There is accumulating evidence that in adult men excessive amounts of gonadotropins induce testicular desensitization to further gonadotropin stimulus. We evaluted the effects of endogenous gonadotropins and of repeated doses of exogenous human chorionic gonadotropin (hCG) on steroidogensis by studying prepubertal and prepubertal boys. The boys received either two intramuscular injections of hCG 4 days apart (protocol I) or four injections at 3- to 4-day-intervals (protocol II). In protocol I, serum testosterone, 17α-hydroxyprogesterone, and estradiol were measured basally and for 6 days after the second injections, and in protocol II, before each injection and 4 days after the last injection. In the prepubertal boys, serum testoterone increased from very low basal levels to 10.3 (protocol I) and 8.3 nmol/liter (protocol II). In protocol I the increase after the first injection was 64-fold and in protocol II there was an increase after each injection to a final level 144-fold of the basal. No significant changes were seen in the estradiol levels. In the pubertal boys at genital stage G2, the serum testosterone levels increased after the first two injections, but at genital stage G3, the levels increased only after the first injection. Maximal testoterone increases were 27- and 8-fold, respectively. In pubertal boys estradiol levels inncreased progressively throughout the stimulation. The major testosterone response was after the first does of hCG and repeated doses, at least in the pubertal boys, increased estradiol but not testosterone levels, thus causing an estrogen-mediated 17,20-lyase block. We there fore suggest that a single-dose hCG test deserves further evalution for diagnostic use.