Little information is available regarding the time, rhythm, number, and appropriate dosage of human chorionic gonadotropin (HCG) for adequate testing of testicular function in human. The time course of the effect of two, three, or seven HCG injections at intervals of one, five, and two days, respectively, on the plasma levels of testosterone was studied in 11 boys. The first injection induced a progressive and modest rise of T. The second given one day later had little additive effect, maximal values being seen 72 to 120 hr later. In the prepubertal boys to whom several HCG injections were given, testosterone levels reached comparable levels after four injections every five days or seven injections every other day. Although the number of subjects studied was relatively small, these results give some rational basis for the following HCG test: two or four injections at four-day intervals.