PREPOSITIONS: fat, 8wL 139
suspect a sort of zeugma, the constr. being half d-Tro
TTvev/xaTtm/ ax. e^epaTreuoi/TO, like laOijvai OLTTO TOJV vocrcoi/
just before. Rev. xii. 6 is perhaps the only place
where the difference from v6 really vanishes.
But there is an extension of the use of oVo, where
it is used of cause, like the Latin prce. There is
nothing to surprise one in a use like those in Matt,
xiv. 26, Acts xii. 14, or again xx. 9 : but Matt, xviii.
7 is plainly Hellenistic : and the use in Luke xix. 3,
John xxi. 6, Acts xxii. 11 seems not to be quite good
Greek.* Winer seems to think that this last exten-
sion of meaning only occurs in " negative combina-
tions," i.e. where cwro indicates what prevents a thing
being done, not what causes it.
The constr. to express local distance (John xi. 18,
etc.), is late but pure Greek : in a better age the
measure of distance comes in the ace. before OTTO, so
that that would run ws crraSiovs te' ac/>' 'Ie/o. Perhaps
the word, though bearing quite its commonest sense,
is used with rather unusual freedom of constr. in
Acts xvi. 33, Horn. ix. 3, 2 Cor. xi. 3, Col. ii. 20
(where ajroOavtlv awo seems to be a clearer equivalent
to the aTroOaveiv c. dat. of Rom. vi. 2 etc.). If we
decline to rank Heb. v. 7 with these, it is on account
of the limitations of N. T. use of e^Xa^eta, not of aVo.
Ata c. gen. in a local sense is used of extension or
motion through, not of the limits of intervals. Of
time, on the contrary, it has the latter sense in Mark
ii. 1, Acts xxiv. 17, Gal. ii. 1 : so Matt. xxvi. 61=
Mark xiv. 58, " at an interval of three days," i.e.
* In Plut. Timol. c. 27 we have ffvvoirrov ovdev 3\v OLTTO r&v
Tro\/jLiwv : but further on in the same c. read viro T&V ap
. . . els "Xjeipas eXdelv rots KapXT/Scw'ocs ov