MISSING GENERATIONS OF SPERMATOCYTES AND SPERMATIDS IN SEMINIFEROUS EPITHELIUM CONTRIBUTE TO LOW EFFICIENCY OF SPERMATOGENESIS IN HUMANS

Daily sperm production per gram parenchyma (DSP/g) in humans is only 25 or 35% of that for most species including rats and nonhuman primates. To explain the low efficiency of spermatogenesis in humans, the number of generations of germ cells (spermatocytes and spermatids) and the number of these germ cells in each generation were determined for each spermatogenic stage in men with varied efficiencies. Testes were obtained at autopsy, fixed by vascular perfusion with glutaraldehyde, further fixed in osmium, and embedded in Epon 812 before 0.5-mum sections were stained with toluidine blue. Tubular cross sections were photographed, and spermatogenic stages were determined by two observers. Testes were divided into three groups on the basis of DSP/g. The number of generations of spermatocytes and spermatids was greater (p < 0.05) in the high (2.01 +/- 0.05) and intermediate (1.77 +/- 0.04) than in the low (1.45 +/- 0.15) DSP/g group. All groups had a lower number of generations of spermatocytes and spermatids compared to the optimum value of three. The number of these generations per cross section was related (r = 0.85; p < 0.01) to DSP/g in these men. The number per cross section of spermatocytes, spermatids, and the combined number of germ cells was higher (p < 0.01) in the high than in the low DSP/g group. The combined number of germ cells per cross section was related (r = 0.85; p < 0.01) to DSP/g. The combined number of germ cells was higher in the high versus the low DSP/g group in stages I through V, but this difference was significant only in stages IV and V. These data reveal that missing generations of germ cells along the length of human seminiferous tubules (noted by those missing in tubular cross sections), coupled with a paucity in number of cells within each generation, contributes to the low efficiency of spermatogenesis compared to that of most other species.