origin to Egypt is far inferior, in point of being up to date, to that given in the new edition of Dr. Scrivener. For this Mr. Miller has been fortunate enough to secure the help of two such authorities as the Rev. A. C. Headlam and the Rev. G. Horner, who have corrected and made large additions to what was, when it was written, the best English account of the Egyptian version, viz. that written by the late Bishop of Durham for the last edition of Scrivener's Introduction.

The first of the versions with which Dr. Gregory deals is the Syriac, and in his short account Dr. Gregory refers to the many points of interest raised by the different forms in which this version is found; as, for instance (i.) the date at which a Syriac version was first made; (ii.) the earliest allusions to it; (iii.) the relation of the Peshitto and Curetonian, in regard to which he expresses agreement with Westcott and Hort, and differs from Dr. Scrivener; (iv.) the so-called Karkaphensian, which Mr. Gwilliam has shown to preserve nothing more or less than a Syrian Massorah; (v.) the Jerusalem Syriac; (vi.) the Philoxenian, and the later revision of it by Thomas of Harkel; (vii.) the question of the translation into Syriac of some of the Catholic Epistles and the Apocalypse. On this last point the researches of Professor Gwynn and J. H. Hall are utilised ; though we may perhaps note that Professor Gwynn has published further investigations on this point in the Transactions of the Royal Irish Academy. It is a point of some importance, because it affects the canon of the New Testament which the Peshitto recognised. We may notice here Dr. Gregory's statement that Tischendorf, in using Schaaf's edition, used one which 'minime sufficit in N. T. textu ex Syriaca versione emendando' (p. 820), which supports what has been already said as to the authorities on which Tischendorf relied for the evidence of the versions. This short general account of the version is followed by a list of MSS. of the different classes of the version-the Peshitto, Curetonian, &c.; and these are numbered for each of the great component parts of the New Testament as it first circulated, viz. the Gospels, Acts and Catholic Epistles, St. Paul, Apocalypse. We may notice that the Curetonian version no longer rests on one MS., though little is said for little has yet been given to the world-about the second MS., recently discovered on Mount Sinai. It has leaked out, however, that one or two of its peculiar readings will be found of a very startling character. The MSS. (eight in all) of the so-called Jerusalem Syriac are given, including the fragments recently acquired for the Bodleian Library.

The whole amount recovered of this version is smaller than we could wish, especially as it is somewhat interesting. When we pass to the Peshitto the number of MSS. is very largely increased, as Dr. Gregory notices no less than one hundred and twenty-two MSS. of the Gospels, fifty-eight of the Acts and Catholic Epistles, sixty-six of St. Paul's Epistles. The first MS. illustrates one of the difficulties which textual critics have to face, in that its text is for the Gospels Philoxenian or Harklean, while the rest of it gives the Peshitto version. Why, under these circumstances, we may ask, does it appear amongst the Peshitto MSS. of the Gospels? In regard to MS. 13, we wonder why Dr. Gregory says that it inserts John vii. 50 (sic)-viii. 12 from the Harklean version. Mr. Gwilliam 1 implies it is a Peshitto translation. It is certainly true that most MSS. of the Peshitto do omit the passage.

This list of MSS. could be increased, as will be seen iater; thus, at St. Petersburg, Bibl. Caes. Orient. 621 (Catalogue of 1852), a Syriac lectionary giving parts of St. Paul's Epistles, is omitted. But anyone looking through them will notice at once the number of early MSS. i.e. dating from the fifth and sixth centuries-in which the version is preserved. Mr. Gwilliam, however, in dealing with the materials for a critical edition in the article just referred to, does not think that many important divergences will be found from the ordinary printed text. It is, therefore, a fortunate accident, though not of any great critical importance, that so many early MSS. of the version survive.

The next section is devoted to the Philoxenian or, as it is called in its revised form, the Harklean, version. Of this, twenty-nine MSS. are enumerated for the Gospels, ten for the Acts and Catholic Epistles, five for St. Paul's Epistles, and six for the Apocalypse. It is chiefly in regard to three of the Catholic Epistles and the Apocalypse that this version is important. An edition of the Apocalypse is promised by Dr. Gwynn, who has done much to investigate the history of the Syriac translation of these disputed parts of the New Testament.

The Egyptian version in its various forms follows next in order, but what we have to say on this we will keep till we come to notice the section in Dr. Scrivener's Introduction dealing with the same subject. We may, however, remark here (i.) that it is not obvious why 33 a is not 19 a, seeing that it is described as a copy of the MS. numbered 19; (ii.) that Dr. Gregory has followed Dr. Lightfoot's list of MSS. given

1 Studia Bibl. iii. 54.

in Scrivener (edition three) in two mistakes as to press marks, for the MSS. numbered 6 and 7 of the Acts and Catholic Epistles are Parham 124 and 125, not 120 and 121 (this is corrected in the new edition of the Introduction); (iii.) that probably it is not the case that the Epistle to the Hebrews follows Philemon in MS. 7, as this statement has been cut out in the new edition of Scrivener's Introduction. For the Memphitic, or, as we are now to call it, the Bohairic version of the Gospels we have a list of sixty-four MSS., but one third of this number are very late MSS. of little or no critical value. For the Sahidic we have a list of ninety MSS., but many of these numbers represent very brief fragments. We may here express our regret that the edition of the Egyptian versions which M. Amélineau was prepared to undertake for the Clarendon Press has, so far, come to nothing. As Dr. Lightfoot said in the third edition of Dr. Scrivener's Introduction, such an edition is one of the most pressing needs of Biblical criticism.

The Ethiopic version follows next in order. As to the date of this authorities are divided. While Dillmann would assign it to the fourth or fifth century, and regards Abyssinia as its place of origin, Gildemeister, on what seem to Dr. Gregory insufficient grounds, assigns it to the sixth or seventh century, and supposes it to have been the handiwork of Syrian Monophysites. Here, again, the need of a critical edition is expressed and the insufficiency of those used by Tischendorf emphasised. Dr. Gregory enumerates IOI MSS. of the version, the oldest apparently being MS. 60, which represents the oldest form of the Ethiopic version. The account given in Scrivener's Introduction is from the pen of Professor Margoliouth, who does not attempt to furnish any list of MSS., though the brief review which he gives is a very clear summary of the most important points of interest.

The Armenian version is described briefly (pp. 912-15). This dates from the beginning of the fifth century, according to Dr. Gregory, while Mr. Conybeare, who writes the account in Scrivener's Introduction, says: 'It would appear (from the evidence of Moses of Khorene) that the Bible was translated twice into Armenian before the end of the fourth century, by Mesrop from Greek, and by Sahak from Syriac' (ii. p. 151). Zohrab, who edited the New Testament in Armenian at the end of the eighteenth century, affirmed that the Apocalypse was not translated till that century, while Mr. Conybeare

1 Paris, Bibl. Nat. Ætk, 22.

states that 'the book of the Revelations' (sic) was translated twice by Mesrop and his disciples.

Dr. Gregory enumerates sixty-four MSS. of the Armenian version, and refers to the existence of many more in the Armenian convent at Venice and the library at Etzschmiadzin. In regard to the date of the MSS. available Mr. Conybeare remarks: 1

'Old Codices of the Armenian Gospels are very common, and the present writer knows of as many as eight, none of them later than the year 1000; of four of these he has complete collations. The rest of the New Testament is only found in Codices of the whole Bible, which are rare and always written in minuscules, never in uncials, as are the Gospels. He knows of no copies of the whole Bible earlier than the twelfth century.'

Very

The version which follows is the Georgian little indeed is said about this, and the only MSS. to which reference is made are three in the Vatican Library. Neither Dr. Gregory nor Mr. Conybeare refers to what is probably the best account of Georgian literature available, viz. Professor Tsagareli's Information about the Monuments of Georgian Literature, written in Russian, and published at St. Petersburg in 1886. In this, and another book by the same author on the same subject, we are told of a manuscript of a Psalter on papyrus ascribed to the seventh or eighth century; also of a manuscript of the four Gospels and Psalter in the same library on Mount Sinai, referred to the eighth or ninth century. The earliest dated manuscript of the Gospels is ascribed to the year 936. Another manuscript of the tenth century is the so-called Themu Gospel. The earliest manuscript of the whole Bible in Georgian, though part of that is mutilated, is a manuscript in two volumes which is ascribed to the year 978, and is in the Iberian monastery on Mount Athos. The great centres of literary activity outside Georgia seem to have been the monasteries of the Holy Cross and St. Saba in Palestine, Mount Sinai, and Mount Athos. In his Catalogue of manuscripts in chronological order Professor Tsagareli mentions fourteen manuscripts of the whole or parts of the Gospels, eleven of the Apostol, and two of other parts of the New Testament, viz. one uncial MS. of the Ephesians, and a tenth-century manuscript of St. James and St. Peter, besides others of importance. Professor Tsagareli points out that the version was of gradual growth, and that the manuscripts show many traces of different recensions. To the other authorities quoted in Dr. Gregory

1 Scrivener, ii. F. 152.

may be added articles by Brosset in the Journal Asiatique, ii. and ix. We may also notice, as a correction of Dr. Gregory's account, that the Moscow edition of 1743 (1723 as he gives it by a slip) was not the earliest edition of the version, as the Gospels were printed at Tiflis in 1709.

To the Persian version Dr. Gregory devotes a brief introduction, but he quotes twenty-seven manuscripts. This version varies very much in critical value, not that any manuscript of it has very great worth. One class of MSŚ. preserves a version made from the Greek, another represents a translation from the Peshitto, while there is also a very late translation from the Latin made near the middle of the eighteenth century. Judging from the accounts given by both the authorities under review (the last edition of Scrivener merely reproduces the language of the earlier), we should say that it would be worth while for some Persian scholar to devote a little time to the study of these versions, for it does not appear that much of it has been edited in a critical way, although there is material both in the Old and New Testaments for such an edition.

The different classes of the Arabic version have been examined by Guidi in Atti della R. Accademia dei Lincei, and divided into five classes or recensions-(i.) those made from the Greek; (ii.) those corrected by the Peshitto; (iii.) those which show greater or less traces of Coptic influence; (iv.) manuscripts containing two recensions made in the thirteenth century; (v.) those in which elegance of style is the marked characteristic. We get an excellent, though naturally brief, account of the version in the new edition of Scrivener's Introduction, from the pen of Professor Margoliouth. In this we have reference made to the recensions already described, and to the various printed editions with the recensions contained in them. Professor Margoliouth also mentions a fragment of the ninth century contained in Professor Rendel Harris's Biblical Fragments from Sinai, in which the type of text 'agrees with none that have been published, and was probably older than any of them.' He also incidentally answers a question asked by Dr. Gregory as to the present whereabouts of the MSS. used by Erpenius for his edition, by giving the press marks of two of them in the University Library at Cambridge, to which Lagarde said they had been sold. We may here quote Professor Margoliouth's summary of the results of the study of this version:

'The repeated revision and correction which these translations have undergone, while they give evidence of the industry and zeal of