In women, a single follicle is generally ovulated each cycle despite the perfusion of both ovaries by a common blood supply. Accordingly, we questioned whether the ovary containing the dominant follicle may be secreting a substance that suppresses the responses of other follicles to gonadotropins. Ovarian venous blood (5 ml) was collected from six women undergoing laparotomy for indications not related to ovarian dysfunction on days 12–14 after the onset of their last menstrual period. Serum was fractionated by ammonium sulfate precipitation, dialyzed against buffer with 10,000 molecular weight exclusion membranes, and thereafter sequentially eluted through Concanavalin A and Sephadex G-50 columns. The activity of the eluent was assessed as inhibition of ovarian weight increase and serum 17β-estradiol levels in 23-day-old, hypophysectomized, diethylstilbestrol-treated rats (HIFR) challenged with human menopausal gonadotropin (hMG). Sephadex G-50 fractions (elution volume/void volume 1.42–1.55) from patient 1 produced a decrease in ovarian weight (mean % SD, 59 % 0.5 vs. 89.1 % 2.6 g) and a significant (P < 0.01) decrease in serum 17β-estradiol levels (< 25 vs. 215.5 % 12.7 pg/ml). Although peripheral and ovarian venous blood collected from the ovary contralateral to the site of ovulation demonstrated similar Sephadex G-50 elution profiles, when representative fractions were tested by bioassay, no reduction in ovarian weight or serum 17<-estradiol levels was found. In addition, ovarian venous serum from the ovulatory ovary of patients 2 and 3 had a similar Sephadex G-50 elution profile with fractions (elution volume/void volume = 1.48–1.60) which suppressed rat ovarian weight and serum 17β-estradiol concentrations in the hMG-HIFR assay. When active fractions from the G-50 eluents were heated to 56 C or trypsin digested, they lost their ability to suppress ovarian weight and 17β-estradiol secretion in response to hMG stimulation. Estimations of molecular weight by gel permeation ranged between 14,000–18,000 for patients 1–3. Bioassay results from representative fractions obtained by ampholyte displacement chromatography suggested that the isoelectric point of active material was between pH 5.8–6.5 for patients 1–3. Similarly processed samples from three anovulatory patients contained no inhibitory activity in the bioassay. We report the identification of a heat- and trypsin-labile substance secreted directly into the venous drainage from the ovary containing the dominant follicle which suppresses the follicular response to gonadotropins. That this protein is not secreted in large amounts by anovulatory ovaries was evidenced by the failure of the bioassay to detect inhibitory activity in the venous drainage of the contralateral ovary of patients 1–3 as well as the ovarian venous effluents from three anovulatory women. Although confirmatory data await further studies, it is tempting to speculate that this potential intra- and/or interovarian regulator of folliculogenesis may mediate dominance of the preovulatory follicle by an active process, such that after the selection of the dominant follicle, the gonadotropin responsivity of other follicles on the same and contralateral ovaries is suppressed. (J Clin Endocrinol Metab54: 1091, 1982)