64 Charles E. Bennett. 



suggests that the elided vowel is i, i.e. Kan. Hall {Proceed- 

 ings Am. Or. Soc., x., p. clviii.) suggests ko, re {i.e. kuI re) in 

 the sense of the usual re kul ; but this is impossible. 



Assuming with Deecke that Kan was the full form of this 

 word we are not justified in assuming that this developed 

 to *Kd(Tc and then to /ca? (before vowels), since in that event 

 we should not find /cart and /ca? side by side. Nor can we 

 explain /cat as developed from Kart, through the medium of 

 *Kda-i, since the secondary <r of the latter form would not have 

 disappeared, but would have remained. 



3. ISe occurs Coll. 60, 12, 24, used like the apodotic Be to 

 introduce the conclusion of a conditional sentence. In 60, 

 26 it has the force of the simple 8e. 



4. t 'and' is found Coll. 60, 24. 



5. Trdi Coll. 60, 4; 60, 12 ; 71 is most naturally explained 

 like the Attic ttj] as an instrumental which has assumed 

 the I secondarily. Meyer, Gr. Gr.,^ § 388. 



194 



