47 It may, however, have been accidentally lost. See Moffatt, INT., pp. 238 ff, where the question is discussed. 48 In this connexion it is important to remember that even early orthographic peculiarities have been accurately preserved. “I have been much struck by the number of cases in which the old uncials preserve spellings which can be proved current in the time of the autographs, but obsolete long before the fourth century. Faithful in minutiae, they might reasonably be expected to be faithful also in greater matters” (J. H. Moulton, in an article in the _Classical Review_, March, 1915, reprinted in _The Christian Religion in the Study and the Street_, 1919, p. 153). See also the _Prolegomena_, pp. 42‐56. 49 The italics are ours. 50 Plummer, ICC., St. Lk., pp. xlviii ff.; Harnack, _Luke the Physician_, p. 104 f.; Moffatt, INT., p. 278 f.; Hawkins, HS., 2nd Ed., pp. 15 ff. 51 There is a well‐known difficulty of punctuation in verse 35. Ought we to put a comma, with WH., after κληθήσεται? If we do so, the subj. is τό γεννώμενον, and ἅγιον is part of the predicate. If we omit the comma, the whole phrase τὸ γεν. ἅγιον is the subj., and the pred. is κληθ. υἱὸς θ. (cf. RV. marg.). Most critical editors of the Greek text omit the comma. It is probable, as the WH. type shows, that Dr. Hort was influenced by his belief that ἅγιον κληθ. went together as a quotation or reminiscence of the OT., and, if the passage comes from St. Luke, this is a strong argument. On the other hand, it can be argued that if the words are a Greek rendering of an Aramaic phrase it is improbable, if not impossible, that the participle should stand alone as the subj. It is not possible, of course, to settle the question by appealing to manuscript authority, as the early MSS. were practically devoid of punctuation marks. In our own case, we are unable to use either of the arguments cited, since each rests upon the assumption of the Lukan origin of Lk. i. 34 f., which is the very point we are discussing. While then we follow the WH. text we have to leave the question of punctuation an open one. If the comma should be omitted we lose the difficulty of τό γεννώμενον noted on p. 61, and we lose also the argument from its construction, sketched on p. 64.