120 Professor Bu'nLER on the Sanscrit Llngiuds. 



responding Greek verb ap-ap-io-Ku) prove must have meant 

 in the primitive Indo-European language ' to fit or to make 

 to fit.' The etymological meaning of arni=ani would there- 

 fore be "a thing to be fitted (into some other thing.") 

 (Compare also ara 1 a spoke' which is derived from the same 

 root). Katu 'sharp' probably stands for kartu, like natayati 

 for nartayati, and seems to be derived from the root krit 

 ' to cut.' 



But even if Dr. Caldwell's opinion regarding the origin 

 of all the sixteen words were correct, and even if they were 

 found in the most ancient dialect of Sanskrit, is it likely 

 that so small a number of words should have caused the 

 production of thousands of Sanskrit Unguals ? I think there 

 is a better way of solving our problem, and it is indicated 

 by the third point which I brought forward against Dr. 

 Caldwell's opinion. As the Bactro-Indian language con- 

 tained at least three Unguals, the consonantal and vocalised, 

 * t and sh, and as it can be shown that by far the greater 

 number of Sanskrit lingual mutes and nasals is produced 

 either by the direct change of %' 'sh' into Unguals or by the 

 change of dentals into the corresponding Unguals through 

 the influence of %' '11 'rf, sh, we are driven to conclude, 

 that the latter letters originally caused the production 

 of the lingual t, th, d, dh, and consequently that lingua- 

 lisation is entirely an Aryan proceeding rooted in the 

 ancient phonetic system of the language. On further in- 

 vestigation it will however appear, that the letter 'h' and in 

 rare cases also T undergo similar changes as r and sh, and 

 that in a very limited number of words Unguals have 

 been substituted for dentals without any apparent cause 

 whatever. 



I shall now turn to an examination of the facts in order 

 to show that they fully bear out my assertions. 



