206 



Proceedings of the Royal Irish Academy. 



du xii siecle. En realite il est ecrit de plusieurs mains, et j 'incline a 

 croire qu'il est plutot du xi siecle." 



We must now take into consideration the textual character of 

 these six mss. 



The unique associations that are recalled by the Book of Armagh, and 

 its relatively great antiquity — having been transcribed at least 250 years 

 earlier than any of the other mss. — have biassed the judgment of some 

 scholars when dealing with its peculiarities. The most disturbing of 

 these are its omissions. It does not contain the Epistola, yet the title 

 prefixed to the Confessio, Lihri Sancti Patricii, implies that it must 

 have been, originally at least, included in the contents of the exemplar 

 which Eerdomnach copied- But the absence of the Epistola is less 

 perplexing than the omissions — for such we must regard them — in the 

 Confessio. 



Assuming that the other mss. present the Confessio in its entirety, 

 there are five lacunse in A. The first begins with § 26 of the present 

 edition, and covers nearly 20 lines; then we have lines of A, 

 followed by a gap of 30 lines; then 5^ lines of A; then a gap of 15 

 lines. The next section of A is that in which the scribe betrays his 

 impatience of the string of texts that he saw before him ; so that in 

 calculating its length we are entitled to allow for the texts when 

 written in full. We reckon it then at 39 lines, or a little more. We 

 then have a lacuna of about 88 lines, followed by 11^ of A, and finally 

 a gap of about 37 lines before the concluding paragraph of A, which 

 covers 6 lines. It will be noticed that of the short sections of A two 

 occupy each 11^ lines, and two others 5^ and 6 respectively, and that 

 tlie length of the lacunse are, roughly speaking, multiples of 5 and 6. 

 I have estimated them respectively as 20, 30, 15, 88, 37. 



It would be unreasonable to suppose that each page of the exemplar 

 would contain exactly the same number of letters and letter spaces ; 

 and therefore the figures just quoted point to the hypothesis 

 that the exemplar from which the Confessio was copied into the 

 Book of Armagh was written on very small folios, possibly not in a 

 very neat hand, so that each folio did not contain more words than five 

 or six lines of the present edition, and that a considerable number of 

 leaves had been lost. The note at the end : Sue usque uolumen 

 quod Patricius manu conscripsit sua, cannot be pressed as proof that 

 Ferdomnach had before him the actual autograph, since it might have 

 been merely repeated from a copy of the autograph. But all the facts 

 can be explained by the supposition that we have, in the Book of 

 Aimagh, all that remained of the autograph, or what was thought to 



