Quantitative structure-activity relationships have been formulated for the inhibition of Leishmania major dihydrofolate reductase (DHFR) and for inhibition of promastigote cell growth by a series of 4,6-diamino-1,2-dihydro-2,2-dimethyl-1-(3-substituted-phenyl)-s-triazine s. The inhibition of DHFR is best correlated by a modified variable for hydrophobicity of the 3-X substituent (pi'3), an alkoxy group indicator variable (IOR), a disposable parameter (beta) obtained by iteration, and a variable that parameterizes steric effects (MR) in the equation, log 1/Ki = 0.65 pi'3 - 1.22 log (beta X 10 pi'3 + 1) - 1.12IOR + 0.58MRY + 5.05 (r = 0.965). The EC50 values for triazine inhibition of L. major cell growth in culture are correlated by the equation log 1/EC50 = 0.21 pi 3 + 0.44 log 1/Ki + 0.53 (r = 0.960). When compared to DHFR from human, other vertebrates, and E. coli, L. major DHFR differs in that it optimally binds triazine congeners that are much more hydrophobic. Furthermore, in contrast to other DHFR's studied, triazine binding to L. major DHFR does not seem to be influenced by the electronic characteristics of the 3-X substituent of the parent triazine molecule. However, L. major DHFR is more sensitive to the steric effects and polarizability of the 3-X substituent. Our results indicate that triazines inhibit L. major promastigote growth via direct inhibition of DHFR as is shown by the good correlation between log 1/Ki values for inhibition of the purified enzyme and log 1/EC50 values for inhibition of cell culture growth. Two lipophilic, sterically large analogues of this triazine series showed selectivity for L. major DHFR over human DHFR. Further optimization of the MR and IOR terms in the above QSAR equations may provide even more selective inhibitors.