Growth curves of the 5 commonly used Ames Salmonella tester strains have been measured turbidimetrically in semi-solid agar. Lag times, doubling times and maximum cell densities have been calculated for each of the 5 strains. The time dependence of reversion has been studied in the standard plate incorporation assay using 1-h pulsed doses of (a) bromoethane, a volatile chemical mutagen, and (b) 1-h exposures to visible light. Essentially no reversion takes place during the first 4 h after plating. Reversion is detectable between hours 4 and 16. The cumulative or integrated revertants versus time curve has the characteristics of a growth curve. Conversely the derivatives of the growth curves resemble the curves obtained in the pulsed mutagenicity studies. Thus, the reversion rate in any given 1 h interval is proportional to the growth rate during that same interval. These results suggest that mutagenic chemicals must be present during the bacterial growth cycle (about 4-16 h after plating) in order to revert the tester strains. Short-lived chemical mutagens, then, should produce enhanced results if plated 6-8 h after the bacteria. We have confirmed this for N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), 9-aminoacridine and 2-aminoanthracene (with S9).