The contraceptive regimen consisting of levonorgestrel and quinestrol (EP-1) has been shown to be effective in several types of wild rodents. In the present study, we investigated the effect of EP-1 and its two components on fertility and spermatogenesis to elucidate the mechanisms underlying its contraceptive effect. Sprague-Dawley rats were treated with 0.33 mgkg(-1) quinestrol (E group), 0.67 mgkg(-1) levonorgestrel (P group) or their combination (EP group) for 7 days and then killed on Days 21 or 42 after treatment for tissue analysis. On Day 21, the weight of the cauda epididymis decreased significantly, while the weight of the adrenal gland increased significantly in the E and EP groups compared with the weights in the control group. In addition, there was a significant decrease in sperm number in the E and EP groups compared with the control group and there was less staining for the androgen receptor and Wilms' tumour nuclear protein 1 in the E and EP groups. The primary defects in E- or EP-treated rats were abnormal spermiogenesis, lack of elongating spermatids, and pachytene spermatocyte arrest. Analysis of MutL homologue 1 revealed that EP treatment inhibited chromosome recombination during meiosis, but did not cause obvious genetic abnormalities. These data demonstrate that quinestrol, alone or in combination with levonorgestrel, induces subfertility in male rats mainly by interfering with germ cell differentiation. Thus, EP-1 or E alone may be effective contraceptive regimens for fertility control in rodents.