The Greek Diminutive Suffix -loxo- -lOxri-. 147 



from the adjectives through the meaning ' Hke, but not equivalent 

 to,'^ since the neuter gender of diminutives argues their origin from 

 substantivized adjectives.'^ The German deterioratives Hke kindisch 

 and weibisch again have no connection with either the Greek or 

 Slavic, but are of demonstrably late origin,^ and must have been 

 patterned after words hke ' bauerisch,' in which the deteriorative 

 element comes from the primitive.* Similarly the meaning ' coming 

 from,' which is productive in Germanic and Slavic adjectives, and 

 might be conceived as due to syncretism with -ko-, is explained by 

 Brugmann, Gr. 2. i^. 666, as a new independent developmicnt in 

 words like O. H. G. irdisc, originally ' of an earthly nature,' but 

 in many combinations easily conceived as ' coming from the earth.' 

 It is the concern of the grammar of the different languages to 

 determine the relation of their own usage to the I. E. original, 

 and we may henceforth confine ourselves to the problem of the 

 origin of the Greek uses. 



II. ITS SANDHI, ACCENT, AND GENDER. 



12. About the combination of -icrxo- -ktky]- with the stem-finals 

 Httle need be said. Practically the whole thing can be summarized 

 in one statement : The final two syllables of any combination of 

 stem-suffix and -1,(7x0- -htxyj- result in -ktxo- -tcTKY)- with short i,^ as 

 is shown by the fact that the accentuation of the masculines is 

 always -iayiot;, whereas a contraction of -i-Caxo<; > -i(7xo<; would have 

 been revealed by the circumflex accent in the nominative and ac- 

 cusative, e. g. -T(7)i0i;. The suffix, then, was ordinarily added to 

 consonant stems : 'Av^pt(7xo? : avBp-, /"/jvi'Txg? : yy]^, [xeipaxicrxv] : jjlsT- 

 pa'^, 7ioBi(7xoc : tcoB-. The stem-finals 0, Yj/a, i, 10, ta, eto,® u, to were 

 propped or elided before -1(7X0-' : avQ^pwmcrxo^ : av8>pa)7co?, >.£xavtaxY) : 



1 So Belie, Arch. f. si. Phil. 23. 151. ^ Cf. Brugmann. 1. e. 



3 Cf. Wilmanns, op. cit. 469. * Cf. Brugmann, op. cit. 683. 



^ That e. g. -icxo- almost never contracts with a stem-vowel to form 

 -eiaxo- is mentioned by Choeroboscus Anec. Ox. 2. 268 ovdenote to ax iv 

 GvTiXrjxpsi ouiu £](ovai zrji' et diq)ihoyYof nQorjyovfj.syrji', ;fw^«V f^iag Ac'^fWf xat 

 ravTi^g ctfjLifiijia'k'ko^ivrig • oiov^ 'HQuxXeiaxog 'HqaxXelaxog ' naQadeiyf^ara <fe zcevra • 

 puaiXLaxog • juHQaxiaxog. Similarly Herodian 2. 445. 19. 



* -aio- on the other hand seems to have lost only the 10, but not the 

 a : 'E^ixni'exog : 'Eofxalog, 'HQutaxog : 'HQatog. For 'Igaviexog (probably not : 

 Elqrivalog) cf. § 105. 



' For the probable cause of this see note to § 6. 



