________________
No. 16.]
by an ordinary y (but without the top-stroke) attached below the upper y, instead of by the usual subscript form which we have in the preceding word rajyan: I cannot quote any similar instance in so late a record; and it seems to be here a freak.- The language is Kanarese, of the archaic type, in prose. The record gives us, in line 1-2, ballaha, as a Prakrit form of the Sanskrit vallabha; in line 4-5, paṭṭagara, as a variant of patakára, a weaver;' in line 5, seniya, a (head)-man of a guild,' from seni, = éréni, a (3), with which we have to compare náda, 'a (head)-man of a district,' from nád, nádu, + a (3); in lines 7 and 9, sampu, which seems to mean 'a length (of cloth or silk),' and to be another variant of chápu, 'stretch, length, extent,' júpu, the measure of a long stride,' and dapu, stretch, etc., the measure of a stride;' and in line 8, kelagu (with the copulative affix), as a variant of kelagu, under, down, below.'The orthography does not present anything calling for comment.
4
SOME RASHTRAKUTA RECORDS.
165
The inscription refers itself to the reign of a king whom it mentions by only the biruda Śriballaha, that is to say Srivallabha. On palæographic grounds, it is to be placed in the last quarter of the eighth century A.D. For that reason, coupled with the locality to which it belongs, it is unquestionably a Râshtrakûța record. And this king Śrivallabha is, in all probability, to be identified with the Rashtrakuta king Dhruva: the only alternative is that he is Dhruva's son Govinda III.; but, in spite of what has previously been thought, it now appears very questionable whether Govinda III. was so specifically well known by the biruda Srivallabha as was his father Dhruva. The object of the inscription is to record that the head-man of the guild of the weavers of the múrumkéri of Purigere-(Lakshmeshwar)
1 It was the ancient way of forming the subscript y; see the Junagadh inscription of Rudradaman, in the word mahakshatrapasya near the end of line 3, and in other places (Archeol. Sure. West. Ind. Vol. II. p. 128, and Plate).
1 For an instance of the word ndda, see page 71 above, note 2.
When I first brought this inscription to notice, I treated it as a record of the reign of Govinda III. At that time, in dealing with the Rashtrakutas I was chiefly following the lend of Dr. Bühler. His Table of the Rashtrakutas shewed the biruda Śrivallabha for only Govinda III.; see Ind. Ant. Vol. VI. p. 72, and his remarks (ibid. p. 64) in his introduction to the Rådbanpur grant which he was then editing, and his translation (ibid. p. 71) of the passages from which he took the biruda. And, as a matter of fact, it is only recently,since the time when the collotype of this record, now issued, was prepared and titled, that it could be recognised that this biruda, when used in a Rashtrakuta record, referable to an indefinite date in the period A.D. 775 to 800, in the special manner in which it is used in this record, does not by any means necessarily denote Govinda III. On this point, see further on, under the use of the biruda Śrivallabba in the Rashtrakuta records.
This word murum-kéri, —or murun-kéri, as actually written in this record, with the guttural nasal instead of the anusvára,- would mean, by literal translation, three streets.' But it seems to be a technical expression, the exact purport of which is to be found in connection with the wider meaning of 'quarter, quarters, a division of a town,' which kéri has in, for instance, holegéri, the Holer's quarters,' the well known expression for that part of a village (usually outside the village itself) in which the Mahars, Mange, and other low-caste people dwell. I do not at present find anything, helping to explain it, in any of the other records at Lakshmeshwar. A proverbial saying, which may or may not indicate some clue, is given in the Rev. Mr. Kittel's Kannada-English Dictionary, under náru, namely náru dru iddarú kéri beku, "though there be fully a hundred (persons), a street in necessary;" and it is explained to me by Mr. Kittel as meaning that a hundred persons, or more, may be a large number, but, if their houses are erected unsystematically, one here and one there, there is no proper village, and a street, along which houses are built in rows, is necessary to constitute a regular village. It seems likely that we have a synonym of marumkéri in another technical expression, múrumpura, of which, also, the exact purport is not apparent. Mürumpura would mean, by literal translation, 'three towns;' but the exact bearing of it is, no doubt, to be explained in connection with the more special meaning, which pura has, of a division of a town, a ward,' particularly in the actual names of such divisions or wards. There was a mûrumpura at Balagami. A record there, of A.D. 1129, likens, the pancha-mathamgal or five mathas of that place, which it specifies as the shrines of Hari (Vishnu), Hara (Siva), Kamalasana (Brahman), Vitaraga (Jinêndra), and Bauddha (Buddha), to the fivefold string of pearls of the Earth, and likens the mdrum-purangal, which it calls alliya marum-kangal or "the three eyes of that place," to three strings of pearls on the neck of that same lovely woman (the Earth), who is thus superior to even the perfect Lakshmi (P. 8. O.-C. Inacrs. No. 178, lines 48 to 46, and see Mysore Insers. p. 90). Also, a record of A.D. 1181, at the same place, mentions a certain Sâvideva, who is described as- nagarapanchamatha-múrumpurada saudo(?)re-herggade, "the Saudore(?)-Hergade of the sagara and the