Reduced caloric intake has been shown to inhibit reproductive cycles in females of several mammalian species. Previous studies have shown that increased negative feedback efficacy of estrogen on gonadotropin secretion may be responsible. The present study was designed to test the alternate hypothesis that caloric restriction alters the positive feedback efficacy of estrogen on gonadotropin secretion. Adult, cycling female rats were placed on reduced food intake (R) equal to 50% of that consumed by ad/libitum-fed controls (C). When R rats stopped cycling, both R and C rats were ovariectomized (OVX) and immediately implanted subcutaneously with a Silastic capsule containing 100 microg 17beta-estradiol (E2). Blood samples were obtained at 0900-1000 hr and 1600-1730 hr on Days 2, 4, 6, 8, 10, 12, and 14 after OVX and implantation. Follicle-stimulating hormone (FSH), luteinizing hormone (LH), and E2 were measured by radioimmunoassay in duplicate aliquots. Results indicate that underfed female rats retain the ability to respond to elevated estrogen levels with an afternoon surge of gonadotropin which is present for at least 14 days for LH. By contrast, FSH surges in R rats became progressively smaller and were no longer significant after Day 10. The present results also demonstrate that the response of R rats to elevated estrogen levels is significantly greater than that of C rats on Days 2-4 for FSH and 2-14 for LH. It is concluded that an inability to respond to elevated estrogen levels with an afternoon LH surge is not the cause of the cessation of normal estrous cycles. The progressive decrease in the afternoon surge of FSH may be, at least partly, responsible for the decreased follicular development observed in underfed rats. Possible explanations of the enhanced LH response to the positive feedback of estrogen are discussed.