186 IVa/ier Petersen, 



minutive ' meanings. The latter uses along with its use to desig- 

 nate similarity are the only ones that can be found in the Classical 

 period, but since -ko- in Greek was not a hving ' diminutive ' 

 suffix nor often a suffix of similarity, there rarely was any con- 

 tact with -icr/.o-, and further syncretism of the two suffixes was 

 out of the question. Any spread of the latter to meanings more 

 remote from its original one must be due to different causes, either 

 syncretism with Greek -lov or congeneric association or both. 



86. That, however, such meanings as ' belonging to,' ' coming 

 from,' ' made of ' were ever consciously connected with the suffix 

 is not at all probable. The vague feeling of equivalence of -tov, 

 which certainly had all of these adjectival meanings in addition 

 to its ' diminutive ' uses, caused certain words of analogical rela- 

 tion between primitive and derivative to be formed by -1,(7x0- if a 

 certain pattern in -lov existed, or one word in -itxo- brought forth 

 another one which was synonymous or belonged to the same con- 

 generic group even though the new word did not show a mean- 

 ing previously found in the suffix. But there is no case where 

 the freedom of forming new words with suffixal meanings outside 

 of the limits just mentioned shows that these adjectival meanings 

 were actually felt. I arrange the words in question then accord- 

 ing to the apparent meaning of the suffix, though it will be clear 

 that the cause of its addition was something else than a feeling 

 for such meanings. 



87. In the first place there are quite a number of words in 

 -iav.0- which are indisputably equivalent to their primitives, aside 

 from the names of vessels mentioned g 41. It might be supposed 

 that all of these, hke a number of similar words in -lov, went back 

 to the meaning ' belonging to the category of,' that e. g. XsxavifTXY] 

 was a vessel ' belonging to the category of Xsxav/],' ' a vessel of 

 the plate kind.' But there is no evidence whatever for a feehng 

 for such usage at any stage of the language, and these words are 

 due to congeneric attraction, following other words in which -loxo- 

 implied similarity or faded diminutives. Similarly the following 

 words : d^qqiaycog (CIA. 2. 678 B 73) B£pp[i](7xoi llll (378-366 B. C.) 

 seems to be equivalent to Bsppi? ' leather covering or coat,' and if 

 so, it followed /iTcovtcrxos (§ 40) in its ending as well as gender, 

 as did also TQiplaxog = 6 Tpifxixo?. Cf. Hes. Tpuxicrxov • iixoctiov. 

 'Ac-TtsvBioi. Xiiiiriaxog ' a kind of fillet,' seems to have been in- 

 fluenced in its ending and gender by the congeneric (TTscpavtcrxo? 

 'wreath or crown.' Its primitive, however, is not found, but we 

 may refer it to a *>.rj[j.vCov on the basis of Bi,>.r,[j.viov, or else to 



