¹ Especially by Rendel Harris, F. C. Burkitt, Corssen, Blass, and Zahn.

² This began with the publication of the palaeographically Calabrian group of minuscules, 13, 64,124,346, by Ferrar (hence called the Ferrar group) and Abbott, 1877, and the investigation of the group by Martin in France, 1885, and Rendel Harris in England, 1893. It was carried further by W. Bousset, “Textkritische Studien,” 1894, and by K. Lake, “Codex 1 of the Gospels and its Allies,” 1902.

³ See Schmiedel, P. W., ‘Der Ertrag der Arbeit Hermann v. Soden's am Text des Neuen Testaments’ in Theologische Blätter, 1922Google Scholar, No. 10.

⁴ Formerly, though we talked about this text, we knew very little about it. In practice the Antiochian text meant the editions of Stephanus and Elzevir. It was known that in many cases these editions did not really represent the Greek mediaeval text, but there was no clearness on the subject; nor could any clearness ever have been attained without a methodical investigation of almost all existing MSS. This investigation von Soden made. The result is that we now know that the K-text is found in at least three forms, K^l, K^x, K^r, and it would not be difficult to take a few MSS., typical of these three forms, and from them produce a usable edition of the K-text. Complete accuracy would be unattainable without immense labor, but an edition that would be of enormous help to collators and investigators of the text could be made with relative ease. This addition to our knowledge serves to define the problem as to the K-text, with which von Soden left us. What is the date of the K-text? I believe that there is no proof of a pure K-text before the sixth century. But there are fifth century MSS., such as A (which in the gospels is mainly of the K-type), which contain K-readings. Are K-readings the proof that a K-text existed from which they were taken or are they the material from which the K-text was made? That is the real problem, although unfortunately von Soden did not see it and assumed that K-readings imply a K-text, equal, as he thought, in age to the H-text and the I-text. How is it possible to solve this riddle? Not from general probabilities; but by investigating the text of the early Greek fathers, which may hold the key. At present we have no good edition of the texts of Athanasius, the Cappadocians, Chrysostom, or Cyril. Von Soden assumed that we know much more about these texts than we do. Possibly it may not be worth while to do so much in order to learn so little, but patristic evidence is the key to the problem of the K-text, and a key which can be gained by simple methods involving long, but not difficult, work. A small percentage of the energy wasted on doctors’ dissertations which begin nowhere and return to their starting-point would soon accomplish real results in this field.

⁵ This codex is described by Gregory, with bibliographical references, in Textkritik, I, 1900, p. 257, his information having been derived from Oskar von Gebhardt.

⁶ Or is it with an early K-text? I think not, but the character of the non-western element in D has never been fully studied. That D is conflate cannot be doubted by any one who has analyzed it.

⁷ I omit some MSS. which on von Soden's showing ought to be included, because their text is not published and I cannot reconstruct it from his apparatus.

⁸ I also included cod. 22, but its affinity is much less clear, at least in Mark i, and I have therefore neglected it in this article. Von Soden reckons it among the 1^η MSS.; this may be right in Matthew, but the question requires further investigation.

⁹ 565 reads και παντες εβαπτιζοντο.

¹⁰ εν τῃ ερημῳ ℵBDL 33 boh latt. The evidence of fam¹³ is divided. 69 124 read εκει, 13 346 543 788 826 828 read εν τῃ ερημῳ. εκει sine addit. is found elsewhere only in syr^sin arm and a few late Greek MSS. of which KII are the prominent ones.

¹¹ The omission of ο is the K¹ reading, against all other groups.

¹² A unique reading, probably an accident.

¹³ The text is confused: θ reads και εκαλεσεν αυτους και ευθεως αϕεντες 565 και αϕεντες, fam¹³ reads και ευθυς εκαλεσεν αυτους και ευθεως αϕεντες. The reading of 565 may be either that of the family or due to homoeoteleuton; in the latter case it seems based on the text of fam¹³. The reading of θ is an accident, or an attempt to correct to the text of ς by taking out the wrong ευθυς.

¹⁴ So von Soden, but not Belsheim or Hoskier.

¹⁵ Fam¹⁸ has ευθυς πανταχου εις ολην. This seems to be a conflation of ευθυς (found by itself in AD etc ς) and πανταχου (found by itself in no Greek MS. but in be q boh). But the apparent conflation is found also in BCL.

¹⁶ The reading of θ may be accidental; but cf. the reading of D εξελθων εκ της συναγωγης, omitting ευθυς.

¹⁷ May be accidental, but it is the reading of ℵBD.

¹⁸ Only in D (which has a widely variant text) and in the Latin; perhaps a reminiscence of Lk. 4, 41.

¹⁹ So B, but according to von Soden no other MS.

²⁰ Probably accidental; no other support.

²¹ Note that B has the conflation κυριε οτι εαν. Also von Soden's statements seem inconsistent with Hoskier and the text of θ, nor is the text of 565 quite certain.

²² και ειπε 28.

²³ I have been as careful as time would allow, but I make no claim to complete accuracy in this specimen.

²⁴ And of these one (γεργεσηνων) in Mark vi is not an uncommon reading, and, being found in L Δ boh, may be late Neutral (or “Alexandrian,” in the sense of WH).

²⁵ Матеріалы по Археологіи Кавказа, издаваемые Графинеӥ Уваровоӥ, Выпускъ Х, Москва, 1907; see Beermann, G. and Gregory, C. B., Die Koridethi Evangelien, Leipzig, 1913, pp. 523–524Google Scholar.

²⁶ Plate I.

²⁷ Plates II-L.

²⁸ Plates II-XI.

²⁹ Dr. Blake examined the MS. repeatedly during his stay in Tiflis (July 1918- May 1920).

³⁰ Beermann and Gregory, pp. 599 f.

³¹ Gregory, though noting this point (cf. Beermann and Gregory, pp. 600, 607), does not seem to grasp its importance.

³² Facsimile, plate I, col. 2, 1. 13.

³³ Ibid., passim.

³⁴ Especially M.

³⁵ A convenient list of classified errata is contained in von Soden's remarks, pp. VIII-X of Russian edition.

³⁶ Beermann and Gregory had inklings of this, but do not carry out their train of reasoning to the end (p. 607).

³⁷ P. Franchi de’ Cavalieri and H. Lietzmann, Specimina Codicum Vaticanorum Graecorum, Bonn, 1910, plate 3.

³⁸ Ibid., plate 6.

³⁹ Beermann and Gregory, pp. 614–615, 625.

⁴⁰ I doubt if this be true to the extent that Gregory thinks (Beermann and Gregory, p. 607).

⁴¹ Beermann and Gregory, pp. 569–581.

⁴² Ibid., p. 577.

⁴³ Among other things, the coarseness of the material speaks against this.

⁴⁴ Beerman practically admits this (p. 580).

⁴⁵ H. Я. Mappъ, Грузинскіл Приписки Коридетскаго Евангелія. ИAH, 1911, стр. 211–240.

⁴⁶ Beermann and Gregory, pp. 583–584.

⁴⁷ See O. von Lemm, ИAH 1911, стр. 458–459.

⁴⁸ On the stay-at-homeness of Coptic, cf. Peeters, P., Traductions et traducteurs dans l'hagiographie orientale, Acta Bollandiana 40, 1922, p. 246Google Scholar.

⁴⁹ Lemm, O. von, Zur Bekehrung der Iberer zum Christentum, Kleine koptische Studien LX (ИAH 10 [1899], 403 ff.)Google Scholar: Iberica (Записки Имп. Акад. Наукъ, фил. ист. кл. Cep. viii, Томъ. 7, No. 6, 1906).

⁵⁰ A. A. Пагарели, Памятники грузинскоӥ старины въ Святоӥ Землѣ и на Синаѣ. Сборникъ Имп. Палест. Общества, Выпускъ. 10. Спб;. 1888 Г.

⁵¹ Ibid., pp. 27 f.

⁵² Procopius, de aed. 5, 9. 6–7 (III, 2. 164, 16–17 Haurey).

⁵³ Кіевъ 1895. This reference is taken from К. Кекелидзе, Іерусалимскіӥ Канонаръ vii вѣка, Тифлисъ, 1912, стр. 34–35.

⁵⁴ Кекелидзе, l.c., p. 33, cf. his Древнегрузинскіӥ Архіератиконъ, Тифлихъ, 1912, стр. ix–xiv.

⁵⁵ Кекелидзе. Iep. Канонарь, p. 1, note 2.

⁵⁶ H. Марръ, Житіе св. Гриторія Хандзт' ійскаго, Спб. 1911 (TP VII), text 12, 50–53, translation 97, 47–50.

⁵⁷ See (Peeters, P.), Acta Bollandiana, 32, 1913, pp. 236 f.CrossRefGoogle Scholar (St. Ilarion d'lbérie).

⁵⁸ See Peeters, P., Acta Bollandiana, 40, 1922, pp. 282–283Google Scholar.

⁵⁹ It was given to Czar Nicholas in 1829 by the Metropolitan Sylvester, who states that tradition connects it with the Empress Theodora.

⁶⁰ It runs, according to Belsheim (p. 5): εγραϕη και αντεβληθη ομοιως εκ των ἰεροσολυμων παλαιων αντιγραϕον. Doubtless it is accentuated, but Belsheim always omits accents.