1863.] Note on the Tuxila Inscription. 155 
stance remarkable being that the word occurs several times in each 
document and always with perfect uniformity, puya in the one and 
pus'a in the other, with the most vexatious constancy. In an un- 
published inscription on a steatite vase in the Peshawar museum,* the 
word appears as puya, not pus'a, and if this be permitted to incline 
the balance of probability in favour of that reading, I feel disposed to 
trace its origin to the Sanskrita root pej “to worship ;” the 7 of the 
word being, by local or dialectic peculiarity, changed to y. In the 
present day they in Uriah is invariably pronounced as 7, and in Bengali 
it assumes the two forms of & and @, the former being pronounced as 7 
and the latter as y, though in the alphabet there is a separate symbol 
for j, and the 4 whether with or without the dot at the bottom, holds an 
only place at about the end of the series. The panditas of Bengal 
observe this distinction even in reading Sanskrita, and frequently 
quote an ancient verse in support of their practice.y The Nipalese 
_ follow the Uriah and pronounce their y, whether initial, medial or final, 
asj. According to Yajur Vedic scholars, the initial y should invariably 
be pronounced as 7, hence the name of their Veda, though written with 
ay,is pronounced as Jajur. In the Kapur di Giri inscriptions the y has 
- often been used in the place of j and the word Rajé changed to Rayo 
offers a notable instance of the convertibility of the y andy. These 
analogies would go a great way to support my assumption of puye 
being a corruption of pwa, and if this be admitted the meaning 
of puyaye, would be “ for the purpose of worship.” The supposition 
that itis a corruption of punya has not the support of analogy, 
for we know of no rule of orthoepy Sanskrita, Pali or Prakrita, which 
alone can govern the phonology of the Bactro- Pali, by which the sonant 
n can be elided for the benefit of the liquid y. The rvle in the Pré- 
krita is invariably to drop the semi-vowel and double the surd or 
the sonant with which it may be associated. 
General Cunningham reads the four last sylables at the end of the 
third line puya yuta. I wish to alter them to puyaye cha to preserve 
the syntactic connexion of the clause in which they occur with that 
which precedes it. The last letter of the line is partially obliterated, 
and taking it therefore, for the sake of concord as ch and not as ¢ 
[* Vide ante, p. 151.—Kps. J 
t Weal qeiea amie qa ae ! 
qietel q Wael aq AaRICIAA AT It 
