The inhibitory activity of pyrimethamine (PMA) is 1/290 of the activity of trimethoprim (TMP) against E. coli as evaluated from a plot of C·ko/ko-kapp vs. C. Even at high concentrations the effect of PMA in contrast to TMP seems to be bacteriostatic. Combinations of TMP and PMA reveal an additive effect. In PMA-treated cultures, the slope of the logarithmic growth curve decreases after an initial inhibited growth, and a second steady state is established. This second steady state has a different reason than the one observed in TMP-treated cultures; whereas the second phase in TMP-inhibited cultures depends on the number of germs, in the case of PMA it depends on the number of generations. Using prewashed cell cultures, it was shown that there is no influence on the two steady states in PMA-inhibited cultures; for TMP, however, it was shown that the presence of the first phase is due to an antagonist excreted into the culture medium. These observations hint at differences in the mode of action of TMP and PMA in addition to differences in the affinity to the target enzyme dihydrofolate reductase. Combination of PMA with sulfamethoxazole (SMZ) at concentrations where both drugs are acting only bacteriostatically leads to effects considerably greater than would be expected from simple additivity. The kill rate observed is the same as observed for TMP/SMZ combinations despite of the considerable lower activity of PMA compared to TMP. It is justified to speak of a synergism in action of PMA and SMZ. The results support the assumption that it might be possible to select drug combinations considering the best pharmacokinetical fit and not necessarily the most effective drugs in the series studied