the postnatal study or in the long-term study, the occurrence of tumors was found in identical organ systems after 8 compounds which is an 80% agreement. In mice, this identity of organs was demonstrated only in 62.5% of the compounds tested. The present results demonstrated that the use of postnatal carcinogenicity test in rats offers the determination of possible carcinogenic effect of the compound tested with a high probability, under economically more advantageous conditions, and with almost the same qualitative results as with the use of the long-term tests for carcinogenicity.