^{page 357 note 1} J.R.A.S., 1890, p. 639. In the following pages, this article is referred to Bh.

^{page 358 note 1} Rapson, , “Indian Coins” [Bühler's Grundriss, ii, 3b], § 81Google Scholar, for references to the various attempts made to explain these coin-legends.

^{page 359 note 1} It may be mentioned that in these copies the restorations within brackets are in every case probable, and seem to be justified by the traces which remain of the letters. A note of interrogation has been substituted whenever the visible traces do not suggest a probable restoration. It is interesting to note that Colonel Shepherd's coin alone of all the available coins of Nahapāna and Castana preserves the letter of ) distinct, and thus verifies a conjecture which had been made before its appearance.

^{page 360 note 1} For the initial letter z, cf. the Greek transliteration of the word kṣatrapa on the coins of Caṣṭana (inf., p. 370). The representation of r by l both in Sanskrit and in the Prakritic dialects is common enough; cf. Wackernagel, , Altindische Grammatik, § 191Google Scholar.

^{page 360 note 2} Cf. Archaeological Survey of Western India: Kāṭhiāwāḍ and Kachh, p. 16.

^{page 362 note 1} I must altogether retract an opinion expressed by me in a note to Bhagvānlāl's paper (p. 648) that “It is scarcely too much to say that no identical or even very similar combinations of these Greek letters are to be found on different specimens.” It was only after the systematic examination of many hundreds of specimens that it was seen that these fragments are, in most cases, parts of the same original.

^{page 363 note 1} Roman coins belonging to the period from Augustus to Antoninus Pius were, as we know, plentiful in India (to the references given in Indian Coins, § 14, add Hill, , Num. Chron., 1898, p. 304)Google Scholar, and there are undoubted instances of Roman influence on Indian coinage–e.g., in the bust on the small bronze coins of Kozola Kadaphes, and in the style of the figures of divinities represented on the gold coinage of the Kuṣana monarchs Kaniṣka and Huviṣka.

^{page 365 note 1} Ind. Ant., 1897, p. 153.

^{page 366 note 1} “On the Sáh, Gupta, and other Ancient Dynasties of Kattiawar and Guzerat”: Journ. Bomb. Br. R.A.S., 1862, p. 1.

^{page 368 note 1} Thomas (Arch. Surv. West. Ind., Kāthiāwāḍ and Kachh, p. 46 ff.) contended with much learning and ingenuity that the Greek title was to be restored here. In J. R. A. S., 1881, p. 526, he suggested —no doubt on the analogy of the coins of Miäus.

^{page 370 note 1} As the types of the silver coinage of the Kṣatrapas remain the same from this time to the end of the dynasty, it will not be necessary to repeat the description of them in every case.

The Rev. H. R. Scott informs me that there is a specimen of Caṣṭana's coinage, similar to the one here described, in the Watson Museum at Rajkot. It is “thinner than the ordinary Kṣatrapa coin, a little broader, and a few grains less in weight.”

^{page 371 note 1} The reading of the second syllable as -ṭa is not quite certain on any of the coins, but it is more probable than any other. The last syllable -sa on this coin is also very peculiar. Want of space is, perhaps, a sufficient explanation of these abnormal characters.

^{page 371 note 2} Dowson's, original reading, J.R.A.S., 1863, p. 221Google Scholar.

^{page 371 note 3} There is no certain trace of a Kharoṣṭhī inscription. It is inferred on the analogy of other coins. There is certainly not room enough for the whole inscription. It may possibly have been Raño Caṭanasa or even Raõ Chatrapasa Caṭanasa.

^{page 371 note 4} The actual name of Caṣṭana does not occur on this coin, but there can scarcely have been another Ghsamotikaputra.

^{page 373 note 1} Cunningham, , Coins of Ancient India, p. 94, pl. xGoogle Scholar.

^{page 374 note 1} “Of Rudradāman'a son, Dāmajadaśrī, we have two coins, neither with date, and both call Dāmajada Mahākṣatrapa.” (Rev. H. R. Scott.)

^{page 375 note 1} For the facts see the following paragraph on Jīvadāman, son of Dāmajada.

^{page 375 note 2} The date of this inscription is 103 as stated in Bh. There is no doubt on the subject, as the date is given both in words and figures. The reading on the facsimile of this inscription, given in Bhāvnagar Inscriptions, Pl. xvii, is undoubtedly [va]rṣe triuttaraśate = 103. Bühler, however, read dvyuttaraśate = 102. His statement (Ind. Ant., x, p. 157) that the date (102) is no new one, as it occurs on the coins, was no doubt the result of his own observation; but it appears not to have been published. The Rev. H. R. Scott tells me that a coin in the Watson Museum at Rajkot appears to be dated 102.

^{page 375 note 3} Rudrasīha seems to be the regular form of the name on the coins, though the vowel ī is not always visible. The same form is found in inscriptions, cf. Bühler, , Ind. Ant., x (1881), p. 157Google Scholar.

^{page 377 note 1} Two in the B.M.: Cunningham, 94, 5–7, 679; and Bh., 91.

^{page 378 note 1} So far as has been noticed, only one other instance is known, on a coin of Mkṣ. Svāmi Rudrasena, son of Mkṣ. Svāmi Rudradāman, published by Newton, in the Journ. Bomb. Br. R.A.S., 1862, p. 9, tig. 9Google Scholar.

^{page 383 note 1} “We have a specimen in this Museum in fairly good preservation, with inscription as figured in your plate, in well-shaped letters ‥‥ date 14x.” (Rev. H. R. Scott.)

^{page 385 note 1} It is extremely doubtful, however, whether such a gap exists (v. sup., p. 380).

^{page 386 note 1} Journ. Bomb. Br. R.A.S., 1862, p. 9.

^{page 386 note 2} Generally after the time of Caṣṭana; but the earlier form still appears occasionally on coins of a later date (inf., p. 397).

^{page 390 note 1} The Rev. H. R. Scott says: “We have a coin of Dāmasena's son Dāmajadaśrī which appears to be dated 173, in which case the gap mentioned by the Paṇḍit would be filled up and no room left for īśvaradatta between Vijayaaena and this king.”

^{page 394 note 1} Cf. Thomas, , J.R.A.S., 1850, p. 58Google Scholar.

^{page 396 note 1} Confirmed by the Rev. H. R. Scott's account of the coins in the Rajkot Museum.

^{page 398 note 1} J.R.A.S., 1882, p. 374.