Exposure of prepubertal, pubertal, and adult mice to 0, 8, 16, 32, 40, or 48 mg 1,3-dinitrobenzene (m-DNB)/kg body weight and measuring responses 1-25 d posttreatment (dpt) demonstrated significant effects on testicular function only at 48 mg/kg dosage. m-DNB had no effect on body or testis weights with the exception of reduced adult mouse testis weights at 22 dpt with 48 mg/kg (p less than .05). None of the exposures resulted in detectable levels of germinal epithelial cells in the ductus epididymis. Exposure of prepubertal and pubertal mice to m-DNB caused only minimal nonsignificant changes in the relative percent of testicular cell types present up to 25 dpt. The adult mice testicular cell type ratios, in particular the round and elongating spermatid populations, changed significantly at doses of 48 mg/kg. Also, a reduction in the percent tetraploid cells occurred at d 1, suggesting these cells may be a primary target of m-DNB action. Caput and caudal sperm from mice exposed to m-DNB prior to puberty did not demonstrate an increased susceptibility to DNA denaturation when analyzed by the sperm chromatin structure assay. However, in pubertal mice, m-DNB exposure further exaggerated the abnormal chromatin structure that normally characterizes sperm during the onset of sperm production. In adult mice, 48 mg/kg resulted in increased susceptibility to DNA denaturation of caput sperm chromatin at 11 dpt (p less than .05) and in caudal sperm at 22 dpt (p less than .01). The abnormal chromatin structure of cauda sperm from adult mice was highly correlated with sperm head morphology abnormalities (ABN; 0.82 to 0.95, p less than .01, 11 and 22 dpt, respectively), but showed lower correlations with dose (0.60 to 0.79, p less than .01, 11 and 22 dpt, respectively). For pubertal mice, a positive relationship was also observed between the variation of sperm chromatin structure abnormalities and ABN. The effect of m-DNB on testicular function in prepubertal and pubertal mice appear to be less pronounced than in adult mice. Furthermore, following exposure to the same dosage, the effect of m-DNB is less severe in adult mice than that observed for adult rats as reported in the companion paper.