The Complete Works of Archimedes, Vol. 2

Henceforth it is probable that the Florentine codex and the codices BC are related, and this is argued by other most certain documents. First, we saw that the codex of Valla was mutilated at the beginning, and for that reason codex C lacks the greater part of the letter prefixed to Book I of On the Sphere and Cylinder; also in the Florentine codex the first page was written by a different hand and with many lacunae, and in B also the same place is similarly corrupt (cf. vol. I, p. 3, note); the scribe added: "The preface of Archimedes' On the Sphere and Cylinder is missing; for the first page of the copy was illegible, as you can see." Furthermore, in vol. III, p. 4, 18, in all codices there is a lacuna between σύγκειται and τῇ ΑΒΓΔ; the scribe of codex B added: "One whole page or even two are missing." Finally, the numbers of the propositions of the book on conoids are confused and corrupt in exactly the same way in all codices (cf. Quaest. Arch. p. 123 sq.). Therefore, since the Florentine codex cannot have been copied from the Parisian ones, it remains that it is either that very codex of Valla or derived from it. I shall now explain why I believe this to be true.

From the note of Georges d'Armagnac quoted above from codex C, it appears that the first and last pages of the codex of Valla were indeed inscribed with letters, but could not be read due to age and obscurity. But in the Florentine codex, the first page was initially completely blank, and only later was the omitted part of the letter added by another hand¹⁾, nor is there any trace of earlier writing later deleted, nor was anything ever written in the lacunae. And the extreme part of the Heronian fragments is written neatly and clearly, like the rest of the codex, and nothing was written after it. I said above that the last fragment in the Florentine codex is slightly

1. Nicolaus Anziani, the Laurentian librarian, thinks the first page was written by the same hand as the rest of the codex. To me, however, the handwriting seems different, although it does not differ much, and H. Menge and Carolus Graux agree with me, the latter of whom assigns this part to the sixteenth century. Certainly, this is of interest, that in this part almost all accents and breathings are present.