40 LANGUAGE OF THE NEW TESTAMENT.
for eXeeT. Here again in the great majority of N. T.
passages we have forms that would suit with either
pres., the aor. imp. cXei/erov being very frequent.
There is no doubt about eXXoya in Philem. 18 ; but
there is little evidence for cXXoyaTat (more for the
anomalous eXXoyaro) in Rom. v. 13.
There is, at least in one or two cases, similar un-
certainty between the forms e^ovScveo) (or efou#ei/e'w)
and -voco.
It is judged that verbs in -oo> always make the
inf. in -olv not -ow, except irXrjpovv in Luke ix. 31.
Certain other forms from stems in o raise syntactical
questions, for which see below (p. 107-8). But we
may notice here the Si&o which in Rev. iii. 9 seems
to stand for Si'Soyu, and the (not unnatural) formation
aTroStSow which is not improbable in xxii. 2. In
all these points, it is hard to draw the line between
questions of inflexion and questions of orthography
the latter of which we do not think it needful to
discuss. And in nearly all, the question iscomplicated
with that of uncertainties of reading. For a full
discussion of these, we must refer to Westcott and
Hort's Appendix II. We can only say, as a summary
of the conclusions there arrived at, what were the
general habits of the chief groups of MSS. in re-
producing or disguising what we may regard as the
spelling of the N. T. writers.* Apparently, those
MSS. which transmit the text with least modifi-
cation transmit the spelling with least modification
too; though here we have to allow a good deal for
* We must remember, when we use this phrase, that in the
case of St. Paul at least, the most that we can arrive at is the
practice, not of the Apostle, but of his various amanuenses.