Objective To analyze the effect of thalidomide on the progression of endometriotic lesions experimentally induced in rats and to characterize the pattern of cell proliferation by immunohistochemical Proliferating Cell Nuclear Antigen (PCNA) labeling of eutopic and ectopic endometrium. Methods Fifteen female Wistar rats underwent laparotomy for endometriosis induction by resection of one uterine horn, isolation of the endometrium and fixation of a tissue segment to the pelvic peritoneum. Four weeks after, the animals were divided into 3 groups: control (I), 10mg/kg/day (II) and 1mg/kg/day (III) intraperitoneal thalidomide for 10 days. The lesion was excised together with the opposite uterine horn for endometrial gland and stroma analysis. Eutopic and ectopic endometrial tissue was submitted to immunohistochemistry for analysis of cell proliferation by PCNA labeling and the cell proliferation index (CPI) was calculated as the number of labeled cells per 1,000 cells. Results Group I showed a mean CPI of 0.248 ± 0.0513 in the gland and of 0.178 ± 0.046 in the stroma. In contrast, Groups II and III showed a significantly lower CPI, that is, 0.088 ± 0.009 and 0.080 ± 0.021 for the gland (p < 0.001) and 0.0945 ± 0.0066 and 0.075 ± 0.018 for the stroma (p < 0.001), respectively. Also, the mean lesion area of Group I was 69.2 mm2, a significantly higher value compared with Group II (49.4 mm2, p = 0.023) and Group III (48.6 mm2, p = 0.006). No significant difference was observed between Groups II and III. Conclusion Thalidomide proved to be effective in reducing the lesion area and CPI of the experimental endometriosis implants both at the dose of 1 mg/kg/day and at the dose of 10 mg/kg/day.