Papers of the American School of Classical Studies at Athens (4.1885-1886)

NOTES ON ATTIC VOCALISM.

269

On the other hand, E continues to be written occasionally for e
until the latter part of the fourth century,1 — an improbable ortho-
graphical survival, had e become in any degree diphthongal. This
occurs most frequently in the word iU, which is equally written h
when a consonant follows ; the prosodial treatment of the preposition
in Comedy makes it certain that in mature Attic the longer form had
entirely superseded the shorter. The spellings e (mjky and d a-TrjArj
for iv a-Tykjj2 are interesting as showing that the syllable -evs became
-is in rapid utterance as well in Attic of the fourth century as at far
earlier periods of the Greek language.

B. Confusion of I with rj. That e had in no degree become diph-
thongal in the fourth century is further testified by its interchange-
ableness, in certain cases, with the open <?-vowel. Thus between 378
and 324 the nominative plural ending of nouns in -eus was suffering a
gradual change from -jys to -eis : during this time we find both spell-
ings even in the same inscription.3 Perhaps the difference in these
cases is a merely orthographic one ; for the frequent spellings
-o?s,4 -«"s,s of this same termination show clearly enough how easy
to an Athenian ear was the confusion, at this time, between e, e, and
the narrow (or aboriginal) -q. The augmented forms of ipyd^oficu
furnish another example of this, beginning interchangeably with cl-
ot f)- during the fourth century. Here, again, the difference was
probably only in the spelling. It should be kept in mind, however,
that in Attic this confusion of c and -q is only occasional, — the two
vowels were at no time properly equivalents, and the contractions
producing them are quite different. Plurals in -i}s are, of course,
not from contracted -hs, but from -ijts; for the dual of yeVos, yeVi? is
as much a violation of Attic as yeVee.6 Yet, were we in possession

1 The latest examples are perhaps II. 804 A, 13, 33, is rh and aTroHiaev.
■ n _86 (376-365); 553 (403).

3 Aieis, KapOauTs, 'Eoriai^s, XaKKi[8ijs], II. 17 (378); KudaOrivattTs, naiaciTjs,
865 (after 400); [-A/u£|arr««[s], 'AfaKaiijs, et al, 1006 (bef. 350); v. Meisterh.,
p. 56.

4 Cf. -k\eris Tor -xMjs, even in the fourth century, Meisterh., p. 57. 'AypuKer/s,
L 33S (40S); ['AAaVer, KoAA[u]t€6s, -f^aie'ei, beside Bar^j, II. 870 (circa 350).

5 KoAAuTekj, beside 'AKaieTs, etc., EL 872 (341), etc. Cf. Meisterh., p. 55.

. 6 o-kc'AE, Et. 652 A, 24; £ivyE, id. b, 26. w6\rt (Isoc. 8, 116), <pi<rv (Plat. Hep.
410 e), and similar forms, if genuine, are the result of a desire to differentiate the
dual from the dat. sing, when the diphthong ei had become identical in sound