290 Proceedings of the Eoi/al Irish Academy. 



It is not indeed easy to classify the tituli of A. On the whole, they 

 resemble those of Jerome's Eomau and G-allican Psalters. They are in 

 many cases identical, or nearly identical, with one or both of these against 

 the Old Latin version of Sabatier ; but, so far as I have oljserved, they never 

 take the side of the Old Latin against them. But their resemblance to the 

 G-allican and Eoman series is not very close. An examination of 73 successive 

 psalms (xxxiii-cv) gives the following results : —A agrees with both Eoman 

 and Galilean 14 times; it agrees with the Galilean against the Eoman 10 

 times, and with the Eoman against the Galilean 14 times (some of the agree- 

 ments not amounting to identity). But, on the other hand, it stands alone 

 35 times, 15 times where the Eoman and Galilean are identical, and 20 where 

 they differ. The general character of its tituli, nevertheless, is Septuagintal 

 and Hieronymian. In other words, we should not be surprised to find such 

 titnU as these either in a Galilean or a Eoman Psalter ; but they would be 

 out of place in the Old Latin. They are utterly incongruous in a Psalter 

 according to the Hebrew verity. It was, perhaps, a sense of their unsuitability 

 to such a position that led the scribe of the Bamberg Psalter to discard them, 

 with the appended headings and liturgical notes, in favour of the rubrics proper 

 to the version from the Hebrew. We cannot but suspect, therefore, that they 

 were not in the copy of the Psalter brought from Italy, which was the main 

 source of the text of this part of the Amiatine Bible. 



And now we turn to the Cathach Psalter (G). Here the titiili, like the 

 headings, are very similar to those of the Codex Amiatiuus : a similarity all the 

 more remarkable because both series are of an abenant type. The differences 

 between them are obviously due to the blundering of scribes on the one side 

 or the other.' Thus from tlie tituli we have fresh proof of the close connexion 

 between the two manuscripts, in spite of the fact that their texts are entirely 

 unlike. The tituli, with the headings and notes, must somehow have found 

 theu" way from one of the two texts to the other. But in C they are at home, 

 for, as we have seen, it is fundamentally Galilean. These rubrics may have 

 stood for many generations in the line of manuscripts from which it sprang. 



' A good many of the tituli of C are lost or illegible ; but some 63 are available for 

 comparison. Of these, 10 differ from A, but have the support of R : proi3iimably thuy 

 agree with a, the common parent of A and R. Of the remaining diflf'erences, all but the 

 following are too trivial to call for mention. In the rubrics of Pss. xxxiii {cum inmotauit 

 for commutduit), Ixxviii (where o repeats the titulus of Ivi), Ixxxvii (where a puts the first 

 clause of the titulus after the heading), C gives the correct text. C omits the titnlus of 

 Ps. xlvi. It omits also the first clause of the titnlus of Ps. xcv, the word dauid in that 

 of Ps. Ixxxv, the words in tituli and in (tert.) in that of Ps. lix, where also it has et syriam 

 sabba for syriae sabaa ; it inserts the word psalmus in the tihdus of Ps. Ixxxiii, and 

 substitutes /rtm«J/' for Iwminis in that of Ps. Ixxxix. In one or two of these readings C is 

 possibly superior to a. 



