of time fixed at ten years, and that he adopted this notion purely for organizational purposes that arose, for the most part, from the lack of complete death dates for the biographees and from the wide discrepancies in those dates—especially during the first three centuries.

My study has shown that this arrangement has no connection with adab al-ṭabaqāt (the biographical stratum genre); it is better classified under adab al-tanzīm ʿalā al-sinīn (the practice of organizing material by calendar years)(1).

From all that has been said, it is clear that al-Dhahabi used the term “stratum” to denote people who resemble one another, first, in actual encounter—that is, in the teachers from whom they received transmission—and, second, in being close in age with respect to birth and death in a way that does not contradict the possibility of meeting. This method allows both disparity in the death dates of those he records and fluctuation in the total number of strata.

How, then, are we to explain the great disparity in the division of strata by a single author such as al-Dhahabi? In Maʿrifat al-Qurrāʾ he sets out seventeen strata, whereas in al-Siyar he establishes nearly forty. In our view, the primary reason lies in the type of figures treated within a given book. A work like al-Tadhkira, which contains only the leading ḥuffāẓ, can be arranged in fewer strata than others because of the calibre of those included: all, or most, possess high isnād levels, so that the time span between one stratum and the next becomes wider and the total number of strata is reduced—without in any way violating the principle of actual encounter.

By contrast, al-Siyar embraces almost all the figures found in al-Tadhkira in addition to many of lesser standing, which obliges al-Dhahabi to increase the number of strata.

Naturally, it is not necessary that every single person in a given stratum actually met