2,4-Diamino-5-[p-(m-fluorosulfonylbenzamido)phenoxypropyl]-6-methylpyrimidine (2) and three variants in the benzamido moiety have been synthesized via the intermediate 2-amino-6-methyl-5-(p-nitrophenoxypropyl)-4-pyrimidinol and 5-(p-aminophenoxypropyl)-2,4-diamino-6-methylpyrimidine (6); the key reaction was azide displacement of the chloro atom of 2-acetamido-4-chloro-6-methyl-5-(p-nitrophenoxypropyl)pyrimidine (5) followed by reduction of the azidopyrimidine (7) to the 4-aminopyrimidine since the usual ammonia displacement caused cleavage of the nitrophenoxy side chain. Evaluated as an irreversible inhibitor of dihydrofolic reductase, 2 met all the criteria needed for in vivo evaluation; 2 had Ki = 0.003 μM, showed good irreversible inhibition of L1210 dihydrofolic reductase at a 2Ki concentration, and showed no significant irreversible inhibition of dihydrofolic reductase from normal mouse liver, spleen, or intestine at a concentration of >70Ki. However, 2 showed poor penetration of the L1210 cell wall in culture and hence was inactive in vivo. In contrast, N-[p-(4,6-diamino-2,2-dimethyl-1,2-dihydro-s-triazhi-1-yl)hydrocinnamoyl]sulfanilyl fluoride (12; slowed good penetration of the L1210 cell wall in culture and good in vivo activity, but 12 was not a selective irreversible inhibitor of L1210 dihydrofolic reductase since it also inactivated the enzyme from mouse liver, spleen, and intestine. Future st tidies to combine the selectivity of 2 with the in vivo effectiveness of 12 are discussed. © 1969, American Chemical Society. All rights reserved.