258 Proceedings of the Royal Iriah Amdemij. 



■which the library possihly strffered much from neglect, and from its 

 many journeys from place to place. Bnt this is not all. The C marks 

 were usually insciihed in two places — on one of the fly-leaves, and on 

 the front of the volume, across the edges of the pages. The B marks 

 on the other hand, were entered, in most cases, on the fly-leaves only. 

 And experience proves that it was just in that place that they were 

 least likely to survive the ordeals of mis-use and re-binding. Thus 

 the objection which has been stated is satisfactorily disposed of. "We 

 may rest assured that A, B, and C refer to the same library. This 

 library su:ffered losses, and it received accessions ; it was more than 

 once re-arranged, but the bulk of the books remained thi'ough all its 

 vicissitudes. 



This being assumed, we can now draw a further inference. A 

 volume in the present library may have lost its A press-mark, and 

 yet we may be confident that it was actually in the library which 

 A represents. Thus, for instance, IN'o. 4 in the table retains the 

 marks of B and C. We conclude at once that it was in A, though 

 the A press-mark is gone. The only other possible supposition is, 

 that when A was written the library had a copy of the book, that this 

 copy was subsequently lost, and that another copy was procui'ed to 

 take its place in the interval between the compilation of A and B. 

 That this happened, in a large number of instances, is highly im- 

 probable. It is considerations such as these which justify us in 

 including I^os. 4, 5, 7, 8, 9, 11, 12, 14, 15, 16, 17, 19, 21, 22, 23, 

 24, 25, 27, 29, 33, 34, 37, 40, 42, 43, 44, 45, 46, 48, 49, 51, 53, 55, 

 56, 61, 64, 66, 67, 69, 70, 72, 74, 75, 76, in our table. 



But again. Most of the volumes, which we have succeeded in 

 identifying, are now kept in the cases marked BB, CC, DD. Xow 

 many other volumes in these presses answer to the descriptions of 

 volumes given in A, as may be seen from a glance at the foot-notes 

 on pp. 218-239. The presumption clearly is, that the bulk of these 

 volumes also belonged to the library catalogued in A, B, C, though 

 we are precluded from dogmatising in any individual case. 



But we must now state the second conclusion which oiu' table 

 seems clearly to warrant. It is this : that the library, of which D is 

 a catalogue, was a collection completely distract fi'om the A, B, C 

 library. In no more than fourteen instances does D record a volume 

 with the same title as one of those in oui' first column. In no single 

 case in which a volume in col. 4 is proved to be identical with one in 

 col. 5 is that volume shown to be identical with one mentioned in 



