White —Latin Writings of St. Patrick. 207 



be the autograph, in the year 807 ; and that it was then thought well 

 to copy the precious document into the official Armagh repertory, 

 without at all implying that other copies were redundant or interpo- 

 lated. The marginal notes incertus liher and Z (however explained) 

 certainly support the theory that Ferdomnach's exemplar was a very 

 old one. 



The truth is, that if these lacunae had occurred in any other ms. 

 than the Book of Armagh, it would never have occurred to anyone to 

 speak of the other mss. as interpolated. There is absolutely no 

 difference in style between the passages that are and that are not in A. 

 St. Patrick's style is abrupt ; but the Confessio, as read in the other 

 MSS., is much less violent in the transitions of thought than as read in 

 A. It must, however, be noted that the proofs drawn from the 

 manuscript itself in support of the view that the Confessio is 

 deliberately abridged in A cannot be pressed. Graves urged as bearing 

 on this the occurrence here and there of Z in the margin, and the 

 words in § 40 : et caetera^ reliqua usque dicit saeculi, reliqua sunt exempla. 

 But this Z never occurs at the beginning or end of a lacuna ; it is found 

 only where there is a textual or exegetical difficulty ; and the disincli- 

 nation to copy out at length familiar texts is not unusual in transcribers 

 of catenae.^ 



The fact that Ti'rechan actually quotes from a portion of the 

 Confessio which is not in A, seems a conclusive argument against 

 the interpolation theory : Extendit [^expendif] Fatricius etiam praetium 

 xu animarum hominum^ ut in scriptione sua adjirmat, de argento et auro, 

 ut nullum malorum hominum inpederet eos in uia recta transeuntes totam 

 Hiherniam (Bk. of Armagh, fol. 10, verso, b. 1. 34). See Conf. § 53 

 Censeo enim non minimum quampretiumquindecim hominum distrihui illis. 



It has been just stated that the marginal notes in A cannot be 

 urged as a proof that the text is an abridged one ; but at the same 

 time when we have set out our apparatus criiicus, we find that there 

 is a considerable number of places in which A omits words and 

 clauses which are found in all the other mss., and which in most cases 

 seem necessary to complete the sense. 



The following examples do not include the two omissions of 



^ Graves explained this Z as ^TjTerre : see Todd, St. Patrick, p. 348 ; but the 

 facts just stated are rather in favour of a conjecture which has been suggested to 

 me, that it stimds for Zahdtis, i.e. Diabolus, and that the scribe means to suggest 

 that he had been at work to cause whatever error or confusion occurred in the book. 

 Professor Bury calls it "the mark of query." Eng. Hist. Review, July, 1903. 



