J 862.] On some Bdbtr.o-BuddMsl Belies from Mdical Pindi. 179 



as far back as the 7th century. In its simple form it must have 

 been in use long before that time, and as the Jogis, as a sect, are of very 

 ancient date and notices of their rigorous penances occur in books 

 many centuries before the commencement of the Christian era, it 

 would not be too much to suppose that the term haiisa was well 

 known at the time when the Bactrians held sway in Western 

 India. If this be admitted, bearing in mind the well-established 

 fact of the Buddhist having borrowed most of their terminology 

 from the Hindus, it would not be unreasonable to suppose that the 

 duck under notice, was placed in the monument as an emblem of the 

 superior intelligence of the saint whose memory it was to perpetuate. 

 The inscription (Fig. 11) is in Arian characters, its language being 

 Pali, similar to that of the Kapur-di-giri edicts of As' oka, and the 

 Wardak record of the time of Huvishka. The letters have been 

 punched on the gold leaf, and are in an excellent state of preservation, 

 but several of them are peculiar in shape, and the difficulty of ascer- 

 taining their phonetic values throws much doubt on the meaning of 

 the whole record. Moreover in the Arian alphabet, as far as } 7 et known, 

 four different letters either by themselves or with their vowel- 

 marks, appear very much alike, and they constantly lead to misappre- 

 hensions and mistakes. They are all formed of an oblique line bending 

 to the left with a top stroke more or less curved. The letters alluded to 

 are v, r, t, and b. Of these v perhaps is the most characteristic with its 

 perfectly horizontal top line, and yet it is liable to be mistaken for an 

 r ; the r is liable to be confounded with t and b, and the t has a strong 

 tendency to merge into b. The I too in the first line of the Kapur-di-giri 

 inscription has some resemblance to b. The v stands at the fourth 

 remove from b and is not often liable to be mistaken for it, nor for a t, 

 and yet when the horizontal top stroke is modified by a perpendicular 

 stroke at its end to indicate the long vowel a, nothing save the context 

 is left to guide the decypherer to their values, and even that dubious 

 guide fails him whenever he has an unknown proper name with any 

 of these letters before him. I feel myself, therefore, in my reading of 

 the record, freely open to correction, and if I publish it in its tentative 

 form, it is only to provoke enquiry, and to assist the researches of others 

 into a subject fraught with the deepest interest in connexion with 

 the history of Bactrian domination in India. I presume not to apply 

 the "verifying faculty" so as to convert the plausible into the certain. 



