Theory and practice of stereological techniques applied to the estimation of cell number and nuclear volume in the testis.

The historical background to contemporary approaches to the estimation of cell/nuclear number and volume in the testes is reviewed. The limitations of older geometric model-based approaches to the estimation of cell/nuclear number are discussed, and the need for absolute estimates of cell number rather than ratio estimates is examined. The physical and optical disector approaches to the direct estimation of numerical density and, hence, absolute cell number are presented together with data illustrating their operational efficiency in the testis. New approaches to the direct estimation of nuclear/cell volume, using the point-sampled intercept family of methods, are presented and illustrated, using the example of the Sertoli cell nucleus. The use of both classical transverse and the newer vertical section approaches is explored. Estimation of Sertoli cell/nuclear volume in the volume (point-sampled intercept procedure) and number (nucleator and rotator methods) distributions on both conventional transverse and vertical sections is discussed. The use of transverse sections of the testis is shown to produce a consistent bias in the estimation of Sertoli cell nuclear volume in 120-day-old animals, with all the estimators. Comparison of the Sertoli cell nuclear volume (measured on vertical sections) in the volume and number-weighted distribution suggests a coefficient of variation of volume in the number distribution of 0.4-0.5, suggesting either a random or stage-dependent variation in Sertoli cell nuclear size which requires further exploration.