^{page 99 note 1} It may be observed that it is often much easier to improve upon a previous treatment of an epigraphic record than it is to produce an original treatment. The inscriptions are not accompanied by commentaries, as the literary works are: and there are frequently many more subsidiary points to be considered than are apparent at first sight, or can be fully examined all at once. The first interpretation of an inscription becomes a commentary on it: but, as we know well, even the commentaries on the literary works are not all final: and the first treatment, or even the second or third, of an inscription is not necessarily final, even though it may be the work of a scholar of repute. The inscriptions, especially the more ancient ones, always remain open to further examination in the light of later discoveries: and there is perhaps no line of work in which more advance has been made during quite recent years towards a settlement of many points previously undeterminable and even unrecognizable.Google Scholar

^{page 101 note 1} The inscription 1, A, shews that pitu = pitṛi stands here as an ēkaśēsha of mātāpitu, ‘mother and father.’Google Scholar

^{page 101 note 2} I follow Professor Bühler's translation of this record. But kāṇīṭha seems a somewhat peculiar form for kanishṭha; and we possibly have here a term denoting some office or avocation: “Utara, son of Pigaha, (is) the kāṇīṭha (in this matter).”Google Scholar

^{page 101 note 3} The text in EI, 2. 326, accidentally omits the cha both before and after Siva[sha]. The words are duly shewn in VOJ, 6, 1892. 155.Google Scholar

^{page 102 note 1} In Pāli, apt + ēva becomes appēva; but ēva is liable to become va after a long vowel: see Childers' Dictionary under api and ēva. The form which we have here seems to be a kind of compromise between those two practices: the ē of ēva was elided, and the i of api was lengthened.Google Scholar

^{page 103 note 1} Possibly, the wrong words were italicized and bracketed in EI, 2. 327,— “(has been ordered)” instead of “(the preparation of)”. On the other hand, however, in VOJ, 6, 1892. 155, we have “(has been defrayed the expense of) the preparation of a casket.’.Google Scholar

^{page 103 note 1} In Pallava records, we have the same verse with the various reading pibati instead of haratiGoogle Scholar (IA, 5. 52, line 32), and also with the further various reading kilvisham instead of dushkṛitam (ibid., 156, line 35). It comes, no doubt, from some law-book.

^{page 106 note 1} It is this point that fixes the conclusion that paṇati stands here for praṇatiḥ, and not for paṁṇatti, = prajñaptiḥ, which would give two short-syllable instants too many. As regards another detail, it may be added that composition of words is not permissible in passing from the second to the third pāda: this is a second reason for which we cannot take majusaṁ-paṇati as a compound.Google Scholar

The line might be set right by transferring paṇati to stand after cha: but that involves taking a liberty with the text; and it would spoil the construction.

^{page 107 note 1} In the way of other metrical curiosities, it may be observed that we have a Vasantatilaka verse which is irregular in the first pāda in the Garigdhār inscriptionGoogle Scholar (Gupta Inscriptions, 75, line 19):— Yātēshu|chaturshu|kṛitēshu| etc. Also, that in the inscription on the boar at Ĕraṇ we have a passage which distinctly begins as an Āryā but passes into prose (ibid., 159, line 1):— Varshē| prathamē| pṛithivīṁ| pṛithukī|rttau, etc. Further, that in the Bijayagaḍh inscription (ibid., 253) the word siddham should not be separated by a mark of punctuation from kṛitēshu etc.: lines 1 and 2 then form a verse in the Āryā metre, except that in the last pāda we must scan ētasyām as ēta|syāṁ|, taking the liberty of shortening the long ā: that this passage is a verse has been suggested by Professor Kielhorn in IA, 26, 1897. 153, and note 38.

I think that Professor Bühler somewhere pointed out that in the passage beginning with pradāna-bhuja-vikrama in line 30 of the Allahābād inscription of Samudragupta (Gupta Inscriptions, 9) we have a verse in the somewhat rare Pṛithvībhara metre. But the point may be mentioned here, in case it has not been previously notified.

² Perhaps Mr. Rouse can adduce other instances from this source.Google Scholar