Key points are not available for this paper at this time.
SYNOPSIS A general problem encountered in quantitative histology is the error caused by the inclusion of fragmented cells in the observed count from histological sections. The result of including such fragments is an overestimate of the number of cells. With this problem in mind, a method is given for estimating cell numbers in systems where cells are mononuclear with single nucleoli. The method involves applying a correction to the difference in observed counts from adjacent histological sections of different thicknesses. The proposed method is based on the same formulation of this cell counting problem as is the method of Floderus, but a major distinction, however, is that the present method does not require the estimation of the mean particle diameter or vertical height of the smallest visible fragment and thus eliminates a potential source of error. The correction formula is derived by considering a simple model for this cell counting problem. A hypothetical cell population is studied to show the theoretical equivalence of the proposed method and that of Floderus and also to show the potential bias caused by using Abercrombie's formula which does not take into account the effect of invisible fragments. The new method is compared with other available techniques and is empirically evaluated by extensive counting in a human superior cervical sympathetic ganglion.
Ebbesson et al. (Wed,) studied this question.