1076. THEOLOGICAL FRAGMENTS 7

the question which is the earlier critics are still divided. Besides these two, however, for chaps. vi. 9—xiii. 8, there is found in three cursives a third Greek recension, which may be said to occupy an intermediate position between BA and ℵ; from chap. viii onwards this text agrees closely with the Syriac.

Now the present fragment from the second chapter is clearly to be distinguished from BA on the one hand and ℵ on the other; the obvious question then arises, Can it belong to the third recension partially preserved in the cursives (C)? This view appears to be highly probable. The relation of 1076 and C to BA and ℵ respectively is closely similar. Both 1076 and C belong to the ℵ type, but are more concise, while at the same time they occasionally add points of their own. In ii. 3, for instance, the elaborate forms of address in ℵ disappear in 1076, just as in vi. 11 they are omitted in C (cf. note on l. 15). On the other hand, the insertion of "and he lost all his possessions" original: "καὶ ἀπώλεσεν πάντα τὰ ὑπάρχοντα αὐτοῦ" in 1076 32-5 (ii. 8) has parallels in C, e.g. in vi. 15 the addition of "from the unclean spirit" original: "ἀπὸ τοῦ πνεύματος τοῦ ἀκαθάρτου" and "he loves her" original: "φιλεῖ αὐτήν". These like characteristics strongly suggest a common origin; and corroborative evidence for this theory is supplied by the Old Latin version. A peculiarity of that version is that while generally following ℵ it occasionally reflects C. Thus in vi. 15 hoc daemonium this demon corresponds to "from the unclean spirit" and diligit eam he loves her to "he loves her." It therefore seems highly significant that just in the same way in ii. 8 the Old Latin alone of the versions reproduces the phrase "and he lost all his possessions" of 1076 with et perdidit substantiam suam and he lost his substance.

The fragment consists of the lower part of a vellum leaf, on which the text was written in two columns in carefully formed, large round uncials, which may date from the sixth century. Hands of a similar type on papyrus are shown e.g. in Amh. II, Plate 24. One side of the leaf has shrivelled, with the consequence that the letters are here considerably reduced from their original size. The ink is of the brown colour commonly found in the Byzantine period. Rulings were made in the usual way with a hard point, which has left a dark mark upon the surface. Punctuation was effected by dots in the medial position, accompanied by a short blank space; in l. 30 the dot was omitted or has disappeared. A new section is indicated by a marginal sign at l. 11. Small curved marks resembling circumflex accents do duty for rough breathings (ll. 5, 22) as well as diaereses (ll. 34, 36).