The Markandeya Purana

by Frederick Eden Pargiter | 1904 | 247,181 words | ISBN-10: 8171102237

This page relates “the geography of bharata” which forms the 57th chapter of the English translation of the Markandeya-purana: an ancient Sanskrit text dealing with Indian history, philosophy and traditions. It consists of 137 parts narrated by sage (rishi) Markandeya: a well-known character in the ancient Puranas. Chapter 57 is included the section known as “exposition of the manvantaras”.

Canto LVII - The Geography of Bhārata

Mārkaṇḍeya mentions the nine divisions of Bhārata, one of which is India—He mentions the seven mountain ranges in India (exclusive of the Himālaya Mountains) and names twenty-two separate hills— He mentions the chief rivers in India, grouping them according to the mountain ranges out of which they rise—He mentions the chief peoples in India and on its borders, arranging them according to the main natural divisions of the country—and he concludes with general descriptive remarks and an encomium on India as the sole land of action.

Krauṣṭuki spoke:

Adorable Sir! thou hast fully described this Jambudvīpa. Just as thou hast declared it, merit-producing action exists nowhere else, nor action that tends to sin, except in Bhārata, O illustrious Sir! And from this land both Svarga is attained, and final emancipation from existence, and the medium end also. Verily nowhere else on earth is action ordained for mortals. Therefore tell me, O brāhman, about this Bhārata in detail, and what are its divisions, and how many they are, and what is its constitution accurately; it is the country,[1] and what are the provinces and the mountains in it, O brāhman?

Mārkaṇḍeya spoke:

Hear from me the nine divisions of this country Bhārata; they must be known as extending to the ocean, but as being mutually inaccessible. They[2] are Indradvīpa, Kaśerūmat,[3] Tāmravarṇa,[4] Gabhastimat, and Nāgadvīpa, Saumya, Gāṇḍharva,[5] and Vāruṇa; and this is the ninth dvipa among them, and it is surrounded by the sea.[6] This dvīpa is a thousand yojanas from south to north.[7] At its east end are the Kirātas,[8] and at the west the Yavanas.[9] Within it dwell brāhmans, kṣatriyas, vaiśyas and śūdras, O brāhman. They accomplish their purification with the occupa tions of sacrifice, meditation, trade, &c.; and they seek their mutual business through these occupations, and they gain Svarga or final emancipation from existence, merit and sin then.

The seven mountain ranges[10] in it are Mahendra,[11] Malaya,[12] Sahya,[13] Śuktimat,[14] the Ṛkṣa mountains,[15] and Vindhya,[16] and Pāripātra.[17] And there are other hills besides them in thousands, which are situated 12 near them. Their summits are broad and lofty, and are delightful and spacious;—Kolāhala,[18] and Vaibhrāja,[19] Mandara,[20] the hill Durdura,[21] Vātasvana,[22] and Vaidyuta,[23] Maināka,[24] and Śvarasa,[25] Tuṅgaprastha,[26] Nāgagiri,[27] Rocana,[28] the hill Pāṇḍara,[29] the hill Puṣpa,[30] Durjayanta,[31] Raivata,[32] and Arbuda,[33] Ṛṣyamūka,[34] and Gomanta,[35] Kūṭaśaila,[36] Kṛtasmara,[37] and Śrīparvata,[38] and Kora,[39] and other mountains in hundreds. By them the people, both Mlecchas and Āryas, are mingled together according to their divisions.

The chief rivers of which those people drink, hear them from me duly. Gaṅgā, Sarasvatī,[40] Sindhu,[41] and Candrabhāgā[42] also,[43] and Yamunā, and Śatadru,[44] Vitastā,[45] Irāvatī,[46] Kuhu,[47] Gomatī,[48] and Dhūtapāpā,[49] Bāhudā,[50] and Dṛśadvatī,[51] Vipāsā,[52] Devīkā,[53] Rankṣū,[54] Niścīrā,[55] and Gaṇḍakī,[56] and Kauśikā[57] are the rivers[58] which flow from the slopes of Himavat, O brāhman.

The Vedasmṛti,[59] Vedavatī,[60] Vritraghnī,[61] and Sindhu,[62] Veṇvā,[63] and Ānandinī[64] also, Sadānīrā,[65] and Mahī,[66] Pārā,[67] Carmanvatī,[68] Nūpī,[69] Vidiśā,[70] and Vetravatā,[71] Śiprā,[72] and Avarṇī[73] also are known[74] as those connected with the Pāripātra mountains.

The Śoṇa,[75] and Mahānada,[76] Narmadā,[77] Surathā,[78] Adrijā,[79] Mandākinī,[80] and Daśārṇā,[81] and Citrakūṭā[82] also, Citrotpalā,[83] and Tamasā,[84] Karamodā,[85] Pīśācikā,[86] and Pippaliśroṇī[87] also, Vipāśā,[88] the river Vañjulā,[89] Sumerujā,[90] Śuktimatī,[91] Śakulī,[92] Tridivā in regular order,[93] Vegavāhinī[94] also[95] flow from the slopes of the Vindhya[96] Mountains.

The Śiprā,[97] Payoṣṇī,[98] Nirbindhyā,[99] Tāpī,[100] and Niṣadhāvatī,[101] Veṇyā,[102] and Vaitaraṇī,[103] Sinībālī,[104] Kumudvatī,[105] Kavatoyā,[106] Mahāgaurī,[107] and Durgā,[108] and Antaḥśirā;[109] those rivers[110] flow from the slopes of the Ṛkṣa[111] Mountains, have holy waters and are bright.

The Godāvarī,[112] Bhīmarathā,[113] Kṛṣṇā,[114] and another[115] Veṇyā,[116] Tuṅgabhadrā,[117] Suprayogā,[118] Vāhyā,[119] and the river Kāverī;[120] these noble rivers[121] issue from the slopes of the Saliya[122] Mountains.

The Kṛtamālā,[123] Tāmraparṇī,[124] Puṣpajā,[125] Sūtpalāvatī;[126] these are rivers[127] which rise in the Malaya Mountains, and have cool water.

And the Pitṛsomā,[128] and Ṛṣikulyā,[129] Ikṣukā,[130] and Tridivā,[131] Lāṅgūlinī,[132] and Vaṃśakarā[133] are known to spring from the Mahendra[134] Mountains.

The Ṛṣikulyā,[135] and Kumārī,[136] Mandagā,[137] Mandavāhinī,[138] Kṛpā,[139] and Palāśinī[140] are known to spring in the Śuktimat[141] Mountains.

All the rivers[142] possess holy merit; all are rivers flowing into the ocean; all are mothers of the world;[143] they are well-known to cleanse from all sin.

And others, small streams, are mentioned in thousands, O brahman, those which flow only during the rainy season, and those which flow at all seasons.

The Matsyas,[144] and Aśvakūtas,[145] and Kulyas,[146] the Kuntalas,[147] the people of Kāśi,[148] and the Kośalas,[149] and the Atharvas, and Arka-liṅgas,[150] and the Malakas,[151] and Vṛkas[152]; these[153] are well known generally as the peoples who inhabit the Central Region.[154]

Now along the northern half of the Sahya mountains[155] that region, in which the river Godāvarī flows, is a delightful one compared even with the whole earth.; Govardhana is the charming city of the high-souled Bhārgava race.[156]

The North-western peoples are these —The Vāhlīkas[157] and the Vāṭadhānas,[158] and the Ābhīras,[159] the Kālatoyakas,[160] and the Aparāntas,[161] and the Śūdras,[162] the Pallavas,[163] Carmakhaṇḍikas,[164] Gāndhāras,[165] and Grabalas,[166] the Sindhus,[167] Sauvīras,[168] and Madrakas,[169] and the people who dwell along the Śatadru,[170] the Kaliṅgas,[171] the Pāradas,[172] the Hārabhāṣikas,[173] the Māṭharas,[174] and the Bahubhadras,[175] the Kaikeyas,[176] the Daśamālikas,[177] and the settlements[178] of Kṣattriyas, and the families of Vaiśyas and Śūdras.[179]

The Kāmbojas,[180] and the Daradas,[181] and the Varvaras,[182] the Harṣavardhanas,[183] and the Cīnas,[184] the Tukhāras[185] are the populous[186] races of men outside.[187]

And the Ātreyas,[188] the Bharadvājas,[189] and Puṣkalas,[190] the Kuśerukas,[191] the Lampākas,[192] the Śūlakāras,[193] the Culikas,[194] and the Jāguḍas,[195] and the Aupadhas,[196] and the Animadras,[197] and the races of Kirātas,[198] the Tāmasas,[199] and the Haṃsamārgas,[200] the Kāśmīras,[201] and the Tuṅganas,[202] the Śūlikas,[203] and the Kuhakas,[204] the Urṇas,[205] and Daryas;[206] these are the peoples of the Northern countries.

Hear from me the peoples who inhabit the Eastern countries. The Adhrārakas,[207] the Mudakaras,[208] the Antargiryas,[209] the Vahirgiras,[210] and the Pravaṅgas[211] also, the Rangeyas,[212] the Mānadas,[213] the Mānavartikas,[214] the Brahmottaras,[215] the Pravijayas,[216] the Bhārgavas,[217] the Jñeyamallakas,[218] and the Prāgjyotiṣas,[219] and the Madras,[220] and the Videhas,[221] and the Tāmraliptakas,[222] the Mallas,[223] the Magadhas,[224] the Gomantas,[225] are known as the peoples in the East.[226]

Now the other peoples who dwell in the Southern Region[227] are the Puṇḍras,[228] and Kevalas,[229] and Golāṅgulas[230] also, the Śailūṣas,[231] and Mūṣikas,[232] the Kusumas,[233] the Nāmavāsakas,[234] the Mahārāṣṭras,[235] Māhiṣakas[236] and Kaliṅgas[237] on all sides,[238] Ābhīras,[239] and Vaiśikyas,[240] Āḍhakyas,[241] and the Śavaras,[242] the Pulindas,[243] the Vindhyamauleyas,[244] the people of Vidarbha[245] and the Daṇḍakas,[246] the Paurikas,[247] and the Maulikas,[248] the Aśmakas,[249] Bhogavardhanas,[250] Naiṣikas,[251] Kuntalas,[252] Andhas,[253] Udbhidas,[254] Vanadārakas[255]; these[256] are the peoples of the countries of the Southern region.[257]

Hear from me the names of the Western peoples. The Sūryārakas,[258] the Kālibalas,[259] and the Durgas,[260] and the Anīkaṭas,[261] and the Pulindas,[262] and the Sumīnas,[263] the Rūpapas,[264] and the Svāpadas,[265] and the Kurumins,[266] and all the Kaṭhākṣaras,[267] and the others who are called Nāsikyāvas,[268] and the others who live on the north hank of the Narmadā,[269] the Bhīrukacchas,[270] and the Māheyas,[271] and the Sārasvatas[272] also, and the Kāśmīras,[273] and the Suraṣṭras,[274] and the Avantyas,[275] and the Arbudas[276] also. These are the Western peoples.

Hear the inhabitants of the Vindhya Mountains.[277] The Sarajas,[278] and Karūṣas,[279] and the Keralas,[280] and Utkalas,[281] the Uttamarṇas,[282] and the Daśārṇas,[283] the Bhojyas,[284] and the Kiṣindhakas,[285] the Tośalas,[286] and the Kośalas,[287] the Traipuras,[288] and the Vaidiśas,[289] the Tumburas,[290] and the Tumbulas,[291] the Paṭus,[292] and the Naiṣadhas,[293] Annajas,[294] and the Tuṣṭikāras,[295] the Vīrahotras,[296] and the Avantis.[297] All these peoples dwell on the slopes of the Vindhya Mountains.

Next I will tell thee also the names of the countries which rest against the Mountains.[298] The Nīhāras,[299] and the Haṃsamārgas,[300] the Kurus,[301] the Gurgaṇas,[302] the Khasas,[303] and the Kuntaprāvaraṇas,[304] the Urṇas,[305] the Dārvas,[306] the Sakṛtrakas,[307] the Trigartas,[308] and the Gālavas,[309] the Kirāṭas,[310] and the Tāmasas.[311]

And in this Bhārata is established the law of the four ages, the Kṛta, Tretā and the two others. Such is this country Bhārata, constituted with a four-fold conformation.[312] On its south and west and east is the great ocean; the Himavat range stretches along on its north, like the string of a bow.[313]

Then this country Bhārata is filled with every kind of seed, O brāhman. It has the supremacy of Brahmā, the lordship of the Ruler of the Immortals, the divinity of the gods, and the mortal nature of men.[314] It has various kinds of wild animals, cattle and aquatic animals;[315] and all creeping things likewise. And from it are produced[316] all immovable things, together with things good or bad. No other land of action exists among the worlds, O brāhman. Even among the gods, O saintly brāhman, this is ever in truth[317] the wish — “Oh, that we shall become men on the earth, when we fall from our divine condition! A man indeed does actions that the gods and demons cannot do!” Those who are involved in the fetters of such action, who are eager to proclaim their own actions,[318] and who are possessed of a small portion of happiness perform no action at all.

Footnotes and references:




This and the three following verses agree closely with the Kūrma Purāṇa canto xlvii, verses 22—25.


The dictionary gives the word as kiśeru-mat; the Kūrma Purāṇa as kaseruk-mat (canto xlvii, verse 22), in preference. Another form is said to be Kasetu.


The Kūrma Purāṇa gives the word as tāmra-parṇa in preference (canto xlvii, verse 22). It is Ceylon.


Or, gandharva, Kūrma Purāṇa, canto xlvii, verse 23.


This is understood to mean India, as the following verses show; see Wilson’s Viṣṇu Purāṇa, Edn. Fitz Edward Hall, Book II, chap. iii, note on the similar passage. But this Pnrāna states clearly enough (see verse 59 below) that India is not surrounded by the sea, but bounded by it only on the east, south, and west, and only partially so on the east and west, for verse 8 places the Kirātas and Yavanas there respectively.


The yojana is defined in canto xlix, verse 40, to be about 40,000 feet; this length therefore is 7,576 miles.


The Kirātas are the uncivilized tribes of the forests and mountains; here the word appears to denote all the races with the Burmese type of features along the eastern limits of India.


The Greeks originally, and afterwards the Mohammedans.


For the notes in this Canto I have consulted, Wilson’s Viṣṇu Purāṇa (Edn. Fitz Edward Hall), General Cunningham’s Ancient Geography of India (1871), his Archaeological Survey of India Reports, besides other works and maps.


“Mahendra is the ohain of hills that extends from Orissa and the northern Circars to Gondwana, part of which near Gañjam is still called Mahindra Malei or hills of Mahindra.” Wilson’s Viṣṇu Purāṇa, Book II, chap, iii, note. The rivers which flow from these hills are named in verses 28 and 29, but only a few of them have been identified. This range then appears to be the portion of the Eastern Ghats between the Godaveri and Mahānadi rivers, and the hills in the south of Berar. See, however, note on the Śukti-mat range on the next page.


This is the southern portion of the Western Ghats. Only four rivers are mentioned in verses 27 and 28 as rising in these hills, and none of them appear to have been identified; but as the River Kaveri is said in verses 26 and 27 to rise in the Sahya mountains, the Malaya mountains can be only the portion of the Western Ghats from the Nil-giris to Cape Comorin.


The Sahya mountains are the Northern portion of the Western Ghats, and, as appears from the rivers which rise in them (see verses 26 and 27), extend from the River Tapti down to the Nil-giris.


This range is not definitely identified, nor the rivers which are said in verses 29 and 30 to rise in it-

General Cunningham says the River(?) Suktimati “derived its name from the Suktimāl (sic) mountains, in which it had its source”; asserts that the river must be the same as the Mahānadī; and infers that the Śukti-mat mountains must “correspond with the high range of mountains to the south of Sehoa and Ranker, which gives rise to the Mahanadi, the Pairi and the Seonath rivers, and which forms the boundary between Chattisgarh and the feudatory state of Bastar.” (Arch. Survey Reports, vol. XVII, pp. 24 and 69; and map at end.) His premises seem to me unsafe; and his conclusion confounds the Śukti-mat range with the Mahendra range, and must he incorrect, for the latter range appears to be identified beyond doubt.

Mr. Beglar proposes to identify the R. Śukti-matī with the Sakri (which is a tributary of the Ganges, and flows northwards about 35 miles east of Gaya), to connect the river with the Śukti-mat range, and apparently to identify the range with the hills in the north of the Hazaribagh district. He proposes to strengthen this position by identifying the rivers Kiyul (another tributary of the Ganges, east of the Sakri) and Kaorbari (which I do not find, but which seems to be another small tributary) with the Ṛṣi-kulyā and Kumārī, which rise in the Śukti-mat mountains; see verses 29 and 30. (Arch. Survey Reports, vol. VIII, pp. 124 and 125). But the Śukti-mat range and Śukti-matī river do not seem necessarily connected; neither this nor the Vāyu Purāṇa makes the river rise in the Śukti-mat range, (see verse 23); Sakri does not appear the natural equivalent for Śukti-matī, (there is besides another river Sakri, a tributary of the Seonath,) nor Kiyul and Kaorhari of Ṛṣi-kulyā and Kumārī; Sakri corresponds better with Śakulī (see verse 23); and the hills in the north of the Hazaribagh district are not remarkable, and are rather the termination of the Vindhya range than a separate mountain system.

The only mountains, which have not been appropriated to the Sanskrit names, are the Aravalli mountains and the southern portion of the Eastern Ghats, so that this range might be one of these two; and if the former are rightly included in the Pāripātra Range, (see note ‡, next page) the Śukti-mat range might be the southern portion of the Eastern Ghats and the hills of Mysore. If, however, the Śukti-mat range must he placed in Berar, the Mahendra range will be restricted to the Eastern Ghats.


These are said to be the mountains of Gondwana, see Wilson’s Viṣṇu-Purāṇa, Book II, chap, iii, note, Judging from the rivers which are said in verses 21 to 25 to rise in the Vindhya and Rikṣa Ranges, it appears this range consists of the hills which form the water-shed between the Narbudda, Sone and Mahanadi on one side, and the Tapti and northern tributaries of the Godavari on the other side; that is, it comprises the Satpura Hills, and the hills extending through the middle of Berar and the south of Chutia Nagpur nearly into West Bengal.


For vindhaś read vindhyaś. This does not denote the whole of the modern Vindhya Range, but only the portion of it east of Bhopal, and also the water-shed hills which extend from it into Behar, as will appear from a comparison of the rivers which rise in it according to verses 21—23.


Called also Pāriyātra. This is the western portion of the modern Vindhya Range, west of Bhopal, as appears from the rivers which rise in it according to verses 19 and 20. Prof. Wilson says (Viṣṇu Purāṇa, Book II, chap, iii, note) “the name, indeed, is still given to a range of mountains in Guzerat (see Colonel Tod’s Map of Rajasthan),” and that may be considered an offshoot of the main range. If the Vāyu Purāṇa is right in reading Varṇāśā instead of Veṇvā in verse 19, this range would also probably include the Aravalli mountains in Rajputana. If this be so, the configuration of this range, a curve around the west and south of Malwa, would suggest a derivation for both the names, viz,, Pāripātra, from pari + pātra, “the mountains shaped like an enclosing receptacle”, or “the mountains which form a protection around;” or Pāriyātra, from pari + yā, “the mountains which curve around” The name may thus still survive in the Pathar range, which lies between the rivers Chambal and Banās. (Arch. Survey Reports, vol. VI, p. 1 and map; and vol. XIY, p. 151).


Mr. Beglar proposes to identify this hill with the Kawa Kol range, which is east of the R. Sakri (a tributary of the Ganges about 35 miles east of Gaya); but there does not appear to be anything about the range agreeing with the description in the text. (Arch. Survey Reports, vol. VIII, pp. 123 to 125, and map at end).


This as a mountain is not mentioned in the dictionary, and I do not find any such mountain. The Vāyu Purāṇa reads Vaihāra instead (xlv. 90), which is a synonym for a famous hill called Vaibhāra or Baibhāra, near Raja-griha, and about 28 miles north-east of Gaya. (Cunningham’s Anc. Geog. of India, vol. I, p. 452, map, and p. 463; Arch. Survey Reports, vol. I, p. 21 and plates III. and XIV.; vol. III., p. 140).


Mandara, the famous mountain, is situated about 35 miles south of Bhāgalpūr, in Behar (Arch. Survey Reports, vol. VIII., p. 130). There was another mountain of this name in the neighbourhood of Śveta-giri and Kailāsa; see MahāBhārata, Sabhā-P. li. 1858; Vana-P. cxxxix. 10820-30; and Anuśās-P. xix. 1434. This Mandara seems to have been the famous mountain originally.


This form is not in the dictionary, but it occurs in the Rāmāyaṇa Sundara-K. xcv. 25, and is the same as Dardura, which is the usual form, and Dardara, which is mentioned in the dictionary. It was a mountain or group of mountains, in the extreme South of India. It is mentioned in the MahāBhārata, Sabhā-P. li. 1891-3, where the context suggests that it was on the borders of the Cola and Pāṇḍya kingdoms; and it appears to be intended in the Anuśās-P. clxv. 7658, by the name Dururduda, which violates the metre. Malaya and Dardura are mentioned as the two highest mountains in the extreme South in the Raghu-V. (iv. 51.) In a paper on the Geography of Rama’s Exile, in the Journal, R. A. S., April 1894, p. 262, I have proposed to identify this mountain with the Nilgiris.


This as a mountain is not mentioned in the dictionary. The Vāyu Purāṇa reads Pātandhama instead (xlv. 91). I do not find either. But Mr. Beglar found a hill Bathan or Bathani in south Behar, and mentions a hill called Banthawa or Pandhawa in Buddhist records. These names might be easy corruptions of Pātandhama. (Arch. Survey Reports, vol VIII., p. 46).


This as a mountain is not mentioned in the dictionary, and I do not find it. Is it to be connected with Baijñath or Vaidya-nath, the famous place of pilgrimage, near the R. Karma-nāśā, south of Ghazipūr? There does not seem, however, to be any prominent hill there. (Arch. Survey Reports, vol. VIII, p. 137; and vol. XIX, p. 27). Or, should the reading be Vaidūrya? This seems preferable. The Vaidūrya Mts. are the Satpura Range; compare MahāBhārata, Vana-P. lxxxviii. 8343; Ixxxix. 8354-61; and cxxi; and Journal, R. A. S., April 1894, p. 245.


There were three mountains of this name. One is the famous Mountain in the north; it is called a son of Hima-vat by the apsaras Menā, and was a part of the great Hima-vat Range. It was near Kailāsa, Gandha-mādana and Śveta-giri (see M-Bh„ Sabhā-P. iii. 58-60; Vana-P. cxxxv. 10694-5; cxxxix. 10820; and clviii. 11540; and Bhīṣma-P. vi. 237): and from Vana-P. cxlv. 11054-64; and Harivaṃśa cxxxiii. 7598-7605, it appears to have been situated near the eastern sources of the Ganges; hence this Maināka probably denoted the group of hills in the north of the Almora district; but these passages are not quite consistent. The Rāmāyaṇa in Kiṣk. K. xliv. 35-37 places lake Mānasa on Mt. Krauñca, (which is called Maināka’s son, in Harivaṃśa, xviii. 941-2,) and Maināka beyond Krauñca; but those geographical cantos, xl to xliv, seem to be a late interpolation.

Another Maināka is the fabulous mountain situated in the sea, midway between the southern point of the Indian Peninsula and Ceylon; see Rāmāyaṇa, Sund. K. vii; and in this connexion see canto lii., verse 13.

And the third Maināka is in Western India, apparently in Khandesh as it seems to be connected with Cyavana; Bee MahāBhārata, Vana-P. lxxxix. 8364-5. In Gen. Cunningham’s Arch. Surv. Reports (vol. VIII, p. 124) the R. Sone is said to be called Maināka-prabha, from the mountain in which it rise, but Maināka is probably a mistake for Mekala.


This as a mountain is not mentioned in the dictionary; and I do not find it. The Vāyu Purāṇa reads Sasurasa, or Susarasa instead (xlv. 90); I do not find any such hill. Surasa is one of the mountains mentioned in canto lv. verse 9.


I do not find this. The Vāyu Purāṇa reads Gantu-prastha instead (xlv. 91) which seems a mistake.


I do not find this. In a list of mountains in Harivaṃśa, clxviii., 9499-9505 are mentioned Nāga and Nāga-rāṭ. Perhaps these mountains may be placed in the Nāga country, near Nāgpūr in the Central Provinces.


I do not find this.


The Vāyu Purāṇa reads Pāṇḍura (xlv. 90); neither is mentioned in the dictionary as a mountain. Should we read Pāṇḍava instead? There are two hills now which are called Pāṇḍua Hill or the Pāṇḍus’ Hill, one found by Mr. Carlleyle, north-west of Bairāt (or Vairāṭa) in Alwar (Arch. Survey Reports, vol. VI, pp. 95-101); and the other by Mr. Beglar, north of Hatta and near the R. Ken in Bandelkhand, where pilgrimages are still made (id. vol. VII., p. 56).


I do not find this. Puṣpaka is one of the mountains mentioned in canto lv. verse 13. A river called Puṣpa-jā is said to rise in the Malaya Mts. (see verse 27, below).


I do not find this. The Vāyu Purāṇa reads Uj-jayanta instead (xlv. 92), which Gen. Cunningham identifies with Girnar hill, which is situated on the east Bide of Juna-gaḍh in the peninsula of Gujarat (Anct. Geog., vol. I. p. 325). It was in Surāṣṭra (Mahā-Bhārata, Vana-P. lxxxviii. 8347-9). There are also the Ajanta Hills, north-east of Auraṅgabad (Arch. Survey Reports, vol. IX., p. 121) which seem to be the same as the Ajayanti Hill (Anc. Geog. of India, vol. I. p. 555).


Raivata, or Raivataka, was near Dvārakā or Kuśa-sthalī, which was near the extreme western promontory, and was the capital of the country Ānarta, in the peninsula of Gujarat; for in the MahāBhārata, a festival on this mountain is described, in which the citizens of Dvārakā went out there with their families, in thousands, on foot or in carriages (Ādi-P. ccxix. 7906-17); and it i s called the ornament of the gate of that city in the Harivaṃśa (cxiii. 6361-70; and cxiv. 6410-15.). The Girnar hill mentioned in the last preceding note is sometimes identified with this hill, but Mt. Girnar is about 110 miles from Dvārakā, and this distance is incompatible with either of the passages quoted. There does not appear to be any mountain close to that city, but the Baradā group of hills is not far from it, and they are the only hills that comply with the conditions (see Arch. Surv. of W. India, by J. Burgess, Kāṭhiāwāḍ, pp. 12,15, 84 and 154). Raivata is not necessarily a single mountain, for the Himavat, Vindhya and other ranges are often spoken of in the singular. I would therefore propose to identify Raivata with the Baradā Hills in Hālār, the western corner of the peninsula.


The modern Mount Abu, at the south end of the Aravalli range. Vasiṣṭha is said to have had his hermitage there (Mahā-Bhārata, Vana-P. lxxxii. 4097-8). The country around Arbuda was noted for its breed of horses (id. Sabhā-P. 1. 1851.)


Ṛṣyamūka is in the Dekhan. It was the scene of Rama’s meeting with Sugrīva and Hanūmān. I have proposed to identify it with the range of hills which stretches from Ahmadnagar to beyond Naldrug and Kalyani, dividing the Mañjira and Bhima rivers (Journal, R. A. S., April, 1894, p. 253).


There are two hills of this name. One is mentioned in the Harivaṃśa as situated in a gap or opening (vivara) of the Sahya Mts. (xcvi. 5331-40). It was three or four days’ journey by swift chariot from Karavīra-pura (xcvi. 5325-40; and c. 5650-52), i.e., probably 100 or 120 miles in a hilly country: and that city, which was the capital of the country Padmāvata, was on the Sahya Mts. on (and therefore near the source of) the R. Veṇvā, and presumably near Śūrpāraka (xcv. 5212, 5228-31; and xcvi. 5283-5322). This R. Veṇvā would therefore appear to be the river Purna (or one of the other small rivers south of Surat), flowing into the Gulf of Cambay, south of the R. Tapti; Karavīra-pura would have been near its source on the Western Ghats, and Gomanta would probably be the hills S. or S.-E. of Nasik. The other Gomanta is the hill of Gwalior. Gen. Cunningham says it was originally called Gopācala and Gopa-giri, Gopāhvaya, and later, Go-manta (Arch. Survey Reports, vol. II. pp. 372, 373). The Vāyu Purāṇa mentions Go-dhana instead (xlv. 91); but I have met with no hill of this name elsewhere; it suggests Go-vardhana, but Go-vardhana does not suit the metre.


I do not find this.


I do not find this. Is this to be connected with the Kāramār hill, in Gāndhāra? (Arch. Survey Reports, vol. II. pp. 92 and 106, and map at p. 87; vol. XIX. p. 126).


Śrī-parvata, or Śrī-śaila, is the name of a lofty rock which over-hangs the R. Kṛṣṇa in the Kurnool District; it is the site of a famous temple called Mallikārjuna, one of the twelve great Liṅga shrines (Arch. Surv. of S. India, by R. Sewell, Vol. I. 90; Arch. Surv. of W. India, by J. Burgess, p. 223). The Agni Purāṇa places Śrī-parvata on the R. Kāverī, and says it was dedicated to the goddess Śrī by Viṣṇu, because she had once performed some austerities (cxiii. 3,4). But Mr. Beglar, in a list of tīrthas where portions of Pārvatī’s body are fabled to have fallen when she was destroyed at Dakṣa’s sacrifice, mentions “Sri Parvat, near the Karatoya River.” This may perhaps be the river mentioned in verse 25, for I do not think there is any such hill near the other Karatoyā in North-East Bengal.


The Vāyu Purāṇa reads Kāru, or Ketu (xlv. 92) and a mountain Koṇva is mentioned in the Bhāgavata-P. (V. xix. 16); none are mentioned as mountains in the dictionary. There is a hill called Kolia in Mewat (Arch. Survey Reports, vol. XX. p. 133).


The modern Sursooty, between the Jumna and Sutlej. For a clear description, see Arch. Survey Reports, vol. II. 214, &c; and XIV. pp. 87-90 and Plate XXVI. There can be little doubt that in ancient times it was a very much larger river than it is now; see an interesting paper in the Journal, Beng. Asiat. Socy. 1886, Part II. p. 340; but in later times it perished, as it does now, in the sands of the desert, and Vinaśana was the name of the place where it disappeared (Mahā-Bhārata, Vana-P. lxxxii. 5052-5; and Śalya-P. xxxviii. 2119-20.). South and East of it was the Driṣadvatī, and between them lay the sacred region called Brahmāvarta (Manu II. 17, 18) and Tri-piṣṭapa (Mahā-Bhārata Vana-P. lxxxiii. 5074 and 7075) and also apparently Brahma-kṣetra (ibid., 5076). The name Saras-vatī however, was given to the seven rivers Suprabhā, Kāñcanākṣī, Viśālā, Manoramā, Ogha-vatī, Sureṇu and Vimalodakā (id., Śalya-P. xxxix. 2188—2216)


The Indus. As to its ancient course through Sindh, see Journal, Beng. Asiat. Socy., 1886, Part II. p. 323.


The R. Chenab, in the Puñjab. It was also called the Asīknī, the Greek Akesines.


Or, “and another Candra-bhāgā.” There were two rivers of this name (Mahā-Bhārata, Bhīṣma-P. ix. 322 and 327), but I have found no data to identify the second.


The R. Sutlej; the Greek Hyphasis. In ancient times this river probably did not join the Beas, as it does now, but pursued an independent course to the confines of Sindh. It flowed South-West from where it issues from the Himalayas, into the channel called the Naiwal and then along the dry bed called the Hakra or Ghaggar, at a distance of 30 to 50 miles south of, and more or less parallel to, its present course. See Journal, Beng. Asiat. Socy., 1886, Part II. p. 332.


The modem R. Jhelam, in the Pañjab; the Greek Hydaspes.


The modern R. Ravi, in the Pañjab; the Greek Hydraotes.


This does not appear to be known, though it is also mentioned by the Vāyu (xlv. 95) and Kūrma Purāṇas (xlvii. 27), both of which read Kuhū. As it is mentioned in conjunction with rivers in the Pañjab, is it to be identified with the Kubhā (Ṛg-V. x., 75. 6.), the Greek Kōphēn, the modern Kabul river? (Cunningham, Anc. Geog. of India, I. 37).


The modern Goomti, which joins the Ganges on the left bank below Benares. There was, however, another and older Go-matī (Ṛg-V. a. 75. 6), which is probably the modern R. Gomal, a western tributary of the Indus (Muir, Sansk. Texts, II. 357).


Gen. Sir A. Cunningham says this is a name of the Go-matī (Arch. Surv. Repts, I. 315). The text is Go-matī Dhūta-pāpā ca; and the Vāyu (xlv 95), Kūrma (xlvii. 27), Vaṛāha (Ixxxv.) and Viṣṇu Purāṇas all read the same. The two words are also linked together in the MahāBhārata (Bhīṣma -P. ix. 325), but not, I believe, in the Rāmāyaṇa, where the Go-matī is generally called “crowded with cattle.” Dhūta-pāpā then either means the Go-matī, and the translation would be, “and the sin-cleansing Go-matī;” or it denotes some tributary of that river.


There were two rivers of this name, this one (see M-Bh., Bhīṣma-P. ix. 337), and another in the Dekhan (ibid., 322; Anuśās.-P. clxv. 7653: and Rāmāyaṇa, Kiṣk. K. xli. 13). This river is mentioned in various passages (M-Bh., Vana-P. Ixxxiv. 8045-6; lxxxvii. 8323; xcv. 8513; Śānti-P. xxiii. 668, &c.; Anuśās-P. xix. 1408-11; and Harivaṃśa xii. 710), and from these it appears to have been a considerable river between the Go-matī and Ganges, in or near the territory of Ayodhyā, and having its source well up in the Himalayas. The only river which satisfies these conditions is the modern Ramgaṅgā, which joins the Ganges on the left, near Kanauj; and this river therefore is probably the Bāhudā.


Or, better, Dṛṣad-vatī; the famous river between the Saras-vatī and Jumna. It was the southern and eastern boundary of Brahmāvarta (Manu ii. 17). For a full description, see Cunningham, Arch. Surv. Repts., II. 214, &c.; and XIV. 87-90, and plate xxvi. See also note under Saras-vatī in verse 16.


Read Vi-pāśā, for Vi-pāsā. It is the modem R. Bias, in the Pañjab, the Greek Hyphasis. It is now a tributary of the Sutlej, but was probably altogether separate in olden times, for the Sutlej then had an independent course considerably to the south-east.


There are two Devikās, one in the Dekhan Rāmāy. Kiṣk. K. xli. 13), and this river (M-Bh., Bhīṣma-P. ix. 324; Anuśās-P. xxv. 1696-7; and Vana-P. ccxxi. 14229). From the second of these passages it appears that the northern Devikā was near Kashmir, and it may probably be identified with the modern river Deeg, a, tributary of the Ravi on its right bank. The Devikā, which is mentioned in Vana-P. lxxxii. 5044-9, seems to be a lake, and may be the same as Deviled Sundarikā hrada in Anuśās.-P. xxv. 1707-8.


I do not find any river of this name mentioned elsewhere. The Vāyu Purāṇa reads Ikṣu (xlv. 96), and this occurs in the M-Bh. (Bhīṣma-P. ix. 324); but I have found no data to identify it. Probably, however, we should read Vakṣu or Vankṣu, which is the Oxus.


This is not in the dictionary. The Vāyu Purāṇa gives the same name (xlv. 96), and the Varāha reads Nisvīrā (lxxxv) j while other readings are Niścitā, Nirvīrā, and Micitā. The Niścitā and two other rivers, the Nicitā and Nīvārā, are mentioned in the Bhīṣma-Parva list (ix. 326, 328), and the Nirvīrā in M-Bh., Vana-P. lxxxiv. 8116-9, but there appears to be nothing to identify them beyond that the Nirvīrā is connected with the Kauśikī (see note ** below) in the last passage and its context.


The R. Gandak, which flows into the Ganges on its north bank near Patna. It has shifted its course considerably; and formerly it flowed east of its present course, through the middle of the districts of Champaran, Muzaffarpur and Darbhanga.


Or, generally, Kauśikī; the modem R. Kosi, which flows into the Ganges on its north bank, through the district of Purnea. It has shifted its course very remarkably. Formerly it flowed east of its present position.


Or, as the text may be read, “and the Apagā flow,” &c. There is a river called the Apagā in Kuru-kṣetra (M-Bh. Vana-P. lxxxiii, 6038-40; Cunningham’s Arch. Surv. Repts., XIV. 88, and plate xxvi). The Kūrma Purāṇa reads Lohinī ceti instead (xlvii. 28); and the Vāyu (xlv. 96) and Varāha Purāṇas Ixxxv) mention the Lohita. The Lohita is the Brahma-putra, which till last century flowed round the south side of the Garo Hills, and then southward through the districts of Maimansingh and Dacca. Lohinī, though fem., no doubt means the same. The Vāyu Purāṇa reads Kauśikī ca tritīyā tu instead (xlv. 96), which may mean the “third Kauśikī,” for there seem to be three rivers of this name (see M-Bh., Vana-P. ccxxi. 14231); or may refer to a river Tritīyā which is mentioned in the M-Bh. (Sabhā-P. ix. 373); but I would suggest as preferable, Kauśikī Karatoyā tu, or Kauśikī ca tri-srotās tu. The Karatoyā is the modern Kuratee in the Bogra District in North Bengal; and Tri-srotas or Triḥ-srotasī (see M-Bh., Sabhā—P., ix. 375ṛ is, I believe, the ancient name of the modern Teesta, which is east of that; both now flow into the Brahmaputra; but the first formerly flowed into the delta, before the Ganges and Brahmaputra shifted their courses. (Cunningham, Arch. Surv. Repts., XV. 127 and 131, and plates i. and xxxiii. The Varāha Purāṇa adds the Cakṣuṣ-matī (lxxxv), an unknown name.


Or Veda-smṛtā mentioned in the M-Bh., the former in Anuśās -P. clxv. 7651, and the latter in Bhīṣma-P. ix. 324; and the Veda-smṛti is also mentioned in the Bhāgavata-P. (V. xix. 17); but I have found nothing to identify it.


Or Vedasinī, or Vetasinī. I have not met with these two names elsewhere; the Veda-vatī is mentioned in the M-Bh., (Bhīṣma-P. ix. 324; Anuśās.-P. clxv. 7651), but there appears to be nothing to identify it.


Or Vrata-ghnī, as the Kūrma Purāṇa reads (xlvii. 28). I have not met with either name elsewhere, and the river is not known apparently.


This is most probably the modem Kali Sindh, a tributary of the R. Chambal, though it may also be the Sindh, which is a tributary of the Jumna, between the Chambal and Betwa. The former is the more probable, because it is a large river and rises well up in the Pāripātra range, and suits the following incident better. This Sindhu was a river of much note, and on it was a great tīrtha, where Agastya met Lopā-mudrā, daughter of the King of Vidarbha, and she chose him for her husband (M-Bh., Vana-P. xcvi, xcvii; and cxxx. 10541). The name of this tīrtha may have been Sindhūttama, (id. Ixxxii, 4082-4095; and Anuśās-P. clxv. 7650); but if so, it must be distinguished from the great tīrtha Sindhūttama, which was on the Indus (Vana-P. Ixxxii. 5021).


This name is not in the dictionary, but it occurs several times, and is a variation of Veṇā. There is a river of this name in the Dekhan (see verse 24, note to Veṇyā), and one in Western India (see verse 26, note to Veṇyā), but I have not met with any river of this name in North India. Both the Vāyu (xlv. 97), and the Burma (xlvii. 29) Purāṇas read Varṇāśā instead; the Varāha reads Parṇā instead (lxxxv), and the Burma offers Parṇā and Parṇāśā in a note, (loc. cit.) The Varṇāśā or Parṇāśā is the modern Banās, and there are two rivers of this name; one a tributary of the Chambal, rising near Udaypur (Cunningham, Arch. Surv. Repts., VI, plate i.), and the other, a stream rising near Mt. Ābu and flowing into the Rann of Kachh; the former is the larger, and is probably the river meant in the text. Cunningham writes the name Parṇāśā (id. VI. 157) and Parṇa-nāśā (id. XV. 132), but the latter form seems doubtful. Devāvṛdha is said to have married one of these rivers (Hari-Vaṃśa. xxxviii. 1999, and 2004-10), probably the second.


Or, Sānandinī. The Vāyu (xlv. 97) and Kūrma (xlvii. 29) Purāṇas read Candanā instead, and the latter proposes Bandhanā and Sābandhanā in a note. The Varāha reads Candanābhā nāśadācārā (lxxxv.) for this and the next river, but not very intelligibly. None of these names appear to be identified.


The river “that is always filled with water.” The inclusion of this name among the rivers that rise in the Pāripātra Mountains is strange yet the Kūrma Purāṇa places it in the same group (xlvii. 29, note). I have met with no river Sadā-nīrā except that in North India. A river Sadā-nīrā-mayā is mentioned in Bhīṣma-P ix. 340, but there is nothing to identify it. The Vāyu Purāṇa reads Satīrd and Sadātīrā instead (xlv. 97), but I have not found these names elsewhere.

A few remarks may be offered about the Sadā-nīrā in North India. Sāyana says it is the Karatoyā, the modern Kurattee (see verse 18, note††), but it is stated in the Śata-patha Brāhmaṇa (I. iv. 1), that the Sadā-nīrā was the boundary between Kosala and Videha. It is therefore identified with the R. Gandak by Dr. Eggeling (loc. cit., note) and Muir (Sansk. Texts, II. 419-422). But the old stream of the Gandak flowed through the districts of Champaran, Muzaffarpur and Darbhanga, i. e., through the middle of the Videha country; and the Gaṇḍakī and Sadā-nīrā are mentioned as distinct rivers in M-Bh., Sabhā-P xix., 794. The Sadā-nīrā can hardly, therefore, be the same as the Gandak, and is more probably the modem Rapti, a tributary of the Sarayū, and the midway position of the Rapti eminently satisfies the position of a boundary.


The R. Mahī, which rises in Malwa and falls into the Bay of Cambay. The Vāyu Purāṇa has a variant, Mahatī (xlv. 97), and the Varāha reads Rohi (lxxxv.); both seem incorrect. The Mahitā mentioned in M-Bh., Bhīṣma-P. ix. 328, appears to be this river.


Or Parā, according to the Vāyu Purāṇa (xlv. 98). This is said to be the modern R. Parbatī, which rises in Bhopal and falls into the Chambal (Cunningham, Arch. Surv. Repts., II 308 and Rennell’s Atlas of 1781).


For Carman-vatī, read Carmaṇ-vatī. The R. Chambal, the largest tributary of the Jumna.


This is not in the dictionary. The Kūrma Purāṇa mentions the Surā and the Sūryā (xlvii. 29), but I have found no other mention of them, and they do not appear to be known.


This must, no doubt, be connected with the town Vidiśā, which was on the R. Vetra-vatī (Megha D. i. 25) the modern R. Betwa (see next note). Vidiśā appears to be the modern town Bhilsa. The R. Vidiśā therefore was probably the small tributary which joins the Betwa on its left bank at Bhilsa.


The modern R. Betwa, which rises near Bhopal and flows into the Jumna. There was another river of this name in Western India (Hari-Vaṃśa clxviii, 9514-6). The Varāha Purāṇa reads Veda-trayī wrongly (lxxxv).


This is the river on which Ujjayinī, the modern Ujjain, stands (Megha D. i. 31, 32). Another Śiprā is mentioned in verse 24.


This is not in the dictionary, and I have not found it elsewhere. The Vāyu Purāṇa reads Avantī instead (xlv. 98), which is preferable, and would be the river of the Avanti country (see notes to verses 52 and 55, below). The R. Avanti therefore is probably the river which rises near Mhow and flows into the Chambal. The Varāha Purāṇa reads Vapantī (lxxxv.) erroneously.


For smatāḥ, read smṛtāḥ.


The R. Sone which rises near the source of the Narbada and flows into the Ganges above Patna. It was also called Hiraṇya-bāhu and Hiraṇya-vāha; the Greek Erannoboas. For changes in its course, see Cunning. Arch. Surv. Repts., VIII. 4-24.


Or, Mahā-nadī. It flows through Orissa into the Bay of Bengal. The main stream is now considered to be the river which rises near Ranker, but that cannot be the source meant in the text, for it would belong to quite a different water-shed. The Mahānada here must designate the branch now called the Hasdu or Hestho, which rises near the source of the Sone. (Cunning., Arch. Surv. Repts., XVII. plate i.). The Varāha Purāṇa omits the Mahānadī altogether, and reads Jyotī-rathā instead (lxxxv). This river, which is also called Jyoti-rathyā (M-Bh., Vana-P. lxxxv. 8150) and Jyoti-rathā (Hari-Vaṃśa clxviii 9510-12) ia said to be a tributary of the Śoṇa in the former passage, and is placed in the Dekhan in the latter. It is, therefore, probably the modern Johila, the southern of the two sources of the R. Sone.


The modern Narbada or Nerbudda, which rises near the Sone and flows into the Gulf of Cambay.


This is not in the dictionary, and I have not met the name elsewhere; it is a synonym of Jyotī-rathā? (See last page, note §§). The Kūrma Purāṇa mentions the Su-rasā (xlvii. 30), and so also the Varāha (lxxxv); instead of this and the next river the Vāyu Purāṇa reads Su-mahā-drumā or, Surahādrumā (xlv. 99); but I have not met with any of these names elsewhere, except Surasā in the Bhāgavata-P. (V. xix. 17)


This is not in the dictionary, but is mentioned in M-Bh., Anuśās-P. clxv. 7648. I have found nothing to identify it.


The R Mandakin, which flows near Mt. Citrakut into the R. Paisuni, a tributary of the Jumna between the Ken and the Tons (Cunning., Arch. Surv. Repts., XXI. 11). Mr. Beglar’s proposal to identify it with the R. Reur, a southern tributary of the Sone (Ibid. XIII. 42—54) depends upon his identification of Mt. Citra-kūta with Ramgarh hill in Chhattisgarh, and is untenable (see Journal, R. A. S., April, 1894, page 240). The river Reur, or Rer, is also called Araud, and all these forms appear to point to Eraṇḍā as the original name.


The river of the country Daśārna, the modern R. Dasān, between the Betwa and the Ken.


This is not in the dictionary. It is no doubt to be connected with Mt. Citra-kūta, the modern Citrakut (see Journal, R. A. S. Aprū, 1894, page 239), and is probably the stream which flows round the south and east of the modern Mt. Citrakut, past Karwi into the Jumna.


This is not in the dictionary, but a Citropalā is mentioned in M-Bh Bhīṣma-P. ix. 341. Cunningham says Citrotpalā is the name of the modern main-stream of the Mahānadī below its junction with the Pairi (Arch. Surv. Repts., VII. 155, and XVII. 70); but that river as mentioned already (page 295, note §§) would belong to a different water-shed.


Or Tāmasī, as the Kūrma Purāṇa reads (xlvii. 30). It is the R. Tons which flows into the Ganges on the right bank below Allahabad.


This is not in the dictionary, and I have not found the name elsewhere. The Vāyu Purāria xlv. 100) and the Varāha (lxxxv.) read Karatoyā instead. Should we read Karma-nodā, as a synonym of Karma-nāśā? The river meant is no doubt the modern Karamnasa, which flows into the Ganges on the right bank just above the Sone.


I have not met with this river elsewhere. Piśāca was a name given to various races, chiefly barbarous hill tribes (Muir, Sansk. Texts, II. 59). In this place it would, no doubt, mean the tribes inhabiting Rewah and Chuta Nagpore, and the Pīśācikā is probably one of the southern tributaries of the Sone, such as the Her (see page 296, note §), or Kanhar.


Or Pipyalā śroṇī, as the Vāyu Purāṇa reads (xlv. 100); or Pippalā, as the Varāha reads (lxxxv). I have not found any data to identify it, but have seen the name assigned to the modern river Paisuni or Parsaroni, a tributary of the Jumna between the Ken and the Tons (Arch. Surv. Repts., XXL 11), and these words may well be corruptions of Pippali-śroṇī.


This appears to be the river mentioned in MahāBhārata, Anuśās-P. xxv. 1733 and perhaps 1710-11 also. It is probably the modern Bias which flows past Saugor and joins the R. Ken, a tributary on the right bank of the Jumna. (Cunning., Arch. Surv. Repts., XXI. 157, and plate xxxiv). The Ken or Kiyān, an important stream, does not appear to be mentioned; it is said to be a corruption of Karṇa-vatī (Ibid. 156; and II. 446), though.Lassen gives Kāyama as its ancient form (Ind. Alt., Map). Was Vi-pāśā the ancient name of this whole river? The Vi-pāśā in the Pañjab is mentioned in verse 18. The Varāha Purāṇa reads Viśālā (lxxxv), and the Kūrma mentions this name as a variant (xlvii. 31). There are several rivers of this name, and the river here meant is no doubt the Saras-vatī Viśālā at Gaya (Mahā-Bhārata, Śalya-P. xxxix. 2188-9, and 2205-6), probably the modern Lilajan which flows past Bodh Gaya.


I have not found this name elsewhere. The Varāha Purāṇa reads Vañjukā (lxxxv), the Kūrma Mañjulā (xlvii. 31), and the Vāyu Jambulā (xlv. 100). Of these names I have met only with Mañjulā elsewhere (Mahā-Bhārata, Bhīṣma-P. ix. 341), but with no data to identify it. The river meant is probably that on which Gaya stands; its eastern source is called the Mohana, its middle portion the Phalgu, and the eastern branch, into which it divides, the Jumna.


The Vāyu Purāṇa reads Siterajā (xlv. 101), and the Varāha Vīrajā (lxxxv). I have not met with any of these names elsewhere, but the MahāBhārata mentions three rivers Vīrā (Bhīṣma-P., ix. 329), Vīra-vatī (ibid., 332) and Vīrankarā (ibid., 333), which are all distinct. The Matsya Purāṇa reads two names instead, the Śunī and Lajjā (cxiii. 26), probably erroneous.


This river has been much written about but does not seem to be identified safely yet. See p. 285, note‡; and also Cunning., Arch. Surv. Repts., IX. 55. It is mentioned in the Harivaṃśa (clxviii. 9509-13) and is said there to he in the Dekhan; it seems to be meant by the name Mukti-matī in MahāBhārata, Bhīṣma-P. ix. 342; and perhaps it is referred to in Harivaṃśa xxxvii. 1980-7. These passages, however, may allude to two rivers of this name. It was the river on which stood Śukti-matī, the capital of Cedi; see note to Cedi in canto lviii, verse 16.


The Vāyu Purāṇa reads Mahruṇā or Makṣaṇā (xlv. 101), and the Varāha Pankinī (Ixxxv), but I have not met with any of these names elsewhere. The Śakulī, however, may probably be identified with the It. Sakri, which flows into the Ganges on the south, about half-way between Patna and Monghyr (Cunning., Arch. Surv. Repts., VIII. plate i; and XV. plate iv). There is also another Sakri which is a tributary of the It. Seonath, a tributary of the Mahānadī (id. XVII. plate i), but that rises rather in the Rikṣa Mts. The Bhīṣma-P. list mentions a river called Makarī (ix. 331); and the Matsya Purāṇa reads Mukuṭā instead (cxiii. 26).


The text Tridivā-kramu seems wrong, and I have adopted the reading of the Vāyu Purāṇa Tridivā kramāt, which is preferable. The word kramāt, if right, would indicate that the rivers are mentioned in regular order from west to east. The Tridivā is also mentioned in the MahāBhārata, (Bhīṣma-P. ix. 324; and Anuśās.-P. clxv. 7654), but no data are given to identify it. It may be noticed there is a river called the Krumu (Ṛg.-V., X. 75.6), which is probably the modern R. Kuram, a tributary of the Indus, south of the Kabul R. (Muir’s Sansk. Texts, II. 357); but it cannot be intended here. Another Tridivā is mentioned in verse 28.


This is not in the dictionary, but it occurs in MahāBhārata, Sabhā-P. ix. 371, The Vāyu (xlv. 100) and Varāha (lxxxv) and Kūrma Purāṇas (xlvii.31) read Bālu-vāhinī instead, and the last gives Ratna-vāhinī as a variant. I have riot met with either of these names elsewhere.


The Varāha Purāṇa adds another river Rātrī (lxxxv), but I have not met with it elsewhere.


The text reads Skandha, which is clearly wrong. The Vāyn (xlv. 101) Kūrma (xlvii. 31) and Varāha Purāṇas (lxxxv) read Rikṣa. There is certainly some confusion in this group of rivers, for the Mandākinī, Daśārṇā, and Tamasā rise in the Vindhya watershed, while the Śoṇa, Mahānada and Narmadā rise rather in the Rikṣa Mts; but the rivers mentioned in verse 24 rise in the Rikṣa Mts, so that the proper reading here should no doubt be Vindhya. The Agni Purāṇa says the Narmadā rises in the Vindhya Mts. (cxviii. 7); so that perhaps this river and also the Sone and the Hasdu branch of the Mahānadi, which all rise close together near Amara-kaṇṭaka, may have been considered to belong to the Vindhya watershed. There seems to have been some vagueness in this matter, for the Utkalas and (Dakṣiṇa) Kosalas are classed among the races who inhabited the Vindhya Mts. inverses 53 and 54.


One śiprā has been mentioned already in verse 20, and the Harivaṃśa says there is a Śiprā in the southern region (clxviii. 9509). The Vāyu Purāṇa reads Madrā instead (xlv. 102); and the Kūrma (xlvii. 32) and Varāha (lxxxv) Śighrodā. I have not found either of these names elsewhere, but a river Śīghrā is mentioned (Mahā-Bhārata, Bhīṣma-P. ix. 336) and another called Śivā (ibid., 332). The Matsya Purāṇa reads Kṣiprā (cxiii. 27).


The Payoṣṇī was in the southern region (Mahā-Bhārata, Vana-P. lxxxviii. 8329-35); it was the river of Vidarbha (ibid. cxx. 10289-90), and was separated from the Narmada by the Vaidūrya Mts. (ibid. cxxi. 10306-7). It was the modern river Purna (the tributary of the Tapti) together with the lower part of the Tapti into which the Puma continues. A careful consideration of King Nala’s remarks (ibid. lxi. 2317-9) with a map will show that the view described could only have been obtained from a position on the Satpura Mts. about longitude 75°; hence the Payoṣṇī visible from there could be only the lower part of the Tapti. Such was considered the main stream in old times, and it was a famous and sacred river. Gen. Cunningham’s proposal to identify the Payoṣṇī with the Pahoj, a tributary of the Jumna between the Sindh and Betwa, (Arch. Surv. Reports, VII. plate xxii.) is untenable as regards this famous river; but there were two rivers of this name (Mahā-Bhārata, Bhīṣma-P. ix. 324 and 327), and the Pahoj may be the other Payoṣṇī. The Varāha Purāṇa reads Payollī (lxxxv), which seems a mistake.


Or Nir-vindhyā; or according to the Vāyu Purāṇa, Nir-bandhyā (xlv. 102). One river Nirvindhyā is mentioned in the Megha-D. (I. 28 and 29, commentary) as lying between the R. Vetra-vatī (or Betwa) and Ujjayinī (Ujjain), and (if the Pārā is rightly identified with the modern Parbati, see note to verse 20) must be the modern Parwan which is west of the Parbati; but that river rises in the Vindhya Range according to the Megha-Dūta, and belongs to the Pāripātra watershed according to verses 19 and 20 above; on either view it is out of place here. There was, however, another large river of this name in the Dekhan, for it is mentioned along with the Payoṣṇī, the Tāpī and the Godāvarī and its tributaries in the Bhāgavata-Purāṇa (Y. xix. 17), and judged by its position there, it may be the Pen-ganga a tributary of the Warda.


See note to Payoṣṇī above. This is the upper part of the modern Tapti before it joins the Purna. This branch was hardly known in early times; it does not appear to be named in the Maha-Bhārata or Ramayaṇa, nor is it mentioned in the copious list in the Bhīṣma-P. (ix). The reason was, no doubt, it was hidden amid hills and forests.


Or Niṣadhā, as the Vāyu Purāṇa reads (xlv. 102). I have not met with this name elsewhere, but it naturally suggests a connexion with Niṣadha, the realm of Nala. As regards Niṣadha, see the note to verse 54 below. This river then may be one of the small tributaries of the Narmadā or Tapti, which rise in the middle part of the Satpura Range. The Kūrma Purāṇa reads Mahānadī instead (xlvii. 32), which may mean the Mahānadī in Chhattisgarh and Orissa, but is unsatisfactory, as it has mentioned that river before (ibid., 30). The Matsya Purāṇa reads Ṛṣabhā instead (cxiii. 27), which I have not met elsewhere.


This form is not in the dictionary. The Vāyu Purāṇa reads Veṇvā (xlv. 102); the Kūrma reads Vinnā and gives Veṇyā and Cintā as variants (xlvii. 32). Cintā is no doubt an error. The other names are merely different forms of the same word. The river is called Veṇvā in the Harivaṃśa (clxviii. 9509-10) and also in the MahāBhārata (Sabhā-P. xxx. 1118), but in the latter poem it is generally called Veṇā, and this seems the proper term (Bhīṣma-P. ix. 335; Anuśās-P. clxv. 7648; Vana-P. lxxxviii. 8328, clxxxix. 12909, and Ixxxv. 8176-7; whether the same river is also meant in line 8175 is not clear). Prom the passage last cited it appears the Veṇā is the river which joins the Godavari and Varadā (the modern Warda), that is, the modern Wain-ganga and its continuation the Pranhita. The Varāha Purāṇa reads Veṣṇāpāśā (lxxxv) which seems a mistake. This river appears to be also called Su-veṇā (Mahā-Bhārata, Vana-P. clxxxix. 12909) in contra-distinction to the Kṛṣṇa-veṇā (ibid.; and also id. Vana-P. lxxxv. 8180-1; Bhīṣma-P. ix. 335, and Anuśās.-P. clxv. 7648; and Harivaṃśa clxviii. 9509-11) which appears from the second passage to be a tributary of the Veṇā, and which I have proposed to identify with the western tributary rising near Deoghar and Seoni (Journal, R. A S., 1894, p. 244). Another river of this name is mentioned in verse 26, and a Veṇvā in verse 19.


This is no doubt the modern Bytarai, which flows through the north of Orissa; and if it is rightly classed here, the Rikṣa Range must include the hills which stretch along the south of Chuta Nagpore.


The Vāyu Purāṇa reads Śitibāhu (xlv. 102), the Kūrma Balākā (xlvii. 32) and the Varāha Vedipālā (lxxxv). None of these rivers are mentioned in the dictionary, but the name Śinibāhu is given. I have not found any of these names elsewhere, except Balākā in MahāBhārata, Anuśās.-P. xxv. 1706-7, which may be a river, but appears from the context to be in Northern India. Perhaps the reading should be Śilāvatī or Śilāvatī, which seems to have been the ancient name of the modem river Selye; this after uniting with the Rūpnarain is the river on which Tamluk, the ancient Tāmra-liptaka (see verse 44 below), is situated; and which may well find mention here. Perhaps the name Balākā may be connected with the modern river Barākar, a tributary of the Damudā; these two combined form the largest river in Western Bengal, and flow close to Tamluk. Tamluk was a famous port, and it would be strange if the rivers near it were overlooked. The MahāBhārata mentions a river Śata-balā (Bhīṣma-P. ix. 328). The Matsya Purāṇa reads Viśva-mālā instead (cxiii. 37). I have not met either name elsewhere.


I have not met with this name elsewhere. It may be the Subarna-rekha or one of the small rivers in the north of Orissa; or may we conjecture Damud-vatī, and identify it with the It. Damudā in West Bengal? See the last note.


One river of this name in North Bengal has been mentioned in verse 18 note††; and there was another of the same name in the north of India (M.-Bh, Anuśās.-P. xxv. 1699); neither can be meant here. I have not found any Karatoyā elsewhere, which rises in the Ṛkṣa range. The Vāyu Purāṇa reads Toyā instead (xlv. 103), and so also the Varāha (Ixxxv); but I have not found this name elsewhere. Perhaps the reading should be Karabhāca. Karabhā or Kapiśā is the name of a river on the confines of Utkala and Kaliṅga (Raghu-V. iv. 38, commentary), but no details are given to identify it. The name Kapiśā suggests identification with the modem Cossye or Kansai (the chief river in the Midnapur district) which is said to be modified from Kaṃsavatī, but may well be a corruption of Kapiśā-vatī.


This is also mentioned in MahāBhārata, Bhīṣma-P. ix. 341. It is no doubt a synonym of Brahmāṇi and Brāhmaṇī, all being names of Durgā; it would then be the modern R,. Brahmṇni in Orissa.


There are two rivers of this name mentioned in the Bhīṣma-P. list (ix. 337 and 341) in the MahāBhārata, and the second is that intended here, as it is placed with the Mahā-gaurī; but I have not met with the name elsewhere. It may be a synonym of the small river Brāhmaṇi which flows through the Moorshedabad district into the right bank of the Bhāgīrathī branch of the Ganges.


This is not in the dictionary. The Vāyu (xlv. 103) and Kūrma (xlvii. 33) Purāṇas read Antaḥ-śilā. The Varāha reads Antyāgirā (lxxxv), which is no doubt an intended synonym. I have not met with any of these names elsewhere, but Antra-śilā is mentioned (Mahā-Bhārata, Bhīṣma-P. ix. 337). Antaḥ-śilā seems to be the correct form; and if the name is descriptive, the river is probably one of the northern tributaries of the Mahānadi; all of which are encompassed with hills. See however a people called Antar-giryas in verse 24 below.


The Varāha Parāṇa mentions also Maṇijālā Śubhā (lxxxv); I have not found the former name elsewhere; but the Śubhā is mentioned in the Harivaṃśa (clxviii. 9509-10), and a river Maniṅgā is mentioned in the Bhīṣma-P. list (ix. 342). There are no data to identify them, except that the passage in the Harivaṃśa places the Śubhā in the Dekhan.


The text reads Bindhya or Vindhya, and yet makes the next group of rivers also rise in the same range. The Vāyu (xlv. 103) and Kūrma (xlvii 33) and Varāha (lxxxv) Purāṇas read the same; but the proper reading must be Ṛkṣa as the Viṣṇu Purāṇa says (Bk. II. Chap, iii), for the Tāpī, Veṇyā and Vaitaraṇī certainly do not rise in the former mountains but in the latter. The Agni Purāṇa wrongly groups the Tāpī and Payoṣṇī with the Godavari and other rivers as rising in the Sahya Mts.


The modern Godaverī. This river was famous from the earliest times. Jana-sthāna, the scene of Rāma’s first conflict with the Rākṣasas was the country on both its banks between its tributaries the Mañjira and Pranhita (see Journal, R A. S, 1894, p. 247).


Or Bhīma-rathī as the Vāyu (xlv. 104) and Varāha (lxxxv) Purāṇas read: Bhīma-rakṣī which the Kūrma gives (xlvii. 34) seems incorrect. The former is the name as given in the MahāBhārata (Vana-P. lxxxviii. 8328; Bhīṣma-P. ix. 327; and Anuśās.-P. clxv. 7653). This is the modern Bhima, the tributary of the Kṛṣṇā, rising near Poona. The Varāha Purāṇa adds immediately Marathī (lxxxv); is it a mistaken repetition of the last three syllables of the preceding river? I have found no such river.


The modern Kistna. This river received very little notice in ancient times, and was almost unknown compared with the Godāvarī and Kāverī. Besides its inclusion in the Bhīṣma-P. list (ix. 340), it is doubtful if it is so mentioned in the MahāBhārata, or Rāmāyaṇa. It is omitted from the lengthy account of Sahadeva’s conquests in the south (Sabhā-P. xxx), and the detailed pilgrimage itineraries (Vana-P., Tirtha-yātrā-P.) and other geographical discourses. It does not occur in the story of Raghu’s conquests even in the late poem, the Raghu-Vaṃśa. The reason seems to be that the country through which it flows was nearly all forest in ancient times.


For tathāparā of the text the Vāyu Purāṇa reads ca vañjulā (xlv. 104) and the Kūrma ca vaśyatā or cavatsarī (xlvii. 34). I have not found these names elsewhere. Perhaps we should read some name like Mañjirā, as the large southern tributary of the Godāvarī is now called, though its earliest name was apparently Mandākinī (Rāmāy., Yuddha K. cx. 38, 39; Journal R. A. S., 1894, p. 250).


This form is not in the dictionary. The Varāha Purāṇa reads Veṇā lxxxv); the Kūrma Vend or Varṇā (xlvii. 34); and the Vāyu. Vaiṇī (xlv. 104)—all mere variations, the proper name no doubt being Veṇā. This is the third river of this name mentioned here, see verses 19 and 24. It is probably the same as the Vīṇā in the Bhīṣma-P. list (ix. 328), and the Veṇṇā in the Bhāgavata Purāṇa (Y. xix. 17). Is it to be identified with the R. Penner which is between the Kistna and Kaveri; though the Sanskrit name of the Penner is said to be Pinākā (Arch. Surv. of S. India, by R. Sewell, I. 123 and 129)?


The modern Tumbhudra, the large southern tributary of the Kistna, consisting of the combined streams of the Tuṅga and Bhadra.


This is not in the dictionary, but it is also mentioned in the Bhīṣma-P. list (ix. 328) and in the Vana-P. (ccxxi. 14232), and was a large and known river. Though not apparently identified, it is probably one of the large western tributaries of the Kistna.


This is not in the dictionary, but the Varāha Purāṇa agrees (lxxxv) and the Matsya (cxiii. 29). I have not found the name elsewhere and it does not appear to be identified. The Agni P. reads Vāradā (cix. 22), the large southern tributary of the Kistna called Varada or Vedavati.


The modern Cavery or Coleroon in south India. It was better known than the Kistna in ancient times. It is mentioned in the MahāBhārata, (Vana-P. lxxxv. 8164-5; clxxxix. 12910; and Bhīṣma-P. ix. 328) and Rāmāy. (Kiṣk. K. xli. 21 and 25). King Jahnu is said in the Harivaṃśa to have married this river, and made the Ganges his daughter (xxvii. 1416-22; and xxxii. 1757-61).


The Matsya (cxiii. 29) and Varāha (lxxxv) Purāṇas add the Vañjulā; as to which see verse 26 note ¶.


The text reads Bindhya or Vindhya here, after having read it already in verse 25; and offers Sahya a as variant in a note. The latter is manifestly the proper reading, and agrees with the Kūrma (xlvii. 34) and Vāyu (xlv. 104) Purāṇas.


This is not in the dictionary. The Agni Purāṇa agrees with it (cxviii. 8); the Kūrma reads Ṛtu-mālā (xlvii. 35), the Varāha Śata-mālā (lxxxv), and the Bhāgavata Kata-mālā (Y. xix. 17). Is it to be identified with the Vedamali which flows out north of Cochin? The people of Kaccha or Cochin are mentioned in canto lviii. verse 28.


This is mentioned as a place of pilgrimage in the MahāBhārata (Vana-P. lxxxviii. 8340), and the Raghu-V. says (iv. 49 and 50), that the vanquished Pāṇḍya kings gave Raghu the choicest pearls from the sea at the mouth of the R. Tāmra-parṇī, where (the commentator adds, it is well-known) pearls were produced. This river then was in the Pāṇḍya country and flowed into the G. of Manaar. It is the modem Chittar, the river of Tinnevelly (Arch. Surv. of S. India, by R. Sewell, I. 303).


The Vāyu Purāṇa reads Puṣpa-jāti (xlv. 105), the Kūrma Puṣpa-vatī (xlvii. 35), and the Varāha Puṣpā-vatī (lxxxv). I do not find any of these names elsewhere. A tīrtha Puṣpa-vatī is mentioned (Mahā-Bhārata, Vana-P. lxxxv. 8154-5), but it was situated between Dakṣiṇa Kosala (Chhattisgarh) andCampā (Bhagalpur). A river Puṣpa-veṇī is mentioned (id., Bhīṣma-P. ix. 342), which is joined with a river ĪJtpalā-vatī and therefore is probably the same as the river in the text. A Puṣpa-vāhinī is mentioned as situated in the south in the Harivaṃśa (clxviii. 9510-2).


This is the same as the Utpalā-vatī mentioned in the last note (Mahā-Bhārata Bhīṣma-P. ix. 342) and the Utpalā (Hari-Vaṃśa, clxviii. 9510-2); and the Vāyu (xlv. 105) and Kūrma (xlvii. 35), Matsya (cxiii. 30) and Varāha (lxxxv) Purāṇas read Utpalā-vatī.


There are only six noteworthy rivers rising in the Malaya Mts., viz., the Vaigai, Vaippar and Chittar on the east, the Amaravati (a tributary of the Kaveri) on the north, and the Ponani and Peri or Veda-mali on the west. The Chittar is the Tāmra-parṇī, hence the three others named in the text must be found among the five remaining modem rivers.


Not in the dictionary. The Vāyu Purāṇa reads Tri-sāmā (xlv. 106), and the Agni P. also (cxviii. 8); while the Varāha reads Tri-yāmā (lxxxv). I have met only with the Tri-sāmā elsewhere (Bhāgavata Purāṇa, V. xix. 17). It is probably one of the small rivers on the Eastern coast, for the interior behind these mountains was not well known. The Matsya Purāṇa reads. Tri-bhāgā (cxiii. 31) which I have not met elsewhere.


This is the river on which Gañjam stands, and it bears the same name still. It is mentioned in the Bhīṣma-P. list (ix. 343). The Vāyu Purāṇa reads Ritu-kulyā (xlv. 106), by mistake. Another Ṛṣi-kulyā is mentioned in verse 29.


This is not in the dictionary. The Vāyu (xlv. 106) and Varāha Purāṇas (lxxxv) read Ikṣulā; and Ikṣudā which the Matsya Purāṇa reads (cxiii. 31) is a variant. I have not found any of these names elsewhere, but the R. Ikṣu is mentioned (Mahā-Bhārata, Bhīṣma-P. ix. 324). It is probably one of the email streams on the Eastern coast.


This is the second Tridivā, see verse 23; but I have not found two rivers of this name mentioned anywhere else. Instead of Tridivā ca yā as in the text, the Matsya Purāṇa reads Tridivācalā (cxiii. 31).


This is the modem Languliya, on which Chicacole stands, between Vizianagram and Caliṅgapatam. The Varāha Purina reads Mūlinī or Lāmū-linī (lxxxv) and the Matsya Mūlī (cxiii. 31); I have not found these names elsewhere and they seem incorrect. The Lāṅgalī mentioned in the MahāBhārata (Sabhā-P. ix 374), is probably this river.


The Varāha Purāṇa reads Vaṃśa-varā (lxxxv), and the Vāyu Vathśa-dharā (xlv. 106); the latter is the correct name. It is the modern Bansdharā, the river on which Caliṅgapatam stands.


The Kūrma Purāṇa omits this group of rivers altogether, and puts three of them Tri-sāmā, Ṛṣīkā and Vaṃśa-dhāriṇī into the next group (xlvii. 36). The Matsya Purāṇa mentions three more rivers, the Tāmra-parṇī,Śaravā and Vimalā (cxiii. 31), but all these seem doubtful. A Tāmrā-parṇī has been mentioned in verse 28. A Sarāvati is named in the Bhīṣma-P. list (ix. 327), and a Vimalā or Vimalodā in various passages (e. g. Śalya-P. xxxix. 2214-5; Harivaṃśa, clxviii. 9517-8); but they are in Northern and Western India.

In note †† on page 284, the Mahendra mountains are said to be “the portion of the Eastern Ghats between the Godavari and Mahanadi rivers and the hills in the south of Berar,” but this proposition must be modified on a full consideration of all the foregoing identifications. Gondwana as used by Wilson was applied to a very wide tract in Central India. The Mahendra Mts. cannot extend as far west as Berar nor beyond the Wain-ganga; and must be limited to the hills between the Mahanadi, Godavari and Wain-ganga, and may perhaps comprise only the portion of the Eastern Ghats north of the Godavari. It is in this last tract only that the name has survived. See Raghu Vaṃśa, iv. 43.


This is the second Ṛṣi-kulyā, see verse 28. The Vāyu Purāṇa reads Ṛṣīkā (xlv. 107), and the Varāha Ṛṣikā (lxxxv), and the Matsya Kāśikā (cxiii. 32). I have not met with these names elsewhere.


The Vāyu Purāṇa reads Su-kumārī (xlv. 107), and the Varāha Lūsatī I have not found these names elsewhere, but the Kumārī is mentioned in the Bhīṣma-P list (ix. 313).


This is mentioned in the Bhīṣma-P. list (ix. 340). The Varāha Purāṇa reads Wanda-gāminī (lxxxv); and for this and the next river the Kūrma reads Gandha-mādana-gāminī (xlvii. 36), which is probably erroneous.


This is mentioned in the Bhīṣma-P. list (ix. 340), but hardly in the same connexion.


The Vāyu Purāṇa reads Kūpā (xlv. 107), and the Kūrma Kṣiprā or Rūpd (xlvii. 36). I do not find any of these names elsewhere. A river Krityā is mentioned in the Bhīṣma-P. list (ix. 326), but that appears from its context to be in north India.


This is mentioned in the Bhīṣma-P. list (ix. 330), but in so different a connexion that the references appear to be to two separate rivers. The Matsya Purāṇa reads Pāśinī (cxiii. 32), which however I have not met elsewhere.


These Mts. are but very rarely mentioned, and in page 285 note ‡ I have noticed what has been written about them. They were in the Eastern region, for Bhīma in his conquests in that quarter marched from Hima-vat towards Bhallāta and conquered the Śukti-mat Mountain (M-Bh., Sabhā-P. xxix. 1079). Though Bhallāta does not appear to have been identified, the only noteworthy hills in the east which have not been assigned to the other great ranges are the Garo, Khāsi and Tipperah Hills which bound Bengal in that direction. Can these be the Śukti-mat Mts.? There seems to be no improbability in this, for the river Lohita or Brahma-putra and the country Kāmarūpa, which is in the Assam Valley, were known. If this identification is satisfactory, the R. Kumārī may be the modem Someśwarī which flows southward between the Garo and Khasi Hills (both being names of Durgā); and the Kṛpā may perhaps be the Kapili which flows into the Brahma-putra a little above Gauhati, the ancient Kāmarūpa; the other streams are not recognizable.


Saras-vatyaḥ. Or should this mean only the rivers called Saras-vatī? There were seven rivers specially distinguished by this name (M.-Bh, Śalya-P. xxxix. 2188-9), namely, 1. the Suprabhā among the Puṣkaras (ibid. 2198-2200), that is, near Ajmir; 2. the Kāñcanākṣī in Naimiṣa forest (ibid. 2201-4), which was on the Go-matī; 3. the Viśālā at Gaya (ibid. 2205-6); 4. the Mano-ramā, the swift stream flowing from Himavat in the north part of Kosala (ibid. 2207-10); 5. the Ogha-vatī, which seems to be in Kuru-kṣetra (ibid. 2212-3); 6. the Su-reṇu, which seems to be in Kuru-kṣetra or near Gaṅgā-dvāra (ibid. 2211-4); and 7. the Vimalodā or Vimalodakā at Haimanta-giri (ibid 2214-5).


Viśvasya mātaraḥ; compare M-Bh., Bhīṣma-P. ix. 844.


The people and their country both went by the name Matsya. This country was part of the region called Brahmarṣi (Manu ii. 19). It was south or south-west of Indra-prastha, the modern Delhi (M-Bh., Sabhā P., xxx. 1105-6; the mention in ibid. xxix. 1083 may be a mistake; but the Matsyas are named twice in the Bhīṣma-P. list, ix. 317 and 348, unless one name be a mistake for Vatsa); and it was west of Śūrasena, which was the country round Mathura, the modern Muttra (Virata-P., v. 141-5; see note in canto lviii. verse 7): hence Matsya comprised the modem Alwar State and the land around that. It appears to have extended up to Kuru-kṣetra, because no other country which could intervene is mentioned in Manu ii. 19. Its capital was Upaplavya or Upaplava (Śalya-P., xxxvi. 1973-6) which was l ½ or 2 days’ journey by chariot from Hāstinapura (Udyoga-P., lxxxiii. 3010-17; lxxxv. 3040; and lxxxviii. 3101). Cunningham says Matsya was the country west of Agra and north of the R. Chambal, i. e., the whole of Alwar with portions of Jaypur and Bharatpur; and its capital was Vairāta, the modern Bairat (Arch. Surv. Repts., II. 242; and XX. 2, and plate i).

The Vāyu Purāṇa reads Vatsas instead (xlv. 110). Vatsa or Vātsya was in the region east of Delhi (Sabhā-P., xxix. 1084), and king Vatsa who is said to have given his name to the country was grandson of Divodāsa, king of Benares (Hari-Vaṃśa, xxix. 1587, 1597; and xxxii. 1753). Kauśāmbī was the capital, and it has been identified by Cunningham with the modern Kosam which is on the north bank of the Jumna about 31 miles above Allahabad. Hence the country was also called Kauśāmba (Arch. Surv. Repts., I. 301-310). Yatsa or Kauśāmba therefore comprised the lower part of the Ganges and Jumna Doab and also probably the tract south of that, on the other side of the Jumna.


The Vāyu Purāṇa reads Kisaṣṇas, Kisaṣṭas or Kisadyas instead (xlv. 110); but none of these names are in the dictionary. The Matsya reads Kirātas (cxiii. 35) but they are out of place here. The text reads Matsyāśvakūṭāḥ kulyāśca, but I would suggest instead Matsyāś ca Kanyākubjāś ca, thus reading Kanyā-kubjas or Kānya-kubjas instead of Aśvakūtas and Kulyas. Kanyā-kubja or Kānya-kubja is the modern Kanauj, on the Ganges about 50 miles above Cawnpore; it was a famous city all through Indian history. People called Sukuṭyas are mentioned (Bhīṣma-P. ix. 347), Aśvakas (ibid. 351), and Aśvātakas (ibid. l i. 2105).


This is not in the dictionary as a people; the word occurs in Vana-P., (cxxv. 10408), but does not appear to mean a people there. See the last note.


This country is said by Muir to be one of the Piśāca countries (Sansk. Texts, II. 59), but there were three people of this name, one in the Dekhan (Bhīṣma-P., ix. 367), who are mentioned in verse 48 below; and two others elsewhere (ibid. 347 and 359). Those mentioned in verse 347 are the people meant here, for they are grouped with the people of Kāśi and Kosala; and they probably occupied the country near Chunār (south of Benares), which Cunningham calls Kuntila (Aroh. Surv. Repts., XI. 123). The third people were probably in the West.


Benares, the ancient Vārāṇasī. It was the capital of an ancient and famous kingdom. According to the Rāmāyaṇa Kāśi was a kingdom (Ādi-K., xii. 20) while Prayāga and the country all around it was still forest (Journal, R. A. S., 1894, pp. 237-239). Its sacred character dates from comparatively late times, for it was one of the exploits for which Kṛṣṇa was extolled that he burnt it for a succession of years and devastated it (Udyoga-P., xlvii. 1883; and Harivaṃśa, clxi. 9142-3). For some vicissitudes in its early history, see Harivaṃśa, xxix. and xxxii.


Kośala, Kosala or Uttara Kosala, with its capital Ayodhyā, is the modem Oudh. Gen. Cunningham says it meant more particularly the country north and east of the R. Rapti (Arch. Surv. Repts., I. 327; and XVII. 68); but it seems rather to have denoted the country stretching from the Rapti on the east (see page 294 note ‡) to the confines of the Kuru and Pāñcāla kingdoms on the west. Northward it was bounded by the tribes that inhabited the slopes of the Himalayas, and southward by the kingdom of Benares. It was distinguished from another Kosala, whioh was called Dakṣiṇa or Mahā Kosala and which is mentioned in verse 54.


These two names are not in the dictionary, and I have not found them elsewhere; they seem to be mistakes. The Vāyu Purāṇa reads instead of them atha pārśve tilaṅgāś ca (xlv. Ill); but this is doubtful and unsatisfactory, for the Tilaṅgas are mentioned as a southern people in canto lviii. verse 28. The Matsya reads and Āvantas and Kaliṅgas (cxiii. 36), but these are hardly satisfactory; the former are mentioned in verses 52 and 55, and the latter in verses 37 and 46 below. Perhaps Arka-liṅgas may be meant as a synonym of Sūrya-vaṃśas, the Solar Race, yet this again is hardly satisfactory, for that race reigned in Kosala, whioh has just been mentioned separately. There is a low group of Brāhmans in Behar called Atharvas (Risley’s Tribes and Castes of Bengal, I. 26).


This is not in the dictionary and seems erroneous. The Vāyu Purāṇa reads Magadhas instead (xlv. Ill), the people of Magadha or South Behar but this is unsatisfactory for the Magadhas are mentioned in verse 44. The Matsya reads Mūhas (cxiii. 36), which I have not met elsewhere. The reading should be Malajas probably. They are mentioned in the M-Dh. (Bhīṣma-P. ix. 357), and Rāmāyaṇa (Ādi-K. xxvii. 16-23), and from the course described in the latter poem as taken by Viśvā-mitra and Kama it appears they were neighbours of the Karūṣas, (see note to verse 53), and occupied the district of Shahābād, west of the Sone, for Viśvā-mitra and Kāma crossing from the Sarayū to the south of the Ganges entered that district, which had been inhabited by the Malajas (ibid., 8-16.)


The Vṛkas are named in Bhīṣma-P., li. 2106, and a king Vṛka is alluded to in the Hari-Vaṇśa (xiii. 760-61); but there is nothing to identify them. The Matsya Purāṇa reads Andhakas (cxiii. 36). The Andhakas were a subordinate family of the Vādava race, and are often mentioned in the MahāBhārata (e.g. Udyoga-P. lxxxv. 304), and Harivaṃśa (xxxv. 1907-8; and xxxix. 2041; and xciv. 5190-5204), but they dwelt in Surāṣṭra in the West and appear to be out of place here. The most probable reading seems to be the Vrajas, the people of Vraja (or Vṛji, as it was also called)* the modern Braj, the country north-west of Mathura or Muttra.


This is a short list. Besides these the Vāyu Purāṇa has two lines at the beginning of this group, viz., “the Karus, the Pāñcālas, and the Śālvas, and the Jāṅgalas, the Śūrasenas, the Bhadrakāras, the Bodhas and the lords of Śata-patha” (xlv. 109 and 110); much like a passage in the MahāBhārata (Bhīṣma-P., ix. 346-7). The Matsya has the same lines, but gives the last two names as Bāhyas and Paṭaccaras (cxiii. 35, 36). For the Kurus, see canto lviii. verse 9; for the Pāñcālas, canto lviii. verse 8; for the śālvas, canto lviii. verse 6; the Jāṅgalas are no doubt the people of Kuru-jāṅgala, see note to Kuru, canto lviii. verse 9; for the Śūrasenas, see canto lviii. verse 7. The Bhadrakāras are mentioned in the MahāBhārata (Sabhā-P., xiii. 590) and may perhaps be the same as the Bhadras mentioned in Vana-P. ccliii. 15256; they appear from these passages to have been situated on the west bank of the Jumna, somewhere between Delhi and Muttra. The Bodhas are mentioned in the MahāBhārata (Sabhā-P., xiii. 590; Bhīṣma-P., ix. 347) and are probably the same as the Bodhis (Rāmāy., Ayodh-K. lxx 15), who appear to have been situated on the eastern confines of the Pañjab. I have not met Bāhyas elsewhere; it seems erroneous. Śata-patha seems to be erroneous, and Pataccara is much better. The Pataccaras are mentioned in the M-Bh. (Sabhā-P. xiii. 590-1; xxx. 1108; Virāta-P., i 11-12; &c.,) and appear from the second of these passages to have occupied the tract south of the Apara-matsyas, that is, probably the country south-west of Gwalior.


Madhya-deśa, the whole of the Ganges basin from the Pañjab as far east as the confines of Behar; but Manu restricts it and defines its limits thus (ii. 21) — north, the Himalayas; south, the Vindhya Range; west, Vinaśana which is where the R. Saras-vatī perishes in the desert (M-Bh., Śalya-P. xxxviii. 2119-20); and east, Prayāga or Allahabad.


The text Sahyasya cottare yās tu seems incorrect; these words cannot well go with the preceding verse, for no people north of the Sahya Mts. and south of the Pāripātra Mts. could be within Madhya-deśa; and they do not agree with the following words. The Vāyu Purāṇa reads sahyasya cottarārddhe tu, which I have adopted. That Purāṇa agrees and is a little fuller—“Now along the northern half of the Sahya Mts., where the river Godavari flows, that region is a delightful one within the whole of this earth. This paradise named Go-vardhana was built there by Sura-rāja for the sate of Rāma’s spouse; the trees and herbs there were brought down from above by the Muni Bharadvāja for the sake of Rāma’s spouse. He made a delightful wooded tract the private part of the palace (antaḥ-pura)” (xlv. 112-114). The Matsya Purāṇa reads Sahyasyānantare caite and is similar, but varies at the second sentence, thus—“Where for the sake of Rāma’s spouse the hill named Go-vardhana, Mandara, Gandha-mādana, trees from Svarga and heavenly plants (oṣadhīḥ, acc.) were brought down by the Muni Bharadvāja for the sake of the spouse; hence that region excels in flowers, therefore it has become delightful.” (cxiii. 37-39). The Rāma alluded to here must be Ramā Jāmadagnya or Paraśu-Rāma, who dwelt in this region; see the next note.


These people are here placed on the east side of the Sahya Mts. among the sources of the Godavari. This region and the country west of it on the other side of these mountains and the tract northwards to the Narmada are connected in many a story with Bhṛgu, his son Cyavana and his descendants Ṛcīka, Jamadagni and Paraśu-Rāma (e. g. M-Bh., Ādi-P. clxxviii. 6802-10; Vana-P. cxxi and cxxii with Śata-patha Brāhmana IV. i. 5; Vana-P. Ixxxix. 8364-5; cxv. 10150-2; Śānti-P. xlix. 1778—82; Vana-P. xcix. 8681—2 with Śānti-P. ii). The Bhārgavas were however a numerous race and spread into other regions; they are also mentioned as one of the eastern peoples in verse 43 below. They held a high position and appear to have been numerous in king Kṛta-vīrya’s kingdom at Māhiṣ-matī, and after his death their wealth, it is said, brought down on them the hostility of the Kṣattriyas (Ādi-P. clxxviii. 6802-15). Go-vardhana (masc.) as a city is not in the dictionary. I have not found it alluded to elsewhere.


Vāhlīka or Bāhlika or Bālhika is said in the dictionary to be the modern Balkh, and in Lassen’s map to be the ancient Bactriane; but there was another country, if not two tribes of this name, in the Pañjab. The name is written Vāhlīka or Vāhlīka and there may have been a distinction between the two words, for both are mentioned in the Bhīṣma-P. list (ix. 354 and 361). These were not uncommon names, and there were two princes called Vāhlīka between Parīkṣit and Bhīṣma in the Lunar Dynasty, and the later prince is styled a king (M -Bh., Ādi-P, xciv. 3745, and 3750-51; and xcv. 3798-3800). The Vāhlīkas are mentioned twice in the Rāmāyaṇa and are placed in the western region (Kiṣk. K. xliii. 5), and also in the northern region (ibid., xliv. 13). A distinction as between two people of this name is also indicated by the mention of two independent kings of the Vāhlīkas reigning contemporaneously in the MahāBhārata (Ādi-P. clxxxvi. 6992; Sabhā-P. xxxiii. 1266 and 1272; Udyoga-P. iii. 74 and 77). One of these two peoples was closely connected with the Madras, for Salya, king of Hadra, is also called lord of the Vāhlīkas (Ādi-P. cxiii. 4425-40; and lxvii. 2642), and his sister Mādrī is called Vāhlīkī also (ibid., cxxv. 4886); and an ancient eponymous king Vāhlīka is placed in the same Krodha-vaśa gaṇa with the eponymous kings Madraka and Suvīra (ibid., lxvii. 2695-6). The other people of this name appear to have been closely connected with the Daradas who were a mountain-tribe in the north of the Pañjab (see note to verse 38), and are the modern Dards of Dardistan; for an ancient king Darada the Vāhlīka is mentioned who did not belong to the Krodha-vāsa group (Ādi-P. lxvii. 2694), and the Vāhlīkas are linked with the Daradas (Bhīṣma-P. cxviii. 5484) and are mentioned with the Kāmbojas and Yavanas and other ultra-Pañjab tribes (Droṇa-P. cxxi. 4818; see also Sabhā-P. xxvi. 1031-2). If these inferences may be trusted, one Vāhlīka or Vāhlīka was situated in the plains of the Pañjab alongside Madra-deśa and very possibly south of it (see Rāmāy., Ayodh-K. lxx. 16-19, with note to Madrakas in verse 36, and note to Kaikeyas in verse 37) i. between the rivers Chenab and Sutlej; and the other among the lower slopes of the Himalayas, very possibly between the Chenab and Bias. The name Vāhlīka appears to have been altered in later times to Bāhīka seemingly by a punning resemblance to vahis, “outside,” because they were shut out by the Saras-vatī, Kuru-kṣetra and other natural features from the central country which remained true to Brahmanism. The people of the Pañjab were then collectively called Ārattas or Bāhīkas, and they and all the tribes beyond were stigmatized as impure and contemptible by the arrogant and intolerant brahmans of Madhya-deśa (Muir’s Sansk. Texts. II. 482, and M-Bh., Karṇa-P. xliv. 2026 &c.; see also Cunningham’s Arch. Surv, Repts., II. 6, 14,17, 195, &c).


This people is mentioned in several passages in the M-Bh. (Sabhā-P. 1. 1826; Udyoga-P. iii. 86; Bhīṣma-P. ix. 354; and Droṇa-P. xi. 398), and their name appears to he derived from an eponymous king Vātadhāta, who was of the same Krodha-vaśa group as the eponymons kings of the Vāhlīkas, Madras and Sauvīras (Ādi-P. lxvii. 2695-9). No doubt therefore they dwelt alongside those tribes. Their country Vāṭadhāna was part of the territory stretching from Pañca-nada to the Ganges, over which the hosts assembled on the Kauravas’ side spread at the beginning of the great war (Udyoga-P. xviii. 596-601), and it was in the western region (Sabhā-P. xxxi. 1190-1). Prom these data it may be inferred that the Vāṭadhānas inhabited the country on the east side of the Sutlej, southward from Ferozpur. Mann declares a Vāṭadhāna to be the offspring of an outcaste brahman and a brahman woman (x. 21), but that is no doubt an expression of the same arrogance which in later times stigmatized all the Pañjab races as outcastes (as mentioned in the last note), for Vāṭadhāna dvijas were among the people whom Nakula conquered (Sabhā-P. xxxi. 1190-1). The Vāyu Purāṇa reads Vādhadhānas (xlv. 115) erroneously.


They were an aboriginal tribe and are called mlecchas and dasyus. (M-Bh.,Vana-P. clxxxviii. 12838-40; and Mausala-P. vii. 222, &c.) They were scattered over various tracts and gained their livelihood in various ways. Three divisions of them are mentioned in the M-Bh. (Sabhā-P. xxxi. 1192), those who dwelt along the river Saras-vatī, those who lived by fishing, and those who inhabited the mountains. The first group occupied the north portion of the desert as far east as Vinaśana on that river, for it is said the river perished there because of her hatred of them (Śalya-P. xxxviii. 2119-20), and as far west as Pañca-nada (Mausala-P. vii. 220-242; and viii. 270; where Pañca-jana is probably a mistake for Pañca-nada); this group is probably the tribe of Ābhīras mentioned in Bhīṣma-P. ix. 354, and Rāmāy., Kiṣk. K. xliii. 19. The second group must no doubt mean those who lived by the sea and not simply on rivers, hence it would have inhabited the coast along the Rann of Kachh and the delta of the Indus; and it is no doubt the tribe of Ābhīras mentioned in Kiṣk. K. xliii. 5. The context indicates that the third group were the mountaineers of the Aravalli Range and the hills of Malwa; but there was another section of this group which appears to have occupied the hilly tracts in the north or west of the Pañjab, for it is classed with the Daradas and Kāśmīras (Bhīṣma-P. ix. 375; and see note to verse 38), and with the Pāradas (Sabhā-P. 1. 1832; and see note to verse 37). The Ābhīras were closely connected with the Śūdras in these three groupings (Sabhā-P. xxxi. 1192; and see note to next verse). The descendants of all these Ābhīras are the modern Ahirs who are scattered widely over Hindustan proper. Another body of Ābhīras was found in the Dekhan (see verse 47). The Ābhīras are said in the Harivaṃśa to have been dominated by the Daityas and Dānavas in ancient times, and to have been the chief inhabitants of the country from the Jumna to the peninsula of Gujarat (xciv. 5142-80).


So also in the Matsya Purāṇa (cxiii. 40). This word is written Kālajoṣaka in the MahāBhārata (Bhīṣma-P., ix. 354), but I have not found either name elsewhere. Possible readings might be Bāla-jartikāḥ or Bāla-kāthikāḥ. The former, however, seems inappropriate; the Jartikas are the same as the Bāhīkas (Karṇa-P. xliv. 2033; Arch. Surv. Repts., 11.13 and 195) who are noticed in the note to “Vāhlīkas” (page 311). The “Bālas and Kāthis” would be a preferable reading. The Bālas, according to Cunningham, occupied the northern portion of Sindh and were ousted from there about the middle of the seventh century A D. and moved south-east. The Kāthis (the Kathæi of the Greek writers), according to the same authority occupied the Rechna Doab between the Chenab and Ravi rivers and also probably the northern portion of Sindh; they have retained the former territory, but those who held the latter were driven from it about the middle of the seventh century A.D. and settled in the peninsula of Gujarat where they have given the name Kāthiāwāṛ to a district (Arch. Surv. Repts-, II. 33-37).


Aparānta means “living at the western border.” A people of this name is mentioned in the MahāBhārata (Bhīṣma-P., ix. 355), and allusion is often made to Aparānta and the Aparāntas (e.g., Vana-P., ccxvii. 7885-6; and Śānti-P., xlix. 1780-2); but the word, though it no doubt designates a people living in the extreme west, yet seems to have a general meaning in most passages (see verse 52 below), and those passages which use it in a restricted sense do not agree— thus Aparānta is stated to be a country in the middle of the sea (Raghu-V., iv. 58, commentary), yet the Aparāntikas in canto lviii., verse 34 are placed in the tortoise’s tail, that is, north of Sindhu and Sauvira which are placed by verse 30 in the right hind foot. Cunningham, judging from the spots where coins have been found, was inclined to locate them in Northern Sindh and parts of West Rajputana (Arch. Surv. Repts., XIV. 136,137) which will satisfy canto lviii., verse 34. The Vāyu Purāṇa reads Aparītas instead (xlv. 115) which seems erroneous. The Rāmāyaṇa mentions simply the Aparas (Kiṣk.-K., xliii. 23). The Matsya Purāṇa reads Purandhras (cxiii. 40) erroneously.


The Śūdras are often linked together with the Ābhīras (Mahā-Bhārata, Sabhā-P., xxxi. 1192; Bhīṣma-P., ix. 375; Droṇa-P, xx. 798; Śalya-P., xxxviii. 2119-20). They appear to have been considered dasyus (Śānti-P., clxxi. 6372, and clxxiii. 6446) and mlecchas (Vana-P., clxxxviii. 12838-40 where Śūra seems a mistake for Śūdra); yet their women are alluded to in rather favourable terms (Sabhā-P., 1. 1829). They were divided into the same three groups as the Ābhīras, viz., men of the plains, men of the sea-coast, and men of the hills (Sabhā-P., xxxi. 1192), inhabiting much the same regions (see note to Ābhīras in verse 35; for Stir as in Rāmāy., Kiṣk.-K., xliii. 19 read probably Śūdras), hence it would seem these two people were considerably intermixed and were probably closely connected aboriginal races. One group of the Śūdras was known to the Greeks as Sudrakæ, and is placed by Cunningham in the middle of the triangle of the Pañjab (Anc. Geog., I. 214-218; and Arch. Surv. Repts., II.).


This is no doubt a mistake for Pahlavas, which the Vāyu Purāṇa reads (xlv. 115) and which occurs in canto lviii., verse 30; though Pallavas are mentioned in the MahāBhārata (Vana-P., li. 1990). The Pahlavas are understood to be the Pehlavi or ancient Persians. Two people of this name are mentioned in the Bhīṣma-P. list (ix. 355 and 375), but there appear to be no data to make a distinction as the allusions to the Pahlavas are generally vague, unless it be supposed there was a Pahlava colony in the Pañjab; and this supposition would suit this verse, for the Persians were altogether outside India. The Harivaṃśa says King Sagara defeated a great confederation of Pahlavas and other people, abrogated their laws, degraded them and made them wear beards (xiii. 763-4; and xiv. 775-783); but this seems a late fable, on a par with their being called dasyus (id., cxv. 6440-3). The Rāmāyaṇa has an absurd fable about the creation of the Pahlavas and other foreign races by Vasiṣṭha (Ādi-K., lv. 18-20; and lvi. 2-3); when contending with Viśvā-mitra he made his cow create Pahlavas. Śakas, Yavanas, &e., and Viśvā-mitra destroyed them all in succession.


Or Carma-maṇḍalas as in the Bhīṣma-P. list (ix. 355) which this Purāṇa has followed closely in enumerating this group of races. The Vāyu, however, reads the same as in the text (xlv. 115). I have not found any of these names elsewhere; but the name suggests identification with Samarkand. The Matsya Purāṇa reads Ātta-khaṇḍikas or Cātta-khaṇḍikas (cxiii. 40) which appear to be erroneous.


Gāndhāra was the whole of the lower basin of the Kabul river, (Arch. Surv. Repts., II. 15, and map to p. 87). It was a famous country in ancient times, its kings ranked with the highest Indian Sovereigns, and its princesses married into the noblest royal families. The passages in which the Gāndhāras are pronounced bad and impure (e.g., Śānti-P., lxv. 2429-31; and ccvii. 7560-1; and Karṇa-P. xliv. 2070) betray the interpolated sentiments of a later age (Muir’s Sansk. Texts, II. 482).


This is not in the dictionary. The Vāyu Purāṇa reads Yavanas instead (xlv. 116), and so also the Matsya (cxiii. 41); this word is supposed to have denoted the Greeks originally, but the Yavanas appear to have been known in India long before Alexander’s time. The Bhīṣma-P. list mentions the Giri-gahvaras in this region (ix. 375).


Primarily Sindhu meant the country along the Indus, but it has generally denoted the lower portion of that country, that is, the modern Sindh more or less closely. It is placed by canto lviii., verse 30 in the Tortoise’s right hind foot, and it stretched down to the peninsula of Kāthiāwār which is called the territory of Sindhu-Rāja in the Harivaṃśa (csdv. 6407-12) The Sindhus are mentioned frequently in the MahāBhārata, and are named twice in the Bhīṣma-P. list, first in connexion with the Pulindas (ix. 348), and again in conjunction with the Sauvīras (ix. 361), but there do not appear to be any other passages which tend to shew a division. Sindhu had a well-known breed of horses (Droṇa-P., xxiii. 973).


The Sauvīras claimed descent from an eponymous king Suvīra of the same Krodha-vaśa group as the Madras and Vāhlīkas (Mahā-Bhārata, Ādi-P., lxvii. 2695-6); but the genealogy of Suvīra in the Harī-Vaṃśa (xxxi. 1679) is fanciful. Sauvīra was closely connected with Sindhu, for the two are often coupled together, and Jayad-ratha king of Sindhu was also lord of Sauvīra and the Sauvīras, and is styled Saindhava and Suvīra indifferently (Vana-P., cclxiii. 15576-81; cclxvi. 15618 and 15635-7; and cclxvii. 15639-51). Cunningham identifies Sauvīra with the country between the Indus and Jhelam, that is, the Sindh-Sagar Doab (Arch. Surv. Repts., II. 4-6, 14 and 23); that hardly agrees with the position assigned to it in the Tortoise’s right hind foot, but the collocation of races in that region is rather confused. May we suppose that Sauvīra was rather the name of the people while Sindhu more properly denoted the territory? The Sauvīras might then have inhabited the northern part of Sindhu, and Sauvīra would have been that portion only of the larger area. This theory seems to satisfy the conditions generally.

Along with the Sauvīras and Saindhavas the Kūrma Purāṇa mentions (xlvii. 40) “the Hūṇas (or Kūṇas), the Mālyas (or Śālvas), the inhabitants of Bālyā (or Kalpa).” For the Hūṇas, see note to canto lviii., verse 45; Kūṇa seems erroneous; for the Śālvas, see canto lviii. verse 6; Mālya, seems erroneous; Bālyā and Kalpa seem unidentifiable.


The Madras or Madras claimed descent from an eponymous king Madraka of the same Krodha-vaśa group as the Sauvīras and Vāhlīkas (Mahā-Bhārata, Ādi-P., lxvii. 2695-6; and see note on page 311); but his genealogy in the Harivaṃśa (xxxi. 1679) seems fanciful. Cunningham places Madra between the Jhelam and Ravi rivers, that is, in the Chej and Rechna Doabs (Arch. Surv. Repts., II. 4, 8, 14 and 39), yet he also says it is the country between the Jhelam and Bias, or between the Chenab and Bias (ibid., 196); but it could hardly have comprised much of the Chej Doab for the Kaikeyas occupied the greater part of that (see note to next verse). The capital of Madra was Śāhala (Sabhā-P., xxxi. 1197) which Cunningham has identified with the modern Sangala, on the R. Apagā which is the modern Ayak (Arch-Surv. Repts., II. 195-6 and Karṇa-P., xliv. 2033). The R. Irāvatī flowed through Madra-deśa, but near the eastern border (ibid., 2038-41; and Matsya Purāṇa cxiv. 7 and 15-18.) Madra then was the country around Sangala, with the tracts on either side watered by the Chenab and Ravi. It was a famous kingdom. The weird story told about king Yyuṣitāśva’s queen (Ādi-P., cxxi. 4695-4714) no doubt means her sons became Madras and did not originate the Madras. In later times the brahmans of Hindustan pronounced the Madras, like the Gāndhāras, base and impure (Śānti-P., ccvii. 7559-61; and Harivaṃśa, xiv. 784); see especially Karṇa-P., xliv. 2033-53, where the Madras are abused in good set terms. The Vāyu Purāṇa reads Bhaārakas erroneously (xlv. 116).


That is, the Sutlej, Śatadru-ja. But this seems mistaken, for the Vāyu Purina reads Śākas and Hradas (xlv. 116); and the Matsya Śakas and Druhyas (cxiii. 41). The Śakas, therefore, are no doubt one of the people meant (see note to canto lviii., verse 6). Hrada seems erroneous. The Druhyas may be connected with Yayāti’s son Druhyu who was king of the West, (Hari-Vaṃśa xxx. 1604 and 1618), but I have not met with them elsewhere.


This seems erroneous. These people are mentioned in verse 46, and there seems to be no ground for thinking any Kaliṅgas lived in North India; yet Kaliṅgas are mentioned in such a connexion (Bhīṣma-P., ix. 376; and Ixxi. 3132; and see note to Arkaliṅgas in verse 33), and a town Kaliṅga-nagara is mentioned in the Rāmāy. on the west of the Go-matī and not far from it (Ayodh. K., lxxiii. 14,15). A tribe called Kuliṅgas is alluded to (Droṇa-P., cxxi. 4819). The Vāyu Purāṇa reads Kulindas instead (xlv. 116), which is no doubt right. Cunningham says the Kulindas or Kunindas are the modern Kunets who occupy Kullu and the Simla hills and the slopes below, along both sides of the Sutlej (Arch. Surv. Repts., XIY. 116 and 125-130). The Kulindas extended further east along the southern slopes of the Himalayas as far as Nepal, for they were the first nation which Arjuna conquered in his Northward march from Indra-prastha (Sabhā-P., xxv. 996), and they also occupied the hills north of Mandara, that is, the Almora hills (id., li. 1858-9; and note * to page 287 above); indeed the name appears to have comprised a considerable body of hill tribes, for “all the countries of Kulinda” are spoken of (Vana-P., clxxvii. 12350).

The Matsya Purāṇa reads Pulindas (cxiii. 41). The Pulindas were a rude tribe inhabiting the Himalayas and intermixed with Kirātas and Taṅgaṇas (Vana-P., cxl. 10863-5; and Droṇa-P, cxxi. 4816-7; and see notes to verses 40 and 41) j they were considered mlecchas (Vana-P., clxxxviii. 12838-40), and are declared to have become degraded because of the extinction of sacred rites, &c. (Anuśās.-P., Ivii. 2103, &c.). There was another body of Pulindas in Central India (Sabhā-P., xxviii. 1068; and xxx. 1120; and Śānti-P., ccvii. 7559).


The Pāradas are generally mentioned with hill tribes (Sabhā-P., 1.1832; li. 1869; and Droṇa-P., cxxi. 4819). They appear to have been a hill tribe like the Kulindas and Taṅgaṇas (see note to verse 41 below) and to have dwelt in the western portion of the Himalayas (Sabhā-P., li. 1858-9), though they are placed in the tortoise’s right hind foot by canto lviii., verse 31; the races placed there, however, are strangely confused. Manu says they were Kṣattriyas and became degraded because of the extinction of sacred rites, &c. (x. 43-44); and the Harivaṃśa says king Sagara degraded them and ordered them to wear long hair (xiii. 763-4; and xiv. 775-83), and they were mlecchas and dasyus (id., cxv. 6440-42). The Vāyu Purāṇa reads Paritas instead (xlv. 116).


This is not in the dictionary. The Vāyu Purāṇa reads Hāra-pūrikas instead (xlv. 116), and the Matsya Hāra-mūrtikas (cxiii. 41); but I have not found any of these names elsewhere. Should the reading be Hāra-hūṇakas? The Hāra-hūṇas are mentioned in the MahāBhārata, as a people outside India on the west (Sabhā-P., xxxi. 1194; 1. 1844; and Vana-P., li. 1991).


The Vāyu Purāṇa reads Ramaṭas instead (xlv. 117), and the Matsya Rāmaṭhas (cxiii. 42), and the Kūrma mentions a people called Rāmas (xlvii. 41). The reading should, no doubt, be Ramaṭas, Ramaṭhas or Rāmaṭhas; they were a western people mentioned in the MahāBhārata (Sabhā-P., xxxi. 1195; Vana-P., li. 1991; and Śānti.-P., lxv. 2430). The Ramaṇas (Bhīṣma-P., ix. 374) may be the same people. There are, however, no sufficient data to identify any of them.


The Vāyu Purāṇa reads Raddha-kaṭakas instead (xlv. 117); the MahāBhārata mentions the Bāhu-bādhas (Bhīṣma-P., ix. 362) and the Bālabhadras (Karṇa-P. vi. 153); and the Matsya Purāṇa reads Kaṇṭakāras (cxiii. 42); but none of these seems satisfactory and I have not met any of them elsewhere.


These people were called Kekayas, Kaikayas and Kaikeyas. An eponymous ancestor Kaikeya is assigned to them by the Harivaṃśa (xxxi. 1679), but his genealogy seems fanciful. They were a powerful and famous nation, and were noted bowmen (Sabhā-P., iv. 126; and Vana-P., cclxvii. 15654). They inhabited the Pañjab and appear to have joined the Madras, for the two are sometimes coupled together (Sabhā-P., li. 1870; and Droṇa-P., xx. 799); and canto lviii. places them both in the tortoise’s left side (verses 42 and 45). Their capital was Rāja-gṛha (Rāmāy., Ādi-K., lxxix. 35—44) or Giri-vraja (id., Ayodh. K., Ixxi. 1; and lxxii. 1). Lassen places the Kaikeyas between the Ravi and Bias rivers. Cunningham, however, dissents and places them on the line of the Jhelam, west of the Bāhīkas, and proposes to identify Giri-vraja with Girjāk, which was the ancient name of Jalalpur on that river (Arch. Surv. Repts., II. 14); and this agrees with the Rāmāy. (Ayodh. K., lxx. 16-19).


These people are mentioned in the MahāBhārata (Bhīṣma-P., ix. 374). The Vāyu Purāṇa reads Daśa-mānikas (xlv. 117). Does the word mean “the ten tribes of Mālikas”? The Mālikas may perhaps be identified with the Malli (Arch. Surv. Repts., II. 37). The Matsya Purāṇa reads Daśa-nāmakas (cxiii. 42), which seems mistaken.


Upa-ni-veśa, a word not in the dictionary. It seems to be synonymous with ni-veśa which appears to be the word meant in the corresponding passage in the Bhīṣma-P. list (ix. 374, kṣattriyā yoniveśāśca; but kṣattriyopaniveśāś ca would be preferable). Ni-veśa is used elsewhere in the MahāBhārata (e.g., Sabhā-P., xiii. 615; and xix. 798), and in the Rāmāy. (Kiṣk.-K, xliii. 24), and appears to denote a military colony or settlement in a foreign country.


That is, Śūdras as a caste, and not as a race; as a race they have been mentioned in verse 36.


The Kāmbojas were in the extreme north of the Pañjab beyond the Indus, and were classed with the Daradas (Sabhā-P., xxvi. 1031), with Yavanas and Śakas (Udyoga-P., xviii. 590), and with Cinas (Bhīṣma-P., ix. 373). Their country was famous for its large and fleet breed of horses which are often mentioned (Sabhā-P., 1. 1824; Bhīṣma-P., Ixxi. 3131; Droṇa-P., xxiii. 972; cxxi. 4831-2;—also Rāmāy., Ādi-K., vi. 24; and Sundar.-K., xii. 36;—and Raghu-V., iv. 70). Lassen places Kāmboja doubtfully south of Kashgar and east of the modern Kafiristan (Ind. Alt., map). They were Aryans by language (Muir’s Sansk. Texts, II. 36S-9). Manu says (x. 43-44) they were Kṣattriyas and became degraded through the extinction of sacred rites, &c.; they are called mlecchas (Vana-P., clxxxviii. 12838-40) and said to have evil customs (Śānti-P., ccvii. 7560-61). The Harivaṃśa says they were degraded by King Sagara and ordered to shave the whole of the head like the Yavanas (xiii. 763-4; and xiv. 775-83). The Rāmāy. has an absurd fable about their origin (Adi K. lvi. 2; see page 314 note *).


Darada is the modern Dardistan, the country north of Gāndhāra and north-west of Kāśmīr. This region satisfies all the allusions to the Daradas. They were a hill people (Droṇa-P., cxxi. 4835—7 and 4846-7; neighbours of the Kāśmīras ( id., Ixx. 2435), of the Kāmbojas (Sabhā-P., xxvi. 1031), and of the Cīnas and Tuṣāras (Vana-P., clxxvii. 12350); they fought largely with stones and were skilled in slinging stones (Droṇa-P., cxxi. 4835-47). Mann says they were Kṣattriyas and became degraded because of the extinction of sacred rites, &c. (x. 43 and 44); and like the Pāradas and others they were considered mlecchas and dasyus (Hari-Vaṃśa, cxv. 6440-6442).


This word is also written Barvara and Barbara, and often means any barbarous race. The Varvaras are generally mentioned in conjunction with the Śakas or Yavanas (Sabhā-P., xxxi. 1199; Vana-P., ccliii. 15254; Śānti-P., ccvii. 7560-61; &c); and from these allusions it appears they were mainly a western or north-western race; but Varvaras were also to be found in the east or north-east of India (Sabhā-P., xxix. 1088), and seemingly also in the south (Vana-P., li. 1989) like the Śavaras (Śānti-P., lxv. 2429). The name no doubt represents the rolling of the letter r or rough and unknown speech; hence it would be applied to various rude tribes.


This is not in the dictionary. The Vāyu Purāṇa reads Priya-laukikas instead (xlv. 118), but I have not found either word elsewhere. Canto Iviii mentions certain Bhoga-prasthas (verse 42) in the north. All these names seem suggestive, and may perhaps be equivalents of Utsava-sanlceta (people who have no marriage and practise promiscuous intercourse, utsava meaning affection, and saṅketa, a gesture of invitation) a people mentioned in the M.-Bh. in the north among the hills (Sabhā-P., xxvi. 1025) and west (id., xxxi. 1191); though also in the south (Bhīṣma-P., ix. 363); and the Raghu-V., places them in the Himalayas (iv. 78). This derivation of Utsava-saṅketa is given in a note to the commentary on Raghu-V., iv. 77.


The Chinese; but Cīna comprised the country of Thibet along the whole range of the Himalayas, for the Cīnas are linked with the Kāmbojas in the north-west (Bhīṣma-P., ix. 373), they are frequently mentioned among the retinue of Bhaga-datta king of Prāg-jyotiṣa, in the east (e.g., Udyoga-P., xviii. 584-5; see note to verse 44 below), and they were near the sources of the Ganges in the country midway between those regions (Vana-P., clxxvii. 12350; Śānti-P., cccxxvii. 12226-9). The country had a valuable breed of horses (Udyoga-P., Ixxxv. 3049). In the MahāBhārata the Cīnas are always spoken of with respect and even admiration (Udyoga-P., xviii. 584—5), and one of their kings called Dhautamūlaka is classed among eighteen famous ancient kings who extirpated their kinsmen (id., Ixxiii. 2730); hence Manu’s remark that the Cīnas were Kṣattriyas and became degraded because of the extinction of sacred rites, &c. (x. 43 and 44) betrays the sentiments of a later age. The Vāyu Purāṇa reads Pinas (xlv. 118) erroneously. The Rāmāy. mentions also Āpara-cīnas (Kiṣk.-K., xliv. 15), “the further Cīnas.”


For tu khārāś read tukhārāś. The Tukhāras are mentioned in the MahāBhārata (Sabhā-P., 1. 1850) and Rāmāy. (Kiṣk.-K., xliv. 15). The Vāyu Purāṇa reads Tuṣāras (xlv. 118), and they are mentioned in the MahāBhārata (Vana-P., li. 1991; Śānti-P., Ixv. 2429; &c.). The two names seem to mean the same people. They were an outside northern race bordering on the Himalayas (Vana-P., clxxvii. 12350). In the Harivaṃśa they are classed along with Śakas, Daradas, Pahlavas, &c., and considered to be mlecchas and dasyus (cxv. 6410-42), they are ranked with wild hill-tribes as originating from king Vena’s sins (v. 310-11), and are said to have been repressed by king Sagara (xiv. 784). Lassen identifies them with the Tochari, and places them on the north side of the Hindu Kush (Ind. Alt., map). The Rāmāy. has an absurd fable about their origin (Ādi.-K., Ivi. 3; Bee page 314, note *).


Bahulā. The Vāyu Purāṇa reads Pahlavas or Ratna-dhāras instead (xlv. 118), probably erroneously; the former have been mentioned in verse 36. I have not met with the latter word elsewhere, but it may be noticed that great quantities of precious stones were found among the Tukhāras and other northern nations (Sabhā-P., 1. 1849-50).


Vāhyato-narāḥ. The Vāyu Purāṇa reads Vāhyatodarāḥ or kṣatodarāḥ instead (xlv. 118) erroneously.


This tribe is mentioned in the Bhīṣma-P. list (ix. 376). The Harivaṃśa Bays that king Raudrāśva’s ten daughters all married the ṛṣi Prabhā-kara of Atri’s race and gave rise to the Atreyas (xxxi. 1660-68); and Atreyas are mentioned as a family of brahmans dwelling in Dvaita-vana (Mahā-Bhārata, Vana-P., xxvi. 971) which was a forest and lake near the Saras-vatī (ibid., clxxvii. 12354-62). The Matsya Purāṇa reads “the Atris” (cxiii. 43), which is the same. Are they to be identified with the Ātreya gotra of brahmans (Risley’s Tribes and Castes of Bengal, I. 27) formerly living perhaps in Sirmour or Garhwal; or to be connected with R. Ātreyī (Sabhā-P., ix. 374) the modern Atrai in North Bengal? The former seems more probable.


Or Bhāradvājas; they are named in the Bhīṣma-P. list (ix. 376). Bharadvāja is often mentioned in the MahāBhārata in connexion with the upper part of the Ganges near the hills (e.g., Ādi-P., cxxx. 5102-6; clxvi. 6328-32; Vana-P., cxxxv. 10700—728; and Śalya-P., xlix. 2762—2824). These were no donbt his descendants, living in Garhwāl or Kumaon. The name Bharadvāja is given to various caste divisions (Risley’s Tribes and Castes of Bengal, I. 96.)


The Vāyu (xlv. 119) and Matsya (cxiii. 43) Purāṇas read Prasthalas and they are no doubt the same as the Proṣakas (Bhīṣma-P., ix. 376) all being placed in the same connexion. If Cunningham is right in identifying Lampāka with Lamghan (see second note below), Puṣkala suggests Puṣkalāvatī or Puṣkurāvatī (Rāmāy., Kiṣk.-K., xliii. 23), the ancient capital of Gāndhāra (Anc. Geog., I. 49), but the Gāndhāras have been mentioned in their proper place in verse 36 above. I have not met the name Proṣakas any where else.

Prasthala was a country closely connected with Trigarta, for Suśarman king of Trigarta is also called lord of Prasthala (Virāṭa-P., xxx. 971; Bhīṣma-P., lxxv. 3296; lxxxviii. 3856; and Droṇa-P., xvii. 691), and Trigarta comprised the territory from Amballa and Pattiala to the R. Bias (see note to verse 57). Prasthala was also near the Pañjab (Droṇa-P., xvii. 691; and Karṇa-P., xliv. 2063-70), and in the second of these passages its people are classed along with the Pañjab nations, and all according to the ideas of a later age were pronounced degraded (Muir, Sansk. Texts. II. 482). Prom these data it seems Prasthala must have been the district between Perozpur, Pattiala and Sirsa. If this position be right, the Prasthalas do not fall into the group of northern peoples named in the text, and the correct reading cannot be Prasthalas.


The Vāyu Purāṇa reads Kaserukas (xlv. 119); and the Matsya Daserakas (cxiii. 43), I have not met the first form of name elsewhere; but the Daśerakas, or Dāśerakas, or Dāserakas are mentioned as joining in the great war in the MahāBhārata (e.g., Bhīṣma-P., 1. 2080; cxviii, 5483; Droṇa-P., xi. 397; and xx. 798); and they appear to have comprised several bands, as the word gaṇa is nearly always added to the name; but there are no data to identify them.


This name occurs in the MahāBhārata (Droṇa-P., cxxi. 4846—7) and there the Lampākas are described as a mountain tribe, like the Daradas and Pulindas, who fought largely with stones and were skilled in slinging stones; but otherwise there are very few references to them in the MahāBhārata Lassen identifies Lampāka with the Lambagæ and places them south of the Hindu-Kush, in modern Kafiristan. Cunningham says Lampāka is the modern Lamghan, north-east of Kabul (Anc. Geog., I. 17 and 27), which agrees with Lassen. The Matsya Purāṇa reads Lampakas (cxiii. 43), no doubt by a mistake.


The Vāyu Purāṇa reads Stanapas or Tanapas (xlv. 119). I have not met any of these names elsewhere, but the latter words resemble the Stana-yoṣikas (Bhīṣma-P., ix. 376), and also Tanayas (ibid., 371), whose grouping however is different. The Matsya Purāṇa reads Talagānas (cxiii. 43), which seems erroneous. Perhaps the Śulakāras may be identified with the Sunuwārs, a cultivating tribe of Nepal, forming part of the highest class (Risley’s Castes and Tribes of Bengal, II. 281).


The Vāya Purāṇa reads Pīḍikas instead (xlv. 119). Canto Ivīīi. verse 37 places the Cūlikas in the Tortoise’s tail at the westernmost part of India. I have not met with either name elsewhere. The Matsya Purāṇa reads Sainikas (cxiii. 43), “Soldiers.’'


Or Juguḍas according to the Vāyu Purāṇa (xlv. 119). The Jāguḍas are mentioned in the MahāBhārata (Vana -P., li. 1991). The Matsya Purāṇa reads Jāṅgalas (cxiii. 43), which is of no help, for it cannot refer to Kuru-jāṅgala (see note to Kurus, canto lviii, verse 9), and I have not met with- any other Jāṅgala; bnt the same Purāṇa mentions the Jaguḍas as a people through whose country the Indus flows, so that they appear to be north or east of Kashmir (cxx. 46—18).


The Vāyn Purāṇa reads Āpagas instead (xlv. 120). I have not met with either name elsewhere. Should the reading be Āpavas, the descendants of Vasiṣṭha? Ātreyas and Bharadvājas have been mentioned, and Gālavas are named in verse 57.


Or Ānimadras or Cānimadras. The Vāyu Purāṇa reads Cālimadrāś ca (xlv. 120). None of these names are in the dictionary, and I have not met any of them elsewhere.


The word Kirāta is no doubt the same as the modern names Kirāti and Kirānti, which mean “a native of the Kirānt-des or mountainous country lying between the Dud-Kosi and the Karki rivers in Nepal. The term includes the Khambu, Limbu and Yākhā tribes; and the Danuār, Hayu and Thāmi also claim to be Kirjinti,” but their claim is disputed by the first three tribes which are superior (Bisley’s Castes and Tribes of Bengal, I. 490). But formerly they had a much larger range and were spread along the greater part of the southern side of the Himalayas, for Arjuna encountered them in his northern expedition (Sabhā-P., xxv. 1002), Bhīma in his eastern (id., xxix. 10S9), and Nakula in his western expedition (id., xxxi. 1199). They formed a group of closely allied yet distinct tribes or clans, for two separate Kirāta kings are named (Sabhā-P., iv. 119 and 120), seven kings are alluded to (id., xxix. 1089), “all the Kirātas" are spoken of (Vana-P., li. 1990), and they are mentioned thrice in the Bhīṣma-P. list (ix. 358, 364 and 376). Their chief territory was among the mountains Kailāsa, Mandara (see page 287 note*) and Haima (Anuśās.-P., xix. 1434), that is, the region around Lake Mānasa. They were allied to the Taṅgaṇas (see next verse) and Pulindas (see page 316 note †) for the three people inhabited one large kingdom ruled by Subāhu, who was king of the Pulindas (Vana-P., cxl. 10863-6) and is also styled a Kirāta (id., clxxvii. 12349). The tribes differed much in material condition, for some were civilized and open to friendly intercourse (Vana-P., cxl. 10865-6; and Udyoga-P., lxiii. 2470), and others were clad in skins, lived on fruit and roots and were cruel (Sabhā-P., li. 1865). Their women were used as slaves (ibid., 1867). The Rāmāy. describes them as wearing thick top-knots (Kiṣk.-K., xl. 30). Manu’s remark that the Kirātas were kṣattriyas and became degraded because of the extinction of sacred rites, &c. (x. 43 and 44) reflects the opinions of a later age.


The same people are mentioned again in verse 57, but I have not found the name elsewhere, and it is not in the dictionary. The Vāyu Purāṇa reads Tomaras (xlv. 120), and the Bhīṣma-P. agrees (ix. 377). The Matsya Purāṇa mentions the Tomaras and the Haṃsa-mārgas as two tribes through whose countries flows the R. Pāvanī, one of the three large rivers which rise in the middle of the Himalaya mountain-system and flow eastward (cxx. ,57-59). The river is doubtful, but the passage places the Tomaras and the Haṃsa-mārgas in the east of Thibet.


“The duck-fowlers.” They are mentioned again in verse 56, and also in the Bhīṣma-P. list (ix. 377); and seem to be the same as the Haṃsa-padas (Droṇa-P., xx. 798) and perhaps Haṃsa-kāyanas (Sabhā-P., li. 1870); but there appear to be no data to identify them, except that they were a people in eastern Thibet as explained in the last note.


The people of Kashmir. They are named twice in the Bhīṣma-P. list (ix. 361 and 375).


Or better, as the Vāyu Purāṇa reads, Taṅgaṇas (xlv. 120); the Rāmāy. calls them Ṭaṅkaṇas (Kiṣk.-K., xliv. 20). They were a mountain tribe and are mentioned rather often in the MahāBhārata, where two sections are spoken of, the Taṅgaṇas and Para-tagaṇas (Sabhā-P., li. 1859; Bhīṣma-P., ix. 372; and 1. 2083), that is, “the nearer” and “the further” Taṅgaṇas. They were intermixed with the Kirātas and Pulindas (or Kulindas), for they all inhabited a large kingdom ruled over by Subāhu, which was in the middle portion of the Himalayas (Vana-P., cxl. 10863-5; Sabhā-P., li. 1858-9); . and they are also linked with the Ambaṣṭhas (Droṇa-P., cxxi. 4819). They are said to have occupied the upper part of the valley of the R. Sarayū (diet.). Like other hill tribes they fought largely with stones and were skilled in slinging stones (Droṇa-P., cxxi. 4835-47).


This resembles Śūlakāras in the last verse. The Vāyu Purāṇa reads Cūlikas (xlv. 121), which has also been mentioned in that verse. The Śulikas are mentioned in the Matsya Purāṇa as a people through whose country flows the R. Cakṣu, one of the three large rivers which rise in the middle of the Himalaya mountain-system and flow westward (cxx. 45, 46). Cakṣu may perhaps he meant for Vakṣu, which is the Oxus; if so, the Śūlikas would be a people on the Oxus in Turkestan.


Kuhaka means a juggler. The Vāyu Purāṇa reads Cāhukas or Ahukas or Āhukas (xlv. 121); Āhuka was the name of a family of the Andhakas (e,g., MahāBhārata, Udyoga-P., lxxxv. 3041; and Harivaṃśa, xxxviii. 2017-24), but they were in the west and cannot be meant here. I have not met with any of these words elsewhere as the name of a people in the north. The proper reading may be Kuhukas. Kuhuka would be the same as Kuhu, and the Kuhns are mentioned in the Matsya Purāṇa as a people oh the line of the Indus (cxx. 46-48).


These people are mentioned again in verse 57. A country Ūrṇa-deśa is placed by Lassen on the Sutlej north of Garhwal (Ind. Alt., map). The whole of the upper Sutlej valley is now called Nari-khorsum or Hun-des. The Vāyu Purāṇa reads Pūrnas here (xlv. 121), which seems erroneous.


These appear to be the same as the Dārvas in verse 57. They were a northern people and are generally associated with the Trigartas and Daradas (Sabhā-P., xxvi. 1020; and li. 1869) and other tribes in the north of the Pañjab (Bhīṣma-P., ix. 362) A river or town called Darvī is mentioned (ibid.), and a tīrtha Darvī-saṅkramaṇa is placed between the sources of the Jumna and Indus (Vana-P., lxxxiv. 8022-4); and this tract perhaps was their territory. But Lassen places the Dārvas between the Indus and Jhelam in the north-west of Kashmir (Ind. Alt., map).


This seems incorrect. The Vāyu Purāṇa reads Andhravākas (xlv. 122), which is hardly acceptable, the Āndhras being properly in the South, rather than in the East, and being presumably intended in verse 48 (see note to Andhas). The Matsya Purāṇa reads Aṅgā vaṅgā instead (cxiii. 44), which is preferable, but these nations are mentioned below (see page 325 note ‡ and page 326 note *).


The Matsya Purāṇa reads Madgurakas (cxiii. 44), and the Vāyu Sujarakas (xlv. 122). I have not found any of these names elsewhere, except that Madguras, “divers,” are mentioned in a totally different connexion in the Harivaṃśa (xcv. 5233-9). Seemingly the word should be connected with Modā-giri in the Eastern region where a kingdom once existed (Mahā-Bhārata, Sabhā-P., xxix. 1095) j is it to be identified with the modern Mungir (commonly Monghyr) on the Ganges in Behar, where there is a small outcrop of hills. Cunningham says Mudgala-puri, Mudgalāśrama (to which the Matsya Purāṇa reading approximates) and Mudga-giri were the old names of Mungir; and an earlier name was Kaṣṭa-haraṇa-parvata (XV. 15 and 18), but this last is open to the objection that no name can well be older than that preserved in the MahāBhārata The Mudgalas are mentioned in Droṇa-P., xi. 397.


Or Antar-giri. as the Matsya Purāṇa reads (cxiii. 44). They are mentioned in the Bhīṣma-P. list (ix. 357). The name, no doubt, means “those who dwell amid the hills,” and as the people are placed in all these passages in proximity to the Aṅgas, it seems reasonable to identify Antar-giri with the Hajmahall hills (in the modern district of the Santhal Parganas) which form a marked natural division between Anga and Vanga. In the only other passage where I have found this name (Sabhā-P., xxvi. 1012) Antar-giri, Vahir-giri (see next note) and Upa-giri are mentioned in obvious contradistinction and are placed in the Northern region; it is doubtful, therefore, whether they denote the tracts mentioned here; and they may perhaps refer to some portion of the slopes of the Himalayas.


Or Vahir-giri as the Matsya Purāṇa reads (cxiii. 44). The name, no doubt, means “those who dwell outside the hills,” and these people are mentioned along with the Aṅgas and Malajas in the Bhīṣma-P. list (ix. 357). If we may identify Antar-giri with the Rajmahall hills (see the last note), Vahir-giri might well designate the outskirts of those hills bordering on Anga, that is, the southern portions of the Bhagalpur and Monghyr districts and the lands bordering thereon to the south in the Santhāl Parganas and Hazaribagh.


I have not met this name elsewhere, though it is stated in the dictionary to be the name of a people and analysed thus, Pravaṃ-ga=Plavaṃ-ga; I would suggest, however, that it should be read here as Pra-vaṅgas, “those who are in front of the Yaṅgas,” i.e., the Aṅgas. The Matsya Purāṇa reads Aṅgas and Vaṅgas (cxiii. 44). The Aṅgas are clearly meant. Anga was a distinct and settled country in early times, and its princes were allied with Aryan royal families (Mahā-Bhārata, Ādi-P., xcv. 3772 and 3777; and Rāmāy., Ādi-K., x. 1-10). This people are said to have been so called after an eponymous king Anga; he, Vanga, Kaliṅga, Puṇḍra and Suhma are described with considerable circumstantial detail as the five sons of king Bali’s queen (Bali being king of the Eastern region) by the ṛṣi Dīrgha-tamas (Mahā-Bhārata, Ādi-P., civ. 4217-21; and Harivaṃśa, xxxi. 1684-93). Anga comprised the modern districts of Bhagalpur and Monghyr, excluding the extreme north and south portions. The ancient name Anga dropped out of use and Bihar (of Buddhist origin) has usurped its place; is the word Pra-vanga here significant of the change? The capital was first called Mālinī, and that name is said to have been superseded by the name Campā in honour of a king Campa, Loma-pāda’s great grandson (Hari-Vaṃśa, xxxi. 1699 and 1700; and MahāBhārata, Śānti-P., v. 134-5), but the Rāmāy. makes a punning connexion between this name and the groves of campaka trees around the town (Ādi-K., xvii. 23); it is the modern Bhagal-pur on the south bank of the Ganges (Vana-P., lxxxv. 8156). The tract near Campā was called Sūta-viṣaya (Vana-P, cccvii. 17150-51), that is, “the land of bards or charioteers.” The Aṅgas are mentioned twice in the Bhīṣma-P. list (ix. 353 and 357); it does not appear why.


This is, no doubt, a mistake for Vangeyas which the Vāyu Purāṇa reads (xlv. 122), and Vaṅgas which the Matsya mentions (cxiii. 44). The Vaṅgas or Vangeyas were the people of Vanga or Banga, the original of the modern Bengal. Vanga was a distinct country in early times and is frequently mentioned, though the references to it very rarely convey any definite information. It lay beyond Anga, to the south-east; and was connected with Kaliṅga, for the Aṅgas, Vaṅgas and Kaliṅgas are constantly linked together as people closely allied by race and position, (e.g., Droṇa-P., lxx. 2436). And the Vaṅgas are said to have been so called after an eponymous king Vanga who was Anga’s and Kaliṅga’s brother (see last note). Vanga comprised the northern portion of Western and Central Bengal, i.e., the modern districts of Birbhum, Moorshedabad, Bardwan and Nuddea. Its capital in early times does not appear to be mentioned. In later times the name was extended over the whole of Central Bengal, for the Raghu Vaṃśa describes the Vaṅgas as dwelling in the islands of the Ganges delta, warring chiefly in boats, and transplanting their rice seedlings into the fields just as at the present day (iv. 36, 37). In those early times the upper part of the delta consisted of numerous islands separated by large rivers, and the southern part could not have been formed.


The Vāyu Purāṇa reads Māladas (xlv. 122) which appears preferable, and this may mean the people of the modern district of Maldah, in which the old cities of Gauṛ and Paṇḍua are situated, while the town Maldah itself is old (Cunningham, Arch. Surv. Repts., XV. 77). The Maladas are mentioned as an eastern people in the MahāBhārata (Sabhā-P., xxix. 1081-2; and also Droṇa-P., vii. 183), but without data enough to say where they were.


The Vāyu Purāṇa reads Māla-vartinaḥ (xlv. 122). The MahāBhārata mentions the Māna-varjakas (Bhīṣma-P., ix. 357) and they appear from the context to be the people meant here. The name seems intended to carry a meaning, either “people who live decorously” according to the text, or “people who are devoid of decorum” according to the last word. Does it refer to a wild tribe in a state of nature? Or does Māna-vartika (Māna-vartin) mean Mān-bhūm (Māna-bhūmi) a district in West Bengal? The Mālavānakas mentioned in the Bhīṣma-P. list (ix. 367) belong to a different group altogether.


The Matsya Purāṇa reads Suhmottaras (cxiii. 44), which is preferable. This means the “people north of Suhma.” Suhma was a well-known country. It was generally classed with Puṇḍra (e. g., MahāBhārata, Ādi-P., cxiii. 4453) and both of them are declared to be closely allied to Anga, Vanga and Kaliṅga by being derived from five eponymous kings of those names who were brothers (Mahā-Bhārata, Ādi-P., civ. 4217-21; Harivaṃśa, xxxi. 1684-93). Suhma was near the sea (Sabhā-P, xxix. 1099; and Raghu-V., iv. 34 and 35), and Dāma-lipta (Tāmra-lipta, the modern Tamluk, see next verse) is said to be within its borders in the Daśa-kumāra-carita (Story of Mitra-gupta). Suhma therefore corresponded with the modem districts of Midnapur and Baṅkura and perhaps also Purulia and Manbhum in West Bengal. Suhmottara would be the tract north of that, and was probably the same as Pra-suhma (Sabhā-P., xxix. 1090). The Mahyuttaras of the Bhīṣma-P. list (ix. 358) seem to be the same people under an error in the spelling.

The reading Suhmottarāḥ, however, is hardly satisfactory in omitting the Suhmas and referring indefinitely to the people north of them, and I would suggest that the proper reading should be Suhmotkalāḥ, “the Suhmas and Utkalas.” The Utkalas were well-known (though not I believe mentioned often in the MahāBhārata), and were a rude tribe of very early origin, for they do not appear to have had any close affinities with the races around them, and the Harivaṃśa throws their origin back to the fabulous time of Ilā (x. 631-2). Their territory reached on the east the R. Kapiśā (Raghu-V., iv. 38), which Lassen identifies with the modem Subarṇa-rekhā near the northern boundary of Orissa (Ind. Alt., Map), but which I propose to identify with the R. Gossye in Midnapore (see page 301 note †); and on the west they touched the Mekalas, for the two people are coupled together in the MahāBhārata (Bhīṣma-P., ix. 348; see also Droṇa-P., iv. 122, and Karṇa-P., xxii. 882) and Rāmāy. (Kiṣk.-K., xli. 14), and the Mekalas were the inhabitants of the Mekala hills, i.e., the hills bounding Chhattisgaṛh on the west and north. Northward dwelt the Puṇḍras and southward the Kaliṅgas. Hence Utkala comprised the southern portion of Chuta Nagpur, the northern Tributary States of Orissa and the Balasore district. Various derivations have been suggested of the name Utkala, but I would only draw attention to some of the above passages where Utkala and Mekala are placed together as if their names possessed something in common. See also in verse 53.


The Matsya Purāṇa reads the same (cxiii. 44), but I have not found them mentioned elsewhere They appear from the context to be the same as the Prāvṛṣeyas of the Bhīṣma-P. list (ix. 358).


These are mentioned in the Bhīṣma-P. list (ix. 358), and were perhaps an off-lying branch of the Bhārgava race in the East; see note to verse 35. The Harivaṃśa mentions a prince called Bhārga or Bhārgava, who founded Bhṛgu-bhūmi or Bhārga-bhūmi; and as he was a grandson of Divodāsa king of Benares, his country may perhaps have been in the Eastern region (xxix. 1587 and 1597; and xxxii. 1753). The Bhīṣma-P. list names also Bhārgas here (loc. cit).


The Vāyu Purāṇa reads Geyamarthakas (xlv. 123), and the Matsya Geyamālavas (cxiii. 44), and the Bhīṣma-P. list omits the corresponding name (see ix. 358). None of these names are in the dictionary, and I have not met any of them elsewhere.


Prāg-jyotiṣa was a famous kingdom in early times and is often mentioned in the MahāBhārata The references to it, however, are rather perplexing, for in some passages it is called a Mleccha kingdom ruled over by king Bhagadatta, who is always spoken of in respectful and even eulogistic terms (e.g., Sabhā-P., xxv. 1000-1; and 1. 1834; Udyoga-P., clxvi. 5804; and Karna-P., v. 104-5), and in other passages it is called a Dānava or Asura kingdom ruled over by the demons Naraka and Muru (Vana-P., xii. 488; Udyoga-P., xlvii. 1887-92; Harivaṃśa, cxxi. 6791-9; cxxii. 6873, etc.; and clxxiv. 9790; and Annotations to Kiṣk.-K., xliii. in Gorresio’s Rāmāyaṇa); while in some other passages the allusions seem mixed (e.g., Sabhā-P., xiii. 578-80, which seems to call Bhagadatta a Yavana; and as to this, see id. 1. 1834—6). The second class of passages occur, I believe, only in descriptions of Kṛṣṇa’s exploits; they are spoilt by hyperbolical laudation and are probably later than the first class. Prāg-jyotiṣa was placed in the North region (Sabhā-P., xxv. 1000; and Vana-P., ccliii. 15240-2), but was also considered to be in the East as in the text here. North of it seemingly lay tracts called Antar-giri, Vahir-giri and Upa-giri (Sabhā-P., xxv. 1000— xxvi 1012) which appear to be the lower slopes of the Himalayas and the Terai; and it was close to the mountains for Bhagadatta is called Śailālaya (Strī-P., xxiii. 644). It bordered on the Kirātas and Cīnas for they formed his retinue (Sabhā-P., xxv. 1002; Udyoga-P., xviii. 584-5). He also drew his troops from among the people who dwelt in the marshy regions near the sea, Sāgarānūpa (Sabhā-P, xxv. 1002; xxxiii. 1268-9; and Karṇa-P., v. 104-5), and it is even said he dwelt at the Eastern Ocean (Udyoga-P., iii. 74); these marshy regions can only be the alluvial tracts and islands near the mouths of the Ganges and Brahma-putra as they existed anciently. These data indicate that Prāg-jyotiṣa comprised the whole of North Bengal proper. The Raghu Vaṃśa places it seemingly beyond the Brahma-putra (iv. 81); but Kālidāsa was a little uncertain in distant geography. Its capital was called Prāg-jyotiṣa also. Although the people were mlecchas, the Rāmāyaṇa ascribes the founding of this kingdom to Amūrta-rajas, one of the four sons of a great king Kuśa (Ādi-K., xxxv. 1-6). Amūrta-rayas, as the name is generally written in the MahāBhārata, is mentioned there simply as father of the famous king Gaya (e.g., Vana-P., xcv. 8528-39; and Droṇa-P., lxvi. 2334, &c).


This seems an impossible name here (see verse 36). The Vāyu Purāṇa reads Muṇḍas instead (xlv. 123) which is permissible. The Muṇḍas are a large Drāviḍian tribe in Chuta Nagpur (Risley’s Tribes and Castes of Bengal, II. 101), and are named in the MahāBhārata (Bhīṣma-P., lvi. 2410).

The Matsya Purāṇa however reads Puṇḍras instead (cxiii. 45) and the Bhīṣma-P. list also mentions them in this region (ix. 358). This is the best reading, for the Puṇḍras were held to be closely allied to the Aṅgas, Vaṅgas and Suhmas (see page 325 note ‡, page 326 note * and page 327 note *), and should rightly be placed here along with those races, rather than in the South according to verse 45. The name occurs in various forms, Puṇḍraka (Sabhā-P., iv. 119), Pauṇḍra, (Ādi-P., clxxxvii. 7020), Pauṇḍraka (Ādi-P., clxxxvi. 6992; Sabhā-P., xxxiii. 1270) and Pauṇḍrika (Sabhd-P., li. 1872). They appear to be used often as if equivalent (e. g., Sabhā-P., xiii. 584), and yet a distinction seems to be made between Puṇḍras and Pauṇḍras for they are separately mentioned in the Bhīṣma-P. list (ix. 358 and 365); and Puṇḍras, Puṇḍrakas and Pauṇḍrikas are all mentioned in one passage (Sabhā-P., li. 1872-4). All, however, appear to have composed one people, and they were not a barbarous nation. From the arrangements of names and descriptions given in various passages (Ādi-P., cxiii. 4453; Sabhā-P., xiii. 584; xxix. 1091-7; Vana-P., li. 1988; Āśvamedh.-P., IxxxiL 2464-5) it appears the Puṇḍras had the Kāśis on their north, the Aṅgas, Vaṅgas and Suhmas on their north-east and east; and the Odras on their south-east; hence their territory corresponded to the modern Chuta Nagpur with the exception of its southern portions. Their bounds on the south were no doubt the land of the Utkalas (see page 327 note *). In one passage (Ādi-P., lxvii. 2679) it is stated an ancient king Balīna reigned over both Pauṇḍra and Matsya; this suggests that their territory extended to near the R. Chumbal anciently (see page 307 note *), and tends to part them from the Aṅgas, Vaṅgas, &c.


Videha was a famous country in early times. Cunningham says it appears to have comprised the northern portion of North Behar from the R. Gan dak to the R. Kauśikī or Kosi (Arch. Surv. Repts., XVI. 34 and map); but its western boundary was the Sadā-nīra (see page 294 note ‡), and it seems Videha extended from the Rapti to the Kosi. Northwards it extended close to the Himalayas, and on the south it was bounded by a kingdom, the capital of which was Yaiśālī (Rāmāy., Ādi-K., xlvi. 10-11; and xlviii. 21-25), or the modern Besarh which is about 27 miles north of Patna (Arch. Surv. Repts., I. 55; and XVI. 6 and 34). The capital was Mithilā (Rāmāy., Ādi-K., xlix. 9-16; and M.-Bli., Śānti-P., cccxxvii. 12233-8), and this name often designated the country itself, especially in the Rāmāyaṇa. The people were called Videhas (or Videgha, as the earlier form was, see Śata-P. Brāh. I. iv. 14), and also Mithilas (Vana-P., ccliii. 15243). Its kings, who were often highly educated (Śānti-P., cccxxvii. 12215-25), are generally called Janaka, which seems to have been the ordinary royal title (Vana-P., cxxxiii. 10637). Cunningham says the capital was Janakpur, which is now a small town just within the Nepal border, north of where the Mozufferpur and Darbhanga districts meet (Arch. Surv. Repts., XVI. 34 and map), but I have not met this name in Sanskrit works.


Or Tāmra-liptas. The country and people are often mentioned in the MahāBhārata, and both forms of the name are used (Ādi-P., clxxxvi. 6993; Sabhā-P., xxix. 1098; and Droṇa-P., lxx. 2436). The name was modified into Tāma-liptaka which the Vāyu Purāṇa reads (xlv. 123), and Tāma-lipta (which occurs in canto lviii. verse 14), and Dāma-lipta (see Story of Mitra-gupta in the Daśa-kumāra-carita), and corrupted into the modern Tamluk. The town Tamluk is in the Midnapur district near the mouth of the Rupnarayan River. It used to be a famous port during the middle ages of Indian history. The country Tāmra-liptaka corresponded therefore to the eastern part of the present district of Midnapur.


The Vāyu Purāṇa reads Mālas (xlv. 123). This people appear to be the Mālēs (properly Māls) and Māl Pahāriyas, two Drāviḍian tribes which now inhabit the Rājmahall and Rāmgarh hills in Western Bengal (Risley’s Tribes and Castes of Bengal, II. 51 and 66). The Matsya Purāṇa reads Śālvas (cxiii. 45) erroneously.


Or Māgadhas. Magadha comprised the present districts of Gaya and Patna. It was a famous kingdom from the earliest times. The Rāmāyaṇa says it was founded by Vasu one of the four sons of a great king Kuśa (Ādi-K., xxxv. 1-9); and the MahāBhārata says it was established by Vṛhad-ratha, who was son of Vasu king of Cedi (Ādi-P., lxiii. 2361-5; and Harivaṃśa, xxxii. 1805), but who is also called an Anga (Śānti-P., xxix. 921-31.) One appears to be an eastern account and the other a. western account, but there may be truth in both accounts, for there was an interval of eight or twelve generations between the two periods spoken of. Both agree that Giri-vraja was made the capital by the founder of the kingdom, the former says by Vasu (loc. cit.), and the latter says by Vṛhad-ratha (Hari-Vaṃśa, cxvii. 6598; Sabhā-P., xx. 798-800). Cunningham has identified Giri-vraja with the modern Giryek on the Pañcana river about 36 miles north-east of Gaya (Arch. Surv. Repts., I. 16 and plate iii). Rāja-griha appears to have been another name of the capital (Ādi-P., cxiii. 4451-2; and Āśvamedh-P,, lxxxii. 2435-63), but Cunningham identifies it with the modern Raj-gir about 6 miles west of Giryek (Arch. Surv. Repts., I. 20 and plate iii). The oldest name of this country is said to have been Kīkaṭa, which occurs in Ṛg-Veda III. 53.14 (Muir’s Sansk. Texts, II. 362, 363).


The Vāyu Purāṇa reads Govindas (xlv. 123); and the Matsya Gonarddhas (cxiii. 45); and the latter people are mentioned in canto lviii. verse 23, but are placed in the South. I have not met with an Eastern people of any of these names elsewhere.


The Kūrma Purāṇa adds Kāma-rūpa (xlvii. 38), the modern Kamrup or Gauhati in Assam. It is mentioned in the Raghu-Vaṃśa (iv. 84), but not, I believe, in the Rāmāyaṇa nor MahāBhārata.


Dakṣiṇāpatha; this generally means South India below the Vindhya Range, and a line from Amara-kaṇṭaka to the north of Orissa.


This seems to be erroneous, for the Puṇḍras were not properly in the South, and they have been noticed in their appropriate place in the East (see page 329 note *). The Vāyu Purāṇa reads Pāṇḍyas instead (xlv. 124) and so also the Matsya (cxiii. 46), and this is, no doubt, the proper reading, for otherwise this nation, which was the most famous and best known in the South, would be omitted from this list. Pāṇḍya is often mentioned in the MahāBhārata; but not in the Rāmāy., except in the geographical canto (Kiṣk. K., xli. 15 and 25) which is probably an addition to the original poem. It comprised the modern districts of Madura and Tinnevelly. The capital was Mathurā, the modern Madura. The Pāṇḍyas belong to the Drāviḍian family, but the Harivaṃśa makes them, or more probably the royal house, descendants of the Paurava race; it says Pāṇḍya, Kerala, Kola and Cola were four brothers and gave origin to the four peoples of those names (xxxii. 1832—6).


This is; no doubt, a mistake for Keralas, which the Vāyu (xlv. 124) and Matsya (cxiii. 46) Purāṇas read; and the Bhīṣma-P. list twice (ix. 352 and 365; though the first mention is probably a, mistake). They were a forest-tribe (Sabhā-P., xxx. 1174-5) and are placed on the west side by the Raghu-Vaṃśa (iv. 53-54). They are said to be descended from an eponymous king Kerala, and to be closely allied to the Pāṇḍyas, Colas, &c. in the Harivaṃśa (xxxii. 1836). They appear to have- occupied the whole of the west coast from Calicut to Cape Comorin.


“The Cow-tails;” a pure fancy, stories of tailed races being common all over the world. It may correspond, to Gonarddhas in canto lviii, verse 23; but the Matsya Purāṇa reads Colas and Kulyas (cxiii. 46), and the Vāyu Caulyas and Kulyas (xlv. 124); and the proper reading should, no doubt, be Colas and Kolas. The Harivaṃśa makes these two tribes closely allied to the Pāṇḍyas and Keralas (see the last two notes).

Cola was a kingdom in early times (Sabhā-P., li. 1891-3) and is often mentioned in the MahāBhārata (e. g., Vana-P., li. 1988; and Sabhā-P., xxx. 1174, where Codra is, no doubt, a mistake for Cola; also Bhīṣma-P., ix. 367; and Droṇa-P. xi. 398). The Harivaṃśa says king Sagara degraded them (xiv. 784). Cola comprised the modern districts of Tañjore, Trichinopoly, Pudukota and South Arcot.

The Kolas are scarcely ever mentioned; yet they appear to be referred to in Sabhā-P., xxx. 1171, and Āśvamedh.-P., lxxxiii. 2476-7. Their position is uncertain. Are they to be identified with the Koravas or Kurrus, who are a vagrant tribe in Madras (Madras Census of 1891, Report, p. 304).


Canto lviii, verse 20 mentions the Sailikas; and the Vāyu Purāṇa (xlv. 125) and Matsya (cxiii. 47) read Setukus. I have not fonnd any of these names elsewhere, but Śaila occurs in Vana-P., ccliii. 15250, perhaps as the name of a country near Pāṇḍya in the extreme south, so that Śailikas might mean its people. Does Setuka refer to Rāma’s setu or Adam’s bridge and mean the people who live close to it?


The Bhīṣma-P. list mentions these people in the same connexion (ix. 366), and another Southern people called Mūṣakas twice (ix. 366 and 371). Canto lviii mentions instead of them the Ṛṣikas in the South (verse 27), and the Mṛṣikas in the South-east (verse 16). I have not found the latter name elsewhere, but the Ṛṣikas appear to have been well-known, there being one people of that name in the North (Sabhā-P., xxvi. 1033-6; Rāmāy., Kiṣk-K., xliv. 13; and Matsya Purāṇa cxx. 53) and another in the South (Kiṣk-K., xli. 16; and Harivaṃśa, cxix. 6724-6). The Matsya Purāṇa reads Sūtikas (cxiii. 47) which appears to be erroneous.


Canto lviii omits this people and names Kumuda hill (verse 26). The Vāyu Purāṇa reads Kumanas (xlv. 125), and the Matsya Kupathas (cxiii. 47). I have not found any of these names elsewhere. Probably the reading should be Kurumbas or Kurubas. The ancient Kurumbas or Pallavas occupied a territory which comprised the modern districts of Madras, Chingleput, North and South Arcot, Salem and the south-east portion of Mysore, with Kāñcī, the modern Conjeveram, for their capital, and their power attained its zenith about the 7th cent. A. D., or perhaps a century or two later After their overthrow they were scattered far and wide and are numerous now in most of the districts south of the R. Kistna in the middle and eastern parts of the Madras Presidency and in Mysore (Madras Census of 1891, Report, pp. 259 and 289).


This is, no doubt, the same as the Vana-vāsakas of the Bhīṣma-P. list (ix. 366), with which the Vāyu Purāṇa agrees in reading Vana-vāsikas (xly. 325). As this name simply means “Forest-dwellers,” it may include several races, who inhabited the great Southern forests; or it may denote the people of the kingdom called Vana-vāsin, which was founded by Sārasa in the Dekhan (Hari-Vaṃśa, xcv. 5213 and 5231-3). Perhaps they may be identified with the Bañjāris or Lambādis, who are the great travelling traders of South India, and who are supposed to be descendants of Bālin and Sugrīva the Yānara kings in the Rāmāyaṇa (Madras Census of 1891, Report, pp. 186 and 279). The Matsya Purāṇa reads Vāji-vāsikas (cxiii. 47), which seems erroneous.


The people of Mahārāṣṭra, the modern Mahraṭṭas, whom canto lviii also considers to be in the South (verse 23). The name is a late one as I have not found it in the MahāBhārata or Rāmāyaṇa. It was a large kingdom in Hiuen Thsang’s time in the 7th cent. A. D., and Cunningham makes it comprise nearly the western half of the Dekhan between the 16th and 20th parallels of latitude, with its capital at Kalyāni (Anc. Geog. of India, I. 553).

The Matsya Purāṇa reads Nava-rāṣṭras (cxiii. 47), but not well, for this country and people are mentioned in the M-Bh. as one of the kingdoms near the Kurus, and as situated in the south-west of Madhya-deśa or on the borders of Rajputana (Sabhā-P., xxx. 1110; and Virāta-P., i. 11-12); and the Harivaṃśa derives them from an eponymous king Nava, making him and the progenitors of the Yaudheyas, Ambaṣṭhas and Śivis (which were tribes in or near the Pañjab) all sons of king Uśīnara (xxxi. 1674-8). Nava-rāṣṭra is therefore out of place here.


So also in the Bhīṣma-P. list (ix. 366), or Māhiṣikas as the Matsya Purāṇa reads (cxiii. 47). These people are, no doubt, the same as the Māhiṣ-makas (Aśvamedh.-P., lxxxiii. 2475-7), that is, the people of Māhiṣ-matī. Māhiṣ-matī was an ancient and famous city (Sabhā-P., xxx. 1125-63) and was situated on the R. Narmada, at a place where the Vindhyas and the Ṛkṣa Mts. (the Satpura range) contract the valley (Hari-Vaṃśa, xcv. 5218, &c). Mucukunda was its founder according to that passage, and Mahiṣ-mat according to another (id., xxxiii. 1846-7). Their descendant was the great Arjuna Kārtavīrya {ibid., 1850-xxxiv. 1890). Māhiṣ-matī is identified with the modern Maheswar on the Narmadā in the Imp. Gaz. of India (Vol. X, p. 329), but this hardly agrees with the notices in Sanskrit writings; for Maheswar must have lain within the ancient Avanti (see verse 52), and Avanti was held to be sometimes in the South and sometimes in the West, whereas Māhiṣ-matī is never, I believe, placed anywhere but in the South. A more easterly position, such as Mandhātā or near there, seems better. At the time of the great war its king was Nila and his people were called Līlāyudhas (Udyoga-P., xviii. 592-3) or Nīlāyudhas (Bhīṣma-P., lvi. 2414). Its people were afterwards declared to have become degraded because of the extinction of sacred rites, &e. (Anuśās.-P., xxxiii. 2103-4; Muir’s Sanskrit Texts, I. 177). A Māhiṣikī, which seems to be a river, is mentioned in the Rāmāy. in this region (Kiṣk.-K., xli. 16.)


Or Kāliṅgas. Kaliṅga was an ancient kingdom, its kings were famous (Ādi-P., lxvii. 2701), and its princesses married into the Aryan royal families (e. g., Ādi-P., xcv. 3774-5, & 3780; and Śānti-P., iv). Its people were closely allied to the Aṅgas and Vaṅgas, and the three nations are often linked together (e. g., Ādi-P., ccxv. 7820; and Droṇa-P., lxx. 2436), and this connexion is emphasized by the allegation that these three and also the Suhmas and Puṇḍras were descended from five eponymous brothers (see page 325 note ‡). Kaliṅga comprised all the Eastern coast between the Utkalas on the north (Raghu-V., iv. 38) and the Teliṅgas or Telugus on the south. The R. Vaitaraṇi (the modern Byturni) flowed through it, and the Mahendra Mts. (the Eastern Ghats) were within its southern limits (Ādi-P., ccxv. 7820-24; and Raghu-V., iv. 38-43). Kaliṅga therefore comprised the modern province of Orissa and the district of Gañjam and probably also that of Vizagapatam. The Matsya Purāṇa makes Kaliṅga extend as far west as the Amara-kaṇṭaka hills (clxxxv. 12), but Kaliṅga there is, no doubt, an error for Kosala. Certain Kaliṅgas have been mentioned in verse 37 above.


Sarvaśaḥ. This seems to be rather a stereotyped phrase. The Matsya Purāṇa also reads the same (cxiii. 47). Pūrvaśaḥ would be a preferable reading, for the Kaliṅgas occupied a large part of the Eastern coast and do not appear to have inhabited any other part of the Dekhan.


These may have been an off-lying branch of this aboriginal race (see note to verse 35). The Vāyu Purāṇa reads Abhīras here (xlv. 126). The Matsya Purāṇa reads Kārūṣus (cxiii. 48); they are the same as the Karūṣas mentioned in verse 53; they come in their proper position there and are out of place here.


This is not in the dictionary. For saha vaiśikyā read either one word or saha vaiśikyair. The Vāyu (xlv. 126) and Matsya (cxiii. 48) Purāṇas read Eṣīkas or Aiṣīhas; but I have not found any of these names elsewhere.


I have not met this name elsewhere and it is not in the dictionary. The Vāyu Purāṇa (xlv. 126) and the Matsya (cxiii. 48) read Āṭavyas, and this may mean either “forest-dwellers,” or more probably “the people of Aṭavī which is mentioned as a city in the Dekhan, but without any data to identify it (Sabhā-P., xxx. 1176).


The Śavaras are an aboriginal tribe, according to some Drāviḍian, and according to others Kolarian. They are mentioned rarely in the MahāBhārata (Śānti-P., lxv. 2429; clxviii. 6294—6303; clxxiii. 641-5; and ccvii. 7559-61) and Rāmāy. (Ādi-K., i. 59; Araṇya-K., lxxvii. 6-32; &c). They are represented in these passages as dwelling in Central India and the Dekhan, as being wicked Dasyus, and as practising evil customs. They are still found scattered about in those parts and also towards Orissa, under the names Sabar, Saur, Suir, &c. In the Madras Presidency they are found chiefly in the Gañjam and Vizagapatam districts (Madras Census of 1891, Report, p. 254.) “The Savars believe their original condition to have been that of a wandering tribe, roaming through the hills of Orissa and Chota Nagpur, living on the fruits of the forest and acknowledging the rule of no recognized chief.” (Risley’s Castes and Tribes of Bengal, II. 241-246); and this belief agrees, if we extend their range, with the earliest notices of them. See also Cunningham, Arch. Surv. Repts., XVII and XX.


These people are mentioned again in verse 50 as being also in the West, and there appears to have been a Northern branch of them in the Himalayas (see page 316 note †). This Southern branch seems from the MahāBhārata to have occupied the middle portion of the Dekhan (Sabhā-P., xxx. 1120; and Bhīṣma-P., ix. 369; and Rāmāy., Kiṣk.-K., xli.17), and extended eastward where they had a great city (Sabhā-P., xxviii. 1068). They were an aboriginal tribe, for they were mlecchas (Vana-P., clxxxviii. 12838-40); they became out-castes from not seeing brahmans (Anuśās.-P., xxxiii. 2104-5); they are called wicked and are said to have practised evil customs (Śānti-P., ccvii. 7559-61). The Vāyu Purāṇa reads Pulindras (xlv. 126) erroneously.


The Vāyu Purāṇa reads Vindhya-mūlīkas (xlv. 126), which is synonymous, “those who dwell at the foot of the Vindhya mountains,” or “the aborigines of the Vindhya mountains.” I have not met this name as describing any particular people, and taken in its general meaning it would include the races mentioned in verses 53-55 below: but perhaps it may be read as an adjective to “Pulindas.” The Matsya Purāṇa reads Vindhya-puṣikas (cxiii. 48), which seems erroneous.


Vidarbha was one of the most ancient and renowned kingdoms in the Dekhan (Vana-P., xcvi. and xcvii). It comprised the valley of the Payoṣṇi, the modem Puma and the middle portion of the Tapti (see page 299 note † and Vana-P., cxx. 10289-90), and corresponded to the western part of the modern Berar and the valley-country west of that. It is said to have been founded by a king Vidarbha who built a city called Vidarbha (Hari-Vaṃśa, cxvii. 6588 and 6605-8; and Vana-P., lxxii), which seems to have been the same as Kuṇḍina the capital (Vana-P. lxxii: and Harivaṃśa, civ. 5800-7; cxviii. 6661-2). Its most famous king was Bhīṣmaka, who held the title “king of the Dekhan” (id., cxvii. 6590-1). The people were Bhojas (Udyoga-P., xlvii. 1881) or perhaps only the royal family was so called (id., clvii. 5350-1; Sabhā-P., xiii. 585-8), and so also in the Raghu-Vaṃśa with reference to a period many generations anterior (vi. 59 and 69; and vii). The name Bhoja seems to have more than one application, for the Bhojas together with the Andhakas and.Vriṣṇis belonged to the Yādava race (Hari-Vaṃśa, xciv. 5181-5204), and the name appears to have been applied also in a much wider sense to Kṣattriyas descended from Yayāti (Sabhā-P., xiii. 566—71).


The Daṇḍakas are mentioned in the MahāBhārata (Sabhā-P., xxx. 1169) and were the inhabitants of the forest region called Daṇḍakāraṇya. Daṇḍaka originally was the name of the immense forest, where Rāma went in banishment, and which is described in the Rāmāy. as covering the whole of Central India from Bundelkhand on the north to southward of the Godaveri (Journal, B. A. S., 1894, p. 211); but as this forest was gradually cleared away by the spread of the Aryan colonies, its limits diminished till at last Daṇḍaka denoted only the country around the sources of the Godaveri and lower part of the Tapti (Mahā-Bhārata, Sabhā-P., xxx. 1169; Vana-P., lxxxv. 8183-4). It could only have been at this stage that its inhabitants could well have been described by the name Daṇḍakas, and it is no doubt the people of that moderate area who are meant here. To account for the name the Harivaṃśa has provided an eponymous king Daṇḍaka who made Daṇḍakāraṇya (x. 637-9).


The people of Purikā. This may be either the famous town Puri in Orissa, or the town Purikā which Mucukunda is said to have built on the northern slope of the Ṛkṣa Mts. in the kingdom of Māhiṣ-matī (Hari-Vaṃśa, xcv. 5220-8); but the latter seems more probable, judging from the context. For Māhiṣ-matī, see page 333 note ‡. The Vāyu Purāṇa reads Paunikas, (xlv. 127); is this to be connected with Poonah, south-east of Bombay? I have not met it elsewhere.


The Vāyu Purāṇa reads Maunikas (xlv. 127); and Mauley as are mentioned in Sabhā-P., li. 1871; but I have not found any of these names elsewhere.


This as a people is not in the dictionary. The Vāyu Purāṇa reads Asmakas (xlv. 127), but Aśmaka seems to be the proper form. These people are mentioned in the Rāmāy. (Kiṣk.-K., xli. 17) and MahāBhārata (Droṇa-P. xxx vii. 1605-8); and are placed in the middle of India by canto lviii. verse 7. They may have been the descendants of Aśmaka, who was the son of king Kalmāṣa-pāda Saudāsa’s queen Madayantī by Vasiṣṭha (Ādi-P., cxxii. 4736—7; and clxxvii. 6777—91), and who founded the town Paudanya (ibid., 6791). A queen Aśmakī is mentioned in the Lunar line (id., xov. 3766).


I have not found this name elsewhere. Perhaps it may be connected with the Southern Utsava-saṅketas (Bhīṣma-P., ix. 368; and see page 319 note †).


This is not in the dictionary. It much resembles the Nāsikyas of canto lviii. verse 24, who are, no doubt, the people of Nasik, north-east of Bombay; bub they seem to be intended by Nāsikyāvas in verse 51. Naiṣadhas can hardly be meant, for they are named in verse 54 and were not in South India. The Vāyu Purāṇa reads Nairnikas (xlv. 127) which somewhat resembles the Nairṛtas of the Bhīṣma-P. list (ix. 359).


A people of this name have been mentioned in verse 32, as dwelling in Madhya-deśa. The Kuntalas here were in the Dekhan and are the same as those mentioned in Bhīṣma-P., ix. 367, and Karṇa-P., xx. 779. It appears Kuntala lay in the region between Belgaum and Bellary (Arch. Surv. of W. India, No. 5 by J. F. Fleet, p. 6; and No. 10 by J. Burgess, p. 72 note).


The Vāyu Purāṇa reads Andhras (xlv. 127), which is, no doubt, right. Andhas are mentioned in the MahāBhārata (Udyoga-P., xviii. 586; and Bhīṣma- P, x. 357), but mistakenly for Andhakas and Andhras respectively. The Andhras or Andhras were a rude race in early times (Sabhā-P., iv. 119; xxx. 1175; xxxiii. 1270; and Vana-P., li. 1988); but they established a kingdom during the third and seeond centuries B. C. Andhra was a kingdom also in Hiuen Thsang’s time in the 7th century A.D., and comprised the eastern portion of the Nizam’s territories, with its capital at Waraṅgal, according to Cunningham (Anc. Geog. of India). Another capital was Dhenukākaṭa, which is Dharaṇikoṭa near Amarāvatī on the Kistna (Arch. Snrv. of W. India, No. 10 by J. Burgess, p. 32). Andhra is said to be probably the same as Teliṅga (ibid., p. 72 note; and dictionary), and is taken to be the Sanskrit name for Telugu in the Madras Census Report. Telugur is the speech of the region extending from a parallel of latitude a little north of Madras northward as far as Ichapur in Gañjam; it does not penetrate into Mysore nor the western limits of Anantapur and Bellary, but is spoken by many of the inhabitants of the Nizam’s Dominions and the Central Provinces (Report, p. 188).


This is not in the dictionary and I have not met with it elsewhere.


“Wood-splitters.” The Vāyu Purāṇa reads Nalakālikas (xlv. 127), other forms of which are Nalakānanas and Nabhakānanas (diet.).


The Kūrma Purāṇa adds Magadhas (xlvii. 38) mistakenly; they are mentioned in their proper place in verse 41.


For Dakṣinātyās read Dakṣiṇāyās? The Matsya Purāṇa keeps pace with the text as far as the “Daṇḍakas” and then jumps at once, without any intimation, to the Western peoples beginning with the “Kulīyas” who correspond to the “Pulindas” of verse 50.


This is obviously a mistake for Sūrpārakas, which the Vāyu Purāṇa reads (xlv. 128). Sūrpāraka or Śūrpāraka (both forms seem correct, though the dictionary gives only the latter) was the country in the West where Rāma Jāmadagnya dwelt (Vana-P., lxxxv. 8185); though it is also placed in the South (Sabhā-P., xxx. 1169; and Vana-P., lxxxviii. 8337), because it was near the Southern sea in the Western region (Śānti-P., xlix. 1778-82). It bordered on the sea near Prabhāsa (Vana-P., cxviii. 10221-7), which is the modern Somnath in the peninsula of Kāthiāwār; it included the country around the mouth of the Narmadā (Anuśās.-P., xxv. 1736), and the mouth of that river was so specially connected with Rāma that it was called Jāmad-agnya (Matsya Purāṇa, cxciii. 33-34). He built the city Sūrpāraka there (Hari-Vaṃśa, xcvi. 5300), and Dr. Burgess has identified it with the small modern town Supara near Bassein, north of Bombay. The country Sūrpāraka therefore comprised the littoral tract from about Bassein to about the R. Narmadā. (Arch. Surv. of W. India, No. 10 p. 31). The proper reading in Rāmāy. Kiṣk.-K., xliii. 5, should, no doubt, be “the Sūrpārakas also” instead of “the extensive towns” (Gorresio’s Edition, Annotations).


This is not in the dictionary. The Vāyu Purāṇa reads Kolavanas (xlv. 128), but I have not found either name elsewhere. Perhaps this is to be connected with Kalwan, a town about 37 miles north of Nasik (which is mentioned in page 339, note ‖).


This is not in the dictionary and I have not found the name elsewhere; but the Durgalas are mentioned in the Bhīṣma-P. list (ix. 359). Perhaps this is to be connected with Dūṅgar-pur, a town and state about 90 miles north-east of Ahmedabad.


Or Ānīkaṭas, or Cānīkaṭas; neither is in the dictionary. The Vāyu Purāṇa reads Kālītakas (xlv. 128); but I have not found these words elsewhere. These names suggest Calicut, but that is too far south to be admissible here in the Western region.


See note to verse 47 above. This branch would be among the hills south-west of Malwa or the southern portion of the Aravalli hills probably The Vāyu Purāṇa reads Puleyas (xlv. 129), and the Matsya Kulīyas (cxiii. 49); but I have not met with either name elsewhere.


This is not in the dictionary. The Vāyu Purāṇa reads Surālas (xlv. 129), and the Matsya Sirālas (cxiii. 49); but I have not found any of these words elsewhere. Are these two names to be identified with Israel? There was an ancient Jewish colony, the modern Beni-Israel, on the Bombay coast before the 2nd cent. A. D. (Hunter’s Indian Empire, p. 234).


The Vāyu (xlv. 129), and the Matsya (cxiii. 49) Purāṇas read Rūpasas. I have not met either name elsewhere. The Bhīṣma-P. list mentions Rūpa-vāhikas (ix. 351).


This as the name of a people is not in the dictionary; but Śva-paca “dog-cooking,” (= Śva-pāka, which would not suit the metre) occurs, as the name of a degraded tribe (Mann, x. 19 and 51). Tri-śaṅku is said to have associated with the Śva-pākas, when discarded by his father (Hari-Vaṃśa, xii. 721-3). The Vāyu Purāṇa (xlv. 129) and the Matsya (cxiii. 49) read Tāpasas, which resembles the Tāpasāśramas placed in the south region by canto lviii. verse 27, which might mean the descendants of ascetics.


This is not in the dictionary. The Vāyu Purāṇa reads Turasitas (xlv. 129), but I have not met either name elsewhere. The Matsya reads Taittirikas (cxiii. 49), which resembles the Tittiras mentioned in Bhīṣma-P., 1. 2084 but there are no data to identify them.


This is not in the dictionary. The Vāyu Purāṇa reads Parakṣaras (xlv. 129), and the Matsya Kāraskaras (cxiii. 49). The Pāraśavas of canto lviii verse 31 appear to be the same people. I have not met with any of these names elsewhere, except Kāraskaras in Sabhā-P., xlix. 1804, but the Pāraśa-vas might mean a tribe which claimed descent from Paraśu-Rāma.


This is not in the dictionary. The Vāyn Purāṇa reads “Nāsikyas and others” (xlv. 130); this agrees with the Nāsikyas of canto lviii, verse 24 except that the latter are placed in the South. The Nāsikyas are, no doubt, the people of Nasik, which is an ancient and sacred city north-east of Bombay. The Matsya Purāṇa reads “and others who are called Vāsikas” (cxiii. 50), but I have not met this name elsewhere.


The Vāyu Purāṇa (xlv. 130) and the Matsya (cxiii. 50) say " within the Narmadā.”


Or Bhāru-kacchas as the Matsya Purāṇa reads (cxiii. 50). These are, no doubt, the same as the Bhṛgu-kacchas of canto lviii. verse 21. The word is the Greek Barugaza and survives in the modern Bharuch or Broach, a large town near the mouth of the Narmadā (Anc. Geog. of India). The Vāyu Purāṇa reads Bhānu-kacchras (xlv. 130) erroneously. None of these names occur I believe in the Rāmāyaṇa or MahāBhārata.


These are, no doubt, the people who dwelt along the R. Mahi (see page 294, note §) north of Baroda. The Māhikas of the Bhīṣma-P. list (ix. 364) are no doubt the same.


So also the Matsya Purāṇa (cxiii. 50). “The people who dwell along the R. Saras-vatī,” which is, no doubt, the small river of that name that flows into the sea at Prabhāsa, the modern Somnath, in the peninsula of Kāthiāwār (Vana-P., Ixxxīi. 5002-4; and Śalya-P., xxxvi. 2048-51). They are not the same as the Sārasvatas of canto lviii. verse 7, who were in Madhya-deśa. The Vāyu Purāṇa reads instead “Sahasas and Śāśvatas” (xlv. 130); I have not found either name elsewhere, but the Śāśikas of the Bhīṣma-P. list (īx. 354) are, no doubt, the same as the second of these.


This name is altogether out of place here, and the Kāśmīras have been mentioned in their proper position in verse 41. The Vāyu Purāṇa reads Kacchvīyas instead (xlv. 131) and the Matsya Kacchikas (cxiii. 51), which indicate the correct reading. They are the people of Kaccha (see canto lviii. verse 28) the modern Kachh or Kutch.


Su-rāṣṭra is a country frequently mentioned in the MahāBhārata, but the references seldom convey any definite information. It included the peninsula of Kāthiāwār and the country around the G. of Cambay—that is, not quite all the modern territory called Gujarat (Vana-P., lxxxviii. 8344-9). It is very rarely alluded to in the Rāmāy. (see once in Ādi-K., xii. 23). The old name survives in the town Surat near the mouth of the Tapti.


This form is not given in the dictionary; read Āvantyāś for Avantyāś? They are the people of Avanti; see note to verse 55 where the Avantis are mentioned again, and more appropriately, for canto lviii. verse 22 places them in the Tortoise’s right side, i.e., the South: but they were considered to be in both regions.

The Vāyu Purāṇa (xlv. 131) and the Matsya (cxiii. 51) read Ānartas, which is perhaps better, as they are placed by canto lviii. verse 30 in the Tortoise’s right hind-foot. Ānarta was the country which had for its capital Dvārakā or Dvāra-vatī or Kuśa-sthalī, the modern Dwarka on the sea-shore at the extreme west of the peninsula of Kāthiāwār (Śānti-P., cccxli. 12955; Harivaṃśa cxiii. 6265-6). It was Kṛṣṇa’s special kingdom, but it was founded long before (Hari-Vaṃśa, x. 642-9; and xciv. 5163-9).


The people of Arbuda, the modern Mt. Abu near the south end of the Aravalli hills.


For Vindha-nivāsinaḥ read Vindhya-nivāsinaḥ; see verse 55. Vindhya is used here in its general and wider meaning, as denoting the whole mountain-chain from Gujarat eastwards, and not in the precise sense given it in verses 21-23, because the Naiṣadhas, Avantis and other western people are included in this group.


I have not met with this name elsewhere, and it is not in the dictionary. The Vāyu Purāṇa (xlv. 132) and the Matsya (oxiii. 52) read Mālavas, which is no doubt correct. Mālava is the modern Mālwa; and the people are generally mentioned in the MahāBhārata as a tribe rather than as a settled nation (e.g., Sabhā-P., xxxiii. 1270; li. 1871; and Vana-P, ccliii. 15256). Mālava does not appear to have been so extensive however as Mālwa and, as this passage indicates, denoted rather the upper portion of that region bordering on the Vindhyas, west of Avanti.


The name is also written Kāruṣa, Kārūṣa and Kāruṣaka. This people constituted a powerful nation under king Danta-vakra in the Pāṇḍavas’ time (Sabhā-P., xiii. 575-7; and Harivaṃśa, xci. 4963); but they seem to have consisted of several tribes (Udyoga-P., iii. 81) and were not looked upon as closely allied to the races around them, for their origin is carried directly back to an eponymous ancestor Karūṣa, a son of Manu Yaivasvata, in the Harivaṃśa (x. 614; and xi. 658). The position of Karūṣa is indicated by the following allusions. It is linked with Cedi and with Kāśi (Ādi-P., cxxiii. 4796; Bhīṣma-P., ix. 348; liv. 2242; lvi. 2415; cxvii. 5446; and Karṇa-P., xxx. 1231), and with the Vātsyas or Vatsas (Droṇa-P., xi. 396; see page 307, note *); it was not a very accessible country (Sabhā-P., li. 1864); and here it is said to rest on the Vindhyas. Moreover Danta-vakra acknowledged Jarāsandha, king of Magadha (see page 330 note ‡) as his suzerain (Sabhā-P., xiii. 575-7; and Harivaṃśa, xci. 4963). Karūṣa therefore was a hilly country, south of Kāśi and Vatsa, between Cedi and Magadha; that is, it comprised the hilly country of which Rewa is the centre, from about the R. Ken on the west as far as the confines of Behar on the east. I have discussed Karūṣa in a paper on “Ancient Cēdi, Matsya and Karūsa” in the Bengal Asiatic Society’s Journal, 1895, Part I. p. 249.


This must be incorrect, for the Keralas were a well-known people in the South; see note to Kevalas in verse 45. The Vāyu Purāṇa (xlv. 132) and Matsya (cxiii. 52) read Mekalas, which is no doubt right. They are mentioned in the MahāBhārata (Bhīṣma-P., ix. 348; and li. 2103) and occupied the Mekala hills and the hilly country around. The Mekala hills are the hills in which the R. Sone rises (Rāmāy., Kiṣk.-K., xl. 20), and which bound Chhattisgarh on the west and north. These people are often coupled, as here, with the Utkalas.


See page 327, note *. Utkala had a wide extension and falls within this group as well as within the Eastern group.


The Uttamas are mentioned in the Bhīṣma-P. list (ix. 348) and are no doubt the same poople. The Matsya Purāṇa reads Auṇḍrāmāṣas (cxiii. 52). A people called Ūrddhva-karṇas are mentioned in canto lviii. verse 16. I have not, however, found any of these names elsewhere.


This people formed a well-known kingdom in early times (Ādi-P., cxiii. 4149; Vana-P., lxix. 2707-8; and Udyoga-P., cxc.-cxciii), and inhabited the country watered by the R. Daśārṇā, the modern Dasan, a tributary of the Jumna. They are named thrice in the Bhīṣma-P. list (ix. 348, 350 and 363), which seems inexplicable. They are mentioned both in the Eastern and in the Western regions in the accounts of Bhīma’s and Nakula’s campaigns (Sabhā-P., xxviii. 1063-5; and xxxi. 1189); the former of these allusions is correct, but the latter seems to be an error. The capital was Vidiśā, see page 343, note †.


Or better, Bhojas, as the Vāyu (xlv. 132) and Matsya (cxiii. 52) Purāṇas read. This name, as mentioned in page 335, note §, seems to have had more than one application. Bhojas as a Yādava tribe dwelt in Kṛṣṇa’s kingdom in Surāṣṭra; and Bhojas inhabited Mṛttikāvatī, which seems from the various references to it to have been situated somewhere on the northeastern limits of the modern Gujarat (Vana-P., xiv. 629; xx. 791; cxvi. 10172-6; ccliii. 15245; Mausala-P., vii. 244—5; and Harivaṃśa, xxxvii. 1980-7; and xxxviii. 2014). These may be the Bhojas mentioned in the text, inhabiting the extreme western end of the Vindhya range.


Or better, Kiṣkindhakas, as the Matsya Purāṇa reads (cxiii. 52). They are no doubt the same as the Kaiskindhyas of canto lviii. verse 18; but they cannot have any reference to Kishkindhyā in the Rāmāy., for that country lay far to the south of the Godavari (see Journal, R. A. S., 1894, pp. 255, &c.), and is referred to in MahāBhārata, Sabhā-P., xxx. 1122. The Kiṣkindhakas mentioned in the Hari-Varṃa (xiv. 784) may be the people intended here, but there appear to be no data to identify them.


The Matsya Purāṇa reads Stośalas (cxiii. 53). They are not in the dictionary, but Toṣala and Tosala are given doubtfully. I have not however found any of these names elsewhere.


These are the people of Dakṣiṇa Kosala or MahāKosala, the modern Chhattisgaṛh. In the Journal, It. A. S., 1894, p. 246, some reasons have been given for placing the Pañcāpsaras lake, where Rāma spent ten years of his exile, in this region. May one hazard the conjecture that it was in consequence of his long residence here, that a colony from North Kosala invaded this region, established a kingdom here and gave their name to this country? The connexion was ancient (Āśvamedh.-P., lxxxii. 2464-5).


The people of Tripurā, the modern Tewar, on the R. Narmadā. It was a famous city (Sabhā-P., xxx. 1164; Vana-P., ccliii. 15246) deriving its name from a legend that the demons had a triple city Tri-pura here, made of gold, silver and iron, and Śiva destroyed it at the intercession of the gods; this is a favourite subject with the poets (Karṇa-P., xxxiii and xxxiv; and see Aitareya-Brahm., I. iv. 23 and 24, for the story in an older form); see also Cunningham’s Arch. Surv. Repts., VIII. 124; IX. 54-55; and, XXI. 23; but his connexion of Tripurā with Cedi in early times is mistaken, see paper on “Ancient Cēdi, Matsya and Karūṣa” in Journal, Bengal Asiatic Society, 1895, Part I. p. 249.


For Vaidiśas read Vaidiśās. These are the people of Vidiśā. It was a famous town, the capital of the country Daśārṇa, and situated on the R,. Vetra-vatī, the modern Betwa, a little way east of Ujjain (Megha-D., i. 24, 25 and 28). It is probably to be identified with the modern Bhilsa, or rather with Bes-nagar the ancient capital which is close to Bhilsa (Cunningham’s Stupa of Bharhut, 132, &c). The Vāyu Purāṇa reads Vaidikas (xlv. 133), which seems erroneous, though the Vaidiśas are really included among the Daśārṇas in the last verse.


These are mentioned as a wild aboriginal tribe who inhabited the slopes of the Vindhya Mountains in the Harivaṃśa (v. 310-11). The Matsya Purāṇa reads Tumuras (oxiii. 53), and the Tumbumas (Bhīṣma-P., 1. 2084) may be the same.


The Vāyu Purāṇa reads Tumuras (xlv. 133); see the last note. The Matsya reads Tumbaras (cxiii. 53). This is no doubt a tribe closely allied to the last.


This people seems to be the same as the Paṭaccaras, who are mentioned several times in the MahāBhārata; see page 309, note *. The Vāyu Purāṇa reads Ṣaṭsuras (xlv. 133), and the Matsya Padgamas (cxiii. 53); but I have not met either of these names elsewhere.


The people of Niṣadha; the Niṣadhas, as the Vāyu Purāṇa reads (xlv. 133). This country is chiefly known from the story of its king Nala (Vana-P., liii.-lxxix); otherwise it is rarely mentioned. The chief data for fixing its position are Nala’s remarks to his wife when he is banished from his kingdom (id., lxi. 2317-9); and, as stated in note † to page 299, it seems to me the view which he describes could only be obtained completely from a position on the Satpura Mountains about longitude 75° E. The text says also Niṣadha rested on the Vindhya Mountains. Hence it may be inferred that Niṣadha comprised the country south of the Vindhyas between long. 74° and 75°, with Avanti to the north-east and Vidarbha to the south-east. Its capital was probably in the Tapti valley; Damayantī in wandering from it found her way north-eastward to Cedī (see note to Cedi in canto lviii. verse 16).


I have not met this name elsewhere and it is not in the dictionary. The Matsya Purāṇa reads Arūpas (cxiii. 54) which seems erroneous. The Vāyu Purāṇa reads Anūpas (xlv. 134) which is no doubt correct. Anūpa means “a country situated near water,” or “a marshy country.” It was applied to various tracts near the sea, generally in the combination sāgarānūpa, e.g., in Bengal (Sabhā-P., xxv. 1002; xxix. 1101: and xxxiii. 1268-9); in or near the Pānṇya kingdom in the South (Udyoga-P., xviii. 578); in the north and west of the peninsula of Kathiawar (Hari-Vaṃśa, cxiii. 6361-9; and cxiv. 6410-11); and on the western coast generally (Udyoga-P., iii. 81); but the name was more specially applied to a tract on the west coast which constituted a kingdom in the Pāṇḍavas’ time (Sabhā-P., iv. 123). The only country which rests on the Vindhyas and borders on the sea is the tract on the east of the G. of Cambay, north of the Narmada, and this no doubt was Anūpa. It is also indicated that Surāṣṭra, Anūpa and Anarta were contiguous countries, and that Anūpa lay beyond and south of Surāṣṭra (Hari-Vaṃśa, xciv. 5142-80). When the kings of Māhiṣ-matī (see page 333, note ‡) were powerful, the valley of the lower Narmada and Anūpa would naturally fall under their sway, and this no doubt explains why Kārtavīrya is called “lord of Anūpa” (Vana-P., cxvi. 10189-90), and king Nīla also (Bhīṣma-P., xcv. 4210).


I have not found this name elsewhere, and it is not in the dictionary. The reading should no doubt be Tundikeras as the Vāyu Purāṇa reads (xlv. 134). A Tuṇḍikera king is mentioned in the MahāBhārata (Droṇa-P., xvii. 691), and the Tuṇḍikeras (Karṇa-P., v. 138); and the Tuṇḍikeras are said in the Harivaṃśa to be a branch of the Haihaya race (xxxiv. 1895). There is a town called Tendukhera a little north of the Narmadā at nearly long. 79° E.; and as this site suits the text, it may be presumed these people occupied that position in the Narmada valley. The Matsya Purāṇa reads Śauṇḍikeras (cxiii. 54), not quite correctly.


The Vāyu Purāṇa (xlv. 134) and the Matsya (cxiii. 54) read Vītihotras which is no doubt correct. Vītihotra was a famous Haihaya king, and the Vītihotras were a branch of that race (Hari-Vaṃśa, xxxiv. 1895). They are called Vītahotras in the MahāBhārata (Droṇa-P., lxx. 2436). Being Haihayas, they probably occupied a part of the upper Narmada valley.


They have been mentioned already in verse 52. Avanti had the Narmada flowing through it (Sabhā-P., xxx. 1114; and Vana-P., lxxxix. 8354-8) and was on the lower portion of that river, for it is placed in the South in the first of these passages, and in the West in the second passage and in verse 52. It appears to have been bounded by the Ṛkṣa Mountains (Satpura range) on the south (Vana-P., lxi. 2317), but its limits on the north are not clear. Its capital, though not mentioned in the MahāBhārata, was Ujjayinī or Viśālā, the modern Ujjain, in later times (Megha-D., i. 31). Avanti therefore comprised the region of the sources of the Chambal and the country south-westward as far as the Satpura range. Two brothers Vinda and Anu-vinda are often named as the kings of Avanti in the MahāBhārata (Udyoga-P., clxv. 5753$ Droṇa-P., xcix. 3682—92; and Karṇa-P., xiii. 498-9), but they were also Kaikeyas and led Kaikeya troops (ibid., 492-524). May it be inferred a branch of the Kaikeyas had invaded and conquered Avanti?


Parvatāśrayin. These mountains are it seems only the Himalaya range. This group repeats many of the tribes mentioned in verses 40-42,


I have not met this name elsewhere. Are these the modern Newārs, who inhabit the great valley of Nepal and its vicinity, and who were the owners of the country prior to the Gurkhā invasion (Journal, Beng. As. Socy., Vol. LXIII, Part I, 213, 214 and 217). The Vāyu Purāṇa reads Nigarharas (xlv. 135), but I have not found it elsewhere. The Matsya reads Nirāhāras (cxiii. 55) which seems erroneous.


See page 323 note †. The Matsya Purāṇa reads Sarvagas (cxiii. 55) which seems erroneous.


These are probably the Uttara or Northern Kurus, for the Kurus of Madhya-deśa could not properly be described as dwelling among mountains. They seem to have been the stock from which the Kurus of Madhya-deśa separated off, for the period when Dhṛta-rāṣṭra and Pāṇḍu were born is described as a golden age, in which both branches of the Kurus engaged in happy rivalry (Ādi-P., cix. 4337-46); but the wistful recollections of their ancient home idealized it afterwards into a blissful land, where fancy gave itself free scope (Rāmāy., Kiṣk.-K., xliv. 82-115). They seem to have occupied the uppermost valleys of the Indus near its sources, with Kailāsa lying beyond (Vana-P., cxlv. 11025-35); and fervid imagination also placed them close to Mount Meru on its north side (Bhīṣma-P., vi. 207—8; and vii. 254), or in the region Hari-varṣa, and declared men could not enter their sacred land (Sabhā-P., xxvii. 1054-8). They are described as living in primitive happiness, and women had the utmost freedom there (cxxii. 4719-23; and Rāmāy., loc. cit.).


This is not in the dictionary and I have not met it elsewhere. Are these people the modern Gurungs, an important tribe of Tatar race, who dwell now throughout Nepal, but whose territory was formerly the country about Lamzung, Ghandrung and Siklis, west of the great valley of Nepal (Journal, Bengal Asiatic Society, Vol. LXIII. Part I. 213, 217 and 223—229; Risley’s Castes and Tribes of Bengal, I. 304)? The Vāyu Purāṇa reads Taṅgaṇas hero (xlv. 135), after having named them previously (ibid., 120); see verse 41. The Matsya reads A-pathas (csiīi. 55), which seems erroneous.


The Khasas or Khaśas are generally mentioned as a half-civilized tribe outside India, along with Śakas, Daradas, &c. (Sabhā-P., li. 1859; Droṇa-P., xi. 399; and cxxi. 4846-7). They are said to have been defeated and degraded by Sagara (Hari-Vaṃśa, xiv. 784), and were considered mlecchas (id., xcv. 6440-1; see also Muir’s Sansk. Texts, II. 482). The Khasas in the text, however, may perhaps be identified with the Khas, who were formerly a small clan but have developed into the predominant military order of the kingdom of Nepal through intermarriages with brahmans (Journal, Bengal Asiatic Society, Vol. LXIII. Part I. 217-223). See canto lviii. verse 6.


I have not found this name elsewhere. The Vāyu Purāṇa reads Kuśa instead of Kunta (xlv. 136), which does not seem satisfactory. The proper reading should no doubt be Karṇa-prāvaraṇas, “those who cover themselves with their ears,” a people mentioned several times in the MahāBhārata (Sabhā-P., li. 1875; and Bhīṣma-P., li. 2103). They are placed in the South in the story of Sahadeva’s conquests (Sabhā-P., xxx. 1170) and among the Kirātas in the Eastern region in the Rāmāy. (Kiṣk.-K. xl. 29); but it seems permissible to identify them with the Ulūkas, for a story is told about an Ulūka named Prāvāra-karṇa (Vana-P, cxcviii. 13334). The Ulūkas dwelt in the Himalayas (ibid.), and formed a kingdom in the Pāṇḍavas’ time (Udyoga-P., clix. and clx), but it seems impossible to fix their position more definitely than somewhere in Nepal (Sabhā-P., xxvi. 1014-20). The word Karṇa-prāvaraṇa is also used as an adjective, for Hanūmān saw female Rākṣasas in Laṅkā “three-eared and pin-eared, long-eared, ear-less, and one-eyed and one-eared, and having their ears as a covering” (Rāmāy., Sund.-K., xvii. 24); and it was an ancient belief that there were people with immense ears which covered their bodies (Plin., iv. 13; and vii. 2; Mandeville’s Travels, chap. xix). The Matsya P. mentions Kutha-prāvaraṇas and Karṇa-prāvaraṇas (cxx. 56 and 58).


This people have been already mentioned in verse 42.


These appear to be the same as the Darvas mentioned in verse 42; see Cunningham’s Arch. Surv. Repts., II. 15; and XIV. 145.


This is not in the dictionary; but it seems to be the same as Sākṛd-grāhas or Sakṛd-grāhas, who are said to be a terrible mleccha tribe in the North (Bhīṣma-P., ix. 373). There appear to be no data to fix their position unless they may be connected with the Sakṛn-nandā, which seems to be a river in the east of Nepal (Vana-P., lxxxiv. 8137). The text might also be read “and the Kṛtrakas,” but I have found no such name elsewhere.


Or Traigartas. Trigarta was considered to be in the Northern region (Sabhā-P., xxvi. 1026) and also in the Western (id., xxxi. 1189). It is generally mentioned in connexion with the Sindhus, Madras and other Pañjab nations (id., li. 1870; Vana-P., celxīv. 15593-9; cclxx. 15743; Bhīṣma-P., xviii. 688; cxviii. 5485; cxx. 5649; Droṇa-P., vii. 183; and also Harivaṃśa, xci. 4965-70). It was near the Kurus, for the Pāṇḍavas when burnt out of Vāraṇāvata visited the Trigartas and other contiguous nations (Ādi-P., elvi. 6084-7); and also near Matsya and Śālva, for these two kingdoms had often raided into Trigarta (Virāṭa-P., xxx). Prom these indications it appears that Trigarta must have touched the Pañjab on the west, and the Kurus on the south-east, and been close to Matsya (see page 307 note *) and Śālva (see note to canto lviii. verse 6) on the south; hence it must have comprised the country from Amballa and Pattiala to the R. Bias, i.e., the Jalandhar doab and the country south-east of that. Cunningham includes Kangra also (Arch. Surv. Repts., IL 16; and XIV. 116 and 117; and Buddh. Cave Temples, p. 93). At the time of the great war Prasthala belonged to Trigarta (see note on page 321), and so brought the Trigarta territories close to Matsya and Śālva.


These people, no doubt, claimed to be the descendants of the ṛṣi Gālava (Hari-Vaṃśa, xxvii 1463-7; and xxxii. 1767-76), who was a famous son of Viśvā-mitra (Anuśās.-P., iv. 249-259; Udyoga-P., cv-cxviii; and see cantos XX and XXI above), or took their name from him; see similarly the Ātreyas and Bharadvājas of verse 39.


See note to verse 40.


These have been already mentioned in verse 41.




This implies that the Himavat range included also the Sulaiman Mountains along the west of the Pañjab. The simile must refer to a drawn bow, with the string angular in the middle.


Marutas tathā seems incorrect. Head instead martyatā tathā ?


Mṛga-paśv-apsaro-yonis. The meaning of “aquatic animal” is given to ap-sara but not to ap-saras in the dictionary.


For pra-yāti read pra-jātiḥ ?


For sadā eṣa read sadaivaiṣa ?


This seems rather meaningless. For sva-karma-khyāpanotsukaiḥ the MS. reads sva-karma-kṣapanonmukhaiḥ, “who are averse to diminishing the stock of their actions;” but kṣapanotsukaiḥ seems preferable, “who are eager to diminish the stock of their actions.”

Help me keep this site Ad-Free

For over a decade, this site has never bothered you with ads. I want to keep it that way. But I humbly request your help to keep doing what I do best: provide the world with unbiased truth, wisdom and knowledge.

Let's make the world a better place together!

Like what you read? Consider supporting this website: