The microscopic method of age at death determination was introduced by Kerley in 1965 [E.R. Kerley, The microscopic determination of age in human bone, Am. J. Phys. Anthropol, 23 (1965) 149-163.]. However, even though the method has been revised several times, there remain some fundamental issues concerning the reliability of the methods. This is because several basic histological features seen in a cross section of a bone have to be quantified according to the method, but the definition, and hence quantification, of these features leaves room for subjectivity. In a previous study we found that some of these features (osteon fragments and Haversian canals) could not be identified reliably. Only secondary osteons could be identified with a low inter and intra observer error. Furthermore, since the histological features are quantified in only parts of an entire bone cross section, the selection of these parts or areas is a potential source of bias. Finally, unless an unbiased method is used for addressing features on the borders of the selected areas, this will also introduce error. These issues have not been addressed specifically in previous studies. In this study, we used the methods of stereology to choose the regions of interest, as well as for dealing with border phenomena, and we only counted secondary osteons. Our results show a statistically significant increase in the median number of osteons per area unit with increasing age at death. However, this was after exclusion of one outlier. This result is probably due to the limited sample size (N = 24). As such, this study is preliminary, but does warrant applying the described techniques to a larger sample.