de divergence of the latter. The
passage is one where almost every word and syllable might easily
and naturally be altered--as the third Gospel shows that they have
been altered--and yet in the Clementines almost every peculiarity
of the Matthaean version has been retained.

Another quotation which shows the delicacy of these verbal
relations is that which corresponds to Matt. vi. 32 (= Luke xii.
30):--

_Matt._ vi. 32.

[Greek: Oide gar ho pataer humon ho ouranios, hoti chraezete
touton hapanton.]

_Clem. Hom._ iii. 55.

[Greek: [ephae] Oiden gar ho pataer humon ho ouranios hoti
chraezete touton hapanton, prin auton axiosaete] (cp. Matt. vi.
8).

_Luke_ xii. 30.

[Greek: Humon de ho pataer oiden hoti chraezete touton.]

The natural inference from the exactness of this coincidence with
the language of Matthew as compared with Luke, is not neutralised
by the paraphrastic addition from Matt. vi. 8, because such
additions and combinations, as will have been seen from our table
of quotations from the Old Testament, are of frequent occurrence.

The quotation of Matt. v. 45 (= Luke vi. 35) is a good example of
the way in which the pseudo-Clement deals with quotations. The
passage is quoted as often as four times, with wide difference and
indeed complete confusion of text. It is impossible to determine
what text he really had before him; but through all this confusion
there is traceable a leaning to the Matthaean type rather than the
Lucan, ([Greek: [ho] pat[aer ho] en [tois] ouranois ... ton aelion
autou anatellei epi agathous kai ponaerous]). It does, however,
appear that he had some such phrase as [Greek: hueton pherei] or
[Greek: parechei] for [Greek: brechei], and in one of his quotations
he has the [Greek: ginesthe agathoi] (for [Greek: chraestoi])
[Greek: kai oiktirmones] of Justin. Justin, on the other hand,
certainly had [Greek: brechei].

The, in any case, paraphrastic quotation or quotations which find
a parallel in Matt. vii. 13, 14 and Luke xiii. 24 are important as
seeming to indicate that, if not taken from our Gospel, they are
taken from another in a later stage of formation. The characteristic
Matthaean expressions [Greek: stenae] and [Greek: tethlimmenae] are
retained, but the distinction between [Greek: pulae] and [Greek: hodos]
has been lost, and both the epithets are applied indiscriminately to
[Greek: hodos].

In the narrative of the confession of Peter, which belongs to the
triple synop