^{page 818 page 1} Kielhorn's List of the Inscriptions of Southern India, Epi. Ind., vol. 7, appendix, No. 3. We already have an appreciable number of other inscriptions, published since the time when Professor Kielhorn's two Lists were issued: but they add nothing in the present matter to what is to be learnt from the Lists.

^{page 818 page 2} Kielhorn's List of the Inscriptions of Northern India, from about a.d. 400, Epi. Ind., vol. 5, appendix, Nos. 14, 352. The Baijnāth Prasasti, which was supposed to be dated “Śaka-kāla” 726 expired, = a.d. 804–5 (ibid., No. 351) is now known to be dated “Śaka-kāla” 1126 expired, = a.d. 1204–5.

The record on the Multal plates, dated in Kārttika, “Śaka-kāla” 631 (expired), falling in a.d. 709 (ibid. No. 350), not only comes from Southern India, but also is almost certainly a southern record. Professor Kielhorn drew attention, in his introductory remarks to the Northern List, to the fact that the List includes not a few southern records.

Pandit Bhagwanlal Indraji had a Chalukya copperplate record, dated “Śaka 653” (expired), = a.d. 731–32, from Balsād (‘Bulsar’) in the Surat District, Gujarat: see JBBRAS, 16. 5. As it has not been published and he did not state the places mentioned in it (except Mangalapurl, the town whence the charter was issued), we cannot locate it. But all the probabilities are against its having any connection with Northern India.

^{page 820 note 1} The astronomer Lalla used as an epoch the end of the “Śāka” year 420, in a.d. 499: Sishyadhlvriddhida, ed. Sudhakara Dvivedi, p. 10, verse 59; p. 50, verse 18. But he wrote at some later time.

There has been recently brought to notice a Sanskrit work entitled Lokavibhaga, written by an author named Siriihasuri and treating of Jain cosmography, which gives the date of its composition as the twenty-second • year of (the reign of) Simhavarman, lord of Kanchi (Conjeveram), and the year 380 (expired), = a.d. 458–59, “of those who are named Sakas”: see the Annual Report by Mr. R. Narasiriihachar, Officer in charge of Archaeological Researches in Mysore, for the year ending 30 June, 1909, §§ 35, 112. This will have to be thought over, because it is not easy to believe that the era can have been used in such a manner as that, at so early a time, in the eastern parts of Southern India: the earliest instance, as yet established, of its use for civil dating anywhere in that direction, is of a.d. 945 (Southern List, No. 563). I learn from Mr. Narasimhachar that the Lokavibhaga quotes in its chapter 6 some Prakrit verses from a work called Trilōkaprajnapti: this may perhaps throw some light on the matter.

^{page 820 note 2} There is at any rate nothing to mark either him or his successors as Śakas.

^{page 821 note 1} The date in the year 127 is in the Jasdan inscription: see Ind. Ant., vol. 12, p. 32. The other date is in the Mulwasar inscription: see, in order, this Journal, 1890. 652; Bombay Gazetteer, vol. 1, part 1, p. 43; Bhavnagar Inscriptions, p. 43, with plate; and Rapson's remarks in this Journal, 1899. 381, and Catalogue of the Coins of the Andhras, etc., introd., p. 62. The second numeral is certainly 20 (not 30 as given in Bhavnagar Inscriptions): and on the whole the first numeral is probably 100 (not 200 as given there). The name of the king has been read both as Rudrasena and as Rudrasiha.

^{page 821 note 2} Rapson, Catalogue of the Coins of the Andhras, etc., p. 192.

^{page 822 note 1} Ujjain (we know) was a great seat of astronomy: and the Hindu prime meridian was taken through it. But there are indications that Broach, also, was in early times a centre of learning as well as of commerce: notably, the point that the Nagari characters were developed there, as is shown by the fact (see, e.g., Bühler, , Indian Paleography, p. 51)Google Scholar that the earliest specimens of Nagari are found in the signatures of the Gurjara princes of Broach on their copperplate charters, which range from a.d. 620 to 736.

^{page 822 note 2} The Gupta era of a.d. 320 is first found in those parts in the Junagādh inscription of Skandagupta (my Gupta Inscriptions, p. 58; Northern List, No. 446), which contains dates in A.D. 455 and the next two years. And the way in which the first of its dates is stated — Gupta-prahāle gananmh vidhdya; “making the counting in the reckoning of the Guptas”— is a fairly clear indication (though it is in verse) that a distinction was being made between the Gupta era and a local erastill in use.

^{page 821 note 3} Before the arrival of the Greek astronomy, which brought with it the solar year beginning at the vernal equinox to which the Hindus attached the Chaitradi lunar year, the Brahmanical lunar year began with Magha śukla 1, and was attached to a solar year beginning at the winter solstice.

^{page 823 note 1} The astronomers, or some of them, no doubt had also the choice of the so-called Vikrama era beginning in B.C. 58. But its Chaitradi variety had not then been established: at that time its years began only with Karttika śukla 1, near the autumnal equinox, which was quite a secondary point in the year. Sewell and Dikshit tell us (Indian Calendar, p. 40, note 2) that this era is never used now by Indian astronomers: and I cannot find any indications that it ever was so used.

There was of course also the Gupta era of a. d. 320, apparently with Chaitradi years: but there is no evidence that it reached Western India before a.d. 455 (see note 2 on p. 822). There was also the era of a. d. 248 or 249, subsequently known as the Kalachuri or Chedi era, in vise under the Traikutaka kings in Southern Gujarat in at any rate a.d. 456 and 493: but its years began with Bhadrapada śukla 1.