(Chapter 40) Footnote 11:

Vesalius does not, in fact, count correctly in the 1543 edition: his itemization adds up to 303. In the 1555 version, he adds four popliteal sesamoids and comes up with 307. Vesalius’ count is higher than the modern reckoning of 206 because of his method of counting and anatomical error, e.g., he counts the coccyx as four and describes the sacrum as six bones. With the exception of the two patella sesamoid bones, no other sesamoid ossicles are counted in modern anatomy. Whatever the count in the Fabrica, it is substantially less than the canonical medieval number of 248 (Avicenna, Albucasis, Mishnah, Benedetti) which originated in the 2nd century AD or earlier, and less than the 246 proposed in the 1536 Tabulae Sex, where the fourth Table contained the distich Adde quater denis bis centum senaque, habebis Quam sis multiplici conditus osse, semel. See Singer and Rabin 1946, pp. 28f. and note 190.