A NOTE ON THE HADLEY -ALLEN GRAMMAR 



By H. W. Magoun. 



While Professor Allen has made an excellent revision of 

 Hadley's Greek Grammar, it has happened in a few instances 

 that the improvement intended has failed to be realized. Such 

 a case appears in § 199, where he says of feminines in -«> of the 

 third declension : " These stems seem to have formerly ended 

 in -of-'. : hence the voc. sing, in -oT an older form of the nom. 

 in -w : Ia-(fa)" This explanation might help to make clear the 

 voc. sing., which in the third declension is prevailingly like the 

 stem; but it must fail to account for the nom., since the ques- 

 tion arises at once how it happens that the nom. did not take 

 the regular ending -? precisely as other iota-stems do, such as 

 r.6Xi-q, duvatii-:;, etc, which it will be observed are also feminines, 

 and it further produces the grave difficulty of accounting for 

 the dropping of both the digamma and the iota in the oblique 

 cases where such a stem ought to give us a diphthong -of-. It 

 seems very likely that the real explanation may be found in 

 supposing that these are really " yod"-stemSi that is, that the 

 real final letter is the so-called iota which appears under a 

 changed form in the so-called iota class of Greek Verbs, where 

 the old Greek -no appears in Sanskrit as -yd-. If this is the 

 case, it may be possible to explain the whole paradigm. In the 

 oblique cases this "yod" simply dropped, exactly as it did in 

 r/jj??,* Sanskrit trdy-as, and contraction took place precisely as 



* See Brugraann, Grundriss der Vergleiehenden Qrammatik der 

 Indogermanischen SpracJien, Erster Band, §§129 and 130 Cf. also §§ 117' 

 118, 119, etc. For verb forms see Miiller's Handhueh der Klassischen 

 Altertums-Wissenschaft, Zweiter Band, §§ 123 and 124. 



