Hostname: page-component-cd9895bd7-lnqnp Total loading time: 0 Render date: 2024-12-18T14:25:29.191Z Has data issue: false hasContentIssue false
  • English
  • Français

Art. I.—The Kushān, or Indo-Scythian, Period of Indian History, b.c. 165 to a.d. 320

Published online by Cambridge University Press:  15 March 2011

Vincent A. Smith
Get access Rights & Permissions [Opens in a new window]

Extract

The four epochs into which, the political history of ancient Northern India is naturally divided are marked by four imperial dynasties — (I) the Maurya, (II) the Kusana or Kushān, (III) the Gupta, and (IV) the Rājpūt line of Harṣavardhana. The date of the Maurya, the earliest of the four dynasties, was practically determined more than a century ago by Sir William Jones, and we know that Candra Gupta Maurya, the first emperor of India, and grandfather of Aśoka, ascended the throne in or about the year b.c. 321.

Type
Original Communications
Information
Journal of the Royal Asiatic Society , Volume 35 , Issue 1 , January 1903 , pp. 1 - 64
Copyright
Copyright © The Royal Asiatic Society 1903

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

page 1 note 1 In Indian inscriptions and Kharoṣthī coin legends the name is written Kuṣana or Guṣana. The spelling Kushān is authorized by the legends on certain Sassanian coins. (Drouin, , Rev. Num., 1896, p. 170Google Scholar.) The coin legends in Greek supply other variants.

page 3 note 1 For the Vikrama theory, see Cunningham, , Reports, ii, 68, noteGoogle Scholar; iii, 30, etc. The Seleucidan theory is expounded in the same author's “Book of Indian Eras”, 42. M. Sylvain Lévi has ingeniously explained his hypothesis in “Notes sur les Indo-Scythes” (J.A., Nov.–Dec, 1896, Mai–Juin, 1897; reprinted 1897, see especially pp. 62, 66, 82).

M. Boyer, with most of whose reasonings I agree, published his excellent paper “L'Époque de Kaniska” in the J.A., Mai–Juin, 1900.

Mr. D. R. Bhāndārkar developed his peculiar views in a paper read before the Bombay Branch of the Royal Asiatic Society, on the 19th October, 1899, entitled “A Kushaaa Stone-inscription and the question about the origin of the Śaka Era.” These views were adopted by Mr. R. G. Bhāndārkar, in his interesting essay, also published in the Journal of the Bombay Society, and reprinted in 1900, which is entitled “A Peep into the Early History of India from the foundation of the Maurya Dynasty to the downfall of the Imperial Gupta Dynasty, b.c. 322–circa 500 a.d.”

MrRapson, writes in “Indian Coins”, p. 18Google Scholar: “Regarding another possibility, viz., that the dates of Kaniṣka and his successors have to be referred to the second century of the era used by Śodāsa, Liako Kuśuluka, and others, see W.Z.K.M., ix, p. 173 f.” See also MrRapson's, remarks in J.R.A.S., 04, 1900, p. 389Google Scholar. My guess that a.d. 60 or 65 might prove to be the desired date was hazarded in my recently published work, “The Jain Stūpa and other Antiquities of Mathurā”, pp. 5, 46. I withdraw it unreservedly. The conjecture that Kaniṣka founded a special era of his own has been often mentioned, but has never been worked out in a definite shape. Fergusson's, and Oldenberg's, theory is fully stated in I.A., x, 213Google Scholar, in the article entitled “On the Dates of Ancient Indian Inscriptions and Coins”, translated from the eighth volume of the Zeitschrift für Nttmismatik (Berlin, 1881)Google Scholar. Professor Gardner, (B.M. Catal., Indo-Scythian Kings, p. 2)Google Scholar was convinced that this theory “settled the matter”, and that no numismatist could hesitate to accept it. M. Boyer's article above quoted, and his paper “Nahapāna et l'ère Śaka” in J.A., Juillet-Août, 1897, as well as the remarks of the Messrs. Bhāndārkar and M. Lévi, demolish the theory.

page 5 note 1 The Yueh-chi were not Mongoloid in race. The Chinese authors describe them as having a pink and white complexion, and the portraits of the kings on their coins exhibit large noses. They have been “wrongly assimilated to the Tochari and to the polyandric Ye - ta or Ephthalitai, which are different races altogether. These confusions were set right in the first case by Professorde Vasconcellos-Abreu, G. (Le Musćon, 1883)Google Scholar, and in the second by MrSpecht, E. (Journal Asiatique, 1883)Google Scholar” (De Lacouperie, , in Academy, 12 31, 1887Google Scholar).

The modern Chinese pronunciation of the name is said to be Yuèti (Kingsmill). It is given as Yueh-ti by M. de Lacouperie, who states that the original form was Gwet-ti. M. Lévi and other French scholars write Yue-tchi, or Yué-tchi. English scholars incline to the form Yueh-chi, which I have adopted.

MrKingsmill's, interesting and valuable papers, “The Migrations and Early History of the White Huns, principally from Chinese sources” (J.R.A.S., 05, 1878)Google Scholar, and “The Intercourse of China with Eastern Turkestan and the adjacent countries in the second century b.c.” (ibid., January, 1882), unfortunately confound the Yueh-chi with the white Huns or Ephthalites.

page 15 note 1 “Alberuni's India”, transl. Sachau, , i, 302Google Scholar. My quotations are from this work. Cunningham necessarily used Reinaud's “Fragments Arabes et Persans”.

page 15 note 2 Minayeff, (“Recherches sur le Bouddhisme”, p. 229, note)Google Scholar points out that the Śaka year began with Caitra, whereas the Buddhists used the reckoning by the three seasons, in which the year began with winter. He cites Yaśomitra as saying: . The evidence above summarized proves that Buddhist practice varied. Cunningham's table of the Laukika era is based on the Kaśmīr practice, which began the year with Caitra.

page 16 note 1 “Alberuni's India”, transl. Sachau, , ii, 8, 9, 54Google Scholar.

page 16 note 2 “Book of Indian Eras”, p. 6; Epigraphia, Indica, i, 97–118, especially 103. The Laukika era is not expressly named in the inscriptions, but is undoubtedly referred to. If Dr. Fleet is right, as he probably is, in reading the Baijnāth Śaka date as 926, the Laukika equivalent is 4080 = a.d. 1004–5.

page 17 note 1 Cunningham, : “Book of Indian Eras”, p. 41Google Scholar; and N.C. for 1892, p. 44.

page 19 note 1 Boyer, , “L'Époque de Kaniṣka”, p. 533 f.Google Scholar, in J.A., Mai–Juin, 1900; Lévi, , “Notes sur les Indo-Scythes”, reprint, p. 48Google Scholar; Kingsmill, op. cit.

page 19 note 2 Giles, : “A History of Chinese Literature”, p. 102Google Scholar. The Chinese author's name is spelt Ssῠ-ma-Ch'ien by Professor Giles, Seu-ma Ts'ien by M. Boyer, and in various ways by other authors.

page 21 note 1 The extract from the History of the First Han Dynasty is most fully given by M. Boyer (op. cit., p. 541). That work covers the period down to a.d. 23 or 24, the date of the death of the usurper Wang-Mang, whose reign divides the Han period into two parts. Pan-ku, who died in a.d. 92, had exceptionally good sources of information. According to Professor Giles (op. cit., p. 108), his work was completed by his sister Pan-chao. M. Lévi explains that the variant reading Kien-chi (or Kien-cheu) as the name of the capital city is due to the confusion of two almost identical characters. The correct reading is Lan-cheu, in French spelling. The correct name, Tou-mi, for the fifth principality, is given in the History of the Second Han Dynasty. The mention of Kābul in the History of the First Han Dynasty is an obvious blunder, inasmuch as Kābul was not annexed by the Yueh-ehi until the reign of Kadphises I, more than a century after the formation of the five principalities.

M. Sylvain Lévi's identification of Lan-sheu with Puṣkalāvatī (Peukelaïtis), near Peshāwar in India, is absolutely impossible. It rests solely on the fact that long afterwards, in the seventh century, Puṣkalāvatī is described as being, or having been, the Yueh-chi capital, and on the guess that Lan-sheu might possibly be a translation of the Sanskrit Puṣkalāvatī. (“Notes sur les Indo-Scythes”, pp. 49, 82.) The Yueh-chi had not annexed either Kābul or Ki-pin, which lay between the Ta-hia territory and India, at the time when they occupied the Ta-hia capital, Lan-sheu. The Chinese texts clearly indicate that the first capital of the settled Yueh-chi was to the north of the Oxus river, and that their second capital, Lan-sheu, was in the Ta-hia country, that is to say, Bactria, south of and not remote from the river. Balkh is in the required position.

The references to Puṣkalāvatī are collected in the note to Beal's Hiuen Tsiang, i, 109. Mr. Kingsmill writes the name Lan-sheu as Lam-shi-ch'eng, and identifies it with the Δάραψαof Strabo.

page 22 note 1 M. Sylvain Lévi appears to be certainly right in identifying the Se with the Śakas. The Chinese character used to denote the Se nation is regularly used as the equivalent of the Sanskrit syllables -saka in the words upāsaka and mahīśāsaka. (“Notes sur les Indo-Scythes”, p. 50.) My statement that the expulsion of the Śakas from their territory by the Yueh-chi preceded the defeat of the Yueh-chi by the young Wu-sun chief rests on the Sse-ki and the report of Chang-k'ien as summarized by Lévi, M. (Notes, p. 53)Google Scholar, which is perfectly explicit, and nearly contemporary with the event. The notice of Ki-pin in the much later History of the First Han Dynasty (ibid., p. 50) states with less accuracy that “formerly when the Hioungnou conquered the Ta-Yue-tchi, the latter emigrated to the west, and subjugated the Ta-hia; whereupon the king of the Se [Çakas], went to the south, and ruled over Ki-pin”.

page 23 note 1 For instance, ProfessorGardner, Percy in B.M. Catal. of Coins of Greek and Scythic Kings (Introd., p. xxxi)Google Scholar; and Cunningham, , N.C. for 1892, p. 41Google Scholar.

page 23 note 2 “L'Époque de Kaniṣka”, p. 544. M. Sylvain Lévi is, I think, in error when he takes the subjugation of the Ta-hia as the point from which the calculation should be made. He refuses to acknowledge that there were two stages in the conquest of the Ta-hia. M. Boyer has rightly perceived that the political subjugation of the Ta-hia, which quickly succeeded the defeat of the Yueh-chi by the young Wu-sun chief, was separated by a long interval of time from the occupation by the Yueh-chi of the Ta-hia lands and capital south of the Oxus.

page 24 note 1 That is to say, ‘more than a hundred years later’. The Chinese characters indicate ‘une différence en excès’. (Boyer.)

page 24 note 2 M. Sylvain Lèvi informs me that the syllable tai, which he inserted on the authority of M. Specht, does not really form part of the king's name, which is expressed by three Chinese characters only. M. de Lacouperie writes the name Po-ta as Puk-ta, and interprets it as meaning Bactria. The name A-si is, according to him, properly written An-sik, which is said to be the regular equivalent for Arsak. (“Une Monnaie Bactro-Chinoise”, from Comptes Rendus, Acad. B.L. et Inscr., 1890, p. 12.) The French spelling of proper names is retained in my English version.

page 24 note 3 The texts concerning the mission of King-lou have given rise to much discussion, which has been settled by M. Svlvain Lévi, who sums up with the remark that “quelle que soit la recension adoptée comme base, la critique et la tradition n'admettent qu'une seule interprétation: En l'an 2 avant J.-C, le roi des Yue-tchi était bouddhiste, et son zèle travaillait à propager la religion du côte de la Chine” (“Les Missions de Wang Hiuen-Ts'e dans l'Inde”, Paris, 1900, p. 112Google Scholar; reprinted from Journal Asiatiqne Mars-Avril et Mai-Juin, 1900). M. Boyer writes to the same effect: “D'abord quant aux textes, rangés sous deux versions, qui se rapportent à King-lou = King-hien, j'admets qu'il en ressort que, deux ans avant J.-C, la Chine reçut du pays des Ta Yue-tchi des ouvrages buddhiques” (“L'Époque de Kaniska”, in J.A., 05–06, 1900, p. 527)Google Scholar. For some account of Buddhism in Khotan see DrStein, 's “Preliminary Report on a Journey of Archæological and Topographical Exploration in Chinese Turkestan”, London, 1901Google Scholar; and several passages in Hiuen Tsiang's travels.

page 26 note 1 M. Boyer (op. cit., p. 536), quoting the Sse-ki, observes that the Yueh-chi, “ayant trouvé, venant de l'est, sur la rive droite de l'Oxus, une terre fertile (cf. fol. 2 v°), occupée par les Ta-hia, ils livrèrent bataille, et, la victoire gagnée, ‘ensuite eurent leur capitale au nord du Koei-choei [Oxus], et y établirent la cour rovale’”.

page 27 note 1 This date is adopted by M. Lévi (op. cit., p. 53). de Lacouperie, M. (Academy, 12 31st, 1887)Google Scholar states the date as “about 143 b.c.”

page 27 note 2 MrKingsmill, (“The Migrations”, p. 17)Google Scholar clearly perceived that the conquest of Bactria by the invading tribes was “probably gradual rather than sudden”.

page 28 note 1 The chronology of the Northern Satraps and the Indo-Parthian kings will be further discussed in a later section of this paper.

page 28 note 2 The Chinese writers, as the extracts cited prove, distinguish Ki-pin from Kao-fu, or Kābul. I agree with M. Sylvain Lévi, in opposition to M. Specht, that the Ki-pin of the Chinese must be identified with the Kapiśa of Tsiang, Hiuen (J.A., ser. ix, t. vii, p. 161Google Scholar; and ibid., t. x, “Note Additionelle”, pp. 526–531.) Tsiang, Hiuen (Beal, i, 57Google Scholar; Julien, i, 42) states that the kings of Gandhāra spent the autumn and spring in Gandhāra and the summer in Kapiśa. The later pilgrim Ou-k'ong states that the town of Gandhāra is the eastern capital of Ki-pin, where the king resides in winter; he spends the summer in Ki-pin. The Ki-pin of Ou-k'ong is therefore the Kapiśa of Hiuen Tsiang. Similarly, Ou-k'ong places the convent of the Śrāmanera in Gandhāra, while Hmen Tsiang places it in Kapiśa, (Beal, i, 63)Google Scholar. The identification of Ki-pin with Kophēnē, or the Kābul, region (Beal, i, p. c)Google Scholar, must be given up. The kingdom was sometimes spoken of by the name of the province of Gandhāra, where the winter capital was situated, and sometimes by the name of the province of Kapiśa, Northern Afghānistān, where the summer capital was located in the hills.

page 29 note 1 Scholars are now generally agreed that Kujula Kadphises is identical with Kozola Kadaphes. Kadphises, Kujula Kara (Num. Chron. for 1892, p. 66)Google Scholar, although distinguished by Cunningham, is evidently the same person. Inasmuch as that person is undoubtedly the first Kuṣana king who struck coins to the south of the Hindū Kūsh, and it is possible to identify his Indian names with the Chinese, he must be the same as K'iu-tsiu-koh. The subject is well discussed by Boyer, M. in “L'Époque de Kaniṣka”, pp. 550–564 (J.A., Mai–Juin, 1900)Google Scholar.

page 30 note 1 The most convenient summary account of the Kuṣana coinage is that given by MrRapson, in “Indian Coins” (1897), pp. 16seqq.Google Scholar, which furnishes abundant references to all publications on the subject. A table of weights and assays of the gold coins will be found in Cunningham's, “Coins of Mediaeval India”, p. 16Google Scholar. It is just possible, as Mr. Eapson informs me, that the head on the coins of Kadphises I may be copied from that of Tiberius, which is very similar to the Augustan portrait. But the Kusana head most closely resembles the portraits on the coins of Gaius (Caius) and Lucius, the grandsons of Augustus, who died respectively in b.c. 4 and a.d. 2. See Von Sallet, , Zeit. für Num., 1879, pp. 218, 378Google Scholar; Nachfolger Alexander's, p. 180; Von Gutschmid, , Gesch. Iran, p. 136Google Scholar.

page 30 note 2 The Chinese statement that Yen-kao-chin was the son of K'iu-tsiu Kioh, or Kadphises I, precludes the hypothetical insertion of any king between Kadphises I and Kadphises II. Nobody, I think, disputes the identity of the Yen-kao-chin of the Chinese with Hima, or Wema , Kadphises, who is, for convenience, designated as Kadphises II.

page 30 note 3 The Chinese general Pan-tch'ao (Pan-chao) was engaged from a.d. 73 to 102 in reducing the western countries to obedience. In a.d. 73 he obtained the submission of the King of Khotan. In a.d. 90 the Yueh-chi king demanded a Chinese princess in marriage. The demand being resented, the Yueh-chi king raised a force of 70,000 horse under the command of a general named Si, who was defeated by the Chinese after his army had suffered severely in the passage of the Tsung-ling mountains. From that date the king of the Yueh-chi regularly sent tribute to China. (Lévi, op. cit., p. 65; quoting De Mailla, , “Histoire Générate de la Chine”, transl. from the Toung-tien-tan mou, pp. 365seqq.)Google Scholar

page 31 note 1 The unique coin described by Cunningham, (Num. Chron. for 1892, p. 71, pi. xv, fig. 14)Google Scholar, which exhibits a bust with two faces, the left face being bearded, with the Kadphises symbol in front, while the right face is smooth, with the symbol of the Nameless King in front, is conclusive evidence that the two princes were contemporary. Cunningham specifies four details in which the coins of both agree. “The coins of both kings are common in the Kābul Valley, throughout the Panjab, and in N.W. India, as far east as Benares and Ghazipur”.

page 31 note 2 Dion Cassius, ix, 58; in McCrindle, , “Ancient India as described in Classical Literature”, p. 213Google Scholar. The exact date of the Indian embassy is not recorded.

page 31 note 3 The evidence for the war with the King of Pāṭaliputra is the tradition in the Śrī Dharma-piṭaha-saṁpradaya-nidāna. (Lévi, , Notes, reprint, p. 37Google Scholar.)

page 32 note 1 The name of the second king is ordinarily written in the inscriptions as Huviṣka, but the forms Huvaṣka, Huvikṣa, and Huksa also occur in the Mathurā records. The Rājatarañginī uses the form Huṣka. The third king's name, which is commonly written Vasudeva, is written Vāsuṣka in the Sāñci inscription of the year 78, and in one Mathurā inscription of the year 76. He is called Juska in the Rajatarañginl. The Greek legends on the coins give other variants of the Kuṣana names, which evidently presented difficulties of transliteration to Indian and Greek writers.

page 34 note 1 Pliny, , Mist. Nat., xii, c. 18 (41)Google Scholar, transl. McCrindle, , in “Ancient India as described in Classical Literature”, p. 125Google Scholar; Cunningham, , “Coins of Ancient India”, p. 50Google Scholar; Thurston, Catalogue No. 2 of Coina in Government Central Museum, Madras; Rapson, “Indian Coins”, sees. 14, 69, 123.

page 37 note 1 Kielhorn, , “A List of the Inscriptions of Northern India from about a.d. 400”, being an appendix to Epigraphia Indica, vol. vGoogle Scholar. The inscriptions referred to are Nos. 351, 352, and 362. References to the original authorities will be found in Professor Kielhorn's List. On the western origin of the Śaka era see Boyer's, M. paper on Nahapāna and the Śaka era in Journal Asiatique, 06–12, 1897, pp. 120–151Google Scholar. M. Boyer points out that in both the Baijnāth and the Deogaṛh inscriptions the Śaka date is a secondary one. The primary expression of the date at Deogaṛh is in the Vikrama era. At Baijnāth, the first praśasti is primarily dated in the Laukika era (Ep. Ind., i, 103)Google Scholar.

page 38 note 1 The only certainly dated piece in the Bactrian series of coins is the unique coin of Plato, (B.M. Catal., p. 20)Google Scholar, dated in the year 147, referable to the Seleucidan era. The marks on other coins which Cunningham read as dates with the hundreds omitted are not accepted as such by most numismatists.

page 38 note 2 This inscription is briefly alluded to by its discoverer, Dr. Führer, in his Progress Report for the year ending 30th June, 1896. It is discussed, transcribed in English characters, and translated by Biihler in a letter dated April 10th, 1896, of which I possess a printed proof. I am not certain where it appeared. I have not seen a facsimile of this puzzling record.

page 39 note 1 The texts and translations in the Reports are fall of errors. For the script, numerical symbols, language, and formula, cf. “Gupta Inscriptions”, Nos. 72, 76; the Kudā, inscriptions in Arch. Surv. W.I., iv, pp. 12–14, 85, 86Google Scholar; and inscribed statuettes from Bundelkhaṇḍ, in J.A.S.B., lxiv, pt. i, p. 160Google Scholar.

page 40 note 1 Reports, v, 59, pi. xvi, fig. 3; “Notes d'Épigraphie Indienne”, No. iii, p. 11.

page 41 note 1 For the date and coins of Gondophares, see Rapson, “Indian Coins” see. 62; Gardner, , B.M. Catal., p. xlviGoogle Scholar; Von Sallet, , Nachfolger, pp. 157, 221–230Google Scholar; Sylvain Lévi, “Notes sur les Indo-Scythes”, reprint, pp. 67–82. M. Lévi is ingenious, but unconvincing. Von Sallet, relying on the fact that the king's tiara on the coins of Sanabares appears to be copied from the Parthian tiara of a.d. 77 or 78, places Gondophares in a.d. 60–80, a little later than I do.

page 42 note 1 Cunningham, , Num. Chron. for 1892, p. 66, Nos. 1, 2, pi. xiv, figs. 9, 10Google Scholar; Gardner, , B.M. Catal., pl. xxiii, fig. 7Google Scholar.

page 42 note 2 The Panjtār inscription was discovered in 1848 by Cunningham, but both it and the Ohind inscription (No. 48 of my list) were soon afterwards lost. They are known only from the imperfect editions in Reports, v, 58, 61. The Takht-i-Bahāī inscription is in the museum at Lahore. The Loriyān Tangai, Hashtnagar, and Kaladara inscriptions are noticed together by Burgess, in J.I.A. for 01, 1900, p. 89Google Scholar. The Hashtnagar date has been variously read, but is certainly 384. It is really perfectly plain. The date is wrongly given at p. 32 of the J.I.A. for July, 1898.

page 49 note 1 The principal references are:—Gardner, B.M. Catal. Greek and Scythie Kings, Introd.; Rapson, , “Indian Coins”, pp. 5–9Google Scholar; Von Sallet, “Die Nachfolger Alexanders des Grossen”; Cunningham, , “Coins of Ancient India”, p. 87Google Scholar; Bhāndārkar, D. R., “A Kushana Stone Inscription”, pp. 15–25Google Scholar; Bhāndārkar, R. G., “A Peep into the Early History of India”, pp. 17, 24Google Scholar. The spelling of the name of the Satrap Rājuvula varies. I have followed the spelling used by Paṇḍit Bhagvān Lāl Indrajī, who has discussed (with Mr. Rapson's and Dr. Bühler's aid) the inscriptions on the lion capital and the coins of the Northern Ksatrapas, in J.S.A.S., 07, 1894, pp. 524–554Google Scholar. I am now satisfied that the date of Soḍāsa's Mathurā inscription is 72, not 42.

page 50 note 1 “Græco-Roman Influence on the Civilization of Ancient India”: J.A.S.B., vol. Iviii, pt. 1, pp. 149, 172.

page 51 note 1 J.R.A.S., 1899, p. 422. The name is sometimes written Takht-i-Bahi.

page 51 note 2 Cole, : “Second Report of the Curator of Ancient Monuments in India”, 1883, p. cxxGoogle Scholar. When writing my paper in 1889 I unfortunately quoted the Third Report, which accidentally misrepresents the details as to the position of the Kaniska coins at Sanghao. I saw that an error existed, but could not point out its nature. The difficulty has now been cleared up by comparison with the Second Report, which is perfectly intelligible.

page 51 note 3 Cunningham, : Reports, v, 194Google Scholar. The name is also written Jamālgiri.

page 51 note 4 “Preliminary Report on Archaeological Exploration in Chinese Turkestan”, London, 1901Google Scholar.

page 54 note 1 “Dialogues of the Buddha”, pp. 54, 55, 240.

page 54 note 2 My views as to the origin and meaning of the Mahāyāna hare been partly suggested by Minayeff's, “Kecherches sur le Bouddhisme”, Paris, 1897Google Scholar. The Hashtnagar inscription proves that fairly good work of the Gandharian school was done in the fourth century a.d., but the best work is all earlier than a.d. 300.

page 56 note 1 The Yueh-chi continued to be a great power in Asia long after their dominion in India to the east of the Satlaj had been superseded by the conquests of Samudra Gupta about a.d. 340. A curious notice of India by the śramana Kālodaka, translated into Chinese in a.d. 392, enumerates four ‘Sons of Heaven’ (Chinese, t'ien-tzen; Sanskrit, devaputra) as existing in the world, namely, the emperor of China (Tsin) in the east, the emperor of India in the south, the emperor of Rome in the west, and the emperor of the Yueh-chi in the north-west. The Chinese dynasty at that date was that of the Eastern Tsin, a.d. 317–420. (Lévi, , “Notes sur les Indo-Scythes”, p. 64, noteGoogle Scholar.) In the year 392 Candra Gupta II of the Gupta dynasty was emperor of India, and at the height of his power. The Yueh-chi empire was shattered and broken up into small principalities by the shock of the Ephthalite or White Hun invasions in the fifth century. These small states continued to exist up to the time of the Arab conquest of Turkestan. (Drouin, , “Monnaies des Grands Kouchans”, Rev. Num., ser. iii, t. xiv, p. 171Google Scholar.) The White Huns also destroyed the Gupta empire in India.