156 Walter Petersen, 



sponsible for the preservation of quite a number. Yet after all 

 there seems to have been a growth of this use of the suffix in 

 the Alexandrian and Roman ages, a growth which apparently is 

 commensurate with the dechne of its ' diminutive ' uses, and this 

 explains its cause. In the Lyric age -itxo- was the ' diminutive ' 

 suffix, and -lov had not yet developed, while in the Alexandrian 

 age -tov was the ' diminutive ' suffix par excellence, and left to 

 -10X0- its most characteristic function of designating similarity, 

 which correspondingly increased. The Attic period represents a 

 transition : ' diminutives ' were still frequent, but the other function 

 was gaining ground. 



31. Of the two factors present in the notion of similarity, that 

 of likeness in some respects and unlikeness in others, ^ the latter 

 was undoubtedly the most important psychological factor for -itko- 

 originaUy, if its origin was correctly described, since mere ap- 

 proximate identity was what the comparative degree emphasized. 

 This emphasis is found e. g. in p^vtaxo? ' crescent,' ' not a real i. e. 

 full moon,' in contrast to [xt^vyj ' moon,' ^(X'jiki<7v.oq : (^aaiXo'Jr, ' not 

 a real king,' 'a mere kingling,' because king of a small country, 

 %c(,ibi<T/.-q : -ncdq, ' not a real child, but only approximately one,' be- 

 cause too old, ' a young girl,' /itwvioxo? : /itwv, ' not a real tunic ' 

 because too short, ' a short coat.' The suffix, however, is not 

 confined to these cases any more than -lov ; cf. e. g. o'jpaviaoioc : 

 o'jpavoc, ' that which is hke the sky,' 'the roof of the mouth,' or 

 dvi(7)co$ : ovo?, ' that wich is like an ass,' ' a windlass,' in which a 

 point of similarity in contrast to general dissimilarity is emphasized. 



32. That -1(7X0- was actually interpreted as implying similarity 

 rather than as a diminutive suffix is of course most evident when 

 the derivative designates an object of the same or larger size than 

 the primitive, e. g. G-cp-/]yioxos : (j(^'(]t„ a large piece of wood shaped 

 like a wasp's tail, /yivicryioi; : yy\y, the end of a ship's stern turned 

 up like a goose's neck, xaiBioxTj ' a young girl ' : Tj 7:01? ' child,' 

 ' little girl,' ayxtovicrxos ' joint ' in a piece of wood one and one 

 half ell wide : cxyxojv ' elbow.' Such words show what the inter- 

 pretation originally was for oupavt(7xoc ' roof of mouth ' : oapavo? 

 ' heaven,' bmoTto? a certain ornament for the head : I'twTio? ' horse,' 

 Tpo/t<r/.oc ' a small ball of soap,' 'lozenge' : Tpo/61; 'wheel,' bvirrAoc, 

 ' a sea-fish of the cod kind ' : OV05 ' ass,' etc. Yet after the dimin- 

 utive use had once been developed the notion of small size might 

 appear uppermost in some words of this kind, and when the prim- 



1 Cf. Gr. Dims. 98 ff. 



