[1] Conybeare, F.C., ed., The Dialogues of Athanasius and Zacchaeus, and of Timothy and Aquila (Anecdota Oxoniensia, Classical Ser., Pt. 8; Oxford: Clarendon Press, 1898), pp. xix–xxiiGoogle Scholar; more briefly in his ‘Three Early Doctrinal Modifications of the Text of the Gospels’, HibJ 1 (1902–1903), pp. 96–102Google Scholar. The argument was seized upon by the radical, such as Schmeidel, P. W., ‘Mary’, Encyclopaedia Biblica, ed. Cheyne, T.K. (London: Black, 1902) 3, pp. 2961–3.Google Scholar The Sinaitic variant (Joseph; and Joseph, to whom was betrothed a virgin, Mary, begat [masc.] Jesus, who is called Christ) has spawned much discussion, which I have examined briefly in ‘Some Doctrinal Variants in Matthew 1 and Luke 2, and the Authority of the Neutral Text’, CBQ 42 (1980), pp. 52–72Google Scholar, pp. 63–5 for the Sinaitic reading and pp. 65–6 for the readings in DTA.

[2] Burkitt, F.C., ed., Evangelion da-Mepharreshe, The Curetonian Version of the Four Gospels (Cambridge: University Press, 1904) 2, pp. 262–6.Google Scholar See also Metzger, B. M., A Textual Commentary on the Greek New Testament (London/New York: United Bible Societies, 1971), pp. 2–7.Google Scholar

[3] The editions of Nestle and of the British and Foreigh Bible Society, for example, list three variants from DTA.

[4] Conybeare's edition, pp. xxiv, xxv. Textual research of the last eighty years has not been kind to Conybeare's speculation that many of the citations in DTA reach back before the canonical stage of the gospels. To be fair to him, though, it should be remembered that knowledge then of the Diatessaron and of the Caesarean witnesses, sources of the utmost importance for identifying the text type of DTA, was almost nonexistent.

[5] Metzger, B. M., ‘The Text of Matthew 1.16’, Studies in New Testament and Early Christian Literature: Essays in Honor of Allen P. Wikgren (SNT 33; Leiden: Brill, 1972), pp. 16–24.CrossRefGoogle Scholar

[6] The author knew a New Testament with four Gospels, Acts, the Catholic and Pauline epistles, but without the Apocalypse. See DTA 66.

[7] DTA pp. 91–2.

[8] See particularly Conybeare's introduction to DTA pp. xxv–lvii (liv and lvi for Egypt); also Williams, A. Lukyn, Adversus Judaeos: A Bird's-Eye View of Christian ‘Apologiae’ until the Renaissance (Cambridge: University Press, 1935), pp. 67–78.Google Scholar

[9] See Robert Robertson, G., ‘The Dialogue of Timothy and Aquila: The Need for a New Edition’, Vigiliae Christianae 32 (1978), pp. 276–88.CrossRefGoogle Scholar He has kindly supplied further ms variants by letter for Matt. 1. 2–18; 8. 29; 21. 16, 33, 35, 38 and 41. For a collation of variants in DTA ms P and some variants from ms E, see Goodspeed, Edgar J., ‘The Dialogue of Timothy and Aquila: Two Unpublished Manuscripts’, JBL 24 (1905), pp. 58–78.Google Scholar

[10] Fee makes, Gordon D. an eloquent plea for this practice in ‘The Text of John in Origen and Cyril of Alexandria: A Contribution to Methodology in the Recovery and Analysis of Patristic Citations’, Biblica 52 (1971), pp. 357–94.Google Scholar

[11] Generous assistance from the University of British Columbia Humanities and Social Sciences Grant Fund facilitated the purchase of microfilms of New Testament manuscripts.

[12] The term ‘Caesarean’ is as problematic as ‘Western’, the latter group including sources as heterogeneous as the Old Latin and Old Syriac. As early as 1945, Metzger, B. M. (‘The Caesarean Text of the Gospels’, JBL 64 [1945], pp. 457–89Google Scholar, reprinted as ch. 2 of his Chapters in the History of New Testament Textual Criticism [Grand Rapids, Mich.: Eerdmans, 1963], p. 67) wrote that ‘it must be acknowledged that at present the Caesarean text is disintegrating. … it is no longer possible to gather all these several families and individual manuscripts under one vinculum such as the Caesarean text. The evidence of p⁴⁵ clearly demonstrates that henceforth scholars must speak of a pre-Caesarean text as differentiated from the Caesarean text proper.’ Despite this cautionary statement, the term ‘Caesarean’ conveniently isolates an important group of non-Neutral and non-Byzantine witnesses. For this and the other classifications, the grouping of Metzger, Textual Commentary, pp. xxix-xxx, is generally followed.

[13] See Eldridge, Lawrence A., The Gospel Text of Epiphanius of Salamis (Studies and Documents 41; Salt Lake City: Univ. of Utah Press, 1969).Google Scholar

[14] The variety within the Caesarean group is seen very clearly in the charts to be found in Kirsopp Lake, Blake, R. P. and New, Silva, ‘The Caesarean Text of the Gospel of Mark’, HTR 21 (1928), pp. 207–404, esp. pp. 216–46.Google Scholar

[15] Of the 21 distinctive Caesarean readings in DTA, almost a third (30, 31, 36, 41, 63, 65) are harmonizations. The significance of such accommodations in the Caesarean text has provided matter for controversy. See Metzger's summary in ‘The Caesarean Text of the Gospels’, p. 67. A collection of variants convenient for studying the nature of harmonization in this text type appears in Ayuso, Teófilo, ‘Texto Cesariense o Precesariense? Su realidad y su trascendencia en la critica textual del Nuevo Testamento’, Biblica 16 (1935), pp. 369–415Google Scholar. The non-canonical arrangement of material that Conybeare noted may indicate that the author cited an early harmony unrelated to Tatian's, but no traces of such a document seem to have survived: see the lists of Louis Leloir, Le témoignage d'Éphrem sur le Diatessaron (CSCO vol. 227, subsidia tome 19; Louvain, 1962), pp. 1–11. Similar harmonizations have been discovered in the major Western codex, D: see Vogels, Heinrich J., Die Harmonistik im Evangelientext des Codex Cantabrigiensis (TU 36:la; Leipzig: Hinrichs, 1910).Google Scholar

[16] The bilingual codex D is well known for its assimilation of Old Latin readings in the Greek text. For brief comments, with further bibliography, see Metzger, , The Text of the New Testament, 2nd ed. (Oxford: Clarendon Press, 1968), pp. 49–51Google Scholar; and Kenyon, F. G., The Text of the Greek Bible, 3rd ed., rev. Adams, A. W. (London: Duckworth, 1975), pp. 89–95Google Scholar. For Old Latin readings in the Greek codex W, see Sanders, Henry A., The New Testament Manuscripts in the Freer Collection, Part I, The Washington Manuscript of the Four Gospels (Univ. of Mich. Studies, Humanistic ser., 9; New York: Macmillan, 1912), pp. 64–9, 73.CrossRefGoogle Scholar