e as this are usually the result of unconscious acts of the
mind, i.e. of memory. A curious instance of the way in which the
Synoptic parallels are blended together in a compound which
differs from each and all of them is presented in 437 D ([Greek:
to blasphaemounti eis to pneuma to hagion ouk aphethaesetai auto
oute en to nun aioni oute en to mellonti]). Another example of
Epiphanius' manner in skipping backwards and forwards from one
Synoptic to another may be seen in 218 D, which is made up of
Matt. xv. 4-9 and Mark vii. 6-13. A strange mistake is made in 428
D, where [Greek: paraekolouthaekoti] is taken with [Greek: tois
autoptais kai hupaeretais tou logou]. Many kinds of variation find
examples in these quotations of Epiphanius, to some of which we
may have occasion to allude more particularly later on.

It should be remembered that these are not by any means selected
examples. Neither Irenaeus nor Epiphanius are notorious for free
quotation--Irenaeus indeed is rather the reverse. Probably a much
more plentiful harvest of variations would have been obtained e.g.
from Clement of Alexandria, from whose writings numerous instances
of quotation following the sense only, of false ascription, of the
blending of passages, of quotations from memory, are given in the
treatise of Bp. Kaye [Endnote 56:1]. Dr. Westcott has recently
collected [Endnote 56:2] the quotations from Chrysostom _On the
Priesthood,_ with the result that about one half present
variations from the Apostolic texts, and some of these variations,
which he gives at length, are certainly very much to the point.

I fear we shall have seemed to delay too long upon this first
preliminary stage of the enquiry, but it is highly desirable that
we should start with a good broad inductive basis to go upon. We
have now an instrument in our hands by which to test the alleged
quotations in the early writers; and, rough and approximate as
that instrument must still be admitted to be, it is at least much
better than none at all.

CHAPTER III.

THE APOSTOLIC FATHERS.

To go at all thoroughly into all the questions that may be raised
as to the date and character of the Christian writings in the
early part of the second century would need a series of somewhat
elaborate monographs, and, important as it is that the data should
be fixed with the utmost attainable precision, the scaffolding
thus raised would, in a work like the present, be out of
proportion to