Forty-one anovulatory patients consisting of twenty with primary amenorrhea and twenty-one with secondary amenorrhea were investigated by ovarian biopsy and endocrine studies. The ovaries of these patients could be divided into two groups, one with ovarian follicles and the other without ovarian follicles. The ovaries with ovarian follicles could be further divided on the basis of the presence or absence of tertiary follicles; groups with tertiary follicles (high developmental stage) and without tertiary follicles (low developmental stage).<BR>The serial measurements of serum gonadotropins and urinary total estrogens were performed in all anovulatory patients. The pituitary responsiveness to synthetic luteinizing hormone releasing factor (LRF) and the ovarian responsiveness to gonadotropins [human menopausal gonadotropin (HMG) and human chorionic gonadotropin (HCG)] in patients with primary and secondary amenorrhea were tested and compared with those in normal control subjects in the follicular phase. Histological and histochemical examination of the ovaries showed some significant differences between the ovaries of the patients with primary amenorrhea and those with secondary amenorrhea.<BR>Of 20 patients with primary amenorrhea, 10 had no ovarian follicles. By contrast, of 21 patients with secondary amenorrhea, 16 had highly developed follicles. More abundant steroidogenic foci were observed in most ovaries of the patients with secondary amenorrhea. The serum levels of follicle stimulating hormone (FSH) were closely related with the presence or absence of ovarian follicles as Goldenberg et al. (1973) had reported previously.<BR>Both primary amenorrheic patients without ovarian follicles and secondary amenorrheic patients with low developed follicles showed endocrinologic pictures of hypergonadotropic hypogonadism. Serum luteinizing hormone (LH) and FSH responses to LRF in these patients were markedly greater than those in normal subjects of the follicular phase. The patients with secondary amenorrhea could be divided into two groups; one with low developed ovarian follicles and the other with highly developed follicles, according to FSH responsiveness to LRF.<BR>The ovarian volume and the developmental stage of follicle growth were closely correlated with the urinary excretion of total estrogens and the ovarian responsiveness tc gonadotropin stimulation. The ovarian response to gonadotropin in patients with secondary amenorrhea was usually greater than that seen in patients with primary amenorrhea. Then, it seemed that the etiologic lesion of most patients with primary amenorrhea existed in the ovary and that of most patients with secondary amenorrhea in the hypothalamus.<BR>In most patients with highly developed ovarian follicles, induction of ovulation was successfully achieved by HMG-HCG treatment, the total amount of HMG being less than 1500 I.U.. Ovulation could be induced in 33.3% of patients with primary amenorrhea and in 61.5% of patients with secondary amenorrhea. The therapeutic results were in accord with histological findings of the ovary and the responsiveness to gonadotropin stimulation.