manuscript, indicated a distinction in the origin of the translation of that Evangelist.' Bernstein said that it referred to the division of the text into paragraphs for the purpose of public reading in the Church Services.2 Gildemeister, assuming, with Cureton, that the adjective belonged to St. Matthew, took it as a title of the writer, Matthew the chosen.' The word has also been translated 'explained,' and it has been pointed out that the text of the Curetonian shows traces of attempts to explain the language of the Peshitto. Wright, although he translates Evangelion da-Mèpharrěshë (?) 'the Separate Gospels," yet in the case of the Psalter says, 'the word seems here really to mean "of the interpreters" or "translators.' We have seen that the Sinaitic, no less than the Curetonian, abounds in readings which are of the nature of exegeses, or are amplifications of the Peshitto. In view of the ambiguity which attends the use of the term, the inquirer will hesitate to assume that the L (Mepharreshe) Gospels, as we now have them, are so called, as having been 'separated' from a Harmony. The term, of course, denotes as much, in the order of Rabbula, to which we have already referred, but there is no proof that the copies he prescribed were Curetonian-Sinaitic. When we consider the familiarity with Diatessaron language which the wide circulation of that work would produce, and the prevalence of corruptions in the Greek copies of the time, we recognise causes sufficient to account for the appearance in certain localities of texts such as are now preserved in the manuscripts discovered by Cureton and by Mrs. Lewis.

5

To conclude. The Syriac New Testament has come down to our times in two forms. (i) There are manuscripts of a text which was revised and adapted to Greek exemplars by Thomas of Harkel, in A.D. 616. This was based on the older revision by Philoxenus in 508, the manuscripts of which have been almost superseded by those of the Harkleian revision. Early in the fourth century Rab

1 Preface, p. vi.

2 For this and the next two see Journal of Sacred Literature, viii. (1858), pp. 140f, 216f; x. 154, 377.

3 Syriac Literature, quoted in Four Gospels in Syriac, Introduction, Pp. xxiii-iv.

Catalogue Syr. MSS. in the British Museum, i. 116 n.

So Mr. Burkitt (Guardian, p. 707), describes the 'Separate Gospels' as 'written by scribes who very probably were accustomed to hear a Gospel Harmony read in the services of the Church.' Only Mr. Burkitt and his colleagues do not recognize the existence of the Peshitto at that date.

bula,' the friend of Cyril of Alexandria, occupied himself in some critical labours on the text of the New Testament. What precisely he did is not known. Probably he revised the Syriac in accordance with Alexandrian MSS. and was thus the precursor of Thomas and Philoxenus. Their labours resulted in the Harkleian-Philoxenian or Syriac Vulgate. The records of these revisions also suggest that when others are not recorded they were not made. Rabbula and his successors imitated the work of Jerome. The theory of an earlier recension requires us to suppose that the Syrians, who showed little originality in literature, set Jerome the example. This Syriac Vulgate is one form of text.

(ii) There are also extant a large number of manuscripts 2 of the other form of text, the Peshitto version, which preceded the revised Harkleian-Philoxenian text. This is the Old Syriac, and has never been superseded amongst the Eastern Syrians, nor wholly supplanted amongst the Westerns. Its text can be traced back to the days of Mar Ephraim, and there is absolutely no hint, much less record, in Syrian literary history, that it was revised in the third or fourth century; nor is there any proof that it was derived from the extra-Peshitto text which has been found in two fifth-century manuscripts.

It is admitted by all that a Syriac Version of the New Testament has existed from (perhaps) the second century. The place of this version has been taken by the Peshitto from the earliest times. Its text stretches back into the farthest regions of Syriac literature. It is a witness to the best form of the Greek text of the New Testament, that text which has been preserved in all parts of the Christian Church, and is more attested by the earliest Greek Fathers than any other. On the other hand, the Curetonian-Sinaitic text is a witness to the corrupt form of text which prevailed in the West and in Syria, amongst those to whom Greek was not a familiar language. It may be an ancient witness. The corruption of the Greek Testament dates back to very early days, and almost from the first the pure line was accompanied by lines of depraved tradition. Manuscripts like the Curetonian fragments and the Sinaitic palimpsest are valuable for

1 Wright, op. cit. p. 11.

2 The number of ancient codices is so great that, in this respect, the Peshitto rivals even the Greek Testament. See Scrivener's Introduction, 4th ed. vol. ii. pp. 12, 13.

3 As appears from the statements of Caius of Rome, only about a century after the close of the N.T. Canon, quoted in Revision Revised, pp. 323-4. See also Scrivener, vol. ii. chap. ix. §§ 2-5.

the help they supply in tracing the causes of textual corruptions, and the history of the transmission of the text through successive ages. But we must not look to them in their connexion with the text of codex B' (or to any other limited class of documents) for the true text of the New Testament. That text is to be found in the contents of the vast number of codices which are the representatives of various streams of tradition, and is supported by collateral evidence of Versions and Fathers, of which some of the best and clearest is that which is afforded by the testimony of the ancient Syriac Peshitto Version.

ART. VI. THE TROPER AND THE GRADUAL.

1. The Winchester Troper, from Manuscripts of the Tenth and Eleventh Centuries, with other Documents. Edited by WALTER HOWARD FRERE, M.A., of the Community of the Resurrection, Radley. (London: for the 'Henry Bradshaw Liturgical Text Society,' 1894.)

2. Graduale Sarisburiense. Reproduced in facsimile from a manuscript of the thirteenth century, for the Plainsong and Medieval Music Society, by W. H. FRERE. Two Parts, folio. (London, 1892-4.)

3. Graduale ad veram et integram preclare ecclesie Sarum consuetudinem. Folio. (Paris: N. Prevost, 1527-32.) 4. Missale Sarum. Paris, Regnault, 4to, 1527. The same, Burntisland, ed. Dickinson, 1861-83.

5. Music Loan Exhibition, 1885. Catalogue of Manuscripts, &c. By W. J. H. WEALE. (London, 1886.)

6. Bibliotheca Musico-Liturgica. List of Manuscripts drawn up by W. H. FRERE. For the Plainsong Society. Fascic. i. (London, 1894.)

THOSE who had the good fortune to visit the Historical Music Loan Exhibition in the gallery at the Albert Hall in the summer or autumn of 1885 may remember, among the greatest treasures there exhibited, seven manuscripts from the Library of St. Gall, the home of Notker Balbulus, the father of the German school of Sequences. These were one Gradual of the ninth and two of the eleventh century, an early tenth century Troper, a tenth century Troper, Gradual, and Sequences in one, with the letter of Notker to the Bishop of

1 On the real character of this ancient, but often solitary witness, see Miller's Textual Guide, pp. 54-9.

Vercelli, and an Antiphonarium written at the close of the thirteenth century, with an office and sequence for St. Francis added in 1323, besides a sixteenth century choir-book of Masses with sequences, written at a time when four-part singing had been introduced.

The earliest St. Gall Gradual, set in such a striking

manner

'twixt board and board of oaken tree'

each an inch and a half thick, and one of them containing a pair of ivory plaques, at least four centuries old when Notker wrote, and three hundred years older than the ancient music book which they now contain, will not easily be forgotten. And the Tropers, with their lines of music in neumes which an uninformed spectator might mistake for old shorthand or for a representation of some strange telegraphic code, have impressed themselves upon our memories.'

So many of our readers as chanced to see certain of these books from St. Gall, or who may remember a Troper which is, or was, displayed in a show case in the great library at Paris, will have at least some general notion of what a Troper looks like, even if they feel that their ideas of its contents are somewhat indefinite, and they will not make the mistake of confounding it with the Greek hymns and hymnals called troparia, with which it has no connexion beyond a common derivation from the Greek word for ' turning,' although the Western Tropers contained some curious specimens of Greek and had something to do with singing.2

It is only natural, and indeed inevitable, that some perplexity about these volumes should prevail. Even when Sarum Use was almost unknown, or hardly formed, Tropers had ceased to have a living existence. Specimens were to

1 Specimens of neumes and Latin musical notation will be found attached to the late Recorder of Sarum's (J. D. Chambers) article in Julian's Dictionary of Hymnology, pp. 654-5. The St. Gall MS. 359 (cir. 900) has been edited (facsimile) at Paris (Lambillotte, 1851). And MS. 339 (late tenth century) in Paléographie Musicale, i., Solesmes, 1889.

According to Arcadius, the ippos takes its name from drawing the troparia, or short hymns, after it. The тporápia are so called because they turn towards their model hymn, the hirmos. See Neale, Eastern Ch. pp. 830-33, nn. According to Sicardus of Cremona (cir. 1200) the Western Tropalis vel Troparius est liber in quo tropi, id est hymni [et Kyrie eleison] cum prosis et sequentiæ continentur ; et dicitur a трóños, quod est conuersio; quia conuertitur ad introitum,' Mitrale, lib. 5, prolog.

The name 'troparium' continued for some time in use, but the book to which it was applied was merely a collection of Sequences. (F. E. Warren in Julian's Dict. Hymnol. p. 1186 a.)

be found in the Churches here and there, but they were the books cast off by a generation that had passed away. The tropes were discredited by those in authority (and who shall altogether blame those powers that were?), and by the thirteenth century those elements of the old Tropers which had life remaining in them, had transfused themselves into more modern service-books. So thorough had been the clearance that it is only astonishing to us that so many as sixty Tropers are reported to have been seen in Christendom within the observation of modern students of music or liturgiology.

Those which Mr. Frere has introduced to our notice in the volume which he has recently edited for the Henry Bradshaw Society,' are as follows: two Winchester Tropers, the earliest of them being contemporary with St. Dunstan and St. Ethelwold; the Canterbury Troper in the Cottonian collection; and the Dublin Troper (now at Cambridge) in the same volume with Dr. Todd's famous manuscript of the Sarum Custom Book, besides twelfth or thirteenth century Graduals of Worcester and St. Albans containing tropes, and some Continental specimens introduced to give the reader some notion of the French and German schools of music books, and particularly the former, with which our Tropers had the closer affinity.

As the book has been edited for a society whose business is with liturgical texts, the Winchester Troper' and the ' other documents' appear shorn of their musical accompaniment, if we may for a moment adopt this phrase. It is historically most inappropriate, since, as will presently be seen, the text was in this case the accompaniment to the music. However, by dint of producing admirable collotypes and a lucid Introduction, Mr. Frere has provided a volume which has its attractions for those who are interested in the history and archæology of music as well as for the student of servicebooks, whose principal care is for the rites or words recited. The collotype plates consist of thirty-one leaves and three smaller pieces attached, all of them containing musical notation as well as words. Three leaves (slightly reduced) containing Sequences are derived from E, the (Ethelred) Winchester Troper, which, as we understand, was written in the year 979.' Twenty-three plates (the size of their originals) represent various sections of the other (eleventh century) Winchester

1 Ethelred succeeded King Edward the Martyr († March 18) 979. Dedication of Winchester changed by addition of St. Swithun's name, October 20, 980.