We studied the relationship between infection, renal scarring, and antimicrobial therapy in pyelonephritis by an experimental model of the disease. Our specific concern was to identify those aspects of the acute phase of infection associated with the induction of the inflammatory response. Our hypothesis was that the degree of the inflammatory response and the extent of the subsequent lesion in the kidney depended on the rate at which a critical number of microorganisms was reached in the kidney. We proposed that the rapid increase in bacterial numbers in the kidney was the stimulus initiating the inflammatory response and not the total number of microorganisms in the kidney. In the experimental investigation, we treated animals with induced renal infection at varying intervals after the establishment of infection, and we determined the effect of antimicrobial therapy on the bacteriologic, gross, and histopathologic features of the disease. The early events relating to the bacterial invasion of the kidney were important determinants but the hypothesis needed extending to account for the fact that antimicrobial agents, administered after bacterial numbers had reached a plateau, still reduced markedly the damage to the kidney. Two factors seem to be involved in the genesis of the pyelonephritic lesion. The first is the rate of acquisition of a bacterial population by a previously sterile organ, but an additional and important component is the total number of microorganisms in the kidney up to 4 days after challenge. Clinically, the results are relevant in that they demonstrate that renal scarring can be prevented or significantly reduced by prompt antimicrobial therapy.