The Complete Works of Archimedes, Vol. 2

he took care to have it copied. But the statement that Joannes Rhosos¹) was to copy the Archimedes codex did not come to pass. For F cannot have been written by him, as is easily apparent by comparing any codex²) written by him—of which there are very many—with our plate.
Therefore, F was copied from the very codex of Valla in the year 1491 or shortly thereafter, not very long before the codices BC. It remains, therefore, for us to inquire which of these three "sisters," so to speak, is most to be trusted.
Now, that codex B was copied by a learned scribe who corrected many things and also attempted to emend many things incorrectly, I showed with sufficiently firm documents in Quaest. Arch. pp. 128–130, and more instances occur on almost every page. I will add one. III p. 182, 25 in F there is οὑ³) for οὑτως; in B it is read, according to Lebègue, as οὑ, but in the margin by the first hand as οὑτως. Codex C, on the other hand, approaches F more closely, and would approach it even more if it had been diligently collated; for in almost all the places that Henri Lebègue inspected, the same reading that is found in F was also found in C, even though nothing had been noted in the Torellian collation, as anyone who wishes to compare the passages marked with an asterisk in my apparatus with that collation will understand.
But we can demonstrably show that codex F was copied with greater fidelity than the others. For there are certain places where, from Valla's own translation, we know that the same inept errors that F exhibits were in his codex, but which have been corrected in BC⁴):

1. For he is "Papa Ianni," as N. Anziani correctly cautioned me. Concerning Joannes Rhosos, priest of Crete, a tireless copyist, see Gardthausen p. 326 sq.
2. I compared in Florence with codex F itself the Laurentian codices XXXII, 6; LV, 9; LXXXI, 23; LXXXVI, 18.
3. Also in Valla, "cuius circulus."
4. I already indicated this document in Neue Jahrb. Suppl. XII pp. 384–385. Afterward, at my request, Henri Lebègue again collated all the passages where nothing regarding BC was noted in Torellius. What he found in addition to the passages cited above, I will note here, because they most manifestly argue for the incredible negligence of the Torellian collation. III p. 62, 17: τῆς ΖΗ] B, τῆ ΖΗ C (F). p. 68, 13: