1887.] Dr. Mitra— On Ehotibhdva. Ill 



sound is the same ; it is read to express the Sanskrit f! ' This reversed 

 mark is very uncommon, and in ordinary Tibetan writings it is not met 

 with. It is certain too that the Tibetans do not make any distinction 

 between the long and the short i. It is unsafe therefore to rely upon 

 Tibetan texts in this respect. 



1 The reference to the Sanskrit MS. is also puzzling to me. I can- 

 not make out what Babu Saratchandra Das means by * Devanagari 

 character of the 8th or the 9th century A. D.' I am not aware of any such 

 specific character, and I should very much like to know how the vowel- 

 marks are put in it before I can decide its value. The Babu's reading 

 may be correct, but I cannot say as much for the correctness of the MS. 

 Anyway we have on the one side a single MS., and that in a country where 

 the distinction between the long and the short i is very much neglected, if 

 not positively unknown, and on the other eight MSS. from a country 

 where the distinction is carefully observed, and the whole of the Pali 

 texts examined by Mr. Childers. In this state of evidence before me, 

 I cannot venture to come to other than an adverse conclusion. 



1 The derivation given by Babu Saratchandra Das of the term under 

 notice is not his, nor taken from his Tibetan MSS., but founded upon 

 a Sanskrit passage written by Professor Nilmani Mukerji. The passage 

 has been incorrectly transliterated by Babu Saratchandra Das. The 

 word uttarthdh is obviously intended for ityarthah, and the utah should 

 have been written with a long u. 



' The derivation of the word as given by the Professor does not 

 appear to me to be satisfactory. The crucial word is uti, the second 

 member of the compound, and it may be derived from more than one 

 Sanskrit root. Professor Mukerji derives it — from ve, Ida and chvi, but it 

 is not what we find in our Sanskrit dictionaries, and it necessitates 

 recourse to two affixes when one is quite sufficient. Ghvih as an affix is 

 rarely used, and not at all needed here. Ve with hti makes uti, and this 

 is the form most used by our lexicographers and exegetes, and I see no 

 reason to reject it in favour of a derivation which no Sanskrit author 

 has used, and which involves the use of two affixes for a single purpose. 

 It amounts to a preference for a novelty for the sake of novelty only. I 

 go further, and hold that, under the rule of Panini, kriblivashiyoge sampa- 

 dya kartari chvih (5. 4. 50) the affix chvi seems inapplicable in the present 

 instance. The rule requires a complete change of substance (abhuta- 

 tadblidva — and the leading example is Brahmi bhavati, or change into 

 Brahma. The subject has been explained at some length by Professor 

 Taranatha Tarkavachaspati, in a note in his edition of the Siddhanta- 

 kaumudi, but I cannot cite his words from memory. This much, however, 

 is certain that the stringing on a thread does not imply such a change 



