The determination of an empirical correction factor to deal with the problem of nucleolar splitting in neuronal counts

Abstract

Nucleolar counts are the method of choice for determining neuronal numbers. The main problem is the determination of an accurate correction factor for split nucleoli. The difficulties are that small nucleolar fragments are often unrecognizable and that nucleoli may be pushed or rolled rather than cleanly cut by the knife. A widely used method uses an estimate to account for the difficulties, and almost all methods depend on measurements of such things as section thickness and nucleolar diameters. We differ from previous procedures by identifying neurons first and then determining whether the nucleolus in each identified neuron is split or whole. If N is the true number of neurons, n the number of nucleoli counted to estimate N, T the number of nucleoli counted for the correction factor and S the number of nucleoli in T that are split, then N = [(T-S/2)/T] X n. The advantages are that the observations are easily done and that there are no estimates, only a determination of the numbers of whole and split nucleoli for a sample population of neurons.