1840.] from Bactrian and Indo- Scythian coins. 265 



fully confirmed. The i, following consonants, is always re- 

 presented by this line, which, in favour of the figure of some 

 consonants, is rendered also obliquely. In similar manner 

 the e or i in this place is denoted in the old Indian alphabets 

 by a mark above, connected with the consonants. Mr. Grote- 

 fend, following the Greek alphabet, has incorrectly adopted a 

 as the representative of the vowel in W; if this be so, the 

 same sign must represent y in Amyntas, e in Diomedes, 

 and i, if there is i in the Greek, as for instance, with +1 in 

 Antimachos. It is evident, that it always is the same vowel, 

 which did not sound differently in different words, but is always 

 i, and that the native language substituted the same one for 

 those four Greek vowels together; this supposition agrees also 

 much better with the nature of those languages. They all 

 reject the v ; in consequence, we find UJ in Amyntas, as it was 

 the sound, most congenial to the ear. When upon the coins 

 Dijamido is written for Diomedes, we ascribe it to the pro- 

 nunciation of the t) as an i, which already had stolen in ; the 

 vowel, however, in this instance must be long. But because 

 we also do not find the long A differing in orthography from 

 the short A, the objection taken, that our line is likewise sub- 

 stituted for a long i, will be of no consequence. 



The position holds good, when the vowel-line may represent £ . 

 Why was e not written for 6, seeing that we observed before, 

 that T though really 6, is still sometimes put for 6 ? Now just, 

 because the language has an 6, I am inclined to infer, that it had 

 also an e.* He who is acquainted with the analogy of languages 

 used in lands contiguous (to Bactria,) will admit of the inference. 

 In the Sanscrit there is no e, but there is an e ; on the other hand, 

 the Zend has even three es (e, e,e,) the first of which is decidedly 

 a short vowel ; lastly, there is an e in the Pracrit,f which is still 

 more frequently shortened than 6, but being shortened, alter- 

 nates with i. In the Zend, the manuscripts give also sometimes 

 i for 6 ; but otherwise the e is not related to i. Why then was 

 the sign e not written for the Greek e 9 if that sound was to be 



* I have above endeavoured to express the German sound of e" by A 

 ^ Inch it resembles : I shall hereafter leave it as it is. 

 t Vide my Grammar. 



