the GreeJc Text of the Mathematical Collections of Pappus. 63 
scriptis codicibus collati aut a viris doctis illustrati and there 
can be but little doubt of the reference’s belonging to Bernard’s 
copy of Commandine, which is now in the Bodleian, with a 
number of marginal remarks and extracts from the Greek text. 
Bernard, likewise, in his “ Veterum Mathematicorum Synop- 
sis,” cites only the two Savilian manuscripts of Pappus, and one 
belonging to D. Lescuyer, who was a French gentleman men- 
tioned by Mersennus ; Halley likewise, who was also contempo- 
rary with Wallis, makes no mention of any but the two in the 
Savile Library. But all this, however strong, is only negative 
evidence, and Wallis’s expressions are so very precise, that they 
can hardly be reconciled with the probability of any mistake. 
There may still be hopes that the book has been bound up with 
others, so as to have escaped notice. This is known to have 
happened in many instances, and if it has done so in the present 
case, the researches now going on in the Bodleian, for the for- 
mation of new catalogues, are the most likely means of bringing 
this treasure to light. 
There is in the Savilian Library the Greek of the third, and 
of the latter end of the second book, copied from No. 2368 of 
the French King’s Library. These are written very distinctly 
in a large foreign hand, upon 30 leaves of folio paper, about the 
size of foolscap, and the diagrams, drawn on a whiter and finer 
paper, are wafered on to the places to which they belong. What 
the history of this transcription may be, and why it was carried 
no farther, we have no clue for discovering. It is not inserted 
in any catalogue, and is found under the same cover of No. 9., 
with the transcript of the fragment of the 2d book, which Hr 
Wallis had prepared for the press. This seems suspicious af-. 
ter finishing his work, it will occur that he might by mistake 
have put all the MSS. together into the Savile Library, but 
still it is clear that this is not the Bodleian MS. to which he 
referred. There is no water-mark, from which the age of the 
paper can be collected ; but the colour of the ink, and the appear- 
ance of the whole, is too fresh for its having been as old as the 
time of Wallis ; besides (which is more conclusive);, he says that 
of the three manuscripts, fragmentum libri secundi non nisi 
in unico habetur,” and that one is undoubtedly the Savilian 
No. 9. It is clear, therefore, that the Bodleian MS. did not 
