*14. S. Synod. 291 (Mt. e) [xII] 4°, well written, from the monastery τοῦ ἐσφιγμένου on Athos.

*15. Typogr. Syn. 31 (Mt. tz) [dated 1116].

*16. (= Evst. 52).

*18. (Evst. 54).

*17. (Evst. 53).

* 19. (Evst. 55).

*20. (= Evst. 56).

Codd. 21-58 comprise Scholz's additions to the list, of which he describes none as collated entire or in the greater part. He seems, however, to have collated Cod. 12.

21. (Evst. 83).

22. Reg. 304 [x] fol., brought from Constantinople: mut. fine. 23. Reg. 306 [XII] fol., mut. initio et fine.

24. Reg. 308 [XIII] fol., contains a few lessons from the New Testament, more from the Old: mut.

25. Reg. 319, once Colbert's [xi] fol., ill-written, with a Latin version over some portions of the text.

26. Reg. 320 [x11] fol. mut.

27. parts.

Reg. 321, once Colbert's [XIII] fol., mut., and illegible in

28. (Evst. 26).

29. (Evst. 94).

30. Reg. 373 [XIII] 4°, mut. initio et fine: with some cottonpaper leaves at the end.

31. (Evst. 82).

32. (Evan. 324, Evst. 97).

33. Reg. 382, once Colbert's [XIII] 4o.

34. Reg. 383, once Colbert's [xv] 4o, chart. In readings it is much with Apost. 12, and the best copies.

35. (= Evst. 92).

36. (= Evst. 93).

37. (Evan. 368, Act. 150, Paul. 230, Apoc. 84).

38. Vat. 1528 [xv] 4o, chart., written by the monk Eucholius.

39. (Evst. 133).

40. Barberini 18 [x] 4o, a palimpsest (probably uncial, though not so stated by Scholz), correctly written, but mostly illegible. The later writing [XIV] contains lessons in the Old Testament, with a few from the Catholic Epistles at the end.

41. Barb.? [xi] 4o, mut.

42. Vallicell. C. 46 [xv1] 4o, chart., with other matter.

43.

Richard. 2742 at Florence: seems to be the same as Cod. 48 below, and is not (as Scholz states) Evst. 139.

44, 45. Hunterian Mus. Glasgow, having been bought by Hunter at Caesar de Missy's sale (Nos. 1633-4): 45 is dated A.D. 1199.

46. Ambros. 63 [XIV] 4°, bought (like Evst. 103) in 1606, “Corneliani in Iapygia."

56. (Act. 42, Paul. 48, Apoc. 13 and Evst. -) contains only 1 Cor. ix. 2-12. 57. (Apoc. 26, Wake 12, p. 182).

58. Wake 33, at Christ Church, Oxford [dated 1172] fol., 265 leaves, the ink quite gone in parts.

zser (see p. 221) contains four lessons from the Epistles; and de Muralt's Evst. 3p (p. 178) is also a Praxapostolos.

Additional copies are:

+*tisch. f. Bibl. Univers. Lipsiens. 6. F. (Tischend. v) [IX or x], containing Heb. i. 3—12, published Anecd. sacr. et prof. p. 73, &c.

+*Petrop., one leaf of a double palimpsest, now at St Petersburg, the oldest writing [IX] containing Act. xiii. 10; 2 Cor. xi. 21 -23, cited by Tischendorf (N. T. Prol. p. ccxxvi, 7th edition).

+ His new uncial Lectionary at St Petersburg (see p. 220) also contains lessons from all parts of the New Testament; Scholz seems to state the same of Evst. 161, "continet lect. et pericop.," and Coxe of Evst. Cairo 18.

At Lambeth, manuscripts 1190 [x], 1191 [x] 4o, mut. initio et fine, 1194, 1195, 1196, all [x111] 4°, mut. are Lectionaries of the Praxapostolos, which Dr S. T. Bloomfield has collated.

We find Latin versions in 8 uncial and 10 cursive codices; an Arabic version in Evan. 211; 450; Evst. 6; Coxe's Evst. at St Saba, No. 40; Latin and Arabic in Act. 96.

The total number of manuscripts we have recorded in the preceding catalogues are 34 uncial and 601 cursive of the Gospels; 10 uncial and 2281 cursive of the Acts and Catholic Epistles; 14 uncial and 282 cursive of St Paul; 4 uncial and 102 cursive of the Apocalypse; 58 uncial and 183 cursive Evangelistaria; and 7 uncial, 65 cursive Lectionaries of the Praxapostolos. In calculating this total of 127 uncials and 1461 cursives we have deducted 66 duplicates, and must bear in mind that a few of the codices, whose present locality is unknown, may have reappeared under other heads.

Ὁ μὲν θερισμός πολύς, οἱ δὲ ἐργάται ολίγοι.

1 In spite of the utmost care to detect duplicates, I overlooked at p. 193 what I had observed at p. 130, that Scholz's Act. 102, Paul 117 is Tischendorf's uncial K of all the Epistles. Hence it becomes necessary to make the requisite changes in the totals at pp. 200, 207.

CHAPTER III.

ON THE ANCIENT VERSIONS OF THE NEW TESTAMENT IN VARIOUS LANGUAGES.

1.

THE

HE facts stated in the preceding chapter have led us to believe that no extant manuscript of the Greek Testament yet discovered is older than the fourth century, and that those written as early as the sixth century are both few in number, and (with one notable exception) contain but portions, for the most part very small portions, of the sacred volume. When to these considerations we add the well-known circumstance that the most ancient codices vary widely and perpetually from the commonly received text and from each other, it becomes desirable for us to obtain, if possible, some evidence as to the character of those copies of the New Testament which were used by the primitive Christians in times anterior to the date of the most venerable now preserved. Such sources of information, though of a more indirect and precarious kind than manuscripts of the original can supply, are open to us in the versions of Holy Scripture, made at the remotest period in the history of the Church, for the use of believers whose native tongue was not Greek. Translations, certainly of the New and probably of the Old Testament, were executed not later than the second century in the Syriac and Latin Tongues, and, so far as their present state enables us to judge of the documents from which they were rendered, they represent to us a modification of the inspired text which existed within a century of the death of the Apostles. Even as the case stands, and although the testimony of versions is peculiarly liable to doubt and error, the Peshito Syriac and Old Latin translations of the Greek Testa

ment stand with a few of the most ancient manuscripts of the original in the very first rank as authorities and aids for the critical revision of the text.

In a class apart from and next below the Peshito Syriac and Old Latin we may group together the Curetonian Syriac, the Egyptian, the Latin Vulgate, the Gothic, the Armenian and Æthiopic versions, which we name in what seems to be their order in respect to value. Of these the Curetonian will be discussed more fitly hereafter (pp. 236-241); the Egyptian may have been formed, partly in the third, principally in the fourth century; the Latin Vulgate and the Gothic belong to the fourth, the Armenian and possibly the Ethiopic to the fifth. The Philoxenian Syriac, although not brought into its present condition before the beginning of the seventh century, would appear, for reasons that will be detailed hereafter, to hold a place in this class not much lower than the Latin Vulgate.

The third rank must be assigned to the several minor Syriac (so far as their character has been ascertained), to the Georgian and Slavonic, some Arabic, and one of the Persic versions: these are either too recent or uncertain in date, or their text too mixed and corrupt, to merit particular attention. The other Persic (and perhaps one Arabic) version being derived from the Peshito Syriac, and the Anglo-Saxon from the Latin Vulgate, can be applied only to the correction of their respective primary translations.

A

2. The weight and consideration due to versions of Scripture, considered as materials for critical use, depend but little on their merits as competent representations of the original. very wretched translation, such as the Philoxenian Syriac, may happen to have high critical value; while an excellent one, like our English Bible, shall possess just none at all. And, in general, the testimony of versions as witnesses to the state of the text is rendered much less considerable than that of manuscripts of the same date, by defects which, though they cleave to some of them far more than to others, are too inherent in their very nature to be absolutely eliminated from any. These defects are so obvious as to need no more than a bare statement, and render a various reading, supported by versions alone, of very slight consideration.

(1). It may be found as difficult to arrive at the primitive text of a version, as of the Greek original itself: the variations in its different copies are often quite as considerable, and suspicions of subsequent correction, whether from the Greek or from some other version, are as plausible to raise and as hard to refute. This is preeminently the case in regard to the Latin version, especially in its older form; but the Peshito Syriac, the Armenian, the Georgian and almost every other have been brought into discredit, on grounds. more or less reasonable, by those whose purpose it has served to disparage their importance.

(2). Although several of the ancient versions, and particularly the Latin, are rendered more closely to the original than would be thought necessary or indeed tolerable in modern times, yet it is often by no means easy to ascertain the precise Greek words which the translator had in his copy. While versions are always of weight in determining the authenticity of sentences or clauses inserted or omitted by Greek manuscripts1, and in some instances may be employed even for arranging the order of words, yet every language differs so widely in spirit from every other, and the genius of one version is so much at variance with that of others, that too great caution cannot be used in applying this kind of testimony to the criticism of the Greek. The Aramæan idiom, for example, delights in a graceful redundancy of pronouns, which sometimes affects the style of the Greek Testament itself (e.g. Matth. viii. 1; 5): so that the Syriac should have no influence in deciding a point of this kind, as the translator would naturally follow the usage of his own language, rather than regard the precise wording of his original.

(3). Hence it follows that no one can form a trustworthy judgment respecting the evidence afforded by any version, who is not master of the language in which it is written. A past generation of critics contented themselves with using Latin versions of the Egyptian, Ethiopic, &c., to their own and their readers' cost. The insertion or absence of whole

1 This use of versions was seen by Jerome (Praefatio ad Damasum) “Cum multarum gentium linguis scriptura ante translata, doceat falsa esse quae addita sunt." It is even now the principal service they can perform for the critic.