In a recent article, Mitrani (1983) asked the question'Is upward basal cell movement independent of mitosis in the normal epidermis?', that is, whether migration is an active or passive process. Few would argue against mitosis and migration being linked. The question is how are they linked and which of them, mitosis or migration, comes first? Mitrani's paper presented computer simulations of the changes in the number of labelled cells and the total number of cells in the basal and spinous layers in unirradiatcd and irradiated epidermis. In spite of statements to the contrary, no in vivo data were presented. Instead the simulated curves were assessed in relation to data not shown, but published by Etoh, Taguchi and Tabachnick (1975). Since the paper quotes observations made by our group and contains some inconsistencies and, we believe, incorrect interpretations, we felt that some alternative considerations should be presented.

The paper states that cell kinetic data for mouse dorsal epidermis were taken from Etoh et al,(1975)-This noteworthy paper, in fact, presented little or no cell kinetic data and reported studies only on guinea-pigs and not mice. Mitrani then continued to compare simulated data for mice with published data for guinea-pigs. This may not be a problem provided the proliferative populations behave in a similar fashion. The theoretical curves would be dependent not only on the average cell kinetic parameters including the total cell cycle, which was not mentioned and is difficult to assess accurately in any case, but also on the proliferative model assumed. When the theoretical curves illustrated in the paper are in fact compared with all the available experimental data the picture is far from clear. Assuming that the basal layer contains about roo cells/mm, the total cell numbers predicted by active and passive models (based on the assumptions for mouse skin presented Mitrani) can be compared with actual data from pigs and guinea-pigs (Fig. i) and with data from mice (Fig. 2). Neither model fits the results for pig skin. The passive model appears to fit for guinea-pigs whereas the active model fits slightly better for the mouse data. Adoption of different assumptions will strongly influence the position and slope of the theoretical curves. This is particularly the case when the following are considered:(i) the number of cells/mm,(2) the cell cycle times,(3) the proliferative organization (the proportion of the proliferative cell population which function as stem cells),(4) whether or not there are maturing, non-dividing differentiated cells in the basal layer and (5) whether there is more than one kinetically distinct cell population, as well as the assumptions made concerning the way in which cells respond to radiation.