162 Walter Petersen, 



ypLcToTc ^(y.[i.\i.y.-:i'jy.'-jic ava},£}.OY/o)[jivoir. inriiGxoc : l~i:oc, a ' horse- 

 like ' ornament for the head. Crat. Jun. frg. 3. 376 (2). fjrjvioxog: 

 |j.rjvrj, an ornament 'like the moon,' i. e. a crescent-shaped orna- 

 ment for the neck. LXX Isai. 3. 18. Of a covering to protect 

 the head of statues Ar. Av. 1114. ae/.r^viaxoc : Gzlr^^t], a ' moon- 

 like ' ornament, i. e. an ivory crescent on the boots of the Roman 

 senators. Jo. Laur. Mens. p. i. 19, De magistr. 2. 13. Of a 

 crescent-shaped amulet in Hes. TQO'iiaxoQ : 'po/o?, an ear-ring, be- 

 cause 'like a wheel.' LXX Ezek. 16. 12. %ti(nviaito(; : /itojv, 'not 

 a real tunic, but only like one,' because shorter. Ar. Av. 946 ; 

 Antiphan. frg. 3. 17 (3) ; Xen. An. 5. 4. 13 ; Dem. 19. 197 ; CIA. 2. 

 754. 12 (349/4 B. C). 



7. Vases, Vessels, Baskets. 



41. Of the considerable number of words of this group ^ the 

 large majority is of obscure suffixal meaning, and often we can 

 not tell whether primitive and derivative designated the same or 

 different kinds of vessels, or whether the derivative was a dimin- 

 utive. As to those cases where primitive and derivative are equiv- 

 alent we can be sure that there is no fading of the notion 'be- 

 longing to the category of,' since there is no plausible example 

 of such a meaning of -1,(7x0- being actually ahve. We must rather 

 assume that the few words in which the suffix designated simi- 

 larity and which subsequently became equivalent to their primi- 

 tives caused others to follow by analogy. Thus eyivtoy.05 ' a pot ' 

 was something ' hke a sea-urchin's shell,' but the primitive l/Tvo? 

 could also be used of such a vessel, and with this meaning in 

 mind primitive and derivative seemed equivalent, so that they 

 could cause by analogy TTaptovvOi; = G-Ta[j.voi;. In view of the 

 uncertainty of the relation of primitive and derivative in so many 

 words I divide the material into two groups : A) -toy.o- probably 

 designates similarity ; B) there is no evidence that primitive and 

 derivative differed, though I would by no means affirm in every 

 case that they did not differ. 



A. -laxo- implies similarity, ixivioxog : I'/poc, (see above). Poll. 

 10. 95 mentions it with b/yjoc, among toc toO [xayeipou oxeuY). 

 d^vMixlGxoq : Q'uXaxoi;, that which is ' like a bag or sack,' a bread- 

 net or basket. Crat. frg. 2. 237 (i. 6) ; Ar. frg. 2. 1046 (37), 1165 

 (6) "E^rsiT Ixi To5']jov r^xe ':*?]v cTcupiBa T^ajicov Kai O'uXaxioxov xai to 



1 Cf. Gr. Dims. 91 ff. for a similar group of names of vessels in -101/. 



