Inhibin B appears to be the physiological feedback signal for FSH. Herein the determination of serum levels of inhibin B, FSH, LH and testosterone in 148 infertile patients and their association with clinical findings and seminal parameters are reported. A significant negative correlation of FSH and inhibin B (r = -0.60) was found. LH levels showed a significant inverse correlation (r = 0.37), but a weak regression (c0 = 0.01). No correlation with testosterone levels occurred. A significantly positive correlation was observed between testis volume and inhibin levels (r = 0.39) as well as between sperm count and inhibin levels (r = 0.39). To evaluate whether the secretion of inhibin B depends on the nature of damage to the Sertoli cells, inhibin levels in 23 patients with varicocele; eight after cryptorchidism, and 16 after hemiorchiectomy were compared to those of other patients without these diseases, but identical sperm count. No significant differences were found. In 20 men undergoing testicular biopsy, inhibin levels were compared to histology. Although the men with Sertoli-cell-only syndrome had significantly lower levels ((15.83 +/- 12.2) pg ml-1) than those with normal spermatogenesis ((183.8 +/- 112.3) pg ml-1), a distinction between patients with hypospermatogenesis from those with normal spermatogenesis was not possible on the basis of inhibin levels. Between these groups, the distinction was better achieved by the FSH levels (sensitivity of 85%). We conclude that inhibin B levels are a serum marker of Sertoli cell function, but the prediction of the quality of spermatogenesis is not superior to that of FSH levels.