135. Here we have two short readings of three words each (α,β), differing only by the preposition compounded with the verb and by the presence or absence, of the last letter, having therefore a strong prima facie appearance of being derived the one from the other. The documents attesting α are four uncials (two of them our two oldest), three cursives, and at least three versions in different languages, one of them made late in Cent. IV, one early in Cent. V, and the third of age treated as not yet determined, but at least not later than Cent. III. The Vulgate Syriac is on the whole a supporter of α, as it reads προῆλθον and has but one clause: its ending may be due either to modified reduplication of the last word of α or, more probably, to conflation with the last word of β. For β (and the readings evidently derived from it) we have an uncial of Cent. VI, two cursives, and three Old Latin MSS. No true Old Latin MS is in any way favourable to α or δ against β: two, e k, which contain other parts of this Gospel, are absent; as are also the Thebaic and Old Syriac and Jerusalem Syriac versions. The longer reading δ, which is that of the Received Text, is supported by eleven uncials, one of them of Cent. V (or possibly IV) and the rest not earlier than Cent. VIII; all cursives except five; two Latin MSS belonging approximately to the Italian revision, which cannot be younger and is probably not older than Cent. IV; and two versions unquestionably later than Cent. IV.
136. If now we compare the three readings with reference to Transcriptional Probability, it is evident that either δ is conflate from α and β, or α and β are independent simplifications of δ; for the similarity αὐτοῦ and αὐτούς, combined with the relative dissimilarity of both to πρὸς αὐτόν, shews that δ can hardly have been a pas-
