This study was designed to evaluate the correlation between the follicular biophysical and biochemical indicators in spontaneous (N = 11) and stimulated (N = 110) ovulatory cycles. Ovulation was induced with clomiphene citrate in 14 cycles, gonadotropin-releasing hormone (GnRH) in 12 cycles, and human menopausal gonadotropins in 84 cycles. Patients were studied daily, starting on day 10, until sonographic verification of ovulation. Each woman had serum estradiol (E2) and LH measured daily and progesterone measured only 7 days after ovulation. In addition, the ovaries were imaged transvaginally daily and the two largest follicular diameters, volumes, cross-sectional areas, and circumferences were measured in all follicles 10 mm or larger in diameter. Ultrasonographic measurements of follicles from clomiphene-stimulated cycles were significantly larger than those from spontaneous, GnRH-, and human menopausal gonadotropins-stimulated cycles (P less than .05). Serum E2 and progesterone secretion in human menopausal gonadotropins- and clomiphene-stimulated cycles were significantly higher than in spontaneous and GnRH-stimulated cycles (P less than .01). Women treated with human menopausal gonadotropins developed significantly more follicles than with any other treatment (P less than .05). Correlation analysis indicated that biophysical variables alone (follicular diameter, volume, cross-sectional area, or circumference) were good indicators of normal follicular development and predicted the mid-cycle LH surge in spontaneous (r = 0.81, P less than .001), GnRH- (r = 0.78, P less than .001), and clomiphene citrate-stimulated cycles (r = 0.83, P less than .001). However, in human menopausal gonadotropins-stimulated cycles, both serum E2 levels and ultrasonographic evaluation were necessary to decide the best time for hCG administration (r = 0.55, P less than .001).