of the two Ṣaḥīḥs (1) or in Ṭabaqāt al-Shuyūkh (2). All of these works have come down to us except the last.

Our study of these books shows that al-Dhahabī did not apply a single scheme for determining the number of strata across them, nor did he seek proportionality in the number of entries from one stratum to the next within any given work. Likewise, he observed no fixed chronological span for a stratum in all of his writings—Tārīkh al-Islām being outside this arrangement, as will be explained later.

1. Number of strata: al-Dhahabī divided Tadhkirat al-Ḥuffāẓ into twenty-one strata (3); Maʿrifat al-Qurrāʾ into seventeen strata; and Siyar Aʿlām al-Nubalāʾ into roughly forty strata—even though the three books cover the same historical period, from the Companions down to his own time.

2. Number of biographical entries: Within a single work we also find a wide disparity in the number of individuals assigned to each stratum. In Tadhkirat al-Ḥuffāẓ, for example, the twenty-one strata contain, in order, the following totals: 23, 42, 30, 58, 78, 81, 106, 130, 106, 117, 77, 79, 74, 31, 46, 18, 25, 26, 12, 10, and 8. Thus, the count ranges from as few as eight persons to as many as one hundred and seventeen.

The same observation applies to Siyar: Stratum Thirty, for instance, includes 77 biographies, whereas the following stratum contains 130—and so forth.