This study addressed whether reduced Sertoli cell number or manipulation of the neonatal hormone environment has an influence on final Leydig cell number per testis in the rat, by applying neonatal treatments known to affect these parameters, namely administration of a GnRH antagonist (GnRHa) or diethylstilboestrol (DES, in doses of 10, 1 or 0.1 microg per injection). The effect of treatment with either of two 'environmental oestrogens', bisphenol-A (Bis-A) or octylphenol (OP), was also evaluated. Leydig (3beta-hydroxysteroid dehydrogenase immunopositive) cell development and function (plasma testosterone levels) were studied through puberty into adulthood. Treatment with GnRHa impaired testis growth, Leydig cell (nuclear) volume per testis and testosterone levels during puberty, when compared with controls, but final Leydig cell volume/number in adulthood was comparable with controls. As adult testis weight was reduced by 45% in GnRHa-treated rats, the percentage Leydig cell volume per testis was approximately double (p < 0.01) that in controls, and also at day 35. Testosterone levels in adulthood in GnRHa-treated rats were lower (p < 0.01) than in controls but were within the lower end of the normal range. Treatment with DES caused largely dose-dependent suppression of testis growth, Leydig cell (nuclear) volume per testis and testosterone levels up to day 35. Although by adulthood, Leydig cell volume/number per testis was comparable with controls in DES-treated rats, testosterone levels remained grossly subnormal. Neonatal treatment with either Bis-A or OP had little consistent effect on any of the parameters studied except that both treatments significantly elevated testosterone levels on day 18, as did treatment with DES-0.1 microg. The present findings are interpreted in the context of what is known about the hormonal regulation of Leydig cell development. These lead to the conclusion that final Leydig cell number per testis is not determined by the number of Sertoli cells per testis and appears not to be influenced in any major way by gonadotrophins, androgens or oestrogens in the first 2 weeks of postnatal life. This implies that adult Leydig cell number may be determined prior to birth.