122 Timo Flink ἐπ᾽ ἐσχάτων τῶν χρονῶν 442 // 665 // 93 2344 // al K:Bmss ἐπ᾽ ἐσχάτων τοῦ χρόνου – // 621 // 1735 // L596 The oldest manuscript (P72) supports ἐπ᾽ ἐσχάτου χρόνου111. It is found in Egypt and Syria with P72B pair as noteworthy supportive evidence that puts the reading to the second century. However, the amount of witnesses is small, seven in number, in comparison to other possibilities (though this does not exclude it as a good candidate for the initial text). The “best” manuscript (codex 81) reads ἐπ᾽ ἐσχάτου τῶν χρονῶν. This variant reading is also found in Egypt and Syria with the support of the Coptic tradition. It is probable a second century reading via K:S. The third reading ἐπ᾽ ἐσχάτου τοῦ χρόνου is found in the same geographical area, but it is not found in witnesses that would absolutely put it to the second century, although this is possible. Variant readings ἐπ᾽ ἐσχατῷ τοῦ χρόνου, ἐπ᾽ ἐσχάτων τοῦ χρόνου and ἐπ᾽ ἐσχάτων τῶν χρονῶν are poorly attested and should be taken as another set of secondary readings. I agree with Wachtel and Wasserman that the Majority text reading ἐν ἐσχάτῷ χρονῷ is a later normalisation of the text and should be dropped from consideration112. This leaves three readings, (1) ἐπ᾽ ἐσχάτου χρόνου, (2) ἐπ᾽ ἐσχάτου τοῦ χρόνου, and (3) ἐπ᾽ ἐσχάτου τῶν χρονῶν, which enjoy the best external evidence and can content for “originality”. The decision needs to be made on internal grounds, because the external evi- dence is not conclusive between these three variant readings. However, the articular forms are better attested, so the external evidence slightly favours the inclusion of the article. Thus, it is conceivable that the difference between the exclusion and inclusion of the article goes back to the second century. Geographically the readings are found in the same areas. There are no umlaut markings in codex B for this textual variation unit. The omission of the article has the secondary support of the Majority text, while τῶν gives secondary support for τοῦ (and vice versa)113. The question becomes whether to exclude or to include the article, and if it is included, whether to include the singular or the plural form. The external evidence appears to give a slight edge for the inclusion, but it is too close by itself to determine Grundmann, Der Brief des Judas, 45, accepts this reading as “original” on the basis 111 of P72. Wachtel, Der byzantinische Text, 361; Wasserman, The Epistle of Jude, 313. For a 112 contrary view, see Kubo, P72 and the Codex Vaticanus, 144. One might write out this problem as P72 B C (L) Ψ (88 808) al (Byz) against ) A (81 113 307) 431 436 (453 1739) 2200 al Cyr Did Eph Thph (K:SBA).