A quantitative method based upon a stochastic model for the appearance of initiated cells and their clonal growth was used to estimate cell birth and death rates in enzyme-altered liver foci (EAF). gamma-Glutamyltranspeptidase (gamma-GT)-positive foci were initiated in livers of female SPF Wistar rats by a single application of N-nitrosomorpholine. Serial terminations during and after stop of promoter treatment with either phenobarbital (PB) or alpha-hexachlorocyclohexane (alpha-HCH) provided information on the growth and regression of the EAF. Simultaneous labeling index (LI) measurements were obtained via single injections with [3H]thymidine. No significant increases of the LI were observed with PB or alpha-HCH treatment. Since both agents are strong liver promoters we conclude that the growth of gamma-GT-positive foci is mainly due to a decrease in the rate of apoptosis. Indeed, our analysis supports this conclusion but determines that the abrogation of homeostatic control during promoter treatment is subtle. The ratio of cell death and cell birth rate is found to be decreased only slightly during promoter treatment and slightly increased after stop of promotion. For the mathematical analysis, two distinct focal growth scenarios were employed: (i) volume growth, i.e., all cells within individual foci cycle actively with the same rate, and (ii) surface growth where only cells on the surface of foci cycle actively while interior cells are resting. The model based upon scenario (ii) provides a better fit to the data and is more consistent with the experimental observations indicating heterogeneity of cell division rates within individual foci.