xiii. 16, can prove nothing, as they are common to the Curetonian

with the Peshitto, from which version they may very well have been derived.

16 The title to St. Matthew is remarkable; for while (in the subscription) we read, “Gospel of Markos,” and “Gospel of Juchanan” occurs, as in other Syriac MSS., to St. Matthew is prefixed the title “Evangeliom dampharsa Mattai.” The meaning of the second word is doubtful in this application. The root means _divide_, _distinguish_, _separate_—cf. Daniel v. 28. Cureton (Pref. vi) says (1) that the great authority Bernstein suggested “Evangelium per anni circulum dispositum.” This is inapplicable, because the copy is not set out in Church Lessons, although some are noted by a much later hand in the margins. (2) Cureton himself, noticing a defect in the vellum before ܡܬܝ (or ܝܬܡ), would read ܕܡܬܝ (or ܝܬܡܕ), and render “The distinct Gospel of Matthew.” This he understood to indicate that the translation of Matthew had a different origin from the other books, and was “built upon the original Aramaic text, which was the work of the Apostle himself.” But there is nothing to justify the insertion of a ܕ, which is required to connect the title with the following name. The title belongs to the whole work, “Evangeliom dampharsa—Mattai” [Catalogue Brit. Mus. _l. c._]; the other names being preceded by “Evangeliom” only. (3) “Dampharsa” has been rendered “explained” [see the review in “Journal of Sacred Literature,” 1858], viz. from the text of the Peshitto; and this, as we shall see presently, agrees with the character of the Curetonian, for it abounds in deliberate alterations. But (4) from the quotations and references in the “Thesaurus Syriacus” (R. Payne Smith), col. 3304, it seems almost certain that the epithet means “separated,” as opposed to “united in a Harmony.” Such, of course, the Codex Curetonianus is, but further evidence is required to justify the inference that the Curetonian was the offspring of Tatian’s Harmony, and became the parent of the Peshitto, an opinion in large measure contradicted by the character of the translation.

17 “Si nous devons en croire Scrivener, la version syriaque dite _Peshitto_ s’accorde bien plus avec lui [Cod. A] qu’avec (B).” (Les Livres Saints, &c., Pau et Vevey, 1872, Préface, p. iii.) The fact is notoriously true, and of course rests not on Scrivener’s evidence, but on universal consent.

18 The student may also consult:—Evangelienfragmente, F. Baethgen, 1885. Disputatio de cod. Evangg. Syr. Curetoniano, Hermansen, 1859. Lehir’s Etude, Paris, 1859. Dr. Harman in Journal of the Society of Biblical Literature, Boston, 1885. Zeitschrift des Morgenländische Gesellschaft, 1859, p. 472. Dr. Wildeboer in De Waarde der Syrische Evangeliën (Leiden, 1880) gives three pages of the literature of the question.

19 Cureton, Preface, pp. xi, xciii.

20 Brit. Mus. Add. 12,138—_see_ p. 36.

21 So Roediger in Z.M.D.G., b. 16, p. 550, instances ܐܚܢܢ (or ܢܢܚܐ); but it proves nothing, for the form occurs also in old Peshitto MSS.

22 Pages 164-5.

23 Pages 171-2.

24 Some of the Homilies of Aphraates were composed between 337 and 345. Ephraem died A.D. 373. Bickell, Conspectus, p. 18.

25 Page 14.

26 In the following paragraphs we quote from a MS. exhibiting the results of investigations made by the Rev. Dr. Waller, Principal of St. John’s Hall, Highbury, who has most generously permitted us to make use of his labours.

27 For other like cases see Mat. iv. 11, 21; v. 12, 47, in the Curetonian.

28 The forms in which O. T. quotations appear in the Curetonian demand attention, as they seem to suggest similar inferences.

29 E.g. in the transposition of the Beatitudes in St. Matt. v. 4, 5.

30 Since the discovery of the Curetonian version in Syriac by Archdeacon Tattam in 1842 and Canon Cureton, some Textualists have maintained that it was older than the Peshitto on these main grounds:—

1. Internal evidence proves that the Peshitto cannot have been the original text.

2. The Curetonian is just such a text as may have been so, and would have demanded revision.

3. The parallels of the Latin texts which were revised in the Vulgate suggests an authoritative revision between A.D. 250 and 350.

These arguments depend upon a supposed historical parallel, and internal evidence.

The parallel upon examination turns out to be illusory:—

1. There was a definite recorded revision of the Latin Texts, but none of the Syrian. If there had been, it must have left a trace in history.

2. There was an “infinita varietas” (August. De Doctr. Christ., ii. 11) of discordant Latin texts, but only one Syriac, so far as is known.

3. Badness in Latin texts is just what we should expect amongst people who were poor Greek scholars, and lived at a distance. The Syrians on the contrary were close to Judea, and Greek had been known among them for centuries. It was not likely that within reach of the Apostles and almost within their lifetime a version should be made so bad as to require to be thrown off afterwards.

As to internal evidence, the opinion of some experts is balanced by the opinion of other experts (see Abbé Martin, Des Versions Syriennes, Fasc. 4). The position of the Peshitto as universally received by Syrian Christians, and believed to date back to the earliest times, is not to be moved by mere conjecture, and a single copy of another version [or indeed by two copies]. Textual Guide, Miller, 1885, p. 74, note 1.

31 On the order, functions, and decay of the Χωρεπίσκοποι, _see_ Bingham’s “Antiquities,” book ii, chap. xiv.

32 Davidson, Bibl. Crit., vol. ii. p. 186, first edition. The Abbé Martin (_see_ p. 323 note), after stating that this version was never used by any Syrian sect save the Monophysites or Jacobites, goes on to ask “Est-ce à dire que cette version soit entachée de monophysisme? Nous ne le pensons pas; pour l’affirmer, il faudra l’examiner très minutieusement; car l’hérésie monophysite est, à quelques points de vue, une des plus subtiles qui aient jamais paru” (Des Versions Syriennes, p. 162).

33 The asterisks ([symbol] [symbol]) and obeli ([symbol] [symbol]) of this version will be observed in our specimens given below. Like the similar marks in Origen’s Hexapla (from which they were doubtless borrowed), they have been miserably displaced by copyists; so that their real purpose is a little uncertain. Wetstein, and after him even Storr and Adler, refer them to changes made in the Harkleian from the Peshitto: White more plausibly considers the asterisk to intimate an addition to the text, the obelus to recommend a removal from it.

34 “Sacrorum Evangeliorum Versio Syriaca Philoxeniana, ex Codd. MSS. Ridleianis in Bibliotheca Novi Collegii Oxon. repositis; nunc primum edita, cum Interpretation Latinâ et Annotationibus Josephi White. Oxonii e Typographeo Clarendoniano,” 1778, 2 tom. 4to. And so for the two later volumes. Ridley named that one of his manuscripts which contains only the Gospels Codex Barsalibaei, as notes of revision by that writer are found in it (e.g. John vii. 53-viii. 11). G. H. Bernstein has also published St. John’s Gospel (Leipzig, 1853) from manuscripts in the Vatican. In or about 1877 Professor Isaac H. Hall, an American missionary, discovered at Beerût a manuscript in the Estrangelo character, much mutilated (of which he kindly sent me a photographed page containing the end of St. Luke and the beginning of St. John), which in the Gospels follows the Harkleian version, although the text differs much from White’s, but the rest of the N. T. is from the Peshitto. Dr. Hall has drawn up a list of over 300 readings differing from White’s.

35 Martin names as useful for the study of a version as yet too little known, the Lectionaries Bodleian 43; Brit. Mus. Addit. 7170, 7171, 7172, 14,490, 14,689, 18,714; Paris 51 and 52; Rome, Vatic. 36 and Barberini vi. 32.

_ 36 See_ also Syriac Manuscript Gospels of a Pre-Harklensian version, Acts and Epp. of the Peshitto version ... by the Monk John. Presented to the Syrian Protestant College, &c., described with phototyped facsimiles by Prof. Isaac H. Hall [viii-ix], ff. 219 + a fragment at end. _Mut._ at beg. and end, &c. Written in old Jacobite characters. Sent courteously to the Editor.

37 Thus also the termination of the definite state plural of nouns is made in ܐ [final form] for ܐ: the third person affix to plural nouns in ܘ for ܗܘ. In the compass of the six verses we have cited (_below_, p. 39) occur not only the Greek words ܘܝܪܘܣܐ (or ܐܣܘܪܝܘ) (καιρός), _v._ 3, and ܢܘܣܐ (or ܐܣܘܢ) (ναός), _v._ 5, which are common enough in all Syriac books, but such Palestinian words and forms as ܕܝ (or ܝܕ) for ܕܝܢ (or ܢܝܕ), δέ (_vv._ 4, 6, 7); ܒܒܝܢ (or ܢܝܒܒ) _v._ 3, “when;” ܐܗܐ _v._ 3, “repented;” ܐܕܡܐ (or ܐܡܕܐ) for ܕܡܐ (or ܐܡܕ) (_vv._ 4, 6, 8), “blood;” ܥܥܝܢܗ (or ܗܢܝܥܥ), _v._ 4, “to us;” ܓܪܡܐ (or ܐܡܪܓ), _v._ 5, “himself;” ܕܡܝܢ (or ܢܝܡܕ), _v._ 6, “price” (Pesh. has ܛܡܝ (or ܝܡܛ), Hark. ܛܝܡܐ (or ܐܡܝܛ) (pl.) τιμή); ܥܦܝܢ (or ܢܝܦܥ) _v._ 8, “therefore;” ܗܐܘ (or ܘܐܗ), _v._ 8, “this.”

38 Hence the name by which this version is distinguished. For the recensions of Targum and Talmud, _see_ Etheridge’s “Hebrew Literature,” pp. 145-6, 195-7.

39 Dr. Hort’s not very explicit judgement should now be added: “The Jerusalem Syriac Lectionary has an entirely different text [from the Harkleian], probably not altogether unaffected by the Syriac Vulgate [meaning thereby the Peshitto], but more closely related to the Old Syriac [meaning the Curetonian]. Mixture with one or more Greek texts containing elements of every great type, but especially the more ancient, has however given the whole a strikingly composite character” (Introd., p. 157).

40 On these readings, and those of the MSS. mentioned below (p. 34), _see_ “The New Syriac Fragments” (F. H. Woods), in the _Expository Times_, Nov., 1893.

_ 41 See_ the “Life and Times of Gregory the Illuminator, the Founder and Patron Saint of the Armenian Church,” translated by the Rev. S. C. Malan, London, 1868.

42 Kept by the Greeks Oct. 23. Gale O. 4. 22 and other Greek Evangelistaria commemorate this holiday.

43 Dec. 27 in the Western Calendar.

44 So Gale O. 4. 22, with the same Lesson.

45 See _Athenaeum_, Oct. 28, 1893.

46 Anecdota Oxoniensia, “The Palestinian Version of the Holy Scripture;” edited by G. H. Gwilliam, B.D.: Oxford, Clarendon Press, 1893.

47 The full form (ܛܘܒܢܐ or ܐܢܒܘܛ _blessed_) occurs in the scholion to Rom. viii. 15; Wiseman thought it meant the Peshitto; but see “Studia Biblica,” iii. 60 and note.

48 Our specimens show the use in MSS. of _rucaca_ and _kushaia_, here printed with fine points. The dots and dashes of the Nestorian Massorah ore also shown.

49 Studia Biblica et Ecclesiastica, iii. 56.

50 The Codex Babylonicus, A.D. 916, is the oldest Old Testament MS. known at present. Dr. Neubauer, Stud. Bibl. et Eccl., iii. 27.

51 Karkaphta = skull. See also “Thes. Syr.,” col. 3762.

52 Mr. Gwilliam suggests that this may have been the well-known Thomas Heracleensis. M. l’Abbé Martin (Tradition Karkaphienne, ou la Massore chez les Syriens), who carefully studied the subject twenty years ago, suggests Thomas of Edessa, teacher of Mar Abbas. _See_ Mr. Gwilliam’s Essay in “Stud. Bibl. et Eccl.,” iii. pp. 56-65.

53 “How the Codex was found” (Lewis and Gibson), 1893.

54 Of no passage is this judgement more true than of this actual sentence itself, which is hardly quoted in the same way in any three MSS.; see Wordsworth’s Vulgate, Fasc. 1, p. 2.

55 For _Itala_ Bentley conjectured _et illa_, changing the following _nam_ into _quae_; and he wrote to Sabatier almost ridiculing the idea of a “Versio Italica;” _see_ Correspondence, ed. Wordsworth, 1842, p. 569; and “Versio Latina Italica, somnium merum,” in Ellis, Bentleii Critica Sacra, pp. 157-159; Kaulen, Gesch. d. Vulgata, Mainz, 1868, p. 116 f.; Abp. Potter conjectured _usitata_ for _Itala_; _see_ Field, Otium Norvicense, pars tertia, p. 57.

56 Bibliorum Sacr. Latinae Versiones Ant. seu Vetus Italica etc. opera et studio D. Petri Sabatier, 3 vols., Rheims, 1743-1749; a revised edition of this great work, for the Old Test., is in course of preparation under the auspices of the Munich Academy, and the able superintendence of Professor E. Wölfflin.

57 Evangeliarium Quadruplex Latinae Versionis Antiquae, seu Veteris Italicae, editum ex codicibus manuscriptis ... a Josepho Blanchino, 2 vols., Rome, 1749; reprinted by Migne, Patr. Lat. xii, with the works of Eusebius Vercellensis.

58 That is, by scholars who did not live in Italy; Italian Christians would use other names, _vetus_, _antiqua_, _usitata_, _communis_, _vulgata_; Kaulen, p. 118, Berger, p. 6.

59 Published in the _Catholic Magazine_ for 1832-3; since reprinted in his “Essays on various subjects,” 1853, vol. i.

60 We have let these sentences stand as Dr. Scrivener penned them in 1883; since that time the opinion of scholars has become less positive as to the African origin of the Latin version. It is true that the words, phrases, &c., of that version in its earlier forms can be illustrated from contemporary African writers, and from them only; but that is because during this period we are dependent almost exclusively on Africa for our Latin literature; and consequently are able to use only the method of _agreement_ and not the method of _difference_ in testing the origin and characteristics of the Latin New Testament. These characteristics may be the result only of the time and not of the supposed place of writing. Nor can more stress be laid on the use of Greek names in the West than on the use of Latin names (plenty of which could be cited) in the East.

_ 61 See_ Kaulen, p. 130 f., and also his Handb. d. Vulg., Mainz, 1870.

62 “Novum opus me facere cogis ex veteri: ut post exemplaria Scripturarum toto orbe dispersa, quasi quidam arbiter sedeam: et quia inter se variant, quae sint ilia quae cum Graeca consentiant veritate, decernam. Pius labor, sed periculosa praesumptio, judicare de ceteris, ipsum ab omnibus judicandum: senis mutare linguam, et canescentem jam mundum ad initia retrahere parvulorum.” Praef. ad Damasum.

63 “[Evangelia] Codicum Graecorum emendata collatione, sed veterum, quae ne multum a lectionis Latinae consuetudine discreparent, ita calamo temperavimus, ut his tantum quae sensum videbantur mutare correctis, reliqua manere pateremur ut fuerant.” _Ibid._ For a signal instance, see below, ch. ix, note on Matt. xxi. 31.

64 To his well-known censure of Jerome’s rendering of the Old Testament from the Hebrew, Augustine adds, “Proinde non parvas Deo gratias agimus de opere tuo, quod Evangelium ex Graeco interpretatus es: quia pene in omnibus nulla offensio est, cum Scripturam Graecam contulerimus.”

65 Roger Bacon’s writings, however, in the thirteenth century, are the first in which Jerome’s translation is cited as the “Vulgate” in the modern sense of the term. _See_ Denifle, Die Handschriften der Bibel-correctorien des 13. Jahrhunderts, 1883, p. 278.

_ 66 See_ Jaffé, Monumenta Carolina, p. 373, “Jam pridem universos Veteris ac Novi instruments libros ... examussim correximus;” S. Berger’s essay (to be distinguished from his larger work), De l’histoire de la Vulgate en France (1887), p. 3 f.

_ 67 See_ the Oxford “Studia Biblica et Ecclesiastica,” ii (1890), p. 278 f.

68 Fritzsche, “Latein. Bibelübersetzungen” in Herzog, R. E.2 viii. p. 449; Westcott, “Vulgate,” in Smith’s Bibl. Dict. iii. p. 1703; Kaulen, Gesch. d. Vulg., p. 229 f.; P. Corssen, in “Die Trierer Adahandschr.” (Leipzig, 1889), p. 31.

69 Berger, as above, p. 7.

_ 70 See_ the Life of Lanfranc, by Milo Crispinus, a monk of Bec, ch. xv, in Migne, Patr. Lat. 150, col. 55, and his Commentary, _ibid._, col. 101 f.; Mill, Proleg., § 1058; Cave’s remark (Hist. Lit. 1743, vol. ii. p. 148), “Lanfrancus textum continuo emendat,” seems hardly borne out by the facts.

71 His corrected Bible in four vols. is now preserved at Dijon, public library, 9 bis, _see_ below, p. 68, no. 8; also Denifle, Die Hdss. d. Bibel-correctorien des 13. Jahrh. 1883, p. 267; Kaulen, p. 245.

72 His criticisms are preserved in a MS. at Venice (Marciana Lat. class. x. cod. 178, fol. 141); _see_ Denifle, p. 270, who prints them.

_ 73 See_ the quotations in Denifle, p. 277 f., and Hody, p. 419 f.

_ 74 See_ S. Berger, De l’histoire de la Vulgate en France, p. 9 f., 1887, and Revue de Théol. et de Philos. de Lausanne, t. xvi. p. 41, 1883.

_ 75 See_ Hugo’s remark (Denifle, p. 295), “In multis libris maxime historialibus, non utimur translatione Hieronymi.”

_ 76 See_ Vercellone, Diss. Acad., Rome, 1864, pp. 44-51; Hody, pp. 426-430; and Denifle, pp. 295-298. This correctorium is cited in Wordsworth’s Vulgate as _cor. vat._; _see_ Berger, Notitia Linguae Hebraicae etc., p. 32 (1893).

_ 77 See_ W. A. Copinger, Incunabula Biblica, or the first half-century of the Latin Bible, p. 3, London, 1892; and L. Delisle, Journ. des Savants, Apr. 1893.

78 Or to Peter Schoeffer, _see_ J. H. Hessels, in the _Academy_, June, 1887, p. 396; August, p. 104; or to Johann Fust. _See_ the British Museum “Catalogue of Printed Books,” Bible, part i. col. 16.

79 Westcott, Vulgate, p. 1704. This seems to be that of “Thielman Kerver, impensis J. Parvi,” with emendations of A. Castellani.

80 The British Museum possesses a copy (340. d. 1); _see_ the “Catalogue,” part i. col. 1.

81 For details _see_ “Old Lat. Bibl. Texts,” i. p. 51 f.

_ 82 Ibid._, p. 48 f.

83 The critical notes of Lucas Brugensis himself appear to be found in three forms:—

(1) The “Notationes,” published in 1580, and incorporated in the Hentenian Bible of 1583.

(2) The “Variae Lectiones,” printed in Walton’s Polyglott, and taken from the Louvain Bible of 1584. These are simply a list of various readings to the Vulgate, with MS. authorities; he frequently adds the letters Q. N., i.e. “quaere notationes,” where he has treated the subject more fully in (1).

(3) The “Notae ad Varias Lectiones,” also printed (for the Gospels) in Walton’s Polyglott; a _delectus_ of them is given in Sabatier at the end of each book of the New Testament, under the title “Roman. Correctionum auctore Fr. L. Br. delectus.”

_ 84 See_ E. Nestle, Ein Jubiläum der lateinischen Bibel, Tübingen, p. 13 f., 1892.

85 There is a copy in the British Museum, Q. e. 5. It is practically in one volume, as the paging is continuous throughout.

86 He gives a long list of the variations between the Sixtine and Clementine Bibles; Vercellone estimated their number at 3,000. It is to be noticed that the _versing_ of the Sixtine ed. differs considerably from the Clementine as well as from Stephen.

87 The regular form of title, “Biblia Sacra Vulgatae Editionis Sixti V Pont. Max. jussu recognita et Clementis VIII auctoritate edita,” does not appear in any edition known to the writer before that of Rouille, Lyons, 1604. _See_ Brit. Mus. Catalogue, col. 50. The earliest edition with this title known to Masch (Le Long, Bibl. Sacra, 1783, ii. p. 251) is dated 1609; and Vercellone (Variae Lect. i. p. lxxii) names others considerably later as the earliest.

_ 88 See_ Old Lat. Bibl. Texts, i. p. xvi.

_ 89 Ibid._, p. xxv.

_ 90 See_ Fasc. i. p. xv, and Ellis, Bentleii Critica Sacra, Cambridge, 1862.

91 M. Berger, with exceptional kindness, allowed me to see the proof-sheets of his “History of the Vulgate” as they were printed, and to add a large number of MSS. to this list from that source.

92 For the Würzburg MSS., _see_ G. Schepps, Die ältesten Evangelienhandschriften der Universitätsbibliothek, Würzburg, 1887, from which these descriptions are mainly taken.

93 For these MSS., _see_ as before, G. Schepss, Die ältesten Evangelienhandschriften d. Würzb. Univ. B., 1887.

94 My authority for these facts is Brugsch, Grammaire Démotique, p. 4, but what does he mean by the words which I have italicised? “Au nombre des auteurs les plus récents qui nous aient donné des témoignages sur l’existence du démotique il faut citer St. Clément, prêtre de l’église chrétienne à Alexandrie, et qui vivait vers l’an 190 de notre ère, ou environ le temps où régnait l’empereur Sévère. Mais les monuments nous prouvent que _cette date n’est pas la dernière_; il se trouve encore des inscriptions d’une époque plus rapprochée; telle est par exemple une inscription démotique que M. de Saulcy avait copiée en Égypte et qu’il eut la complaisance de me communiquer pendant mon séjour à Paris; elle date du règne en commun d’Aurélius et de Vérus, ce qui prouve que _dans la première moitié du troisième siècle_ le démotique était encore connu et en usage.” L. Verus died A.D. 169.

95 The date, however, is placed very much earlier by Revillout (Mélanges d’Archéologie Égyptienne et Assyrienne, p. 40), who supposes the Coptic alphabet to have been a work commenced by pagan Gnostics, completed by Christian Gnostics, and adopted when complete by their orthodox successors.

96 [That Bahiric is a wrong transliteration is shown by Stern, Zeitschr. für Aeg. Sprache, 16 (1878), p. 23.]

97 [There has been considerable variation in the names given to the different dialects. The terms Thebaic and Memphitic have been commonly adopted as a more convenient nomenclature, but, as will be shown below, the latter name at any rate is incorrect and misleading. Owing to the accident that the Memphitic dialect was the form of Coptic best known and earliest studied in Western Europe, the term Coptic has been sometimes confined to the Bohairic or Memphitic, as distinguished from the Sahidic or Thebaic, and was so used by Tischendorf; this usage also is erroneous and misleading; and the names Bohairic and Sahidic are almost universally employed by scholars at the present day.]

98 Schwartze, whose opinion will not be suspected of any theological bias, infers from the historical notices that “the greatest part of the New Testament writings, if not all, and a part of the Old Testament, especially the Psalms, had been already translated, in the second century, into the Egyptian language, and indeed into that of Lower as well as into that of Upper Egypt” (p. 963).

99 For convenience the following abbreviations will be used: “Z. A. S.” for _Zeitschrift für Aegyptische Sprache_; “Recueil” for the _Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes_; “Mémoires” for the _Mémoires de la Mission Archéologique Française au Caire_; and “Mitt.” for the _Mittheilungen aus der Sammlung der Papyrus Erzherzog Rainer_.

100 Quatremère can only point to a single word accidentally preserved, which according to his hypothesis belongs to the real Bashmuric (Sur la Langue &c., p. 213 sq.).

101 Memphitic (Lightfoot), Coptic (Tischendorf and others).

_ 102 See_ also A. J. Butler’s “Coptic Churches,” vol. ii, Oxford.

103 I have always added 284 to the year of the Martyrs for the year A.D.; but this will not give the date accurately in every case, as the Diocletian year began in August or September; _see_ Clinton, Fast. Rom., ii. p. 210.

104 I have observed Luke xxiii. 17 in at least three wholly distinct forms in different Bohairic MSS.

105 My sincere thanks are due to the late Earl of Crawford and Balcarres, and to Lord Zouche, for their kindness in allowing me free access to their valuable collections of Coptic MSS., and in facilitating my investigations in many ways.

106 The volume, *Parham 102, described in the printed Catalogue (no. 1, vellum, p. 27) as a MS. of the Gospels of St. Matthew and St. Mark, is really a selection of passages taken in order from the four Gospels, with a patristic catena attached to each. The leaves, however, are much displaced in the binding, and many are wanting. The title to the first Gospel is ϯ ⲉⲣⲙⲏⲛⲓⲁ ⲛⲧⲉ ⲡⲓⲉⲩⲁⲅⲅⲉⲗⲓⲟⲛ ⲉⲑⲟⲩⲁⲃ ⲕⲁⲧⲁ ⲙⲁⲑⲉⲟⲛ ⲉⲃⲟⲗϩⲓⲧⲉⲛ ϩⲁⲛⲙⲏϣ ⲛⲥⲁϧ ⲟⲩⲟϩ ⲛⲫⲱⲥⲧⲏⲣ ⲛⲧⲉ ϯ ⲉⲕⲕⲗⲏⲥⲓⲁ, &c. “The interpretation of the Holy Gospel according to Matthew from numerous doctors and luminaries of the Church.” Among the Fathers quoted I observed Athanasius, Basil, Chrysostom, Clement, the two Cyrils (of Jerusalem and of Alexandria), Didymus, Epiphanius, Eusebius, Evagrius, the three Gregories (Thaumaturgus, Nazianzen, and Nyssen), Hippolytus, Irenaeus, Severianus of Gabala, Severus of Antioch (often styled simply the Patriarch), Symeon Stylites, Timotheus, and Titus.

In the account of this MS. in the Catalogue it is stated that “the name of the scribe who wrote it is Sapita Leporos, a monk of the monastery, or monastic rule, of Laura under the sway of the great abbot Macarius,” and the inference is thence drawn that it must have been written before 395, when Macarius died. This early date, however, is at once set aside by the fact that writers who lived in the sixth century are quoted. Professor Wright (Journal of Sacred Literature, vii. p. 218), observing the name of Severus in the facsimile, points out the error of date, and suggests as an explanation that the colophon (which he had not seen) does not speak of the great Macarius, but of “_an abbot_ Macarius.” The fact is, that though the great Macarius is certainly meant, there is nothing which implies that he was then living. The scribe describes himself as ⲁⲛⲟⲕ ϧⲁ ⲡⲓ ⲧⲁⲗⲉⲡⲱⲣⲟⲥ ⲉⲧⲁϥⲥϧⲁⲓ, “I the unhappy one (ταλαιπωρος) who wrote it” (which has been wrongly read and interpreted as a proper name Sapita Leporos). He then gives his name ⲑⲉⲟⲗ ⲡⲟⲩⲥⲓⲣⲓ (Theodorus of Busiris?) and adds, ⲡⲓⲁⲧⲙⲡϣⲁ ⲙⲙⲟⲛⲁⲭⲟⲥ ⲛⲧⲉ ϯⲗⲁⲩⲣⲁ ⲉⲑⲟⲩⲁⲃ ⲛⲧⲉ ⲡⲓⲛⲓϣϯ ⲁⲃⲃⲁ ⲙⲁⲕⲁⲣⲓ, “the unworthy monk of the holy laura of the great abbot Macarius.” He was merely an inmate of the monastery of St. Macarius; see the expression quoted from the Vat. MS. lxi in Tattam’s Lexicon, p. 842. This magnificent MS. is dated A.M. 604 = A.D. 888 and has been published by Professor De Lagarde; but its value may not be very great for the Bohairic Version, as it is perhaps translated from the Greek.

The *Parham MS. 106 (no. 5, p. 28) is wrongly described as containing the Gospel of St. John. The error is doubtless to be explained by the fact that the name ⲓⲱⲁⲛⲛⲟⲩ occurs at the bottom of one of the pages; but the manuscript is not Biblical. Another MS. (no. 13, p. 29) is described as “St. Matthew with an Arabic translation, very large folio: a modern MS. copied at Cairo from an antient one in the library of the Coptic Patriarch.” I was not able to find this, when through the courtesy of Lord Zouche I had access to the Parham collection.

107 The above account has been throughout revised by the Rev. G. Horner, who has collated or examined all MSS. of the Bohairic versions in European libraries.

108 The MSS. 7 and 16 are exceptions.

109 No weight can be given to the abnormal order in no. 12, until we know something more of this MS., which is perhaps a late transcript.

110 It is used in the Apocalypse by Tregelles, and apparently also by Tischendorf in his eighth edition; and in the Rev. S. C. Malan’s “Gospel according to St. John, translated from the Eleven Oldest Versions except the Latin,” London, 1862, all Tuki’s Sahidic fragments of this Evangelist are included.

_ 111 See_ Münter, De Indole, &c., Praef., p. iv. Schwartze (Quat. Evang. p. xx) says, “Praeterquam quod sicut omnes Tukii libri scatent vitiis, etiam angustioris sunt fidei _Rudimenta_, Sahidicis locis partim e versione Arabica a Tukio concinnatis.” I do not know on what grounds Schwartze makes this last statement.

112 This has now been published. By Amélineau, Notice sur le Papyrus Gnostique Bruce. Texte et Traduction, Notices et Extraits de la Bibliothèque Nationale et autres Bibliothèques. Tome xxix. lre Partie. Paris, 1891; and Gnostische Schriften in Koptischer Sprache aus dem Codex Brucianus, von Carl Schmidt, Leipzig, 1892.

113 In the interval between Woide and Zoega, Griesbach (1806) appears to have obtained a few readings of this version from the Borgian MSS., e.g. Acts xxiv. 22, 23; xxv. 6; xxvii. 14; Col. ii. 2. At least I have not succeeded in tracing them to any printed source of information.

Of the use which Schwartze has made of the published portions of the Sahidic text in his edition of the Bohairic Gospels, I have already spoken (p. 108). He has added no unpublished materials.

114 Catal., p. 169: “Si de aetate codicum quaeris, scio equidem non defuisse qui singulos ad saecula sua referre satagerent, qui si aliquid profecerunt, ego sane non obstrepo. Sed quoniam meum sit quacumque in re ignorantiam fateri potius quam quae mihi non satisfaciunt, aliis velut explorata offerre, &c.” But since this was written the publication of Hyvernat’s “Album de Paléographie Copte” has given much assistance; and more may be looked for from the publication of the Paris fragments.

115 Its position was before Galatians, and not, as in the archetype of the Codex Vaticanus, after it.

116 The term “Middle Egyptian” is often used as a general term to include the three varieties of Fayoumic, Lower Sahidic or what is properly Memphitic, and Akhmimic.

117 The writer must express his regret that, owing to the haste with which the additions to this article had to be written, much must have been passed over.

118 “But he prudently suppressed the four books of Kings, as they might tend to irritate the fierce and sanguinary spirit of the barbarians;” Gibbon, ch. xxxvii.

119 “A faithful, a stern and noble Teutonic rendering of the Greek,” is the verdict of Prebendary S. C. Malan (St. John’s Gospel, translated from the Eleven Oldest Versions except the Latin, &c., 4to, 1872, Preface, p. viii). Bishop Ellicott also praises this version as usually faithful and accurate, yet marks an Arian tinge in the rendering of Phil. ii. 6-8.

120 Goth. Version. Paul. Epist. quae supersunt, C. O. Castiglione, Milan, 1834.

121 Skeat, St. Mark, 1882.

122 Matt. iii. 11; v. 8; 15-vi. 32; vii. 12-x. 1; 23-xi. 25; xxv. 38-xxvi. 3; 65-xxvii. 19; 42-66; Mark i. 1; vi. 30; 58-xii. 38; xiii. 16-29; xiv. 4-16; 41-xvi. 12; Luke i. 1-x. 30; xiv. 9-xvi. 24; xvii. 3-xx. 46; John i. 29; iii. 3-5; 23-26; 29-32; v. 21-23; 35-38; 45-xi. 47; xii. 1-49; xiii. 11-xix. 13; Rom. vi. 23; vii. 1-viii. 10; 34-xi. 1; 11-xii. 5; 8-xiv. 5; 9-20; xv. 3-13; xvi. 21-24; 1 Cor. i. 12-25; iv. 2-12; v. 3-vi. 1; vii. 5-28; viii. 9-ix. 9; 19-x. 4; 15-xi. 6; 21-31; xii. 10-22; xiii. 1-12; xiv. 20-27; xv. 1-35; 46-Gal. i. 7; 20-iii. 6; 27-Eph. v. 11; 17-29; vi. 8-24; Phil. i. 14-ii. 8; 22-iv. 17; Col. i. 6-29; ii. 11-iv. 19; 1 Thess. ii. 10-2 Thess. ii. 4; 15-1 Tim. v. 14; 16-2 Tim. iv. 16; Tit. i. 1-ii. 1; Philem. 1-23; but no portion of the Acts, Hebrews, Catholic Epistles, or Apocalypse.

_ 123 See_ p. 10 of the Armenian edition; Venice, 1833. The French translation of this in the “Collection des Historiens de l’Arménie,” Paris, 1869, is untrustworthy in all ways, and especially because the translator both adds to and omits from the Armenian text at random.

124 The true history of which we cannot now make out, for, as given by his contemporaries, it is already obscured by legend and miracle.

125 The translation of this writer in Langlois’ second volume is reliable.

126 Some critics bring down the date of Moses as late as the seventh or eighth century.

127 Dr. Baronean thinks that the varieties of readings in the oldest Armenian MSS. is due to the fact that more than one _sure_ copy was brought from Constantinople on which to base the final revision.

128 This is the conclusion at which P. P. Carékin arrives. See his “Catalogue of Ancient Armenian Translations,” Venice, 1889, p. 228.

129 Among the chief authorities on the Slavonic version are the following:—

(i) Горскій и Невоструевъ, описаніе славянскихъ рукописей Московской Синодальной Библіотеки. Москва, 1855.

(ii) Астафьевъ, Опьітъ исторіи библіи въ Россіи въ связи съ просвѣщеніемъ и нравами. С. Петербургъ, 1892.

(iii) Voskresenski, Характеристческія чертъі гиавнъіхъ редакцій славянскаго перевода Евангелія.

(iv) Voskresenski, Древній славянскій переводъ Апостола и его судьбы до xv вѣка.

(v) Oblak, Die Kirchenslavische Uebersetzung der Apocalypse [in the “Archiv für Slavische Philologie,” xiii. pp. 321-361].

(vi) Prolegomena to the editions of the Codex Marianus and the Codex Zographensis, &c., by Jagić.

(vii) Kaluzniacki, Monumenta Linguae Palaeoslavonicae, vol. i.

130 In the Synodal Library at Moscow this proportion is as nine to two, and in another library as twelve to one. _See_ Описаніе славянскихъ рукописей и т. д. (as above), p. 299.

131 Kaluzniacki, _l. c._, p. xlv, gives instances.

_ 132 See_ Jagić, Codex Zographensis, pp. xxvii ff.

133 The statement that John Bishop of Seville translated the Bible into Arabic in A.D. 719 is disproved by Lagarde (Die vier Evangelien Arabisch, p. xv).

134 Edward Pocock, Professor of Hebrew at Oxford (1648-91) and a great Oriental scholar, should be distinguished from Richard Pococke, an Eastern traveller and Bishop of Meath, who died in 1765.

135 I have been obliged to alter the first paragraph in this chapter because of Dr. Scrivener’s private confession to myself of the great value of Dean Burgon’s services in this province of Sacred Textual Criticism. I am convinced that he could not have continued to maintain an opinion so adverse to the value of early citations as that which he formed when people were not sufficiently aware of the wealth of illustrative evidence that lay ready to their hands. As Editor I owe very much in this chapter, both to the express teaching in Dean Burgon’s great book, and to his method of argument in respect to patristic citations. The Dean did not leave this province at all as he found it.

136 The Revision Revised, by John William Burgon, B. D., Dean of Chichester. John Murray, 1883.

_ 137 See_ some very thoughtful and cautious remarks by the Rev. Ll. J. M. Bebb in the second volume of the Oxford “Studia Biblica (et Ecclesiastica).” Mr. Bebb’s entire Article on “The Evidence of the Early Versions and Patristic Quotations on the Text of the Books of the New Testament” is well worth careful study.

138 “Dated codices, in fact they are, to all intents and purposes.” Burgon, Revision Revised, p. 292. “Every Father is seen to be a dated witness and an independent authority,” p. 297.

139 I am glad to be able to coincide thus far with the judgement of Mr. Hammond, who says: “The value of even the most definite Patristic citation is only corroborative. Standing by itself, any such citation might mean no more than that the writer found the passage in his own copy, or in those examined by him, in the form in which he quotes it. The moment, however, it is found to be supported by other good evidence, the writer’s authority may become of immense importance” (Outlines of Textual Criticism, p. 66, 2nd edition). His illustration is the statement of Irenaeus in Matt. i. 18, which is discussed below, Chap. XI. (Third Edition.)

140 He speaks (N. T., Proleg., § 1478) of Bp. Fell’s “praepropera opinio;” he merely stated as _universally_ true what for the most part certainly is so.

141 Take the case of Irenaeus, in some respects the most important of them all. The _editio princeps_ of Erasmus (1526) was printed from manuscripts now unknown. The three best manuscripts are in Latin only. The oldest of them I saw at Middle-hill, an exquisite specimen of the tenth or eleventh century, _olim_ Claromontanus; another, of the twelfth, is in the Arundel collection in the British Museum; the third once belonged to Vossius.

142 Tischendorf (N. T., Proleg., p. 256, 7th edition) speaks of one Wolfenbüttel manuscript of the sixth century containing the Homilies on St. Matthew, which he designed to publish in his “Monumenta Sacra Inedita,” vol. vii. He indicates its readings by Chrgue.

143 Life of Dean Burgon, by Dean Goulburn, p. 82, note. Murray, 1892.

144 Dampar cod. i.e. “Joh. Damasceni parallela sacra ex cod. Rupefuc. saeculi ferè 8.” Tischendorf, N. T., Preface to vol. i of the eighth edition, 1869. He promised full information in his “Prolegomena,” which never appeared. Here we have a manuscript ascribed to the same century as the Father whose work it contains. One MS. is at Paris (collated by Mr. Rendel Harris, A.D. 1884); another in Phillipps collection at Cheltenham.

145 This important witness for the Old Latin version must now be used with H. Roensch’s “Das Neue Testament Tertullian’s,” Leipzig, 1871, wherein all his citations from the N. T. are arranged and critically examined.

_ 146 See_ Dean Burgon’s Appendix (D) to his “Last Twelve Verses of St. Mark,” pp. 269-287, which well deserves the praise accorded to it by a not very friendly critic. The Dean discusses at length the genius and character of Victor of Antioch’s Commentary on St. Mark, and enumerates the manuscripts which contain it.

147 It should be stated that some of the dates in the two tables just given are doubtful, authorities differing.

148 Since the first edition of this book was issued, Ed. Reuss has published “Bibliotheca Novi Testamenti Graeci, cuius editiones ab initio typographiae ad nostram aetatem impressas quotquot reperiri potuerunt collegit digessit illustravit E. R. Argentoratensis” (Brunsvigae, 1872), to which the reader is referred for editions which our purpose does not lead us to notice. Some of his statements regarding the text of early editions we have repeated in the notes of the present chapter. His enumeration is not grounded on a complete collation of any book, but from the study of a thousand passages (p. 24) selected for his purpose. Hence his numerical results are perpetually less than our own, or even than Mill’s. Professor Isaac H. Hall in Schaff’s “Companion to the Greek Testament and the English Version,” D. I. Macmillan, 1883, has improved upon Reuss, and given a list of editions which as to America is, I believe, exhaustive (_see_ also his “American Greek Testaments—a Critical Bibliography of the Greek New Testament as published in America”—Philadelphia, Pickwick and Company, 1883), and is very full as regards English and other editions. I should like to have availed myself of the Professor’s kind permission to copy that list, but it would have been going out of the way to do so, since these two chapters are simply upon the _Early_ Printed and the _Critical_ Editions of the Text.—ED.

149 “Novum Testamentum Grece et Latine in academia complutensi noviter impressum,” Tom. v.

150 Quite enough has been made of that piece of grim Spanish humour, “Mediam autem inter has latinam beati Hieronymi translationem velut inter Synagogam et Orientalem Ecclesiam posuimus: tanquam duos hinc et inde latrones, medium autem Jesum, hoc est Romanam sive latinam Ecclesiam collocantes” (Prol. Tom. i). The editors plainly meant no disparagement to the original Scriptures, _as such_; but they had persuaded themselves that Hebrew codices had been corrupted by the Jew, the Septuagint by the schismatical Greek, and so clung to the Latin as the only form (even before the Council of Trent) in which the Bible was known or studied in Western Europe.

151 Of these, two copies are in Greek, three in Latin Elegiacs. I subjoin those of the native Greek editor, Demetrius Ducas, as a rather favourable specimen of verse composition in that age: the fantastic mode of accentuation described above was clearly not _his_ work.

Ειπράξεις ὅσιαι ἀρετήτε βροτοὺς ἐς ὅλυμπον, ἐσμακάρων χῶρον καὶ βίον οἶδεν ἄγειν, ἀρχιερεὺς ξιμένης θεῖος πέλει. ἔργα γὰρ αὐτοῦ ἤδε βίβλος. θνητοῖς ἄξια δῶρα τάδε.

152 Tregelles (Account of the Printed Text, p. 7, note) states that he was _elected_ Feb. 28, crowned March 11: Sir Harris Nicolas (“Chronology of History,” p. 194) that he was elected March 11, without naming the date of his coronation as usual, but mentioning that “Leo X, in his letters, dated the commencement of his pontificate before his coronation.”

153 The following is the document (a curiosity in its way) as cited by Vercellone: “Anno primo Leonis PP. X. Reverendiss. Dom. Franciscus Card. Toletanus de mandato SS. D. N. Papae habuit ex bibliotheca a Dom. Phaedro Bibliothecario duo volumina graeca: unum in quo continentur libri infrascripti; videlicet Proverbia Salomonis, Ecclesiastes, Cant. Cant., Job, Sapientia, Ecclesiasticus, Esdras, Tobias, Judith [this is Vat. 346, or 248 of Parsons]. Sunt in eo folia quingenta et duodecim ex papyro in nigro. Fuit extractum ex blancho primo bibliothecae graecae communis. Mandatum Pontificis super concessione dictorum librorum registratum fuit in Camera Apostolica per D. Franciscum De Attavantes Notarium, ubi etiam annotata est obligatio. Promisit restituere intra annum sub poena ducentorum ducatorum.”—“Restituit die 9 Julii, MDXVIII. Ita est. Fr. Zenobius Bibliothecarius.”

154 The Catalogue is copied at length by Tregelles (Account of the Printed Text, pp. 15-18). It is scarcely worth while to repeat the silly story taken up by Moldenhawer, whose admiration of _las cosas de España_ was not extravagantly high, that the Alcalà manuscripts had been sold to make sky-rockets about 1749; to which statement Sir John Bowring pleasantly adds in 1819, “To celebrate the arrival of some worthless grandee.” Gutierrez’s recent list comprehends all the codices named in the University Catalogue made in 1745; and we may hope that even in Spain all grandees are not necessarily worthless.

155 Thus in St. Mark the Complutensian varies from Laud. 2 in fifty-one places, and nowhere agrees with it except in company with a mass of other copies. In the Acts on the contrary they agree 139 times, and differ but forty-one, some of their _loci singulares_ being quite decisive: e.g. x. 17; 21; xii. 12; xvii. 31; xx. 38; xxiv. 16; 1 Pet. iii. 12; 14; 2 Pet. i. 11. In most of these places Seidel’s Codex, in some of them Act. 69, and in nearly all Cod. Havn. 1 (Evan. 234, Act. 57, Paul. 72) are with Laud. 2. On testing this last at the Bodleian in some forty places, I found Mill’s representation fairly accurate. As might have been expected, his Oxford manuscripts were collated much the best.

156 Goeze’s “Defence of the Complutensian Bible,” 1766. He published a “Continuation” in 1769. _See_ also Franc. Delitzsch’s “Studies on the Complutensian Polyglott” (Bagster, 1872), derived from his Academical Exercise as Dean of the Theological Faculty at Leipzig, 1871-2.

157 Reuss says boldly that the Complutensian text “purus et authenticus a veteribus nunquam repetitus est” (p. 25), and gives a list of forty-four places in which the Complutensian and Plantin editions are at variance (pp. 16, 17). He subjoins a list of 185 cases in which the two are in unison against Erasmus and Stephen jointly (pp. 18-21), so that the influence of the former over the latter cannot be disputed.

158 At forty he obtained the countenance of that good and bountiful rather than great prelate, William Wareham, Archbishop of Canterbury (1502-32), who, prosperous in life, was so singularly “felix opportunitate mortis.” It gladdens and makes sad at once an English heart to read what Erasmus writes about him ten years later: “Cujusmodi Maecenas, si mihi primis illis contigisset annis, fortassis aliquid in bonis literis potuissem. Nunc natus saeculo parum felici, cum passim impunè regnaret barbaries, praesertim apud nostrates, apud quos turn crimen etiam erat quicquam bonarum literarum attigisse, tantum aberat ut honos aleret hominum studia in eâ regione, quae Baccho Cererique dicata sunt verius quam musis” (N. T. 1516, Annot. 1 Thess. ii. p. 554).

159 Bishop Middleton may have lost sight of this pregnant fact when he wrote of Erasmus, “an acquaintance with Greek criticism was certainly not among his best acquirements, as his Greek Testament plainly proves: indeed he seems not to have had a very happy talent for languages” (Doctrine of the Greek Article, p. 395, 3rd edition).

160 The title-page is long and rather boastful. “Novum Instrumentum omne, diligenter ab Erasmo Roterodamo recognitum et emendatum, non solum ad graecam veritatem, verum etiam ad multorum utriusque linguae codicum, eorumque veterum simul et emendatorum fidem, postremo ad probatissimorum autorum citationem, emendationem, et interpretationem, praecipue, Origenis, Chrysostomi, Cyrilli, Vulgarii, Hieronymi, Cypriani, Ambrosii, Hilarii, Augustini, una cum Annotationibus, quae lectorem doceant, quid qua ratione mutatum sit. Quisquis igitur amas veram theologiam, lege, cognosce, ac deinde judica. Neque statim offendere, si quid mutatum offenderis, sed expende, num in melius mutatum sit. Apud inclytam Germaniae Basilaeam.” The Vulgarius of Erasmus’ first edition is no less a person than Theophylact, Archbishop of Bulgaria, as appears plainly from his Annotations, p. 319, “nec in ullis graecorum exemplaribus addita reperi [ἐκ σοῦ, Luke i. 35], ne apud Vulgarium quidem, nec in antiquis codicibus Latinis.” He had found out his portentous blunder by 1528, when, in his “Responsio ad Object, xvi. Hispanorum,” he gives that commentator his right name.

161 Yet he could have followed none other than Cod. 1 in Matt. xxii. 28; xxiii. 25; xxvii. 52; xxviii. 3, 4, 19, 20; Mark vii. 18, 19, 26; x. 1; xii. 22; xv. 46; Luke i. 16, 61; ii. 43; ix. 1, 15; xi. 49; John i. 28; x. 8; xiii. 20; in all which passages the Latin Vulgate is neutral or hostile. See also Hoskier, Cod. Ev. 604, App. F. p. 4.

162 Such are ὀρθρινός, Apoc. xxii. ver. 16; ἐλθέ bis, ἐλθέτω, λαμβανέτω τό, ver. 17; συμμαρτυροῦμαι γάρ, ἐπιτιθῇ πρὸς ταῦτα,—τῷ (_ante_ βιβλίῳ) ver. 18; ἀφαιρῇ, βίβλου, ἀφαιρῆσει, βίβλου _secund_., καί ult-τῷ (_ante_ βιβλίῳ) ver. 19; ἡμῶν, ὑμῶν, ver. 21. Erasmus in his Annotations fairly confesses what he did: “quanquam in calce hujus libri, nonnulla verba reperi apud nostros, quae aberant in Graecis exemplaribus, ea tamen ex latinis adjecimus.” But since the text and commentary in Cod. Reuchlini are so mixed up as to be undistinguishable in parts without the aid of a second manuscript (Tregelles’ “Delitzsch’s Handschriftliche Funde,” Part ii. pp. 2-7), it is no wonder that in other places Erasmus in his perplexity was sometimes tempted to translate into his own Greek from the Latin Vulgate such words or clauses as he judged to have been wrongly passed over by his sole authority, e.g. ch. ii. 2, 17; iii. 5, 12, 15; vi. 11, 15 (_see_ under Apoc. 1); vii. 17; xiii. 4, 5; xiv. 16; xxi. 16; xxii. 11, where the Greek words only of Erasmus are false; while in ch. ii. 3; v. 14 (_bis_); vi. 1, 3, 5, 7; xiii. 10; xiv. 5 (as partly in xxii. 14), he was misled by the recent copies of the Vulgate, whereto alone he had access, to make additions which no Greek manuscript is known to support. Bengel’s acuteness had long before suspected that ch. v. 14; xxii. 11, and the form ἀκαθάρτητος, ch. xvii. 4 (where Apoc. 1 has τὰ ἀκάθαρτα) had their origin in no Greek copy, but in the Vulgate. Nor does Apoc. 1 lend any countenance to ch. xvii. 8, καίπερ ἔστι, or to ver. 13, διαδιδώσουσιν. For Erasmus’ πληρώσονται ch. vi. 11, Apoc. 1 has πληρώσωσιν, the Latin _impleantur_; for his σφραγίζωμεν, ch. vii. 3, we find σφραγίσωμεν in Apoc. 1, but the latter omits τῆς ἀμπέλου, ch. xiv. 18, and so does Erasmus on its authority.

163 Tregelles, Account of the Printed Text, p. 19.

164 It sometimes happens that a reading cited in the Annotations is at variance with that given in the text; but Erasmus had been engaged in writing the former for about ten years at intervals, and had no leisure to revise them then. Thus John xvii. 2 δώσει (after Cod. 1, but corrected to δώση in the errata); 1 Thess. ii. 8; iii. 1; 1 Tim. v. 21; Apoc. i. 2; ii. 18; xiv. 10, 13; xxi. 6.

165 The first complete printed English N. T. (Tyndale 1526) followed Erasmus’ third edition rather than his second: cf. Rom. viii. 20, 21 as well as 1 John v. 7, 8.

166 I never saw the Basle manuscripts, and probably Dean Alford had been more fortunate, otherwise I do not think he has evidence for his statement that ’Erasmus tampered with the readings of the very few MSS. which he collated’ (N. T., vol. i. Proleg. p. 74, 4th edition). The truth is, that to save time and trouble, he used them as _copy_ for the press, as was intimated above, where Burgon’s evidence is quite to the point. For this purpose corrections would of course be necessary (those made by Erasmus were all too few), and he might fairly say, in the words cited by Wetstein (N. T., Proleg., p. 127), “se codices suos praecastigasse.” Any wanton “tampering” with the text I am loth to admit, unless for better reasons than I yet know of.

167 Reuss (p. 24) enumerates 347 passages wherein the first edition of Erasmus differs from the Complutensian, forty-two of which were changed in his second edition. In fifteen places the first edition agrees with the Complutensian against the second (p. 30).

168 Besides the weighty insertion of 1 John v. 7, 8, Reuss (p. 32) gives us only seven changes in the third edition from the second: Mill’s other cases, he says, must be mere trifles.

169 Here again Reuss declares “paucissimas novas habet” (p. 36), and names only six.

170 “Non deserit quartam nisi duobus in locis: 1 Cor. xii. 2; Acts ix. 28” (Reuss, p. 37). Reuss had evidently not seen the first edition of the present work.

171 Multis vetustissimis exemplaribus collatis, adhibita etiam quorundam eruditissimorum hominum cura, Biblia (ut vulgo appellant) graece cuncta eleganter descripsi (Andreas Aesulanus Cardinali Aegidio).

172 This is Mill’s calculation, but Wetstein followed him over the ground, adding (especially in the Apocalypse) not a few variations of Aldus which Mill had overlooked, now and then correcting his predecessor’s errors (e.g. 2 Cor. xi. 1; Col. ii. 23), not without mistakes of his own (e.g. Luke xi. 34; Eph. vi. 22). Since Wetstein’s time no one seems to have gone carefully through the Aldine N. T., except Delitzsch in order to illustrate the Codex Reuchlini (1) in the Apocalypse. Reuss (p. 28) notes eleven places in which it agrees with the Complutensian against Erasmus; seven wherein it rejects both books.

173 The title-page runs εν λευκετια των παρησιων, παρα σιμωνι τω κολιναιω δεκεμβριου μηνος δευτερα φθινοντος, ετει απο της θεογονιας α φ λ δ. This book has no Preface, and the text does not contain 1 John v. 7, 8. It stands alone in reading ἀγγελία, 1 John i. 5. Reuss (p. 46), who praises Colinaeus highly, states that he deserts Erasmus’ third edition 113 times out of his own thousand, fifty-three of them to side with the Complutensian, and subjoins a list of fifty-two passages wherein he stands alone among early editors, for most of which he may have had manuscript authority.

174 Wordsworth, Old Latin Biblical Texts, I. xv.

175 Reuss (pp. 50, 51, 54) mentions only nine places wherein Stephen’s first edition does not agree either with the Complutensian or Erasmus; in the second edition four (or rather three) more; in the third nine, including the great erratum, 1 Pet. iii. 11. He further alleges that in the Apocalypse whatever improvements were introduced by Stephen came from the fourth edition of Erasmus, not from the Complutensian.

176 Mill states that Stephen’s citations of the Complutensian are 598, Marsh 578, of which forty-eight, or one in twelve, are false; but we have tried to be as exact as possible. Certainly some of Stephen’s inaccuracies are rather slight, viz. Acts ix. 6; xv. 29; xxv. 5; xxviii. 3; Eph. iv. 32; Col. iii. 20; Apoc. i. 12; ii. 1, 20, 24; iii. 2, 4, 7, 12; iv. 8; xv. 2: β’ seems to be put for α’ Matt. x. 25.

177 Viz. in the Gospels 81, Paul. 20, Act. Cath. 17, Apoc. 1 (ch. vii. 5): but for the Apocalypse the margin had only three authorities, α᾽, ιε᾽, ιϛ᾽ (ιϛ᾽ ending ch. xvii. 8), whose united readings Stephen rejects no less than fifty-four times. _See_, moreover, above, p. 154, note 3.

178 Here, again, my own collation represents Stephen’s first edition as differing from his third in 797 places, of which 372 only are real various readings, the rest relating to accents, or being mere errata. Of these 372 places, the third edition agrees in fifty-six places with π. or πάντες of its own margin, and in fifty-five with some of the authorities cited therein. Stephen no doubt knew of manuscript authority for many of his other changes, though some may be mere errata.

179 Wetstein (N. T., Prol., vol. i. p. 36) instances the readings of Cod. D (indicated as “quidam codex” by Beza in 1565) in Mark ix. 38; x. 50; Luke vii. 35. We may add that Beza in 1565 cites the evidence of one Stephanic manuscript for the omission of ὑμῶν, Matt. xxiii. 9; of two for κατεδίωξεν Mark i. 36; in later editions of two also in Luke xx. 4, and Acts xxii. 25; of three for ἑτέρῳ; Matt. xxi. 30, two of which would be Cod. D and Evan. 9 (Steph. ιβ᾽). In his dedication to Queen Elizabeth in 1565, Beza speaks plainly of an “exemplar ex Stephani nostri bibliotheca cum viginti quinque plus minus manuscriptis codicibus, et omnibus paenè impressis, ab Henrico Stephano ejus filio, et paternae sedulitatis haerede, quam diligentissimè collatum.”

180 But here again we must qualify previous statements. Reuss (p. 58) cites six instances wherein Stephen’s third and fourth editions differ (Matt. xxi. 7; xxiii. 13, 14; xxiv. 15; Luke xvii. 36; Col. i. 20; Apoc. iii. 12): to which list add Mark xiv. 21; xvi. 20; Luke i. 50; viii. 31; xii. 1; Acts xxvii. 13; 2 Cor. x. 6; Heb. vii. 1.

181 Professor Isaac H. Hall, who has the advantage of Dr. Scrivener in actually himself possessing all the ten editions of Beza, as he states in MS. in a copy of his “American Greek Testaments” kindly given to me, says, p. 60, note, that in the edition of 1556 the Greek does not occur, and that Beza’s first _Greek_ text was published in 1565. Beza must have reckoned his Latin amongst his editions when he spoke of his folio of 1565 as his second edition, and must generally have dated from 1556 as the beginning of his labours. The dates of the ten editions given above are extracted from Professor Hall’s list in Schaff’s “Companion to the Bible,” pp. 500-502.

182 Reuss says fairly enough (p. 85) that Beza was the true author of what is called the received text, from which the Elzevir of 1624 rarely departs. He used as his basis the fourth edition of Stephen, from which he departed in 1565, so far as Reuss has found, only twenty-five times, nine times to side with the Complutensian, four times with Erasmus, thrice with the two united; the other nine readings are new, whereof two (Acts xvii. 25; James v. 12) had been adopted by Colinaeus. The second edition of 1582 withdraws one of the peculiar readings of its predecessor, but adds fourteen more. The third edition (1588), so far as Reuss knows, departs from the second but five times, and the fourth (1598) from the third only twice, Matt. vi. 1 (δικαιοσύνην); Heb. x. 17 (add. τότε εἴρηκε), neither of which I can verify. These results, on Reuss’s system of investigation, can be only approximately true (_see_ p. 154, note), and do not include some changes silently introduced into Beza’s Latin version, as suggested in his Annotations.

183 Reuss (p. 109) states that out of his thousand select examples Elzevir 1624 differs from Beza’s smaller New Testament of 1565 in only eight readings, all of which may be found in some of Beza’s other editions (e.g. the small edition of 1580), except one misprint (Rom. vii. 2).

184 Οἱ δοῦλος is disputed by Hoskier (App. C. p. 18, n.), who says that he has seen besides his own copy of 1624 several which read οἱ δοῦλου. He had also inspected mine. “And although he says it reads δοῦλος, I read easily δοῦλοι. The type is rather faulty, that is all.” The point is not worth disputing.

185 “American Additions and Corrections,” p. 50.

186 Professor Hall states (Schaff’s “Companion,” p. 501) that Beza’s editions of 1588 and 1598 were the chief foundations of the Authorized Version of 1611. Archdeacon Palmer (Preface to Greek Testament with Revisers’ Readings, p. vii) refers chiefly to Stephen’s edition of 1550. Dr. Scrivener (to whom Archdeacon Palmer refers), Cambridge Greek Testament, Praef., p. vi, in taking the Elzevir edition of 1624 as the authority for the “Textus Receptus,” says that it rests upon Stephen’s 1550, and Beza’s 1565, 1582, 1589 (= 1588), and 1598 (especially the later editions, and particularly 1598, Authorized Edition of the British Bible, p. 60), besides also Erasmus, the Complutensian, and the Vulgate (Authorized Edition, p. 60). Dr. Scrivener adds in the passage just named that out of 252 passages the “Translators abide with Beza against Stephen in 113, with Stephen against Beza in fifty-nine, with the Complutensian, Erasmus, or the Vulgate against both Stephen and Beza in eighty.”

187 “The Authorized Edition of the English Bible (1611), its subsequent Reprints and Modern Representatives.” By F. H. A. Scrivener, M.A., D.C.L., LL.D., &c., Cambridge, University Press, 1884. Appendix E.

_ 188 See_ Miller’s “Textual Guide,” George Bell & Sons, 1885. Also Dr. Scrivener’s “Adversaria et Critica Sacra” (not yet published).—Postscript.

189 Reuss (p. 56) excepts Matt. ix. 17; 2 Tim. iv. 13; Philem. 6, where Walton prefers the Complutensian reading.

190 Nos. 2 and 3 had been partially used by Beza (American Additions, p. 50).

191 If Ussher lacked severe accuracy in collating his manuscripts, as well as skill in deciphering them, we have not to look far for the cause. In a Life prefixed to Ussher’s “Body of Divinity,” 1678, p. 11, we are told that “in the winter evenings he constantly spent two hours in comparing old MSS. of the Bible, Greek and Latin, taking with his own hand the _variae lectiones_ of each:” on which statement Dean Burgon (Letter in the _Guardian_, June 28, 1882) makes the pregnant comment, “Such work carried on at seventy or more by candlelight, is pretty sure to come to grief, especially when done with a heart-ache.”

192 “Sed, cum aliqui ex editoribus N. T. in analogiis discernendis nimis fortasse curiosi loca Parallela ad infinitum fere numerum auxerint, quorum alia parum definitae similitudinis, alia remotioris sunt argumenti quam quae servatis sanae interpretationis legibus possint adhiberi, satius habuimus Curcellaeum sequi, qui nec parcior est, nec nimis minutus in locis allegandis, nec dissimilia unquam aut prorsus ἀπροσδιόνυσα ad marginem locavit.”—Car. Oxon. (Bishop C. Lloyd) Monitum N. T. Oxonii, 1827.

193 1 John v. 7, 8 is included in brackets. Reuss (p. 130) thinks that the text follows Elzevir 1633 everywhere else but in Luke x. 22. Mill (N. T., Proleg. § 1397) says that it was printed “ad editiones priores Elzevirianas, typis Elzevirianis nitidissimis.”

194 “Stephani Curcellaei annotationes variantium lectionum, pro variantibus lectionibus non habendae, quia ille non notat codices, unde eas habeat, an ex manuscriptis, an vero ex impressis exemplaribus. Possunt etiam pro uno codice haberi.” Canon xiii. pp. 11, 69-70 of the N. T. by G. D. T. M. D. (_see_ below, p. 204).

195 But it goes with Elz. 1624 in Mark iv. 18; 2 Tim. i. 12; Apoc. xvi. 5, and sometimes prefers the readings of Stephen 1550, e.g. Mark i. 21; vi. 29, and notably Luke ii. 22 (αὐτῶν); Luke x. 22; Rom. vii. 2; Philem. 7. Peculiarities of this edition are Εἰ δὲ for Εἶτα Heb. xii. 9; συγκληρονόμοις 1 Pet. iii. 7. Wetstein’s text follows its erratum, Acts xiii. 29 ἐτέλησαν. Mill seems to say (N. T., Proleg. § 1409) that Fell’s text was taken from that of Curcellaeus.

196 Fell imputes the origin of various readings to causes generally recognized, adding one which does not seem very probable, that accidental slips once made were retained and propagated through a superstitious feeling of misplaced reverence, citing in illustration Apoc. xxii. 18, 19. He alleges also the well-known subscription of Irenaeus, preserved by Eusebius, which will best be considered hereafter; and remarks, with whatever truth, that contrary to the practice of the Jews and Muhammedans in regard to their sacred books, it was allowed “e vulgo quibusvis, calamo pariter et manu profanis, sacra ista [N. T.] tractare” (Praef. p. 4).

197 “Considerations on the Biblia Polyglotta,” 1659: to which Walton rejoined, sharply enough, in “The Considerator considered,” also in 1659.

198 Dr. Hort says that “his comprehensive examination of individual documents, seldom rising above the wilderness of multitudinous details, [is] yet full of sagacious observations” (Introd. p. 180).

199 As Mill’s text is sometimes reprinted in England as if it were quite identical with that commonly received, it is right to note the following passages wherein it does not coincide with Stephen’s of 1550, besides that it corrects his typographical errors: Matt. xx. 15; 22; xxiv. 15; Mark ix. 16; xi. 22; xv. 29; Luke vii. 12 _bis_; x. 6; xvii. 1; John viii. 4; 25; xiii. 30-31; xix. 7; Acts ii. 36; vii. 17; xiv. 8; Rom. xvi. 11; 1 Cor. iii. 15; x. 10; xv. 28; 2 Cor. vi. 16; Eph. iv. 25; Tit. ii. 10; 1 Pet. iii. 11; 21; iv. 8; 2 Pet. ii. 12; Apoc. ii. 5; xx. 4. Reuss (p. 149) tells us that Kuster’s edition recalls the Stephanic readings in Matt. xxiv. 15; Apoc. ii. 5.

200 Ellis, Bentleii Critica Sacra, Introductory Preface, p. xv.

201 Ellis, _ubi supra_, pp. xvii-xix. These _Proposals_ were also very properly reprinted by Tischendorf (N. T., Proleg. lxxxvii-xcvi, 7th edition), together with the specimen chapter (Apoc. xxii). The full title was to have been: “Ἡ ΚΑΙΝΗ ΔΙΑΘΗΚΗ Graece. Novum Testamentum Versionis Vulgatae, per stum Hieryonymum ad vetusta exemplaria Graeca castigatae et exactae. Utrumque ex antiquissimis Codd. MSS., cum Graecis tum Latinis, edidit Richardus Bentleius.”

202 This is all the more lamentable, inasmuch as Bentley was not accurate enough as a collator to make it unnecessary to follow him over the same ground. Dr. Westcott confirms my own experience in this respect when in a MS. note inserted by him on a blank leaf of Trin. Coll. B. XVII. 14, he states that “Bentley’s testimony, when he quotes a reading, may always be taken as true; but it is not so when he notes no variation in particular. On an average he omits _one-third_ of the variations of the MSS., without following, as far as I can discover, any law in the selection of readings.”

203 Bp. John Wordsworth would vindicate both Bentley and Walker from the suspicion of lightly taking up and lightly dropping so important a task. Walker, whom Bentley, as is said, called “Clarissimus Walker,” died on Nov. 9, 1741, at the age of forty-eight.—Wordsworth, Old Biblical Texts, I. xxv. p. 65. And for the Latin and Greek Texts collated by him wholly or partially, _see_ pp. 55-63.

204 He continued this work till after 1735. _See_ paper found by Dr. Ince at Christ Church, quoted by Bp. J. Wordsworth, Old Latin Biblical Texts, I. xxv. note 2.

205 Mr. Jebb (Life of Bentley, p. 164) imputes the failure of Bentley’s grand scheme partly to the worry of litigation which harassed him from 1729 to 1738; partly to a growing sense of complexity in the problem of the text, especially after he became better acquainted with the Vatican readings, i.e. about 1720 and 1729. Reuss (p. 172) ought never to have conditioned the ultimate success of such a man by the proviso “si consilio par fuerit perseverantia.”

206 “This thought has now so engaged me, and in a manner inslaved me, that _vae mihi_ unless I do it. Nothing but sickness (by the blessing of God) shall hinder me from prosecuting it to the end” (Bentley to Archbp. Wake, 1716: Ellis, _ubi supra_, p. xvi). A short article in the _Edinburgh Review_ for July, 1860, apparently from the pen of Tregelles, draws attention to “Nicolai Toinardi Harmonia Graeco-Latina,” Paris, 1707, fol. (“liber rarissimus,” Reuss, p. 167), who so far anticipates Bentley’s labours, that he forms a new Greek text by the aid of two Roman manuscripts (Cod. B being one of them) and of the Latin version.

207 Dr. Gregory says that though Mace’s edition had no accents or soft breathing, he anticipates most of the changes accepted by some critics of the present day.

208 I cannot help borrowing the language of Donaldson, used with reference to an entirely different department of study, in the opening of one of his earliest and by far his most enduring work: “It may be stated as a fact worthy of observation in the literary history of modern Europe, that generally, when one of our countrymen has made the first advance in any branch of knowledge, we have acquiesced in what he has done, and have left the further improvement of the subject to our neighbours on the continent. The man of genius always finds an utterance, for he is urged on by an irresistible impulse—a conviction that it is his duty and vocation to speak: but we too often want those who shall follow in his steps, clear up what he has left obscure, and complete his unfinished labours” (New Cratylus, p. 1). Dr. Gregory quotes against Dr. Scrivener, Mace (1729), Bowyer, a follower of Wetstein (1763), Harwood (1776), besides Whitby, Middleton, and Twells: but Dr. S. looked for greater names, and till Middleton, a more advancing study.

209 The full title is “’Ἡ καινὴ διαθήκη. Novum Testamentum Graecum ita adornatum ut Textus probatarum editionum medullam, Margo variantium lectionum in suas classes distributarum locorumque parellelorum delectum, Apparatus subjunctus criseos sacrae Millianae praesertim compendium limam supplementum ac fructum exhibeat, inserviente J. A. B.”

210 They consist of seven Augsburg codices (_Aug._ 1 = Evan. 83; _Aug._ 2 = Evan. 84; _Aug._ 3 = Evan. 85; _Aug._ 4 = Evst. 24; _Aug._ 5 = Paul. 54; _Aug._ 6 = Act. 46; _Aug._ 7 = Apoc. 80); _Poson._ = Evan. 86; extracts sent by Isel from three Basle copies (_Bas._ α = Evan. E; _Bas._ β = Evan. 2; _Bas._ γ = Evan. 1); _Hirsaug._ = Evan. 97; _Mosc._ = Evan. V; extracts sent by F. C. Gross. To these add Uffenbach’s three, _Uffen._ 2 or 1 = Paul. M; _Uffen._ 1 or 2 = Act. 45; _Uffen._ 3 = Evan. 101.

211 It is worth while to quote at length Bengel’s terse and vigorous statement of his principle: “Posset variarum lectionum ortus, per singulos codices, per paria codicum, per syzygias minores majoresque, per familias, tribus, nationesque illorum, investigari et repraesentari; et inde propinquitates discessionesque codicum ad schematismos quosdam reduci, et schematismorum aliquae concordantiae fieri; atque ita res tota per tabulam quandam quasi genealogicam oculis subjici, ad quam tabulam quaelibet varietas insignior cum agmine suorum codicum, ad convincendos etiam tardissimos dubitatores exigeretur. Magnam conjectanea nostra sylvam habent: sed manum de tabulâ, ne risuum periculo exponatur veritas. Bene est, quod praetergredi montem hunc, et planiore via pervenire datur ad codices discriminandos. Datur autem per hanc regulam aequissimam: Quo saepius non modo singuli codices, sed etiam syzygiae minores eorum vel majores, in aberrationes manifestas tendunt; eo levius ferunt testimonium in discrepantiis difficilioribus, eoque magis lectio ab eis deserta, tanquam genuina retineri debet” (N. T., Apparat. Crit., p. 387).

212 See a eulogistic yet discriminating discussion upon Bengel in _Bengel als Gelehrter, ein Bild für unsere Tage_, from the eminent pen of Dr. Nestle, which has been courteously sent to the editor through the Rev. H. J. White.

213 The opposition of Frey and his other adversaries delayed that _opus magnum_ for twenty years (N. T., Proleg., vol. i. p. 218).

214 We here reckon separately, as we believe is both usual and convenient, every distinct portion of the N. T. contained in a manuscript. Thus Codd. C and 69 Evan. will each count for four.

215 Errors of Wetstein’s text will be found in John xi. 31; Acts i. 26; xiii. 29 ἐτέλησαν, from the Oxford N. T. 1675, though Wetstein himself remarks this. He corrects a few obvious misprints of Elzevir 1633, but his note shows that he does not _intend_ to read τῷ in Mark vi. 29. The following seem to be deliberate variations from the Elzevir text: Matt. xiii. 15; xxi. 41; Mark xiv. 54; Luke ii. 22; xi. 12; xiii. 19; 1 Cor. i. 29; v. 11; xii. 23; xiv. 15; Phil. iii. 5; 1 Tim. iii. 2, 11 (yet not Tit. ii. 2); Philem. 7; 1 Pet. i. 3; iii. 7. All these deliberate variations are found in Von Mastricht’s edition of 1735, which seems to have been used by Wetstein as the basis of his text; and in all of them (except Matt. xxi. 41; Luke xi. 12, and Phil. iii. 5) Fell’s text agrees with Wetstein’s. In Matt. xiii. 15; Mark xiv. 54; 1 Cor. i. 29; v. 11; xii. 23; xiv. 15; Phil. iii. 5; 1 Pet. iii. 7, the Elzevir editions vary. (American Additions and Corrections, p. 51.) He spells ναζαρέτ uniformly, except in John i. 46, 47. Reuss (p. 183) adds nine changes made by Wetstein in the text for critical reasons: Matt. viii. 28; Luke xi. 2; John vii. 53-viii. 11; Acts v. 36; xx. 28; 1 Tim. iii. 16 (δ); Apoc. iii. 2; x. 4; xviii. 17.

216 One other specimen of Matthaei’s critical skill will suffice: he is speaking of his Cod. H, which is our Evst. 50. “Hic Codex scriptus est literis quadratis, estque eorum omnium, qui adhuc in Europa innotuerunt et vetustissimus et praestantissimus. Insanus quidem fuerit, qui cum hoc aut Cod. V [p. 144] comparare, aut aequiparare voluerit Codd. Alexandr. Clar. Germ. Boern. Cant. [Evan. AD, Paul. ADEG], qui sine ullo dubio pessimè ex scholiis et Versione Latinâ Vulgatâ interpolati sunt” (N. T., Tom. ix. p. 254).

217 In using Matthaei’s N. T. the following index of manuscripts first collated by him will be found useful: a = Evan. 259, Act. 98 (a 1), Paul. 113 (a or a 2), Apost. 82 (a 3): B = Evst. 47: b = Apost. 13: c = Act. 99, Paul. 114, Evst. 48: d = Evan. 237, Act. 100, Paul. 115: e = Evan. 238, Apost. 14: f = Act. 101, Paul. 116, Evst. 49: g = Evan. 239, Act. 102, Paul. 117: H = Evst. 50: h = Act. 103, Paul. 118: i = Evan. 240, Paul. 119: k = Evan. 241, Act. 104, Paul. 120, Apoc. 47: l = Evan. 242, Act. 105, Paul. 121, Apoc. 48: m = Evan. 243, Act. 106, Paul. 122: n = Evan. 244, Paul. 123: o = Evan. 245, Apoc. 49: p = Evan. 246, Apoc. 50: q = Evan. 247, Paul. 124: r = Evan. 248, also Apoc. 502, Apoc. 90: s = Evan. 249, Paul. 76: t = Apoc. 32, Evst. 51: tz = Apost. 15: V = V: v = Evan. 250, Apost. 5: x = Evan. 251, Act. 69, Paul. 74, Apoc. 30 (from Knittel); z = Evan. 252: 10 = Evan. 253: 11 = Evan. 254: 12 = Evan. 255: 14 = Evan. 256: 15 = O, 16 = Evst. 56, Apost. 20: 17 = Evan. 258: 18 = Evan. 99: 19 = Evst. 57: 20 = Evan. 89: ξ = Evst. 52, Apost. 16: χ = Evst. 53, Apost. 17: ψ = Evst. 54, Apost. 18: ω = Evst. 55, Apost. 19: Frag. Vet. = part of H: Gpaul. It should be noted, that in several of these cases different MSS. are included under one letter: e.g. c = Evst. 48 is a different MS. from c = Act. 99.

218 The copies of Chrysostom’s homilies on the Gospels freshly collated by this editor are noted 1, 2, 3, 4, 5, 6, 7, 8, 9, 13, α, β, γ, δ, ε, ζ, η, θ, λ, μ, π, ρ, φ: those on St. Paul’s Epistles are noted 1, 2, 3, 4, 5, 6, 7, 8, 9, α, β.

219 Reuss (p. 207) calculates that, besides misprints, Matthaei’s second and very inferior edition differs in text from his first in but twenty-four places, none of them being in the Gospels.

220 “Textui ad Millianum expresso” says Reuss (p. 151), which is not quite the same thing: _see_ p. 203, note 2.

221 “Conscius sum mihi, me omnem et diligentiam et intentionem adhibuisse, ut haec editio quam emendatissima in manus eruditorum perveniret, utque in hoc opere, in quo ingenio non fuit locus, curae testimonium promererem; nulla tamen mihi est fiducia, me omnia, quae exigi possint, peregisse. Vix enim potest esse ulla tam perpetua legentis intentio, quae non obtutu continuo fatigetur, praesertim in tali genere, quod tam multis, saepe parvis, observationibus constat.” (Lecturis Editor, p. v. 1788.) Well could I testify to the truth of these last words!

222 “Symbolae Criticae ad supplendas et corrigendas variarum N. T. lectionum Collectiones. Accedit multorum N. T. Codicum Graecorum descriptio et examen.”

223 Yet Tischendorf (N. T., Proleg., p. xcvii, 7th ed.) states that he only added two readings (Mark vi. 2, 4) to those given by Wetstein for Cod. C. From Cod. D too he seems to have taken only one reading, and that erroneously, επηγειραν, Acts xiv. 2.

224 In the London edition of 1809 ἄλλοι is printed for the first οὗτοί, Mark iv. 18. Griesbach also omits καί in 2 Pet. i. 15: no manuscript except Cod. 182 (ascr) is known to do so.

225 “Dissertatio critica de Codicibus quatuor Evangeliorum Origenianis,” Halae, 1771: “Curae in historiam textus Graeci epistolarum Paulinarum,” Jenae, 1777.

226 “Commentarius Criticus in textum Gr. N. T.,” Part i. 1798; Part ii. 1811.

227 The following specimen of a reading, _possessing no internal excellence_, preferred or favoured by Griesbach on the slightest evidence, will serve to illustrate the dangerous tendency of his system, had it been consistently acted upon throughout. In Matt.