^{page 295 note 1} Which I hope eventually to publish in continuation of my study of the Indian village communities and their tenures.

^{page 295 note 2} Or in the spoken form Chatrī. This, in fact, is the equivalent of ‘Kshatriya,’ and not the word Khatrī, which is also in use but indicates quite another caste. The latter has no real connection with the old military order, though sometimes attempts are made to assert such a connection. The objection to be called ‘Rājput’ is quite modern, the reason being that ‘Rājput’ now applies to a large group of caste-men who have become agricultural, and have even taken to cultivating with their own hands (saving only the touch of the plough—not that!). The higher families therefore desire some distinguishing name, and naturally assert that of the twice-born caste of old. Yet when the name Rājput was first used, it certainly was in a laudatory sense, meaning the royal or ruling race.

^{page 297 note 1} I permit myself to use the familiar form, for simplicity of printing, instead of the more accurate Kṛṣṇa.

^{page 298 note 1} Moreover, where some mythic story is introduced, impossible to connect with real persons, it is very likely to be an allegorical way of representing some real occurrence, which could not be explained.

^{page 298 note 2} Excluding, that is, the originally Moslem tribes and those non-Aryans who are conventionally called ‘aboriginal.’

^{page 300 note 1} In a quite recent authority I have seen it objected that the Hūṇa must be referred, not to the Huns of the fifth century, but to some other ‘Kshatriya’ race, because they appear in a text of Manu or in the Purāṇa, etc. But is there sufficient reason to believe that these texts, whether by origin or later recension, are earlier than the first few centuries of our era? So, too, when strange names in Hebrew appear for articles of commerce imported by King Solomon (1000 b.c.) from India, we immediately set about discovering Sanskrit origins for the words. What Sanskrit - speaking people had by that time so prevailed as to have made their speech the common language of commerce on the West Coast of India? What possible Sanskrit name (to take a single example) could there be for ‘sandalwood,’ when the tree does not grow in or near any country in which a Sanskrit-speaking people had established themselves? Maisur, the home of Santalum album, was not an Aryan country, and only became ‘Hinduized’ at a late date, when it acquired a Brahmanie dynasty. As a matter of fact we find the old Dravidian name of the product adopted into Sanskrit (as well as into every Indian dialect and even into Burmese) when the wood became an article of commerce throughout India. I do not say that the ‘almug’ or ‘algum’ of Scripture can be directly traced to a Dravidian name, as most of the other words can. We must make allowance also for the uncertainty of the kind of wood intended, and for the possibility of a name being transmuted (or substituted) by trading intermediaries from the coast of Arabia or Africa. But certainly there is no Sanskrit original.

^{page 301 note 1} See Gazetteer Cent. Prov., Introd., p. cxxyii. It was not till after the reign of Akbar that any considerable Hindī-speaking population extended beyond the Narbada districts.

^{page 303 note 1} I must not go into details about this curious country; but I may say in a footnote that the earliest (epic) legend places the kingdom in the hands of a tribe who are (in the Rgveda) closely connected with the Bharata, who take as their leader the (Kuśika) Viswāmitra. But before long we find the dynasty of Jarāsandha (always belonging to the Lunar genealogy, but in reality indicating some fusion of the Kuśika race with the Lunar Aryans) not only ruling Magadha, with its largely Kolarian population, but extending his power all round, overthrowing the Solar princes to the north-west of his dominion, and threatening Mathurā. Then in the course of time—perhaps owing to a general destruction of Lunar princes in the Great War—we find the Aryan dynasty replaced by one whose designation, S'eśnāga (or S'iśunāga), indicates a Turanian, serpent-worshipping, origin, and probably a connection with the Nāgbaṇsi houses which, at an undetermined date, established dominion over the Kolarian inhabitants of Chutiya Nāgpur and Eastern Central Provinces. The advent of these non-Aryan rulers, whether as a new importation or a revival of a power already in India, seems to synchronize with the ‘prophecy’ of the Vishnu-Purāṇa that pure Kshatriyā kings would cease, and the Yavana, Tuṣāra, etc., reign in their stead. At the end of the S'eśnāg times, the Maurya appear; and whether we accept the improbable Buddhist account (Max Müller, Hist. S. Lit., p. 283 ff.) or the Hindu, the originator was certainly a foreigner. He is found in connection with Taxila in the north; and that, in Alexander's time, was the capital, if not of the Takā, still of a serpent-worshipping king. It is apparently from Magadha that the foreign Andhra (Gens Andaræ of Pliny) originate, though they are known chiefly as dominating the northern Telugu country. Being Buddhists, Manu speaks of them with contempt (x, 36), as he does of the Liççhavi (x, 22), though the latter were of great power and dignity (Corp. Ins., iii, 135). Not only was there this strong influx of foreign rulers, but there must have been a large Kolarian (Magh) element in the population. Zimmer mentions that in the Atnarvanveda Magadha is alluded to as a ‘mixed’ race. (Altind. Leb., p. 216; see also p. 35.)

^{page 303 note 2} Mahāraṣṭra probably began to receive an Aryan (Yādava) element almost as soon as the Granges Valley, if not before it. I cannot readily adopt the derivation of the name from mahā = ‘magna regio.’ More probable is the origin from Mahār, the name of a once important Dravidian people whose relies still exist. The country is not mentioned by this name in the Mahābhārata (?), but much later in the Mahāvaṅso, in connection with the sending of Buddhist missions after the third synod, apparently in Asoka's time (Lassen, ii, 246). Varāha Mihira calls the people by this name. When the Mughals conquered the country (sixteenth and seventeenth centuries) they corrupted or changed the name of the people (Mahrātha) into Marhaṭṭa, which means ‘robber.’ Hwen Thsang speaks of the country as having a large capital, which perhaps means the seat of the West Chalukyā dynasty.

^{page 304 note 1} North Kanara District Manual, vol. ii, 25.

^{page 305 note 1} Such a fertile region as Jãlandhar (Trigarta) might (e.g.) have early been inhabited, either at once or by an early expansion, from the Ganges Valley settlement. I refer only to permanent settlements; the early Aryans, being largely pastoral, may have formed temporary encampments on the banks of the rivers.

^{page 305 note 2} No doubt in the course of time this idea of impurity would fade away. Adventurous princes seeking new settlements would soon disregard it, and in any case, could have found domains like those of the ‘Porus’ of Alexander's time, where they were not in contact with impure races. Later on the M. represents the Páṇḍava princes as making alliance with the Bāhikā, Madra, etc., of the Panjāb. Brahmans, too, would be tempted to return in order to extend the sphere of their influence, just as they penetrated into other ‘uncivilized’ regions. (See Lassen, Ind. Alt., ii, 181; Muir, A.S.T., ii, 482.) The Bombay Gazetteer (vol. i, p. 13, note) refers to a similar impurity, except in the case of pilgrimages, attaching to the more distant countries of Anga, Vanga, Kalinga, and Saurāṣṭra: (this last addition is very curious, since in the Rāmayana, Saurāṣṭra is an Aryan land). In Manu's time it would seem that the Panjāb was reckoned as an ‘Aryan country,’ since (ii, 19) ‘Āryāvarta’ takes in the whole land between the Himalaya and the Vindhya, as far as the ocean, both east and west.

^{page 306 note 1} I cannot say I believe in the Yādava approaching Gujarāt, etc., by sea. Why should they ? It is Dot, however, impossible that they sent trading vessels from the Indus mouths to the West Coast. But no other tribes came by sea. As to the formidable nature of the obstacle presented by the Vindhyan Hills, now much modified by road and railway, see my “Ind. Village Community,” p. 45.

^{page 307 note 1} We know as a matter of history how in the eleventh century Maḥmūd, coming through the Western passes beyond the Indus, crossed the river at Uççh, skirted round the desert, touching Ājmer and the Abu country, and thence moved directly on the plain towards the north of Gujarāt and came to Anhilpur, whence he crossed the peninsula to attack Somnãth. A similar route must have suggested itself to many other earlier adventurers. Others would, at possible seasons, have crossed the open desert and the ‘Iriṇa’ or Ruun.

^{page 308 note 1} It is not necessary for my purpose to take any note of the several names that are doubtful or disputed. The ‘dasyu’ or enemies are, I suppose, certain tribes in the Northern hills; for the Niṣāda or Bhīl tribes would hardly have been met with so far north of the Jamna forest region (see Zimmer as to the Parṇaka, Altind. Leb., p. 38). The Ṛgveda does not represent a stage of progress beyond the Ganges—I might say the Sutlej-Jamna. But already there were ‘Ahi’ or snake-worshipping tribes, as there were in the Panjāb, centuries later, in Alexander's time.

^{page 308 note 2} Zimmer, p. 122, collects the places of this mention: Ṛg., i, 108. 8; viii, 10. 5, etc.

^{page 309 note 1} The Tṛtsu are called the sacrifice-loving ‘Kshatriya sons of Sudās’ (ṛg., iii, 20. 7). As to their help from Indra and crossing the Jamna, see Zimmer, p. 126.

^{page 309 note 2} ṛg., vii, 8. 16.

^{page 309 note 3} This appears from the Rāmayana, ii, 71, v. 5. 6 (Lassen). And Zimmer, p. 127, refers to the Mahābhārata for the same. The place assigned in Lassen's map (in vol. ii) is quite opposed to this, and many, many miles too far east. But it is explained that the Matsya (also called Kirāta) afterwards extended further east to the neighbourhood of the Cedi.

^{page 309 note 4} Ṛg., iii, 33; but it is desirable that this should be cleared up.

^{page 310 note 1} The ‘Saraswati’ is so interpreted: Ṛg., vii, 96. 2.

^{page 310 note 2} Ṛg., viii, 74. 15.

^{page 310 note 3} Zimmer, p. 124.

^{page 310 note 4} The name Kuru does not occur in the Veda, but there are references to the Kauravya people: the Kṛvī are actually named; and we learn from the S'atapaṭha Br. that Kṛvī was the ancient name of the Pançāla people.

^{page 310 note 5} It is not of much use to enumerate the tribes and countries noticed, because it is so uncertain what is the date of each portion of the poem—whether any mention is in the original or of later interpolation. Taking the text as we have it, we find that the Panjāb and Sindh tribes are regarded rather as foreigners, though sons of the Aryan Yadu, Anu, etc. The Aryan home is the Madhyadeśa. The Upper Doāb (on both sides of the Jamna) has become the site of the kingdoms of Indraprastha and Hastinapura. The Southern Doāb is Pançāla. Benares (Kāśi) and the Matsya country are long settled. So is the Chambal (Çarmanvati) region, and part at least of what was afterwards called Mālwa. The Narbada Valley is occupied. East of that is the Çedi country and Mahākosala (Raipur and the Çhattīsgarh plain), also Berār (Vidarbha), and perhaps Kānbdeś. So is the Tapti Valley, the Upper Dakhan (Dakshmapatha), and Saurāṣṭra. Little notice is taken of Ayodhya or Kosala; vague reference is made to the Southern as also to the Eastern and North-Eastern Kingdoms, and even to Siṁhala or Ceylon. The serpent-worshipping tribes are in full possession, and make the subject of many legends. The Aryans have gained ground quite as much by marriages and alliances as by force of arms.

^{page 311 note 1} As to Anu's connection with Anarttā, part of Saurāṣṭra, there is a difference. The V.P. associates it and the city Kuśasthali with a son of one of Ikshwāku's brothers, from whom came Revata, whose daughter married Balarāma, brother of Krishna (Wilson, p. 354). But Tṛṇa, a descendant of Anu, gives rise to the (historic) Yaudheya tribe of the Lower Sutlej Valley, who were attacked by Rudradãman, the ‘Western Satrap,’ about 150 AD. In R.'s inscription (at Girnār) they are placed along with the Panjāb Mālava (Malli) and Madra (J.R.A.S., 1897, p. 885 ff.). The M. represents Yudhiṣṭhira as having a son called Yaudheya by a daughter of the king of the S'ibī (Lassen, i, 792).

^{page 311 note 2} This is stated, in so many words, in the V.P. (Wilson, pp. 175 and 177), which first mentions ‘Yavana,’ and a little further explains that the term includes the Sindhu, Sauvīra, Hūṇa, Sālva people, those of S'ākala, Madra, Ambasthā, etc.

^{page 312 note 1} Cf. Manu, ii, 19.

^{page 314 note 1} In chap, vii Manu calls Viswamitra the son of Gādhi, a Kauśika. He attained the Brahmanhood by his great humility.

^{page 314 note 2} And one of the names of Gādhipur or Kanyakubja was Kuśasthala (Lassen, i, 158, note 3).

^{page 315 note 1} See Wilson, V.P., 455, 456, and note. The Rāmayana calls him a Kuśika outright. Bharata has not only been adopted into the Lunar genealogy but also into the Solar; this will be discussed further on.

^{page 315 note 2} Lassen, i, 840.

^{page 315 note 3} I refer to the list of countries and people from the Bhishma Parva which appears in the V.P. (Wilson, pp. 178–185).

^{page 316 note 1} The V.P. further illustrates this distribution when it describes the ‘regions’ of India, and says of Upper India that on the west side of it dwell the Yavana (see note 2, p. 311), on the east side are the Kirāta, and in the middle “the four castes” (explained to mean the Kuru-Paṅçala; see Wilson, pp. 175–7). For the rest of India, the author mentions the regions of Mālwa, Saurāṣṭra, etc., as in the west; and the south includes the Pauṇḍra, Kalinga, etc.

^{page 317 note 1} See Arch. Rep. N. India, vol. ii, p. 16.

^{page 318 note 1} This secondary movement towards the old S'ūrasena region was very likely directed by the traditions of the tribes (see Beanies' Elliot, Gloss., i, 128); but the author did not notice that the occupation of the Yāduvaṭi was the consequence of the S'akã movements, and long centuries after the original S'ũrasena settlement (cf. Arch. Rep., ii, 21, 22). It was in connection with this invasion that (the Yādava) Sālivāhana, from his Dakhan home, made a brave but ultimately unsuccessful stand against the Indo-Scythians, defeating them at Kahror, sixty miles from Multān.

^{page 318 note 2} Arch. Rep., vol; xx, p. 1 ff.

^{page 319 note 1} See Lassen, i, 758. When Arjuna visits Kuśasthali (Dwārka), families of Vṛṣṇi (Krishna's Yādavas), Andhaka (another Yādava), and Bhoja families come out to greet him.

^{page 319 note 2} The V.P. (Wilson, p. 441) remarks on the great number of the Yādava branches.

^{page 319 note 3} See Aitareya Br., viii, 3. 3–14.

^{page 319 note 4} See Rajendralāla, “Aryans in India,” ii, 387. Wilson (V.P., p. 186, note) also concluded that the Bhoja were derived from [perhaps ‘related to’ would be better] the Yādava. Bhojakata, one of their cities on the Lower Narbada, was founded by Rukmi, Krishna's wife's brother, and Bhīṣmaka is called a Bhoja. The uncertainty of origin is reflected in the variations of the genealogists. The M. derives Bhoja (in the remote past) from Druhyu. Others take them, more proximately, from Kunti-Bhoja, father's sister's son of S'ūra. See Lassen, i, pp. 720 and 757, as to the connection; also Wilson, V.P., p. 418 (note 20) and p. 424. It must always remain doubtful whether the Bhoja are not Dravidians, contemporaneous like the Bharata, with the Aryans proper; and whether some were not united by adopting Brahmanic customs and by marriage, and so taken up and grafted on the Aryan (Yādava) stem.

^{320 note 1} Sahasra-arjuna is represented as deriding the Brahmans, “clad in skins,” because “they thought so much of themselves” (Muir, A.S.T., i, 462). From early times we begin to have mention of hermits from the North in the Vindhyan forests; a settlement in the Payoṣṇī (Tapti) Valley is also mentioned. Possibly the ‘heathen’ Haihaya mocked the Brahmans, who by their own clansmen or some local adherents avenged themselves. Paraśurāma is made contemporary with Rāma of Oudh by the V.P., which represents the latter as ‘humbling’ the former! Paraśurāma again appears as miraculously reclaiming the Malabar coast out of the sea; but this latter is a quite late legend to glorify the local Nambūṛi Brahmans. But possibly it is intended that we should take Paraśurāma as one of those sages who, like Vasistha and Viswāmitra, live through whole ages and appear when wanted: he represents a principle, or symbolizes the fact that now and again the priest merges into the warrior or fights his own battles. Even Brahman kings (in Kābul and Sindh) were not unknown to history, at least in the early centuries of our era.

^{page 321 note 1} The ancient (lower) capital was Mahiṣmatī (Mahesar in the Indore State). Some confusion is caused by the attempt to carry the reminiscence of this traditional seat further up the Valley to Maṇḍla: there ia no reason to believe that Garha Maṇḍla, or any other place there, was ever called Mahiṣmatī.

^{page 321 note 2} Ṛg., viii, 5. 37.

^{page 321 note 3} Vol. i, 744, note.

^{page 321 note 4} See Arch. Rep., ix, 77. It will be observed that the genealogies (Table II) make them Yādava, but in another branch. In a later volume (Arch. Rep., xvii, 71) Cunningham says that the Cedi in the oldest Rājim inscription (Raipur District, Central Provinces) do not refer to H., but call themselves sons of Kuru. This may have some special justification, but certainly, in general, the Haihaya are alluded to as ‘Cedi swāmi’ =lords of the Cedi; and the Kalachuri always accounted themselves Haihaya (see Arch. Rep., ix, p. 92, and Corp. Ins., iii, Introd., p. 10). The Gujarãt bards sometimes insert the Kālachuri (in some corrupt forms) among the “thirty-six royal clans.” They were of high rank, and married into families of Mewār, and the Mālwa Pramāra, in the twelfth century.

^{page 321 note 5} I can only just allude to the illustration afforded by Sleeman's account of the ‘Gond princes’ (not that the princes were Gond, but that they ruled over Gondwāna, and left many illegitimate descendants called Rāj Gond). The tradition originates the dynasty in one ‘Jādava-rāya’ (the names suggest a Yādava tribal connection) who was serving a Haihobaṅsi chief. He transferred his services to a local Nāgbaṅsi prince, married the prince's daughter (and only child), and succeeded to the kingdom. This, Sleeman says, was in ‘Samvat’ 415 (presumably Çedi era) or late in the seventh century (Sleeman makes it Vikram: which is not admissible).

^{322 note 1} While speaking of the Yādava and cognate tribes, it is impossible to omit all mention of the Ahīr, apparently the same as the Abhîra, who are certainly a non-Aryan, perhaps early Indo-Scythian tribe, from the north-west. Now, they hold no place in the Panjāb, but in the N.W.P. in Ahīrwāra and the Upper West of India. They are not now esteemed, but once furnished princes both in Nepal and on the West Coast. One list at least, gives ‘Abhīra,’ a place in the ‘thirty-six royal clans.’ (See J.R.A.S., October, 1897, p. 890, and Arch. Rep., ii, 300.) In the N.W.P. the Ahīr are still divided into great, and quite independent, sections—Jadubaṅsi, Nandabaṅsi, and Gwālbaṅsi (Elliot, i, 3). Nanda, with whom Krishna took refuge, was an ‘Ahīr’ chief. The author of the Prabandh Çintāmani (cireâ 1305 a.d.), in relating the story of the Yādava Rāvs of Junāgarh, called Navaghana “the Ahīr Rāna” (Raśmāla, p. 118), while an inscription calls the same family “Yādava.” In the Central Provinces, where the Ahīr are numerous, they seem to be connected with some traditional “Gāoli” of former importance, and their name survives in Gwāliūr, Gāvalgarh, Gol-khanda, etc. The Abhīra appear in the Samudra Gupta inscription, and are the people of Ptolemy's Abhiria.

^{page 323 note 1} I do not mean to imply that even the Eastern Aryans very soon, or very completely, acquired strict caste ideas. Not only have we repeated allusions to sages and heroes marrying ‘serpent’ and other strange, not to say ‘inferior,’ wives, hut a more general laxity long prevailed. Mr. J. F. Hewitt (J.R.A.S., 1889, p. 196) has justly called attention to a conversation in the M. in which. Yudhiṣṭhira says to Nahuṣa that “in human society it is difficult to ascertain one's own caste, because of the promiscuous intercourse among the four orders. Men belonging to all the orders have children by women of all the orders.” This, however, shows that the idea of established ‘orders’ and of the propriety of caste distinctions, existed; and that is more than can probably be said for early times in Western India.

^{page 324 note 1} In the M. we hear of Brahmans in the Payoṣṇi (Tapti) Valley, and even as far as Gokarna on the West Coast. Among the Kshatrapas we notice Ushavadatta (probably ruling during the first quarter of the second century). Though a S'akā by birth, he is believed to have adopted the Brahmanic faith. The inscriptions, with pardonable exaggeration, record how he gave 300,000 cows and other wonderful gifts to Brahmans. He fed, we are told, “hundreds of thousands” of Brahmans every year, which is obvious nonsense, because such a number could not have been in existence in the West. It cannot be supposed that Brahmans accompanied the Yadava from the Indus; so that none could be found but hermit wanderers in small groups from the North. (See Bombay Gazetteer, vol. i, pt. 1, p. 125.)

^{page 325 note 1} I may refer to the tradition or legend of the Kādamba king Mayura-varma, or Mayuraśarma, of N. Kanara, introducing eighteen ‘agrahāra’ of Brahmans, which could not have been before the seventh century. (See Bom. Gaz., vol. i, pt. 2, p. 560.) As regards Mr. Baine's remark (Census Rep., 1891, Parl. Blue Book, p. 141) that the Mahrāthi language is peculiarly Brahmanic, any such feature could have been taken on, just as the introduction of Christianity affected the vocabulary, etc., of many languages that had long been fixed.

^{page 325 note 2} Nor need I argue about S'iva and phallic worship being of Dravidian origin. The whole genesis of S'iva worship and the rival Viṣṇu sect is curious. The Krishna cult seems to have been invented as a bid for popular favour for the Viṣṇu side against the Saivite.

^{page 326 note 1} And it is not necessary to say that caste rules had a certain moral utility they had their good aspect as well as their bad.

^{page 326 note 2} It is notorious to the present day how any ascetic teacher or hermit will gather a following. The effect of shrines and the ‘melas’ connected with them is equally well known.

^{page 327 note 1} We have examples in the Central Provinces and elsewhere of how the Nāga chiefs discarded their ancient symbol, and accepted ‘Rajput’ rank and an impossible genealogy for some Epic or Puranic hero. Such princes adopt caste observation with extreme strictness. After some few generations (backed by wealth and success) they are admitted to marriage alliances with Rājput houses of more established reputation.

^{page 327 note 2} With reference, for example, to the possibility of some real descendant of Pāṇḍu, or other Aryan Kshatriya of the North, finding his way to power in the Southern ‘Pandyan’ kingdom.