We previously demonstrated that androgens alone, in the complete absence of gonadotropins, initiated qualitatively complete spermatogenesis in hypogonadal (hpg) mice. Although germ cell to Sertoli cell ratios were normal in hpg mice with androgen-induced spermatogenesis, testicular size. Sertoli cell and germ cell numbers only reached 40% of those of non-hpg mice, and Sertoli cell numbers were unaffected by androgen treatment started at 21 days of age. We postulated that these observations were due to diminished gonadotropin-dependent. Sertoli cell proliferation during perinatal life while the Sertoli cells still exhibited normal carrying capacity for mature germ cells. In order to test this hypothesis, we examined the effects of administering androgens and gonadotropins to hpg mice during the first 2 weeks of postnatal life when Sertoli cells normally continue to proliferate. The study end-points were Sertoli and germ cell numbers in hpg mice following induction of spermatogenesis by 8 weeks treatment with 1 cm subdermal silastic testosterone implants. Newborn pups (postnatal day 0-1) were injected s.c. with recombinant human FSH (rhFSH) (0.5 IU/20 microliters) or saline once daily for 14 days, with or without a single dose of testosterone propionate (TP) (100 micrograms/20 microliters arachis oil) or human chorionic gonadotropin (hCG) (1 IU/20 microliters). Untreated hpg and phenotypically normal littermates were studied as concurrent controls. At 21 days of age, all treated weaning mice received a 1 cm silastic subdermal testosterone implant and, finally, 8 weeks after testosterone implantation, all mice were killed. As expected, qualitatively complete spermatogenesis was induced in all groups by testosterone despite undetectable circulating FSH levels. Exogenous rhFSH increased testis size by 43% (P < 0.002) but a single neonatal dose of either TP or hCG reduced the FSH effect although neither TP nor hCG had any effect alone. In contrast, a single neonatal dose of TP or hCG increased final seminal vesicle size whereas FSH had no effect. FSH and TP treatment significantly increased absolute numbers of testicular spermatids compared with saline treatment, whereas hCG and TP significantly increased testicular sperm when expressed relative to testis size. Stereological evaluation of Sertoli and germ cell numbers demonstrated a rise in the absolute numbers of Sertoli and all germ cell populations induced by neonatal administration of hormones. When expressed per Sertoli cells the numbers of germ cells in the treated mice were between 85 and 90% of non-hpg controls. We conclude that exogenous FSH treatment during the first 2 weeks of postnatal life, coinciding with the natural time of Sertoli cell proliferation, increases Sertoli cell numbers and thereby the ultimate size of the mature testis and its germ cell production. Thus neonatal gonadotropin secretion may be a critical determinant of the sperm-producing capacity of the mature testis. In addition, neonatal exposure to androgens could be important for the imprinting of sex accessory organs in hpg mice, with the long-term effects of altering the sensitivity of the accessory organs to exogenous testosterone later in life.