What remained then on the Gold Coast were the Gha and Akvambu; and these were afterwards conquered by the Akkim Fantis, themselves eventually reduced by the Ashantis.

In no more than nine or ten villages, lying within nine or ten miles of Fort St. James and Christiansborg, was the Akra language spoken in the time of Protten (A.D. 1794), and of the Ghas thus speaking it each understood the Fanti.

This makes the Gha a decreasing, and, for practical purposes, an unimportant population. At the same time I should be glad to direct the attention of some investigator to their ethnology. Their exact relations to the Akvambu are uncertain. The only work known to me where specimens of the latter language are to be found is out of reach.[18]

Then as to the Adampi. Bowdich states that it radically differs from the Gha; the numerals, which agree, being borrowed from the one tongue into the other. But his collation rests on only seven words.

Again,—Adampi, Tembi, and Tambu are words so much alike as to pass for the same. Yet a Tembu vocabulary in the "Mithridates" differs from a Tambu one in the same work—

ENGLISH. TEMBU. TAMBU.

Sky so giom. Sun wis pum. Moon igodi horamb. Man naa nyummu. ... ibalu numero. Woman alo in. Head knynoo ii. Foot navorree nandi. One kuddum kaki. Two noalee ennu. Three nodoso ettee.

Again—the Tembu is related to the vocabulary of a language called Kouri, which the Tambu is not.

ENGLISH. TEMBU. KOURI.

Sun wis nosi. Man ibalu abalu. Woman alo alu. One kuddum kotum. Two noalee nalee. Three nodoso natisu.

Thirdly, the Tjemba of Balbi's "Atlas Ethnologique" is called Kassenti.

Lastly, the Gha, as far as very short comparison goes, is neither Tambu nor Tembu: nor yet Kouri—though it has a few resemblances to all.

The author of the paper alluded to above is the Rev. Mr. Hanson—himself a Gha by birth. It was laid before the British Association in 1849. Two points characterize the theory that it exhibits; but as the publication of the paper in extenso, is contemplated, I merely state what they are.

1. A remarkable number of customs common to the Jews and the Gha.

2. The probable origin of the latter population in some part of the interior of Africa, north of their present locality, and, perhaps, in the parts about Timbuktu.

The Quaquas.—I am not sure that this name is the best that can be given to the class in question. Hence, it is merely provisional. The language that is spoken by them is called the Avekvom. They constitute the chief population of the Ivory—just as the Krumen do that of the Grain and the Fantis that of the Gold—Coast. Apollonia is the English dependency where we find members of the Quaqua stock.

The Avekvom dialects of the Quaqua tribes seem to belong to a different tongue from that of the Krumen and Fantis; and I imagine that the three are mutually unintelligible. Still, it is difficult to predicate this from the mere inspection of vocabularies; the more so, as no language of the western coast of Africa is less known than the Avekvom—the only specimen of any length being one in the last number of the "Journal of the American Oriental Society." With numerous miscellaneous affinities, it is more Fanti and Grebo than aught else; and, perhaps, is transitional in character to those two languages.

At any rate it is no isolated tongue, as may be seen from the following table, where Yebu means the language of the Yarriba country, at the back of Dahomey, and Efik that of Old Calabar:—

ENGLISH. AVEKVOM. OTHER IBO-ASHANTI LANGUAGES.

Arm ebo ubok, Efik. Blood evie eyip, Efik; eye, Yebu. Bone ewi beu, Fanti. Box ebru branh, Grebo. Canoe edie tonh, Grebo. Chair fata bada, Grebo. Dark eshim esum, Fanti; ekim, Efik. Dog etye aja, ayga, Yebu. Door eshinavi usuny, Efik. Ear eshibe esoa, Fanti. Fire eya ija, Fanti. Fish etsi eja, eya, Fanti. Fowl esu suseo, Mandingo; edia, Yebu. Ground-nut ngeti nkatye, Fanti. Hair emu ihwi, Fanti. Honey ajo ewo, Fanti; oyi, Yebu. House eva ifi, Fanti; ufog, Efik. Moon efe habo, Grebo; ofiong, Efik. Mosquito efo obong, Fanti. Oil inyu ingo, Fanti. Rain efuzumo-sohn sanjio, Mandingo. Rainy season eshi ojo, rain, Yebu. Salt etsa ta, Grebo. Sand esian-na utan, Efik. Sea etyu idu, Grebo. Stone desi sia, shia, Grebo. Thread jesi gise, Grebo. Tooth enena nyeng, Mandingo; gne, Grebo. Water esonh nsu, Fanti. Wife emise muso, Mandingo; mbesia, Fanti. Cry yaru isu, Fanti. Give nae nye, Grebo; no, Efik. Go le olo, Yebu. Kill bai fa, Mandingo; pa, Yebu.

There has been war and displacement here as well as in the Gha country. In the seventeenth century the parts about Cape Apollonia were contended for by two tribes called the Issini (or Oshin) and the Ghiomo. The former gave way to the latter, and having retreated to the country of the Veteres, were joined by that tribe against the Esiep.

A Quaqua prayer is given in the "Mithridates." It is uttered every morning by the tribes on the Issini, after a previous ablution in that river—Anghiume mame maro, mame orie, mame shikke e okkori, mame akaka, mame frembi, mame anguan e awnsan—O Anghiume! give rice, give yams, give gold, give aigris, give slaves, give riches, give (to be) strong and swift.

What is here written about the ethnology of Apollonia is written doubtfully; since here, as at Acra, the simple ethnology of the pure and proper Fantis becomes complicated.

The Cape of Good Hope.—The aboriginal population of the Cape is divided between two great families:—

1. The Hottentot.

2. The Kaffre.

1. The Hottentots.—Of the two families this is the most western; it is the one which the colonists came first in contact with, and it is the one which has been most displaced by Europeans. The names of fourteen extinct tribes of Hottentots are known; of which it is only necessary to mention the Gunyeman and Sussaqua the nearest the Cape, and the Heykom, so far eastwards and northwards as Port Natal. The displacement of these last has not been effected by Europeans. African subdued African; and it was the Kaffres who did the work of conquest here.

Of the extant Hottentots, within the limits of the colony of the Cape, the most remote are the Gonaqua, on the head-waters of the Great Fish River; or rather on the water-shed between it and the Orange River. They are fast becoming either extinct, or amalgamated with the Kaffres; inasmuch as they are the Hottentots of the Amakosa frontier, and suffer, at least, as much from the Kaffres as from their white neighbours.

The Namaquas occupy the lower part of the Orange River, the Great and Little Namaqualand.

The Koranas.—This branch of the Hottentots has its locality on the middle part of the Gariep, with the Griquas to the north, the Bechuana Kaffres to the east, and the Saabs in the middle of them. Their number is, perhaps, 10,000. Their exact relation to the other Hottentots is uncertain. They are a better formed people than the Gonaqua and Namaqua, but whether they be the best samples of the Hottentot stock altogether is uncertain. Probably a tribe far up in the north-western parts of South Africa, and beyond Namaqualand, may dispute the honour with them. These are the Dammaras—themselves disputed Hottentots. Their country lies beyond the British colony, but it must be noticed for the sake of taking in all the branches of the stock in question. It is the tract between Benguela and Namaqualand, marked in the maps as sterile country; in the northern parts of which we sometimes find notices of a fierce nation called Jagas. Walvisch Bay lies in the middle of it. Now some writers make the Dammaras of this country Hottentot; others Kaffre; and that both rightly and wrongly. They are both—partly one, partly the other; since Dammara is a geographical term, and some of the tribes to which it applies are Kaffre, some Hottentot. The Dammaras of the plains, or the Cattle Dammaras are the former; the Dammaras[19] of the hills, the latter. Between the Dammara and the Korana a much nearer approach to Kaffre type is made than is usually supposed.

A branch of the Koranas—those of the valley of the Hartebeest River—deserves particular attention. They caution us against overvaluing differences; and Dr. Prichard has quoted the evidence of Mr. Thompson with this especial object. They are Koranas who have suffered in war, lost their cattle, and been partially expatriated by the more powerful sections of their stock. Hence, want and poverty have acted upon them; and the effect has been that they have become hunters instead of shepherds, have been reduced to a precarious subsistence, and as the consequence of altered circumstances, have receded from the level of the other Koranas, and approached that of the—

Saabs or Bushmen.—These belong to the parts between the Roggeveld and Orange River; parts which rival the sterile country of the map in barrenness. As is the country so are the inhabitants; starved, miserable hunters—hunters rather than shepherds or herdsmen.

The Lap is not more strongly contrasted with the Finlander, than the Korana with the Saab; and the deadly enmity between these two populations is as marked as the differences in their physical appearances. I think, however, that undue inferences have been drawn from the difference; in other words, that the distance between the Korana and the Saab has been exaggerated. The languages are unequivocally allied.

I think, too, that a similarly undue inference has been drawn from the extent to which the Kaffre and the Korana are alike; inasmuch as an infusion of Kaffre has been assumed for the sake of accounting for it. Of this, however, no proof exists.

The Saabs are described as having constitutions "so much enfeebled by the dissolute life they lead, and the constant smoking of dacha, that nearly all, including the young people, look old and wrinkled; nevertheless, they are remarkable for vanity, and decorate their ears, legs, and arms with beads, and iron, copper, or brass rings. The women likewise stain their faces red, or paint them, either wholly or in part. Their clothing consists of a few sheepskins, which hang about their bodies, and thus form the mantle or covering, commonly called a kaross. This is their only clothing by day or night. The men wear old hats, which they obtain from the farmers, or else caps of their own manufacture. The women wear caps of skins, which they stiffen and finish with a high peak, and adorn with beads and metal rings. The dwelling of the Bushman is either a low wretched hut, or a circular cavity, on the open plain, into which, at night, he creeps with his wife and children, and which, though it shelters him from the wind, leaves him exposed to the rain. In this neighbourhood, in which rocks abound, they had formerly their habitations in them, as is proved by the many rude figures of oxen, horses, serpents, &c. still existing. It is not a little interesting to see these poor degraded people, who formerly were considered and treated as little better than wild beasts in their rocky retreats. Many of those who have forsaken us live in such cavities not far from our settlement, and we have thus an opportunity of observing them in their natural condition. Several who, when they came to us from the farmers, were decently clothed and possessed a flock of sheep, which they had earned, in a short time returned to their fastnesses in a state of nakedness and indigence, rejoicing that they had got free from the farmers, and could live as they pleased in the indulgence of their sensual appetites. Such fugitives from civilised life, I have never seen otherwise occupied than with their bows and arrows. The bows are small, but made of good elastic wood; the arrows are formed of small reeds, the points furnished with a well-wrought piece of bone, and a double barb, which is steeped in a potent poison of a resiny appearance. This poison is distilled from the leaves of an indigenous tree. Many prefer these arrows to fire-arms, under the idea that they can kill more game by means of a weapon that makes no report. On their return from the chase, they feast till they are tired and drowsy, and hunger alone rouses them to renewed exertion. In seasons of scarcity they devour all kinds of wild roots, ants, ants' eggs, locusts, snakes, and even roasted skins. Three women of this singular tribe were not long since met with, several days' journey from this place, who had forsaken their husbands, and lived very contentedly on wild honey and locusts. As enemies, the Bushmen are not to be despised. They are adepts in stealing cattle and sheep; and the wounds they inflict when pursued, are ordinarily fatal if the wounded part is not immediately cut out. The animals they are unable to carry off, they kill or mutilate.

"To our great comfort, even some of these poor outcasts have shown eagerness to become acquainted with the way of salvation. The children of such as are inhabitants of the settlement, attend the school diligently, and of them we have the best hopes.

"The language of the Bushman has not one pleasing feature; it seems to consist of a collection of snapping, hissing, grunting, sounds; all more or less nasal. Of their religious creed it is difficult to obtain any information; as far as I have been able to learn, they have a name for the Supreme Being; and the Kaffre word tixo is derived from the tixme of the Bushmen. Sorcerers exist among them. One of the Bushmen residing here being sick, a sorceress was sent for before we were aware of it, who pretended, by the virtue of mystic dance, to extract an antelope horn from the head of the patient."[20]

The Griquas.—The Griquas, called also Baastaards, are a pastoral population, upwards of 15,000 in number, on the north side of the great bend of the Orange River. They are the descendants of Dutch fathers and Hottentot mothers.

A mixture of Griquas and Hottentots occurs also on the Kat River, a feeder of the Great Fish River, in the district of Somerset, and on the Kaffre frontier. Here they are distributed in a series of district locations, amid the dales and fastnesses of the eastern frontier. A great proportion of them are discharged soldiers—so that in reality, like the borderers of old, they form a sort of military colony.

2. The Kaffres.—The British districts in contact with the Kaffre populations are the eastern, and of these Albany and Somerset most especially. The Kaffre nation in most immediate contact with Albany and Somerset is—

The Amakosa.—This is the population which constituted the authority of Hintza, and to which Pato, Gaika, and the other chiefs of the last war belonged. To this, too, belong the troublesome chiefs of the present. Next to the Amakosa, and in alliance with them, come—

The Amatembu, or Tambuki (Tambookies), occupants of the upper part of the river Kei, as the Amakosa are of the lower Keiskamma.

Between the Amatembu and Port Natal lie the Amaponda, or Mambuki (Mambookies), the northern extremity of which reaches the country of—

The Amazulu, or Zulu (Zooloos), the chief frontagers (conjointly with the Mambuki) of Port Natal.

The last division of the Kaffres of the coast is that of—

The Fingos.—In 1835, a numerous population, called Fingos, was found by Sir B. D'Urban in the Kaffre chief Hintza's country, and in a state of abject servitude to the Amakosas. They were from different tribes; darker and shorter than the Amakosas—but still true Kaffres. They were offered land between the lower Keiskamma and the Great Fish River, and were emancipated and brought safe into the colony to the amount of 17,000.[21] Since then, they have served as a sort of military police on the Kaffre frontier; and as shepherds in Australia—whither they have been advantageously introduced.

But, besides the Kaffres of the coast there are those of the interior. These speak a modified form of the Kosa (or Amakosa), called Si-chuana, the name of the people being Bi-chuana. They lie due north of the Koranas; beyond the boundaries of the colony; but not beyond the influence of its missionaries, or the range of its explorers. Litaku, Kurrichani, and other similar towns are Sichuana; the Kaffre civilization being said to attain its maximum hereabouts.

There are plenty of points of contrast between the Kaffre and the typical Negro; so many indeed as to have suggested the doctrine that the former class belongs to some division of the human species other than the African. And these points of contrast are widely distributed, i.e., they appear and re-appear, whatever may be the view taken of the Kaffre stock. They appear in the descriptions of their skin and skeletons; they appear in the notice of their language; and they appear in the history of the Kaffre wars of the Cape frontier—wars more obstinate and troublesome than any which have been conducted by the true Negro; and which approach the character of the Kabyle struggle for independence in Algeria. In investigating these differences we must guard against the exaggeration of their import.

Physically, the Kaffre has the advantage of the Negro in the conformation of the face and skull. His forehead betokens greater capacity; being more prominent, more vaulted, and with a greater facial angle. His teeth, too, are more vertically inserted, and the nasal bones less depressed. I have not heard of aquiline noses in Kaffraria; but should not be surprised if I did.

The cheek-bones of the Kaffre project outwards; and where the cheek-bones so project beyond a certain limit, the chin appears to taper downwards, and the vertex upwards. When this becomes exaggerated we hear of lozenge-shaped crania; the Malay skulls being currently quoted as instances thereof. Be this as it may, the breadth in the malar portion of the face is a remarkable feature in the Kaffre physiognomy. This he has in common with the Hottentot. His hair is also tufted like the Hottentot's: while his lips are thick like the Negro's. Tall in stature, wiry and elastic in his muscles, the Kaffre varies in colour, through all the shades of black and brown; being, in some portions of his area nearly as dark as the Negro, in others simply brown like the Arab. The eye is sometimes oblique; the opening generally narrow.

An opinion often gives a better picture than a description. Kaffres, that have receded in the greatest degree from the Negro type, have been so likened to the more southern Arabs as to have engendered the hypothesis of an infusion of Arab blood.

The manners of the Kaffres of the Cape are those of pastoral tribes under chieftains; tribes which, from their habits and social relations, are naturally active, locomotive, warlike, and jealous of encroachment. Next to marauding on the hunting-grounds of an American Indian, interference with the pasture of a shepherd population is the surest way to warfare.

It would be strange indeed if the Kaffre life and Kaffre physiognomy had no peculiarities. However little in the way of physical influence we may attribute to the geography of a country, no man ignores them altogether. Now Kaffreland has very nearly a latitude of its own; inhabited lands similarly related to the southern tropic being found in South America and Australia only. And it has a soil still more exclusively South-African. We connect the idea of the desert with that of sand; whilst steppe is a term which is limited to the vast tracts of central Asia. Now the Kaffre, and still more the Hottentot, area, dry like the desert, and elevated like the steppe, is partially a karro. Its soil is often a hard, cracked, and parched clay rather than a waste of sand, and it constitutes an argillaceous table-land. Its vegetation has strongly marked characters. Its Fauna has the same.

The language is peculiar. If English were spoken on Kosa or Sichuana principles we should say

bun beam instead of sun beam. loon light ... moon light. srand-son ... grand-son, &c.,

since, in the Kaffre languages throughout, subordinate words in certain syntactic combinations, accommodate their initial letter to that of the leading word of the term.

Their polity and manners, too, are peculiar. The head man of the village settles disputes; his tribunal being in the open air. From him an appeal lies to a chief of higher power; and from him to some superior, higher still. In this way there is a long chain of feudal or semi-feudal dependency.

But the power of the chief is checked by that of the priest. A supposed skill in medicine, imaginary arts of divination, and an accredited power over the elements are the prerogatives of certain witches and wizards. Thus, when a murrain among the cattle, or the death of an important individual has taken place, the blame is laid upon some unfortunate victim whom the witch or wizard points out. And the ordeal to which he must submit, is equal in cruelty to those of the Gold Coast. He is beaten with sticks, and then pegged down to the ground. Whilst thus helpless, a nest of venomous bush-ants is broken over his racked and quivering body. If this fail to extort a confession, he is singed to death with red-hot stones.

This tells us what is meant by Kaffre chiefs and Kaffre wizards.

The wife is the slave to the husband; and he buys her in order that she should be so. The purchase implies a seller. This is always a member of another tribe. Hence the wish of a Kaffre is to see his wife the mother of many children, girls being more valuable than boys.

Why a man should not sell his offspring to the members of his own tribe is uncertain. It is clear, however, that the practice of doing so makes marriage between even distant relations next to impossible. To guard against the chances of this, a rigid and suspicious system of restraint has been developed in cases of consanguinity; and relations must do all they can to avoid meeting. To sit in the same room, to meet on the same road, is undesirable. To converse is but just allowable, and then all who choose must hear what is said. So thorough, however, has been the isolation in many cases, that persons of different sexes have lived as near neighbours for many years without having conversed with each other; and such communication as there has been, has taken place through the medium of a third person. No gift will induce a Kaffre female to violate this law.

Is the immolation of human beings at the death of chieftains a Kaffre custom, as it was one of western Africa? The following extract gives an answer in the affirmative, the only difference being the pretext of the murders. On the "death of the mother of Chaka, the great Zulu chief, a public mourning was held, which lasted for the space of two days, the people being assembled at the kraal of the chief to the number of sixty or eighty thousand souls. Mr. Fynn, who was present, describes the scene as the most terrific which it is possible for the human mind to conceive. The immense multitude were all engaged in rending the air with the most doleful shrieks, and discordant cries and lamentations; whilst, in the event of their ceasing to utter them, they were instantly butchered as guilty of a crime against the reigning tyrant. It is said that no less than six or seven thousand persons were destroyed on this occasion, charged with no other offence than exhausted nature in the performance of this horrid rite, their brains being mercilessly dashed out amidst the surrounding throng. As a suitable finale to this dreadful tragedy, it is said that ten females were actually buried alive with the royal corpse; whilst all who witnessed the funeral were obliged to remain on the spot for a whole year."

Details of Kaffre manners may be multiplied almost ad infinitum; and as their history and habits are likely to fill a Blue Book, a short treatise can only notice their more prominent peculiarities.

However, lest an undue inference be drawn from their contrast to the Hottentot, we must remember that the former has encroached upon the latter, and that such transitional populations as existed have been swept away.

Now comes a coloured population—not indigenous, but the descendants of the slaves of the colony. This consists of—

1. Negroes.

2. Malays from the Indian Archipelago.

3. Malagasi from Madagascar.

To which we must add, as of mixed blood, the offspring of—

1. Negroes and Dutch, English, &c.

2. Malays and Dutch, English, &c.

3. Malagasi and Dutch, English, &c.

This seems to be the limit of the intermixture; since, between the Malays and Negroes, &c., there is but little intermarriage. The possible elements, however, of hybridity are numerous, e.g., Griquas and Negroes, Griquas and Malays, Malays and Kaffres, &c.

The so-called yellow men.—On the 4th of August, 1782, the "Grosvenor" Indiaman was wrecked on the coast of Natal. Of the crew who escaped, some reached the Cape and others remained amongst the natives. In 1790, an expedition was undertaken in search of them.

In this expedition, Mr. Van Reenens, considered that he had discovered a village where the people were descended from the whites, and in which there were three old women who had been wrecked when very young. They could not tell to what country they belonged; were treated as superior beings; and, when offered a safe convoy to the Cape, were at first pleased with the prospect, but eventually refused to leave their children and grandchildren. Now, whatever these old women were, they were not of the crew of the "Grosvenor," and I doubt whether they were Europeans at all.

Again—Mr. Thomson, when at Litaku, heard of yellow cannibals, with long hair, whose invasions were the dread of the country; a statement which merely means that some tribes of South Africa, are lighter coloured, and more savage in their appetite than others.

Lastly, Lieutenant Farewell saw one of these yellow men at Natal, who was described as a cannibal, and who shrunk abashed from the lieutenant.

Be it so. The evidence that "there are descendants of Europeans and Africans now widely diffusing their offspring throughout the country; whose services might be turned to good account in civilizing the native tribes," is still incomplete.

Mauritius.—The coloured population, which is far greater than that of the white, consists in the Mauritius of—

1. True Africans—chiefly from the east coast, and, consequently, of the Kaffre stock; the word being used in its most general sense. Darker than the Kaffres of the Cape, they, nevertheless, recede from the Negro type in the shape of the jaw, lips, and forehead. The hair also is less woolly. They are strong and powerful individuals.

2. Malagasi, or natives of Madagascar.—These are not Africans to the same extent as the Kaffres of the coast. As far back as the time of Reland it was known that the affinities of the Malagasi language were with the Malay and Polynesian tongues of Asia; but it was also known that the similarity in physiognomy was less than that of language. Hence came a conflict of difficulties. The speech indicated one origin, the colour another—whilst the fact of an island so near to Africa, and so far from Malacca, as Madagascar, being other than what its geographical position indicated, is, and has been, a mystery. Some writers have assumed an intermixture of blood; others have limited the Malay element to the dominant population. Lastly, Mr. Crawfurd has denied the inferences from the similarity of language in toto; considering that there is "nothing in common between the two races, and nothing in common between the character of their languages." The comparative philologist is slow to admit this—indeed, he denies it.

The blacks form the great majority of the coloured population. Besides these, however, there are—

3. Arabs.

4. Chinese.

5. Hindus, from the continent of India; convicts being transported to the Mauritius for life, and worked on the roads of the colony.

6. Cingalese from Ceylon—the Kandian chiefs whose presence in their native country was thought likely to endanger the tranquillity of the island, were sent hither.

The whites of the Mauritius are chiefly French; though not wholly of pure blood. The first settlers took their wives from Madagascar. The English form the smallest part of the population.

Rodrigues—occupied by a few French colonists from the Mauritius.

The Seychelles—The same; the coloured population outnumbering the white in the proportion of ten to one. Here there is a Portuguese admixture. From Maha, the chief town of the Seychelles, to Madagascar, is five hundred and seventy-six miles—a fact to be borne in mind when we speculate upon the origin of the population of that island.

* * * * *

The Africans of British America.—Honduras, Belize, the West India Islands, and Demerara.—The usual distribution of the population of these parts is—

WHITE.

1. European whites, born in Europe. 2. Creoles, or whites born in the island.

COLOURED.

a. Pure Blood.

1. Mandingos, from the river-systems of the Senegal and Gambia. 2. Coromantines—from the Ivory and Gold Coast. 3. Whydahs—from Dahomey. 4. Ibos—from the Lower Niger. 5. Congos—from Portuguese Africa.

b. Mixed Blood.

1. Sambos, intermixture of the Negro and Mulatto. 2. Mulattoes—Negro and white. 3. Quadroons—Mulatto and white. 4. Mestis—Quadroon and white.

Such is what I find in Mr. Martin's valuable work on the Colonies, and it is, undoubtedly, a convenient and practical classification. Yet for the purposes of ethnology, it is deficient in detail. Without even guessing at the proportion of American slaves which the different parts of the western coast of Africa may have supplied, I subjoin a brief notice of tract between the Senegal and Benguela.

1. First come the Wolof, between the Senegal and Cape Verde. To the back of these lie—

2. The Serawolli—and around Cape Verde—

3. The Sereres—none of these are truly Mandingo; nor is it certain that many slaves have come from them; such as do, however, are probably Mandingos in the current classification.

4. The Fulahs of Fouta-Torro and Fouta-Jallo possess the higher part of the Senegambian system. Imperfect Mahometans, they are lighter-coloured than either the Wolof or the Mandingo. Notwithstanding the great Fulah conquests—for under a leader named Danfodio this has been one of the encroaching and subjugating families of Africa—there are still American slaves of Fulah blood—though, perhaps, but few. Mr. Hodgson procured his vocabulary from a Fulah slave of Virginia; and what we find in the United States, we may find in the British possessions also.

5. The Mandingos Proper are the Negroes of the Gambia; but the following Africans, all within the range of the old slave trade, belong to the same class.

a. The Susu; whose language is spoken from the River Pongos to Sierra Leone.

b. The Timmani.

c. The Bullom—each in contact with that settlement.

d. The Vey—the written language already noticed.

e. The Mendi—conterminous with the Vey.

f. The Kissi—like the last two, spoken in the country behind Cape Mount, and on the boundaries of Liberia.

South of the Gambia and north of the Pongos, the Mandingo tongues, though spoken in the interior, do not reach the coast. On the contrary, they encircle the populations on the mouths of the Cacheo, Rio Grande, and Nun—and truly barbarous populations these are. Of these the most northern are—

6. The Felup (Feloops)—between the Gambia and Cacheo.

7. The Papel—south of the Cacheo.

8. The Balantes—south of the Papel.

9. The Bagnon—on the Lower Cacheo.

10. The Bissago—islanders off the Cacheo.

11. Nalu (Naloos)—on the Lower Nun.

12. Sapi—ibid.

After these come the Susu, &c.; down to the tribes about Cape Mount and Cape Mesurado.

Between Cape Mesurado and Cape Palmas come—

13. The Krumen. Next to them—

14. The Quaquas, of the Ivory Coast; speaking different Avekvom dialects.

Somewhere hereabouts come the—

15, 16, 17. Kanga, Mangree, and Gien; three undetermined vocabularies of the "Mithridates." Then—

18, 19, 20. The Fanti, Gha, and Adampi (?) of the Gold Coast. We now approach the great marts—

21, 22. Benin and Dahomey; and—almost equal in infamous notoriety—the countries of the Delta, of the Niger, or of the—

23, 24, 25. Ibu, Bonny, and Efik (Old Calabar) Africans; at the back of which lie—

26, 27. Yarriba, and the Nufi country. In Fernando Po the population is—

28. Ediya. About the Bimbia river and mountain—

29. Isubu.

30, 31, 32. The Banaka (or Batanga), the Panwi, and the Mpoongwe take us from the Gaboon to Loango; forming a transition from the true Negroes to the Kaffres.

33, 34, 35, 36. Loango, Congo, Angola, and Benguela—the Kaffre type, both in form and language, is now more closely approached. Below Benguela there has been little or no exportation.

FOOTNOTES:

[12] "Journal of the Geographical Society," 1850.

[13] "United Service Magazine," Dec., 1850.

[14] "United Service Journal," Nov., 1850.

[15] Daniell in "Transactions of the Ethnological Society."

[16] "United Service Journal," Nov., 1850.

[17] Dr. Daniell on the Natives of Old Calabar, "Transactions of the Ethnological Society."

[18] Rask.—Vejledning tel Acra-sproget, paa Kysten Ginea, med et Tillaeg om Akvambuisk.—Copenhagen, 1828. Introduction to the Acra Language, on the Coast of Guinea, with an Appendix on the Akvambu.

[19] "Journal of the American Oriental Society," vol. i. no. 4.

[20] "British Colonies." By M. Martin.

[21] "Journal of the Geographical Society," vol. v. p. 319.



CHAPTER III.

BRITISH COLONIES AND DEPENDENCIES IN ASIA.

ADEN.—THE MONGOLIAN VARIETY.—THE MONOSYLLABIC LANGUAGES.—HONG KONG.—THE TENASSERIM PROVINCES; MAULMEIN, YE, TAVOY, TENASSERIM, THE MERGUI ARCHIPELAGO.—THE MON, SIAMESE, AVANS, KARIENS, AND SILONG.—ARAKHAN.—MUGS, KHYENS.—CHITTAGONG, TIPPERA, AND SYLHET.—KUKI.—KASIA.—CACHARS.—ASSAM.—NAGAS.—SINGPHO.—JILI.— KHAMTI.—MISHIMI.—ABORS AND BOR-ABORS.—DUFLA.—AKA.—MUTTUCKS AND MIRI, AND OTHER TRIBES OF THE VALLEY OF ASSAM.—THE GARO.— CLASSIFICATION.—MR. BROWN'S TABLES.—THE BODO.—DHIMAL.—KOCCH.— LEPCHAS OF SIKKIM.—RAWAT OF KUMAON.—POLYANDRIA.—THE TAMULIAN POPULATIONS.—RAJMAHALI MOUNTAINEERS.—KULIS, KHONDS, GOANDS, CHENCHWARS.—TUDAS, ETC.—BHILS.—WARALIS.—THE TAMUL, TELINGA, KANARA AND MALAYALAM LANGUAGES.

Aden.—The ethnology of the Arab stock would fill a volume. It is sufficient to state that the British political dependency of Aden is, ethnologically, an Arab town.

Far more important possessions direct our attention towards India. Nevertheless, there are certain preliminaries to its ethnology.

Mongolia and China—each of these countries illustrates an important ethnological phenomenon.

The first is a physical one. Cheek-bones that project outwards, a broad and flat face, a depressed nose, an oblique eye, a somewhat slanting insertion of the teeth, a scanty beard, an undersized frame, and a tawny or yellow skin, characterize the Mongol of Mongolia.

The second is a philological one. A comparative absence of grammatical inflexions, and a disproportionate preponderance of monosyllabic words, characterize the language of China.

So much for the simple elementary facts; the former of which will be spoken of under the designation of Mongolian conformation; the second under that of monosyllabic language.

Neither term is limited to the nation by which it has been illustrated. Plenty of populations besides those of Mongolia Proper are Mongol in physiognomy. Plenty of nations besides the Chinese are monosyllabic in language.

All the nations speaking monosyllabic tongues are Mongol in physiognomy; though all the nations which have a Mongol physiognomy do not speak monosyllabic tongues. This makes the latter group, which for shortness will be called that of the monosyllabic nations or tribes—a section, or division, of the former.

Little Tibet, Ladakh, Tibet Proper, Butan, and China, are all Mongol in form, and monosyllabic in language. So are Ava, Pegu, Siam, Cambojia, and Cochin China, the countries which constitute the great peninsula, sometimes called Indo-Chinese, and sometimes Transgangetic.

The extremity however—the Malayan peninsula—is not monosyllabic.

The British possessions of Hindostan are monosyllabic on their Tibetan and Burmese frontiers.

Hong-Kong.—Aden was disposed of briefly. So is Hong-Kong; and that for the same reason. Politically, British, it is ethnologically Chinese.

Maulmein, Ye, Tavoy, Tenasserim, and the Mergui Archipelago.—These constitute what are sometimes called the ceded, sometimes the Tenasserim provinces. They came into possession of the British at the close of the Burmese war of 1825. Unlike our dependencies in Hindostan, they are cut off from connection with any of the great centres of British power in Asia—in which respect they agree with the smaller and still more isolated settlements of the Malaccan Peninsula. The power that ceded them was the Burmese, so that it is with the existing subjects of that empire that their present limits are in contact; though only for the northern part. To the south they abut upon Siam.

The population throughout is monosyllabic; except so far as it is modified by foreign intermixture—of which by far the most important element is the Indian. Everything in the way of religious creed which is not native and pagan is Indian and Buddhist. The alphabets, too, of the lettered populations are Indian in origin.

The population of the continental part of these British dependencies is referable to four divisions—of unequal and imperfectly ascertained value. 1. The Mon. 2. The Siamese. 3. The Avans. 4. The Kariens.

1. The Mon.—Mon is the native name of the indigenous population of Pegu, so that the Mon of Maulmein, or Amherst, the most northern of the provinces in question, on the left bank of the lower Salwin, are part and parcel of the present occupants of the delta of the Irawaddi, and the country about Cape Negrais. The Burmese call them Talieng, and under that designation they are described in Dr. Helfer's Report.[22] The Siamese appellation is Ming-mon; apparently the native name in a state of composition. In the early Portuguese notices a still more composite form appears—and we hear of the ancient empire of Kalamenham, supposed to have been founded by the Pandalus of Mon or Pegu.

None of the lettered languages of the Indo-Chinese peninsula are less known than that of Pegu. At the same time its unequivocally monosyllabic character is beyond doubt. The alphabet is a slight variation of the Avan.

The geographical position of the Mon at the extremity of a promontory, and on the delta of a river, taken along with their philological isolation, is remarkable. They have evidently been encroached upon by the Avans in latter times; whilst, at an earlier period, they themselves probably encroached upon others. Whether they are the oldest occupants of Maulmein is uncertain; it is only certain that they are older than their conquerors.

To the Mon of Pegu the exchange of Avan for British rule, has been a great and an appreciated advantage.

2. The Siamese.—The native name for the Siamese language is Tha'y, and Tha'y is the national and indigenous denomination of the Siamese. It is the Avans who call them Sian or Shan; from whence the European term has been derived through the Portuguese.

The Siamese population is of course greatest on the Siamese frontier; so that, increasing as we go south, it attains its maximum in Tenasserim just as the Mon did in Maulmein. It seems, also, to have been introduced at different times; a fact which gives us a distinction between the native Siamese and the recent settlers.

Like the Mon, the Tha'y, at least in its more classical dialect, is a lettered language; the alphabet, like the Buddhist religion, being Indian. Unlike, however, the Mon, which is the only representative of the family to which it belongs, the Tha'y tribes constitute a vast class, falling into divisions and subdivisions, and exceedingly remarkable in respect to its geographical distribution.

The Siamese of Siam, the kingdom of which Bankok is the capital, form but a fraction of this great stock. The upper half of the river Menam is occupied by what are called the Lau, or Laos. These are partly wholly independent, and partly in nominal dependence upon China; and proportionate to their independence is the unlettered character of their language, and the absence of Indian influences. Nor is this all. The Menam is pre-eminently the river of the Tha'y stock, and along the water-system of the Menam its chief branches are to be found; their position being between the Burmese populations of the west, and the Khomen of Cambojia on the east. This distribution is vertical, i.e., it is characterized by its length, rather than its breadth, and runs from south to north. So far does it reach in this direction that, as high as 28 deg. North lat., in upper Assam we find a branch of it. This is the Khamti. In a valuable comparison of languages, well-known as "Brown's Tables,"[23] the proportion of the Khamti words to the South Siamese is ninety-two per cent.

Of the physical appearance of the Siamese, we find the best account in "Crawfurd's Embassy," the classical work for the ethnology of the southern part of the Indo-Chinese peninsula. Their stature is low; the tallest man out of twenty having been five feet eight inches, the shortest five feet three. The complexion, darker than that of the Chinese, is lighter than that of the Malay; the eye oblique; the jaw square; and the cheek-bones broad.

Tha'y is an ethnological term, and denotes all the nations and tribes akin to the Siamese of the southern, the Khamti of the northern, or the Lau of the intermediate area. The difference between the first and the last of these three should be noticed. Some members of the family are Indianized in religion, and organized in politics. Such are the Siamese of Bankok. Others retain both their independence and their original Paganism. Such are some of the Lau. Mutatis mutandis, the same applies to the next family.

This is the Burmese, to which both the Avans and the Kariens belong; but as it has been already stated that the divisions under consideration are by no means of equal value, the two branches will be considered separately.

3. The Avans.—Avan is a more convenient term than Burmese, inasmuch as it is more definite; the Burmese Empire containing not only very distant members of the great Burmese family, but also populations which belong to other groups. Ava, on the other hand, is the centre of the dominant division.

Whether the Mon, or a family yet to be mentioned, represent the aborigines of Maulmein, it is certain that the Avans of that country are of comparatively recent introduction.

Again, whether the Tha'y, or a family yet to be mentioned, represent the aborigines of Tenasserim, it is certain that the Avans of that country are of comparatively recent origin.

Nevertheless, there are Avans in each; and in Maulmein, although the Mon preponderate in number, they all are able to speak the language of their conquerors. I say conquerors, because the Avans are for all the parts south of 18 deg. North lat., an intrusive population: the end of the eighteenth century being the date, when, under Alompra, an Avan or Umerapura dynasty broke up and subjected, in different degrees, the Mon and Tha'y populations to the south, as well as several others more akin to itself on the east, west, and north.

The kingdom of Ava, next to those of China and Siam, best represents the civilization of those families whose tongue is monosyllabic. This implies that it has an organized polity, a lettered language, and a Buddhist creed; in other words that the influences of either China or India have acted on it. Of these two nations it is the latter which has most modified the Indianized members of the great Burmese stock. In strong contrast with these is the fourth and last branch of the continental population for the provinces in question, the

4. Karien.—The Kariens are partially independent; chiefly pagan; and their language, belonging to the same class with the Avan, is unlettered. They are the first of a long list.

Their geographical distribution is remarkable, like that of the Tha'y. Its direction is north and south; its dimensions linear, rather than broad; and it bears nearly the same relation to the water-system of the Salwin that that of the Siamese does to the river Menam. There are Kariens as far south as 11 deg. North lat. and there are Kariens as far north as 25 deg. North lat. Hence we have them in Maulmein, and in Tenasserim, and in the intermediate provinces of Ye and Tavoy as well. All these, like the Mon, have been eased by the transfer from Avan oppression to British rule; though this says but little. Hence, with one exception, the other members of their family are decreasing; the exception being the so-called Red Karien.

This epithet indicates a change in physiognomy; and, indeed, the physical conformation of the Burmese tribes requires attention. It is Mongolian in the way that the Siamese is Mongolian; but changes have set in. The beard increases; the hair becomes crisper; and the complexion darkens. The Kyo,[24] the isolated occupants of a single village on the river Koladyng, are so much darker than their neighbours as to have been considered half Bengali; and, as a general rule, the nearer we approach India, the deeper becomes the complexion. The Mon, too, of Pegu, are very dark. What is this the effect of? Certainly not of latitude, since we are moving northward. Of intermarriage? There is no proof of this. The greater amount of low alluvial soils, like those of the Ganges and Irawaddi, is, in my mind, the truer reason. But this is too general a question to be allowed to delay us. The Red Kariens are instances of an Asiatic tribe with an American colour; just as the Red Fulahs were in Africa. Such are the occupants of the continent.

5. The Silong.—In the islands of the Mergui Archipelago, there is another variety; but whether it form a class itself, or belong to any of the previous ones, is uncertain. Their language is said to be peculiar;[25] but of this we have no specimen. As it is probably that of the oldest inhabitants of the continent opposite, this is to be regretted.

They are called Silong, are a sort of sea-gipsy; and amount to about one thousand. Of all the creeds of either India or the Indo-Chinese peninsula theirs is the most primitive; so primitive as to be characterized by little except its negative characters. They believe that the land, air, trees, and waters are inhabited by Nat, or spirits, who direct the phenomena of Nature. How far they affect that of man, except indirectly, is unascertained. "We do not think about that," was the invariable answer, when any one was questioned about a future state. Too vague for monotheism, the Silong creed is also said to be too vague for idolatry, too vague for sacrifices.

The Kariens, also, believe in Nat, but, as they believe in their influence on human affairs, they sacrifice to them accordingly.

Little, then, as we know, respecting these two families, we know that the common practice of Nat worship connects them; and this worship connects many other members of the Burmese stock. Consequently it helps us to place the Silong in that group. It also favours the notion of the Tenasserim aborigines being Burmese.

It is the delta of the Irawaddi which isolates the Tenasserim provinces; and the British dependency from which it separates them is—

Arakhan.—We are prepared for the ethnological position of the Arakhan populations. They are Burmese.

We are likewise prepared for a division of them; there will be the Indianized and the Pagan—paganism and political independence going, to a certain degree, together.

We are prepared for even minuter detail; the paganism will be Nat-worship; the Indian creed Buddhism: the alphabet also, where the language is written, will be Indian also. In Captain Tower's vocabulary,[26] only seven words out of fifty differ between the Burmese of Arakhan, and the Burmese of Ava; and some of these are mere differences of pronunciation.

The language itself is called Rukheng by those who use it; but the Bengali name is Mug.

This applies to the Indianized part of the population, the analogues of the Avans and Siamese of Tenasserim, and of the Mon of Maulmein. What are the Arakhan equivalents to the Karien?

The Khyen.—These inhabit the Yuma mountains between Arakhan and Ava. A full notice of them is given by Lieutenant Trant, in the sixteenth volume of the "Asiatic Researches." But as they are chiefly independent tribes, it is enough to state that they form the Anglo-Burmese frontier. It is also added that there are numerous Khyen slaves in Arakhan.

Farther notice of them is the less important, because a closely allied population will occur amongst the hill-tribes of—

Chittagong.—Hindu elements now increase. Even in Arakhan, Buddhism had ceased to be the only creed of western origin. There were Mahometans who spoke a mixed dialect called the Ruinga;[27] and Brahminical Hindus who spoke another called the Rosawn. In Chittagong, then, we must look about us for the aborigines; so intrusive have become the Hindu elements. Intrusive, however, they are, and intrusive they will be for some time to come.

The foot of the hill, and the hill itself, are important points of difference in Indian ethnology. On the lower ranges of the mountains on the north-east of Chittagong are the Khumia (Choomeeas) or villagers; khum (choom) meaning village. These are definitely distinguished from the Hindus, by a flat nose, small eye, and broad round face, in other words by Mongolian characteristics in the way of physiognomy. But the Khumia are less perfect samples of their class than the true mountaineers. These are the Kuki,[28]—hunters and warriors, divided into tribes, each under elective chiefs, themselves subordinate to a hereditary Raja,—at least such is the Hindu phraseology.

Their creed consists in the belief of Khogein Pootteeang as a superior, and Sheem Sauk as an inferior deity; the destruction of numerous enemies being the best recommendation to their favour. A wooden figure, of human shape, represents the latter. The skulls of their enemies they keep as trophies. In the month of January there is a solemn festival.

Language and tradition alike tell us that the Kuki (and most likely the Khumia as well) are unmodified Mugs. The displacement of their family has been twofold—first by Hindus, secondly by Buddhist (or modified) Mugs at the time of the Burmese conquest. The Kuki population extends to the wilder parts of the district of Tippera.

Sylhet.—On the southern frontier we have Kukis; on the eastern Cachari; on the northern Coosyas (Kasia). Due west of these last lie the Garo. I imagine that both these last-named populations are members of the same group—but cannot speak confidently. If so, we have departed considerably from the more typical Burmese of Arakhan and Ava. Still we are within the same great class. The Garo will command a somewhat full notice.

The Cachars depart still more from the more typical Burmese; the group to which they most closely belong being one which will also be enlarged on.

North of the Kasia we reach the western portion of the southern frontier of—

Assam.—Here it will be convenient to take the whole of the valley—Upper as well as Middle and Lower Assam—although parts of the former are independent rather than British—and to go round it; beginning with the Kasia country and the Jaintia mountains on the south-west. I imagine—but am not certain—that the Kasia and Jaintia mountaineers are very closely allied.

Next to the Cachars on the southern, or Manipur, frontier are—

The Nagas.—These are in the same class with the Kuki; i.e., the wild tribes of Manipur, speaking a not very altered dialect of the Burmese.

The Singpho.—This people is said to have come from a locality between their present position and the north-eastern corner of Assam and the Chinese frontier. An imperfect Buddhism, and an unappreciated alphabet of Siamese origin, are the chief phenomena of their civilization.

The Jili.—These are conterminous with the Singpho; to whom they are closely allied, in language, at least; seventy words out of one hundred agreeing in the two vocabularies.

The Khamti come in now. These have been mentioned as Tha'y in their most northern localities. They occupy north-eastern Assam, and are conterminous with the Singpho. The Khamti language, with its per-centage of ninety-two words common to it and the Siamese of Bankok, ten degrees southwards, has only three out of one hundred that agree with the Singpho, and ten in one hundred with the Jili. This shows the remarkable character of their ethnological distribution, and, at the same time, suggests the idea of great displacement.

The Mishimi.—These occupy the north-east extremity of Assam. With the Mishimi we turn the corner, and find ourself on the northern or Tibetan frontier. Here it is the most western tribes which come first; and these are—

The Abors and Padam Bor-Abors.—The first, like the Kuki, on the mountain-tops; the latter, like the Khumia, on the lower ranges.

The Dufla.—Mountaineers west of the Abors, with whom they are conterminous in about 94 deg. East lon.

The Aka.—Mountaineers west of the Dufla, with whom they are conterminous in about 92 deg. East lon. The Akas bound Lower Assam, the eastern part of which lies between them and the Cachari country.

The tribes hitherto mentioned, although sufficiently numerous, represent the mountaineers of the Manipur and Tibetan frontiers only. The native tribes of the valley still stand over. These are—

1. The Muttuck or Moa Mareya, south of the Brahmaputra, and so far Indianized as to be Brahminical in religion. Their locality is the south bank of the Brahmaputra; opposite to that of—

2. The Miri, on the north.—The Miri are backed on the north by the Bor-Abors.

3. The Mikir.—Mr. Robertson looks upon these as an intrusive people from the Jaintia hills: their present locality being the district of Nowgong, where they are mixed up with—

4. The Lalong.—I cannot say whether the Lalong speak their originally monosyllabic tongue, or have learnt the Bengali—a phenomenon which does much to disguise the true ethnology of more than one of the forthcoming tribes; one of which is certainly—

5. The Dhekra, occupants of Lower Assam and Kamrup, where they are mixed up with other sections of the population.

6. The Rabha.—Like the Dhekra, these are Hindus. Like the Dhekra they speak Bengali. Hence, like the Dhekra, their true affinities are disguised. It is, however, pretty generally admitted by the best authorities that what may be predicated of the Garo and Bodo—two families of which a fuller notice will be given in the sequel—may be predicated of the sections in question, as also of—

7. The Hajong or Hojai.—Hindus, speaking a form of the Bengali at the foot of the Garo hills; and who join the Rabha, whose locality is between Gwahatti and Sylhet, i.e., at the entrance of the Assam valley.

The Garo of the Garo hills to the north-east of Bengal now require notice. A mountaineer of these parts has much in common with the Coosya; yet the languages are, perhaps, mutually unintelligible. In form they are exceedingly alike.

Now, a Garo[29] is hardy, stout, and surly-looking, with a flattened nose, blue or brown eyes, large mouth, thick lips, round face, and brown complexion. Their buniahs (booneeahs) or chiefs, are distinguished by a silken turban. They have a prejudice against milk; but in the matter of other sorts of food are omnivorous. Their houses, called chaungs, are built on piles, from three to four feet from the ground, from ten to forty in breadth, and from thirty to one hundred and fifty in length. They drink, feast, and dance freely; and, in their matrimonial forms, much resemble the Bodo. The youngest daughter inherits. The widow marries the brother of the deceased; if he die, the next; if all, the father.

The dead are kept for four days; then burnt. Then the ashes are buried in a hole on the place where the fire was. A small thatched building is next raised over them; which is afterwards railed in. For a month, or more, a lamp is lit every night in this building. The clothes of the deceased hang on poles—one at each corner of the railing. When the pile is set fire to, there is great feasting and drunkenness.

The Garo are no Hindus. Neither are they unmodified pagans. Mahadeva they invoke—perhaps, worship. Nevertheless, their creed is mixed. They worship the sun and the moon, or rather the sun or the moon; since they ascertain which is to be invoked by taking a cup of water and some wheat. The priest then calls on the name of the sun, and drops corn into the water. If it sink, the sun is worshipped. If not, a similar experiment is tried with the name of the moon. Misfortunes are attributed to supernatural agency: and averted by sacrifice.

Sometimes they swear on a stone; sometimes they take a tiger's bone between their teeth and then tell their tale.

Lastly, they have an equivalent to the Lycanthropy of the older European nations:—

"Among the Garrows a madness exists, which they call transformation into a tiger, from the person who is afflicted with this malady walking about like that animal, shunning all society. It is said, that, on their being first seized with this complaint they tear their hair and the rings from their ears, with such force as to break the lobe. It is supposed to be occasioned by a medicine applied to the forehead; but I endeavoured to procure some of the medicine thus used, without effect. I imagine it rather to be created by frequent intoxications, as the malady goes off in the course of a week or fortnight. During the time the person is in this state, it is with the utmost difficulty he is made to eat or drink. I questioned a man, who had thus been afflicted, as to the manner of his being seized, and he told me he only felt a giddiness without any pain, and that afterwards he did not know what happened to him."[30]

In a paper of Captain C. S. Reynolds, in the "Journal of the Asiatic Society of Bengal,"[31] we have the notice of a hitherto undescribed superstition; that of the Korah. A Korah is a dish of bell-metal, of uncertain manufacture. A small kind, called Deo Korah, is hung up as a household god and worshipped. Should the monthly sacrifice of a fowl be neglected, punishment is expected. If "a person perform his devotion to the spirit which inhabits the Korah with increasing fervour and devotion, he is generally rewarded by seeing the embossed figures gradually expand. The Garos believe that when the whole household is wrapped in sleep, the Deo Korahs make expeditions in search of food, and when they have satisfied their appetites return to their snug retreats unobserved."

The Miri are supposed to believe the same of what are called Deo Guntas, brought from Tibet.

Now what is the classification of all these tribes? Preliminary to the answer on this point, there are eleven dialects spoken in the parts about Manipur—besides the proper language of Manipur itself—to be enumerated. These are as follows:—1. Songpu. 2. Kapwi. 3. Koreng. 4. Maram. 5. Champhung. 6. Luhuppa. 7, 8, 9. Northern, Central, and Southern Tangkhul. 10. Khoibu; and 11. Maring. Now these twelve (the Manipur being included) have been tabulated by Mr. Brown, in such a way as to show the per-centage of words that each has with all the others; and not only these, but nearly all the tongues which we have had to deal with, are similarly put in order for being compared. The part of the table necessary for the present use is as follows:—

N. C. S. C M h T T T M B S a a L a a a i u i n S K m u n n n K M s r K n i o K o M p h g g g h a A h m a g J G p n a r a h u k k k o r A b i e r p i a u g p e r u p h h h i i k o m s e h l r r p w n a n p u u u b n a r i e n o i o i u i g m g a l l l u g - Aka 47 20 17 12 15 15 5 11 3 10 3 8 8 8 5 6 10 8 10 Abor 47 20 11 10 18 11 6 15 6 11 5 8 6 8 8 8 10 10 18 Mishimi 20 20 10 10 10 13 10 11 0 11 0 3 5 6 8 6 13 10 8 Burmese 17 11 10 23 23 26 12 16 8 20 6 11 11 11 10 13 13 16 16 Karen 12 10 10 23 17 21 8 15 10 15 8 12 4 12 8 12 12 10 15 Singpho 15 18 10 23 17 70 16 25 10 18 11 11 13 15 13 25 13 20 18 Jili 15 11 13 26 21 70 22 16 10 21 13 11 11 18 20 20 13 20 20 Garo 5 6 10 12 8 16 22 10 5 6 5 8 5 8 13 11 5 5 5 Manipuri 11 15 11 16 15 25 16 10 21 41 18 25 28 31 28 35 33 40 50 Songpu 3 6 0 8 10 10 10 5 21 35 50 53 20 23 15 15 13 8 15 Kapwi 10 11 11 20 15 18 21 6 41 35 30 33 20 35 30 40 45 38 40 Koreng 3 5 0 6 8 11 13 5 18 50 30 41 18 21 20 20 11 10 15 Maram 8 8 3 11 12 11 11 8 25 53 33 41 21 28 25 20 16 23 26 Champhung 8 6 5 11 4 13 11 5 28 20 20 18 21 40 20 20 16 15 25 Luhuppa 8 8 6 11 12 15 18 8 31 23 35 21 28 40 63 55 36 33 40 N. Tangkhul 5 8 8 10 8 13 20 13 28 15 30 20 25 20 63 85 30 31 31 C. Tangkhul 6 8 6 13 12 25 20 11 35 15 40 20 20 20 55 85 41 45 41 S. Tangkhul 10 10 13 13 12 13 13 5 33 13 45 11 16 16 36 30 41 43 43 Khoibu 8 10 10 16 10 20 20 5 40 8 38 10 23 15 33 31 45 43 78 Maring 10 18 8 16 15 18 20 5 50 15 40 15 26 25 40 31 41 43 78

The last eleven dialects are not spoken in any British dependency; and they have only been mentioned for the sake of explaining the table.

All belong to one and the same class; a point upon which I see no room for doubt; although respecting the value of that class I admit that some exists.

For this, the term Burmese is as good as any other—without professing to be better; yet, should it seem too precise, there is no objection to the sufficiently general term of monosyllabic being substituted for it.

The reader, however, may doubt the fact of the affinities. This has been done. Long before the present writer knew of such dialects as the Jili, Mishimi, Aka, Abor, Singpho, and the like, he had satisfied himself that the Garo was monosyllabic, and had so expressed himself in 1844,[32] when Brown's Tables had been published, though not seen by him. It was with surprise, then, that he found the author of them writing, that "it would be difficult to decide from the specimens before us, whether it is to be ranked with the monosyllabic or polysyllabic languages. It probably belongs to the latter."

Again, Mr. Hodgson makes the Garo Tamulian, i.e., polysyllabic; a fact which will be noticed again when the Bodo, Dhimal, and Kocch have been disposed of.

The Kocch, Bodo, and Dhimal is the title of one of that writer's works—a model of an ethnological monograph. This gives us a new class. The Bodo of Hodgson are the wild tribes that skirt the Himalayas, from Assam to Sikkim. West of these, between the river Konki and the river Dhorla are the Dhimal, a small tribe mixed with Bodo; and, southwards, in Kocch Behar, are the Kocch. The two former are so much described together that a separation is difficult. This leaves us at liberty to follow the details of either one population or of both. The history of a Bodo from his cradle to his grave is as follows. The birth is attended with a minimum amount of ceremonies. Midwives there are none; but labours are easy. Neither has the priest much to do with ushering-in the new-comer to the world. A short period of uncleanness is recognized, but it is only a short one; the purification consisting in the acts of bathing and shaving performed by the parties themselves. Four or five days after delivery, the mother goes out into the world; and at that time, the child is named. Any passing event determines this; as there are no family names, and no names taken from their mythology. The account, however, of Mr. Hodgson, in this respect is somewhat obscure, "A Bhotia chief arrives at the village, and the child is named Jinkhap; or a hill peasant arrives, and it is named Gongar, after the titular, or general designation of the Bhotias."

As long as a mother can suckle a child (or children) she continues to do so, sometimes for so long a period as three years, when the last and last but one may be seen sucking together.

The period of weaning is thus delayed; and, notwithstanding the current notion as to the prematurity of marriages in warm climates, that of wedlock is delayed as well: the male waits till he is twenty or twenty-five, the female till between fifteen and twenty. The parties least concerned are the bride and bridegroom; the parents do the courtship. Those of the lady take a payment. This is called a Jan amongst the Bodo, and varies from ten to fifteen rupees. With the Dhimal it is a Gandi, and amounts to a higher sum, ranging from fifteen to forty-five. Failing this, service must be done by the youth; and a wife be earned as Jacob earned Leah and Rachel. This is the Gabor of the Bodo, and the Gharjya of the Dhimal.

Such marriages are easily dissolved, i.e., at the option of either party. In case, however, of infidelity on the part of a wife having caused a divorce, the wedding-money is repaid. Adoption is common, concubinage rare; each being on a level with marriage in respect to the status of the children. Of these, all males inherit alike; but the rights of the female are limited.

The ceremony itself begins with a procession on the part of the bridegroom's friends to the bride's house, two females accompanying them. Of these, it is the business to put red-lead and oil on the bride-elect's hair. A feast follows; after which the husband takes his wife home. Thus far the Bodo forms agree with the Dhimal; but they differ in what follows.

The Bodo sacrifices a cock and a hen in the names of the bridegroom and the bride, respectively to the Sun.

The Dhimal propitiate Data and Bedata by presents of betel-leaf and red-lead.

Both bury their dead, and purify themselves by ablution in the nearest stream when the funeral procession is over. The family, however, of the deceased is considered as unclean for three days.

A feast with sacrifices attends the purification. Before sitting down, they repair once more to the grave, and present the dead with some of the food from the banquet;—"take and eat, heretofore you have eaten and drunk with us; you can do so no more; you were one of us, you can be so no longer; we come no more to you; come you not to us." After this each member of the party takes from his wrist a bracelet of thread, and throws it on the grave.

A ceremonial implies a priesthood. Under this class come the Deoshi, the Dhami, the Ojha, and the Phantwal.

The first of these is the village, the second the district, priest.

The Ojha is the village exorcist; and the Phantwal a subordinate of the Deoshi. The influence of this clerical body, although probably higher than Mr. Hodgson places it, is, evidently, anything but exorbitant.

I cannot find anything in the Bodo and Dhimal superstitions higher than what was found in Africa. Nor yet is anything essentially different. Similar intellectual conditions develop similar creeds, independent of intercourse; a fact which, the more we go into the natural history of religions, the more we shall verify. We read indeed of oaths and ordeals; but oaths and ordeals are by no means, what they have too loosely been supposed to be, appeals to the moral nature of the Divinity. The dhoom test, in Old Calabar, is an ordeal. The criminal tests of the Fantis are the same. Indeed, few, if any tribes, are without them. What the real ideas are which determine such and such-like ceremonies is difficult for intellectual adults to understand. The way towards their appreciation lies in the phenomena of a child's mind; the true clue to the psychology of rude populations.

If we take the Bodo and Dhimal religions in detail we find ourselves in a familiar field, with well-known forms of superstition around us.

Diseases are attributed to supernatural agency; and the medicine-man, exorcist, or Ojha, is more priest than surgeon.

The feticism of Africa re-appears; at least such is my inference from the following extract. "Batho is clearly and indisputably identifiable with something tangible, viz., the Sij or Euphorbia; though why that useless and even exotic plant should have been thus selected to type the Godhead, I have failed to learn."

Euhemerism, or the worship of dead men deified, is to be found either in its germs or its rudiments; at any rate, one of their deities bears the name of Hajo, a known historic personage. But this may be referable to Hindu influences unequivocally traceable in other parts of the Pantheon.

It is the rites and ceremonies of a country that give us its religion in the concrete. All beyond is an abstraction. These, with the Bodo and Dhimal, are numerous. Invocations, deprecations, and thanksgivings are all mentioned by Mr. Hodgson; and they are all attended by offerings or sacrifices; libations attend the sacrifices, and feasting follows the libations.

The great festivals of the year are four for the Bodo, three for the Dhimal.

a. In December or January, when the cotton-crop is ready, the Bodo hold their Shurkhar, the Dhimal their Harejata.

b. In February or March, the Bodo hold the Wagaleno.

c. In July or August, the rice comes into ear. This brings on the Bodo Phulthepno, and the Dhimal Gavipuja.

All these are celebrated out of doors, and on agricultural occasions.

d. The fourth great festival is held at home; its time being the month of October; its name Aihuno in Bodo, and Pochima paka in Dhimal. Here, in the Aihuno at least, the family assembles, the priest joins it, and the Sij, or Euphorbia, represents Batho. This is placed in the middle of the room, has prayers offered to it, and a cock as a sacrifice; whilst Mainou's offering is a hog; Agrang's a he-goat, and so on, through the whole list of the nine nooni madai, or deities thus worshipped. As for the symbols which represent them, besides the Sij, which stands for Batho, there is a bamboo post about three feet high, surmounted by a small cup of rice, denoting Mainou; but the equivalents of the other seven are somewhat uncertain.

The Wagaleno festival was witnessed by Mr. Hodgson and Dr. Campbell. The account of it is something lengthy. I mention it, however, for the sake of one of its principal actors—the Deoda. This is the possessed, who, "when filled with the god, answers by inspiration to the question of the priest as to the prospects of the coming season. When we first discerned him, he was sitting on the ground, panting, and rolling his eyes so significantly that I at once conjectured his function. Shortly afterwards, the rite still proceeding, the Deoda got up, entered the circle, and commenced dancing with the rest, but more wildly. He held a short staff in his hand, with which, from time to time, he struck the bedizened poles, one by one, and lowering it as he struck. The chief dancer with the odd-shaped instrument waxed more and more vehement in his dance; the inspired grew more and more maniacal; the music more and more rapid; the incantation more and more solemn and earnest; till, at last, amid a general lowering of the heads of the decked bamboo poles, so that they met and formed a canopy over him, the Deoda went off in an affected fit, and the ceremony closed without any revelation." This self-excited state of ecstasy is an element of most religions in the same stage of development; and a low level it indicates. In Greece, in Africa, and in Northern Asia, we find it as regularly as we find a coarse and material creed; and to the coarseness of the materialism of such a creed it is generally proportionate.

Witches, and the discovery of them, and the influence of the evil eye are part and parcel of the Bodo and Dhimal superstitions.

Kocch means a population, which possibly amounts to as much as a million souls, extended from about 88 deg. to 93-1/2 deg. East long., and 25 deg. to 27 deg. North lat., and of which Kocch Behar is the political centre. The term is ethnological—not political. It is ethnological, and not political, because, although originally native, it has since been partially abandoned. All the inhabitants of the parts in question once called themselves Kocch; and Kocch they were called by their neighbours the Mech. At this time the country was unequivocally other than Indian; i.e., in the same category with that of the Garo and Bodo. Since then, however, great changes have taken place; so that, just as Wales is partially Anglicized, the Welsh language being replaced by the English, the Kocch—the native tongue—is under the process of being replaced by a Hindu dialect. Nevertheless, just as many a Welshman who speaks nothing but English is still a Welshman, so are the Kocch, who have changed their languages, Bodo, Garo, or something closely akin, in ethnological position.

The extent to which different portions of the once great Kocch nation have abandoned or retained their original characteristics is easily measured.

1. Those who have changed most speak a form of the Bengali, and are imperfect Mahometans; imperfect, because their creed is strongly tinctured with Hinduism. Thus the very epithet which they apply to themselves is Brahminical; RajbansiSuryabansiSun-born. The converted Kocch of the Mahometan creed are chiefly of the lower order of the province of Behar.

2. Those who have changed, but changed less than the Mahometans of Behar, are either Brahminists or Buddhists—speaking the same Bengali dialect as the last. These are chiefly the higher classes of the population of Behar. They are Kocch in the way that the Cornishmen are Welsh. They consider them Rajbansi also. Doubtless, their Hinduism is imperfect; i.e., tinctured with the original paganism.

3. The primitive, unconverted, or Pani Kocch, have either not changed at all, or changed but little. They retain the original name of Kocch; which is not endured by the others. They retain their original tongue, which, according to Buchanan, has no affinity with any of the Hindu tongues. They retain their original customs; and they retain their original paganism. Lastly, Mr. Hodgson attests the "entire conformity of the physiognomy of all—with that of the other aborigines around them." He adds that he cannot improve on Buchanan's account of them, which is as follows:—"The primitive or Pani Kocch live amid the woods, frequently changing their abode in order to cultivate lands enriched by a fallow. They cultivate entirely with the hoe, and more carefully than their neighbours who use the plough, for they weed their crops, which the others do not. As they keep hogs and poultry they are better fed than the Hindus, and as they make a fermented liquor from rice, their diet is more strengthening. The clothing of the Pani Kocch is made by the women, and is in general blue, dyed by themselves with their own indigo, the borders red, dyed with Morinda. The material is cotton of their own growth, and they are better clothed than the mass of the Bengalese. Their huts are at least as good, nor are they raised on posts like the houses of the Indo-Chinese, at least, not generally so. Their only arms are spears: but they use iron-shod implements of agriculture, which the Bengalese often do not. They eat swine, goats, sheep, deer, buffaloes, rhinoceros, fowls, and ducks—not beef, nor dogs, nor cats, nor frogs, nor snakes. They use tobacco and beer, but reject opium and hemp. They eat no tame animal without offering it to God (the Gods), and consider that he who is least restrained is most exalted, allowing the Garos to be their superiors, because the Garos may eat beef. The men are so gallant as to have made over all property to the women, who in return are most industrious, weaving, spinning, brewing, planting, sowing; in a word, doing all work not above their strength. When a woman dies the family property goes to her daughters, and when a man marries he lives with his wife's mother, obeying her as his wife. Marriages are usually arranged by mothers in nonage, but consulting the destined bride. Grown up women may select a husband for themselves, and another, if the first die. A girl's marriage costs the mother ten rupees—a boy's five rupees. This sum is expended in a feast with sacrifice, which completes the ceremony. Few remain unmarried, or live long. I saw no grey hairs. Girls, who are frail, can always marry their lover. Under such rule, polygamy, concubinage, and adultery are not tolerated. The last subjects to a ruinous fine, which if not paid, the offender becomes a slave. No one can marry out of his own tribe. If he do, he is fined. Sutties are unknown, and widows always having property can pick out a new husband at discretion. The dead are kept two days, during which the family mourn, and the kindred and friends assemble and feast, dance and sing. The body is then burned by a river's side, and each person having bathed returns to his usual occupation. A funeral costs ten rupees, as several pigs must be sacrificed to the manes. This tribe has no letters; but a sort of priesthood called Deoshi, who marry and work like other people. Their office is not hereditary, and everybody employs what Deoshi he pleases, but some one always assists at every sacrifice and gets a share. The Kocch sacrifice to the sun, moon, and stars, to the gods of rivers, hills and woods, and every year, at harvest-home, they offer fruits and a fowl to deceased parents, though they believe not in a future state! Their chief gods are Rishi and his wife Jago. After the rains the whole tribe make a grand sacrifice to these gods, and occasionally also, in cases of distress. There are no images. The gods get the blood of sacrifices; their votaries, the meat. Disputes are settled among themselves by juries of Elders, the women being excluded here, however despotic at home. If a man incurs a fine, he cannot pay with purse, he must with person, becoming a bondman, on food and raiment only, unless his wife can and will redeem him."

I must now request particular attention on the part of the reader to the terms which Mr. Hodgson applies to the physical conformation of these northern, or sub-Himalayan tribes; and still closer attention must be given to his nomenclature. He calls the stock in question Tamulian. This connects it with the South Indian. He contrasts it with the Hindu. By this he means the Brahminical elements of the Indian populations.

Let us then see what points he considers to be Tamulian.

1. There is "less height, less symmetry, more dumpiness and flesh."

2. There is "a somewhat lozenge contour (of face) caused by the large cheek-bones."

3. There is "less perpendicularity of features in the front—a larger proportion of face to head—a broader flatter face—a shorter wider nose, often clubbed at the end, and furnished with round nostrils."

4. There is a smaller eye, "less fully opened, and less evenly crossing the face by their line of aperture." In other words, there is the oblique eye, so much considered in the Chinese physiognomy.

5. Lastly, there are larger ears, thicker lips, and less beard.

I submit that all these points are Mongolian; and this is what Mr. Hodgson evidently thinks also.

The whole class has passed beyond the hunter state, if ever such existed. It has passed beyond the pastoral or nomadic state also; if such existed. It is at present—and, perhaps, has always been—an agricultural state of society. On the other hand—the industrial state, the development represented by towns and commerce, has not been attained.

The whole stock is essentially agricultural. Likewise, the agriculture is peculiar. We may explain it by the term erratic. They "never cultivate the same field beyond the second year, or remain in the same village beyond the fourth to sixth year. After the lapse of four or five years they frequently return to their old fields and resume their cultivation, if in the interim the jungle has grown well, and they have not been anticipated by others, for there is no pretence of appropriation other than possessory, and if, therefore, another party have preceded them, or, if the slow growth of the jungle give no sufficient promise of a good stratum of ashes for the land when cleared by fire, they move on to another site, new or old. If old, they resume the identical fields they tilled before, but never the old houses or site of the old village, that being deemed unlucky. In general, however, they prefer new land to old, and having still abundance of unbroken forest around them, they are in constant movement, more especially as, should they find a new spot prove unfertile, they decamp after the first harvest is got in."

Arva in annos mutant et superest ager. This passage is explained by their customs.

In respect to their social constitution, they dwell in small communities of from ten to forty houses; each of which community is under a gra or head. This is Hindu—except that as the Hindu villages are both larger and more permanent, the functionaries, in addition to the headman, are more numerous. This is noted, because the difference in the two sorts of village government seems to be one of degree rather than kind.

And now comes more in the way of classification. The Bodo are Kachars, or the Kachars are Bodo. Their languages are the same, so are their gods, so is their name; since Kachar is a Hindu, and no native term—the native name (i.e., of the Kachars) being Bodo. On the other hand, the Hindu name of the Bodo is Mech. Whoever looks to a map will find that the outline of the Bodo area is very deeply indented; implying either a great original irregularity of area, or great subsequent displacement.

Now follow the Garo. One fourth—fifteen out of sixty—of the words of Mr. Brown's Garo vocabulary is Bodo. The inference? That the Bodo and Garo are in the same category. What is this? Mr. Hodgson makes both Tamulian or Indian. In my own mind both are Burmese. But be this as it may, one fact is certain; viz., that a transition between the tongues of the Indian and the tongues of the Indo-Chinese peninsula exists, and that the lines of demarcation which divide them are less broad and trenchant than is generally supposed.

The Dhimal bring us to Sikkim. The dominant nation of Sikkim are—

The Lepchas.—Their language also is monosyllabic; but it is Tibetan rather than Burmese. They are a Sikkim rather than a British Indian population.

When we have passed the rajahship of Sikkim, we reach that of Nepal. This, again, is independent. Such being the case, the line of frontier between the Hindu populations and the populations of the Bodo and Garo character lies beyond the pale of the British dependencies.

But in proceeding westward, we pass Nepal, and reach Kumaon.

This is British, and, as it extends as far north as the Himalayas, it may contain monosyllabic languages, and tribes speaking them. It may present also instances of intermixture like those which we have already found in Behar—the line of demarcation being equally difficult and undefined. Difficult and undefined it really is—because, although it is an easy matter to take a portion of the Sirmor, Gurhwal, or Kumaon population, and say, "this is Hindu because both language and creed make it so," it is by no means so easy to prove that the blood, pedigree, or descent is Hindu also. To repeat an illustration already in use—many such populations may be Hindu only as the Cornishmen are English.

Now the populations of the Tibetan stock to the west of Nepal, so little known in detail, must be illustrated by means of our knowledge of the tribes of Nepal and Tibet most closely related to them—by those of Nepal on the east, and those of Tibet on the north.

For neither of these areas are there any very minute data. For the aborigines of eastern and central Nepal, we have plenty of information. They are tribes speaking monosyllabic languages, and tribes in different degrees of intercourse with the Hindus; being by name—1. The Magars. 2. The Gurungs. 3. The Jariyas. 4. The Newars. 5. The Murmis. 6. The Kirata. 7. The Limbu; and 8. The Lepchas, common to the eastern boundary of Nepal, to the western part of Butan, and to Sikkim. This, however, will not bring us far west enough for the Kumaon frontier; indeed, for the forests of Nepal west of the Great Valley, we have the notice of one family only—the Chepang. For this, as for so much more, we are indebted to Mr. Hodgson. It falls into three tribes; the Chepang proper, the Kusunda, and the Haju. Its language (known to us by a vocabulary) is monosyllabic; its physical conformation, that of the unmodified Indian.

So much for analogy. In the way of direct information we simply know that the Pariahs, or outcasts, of Kumaon[33] are called Doms. These have darker skins and curlier hair than the Hindus. Are these enslaved and partially amalgamated aborigines? Probably. Nay more; in the eastern part of the province, amidst the forests at the foot of the Himalayas, a community of about twenty families, pertinaciously adheres to the customs of their ancestors, resembles the Doms in looks, and is called Rawat or Raji. Though I have seen no specimen of their language, I have little doubt as to the Rawat of Kumaon being the equivalents to the Chepang of Nepal.

From Konawur we have three monosyllabic vocabularies, the Sumchu, the Theburskud, and the Milchan; but the exact amount to which the Tibetan and the Hindu populations indent each other along the western Himalayas is more than I can give.

Here end the monosyllabic tongues spoken in British India. But they fringe the Himalayas throughout, and occur in the country of Gholab Singh, as well as in the independent rajahships between the Sutlege and Cashmeer. My latest researches have carried them even further westward than Little Tibet; as far as the Kohistan, or mountain country, of Cabul—the Der, Lughmani, Tirhai, and other languages, known, wholly or chiefly, through the vocabularies of Lieutenant Leach, being essentially monosyllabic in structure, and definitely connected with the tongues of Tibet, and Nepal in respect to their vocables.

But this is episodical to the subject—a subject still requiring the notice of a very important phenomenon.

Polyandria[34] is a term in ethnology, even as it is in botany. Its meaning, however, is different. Etymologically, it denotes a form of polygamy. Polygamy, however, being restricted to that particular form of marriage which consists in a multiplicity of wives, polyandria expresses the reverse, viz., the plurality of husbands.

At the first glance, the word polyandria looks like a learned name for a common thing; and suggests the inquiry as to how it differs from simple promiscuity of intercourse; or, at least, how far the Tibetan wife differs from the fair frail one who was always constant to the 85th regiment. The answer is not easy. Still it is certain that some difference exists—if not in form, at least, in its effects. One of these, in certain countries where polyandria prevails, is the law of succession to property. This follows the female line, rather than the male.

Again—the marriage of the widow with the surviving brother of her husband, is polyandria under another form.

What the exact polyandria of Tibet is, is uncertain. I am not prepared to deny its existence even in so extreme a form as that of one woman being married to several husbands, all alive at once. Still, I think it more likely that either the circle of community was limited to certain degrees of relationships, or else that the multiplied husbands were successive, rather than simultaneous. Still, the facts of the Tibetan polyandria require further investigation.

One thing, only, is certain—viz., that as an ethnological criterion the practice is of no great value. Capable, as it has been shown to be, of modification in form, it is anything but limited to either Tibet, or the families allied to the Tibetan. It occurs in many parts of the world. It is a Malabar practice; where it is, probably, as truly Tibetan as in Tibet itself. But it is also Jewish, African, Siberian, and North American; so that nothing would more mislead us in the classification of the varieties of man than to mistake it for a phenomenon per se, and allow it to separate allied, or to connect distinct populations.

Necdum finitus Orestes.—There are several populations which, on fair grounds, have been believed to be in the same category with the Dhekra, i.e., which are Hindu in language and creed, though monosyllabic in blood. The Kudi, Batar, Kebrat, Pallah, Gangai, Maraha, Dhanak, Kichak, and Tharu, are oftener alluded to than described—though, doubtless, a better-informed investigator in such special matters than the present writer could find several definite details concerning them. They seem chiefly referable to Behar and north-eastern Bengal. The Dhungers—in the same class—the husbandmen of South Behar, bring us down to the vicinity of the population next to be noticed; a population which is generally considered with reference to the nations, tribes, and families of Southern rather than Northern India.

The name of this family has already been mentioned. It is Tamulian; and the Tamulian physiognomy has been described. It has been seen to extend as far north as the Himalayas. If so, the nations already enumerated have been Tamulian; and no new class is now approaching. This may or may not be the case. Another change, however, is more undeniable. This is that of language. It is no longer referable to the Chinese type; since separate monosyllables have, more or less perfectly, become agglutinated into inflected forms, and the speech is as poly-syllabic as the other tongues of the world in general. As we approach the south this abandonment of the monosyllabic character increases, and from the Tamul language spoken between Pulicat and Cape Comorin, the term Tamulian—applicable in a general ethnological sense—is derived. Agglutinated (or agglutinate) is also a technical term. It means languages in the second stage of their development; when words originally separate, such as adverbs of time, prepositions, and personal pronouns, have become permanently connected with the root, so as to form tenses, cases, and persons—the union of the two parts of an inflected word being still sufficiently recent and imperfect to leave their original separation and independence visible and manifest. When the incorporation or amalgamation, has become more complete; so complete, as in most cases to have obliterated all vestiges of an original independence; the agglutinate character has departed, the second stage of development has been passed, and the language is in the same class with those of Greece, Rome, and Germany, rather than in that of the tongues in question, and of many others.

To return, however, to the Tamulian family, meaning thereby a branch of the great Mongolian stock, speaking, either now or formerly, a language more or less allied to the Tamul of the Dekhan.

The first members of the class, as we proceed southwards from Behar, are certain hill-tribes of the Rajmahali Mountains—the Rajmahali mountaineers. Their Mongolian physiognomy is unequivocal;—a Mongolian physiognomy but conjoined with a dark skin. They have "broad faces, small eyes, and flattish or rather turned-up noses. Their lips are thicker than those of the inhabitants of the plain."[35]

The flattened nose reminded the writer of the Negro, and the general character of the features of the Chinese or Malay; though it is added that the resemblance is in a great degree lost on closer inspection. At the same time it has been sufficiently recognized to have originated the hypothesis of a descent from one of those nations as a means of accounting for it.

With a slight tincture of Brahminic Hinduism, the Rajmahali mountaineers are Pagans. Bedo is one of their gods; doubtless the Potteang of the Kuki, and the Batho of the Bodo. Gosaik, too, is either the name of a god, or a holy epithet; this, also, being a mythological term current amongst many other tribes of India. Other elements in their imperfectly-known mythology deserve notice. Their priesthood contains both Demauns and Dewassis; the latter form being the Bodo Deoshi. As the names are alike, so are the functions. The Dewassi is an oracular seer. When he vouchsafes to give answers, his inspiration takes the form of frenzy—but he neither hurts nor speaks to any one. He makes signs for a cock, and for a hen's egg as well. The cock's head he wrenches off, and sucks the bleeding neck. The egg he eats. After this he seeks the solitude of the wood or stream; and is fed by the deity. Sometimes he has ridden a snake; sometimes put his hands in the mouth of a tiger with impunity. Trees too large to move, or too thorny to touch, he places on the roofs of houses. He sees Bedo Gosaik in visions; and, in the sacrifices therein enjoined, red paint, rice, and pigeons make a part. From the touch of women he abstains; so he does from the taste of flesh. Either would make his prophecies false.

There are also certain sacrifices that the Maungy (chief?) of each village makes, and in which threads of red silk play a part.

One of their gods—an elemental one—is the god of rain, and the dangers of a drought are averted by praying to him. A ceremony called the Satane determines the chief who takes the office of invoker.

A black stone, called Ruxy, is much of the same sort of fetish with these mountaineers as the Sij with the Bodo. The name, too, Ruxy Nad, suggests the Nat worship of the Silong, Kariens, and others.

The northern half of the Tamulian families are, like the Welsh, the Cornish, and the Bretons of France, members of the same ethnological group, but not in geographical contact with each other. Or, rather, they are, like the Celtic population of Wales and the Scottish Highlands, cut off from one another by a vast tract of intervening Anglo-Saxons. Yet the time was when all was Celtic, from Cape Wrath to the Land's End; and when the original population extended, in its full integrity, over York and Nottingham, as well as over Merioneth and Argyleshire. And so it is with the populations in question. They stand apart from each other, like islands in an ocean; the intervening spaces being filled up by Hindus. At the same time the isolation has been much overvalued, and, I imagine that when greater attention shall have been bestowed upon this important subject, connecting links which have hitherto been unnoticed will be detected.

The next locality where we find a population akin to the Rajmahali mountaineers, is the mountain system of Orissa. These are called by the Hindus Kols (Coolies), Khonds and Surs. Such, however, are no native designations—no more than the classical term Barbarian, or the English word Tartar. The people themselves have no collective name; but, being divided into tribes, have a separate one for each.

I say that this branch of Tamulians is isolated, because I am not able to show its continuity; the range of hill-country which gives rise to the rivers between the Ganges and Mahanuddy being but imperfectly known.

In Orissa, the most northern of the hill-tribes are the Kol of Cuttack. South of these come the Khonds best studied in the neighbourhood of Goomsoor. The following is a list of their gods, and as n seems to stand for d, Pennu is but another name for Bedo, and Gossa Pennu for Bedo Gosaik:—

1. Bera Pennu, or the earth god. 2. Bella Pennu, the sun god, and Danzu Pennu, the moon god. 3. Sandhi Pennu, the god of limits. 4. Loha Pennu, the iron god, or god of arms. 5. Jugah Pennu, the god of small-pox. 6. Madzu Pennu, or the village deity, the universal genius loci. 7. Soro Pennu, the hill god. 8. Jori Pennu, the god of streams. 9. Gossa Pennu, the forest god. 10. Munda Pennu, the tank god. 11. Sugu Pennu, or Sidruja Pennu, the god of fountains. 12. Pidzu Pennu, the god of rain. 13. Pilamu Pennu, the god of hunting. 14. The god of births.[36]