² For earlier editions, before it was realised that the fragments were from the same codex as P⁶⁴, cf. Haelst, van, Catalogue, no. 336 (1).Google Scholar

³ Cavallo, G., Ricerche sulla Maiuscola Biblica (Studi e Testi di Papirologia, 2; Firenze: Le Monnier, 1967).Google Scholar

⁴ This is not strictly a text-division but is intended to draw attention to the list of the Apostles. Merell does not print 11.20–5 of the column, which should be restored as follows: 20 к[αι εкλεξαμενоς] [απ αυτων. τβ. оνς] [кαι απоστоλоυς] [ωνоμασεν: σνομωνα] ο[кαωνоμασεν] 25 [πετρоν кαν ανδρεαν] The omicron of o[v in 1. 24, The omicron of o[v in 1.24, projecting into the margin, is clearly visible in the plate here published.

⁵ On the absence of pagination cf. Turner, E. G., The Typology of the Early Codex (University of Pennsylvania Press, 1977) 74–5.CrossRefGoogle Scholar

⁶ Set-offs can be seen in various places where the leaves have been stuck together, e.g. in P⁴, Fragment B, col. 3, where portions of two lines of writing can be seen in the top left-hand corner, reading ΠENAY and NONTHNX, in reverse script. These are impressions from the first two lines of Fragment D, col. 4, which read: ετπεν αντω εкτει/νоν Хην ιερρα

⁷ On the single-quire codex cf. Turner, E. G., Typology, 55–60.Google Scholar

⁸ Turner, E. G., Typology, 58.Google Scholar

⁹ See Turner, E. G., Typology, 73–4.Google Scholar

¹⁰ Aland, K., Studien zur Überlieferung des neuen Testaments und seines Textes (Berlin: de Gruyter, 1967) 108Google Scholar, mentions ‘ν˚ mit bisher unidentifizierten Buchstabenspuren in 2 Spalten’ but I myself have been unable to detect any traces of writing in two columns.

¹¹ Roberts, C. H. and Skeat, T. C., The Birth of the Codex (Oxford, 1983).Google Scholar

¹² Lists of codices written with two columns to the page are given by Turner, E. G., Typology, 36.Google Scholar

¹³ Ordinary tracing paper is useless for this purpose, as it is too opaque. Shops dealing in drawing-office requisites can supply sheets of clear plastic which can be written on with a special pen.

¹⁴ It must be stressed that the lines in the diagram do not represent lines actually drawn on the papyrus, but serve merely to indicate the position of the lines of writing.

¹⁵ Cf. Johnson, W. A., ‘Column Lay-out in Oxyrhynchus Literary Papyri’, ZPE 96 (1993) 211–15.Google Scholar

¹⁶ For such markings on Greek papyri cf. Turner, E. G., Greek Manuscripts of the Ancient World², pp. 4–5Google Scholar. For the use of guide-lines in Demotic papyri of the Roman period see the extensive study by Tait, W. J., ‘Guide-lines and Borders in Demotic Papyri’, in Papyrus: Structure and Usage (British Museum Occasional Paper 60, 1986) 63–9.Google Scholar In the present codex it is very unlikely that the lines would have been ruled right across the column, since subsequent erasure, after the text had been written, would have been very difficult and time-consuming Probably, therefore, as suggested in the text, only the beginnings of the lines were marked out.

¹⁷ (Jean) Vincent Scheil, O.P., was born 10 June 1858. He studied Egyptology in Paris under Maspero and joined the Mission Archéologique in Cairo in 1883. He subsequently made a career as an Assyriologist. He died in Paris 21 Sept. 1940, cf. Who was Who in Egyptology (ed. Dawson, W. R. and Uphill, E. P.; 2nd ed.; London: Egypt Exploration Society, 1972) 263.Google Scholar

¹⁸ Merell gives the location of the fragments in the Bibliothèque nationale as ‘ñ Gr. 1120, suppl. 2’, van Haelst simply as ‘Suppl. gr. 1120’. ‘Suppl. grec 1120’, however, is the press-mark of the Philo codex. According to Thiede, C. P. in Tyndale Bulletin, 46 (1995) 55Google Scholar, the fragments are now kept in ‘Box 5’, not ‘Box 2’, under the number Suppl. grec. 1120.

¹⁹ Thirty years ago, Kurt Aland invited Colin Roberts and myself to express opinions on the dating of the registered New Testament papyri, then numbering 76, on the basis of photographs which he provided. These opinions, which Aland then published in Studien, 103–6, were, of course, our general impressions, since there could be no question of undertaking detailed research in individual cases. Our agreed dates were, for P⁴, ‘3rd cent.’, and for P^64/67, ‘um 200’, the possibility of their coming from the same manuscript being left open. I cannot now recall why we gave for p^64/67 a slightly later date than that assigned by Roberts in his edition of P⁶⁴ and his note on P⁶⁷. Possibly the obvious similarity of script of both of these with that of P⁴ may have influenced us towards keeping the dates fairly close together. Incidentally, our date for P⁷⁵ was 3rd cent.

I must add at this point that the entry regarding P⁴ in Aland's, Repertorium, where P⁴ appears as NT 4 on pp. 219–20Google Scholar, is misleading because a number of the references given relate to the Philo codex and not to P⁴. Thus, support for a 4th century date for P⁴ includes ‘Grenfell/Hunt, p. 16’, where the reference must be to vol. IX of the Oxyrhynchus Papyri, in which some fragments of Philo are published, the editor remarking (p. 16) that Scheil's date of 6th century for the Coptos codex is impossible, and should be 3rd century. Similarly, the reference ‘Kenyon, p. 145’ is to Kenyon's, The Palaeography of Greek Papyri, on p. 145Google Scholar of which he mentions the Coptos codex, suggesting a date of ‘? 3rd cent’. The references ‘Cohn, p. XLII’ and ‘Wilcken, p. XLII’ both refer to Vol. I of the Cohn–Wendland edition of Philo in which the Coptos codex is discussed, with Wilcken's comment that there is no reason to date it so late as the 6th century. None of the foregoing make any mention of P⁴. Conversely, the entry makes no mention of the fact that Kenyon, Collart and Dain had unanimously agreed on a 4th century date for P⁴.

²⁰ Some examples may be quoted here. The well-known vellum fragment of Demosthenes, de falsa legatione (Pack² 293), usually dated 2nd century, has every other line ruled. The Berlin leaf of Euripides, Cretans, also dated 2nd century (P. Berol. 13217 = BKT V ii 73) was certainly ruled. Both sides are reproduced in BKT, and as the vellum has become partially translucent it can be seen that the lines on the two sides match each other completely. As regards Christian manuscripts, no vellum manuscript of the Gospels, with one exception, is dated earlier than the 4th century. The exception is 0171, dated circ. 300. This comprises two fragments, one from Matthew (in Berlin) and one from Luke (in Florence), and tracings from the latter show that the lines on the two sides match exactly; the manuscript must therefore have been ruled, as no doubt are the later examples.