Chronic exposure to stressors increases HPA axis activity and concomitantly reduces HPG axis activity. This antagonistic relationship between both these axes has been proposed to underlie the inhibition of reproductive function due to stress. Sexual behavior in males may be the most vulnerable aspect of male reproduction to acute and chronic stress and it has been suggested that alterations in sexual behavior during stress are due to the antagonistic relationship between testosterone and corticosteroids. However, only in a few studies has a correlation between the levels of testosterone and corticosterone, and sexual behavior been made. In this study, we evaluated the effects of different stressors, applied both acute and chronically, on masculine sexual behavior and whether or not these effects on sexual behavior are accompanied by changes in plasma levels of corticosterone and testosterone. Additionally, we evaluated the effect of testosterone treatment on the effects of stress on sexual behavior. Sexually experienced male rats were exposed to one of the following stressors: immobilization (IMB), electric foot shocks (EFS) or immersion in cold water (ICW). Sexual behavior and plasma levels of testosterone and corticosterone were assessed on days 1, 5, 10, 15, and 20 of stress. In a second experiment, males were castrated, treated with 3 different doses of testosterone propionate (TP) and exposed to ICW for 20 consecutive days. Sexual behavior was assessed on days 1, 5, 10, 15, and 20 and steroids were evaluated on day 20. Parameters of masculine sexual behavior were modified depending on the characteristics of each stressor. Mount, intromission and ejaculation latencies increased significantly, the number of mounts increased, and ejaculations decreased significantly in males exposed to EFS and to ICW but not in males exposed to IMB. Associated with these effects, testosterone decreased in the EFS and ICW groups on days 1, 15, and 20. However, corticosterone increased only in males exposed to ICW. In castrated males, TP treatment failed to block the effects of stress by ICW on sexual behavior and corticosterone. These results indicate that the effects of stress on sexual behavior depend on the characteristics of each stressor, and these effects, as well as the decrease in testosterone are not necessarily associated with the increase in corticosterone. The fact that testosterone treatment did not prevent the effects of stress on sexual behavior suggests that other mediators could be involved in the alterations of sexual behavior caused by stress.