274 Lassen on the History traced [No. 99. 



common form, which is perhaps only produced by the sharp 

 accentuation. Thus ends likewise the name of Lysias ; at 

 the same place (As. Trans. Vol. iv. PI. xxvi. No. 12. R. R. II. 

 No. 8.) Mr. Prinsep inferred, that A was a j ; Mr. Grotefend ex- 

 presses it by i. But if we restore, according to Mr. Prinsep^s 

 striking remark, the upper line of the penultimate in the name 

 of Diomedes, (As. Trans. PL xxxv. No. 3.) we shall have 

 ■P^IUJA*!, in consequence dajamido, for no body would be 

 inclined to read daimidd ; and on the other hand, daimido 

 must be written T^u/)^, A therefore is a consonant, and TAlHx 

 is to be read hirmajo. In the same manner TAVH1 must be 

 accepted as lisajd. 



Here I must however, remark, that we do not yet know 

 whether *l be not rather written as the initial in Diomedes. Mr. 

 Masson read ^AH) at the end upon the coin of Archelaos, 

 (As. Trans. Vol. iv. PI. xxxv. No. 1.) but the name upon the 

 coin has become illegible. According to the same analogy we 

 must rather expect Lisijo, TAArH ; and here also we cannot come 

 to conclusion, but by recommencing the investigation of the 

 same coins, or by discovery of a new one. 



From these corrected readings some peculiarities of the lan- 

 guage become evident. 



From Hirmajo, being substituted for Hermaios, it follows, 

 that the language did not favour an open diphthong, as ai, and 

 therefore changed the i into the affinative semi-vowel j. It 

 probably did likewise reject au, and it would have been ren- 

 dered in av, if a vowel followed this diphthong. 



From Lisajd and Dajamido or Lisijo and Dijamido, it follows, 

 that this language was averse to admit the use of i, followed by a 

 vowel, even if a consonant preceded the i ; I use the expressi- 

 on ee is averse to' 5 this admission ; for if the three first syllables 

 in Antialkides are expressed by atia, this is probably done only 

 in obedience to the order of a foreign king. So much at least 

 is evident, that the language rejected the hiatus in Lysias and 

 Diomedes ; in what manner it was supplied, must be left at pre- 

 sent undecided. However, the most obvious conjecture is, that 

 a j was evolved from i. Thus the Sanscrit in bhie resolves 

 the long i into u whence bhije ; in ijarti, ij is derived from a 



