We should naturally place this quotation in the second column of our classified arrangement, as presenting a slight variation.  At the same time we should have little hesitation in referring it to the passage in our Canonical Gospel.  All the marked expressions are identical, especially the precise and selected words [Greek:  phronimos] and [Greek:  akeraios].  It is however possible that Ignatius may be quoting, not directly from our Gospel, but from one of the original documents (such as Ewald’s hypothetical ‘Spruch-sammlung’) out of which our Gospel was composed—­though it is somewhat remarkable that this particular sentence is wanting in the parallel passage in St. Luke (cf.  Luke x. 3).  This may be so or not; we have no means of judging.  But it should at any rate be remembered that this original document, supposing it to have had a substantive existence, most probably contained repeated references to miracles.  The critics who refer Matt. x. 16 to the document in question, also agree in referring to it Matt. vii. 22, x. 8, xi. 5, xii. 24 foll., &c., which speak distinctly of miracles, and precisely in that indirect manner which is the best kind of evidence.  Therefore if we accept the hypothesis suggested in ’Supernatural Religion’—­and it is a mere hypothesis, quite unverifiable—­the evidence for miracles would not be materially weakened.  The author would, I suppose, admit that it is at least equally probable that the saying was quoted from our present Gospel.

This probability would be considerably heightened if the allusion to ‘the star’ in the Syriac of Eph. xix has, as it appears to have, reference to the narrative of Matt. ii.  In the Greek or Vossian version of the Epistle it is expanded, ’How then was He manifested to the ages?  A star shone in heaven above all the stars, and the light thereof was unspeakable, and the strangeness thereof caused astonishment’ ([Greek:  Pos oun ephanerothae tois aoisin; Astaer en ourano elampsen huper pantas tous asteras, kai to phos autou aneklalaeton aen, kai xenismon pareichen hae kainotaes autou]).  This is precisely, one would suppose, the kind of passage that might be taken as internal evidence of the genuineness of the Curetonian and later character of the Vossian version.  The Syriac ([Greek:  hatina en haesouchia Theou to asteri] [or [Greek:  apo tou asteros]] [Greek:  eprachthae]), abrupt and difficult as it is, does not look like an epitome of the Greek, and the Greek has exactly that exaggerated and apocryphal character which would seem to point to a later date.  It corresponds indeed somewhat nearly to the language of the Protevangelium of James, Sec.21, [Greek:  eidomen astera pammegethae lampsanta en tois astrois tou ouranou kai amblunonta tous allous asteras hoste mae phainesthai autous].  Both in the Protevangelium and in the Vossian Ignatius we see what is clearly a developement of the narrative in St. Matthew.  If the Vossian Epistles are genuine, then by showing the existence of such a developement at so early a date they will tend to throw back still further the composition of the Canonical Gospel.  If the Syriac version, on the other hand, is the genuine one, it will be probable that Ignatius is directly alluding to the narrative which is peculiar to the first Evangelist.