attain his object? The simplest and briefest course was to re-write as follows the five lines just now submitted to the reader : TESABOMNIBUSDEFEN DEPERICULIS PER POST SANCTIFICETNOSQUACOM . PASTISUMUSMENSAECAE LESTISLIBATIOETACUNC and, singularly enough, this is the very thing which St Gregory—or, if not he, his amanuensis-seems to have done. The rubric in the Corpus MS., although reduced to a mere compendium in three letters—PCO.— is, nevertheless, cut into two, the first letter being on 27 v. (5), the second and third on 27 v. (6)1; but, as the scribe of the Corpus MS. was not a man of trumpery caprices, and as this is the only instance of the kind in the volume, I infer that there was a divided rubric in the exemplar. If this inference of mine be a right inference, it yields us a trustworthy proof of the genuineness of the two readings 'defende' for 'nos defende' and 'mensae caelestis libatio' for 'mensa caelestis,' as also a morally certain corroborative proof that the exemplar was a working copy; for there is no reason to believe that the bisection of a minor rubric would have been either attempted or allowed in a finished duplicate. This, I repeat, is the only instance in the whole of the MS. of a bisected word in a minor rubric. Again. At fol. 31 v., lin. 19, and in the Secreta for the Tuesday in Passion-week, the scribe of the Corpus MS. omitted the final 'per.' It, surely, cannot be a mere chance that in this very Secreta we encounter one of the most important variants proper to that revision of the verbal text of the Sacramentary which is so strikingly attested by the Corpus MS. The book, which has been the scene of St Gregory's own manipulation of the passage, was the very book to lack so matter-of-course an adjunct, and to lack it as a consequence of that manipulation. Can these coincidences be fortuitous? I abstain from describing over again the phenomena of the Whitsuntide masses as they stand displayed in the light of the theory of a working copy, a theory kindled by the rays which those phenomena themselves threw together into focus, and, turning the leaves of the 1 I regret to find that at p. 29 I have in the second foot-note written (5), (4) instead of (6), (5). 2 See above, p. xliv. book' in search of other such peculiarities as have already been encountered, pass on into the Proprium Sanctorum. At fol. 75, lin. 12 there are traces of an erased 'tibi' between the first and second words of the well-known Accepta sit in conspectu tuo.' I turn to the title of the mass in quest of a clue, and what do I find? I find that this is the composite mass in honour of SS. Fabian and Sebastian, the three constituents of which were selections from three pairs of rival candidates; and I note with more pleasure than surprise that, of the two competitors for the place of Secreta, the discarded prayer had for its second word 'tibi'-'Hostias tibi domine,' &c. Again I ask, Can this be chance? Here, as once in the ferial mass for the Wednesday in Whitsun-week, and as twice in the ember mass of the following Saturday, there were two prayers in the exemplar; and, in one of the fits of absence which were so peculiarly his own, the scribe all unthinkingly passed from one prayer to the other. I cannot persuade myself that any book but the derivative of a working copy could over and over again present us with phenomena such as these. Here, however, let me pause to remark, that, safe though it be for us, whose task is analytical, to record St Gregory's work in terms of letters, there is no reason to think that he would observe a method so minute and laborious. A practised eye readily informed him how many lines would accommodate a given prayer, with or without its rubric, with or without its conclusion; and what economy of penmanship would, in the case of this prayer or of that, be needed in order to set the rubric of the next prayer at or near to the end of a line. Again, then, I turn the leaves of the volume; and, confining my attention to Roman festa old enough to have been kept by St Gregory, find nothing to invite notice until I come to fol. 109 v., where in the outer margin is a memorandum suggesting the use of another Secreta than that given in the text, a memorandum, that is to say, analogous to the two on fol. 57. Will it be believed? The mass itself has been analogously treated with the last of the new set of summer ember masses which Gregory had destined for incorporation into Subredaction 1 But, although I abstain from saying over again what has so recently been said about St Gregory's reconstruction of the Whitsun-week masses, I must not therefore neglect to notice the conclusion of the Oratio of that for Tuesday. As at first written in our book, and thereforewe may fairly presume-as originally written in the exemplar, the conclusion was a mere 'per', no regard being paid to the necessary 'eiusdem' or 'in unitate eiusdem'. The little that I have to say about this will be found in a postscript to the present chapter. See below, p. clxxxii. 2 See above, pp. cxxi-cxxiii. D'. It is part and parcel of that batch of text which he re-combined, when, cancelling the erewhile mass of SS. Felicissimus and Agapitus and substituting one mass for two on the neighbouring feast of St Laurence1, he abstracted text of the value of a leaf from the middle of the Proprium Sanctorum. Again I say, This sort of thing cannot be fortuitous. Besides, we see the reason of it. The Secreta which now serves for the joint feast of SS. Sixtus, Felicissimus and Agapitus, is not precisely that of the old separate festum of St Sixtus. It is the same prayer, but the same prayer amplified by the ablative clause 'intercedentibus sanctis tuis' and lengthened-needlessly lengthened, except that the addition enhanced the augmentation to the extent of two whole lines-by the very curious extension 'dnm . nrm. ihm.' This double amplification rendered necessary a compensating deduction of text at some early moment; because, except for such deduction, the newly chosen pair of masses would now have been too long by a line. What the figures were before the change we already know. They were, in terms of letters, (352 +370 + 399 + 398) − (415 + 376) = 1519 − 791 = 728 = 39 × 183; and, in terms of lines, (19+20 +21 +21) − (22 + 20) = 81 − 42 = 39. What St Gregory wanted was, of course, a nett deduction of 40 lines. Clearly, therefore, something must be done, and what that something was the marginal note on fol. 109 v. informs us. ་ He on Accipe q. d. munera' for his St Laurence's Day substituted the Secreta This was (352 +370 + 399 + 398) − (415 +365) = 1519-780-739 = 40 x 1818; and, in terms of lines,— (19+20 +21 + 21) (22+ 19)=8141 = 40. If it be true that the book in which the reviser of the Corpus MS. found material for correcting two manifest stichometrical errors incurred in the prosecution of those changes in the foliation of Redaction D, of which the Corpus MS. is witness,―if, I say, it be true that the standard of revision of the Missal of St Augustine's, Canterbury, was a fair copy of Subredaction D'; the book where the errors stood on record, errors 1 See above, p. cxxiii. the adoption of which would have defeated the very end proposed by those changes in foliation, cannot with any show of probability be set down as anything else than St Gregory's working copy of D. There is, as we have repeatedly seen in the foregoing chapters, very much to encourage the deduction of this inference; there is nothing to set against it; and the minute and varied testimony yielded by the idiosyncrasies of the scribe who transferred the contents of the exemplar to the pages of the Corpus MS., is such as to lift the inference to the level of a conclusion morally certain. POSTSCRIPT. One or two miscellaneous items must here be added. They may perhaps serve as starting-points for future students. I. I have already noted1 the absence of a necessary 'eiusdem' from the conclusion of a prayer in a mass of Gregorian compilation. The only other instance occurs at fol. 28 v., lin. 5. The scribe of the Corpus MS. was so conscientious a workman that I am disposed to see in these exceptions to a rule otherwise observed most carefully a proof in favour of the theory of a working copy. In neither case is it the principal reviser who has made good the defect. In neither case, therefore, must we necessarily think that the defect was made good from a finished transcript. Each correction may, I venture to think, be very plausibly referred to a note introduced into the margin of the exemplar, but overlooked by the scribe. II. On the nineteenth Sunday of the post-pentecostal group, and not infrequently thenceforward, our transcriber omits the rubric of the first prayer of a mass. But, before the point just indicated, it is, with one solitary exception, only at the very foot of a page that he omits any minor rubric whatever. In other words; although, in the course of a hundred and twenty-five pages, he now and then at the extreme end of a page forgets a minor rubric, the fault occurs only once in any other 1 See above, p. clxxx. place than that. That one instance occurs, at fol. 54 v., lin. 9, in the Oratio of the Tuesday mass in Whitsun-week; and I find, to my surprise, that if my computation be correct, the Oratio of that mass must have begun on the last line of a verso page, and that its rubric— if written at all-must have been written at the end of such last line. This is as pretty a confirmation of the figures on page clxxiv as could be desired. The masses for Sunday and Monday represent an aggregate of (53 +40 + 336 + 274 =) 703 letters, or 38 lines, since 703 38 × 18; and I make no doubt that, early in the Tuesday mass, our artist, passing from the 'UIR-' at the foot of one page to the 'TUS' at the head of the next, overlooked the vermilioned 'OR' which adjoined the first moiety of the word. Shall I go a step further, and say that his attention was distracted by the multifarious alterations on which his eye must now have fallen? = III. I cannot yet account for the differences noted at page cxvii between ourselves and Pamelius on the Saturday after Ash-Wednesday, on the eighteenth post-pentecostal Sunday and on the Feast of SS. Cornelius and Cyprian. It may, however, be worthy of note that, as modified in our book, the mass for the Vigil of SS. Peter and Paul has a complement of 370 letters. Any future student who may undertake the herculean task of reconstructing the Proprium Sanctorum of the exemplar may, I think, take it for granted that this mass occupied, precisely, one side of a leaf. IV. I need scarcely remind the reader, though it has seemed needless to insist upon it in my concluding chapters, that the theory of a working copy is strongly recommended by the rubrical peculiarities near the close of the Proprium de Tempore. V. But it may be well to add that Archbishop Egbert's account of the 'plena hebdomada post pentecosten' yields an implicit proof of the authenticity of the post-pentecostal series, a proof impregnable in its conclusiveness. VI. The excerpts from the Antiphonary do not offer me material for argument. There is, however, a passage of the Micrologus (cap. XXXI) which it seems relevant to quote in connexion with the Gospel for the Second Sunday of Advent:-'In Dominica prima de Aduentu Domini quidam legunt Euangelium Erunt signa...Alii |