The total nerve cell numbers in the right and in the left human entorhinal areas have been calculated by volume estimations with the Cavalieri principle and by cell density determinations with the optical disector. Thick gallocyanin-stained serial frozen sections through the parahippocampal gyrus of 22 human subjects (10 female, 12 male) ranging from 18 to 86 years were analysed. The laminar composition of gallocyanin (Nissl)-stained sections could easily be compared with Braak's (1972, 1980) pigmentoarchitectonic study, and Braak's nomenclature of the entorhinal laminas was adopted. Cell-sparse laminae dissecantes can more clearly be distinguished in Nissl than in aldehydefuchsin preparations. These cell-poor dissecantes, lamina dissecans externa (dis-ext), lamina dissecans 1 (dis-1) and lamina dissecans 2 (dis-2), were excluded from nerve cell number determinations. An exact delineation of the entorhinal area is indispensable for any kind of quantitative investigation. We have defined the entorhinal area by the presence of pre-alpha cell clusters and the deeper layers of lamina principalis externa (pre-beta and gamma) separated from lamina principalis interna (pri) by lamina dissecans 1 (dis-1). The human entorhinal area is quantitatively characterized by a left-sided (asymmetric) higher pre-alpha cell number and an age-related nerve cell loss in pre as well as pri layers. At variance with other CNS cortical and subcortical structures, the neuronal number of the entorhinal area appears to decrease continuously from the earliest stages analysed, although a secular trend has to be considered. The asymmetry in pre-alpha cell number is discussed in the context of higher human mental capabilities, especially language.