I had not intended to write anything more about the inscription on the Piprahwa relic-vase, treated by me in this Journal, 1906. 149 ff., until I should have completed my examination of the tradition about the corporeal relics of Buddha, and should be able to offer a facsimile of the record. And it is only recently that the occasion has arisen for presenting sooner any further remarks, as the result of the criticism of my interpretation of the record advanced by M. Senart in theJournal Asiatique, 1906, 1. 132 ff., and by M. Barth in theJournal des Savants, 1906. 541 ff. That two such distinguished scholars should differ from me so radically, is an important matter. And I wish that I had seen M. Senart's remarks sooner; but, though issued early in the year, they did not become known to me until towards the end of September. M. Barth's paper, issued in October or November,— in which he has reviewed all the principal previous treatments of the record and suggestions made regarding it, and has endorsed M. Senart's conclusions except in the grammatical analysis of the compoundsukitibhatinaṁ,— reached me after the writing of this article, but in time for me to make a few additions to it.

^{page 105 note 1} The concluding instalment of this inquiry is held over in consequence of want of space.

^{page 105 note 2} There have been unexpected difficulties in the way of doing this; one of them being, that, of the two casts before me, the, cast that belongs to this Society is the one that should be reproduced, but unfortunately at some time or another it was broken into six pieces. It is confidently hoped, however, that a facsimile can be given at a fairly early date from a fresh cast.

We may defer, until the issue of the facsimile, any further discussion of the period to which the framing of the record should be referred.

^{page 108 note 1} I quote them from Mr. Louis Gray's useful work,Indo-Iranian Phonology, § 314.

^{page 109 note 1} The Aśōka records do not happen, so far, to disclose any use of the wordssva, svīya, svaka, orsvakīya themselves.

^{page 110 note 1} These instances, again—(except the first),— I take fromIndo-Iranian Phonology, § 905.

^{page 111 note 1} This feature had not come to notice when M. Senart wrote.

^{page 112 note 1} They may be called “peculiarities;” but it seems hardly correct to continue to mark them as “irregularities: ” because they were plainly recognized features of Pāli and Prākṛit verse.

^{page 114 note 1} From his footnote on page 134, M. Senart seems to have misunderstood, me on this point. I have not suggested that Śākya was obtained by an erroneous restitution from the Prākṛitsaktya = svakīya. I have traced, separately, the form Sakya fromsvakīya, and the form Śākya fromśākīya.

^{page 116 note 1} See Bṛihat-Saṁhitā, 57/56. 8.

^{page 117 note 1} There is somewhere a good epigraphic instance of this. But I cannot find it on the spur of the moment; and it is not necessary to spend time in searching for it, because the permissibility of such an arrangement is undẹniable.

^{page 120 note 1} We havedhātu in a passage with two meanings in the Harshacharita, Kashmīr text, 370, line 1. From one point of view it there means ‘mountain minerals: ’ from the other it meanslarghūni asthīni, ‘the small bones’ (commentary),— “the ashes” (trans., Cowell and Thomas, 159);—of king Prabhākaravardhana.

^{page 121 note 1} It need hardly be remarked that, withbhagavat simply qualifyingbuddha, any case ofbhagavat might stand either before or after the appositioual case ofbuddha.

^{page 122 note 1} I am using, of course, the customary Dēvadatta, whose lot it has been to be chosen as the subject of so many grammatical illustrations. The Dēvadatta of the Pāli books, though he was a cousin of Buddha, would apparently have done anything to the kinsmen of Buddha rather than protect them.

^{page 123 note 1} We all know that rhyme plays a considerable part in vernacular Indian poetry. It figures in also Sanskṛit lyrical poetry: see remarks byColebrooke,,Essays, 2.58,Google Scholar andWilson,, Sanskṛit Grammar,434,b;Google Scholar and tor some instances. see Colebrooke, 68 f., Wilson, 449, andBrown's,Sanskrit Prosody,22.Google Scholar And we have a two-syllable rhyme, whether intentional or not, in the verse on the Peshāwar vase (see thisJournal,1906.453,714).Google Scholar

^{page 124 note 1} Or, of course, any other suitable word of that class, or some such term aspratishṭhāpita, ‘caused to be set up, erected,’ orkārita, ‘caused to be made.’

^{page 125 note 1} AsBarth,M. has indicated (loc. cit.,551, note I), I myself at first translated the record (this Journal, 1905. 680) under the influence of that understanding of it. But I felt, at the time, that that was a strained translation in view of the corrected order of the words. I did not, however, then see exactly how to improve upon it. As I have said elsewhere (this Journal, 1906. 149), I subsequently obtained the required clue from what M. Sylvain Lévi wrote about the record.Google Scholar

^{page 126 note 1} SeeColebrooke,,Essays, 2.138, No. 9. But, is this only a theoretical variety of these metres?Google Scholar

^{page 129 note 1} This is the grammatical construction according to either view of the case. According to the view of M. Senart and M. Barthbuddhassa, according to my viewsukīti-bhātīnam, is dependent, not onsalīla-nidhāne, but onsalīla. This construction, of case-nouns standing outside a compound and to be construed not with the entire compound but with one of its members, is of frequent occurrence; seeSpeijer's,Sanskrit Syntax, § 231.Google Scholar

^{page 130 note 1} I have made some remarks on this point in this Journal, 1906. 179. Dr. Grierson, however, has suggested to me that the explanation is that there was one large prominent Stūpa, with a great number of miniature Stūpas, like the “model Stūpas” found in large numbers at Bōdh-Gayā (see, e.g., ASI, 3. 87), lying all about the place.

^{page 130 note 2} See remarks by Professor Kielhorn in EI, 8. 30, note 3.