II. THE SOUTH ATLANTIC GROUP.

1. THE AMAZONIAN REGION.

Those two mighty rivers, the Amazon and the Orinoco, belong to one hydrographic system, the upper affluents of the latter pouring their waters for six months of the year into the majestic expanse of the former. Together they drain over three million square miles of land,[335] clothed throughout with lush tropical vegetation and seamed by innumerable streams, offering natural and facile paths of intercommunication. It is not surprising, therefore, that we find linguistic stocks extended most widely over this vast area, each counting numerous members. Of them the most widely disseminated were the Tupi, the Tapuya, the Carib and the Arawak families, and to these I shall first give attention.

1. The Tupis.

Along the coast of Brazil and up the Amazon there is current a more or less corrupted native tongue called the “common language,” lingua geral. It is derived mainly from the idiom of the Tupis, whose villages were found by the first discoverers along the seaboard, from the mouth of the La Plata to the Amazon and far up the stream of the latter. According to their traditions, which are supported by a comparison of their dialects, the Tupis wandered up the coast from the south. Their earlier home was between the Parana river and the Atlantic. There they called themselves Carai, the astute, a term they afterwards applied to the Spaniards, but later were given the name Guaranis, meaning warriors, by which they are generally known. They must have been very numerous, as a careful estimate made in 1612 computed those then living in the modern states of Corrientes and Uruguay at 365,000; a census which could not have been much exaggerated, as about a century later the Jesuits claimed to have over three hundred thousand Christianized and living in their “reductions;”[336] even to-day ninety per cent. of the population of Uruguay have Guarani blood in their veins.

The inroads of the Spaniards from the south and of the kidnapping Portuguese from the east, reduced their number greatly, and many bands sought safety in distant removals; thus the Chiriguanos moved far to the west and settled on the highlands of Bolivia, where they have increased their stock from four or five thousand to triple that number,[337] extending as far south as the Pilcomayo river. On the upper waters of the Parana were the Tapes, a nation so called from the name of their principal village. It is another form of Tupi, and means “town.” They received the early missionaries willingly, and are complimented by these as being the most docile and intelligent of any of the nations of South America.[338]

The Tupi tribes did not extend north of the immediate banks of the Amazon, nor south of the Rio de la Plata. It would appear not improbable that they started from the central highlands where the Tapajoz on the north and the Paraguay on the south have their sources. Their main body followed the latter to the Atlantic, where the Tupis proper separated and moved up the coast of Brazil. This latter migration is believed to have been as late as a few hundred years before the discovery.[339]

Like the Tapuyas, the Tupis have a tendency to dolicocephaly, but it is less pronounced. They are less prognathic, the forehead is fuller and the color of the skin brighter. The hair is generally straight, but Pöppig saw many among the Cocamas of pure blood with wavy and even curly hair.[340]

I have no hesitation in including in the Tupi family the Mundurucus, or Paris, on the upper Tapajoz. Their relationship was fully recognized by Professor Hartt, who was well acquainted with both dialects.[341] They are a superior stamp of men, tall, of athletic figures, light in color, their naked bodies artistically tattooed. Their women are skilled in weaving cotton hammocks, and the men pursue some agriculture, and manufacture handsome feather ornaments.

To the same family belong the Muras and Turas, in the swampy valley of the Madeira in its middle course, “an amphibious race of ichthyophagi,” as they are called by Martius, savage and hostile, and depraved by the use of the parica, a narcotic, intoxicating snuff prepared from the dried seeds of the Mimosa acacioides. At the beginning of this century they were estimated at 12,000 bowmen; but this was doubtless a great exaggeration. Though their dialect differs widely from the lingua geral, the majority of their words are from Tupi roots.[342] Others are related to the language of the Moxos, and in the last century certain of their tribes lived in the immediate vicinity of these, and were brought into the “reductions” of the Moxos Indians by the Jesuit missionaries.[343] The tendency of their migrations has been down the Madeira.

The tribes of this lineage in the extreme south of Brazil were numerous. The Guachaguis, corresponding apparently to the modern Guachis, are said by Lozano to speak a corrupt Guarani.[344] Vocabularies have been obtained by Castelnau and Natterer, which indicate only a remote resemblance. According to their own tradition, they migrated from near the Moxos in the Bolivian highlands.

The Gualachos, who spread from the river Iguaza to the sea coast, spoke a Guarani dialect in which the sounds of f, j and l were present, which, in pure Guarani, are absent. They built thatched houses divided into several rooms, and raised abundant harvests.[345]

The Omaguas and Cocamas, the most western of the Tupis, dwelling within the limits of Ecuador, had evidently profited by their contiguity to the civilization of Peru, as they are described by early travelers as familiar with gold, silver and copper, living in permanent villages connected by good roads, and cultivating large fields of cotton, maize and various food-plants. The art-forms which they produced and the prevalence of sun-worship, with rites similar to those of Peru, indicate the source of their more advanced culture. By some authors the Omaguas are stated to have migrated down the Rio Yupura from Popayan in New Granada, where a tribe speaking their dialect, the Mesayas are alleged still to reside.[346] The peculiar “mitred” skulls of the Omaguas are an artificial deformity prized by them as a beauty.

The Tupi is rich in mythological tales which have been collected by several competent students of their tongue. (Hartt, Magalhaes, etc.) Their religion is a simple animistic nature-worship.

The dead were buried in large urns, usually in localities set aside for the purpose. One such on the island Maraho, near the mouth of the Amazon, has yielded a rich harvest to archæologists.

The general culture of the Tupis was superior to that of any other Brazilian tribes, but much inferior to that of the Incas. They were to a slight extent agricultural, raising maize, manioc, tobacco, which they smoked in pipes, and several vegetables. Some fowls, monkeys and peccaries were tamed and used as food. Their houses were of straw, lattice work and leaves, sometimes plastered with mud. The communal system prevailed, twenty or thirty families occupying one residence. A number of such houses would be erected on some favorable site and surrounded by a palisade of strong poles. These towns were, however, not permanent, and nearly half the year was spent in hunting and fishing expeditions along the streams. They went entirely naked, but wove excellent hammocks from the bark of trees and other vegetable fibres. Devoid of a knowledge of metals, they were in the height of the age of polished stone, many of their products in this direction being celebrated for symmetry and delicacy.[347] The art of the potter was also well developed, and the vases from the Amazon, called igasauas, rank both in symmetry, decoration and fine workmanship among the most creditable specimens of American ceramics.

The language which characterizes this widely distributed stock is polysynthetic and incorporating, with the flexibility peculiar to this class of tongues. It has been the subject of a number of works, but still lacks a thorough comparative treatment. The Jesuit missionaries adopted the Guarani dialect throughout their extensive “reductions,” and translated into it a variety of works for the instruction of their acolytes, some of which have been printed.

TUPI LINGUISTIC STOCK.
2. The Tapuyas.

The Tapuya stock is at once the most ancient and the most extensive now living on the soil of Brazil. Its various tribes are found from s. lat. 5° to s. lat. 20°, and from the Atlantic to the Schingu river. The name Tapuya was applied to them by the Tupis, and means “enemies” or “strangers”—two ideas which are always synonymous in primitive life. They are also called Crens or Guerens, the Old Ones or Ancient People. This seems to have reference to their possession of the coast before the arrival of the Tupi hordes from the south.

By some writers they are believed to have been the earliest constructors of the sambaquis, the shell-heaps or kitchen-middens, which are of great size and numerous, along the Atlantic and its bays. These are supposed to indicate an antiquity of 2,000 years;[348] but the Tapuyas can lay claim to a title to their land far older than that. The skulls and human bones which were discovered by Dr. Lund in the caves of Lagoa Santa in immediate juxtaposition to those of animals now extinct, came from a region occupied by the Tapuyas, and are in all respects parallel to those of the tribe to-day. This would assign them a residence on the spot far back in the present geologic period.

Their appearance is that of an antique race of men. They are of middle height, with long upper and short lower extremities. The face is broad, the eyes small and under prominent ridges, the forehead low and retreating; the sutures are simple, the face prognathic, and the skull decidedly dolichocephalic (73), but of good capacity (1470 cub. cent.), and leptorhinic; the mouth is large and the nose prominent. In color they present a variety of shades of reddish-brown, and their hair, which is coarse, verges rather on the dark-brown than the black.[349] They are not ugly, and the expression of the face, especially in the young, is often attractive. Those of them, however, who distend the lower lip with the large labret or botoque (from which the Botocudos derive their name), cannot be other than hideous to European eyes.

In culture the Tapuyas are reported to stand on the lowest scale. When free in their native woods they go absolutely naked; they have no other houses than temporary shelters of leaves and branches; they manufacture no pottery, build no canoes, and do not know how to swim. When first in contact with the whites they had no dogs, knew nothing of the use of tobacco or salt, and were common cannibals. They have no tribal organizations and no definite religious rites.

To counterbalance all these negatives, I hasten to add that they are hunters of singular skill, using strong bows with long arrows, manufacture polished stone axes and weave baskets of reeds, and, what is rare among the Indians, use tapers made from wild bees-wax and bark fibre.[350] Their marriages are monogamous, though rarely permanent, and they are not devoid of family affection.[351] Though lacking definite religious ideas, they are careful to bury the dead, and have a belief that the spirit of the departed survives and wanders about at night, for which reason they are loth to move in the dark. The soul of a chief may take the form of a jaguar. During a thunder storm they shake a burning brand and shoot arrows toward the sky, to appease by imitation the powers of the storm; and they are much given to semi-religious dances, in which their motions are to the sound of a native flute, which is played with the nose.[352]

Their language is difficult in its phonetics, and presents a contrast to most American tongues by its tendency toward the isolating form, with slight agglutination. A carefully prepared vocabulary of it has recently been published by Dr. Paul Ehrenreich,[353] whose studies on this stock have been peculiarly valuable.

TAPUYA LINGUISTIC STOCK.

The Goyotacas in the province of Goyaz and the regions adjacent include a large number of tribes which Von Martius has shown to have sufficient linguistic affinity among themselves to unite in one group, and connections enough with the Tapuya stem to be regarded as one of its sub-stocks.[354]

GOYOTACA SUB-STOCK.

Another group believed by Martius to be a mixed off-shoot of the Tapuya family belong to what I may call the

TUCANO SUB-STOCK.

All these tribes are found in the vicinity of the river Uaupes, and are distinguished by three vertical lines tattooed or incised on the cheeks. They take their name, as do some other Brazilian tribes not related to them, from the beautiful toucan bird, which is frequently held sacred among them, and is sometimes chosen as the totem of a gens.

I also attach to this stock the Carnijos or Fornio, a vocabulary of whose language has been published by Professor John C. Branner, and which hitherto has not been identified.[355] The following comparison between it and the Tapuya dialects will show the affinity:

CARNIJOS.TAPUYA.
Fire,tŏch,tiaköh.
Eye,i-to,ainthó, kitho.
Nose,d-ereta,d’asigri.
Tooth,i-axi,aiquá, daguoi.
3. The Arawaks.

The Arawak stock of languages is the most widely disseminated of any in South America. It begins at the south with the Guanas, on the head-waters of the river Paraguay, and with the Baures and Moxos on the highlands of southern Bolivia, and thence extends almost in continuity to the Goajiros peninsula, the most northern land of the continent. Nor did it cease there. All the Antilles, both Greater and Less, were originally occupied by its members, and so were the Bahama Islands,[356] thus extending its dialects to within a short distance of the mainland of the northern continent, and over forty-five degrees of latitude. Its tribes probably at one time occupied the most of the lowlands of Venezuela, whence they were driven not long before the discovery by the Caribs, as they also were from many of the southern islands of the West Indian archipelago. The latter event was then of such recent occurrence that the women of the Island Caribs, most of whom had been captured from the Arawaks, still spoke that tongue.

They were thus the first of the natives of the New World to receive the visitors from European climes, and the words picked up by Columbus and his successors on the Bahamas, Cuba and Hayti, are readily explained by the modern dialects of this stock. No other nation was found on any part of the archipelago except the two I have mentioned. The whole of the coast between the mouths of the Orinoco and Amazon appears to have been in their possession at or a short time before the epoch of the discovery.

The Antis or Campas, who perhaps occupy the original home of the stock, own as the centre of their domain the table-land known as El Gran Pajonal, or the Great Grass Field, bounded by the rivers Ucayali, Pachitea and Perene. Their hue is a bistre and their habits wild; some slight tillage is carried on, and the women spin and weave the wild cotton into coarse garments. The taming of animals is one of their arts, and around their huts are seen monkeys, parrots, peccaries and tapirs.[357] It is noteworthy that some of them are skilful blacksmiths, smelting the metal from the native ores, and working it into axes, knives, spear points, etc., of excellent quality.[358]

The names Campas and Antis were used as generic terms, the latter applied to the tribes on the slopes of the Cordilleras and the former to those on the plains. A large number of sub-tribes are named by the older writers, the principal of which were the Choseosos, Machigangas, Pilcosumis and Sepaunabos. The Machigangas lived on the Pilcopata and Vilcanota, and their language has been erroneously stated by Von Tschudi to be an independent stock.[359] The Chunchas and Cholones are by some classed with the Campas, and they are said to have been the possessors of the famous Cerro de Sal, or Salt Mountain, to which the neighboring tribes repaired in great numbers to obtain supplies of this useful article.

The Guanas are a nation who have long lived on the upper Paraguay, in the province Mato Grosso on the river Mambaya, and vicinity. D’Orbigny believed that they were a member of the Mataco group,[360] but they are now recognized as belonging to the Arawak stock. They are noteworthy for their peaceful disposition and unusual intelligence. Hervas speaks of them as the most able nation visited by the missionaries in the whole of America.[361] The traveler Castelnau confirmed this good opinion. He found them living in neat houses and cultivating the land with skill and industry. They raised not only the ordinary food plants, but cotton and sugar cane, pressing the sap from the latter by machinery of their own devising, and moulding the sugar into loaves. Their cotton cloth, dyed of various colors, was highly esteemed for its texture.

Castelnau describes them as occupying four settlements near Albuquerque and Miranda, and comprising the Chualas or Guanas proper, the Terenos, the Laianas, and the Quiniquinaos.[362] Later investigations have shown that of these the Terenos and Quiniquinaos are members of the Guaycuru stock of the Chaco, and that the Chualas and Laianas alone belong to the true Guanas.[363]

The Paiconecas or Paunacas were attached to the mission of the Conception in Bolivia, in 16° south latitude. They numbered about 500 in 1831. In customs and appearance they approached the Chiquitos. Their former home was between the sources of the Rio Blanco and Rio Verde.

The Saravecas, three or four hundred in number in 1831, were attached to the mission of Santa Anna, in Bolivia, and were its handsomest members. Their former homes were in the eastern hills of the Cordillera, about 16° south latitude.

Although these are classed as irreducible stocks by D’Orbigny and others who have followed him, they are both clearly branches of the Arawak stem, as will be seen by a brief comparison.[364]

PAICONECA.SARAVECA.ARAWAK STOCK.
Sun,isese,caame,sese, camu.
Moon,kejere,cache,kejeres, kashi.
Fire,chaki,tikai,yaki, ikii.
Water,ina,une,ine, une.
Eye,ihuikis,nol,nohlo, ikise.

Others could readily be added, but the above are sufficient.

Another important tribe of this stock in this region were the Piros, otherwise called Chuntaquiros and Simirenchis, whose home was about the junction of the Ucayali and Apurimac, and thence along both these rivers. The vocabularies of their tongue obtained by Castelnau and Paul Marcoy leave no doubt of their affiliations. They were largely converted by the Jesuits between 1683 and 1727.

The Wapisianas, or Wapianas in British Guiana, with their sub-tribe the Atorai (Tauri or Dauri), are stated by Im Thurn to speak a tongue wholly different from the Arawak; but an analysis of its expression and an extended comparison place it beyond doubt in this stock.[365]

The Tarumas and Maopityans, who now live in southern British Guiana, but are said to have originally come from the Rio Negro, speak related dialects.

They enjoy a rather high degree of culture, being celebrated for the manufacture of cassava graters, for the hunting dogs which they breed and train, and for the fine pottery they manufacture. Both Schomburgk and Im Thurn regard them as an independent stock; but from a comparison of the fifteen nouns given by the former in their language,[366] I infer that they are an Arawak tribe, speaking a dialect mixed with some Carib and Tupi words, and with frequent vowel elision.

Taruma.
Sun,ouang,(auvan-ialü, Paravilhana).
Moon,piwa,(pia, Baniva, piua, Ouayéoué).
Fire,hua,(hua-to, Carib).
Water,tza,(tuná, Carib).
Head, my,a-tta,(no-totia, Baré).
Eye, my,a-tzi,(a-kussi, Arawak).
Mouth,me-ruku-kanna,(ülle-rukuhu, Arawak).
Nose,assa,(issi-rihi, Arawak).
Hand,ahu,(kx-aua, Bakairi).
Foot,appa,(upu, Galibi).
Bow,tzeika,(takou, Carib).
Star,uingra,(uinari, Baré).

This comparison leaves little doubt but that this mixed dialect is chiefly of Arawak lineage.

The Arawaks wandered as far east as the upper Schingu river, where Von den Steinen found the Kustenau, a distant member of the stem, with various minor tribes, as the Vauras, Mehinacus, etc. Along the river Ventuari the populous tribe of the Maipures have taken a conspicuous place in the annals of the missions. Indeed, the whole stock is sometimes called by their name;[367] but it is well to retain the better known Arawak, which is the appellation of that portion of the tribe in Guiana between the Corentin and Pomeroon rivers. It means “meal-eaters,” and was first applied to them in derision on account of their large consumption of cassava bread.

There is a prevailing similarity in their physical type. The adults are slightly undersized, rarely reaching above five feet six inches, with low foreheads and straight narrow noses. The form of the skull is short and the jaws are not protruding—orthognathic and brachycephalic.[368] The physical force averages less than that of the European, and there is decidedly less power of resisting disease.[369] The Jesuit Eder mentions a peculiarity among the Peruvian Arawaks, (Moxos, Baures). It is that the end of the little finger does not reach to the last joint of the third finger. The absence of this peculiarity he states will reveal a mixture of Spanish blood to the third generation.[370] It would be interesting to learn how widely this is noticeable.

The culture of the Arawak stock was generally somewhat above the stage of savagery. On the West Indian islands Columbus found them cultivating maize, potatoes, manioc, yams and cotton. They were the first to introduce to Europeans the wondrous art of tobacco smoking. They wove cotton into garments, and were skilful in polishing stone. They hammered the native gold into ornaments, carved curious masks of wood, blocked rude idols out of large stones, and hollowed the trunks of trees to construct what they called canoes.

Such is approximately the culture of the existing tribes of the stock. The Arawaks of Guiana also raise cassava and maize, though they depend largely on hunting and fishing. Like the northern tribes, they have well-developed gentile or totemic systems, with descent in the female line.[371] Marriages are by purchase, and the strange custom of the couvade obtains; that is, at the period of parturition the husband takes to his hammock, and is waited on as if he was the sick one. Their houses are usually single, not communal, and are furnished with swinging hammocks, mats, basket-work and pottery.

The Haytian mythology was quite extensive, and the legends of the Arawaks of Guiana have been collected, and are also rich. In all the tribes the dead were generally buried, and often the house of the deceased was destroyed or the spot deserted.

ARAWAK LINGUISTIC STOCK.

The Barés are now found along the banks of the Casaquiare and the Guainia, the Felipe, the Atabapo and some portions of the Rio Negro. They belong to the Arawak stock, their dialect being related to those of the Banivas and Maipures. About the middle of this century the traveller Richard Spruce found them in the regions assigned by Gilii to other tribes, indicating a displacement of the population. He collected a number of vocabularies, offering sufficient evidence in his opinion to establish the relationship of the following bands:[372]

BARÉ FAMILY OF THE ARAWAK STOCK.

To these I would add the Uirinas of the Rio Marari, on the strength of a vocabulary collected by Natterer.

4. The Caribs.

The Carib stock is one of the most extensively distributed in the southern continent. At the discovery its dialects were found on the Lesser Antilles, the Caribby Islands, and on the mainland from the mouth of the Essequibo River to the Gulf of Maracaibo. West of the latter it did not reach the coast, nor has any positive traces of its introduction above the straits of Panama earlier than the conquest been found, in spite of frequent assertions to the contrary. Inland from the Arawaks on the shore of Guiana are a number of Carib tribes, as the Macusi and Woyawoi, so numerous that this region has been thought by some to have been the original home of the stock; but the discovery by Dr. Karl von den Steinen of a tribe, the Bakairi, on the head-waters of the Schingu River, speaking a very pure form of the language,[373] and the recognition of the Carib affinities of the Palmellas on the Rio dos Baures, throw another light on the trend of Carib migrations, strongly supported by a series of other considerations. Thus, it has been satisfactorily shown by Im Thurn that the Caribs in Guiana wandered thither from the Orinoco district, some inland and some along the coast, and probably from the large islands adjacent to the coasts.[374]

These islands in turn were peopled from the mainland to the east, as I have already shown, their earlier population having been Arawak. All the Island, Orinoco and Guiana Caribs can thus be traced back to the mainland of northern Venezuela. In this vicinity was spoken the Cumanagoto dialect, in the province of Cumana or New Andalusia. According to the early missionaries, it was current along the coast for more than a hundred leagues, extending into the province of Caracas and beyond. The tribes who spoke it were the Chaymas, the Cores, the Cumanas, the Quacas, the Parias, the Palenques, the Varrigones, and others.[375] Other dialects to the west are the Opone and Carare, specimens of which were obtained by Lengerke in the vicinity of Bucaramanga, province of Santander.[376]

The sierra which divides the head-waters of the Caura from those of the Rio Branco and other streams flowing into the Rio Negro and Amazon, are peopled on both slopes by wandering tribes of the Carib stock. Near the sources of the Caura, Chaffanjon found the once formidable Guaharibos, now naked and wretched fugitives, fearing the white far more than they are feared by him.[377] On the southern slope, along the Rio Jauapery and neighboring streams, are bands of Crichanas, Ipurucotos (Purigotos), Macuchis, and Jauamerys (Waimiris), all speaking nearly related dialects of the Carib tongue. Dr. Barboza Rodrigues has given a touching picture of their recent struggles with the whites of the adjacent settlements, and the miserable condition to which they are reduced. We owe to the same sympathetic naturalist an interesting description of their customs and language.[378]

The hill tribes of French Guiana are known as Roucouyennes, from the roucou, a vegetable coloring matter with which they paint their skins. They exhale a peculiar odor like that of new leather, probably from the action of the tannin in the roucou on the skin. Naturally they are light in color, and at birth almost white.[379] Marriages of father and daughter, or brother and sister, are not rare among them.[380]

A connecting link between these Caribs of Guiana and the Bakairis of the south is supplied by the Apiacas of the Rio Tocantins, who speak a pure dialect of the stock, midway in character between those of the two extremes named.[381]

The Arubas, who occupied the island of that name off the coast of Venezuela, and whose mixed descendants now speak the Papamiento jargon, are no doubt correctly assigned to this stock by M. Pinart. They were skillful potters, and buried their dead in large urns. The numerous polychromatic petroglyphs they have left and their peculiar character are especially noteworthy.[382]

Sir Robert H. Schomburgk classifies the Carib stock in Guiana as follows, giving a short specimen of each dialect, which differ, he says, among themselves about as much as French and Italian.[383]

CARIB SUB-STOCK IN GUIANA.

The Guaques, who live on the head-waters of the Caqueta or Yapura river, have not been heretofore identified as Caribs; but their dialect, as collected by Presbyter Manuel P. Albis in 1853, leaves no doubt as to its relationship. He describes them as intelligent and kindly, but incorrigible and dexterous thieves, skillful in the collection of wax and the preparation of poisons. Nowhere is the couvade with its associate superstitions more rigidly observed. No woman must be seen by men during her catamenia, and at childbirth she must separate from the household for three months. During all that time her husband strictly observes a diet and seclusion.[384]

The lower Orinoco basin was for a long time the center of distribution of the stock; they probably had driven from it nations of Arawak lineage, some of whom, as the Goajiros, they pushed to the west, where they were in contact with the Carib Motilones,[385] and others to the islands and the shores to the east. The Carijonas and Guaques on the head-waters of the Yapura or Caqueta are now their most western hordes, and the Pimenteiras on the Rio Paruahyba are their most eastern. We can thus trace their scattered bands over thirty-five degrees of latitude and thirty of longitude. The earliest center of distribution which best satisfies all the conditions of the problem would be located in the Bolivian highlands, not remote from that I have assigned to the Arawaks.

The physical features of the Caribs assimilate closely to those of the Arawaks. They are taller in the average and more vigorous, but their skulls are equally brachycephalic and orthognathic. They are beardless, and have the same variability in color of skin. As good specimens of the modern Caribs we may take the tribes of Venezuela. These are spoken of as “the strongest, handsomest and most intelligent of any of the natives in northern South America.”[386] They are tall, straight and symmetrical, the women not less muscular than the men. The hair is sometimes slightly wavy, as Von den Steinen saw among the Bakairi.

The Caribs have had a bad reputation as to culture on account of their anthropophagous tendencies. Indeed, the word cannibal is a mispronunciation of their proper name, Karina. But they were quite on a par with their neighbors, the Arawaks, and in some respects superior to them. For instance, their canoes were larger and finer, and they had invented the device of the sail, which seems to have been unknown to all the other tribes on the continent. To some extent they were agricultural, and their pottery was of superior quality.

The beginnings of picture-writing were in use among them, and the remarkable rock inscriptions still visible on the Orinoco and the Essequibo are attributable to them, and were probably intended as conjurations to the supernatural powers, similar to others which remain in St. Vincent and other islands from the date of the Carib occupation.[387] Their family life was not usually communal, but each household occupied its own dwelling. In some parts, as in the deltas of the Essequibo and Orinoco, and even on the dry savannas, their huts were built on a substructure of piles which lifted them five or six feet from the ground or the water, as the case might be.

The religious rites they observed were often elaborate. Their principal divinities are said to have been the sun, moon and earth, the latter of which was spoken of as the mother of the race. They practiced the couvade, and their priests, called piaye, exercised unlimited power, and were correspondingly feared.

It was the opinion of Von Martius that the Carib, the Tupi-Guarani and the Arawak stocks are traceable to some very ancient common tongue. This view is at first sight strengthened by a wide comparison of vocabularies, but is weakened by an examination of the grammars of the three families, especially their pronominal elements. It is probable that the three ancestral tribes had early and close communication, but not original identity.

The seeming relationship has been rendered more prominent in certain instances by free later borrowings. M. Adam has shown that some of the northern dialects are in the condition of jargons, their grammar on the Carib model, their words drawn from various stocks. Such are the “Island Carib,” which is largely Arawak, and the Boni-Ouyana, described by Dr. Crévaux.[388]

CARIB LINGUISTIC STOCK.
5. The Cariris.

In his enumeration of the tribes of Central Brazil, Von Martius brings together a large number who once dwelt in the provinces of Bahia and Pernambuco, under the general title, “the Guck or Coco stem,” so called from the word which in many of them means “the paternal uncle.”[389] This division has not been endorsed by later research, and it is evident that Von Martius included several quite different stocks under this appellation.

Among these, the most prominent were the Cariris or Kiriri. They are now reduced to about 600 souls, but at one time were a powerful nation, and in 1699 the Jesuit Mamiani published a grammar and other works in their tongue.[390] They were among the more cultivated of the Brazilian tribes, given to agriculture, skilled in dyeing and weaving cotton, employing a primitive spindle and loom, with weapons of several kinds and of superior finish.

The Sabuyas, who dwell near them, speak a closely related dialect; but further affinities have not been verified. They have, indeed, many loan words from the Tupi, and some from the Carib stock, but the ground-work of these tongues is different. Von den Steinen offers some reasons for believing that they moved down the Amazon from a far western residence.[391]

6. The Coroados, Carajas and others.

The Coroados derive their name from the Portuguese word coroa, a crown, the term “crowned” being applied to several native tribes who wore their hair in a peculiar manner. It is not at all an ethnic designation, and I use it to bring into relief the need of some term of greater precision. Thus, there are the Coroados who are neighbors and linguistically related to the Puris, dwelling on the Paruahyba river. By some they have been included among the Tapuyas as alleged relatives of the Botocudos. But not only is there no relationship of language, but physically they are widely apart. The Puris-Coroados are a dark yellow brown, with mesocephalic heads, dark brown oblique eyes, large mouths and thick lips—nowise the type of the Botocudo. They are moreover agricultural in habits, and farther advanced in the arts.[392]

There are other Coroados in the extreme south of Brazil, in the province of Rio Grande do Sul, whither they are said to have wandered from the north. These do not appear to be Botocudos either. They have round heads, dark brown eyes, low foreheads, and are of a light coffee color. They are noticeable for their clean and ornamental huts, and for their skill in hunting, in which they employ arrows five feet in length, with bone points. They pray to certain stars as protective divinities, and like some northern tribes, clean and preserve the bones of the dead.[393]

The Carajas belong to a stock who dwell on the affluents of the river Araguay, in the province of Goyaz in southern Brazil. The traveler Castelnau[394] penetrated to them, and was our earliest source of information about them. They are wild and warlike, with a bad reputation among their neighbors. He was told they had no religion and no rites, but also that they were strictly monogamous and singularly firm moralists, punishing libertinage with the death of both parties; statements which do not accord. Their method of burial was curious. The corpse was interred in an upright position, the head out of the ground. An ample stock of bananas and other food was placed near it, and renewed from time to time. This clearly indicates a belief in life after death. The pure Carajas are markedly dolichocephalic.

The Caraja language is known too imperfectly to permit a proper study of its relationship. It is complex and difficult, and spoken differently by the men and the women. From the scant material at hand I perceive lexical relationship in some important words to the Tapuya stock,[395] but a wide divergence in phonetics and apparently in construction. Its members are as follows:

CARAJA LINGUISTIC STOCK.

A certain number of vocabularies have been obtained by travelers in Brazil from mixed-blood tribes, who spoke dialects sometimes compounded of several native tongues, sometimes of these mingled with Portuguese or negro elements. Such is the dialect of the Meniens, who lived in eastern Brazil near the Villa Belmonte, whose speech was a jargon of the Tapuya and negro languages; and that of the Cames in the interior of San Paulo, who also made use of a barbarous dialect, compounded of the African idioms of runaway slaves, and that of the Botocudos. The Catoquina, a specimen of which was obtained by Spix from a band on the affluent of the Jurua, and the Catoxa or Cotoxo of the Rio Parda, are other examples.[396]

7. The Orinoco Basin; Carib Sub-Stock; Salivas; Arawak Sub-Stock; Otomacos; Guamas; Guayoas; Garuoas; Guaraunos; Betoyas; Piaroas, etc.

The Llanos of Venezuela coincide with the former “Territory of Caqueta,” and embrace a region about forty thousand square miles in extent, covered either with grass and rushes or with dense forests. In the wet season it is a vast marsh, in the dry it is scorched by a burning sun, raising the thermometer daily to over 100° in the shade. Yet the Llanos are but a part of the vast upper water-shed of the northern affluents of the Amazon and those of the Orinoco, which together drain a country larger than the whole of France.

This wide expanse is thinly populated with bands of savages, gaining their subsistence chiefly from the rivers, few of them brought within the range of civilized influences. Linguistically the majority belong to the Arawak and the Carib stocks; but there are numbers of tribes whose affinities are uncertain, or who are apparently of quite another lineage. Scores of names are found in the records of the missions and on the pages of travelers, of peoples who have disappeared or are now known by other designations. Alexander von Humboldt named and located 186 tribes on the Orinoco and its affluents alone; but renounced as hopeless the attempt to give them a linguistic classification.[397] I shall not attempt to unravel the tangled ethnography of this region farther than to mention those tribes concerning whom specimens of language or the statements of European visitors permit a reasonable guess as to their affinities.

Something over a century ago, when Father Gilii wrote, largely from personal knowledge, his description of the tribes on the Orinoco and its affluents, he believed they could be included in nine linguistic stocks,[398] as follows:

1. The Carib in a number of dialects, as the Tamanaca, the Paiura, the Quiri-Quiripa, the Mapuya, the Guanero, the Guayquira, the Palenque, the Maquiritare, the Oje, the Mucuru, and others.

2. The Saliva, to which he assigned the dialects Ature, Piaroa and Quaqua.

3. The Maipure (Arawak), in its dialects Avane, Meepure, Cavere, Parene, Guipunave, and Chirupa.

4. The Otomaca, with one dialect, the Tarapita.

5. The Guama, with its dialect, the Quaquaro.

6. The Guayba, related to the Chiricoa.

7. The Jaruri (Yarura).

8. The Guaraunos.

9. The Aruaca.

This classification can stand as only approximately accurate, but it serves as an excellent starting point.

Beginning with the Carib stock, and basing my list on the works of Codazzi and more recent travelers, especially Crévaux, Coudreau and Chaffanjon, I offer the following as the tribes which may be definitely located as its members:

CARIB SUB-STOCK IN THE ORINOCO REGION.

Even when Codazzi collected his material, more than half a century ago, the once powerful Tamanacas had entirely disappeared, and no tribe of the name existed in the region.[399] The process of dissolution and destruction has gone on since his day with increasing rapidity, so that when Chaffanjon visited the Orinoco and Caura in 1884, he found that immense and fertile region almost uninhabited, the ancient tribes scattered and disappeared, or existing only in wretched remnants, misérables débris, of their former selves.[400] The opportunity is forever lost, therefore, to define the ethnography of this region by original observation, and we are thrown back on the collections and statements of former observers.

The Maquiritares, however, still remain as one of the handsomest peoples on the Orinoco, and remarkable for the skill with which they manufacture canoes sixty or seventy feet long from the trunk of a single tree.[401]

On the river Uaupes, an affluent of the Rio Negro M. Coudreau encountered various tribes, such as the Tarianos or Javis and the Nnehengatus, of whose tongues he obtained brief vocabularies. They indicate a distant influence of the Carib stock, especially the latter, but they seem mixed largely with elements from other sources.[402] They dwell adjacent to the Tucanos, to whom I have already referred as assigned by some to the Tapuyas. (See above, p. [240].)

Gilii’s second group, the Salivas, offers difficulties. There appears to be none of them under that name at present on the Orinoco. Chaffanjon states that the Atures have become extinct.[403] The Piaroas survive, but the tribe so-called to-day speak a tongue wholly unlike the Saliva, and unconnected, apparently, with any other stock;[404] and the modern Quaquas (Guagues) speak a dialect of the Arawak. Yet a hundred and fifty years ago the missionaries estimated the Salivas at four thousand souls. They lived principally on the river Cinareuco, below the Meta, and also on the Rio Etari, where they were in contact with the Carib Amarisanes. They are described as of a kindly and gentle disposition, well-made in body and willing scholars of their spiritual masters. In their heathendom they had the unique custom of disinterring the bones of their dead after the expiration of a year, burning them, and then collecting the ashes to mix with their drinking water.[405] Their language, which was vocalic and nasal, has been preserved in sufficient specimens to serve for comparison. According to Vergara y Vergara, it is still spoken on the banks of the Meta,[406] and Hartmann includes in those who employ it, the Quevacus and Maritzis, at the head of the Ventuari, and the Mayongcong on the Merevari.[407]

The Arawak stock, which Gilii calls the Maipure, had numerous branches in this region. They occupied much of the Orinoco in its middle and upper course, as well as the valleys of its affluents. Gumilla speaks of one of its members, the Caveres, as savage and inhuman warriors, but as the only nation which had been able to repulse the attacks of the down-river Caribs, who were accustomed to ascend the stream in fleets of eighty to a hundred canoes, destroying every village on its banks.[408]

The same authority mentions the Achaguas as possessing the most agreeable and cultured dialect, though he is in doubt whether it is strictly related to the Maipure. This nation, quite prominent in the older annals, still existed in the middle of this century to the number of five hundred on the Rio Muco. They were not civilized, and practiced the customs of polyandry and the destruction of female infants.[409] Cassani refers to them as on the river Ele, and describes them as tattooed and painted, with well-formed bodies and taking great pride in preserving and dressing their magnificent hair.[410]

From a variety of sources at my disposition I have prepared the following list of the

ARAWAK SUB-STOCK IN THE ORINOCO REGION.

The Otomacos remain, as Gilii placed them, an independent stock, with their single dialect, the Tarapita. The Jesuits first encountered them in 1732, amid the forests south of the Orinoco, between the Paos and the Jaruros. In later years they are described as a low grade of savages, given to the eating of earth. They are also said to be monogamous, and the women among them enjoy an unusual degree of consideration, being permitted to take equal part in the public games.[411] Their present locality appears to be on or near the river Meta.

The tribes whom Gilii mentions as the Guamas and Quaquaros lived on the banks of the Rio Apure, and in his day had the reputation of “a numerous and valorous people.”[412] They were not unacquainted with some of the arts, and were particularly skillful in the manufacture of small figures in terra cotta, many of which are to be picked up on the sites of their ancient villages. Now, however, they have been smitten with the fate of their race, and are reduced to a few miserable vagrants, destined to disappear wholly in a few years. Their arts are lost, and the oppression of the whites has driven from them all hopes of bettering their condition.[413]

Of their language I have no specimens. According to Felipe Perez, it is related to the Omagua, and hence should be included in the Tupi stock; but this writer is not always dependable.

The Guaybas (Guahibos) and Chiricoas dwelt originally on the broad plains between the Casanare and Meta rivers; but a number of them were converted in the latter half of the seventeenth century and persuaded to come to the missions. They soon returned to their roving life. Cassani speaks of them as of mild and friendly disposition, but incorrigible vagabonds, “the gypsies of the Indies,” constantly migrating from place to place.[414] They have never lost their love of the wilderness, and it has been their salvation, for they still survive—quite a numerous people—on the left bank of the Orinoco, from the Rio Meta to the Vichada. They are rebellious to all attempts at civilization, and the white man is not safe who ventures into their territory.[415]

Humboldt, in his discussion of the tribes of the Orinoco, refers to the Guahibos as white in color, and founds some speculations on this fact. Their hue is indeed light, at times what may fairly be called a dirty white; but in this respect we are assured by recent and competent authority they do not differ from their neighbors, the Maquiritares and Piaroas. It is not a question of descent, but of climatic surroundings and mode of life.[416]

The home of the Jaruris, Yaruras, or, as they called themselves, Japurin, was on and near the Orinoco, between the rivers Meta and Capanapaco. They depended on hunting and fishing, and were indolent and averse to agriculture. They had few arts, but were friendly in disposition, not given to drunkenness, and usually monogamous. At present they number scarcely a hundred individuals, badly formed, afflicted with contagious disease, and rapidly on the road to extinction. They have lost their trait of sobriety, and a man will readily offer his wife or daughter in exchange for a bottle of brandy. (Chaffanjon.)

The Guaraunos, called by the English Warraus, continue to live in considerable numbers—some say about fifteen thousand—in and near the delta of the Orinoco. They are a thrifty, healthy people, building their houses ingeniously upon piles to protect them from the periodical overflows of the stream. This method of construction, however, was adopted only when they sought as refuge marshy and lonely spots to escape their enemies. Contrary to the statements of most travelers, those who know them best report them as preferring dry uplands, where they make clearings, plantations and houses with singular industry and skill. The favorite wood used in such construction is the temiche (not the moriche) palm, which they call, from its magnificent fronds, “the feathers of the sun,” ya juji.[417]

Humboldt placed their number at the beginning of the century at about six thousand, which is doubtless more correct than the later estimates. He adds that the Guayquiries, who inhabited the peninsula of Araya and the adjacent islands of Margarita, “admit the relationship of their language with that of the Guaraunos.”[418] At the beginning of the last century Gumilla found them living on the south bank of the Orinoco in a most wretched condition and nearly annihilated by their merciless enemies, the Caribs. It is probable, therefore, that they removed from that location to the coast.[419] No other dialect of the tongue, so far as I know, has been discovered, and it seems an independent stock.

In appearance they are dark in hue, of muscular build, hair black, abundant and very fine, noses straight and well-shaped, skull brachycephalic, stature below medium.

The Aruaca mentioned by Gilii were some tribes of the Arawaks who occasionally visited the southern bank of the Orinoco, and whose relations to the Maypures were not known to him. They are also mentioned by other authors.

Having thus reviewed the linguistic stocks named by Gilii, I shall proceed to mention some which escaped his attention.

One of the most interesting of these is the Betoi, or Betoya. This tongue derived its name from a tribe dwelling at the foot of the mountains of Bogota, between the rivers Apure and Tame, and are therefore included by some among the Indians of New Granada. From a number of authorities I find the following members are attributed to the

BETOYA LINGUISTIC STOCK.

Of these, the Piojes and Correguages, of which we have vocabularies, do not show close resemblance to the Betoya, yet undoubtedly some;[420] so I place them in this stock partly in deference to old authorities.

The Piojes derive their name from the particle of negation in their language, this being their usual reply to all inquiries by traders or travelers. They are divided into two bands, speaking the same dialect, one on the Napo and one on the Putumayo, neither knowing anything of the other. Some of their customs are peculiar. For instance, it is their rule that a widow shall take her son, a widower his daughter, to replace the deceased consort.[421] They are somewhat agricultural, and are skillful boatmen.

The Tamas formerly lived on the river Aguarico (Coleti). Dr. Crévaux found them on the Caguo, a branch of the Yapura, and obtained from them a short vocabulary, but enough to mark them as members of the stock.[422] There are also some on the Rio Meta who speak Spanish only. (Perez.)

The Betoya has impressed me as showing some distant affinity to the Choco stock, and it may be that ampler resources on both sides would lead to the establishment of an original identity. The following words from the very scanty number which I have for comparison are noteworthy:

CHOCO STOCK.BETOYA STOCK.
Man,uma-china,uma-soi, emi-ud.
Woman,uerá,ro.
Fire,tŭjoor,toa-tui.
Ear,juru,ca-joro-so.
Nose,jun,ju-saca, jin-quepui.

The Choco do, river, seems related to the Betoya ocu-du, water.

The Macaguages are industrious and agricultural. Both sexes dress alike in cotton tunics dyed in violet color, and suspend bright feathers and strings of beads in ears, nose and lips.[423]

A singular question has arisen as to the relationship of the Betoya and the Yarura languages. Their near connection was affirmed by the early missionaries. In fact, the history of the conversion of the Betoyas turns upon the identity of the two tongues. It was brought about in 1701 by a Yarura Indian, a convert to Christianity, who accidentally discovered that he was understood by the Betoyas.

In spite of this detail, it is evident from an inspection of the vocabularies, that there is absolutely no relationship between the two idioms. I can only explain the contradiction as arising from some ambiguity or similarity of names. The two tribes lived together in the time of Gumilla, making up about three thousand souls.[424]

About the middle of this century some six hundred of the Betoyas dwelt on the head-waters of the river Manacacia.[425]

In the territory of St. Martin, above the falls of the Guaviare and along the Rio Guejar and the Meta, are several tribes asserted to speak related dialects, but of which I have little information. The principal one is that of the Churoyas, of whom Professor Nicolas Saenz has given an interesting sketch and a short vocabulary.[426] They are very ugly, with broad faces, low foreheads, small and oblique eyes, and in color like dried tobacco. Nudity is their usual garb, and the skin is decorated with tattoo marks instead of clothing. According to Perez they number about 1200.[427] Following him and other authorities, I may enumerate the following members of the

CHUROYA LINGUISTIC STOCK.

Whether the Cofanes here named are those of the Province of Quitu who murdered the Jesuit missionary, Raphael Ferrer, in 1602, I have not discovered. Perez describes them as still warlike and seclusive, living in the terminal hills of the Cordillera, and avoiding traffic with the tribes of the lower river.[428]

An examination of the vocabulary furnished by Saenz inclines me to think that the Churoya may be a mongrel dialect, or at least has borrowed freely from neighboring stocks. I subjoin the principal words from his short vocabulary, with some comparisons:

Sun,mshojaint.
Moon,juimit (oamito, Guahiba).
Fire,hijit (chichi, = sun, Carib).
Water,minta.
Bow,piranso (paria, Roucouyenne).
Arrow,funait.
Tobacco,joo.
Plantain,parasa.
Dog,uilg.
Tortoise,ainjachie.
Wind,che.
Skin,begt.

The Piaroas are mentioned by Gilii as a branch of the Salivas, but their language reveals no such connection. They are still found on both banks of the Orinoco above the confluence of the Vichada and near the mouth of the Mataveni. They are savage and superstitious, avoiding contact with the whites; they have had good reason to be extremely distrustful of the advances of their civilized neighbors. They are much given to nocturnal ceremonies, and entertain a great respect for the tapir, who is their reputed ancestor, and also the form which is taken by the souls of the departed.[429]

The Puinavis dwell on the Inirida, an affluent of the Guaviare. A tribe, the Guipunabis, is mentioned by Gilii as belonging to the Maipure (Arawak) stock; but it cannot be the same with the one under consideration, the language of which appears to be without affinities. Latham identified them with the Poignavis of the older writers, and on slight linguistic evidence, believed them connected with the Banivas.[430] My own comparisons do not justify this opinion.

8. The Upper Amazonian Basin.

No portion of the linguistic field of South America offers greater confusion than that of the western Amazonian region. The statements are so conflicting, and the tribal changes apparently so rapid, that we are at a loss to bring modern observations into accord with older statements. Thus, I am entirely unable to accept the linguistic classification of Hervas, which certainly was based on the best information of his day. As a matter of comparison I give it.

List of Languages in the Governments of Maynas and the Marañon (Hervas).

STOCKS.DIALECTS.
1.Andoa.Araro.
Chuudaviño.
Gae.
Guazago.
Murato.
Pabo.
Pinche.
Simigae.
Bobonazo.
2.Campa.Amjemhuaco.
Curano.
Manua.
Nanerua.
Nesahuaco.
Sepaunabo.
Tasio.
3.Chayavita.Cahuapano.
Paranaruro.
4.Comaba.Ginua.
Inuaco.
Ruanababo.
Zepo.
5.Cuniba.Manamabobo.
Mananamabua.
6.Encabellada.Guajoyo.
Guencoyo.
Neocoyo.
Zaparro, or Encabellado.
Ziecoyo.
7.Iebera.Tiputini.
Tibilo.
8.Maina.Chapo.
Coronado.
Humurano.
Roamaino.
9.Muniche.Muchimo.
Otanabe.
10.Pana.Iltipo.
Pelada.
11.Pira.Cusitinavo.
Manatinavo.
Upatarinavo.
12.Simigae.Arazo.
Ijinori.
Nevo.
Oa.
Zaparro.
13.Lucumbia.Putumayo.
Yete.
Ceoqueyo.
14.Urarina.Barbudo.
Itucale.
Mayoruño.
Musimo.
15.Yamea.Amaono.
Nahuapo.
Napeano.
Masamae.
16.Jinori.Acamaori.
Camacori.
Iqueconejori.
Panajori.
Tremojori.

A slight examination of this classification suffices to reveal its general inaccuracy. The Zaparos are included in both the Encabellada and the Simigae stocks. The latter is given both as a stock and as a dialect of the Andoa. In fact, all three of the stocks named belong together as dialects of one. The Pano stock, as we now know it, appears scattered under Cuniba, Urarina and Pana; and the arrangement is incorrect in many other points. While it has a value in preserving the names of some now missing tribes, as a linguistic scheme it is wholly unsafe.

The Zaparos constitute one of the most extended and numerous nations in the upper valley of the Amazon. They dwell near or adjacent to the Jivaros on the south, and as their name is variously spelled Zaparos, Xeberos and Jeberos, they have at times been confounded with them. They differ, however, not only in language, but in appearance and temperament. The Zaparos are lighter in color, smaller in stature, with oblique eyes, large mouths, and expanded nostrils.[431] Their disposition is indolent and easy tempered, and their abilities inferior. This is seen in the construction of their houses and the appearance of their fields, which do not compare advantageously with those of the Jivaros; but they display some ingenuity in manufacturing clothing from the bark of a species of Ochroma, and they are skillful in concocting the urara poison, in making blow-pipes, and are daring boatmen.

In 1632 they lived near the Omaguas, on the river Curary, and their number was estimated by the missionaries at 10,000. At present their main body dwells between the rivers Pastaza and Napo and along the Marañon between the rivers Zamora and Morona. In 1850 Osculati estimated their number at over 20,000, which is certainly in excess of their present representatives. The many small tribes into which they are divided, and the confused orthography of the names applied to them, render it difficult to offer a satisfactory list. It seems tolerably certain that the ancient “Andoas” were the Zaparos of the upper Pastaza,[432] and equally sure that the Encabellada, the Iebera, the Simigae and the Jinori languages, all supposed by Hervas to be independent stocks,[433] were spoken by members of the Zaparo family. The Iquitos are another populous branch, sometimes supposed to be distinct.

The Zaparo language is agreeable to the ear, partaking of the phonetic character of the Brazilian idioms. The Italian traveler, Osculati, has furnished a very satisfactory account of it, both grammatical and lexicographical,[434] and there are vocabularies by other voyagers.

I offer the following alphabetic list of the sub-tribes of the Zaparos, without attempting to define their several positions in the general district referred to:

ZAPARO LINGUISTIC STOCK.

On the mountain slope of the Cordillera, north of the Zaparos and east of the Cañaris, are the Jivaros (Givaros, Xivaros), a wild, warlike tribe, never subjugated either by the Kechuas or the Spaniards. Their homes are about the head-waters of the rivers Pastaza, Santiago, and other affluents of the Marañon. They are rather tall, of light color, with thin lips, aquiline noses, straight eyes, prognathic jaws, hair black or with a reddish tinge.

Some say their various bands number as many as four hundred, named from the streams on which they live. Most of them depend upon hunting and fishing, others pursue agriculture and breed pigs. Their weapons are the sarbacane, the lance, the bow and the shield. They have developed a system of sound-signalling or telegraphy by means of large wooden drums placed at certain distances apart, by beating on which in a peculiar manner the advent of an enemy, his number and direction, can be heralded over hundreds of square miles in a few hours. The Jivaros are celebrated for the preparation of human heads by a process of boiling and drying so as to preserve the hair and soft parts. Many of these trophy heads have been brought to Europe, and their purpose has led to some discussion. It appears that they are prepared both as trophies of victory and out of reverence to departed chiefs.[435] Their houses are built solidly of wood, with wooden doors. They sleep upon wooden frames, and construct tools of the same material.[436]

The principal event in their history was their revolt against the Spanish authorities in the year 1599. They destroyed many settlements and the entire city of Logroño, carrying the women into captivity. Many of them had already been converted to Christianity, and their rites are said still to preserve some reminiscences of such teachings. In recent years many of them have been civilized through the efforts of Italian missionaries.

The language of this important nation, although early studied, has as yet no printed literature. I have found of it only the first five numerals, which do not seem to have connection with any other tongue. They are as follows:

1. Alza; 2. catuta; 3. kala; 4. ingatu; 5. aleyticlon.[437]

From a study of proper names and ethnographic traits, Dr. Hamy has expressed himself with great assurance that the Jivaros belong to the Guarani group of the Tupi stock;[438] but the above numerals do not indicate such relationship, nor do I think that his other arguments establish it. For the present they must be considered an independent stock.

JIVARO LINGUISTIC STOCK.

The eastern neighbors of the Jivaros are the scattered bands of the Maynas, separated by Hervas into two stocks, the Maina and the Chayavita, but so far as I can learn, without sufficient reason. The language is or was spoken at the mission of the Conception on the upper Marañon and in the uplands around Cerros de Mainas. It is singularly harsh and difficult. The natives were wild, and lived by hunting and fishing. Their earlier home was on the upper waters of the Morona and Pastaza rivers.

The following bands are embraced in the

MAINA LINGUISTIC STOCK.[439]

On the Rio Javary there seem to be several independent stocks. One of these is that of the Yameos, who are found in the lower course of the river and also further up the Marañon, near Nauta, and on the Huallaga, where they are called Llameos, Yameos, Lamas, or Lamistas. Formerly they were a numerous and warlike nation, sharply divided into gentile organizations, and carefully refraining from intermarriage in the gens. At an early date we hear of them between the Rio Tigre and the Napo. (Markham.)

The following sub-tribes are stated by various writers to belong to the

LAMA LINGUISTIC STOCK.

Pöppig describes them as agricultural and industrious, and much given to trade and travel.

In appearance, they are small, dirty and Mongoloid, sharply contrasting with the Indians of the Huallaga, who are all tall, strong and well built, with good features.[440]

In conformity to old authorities, Markham classes the Ardas as a sub-tribe of the Yameos. Their home was between the rivers Napo and Masso. On the latter they were in immediate contiguity to the Massamaes (Coleti). There has been published a Doctrina in their language, from which the Lord’s Prayer is quoted by Ludewig.[441] This version has no resemblance to the Pater in Yamea contained in the Mithridates; so for the present I leave the Ardas unclassified.

Higher up the river Javary are a number of tribes speaking related dialects of what I shall call the Peba stock, though there are some reasons to consider it a corrupt dialect of the Omagua, and hence related to the Tupi.

PEBA LINGUISTIC STOCK.

To this list I add the Yeguas, Yaguas or Yahuas, found in the same vicinity, and remarkable for their fine personal appearance, “the most perfect physical type,” says M. Ordinaire, “of all the Indian races.”[442] The vocabulary of their language obtained by Castelnau shows unmistakable affinities to that of the Pebas.[443]

On the Rio Chambira, adjacent to the Yameos and Omaguas, dwelt in the early part of the last century the Itucales and Varinas or Uarunas, who, according to Coleti, spoke allied dialects. The Itucales were noteworthy as the aptest and most biddable converts obtained by the missionaries on the river. They were agricultural and monogamous.[444] Hervas classes them with the Musimos, the Mayorunas and the Barbudos, under the Urarina language; but the last two are members of the Pano stock.

The Ticunas (Tecunas, Tucunas) are found along the lower Javary and the Solimoes, adjacent to the Pebas. They wander about in a state of nakedness, depending on hunting and fishing, and under a loose control of the Brazilian government. Many of them can converse in Kechua, though their own tongue is of a different group. They are given to dances of a sacred character, in which the actors appear in masks. An operation allied to circumcision is practiced on infants of both sexes at the time of assigning them names.[445] One of the several tribes called “Orejones” is thought by Pöppig to belong to the Ticunas.[446]

The tribes in the valley of the Huallaga were first visited by Franciscan missionaries in 1676. In that year Father Jose de Araujo converted a number of the Hibitos (Xibitos) in the Upper Huallaga, and wrote an arte of their language. He found it the same as that of the Chunchos in the Sierra. Their neighbors further down the river, the Cholones, speaking a different idiom, were brought under the instruction of Father Francisco Gutierrez, who composed a work on their tongue. A century later we find these two nations living together at the mission, counting 4800 souls, and occupying that portion of the province of Cajamarquilla between 7° and 8° 30´ s. lat. They were peaceable and agricultural, with fields of cotton and food plants.[447]

This fair scene disappeared in the turbulent life of the next generation, and when the traveler Pöppig visited the Huallaga in 1834 he found the mission in decay, and the natives, much reduced in numbers, had resumed their wild life and again become savages.[448] At present, along the main stream to the north, are the Cocamillas, the Aguanteas, and the Puinahuas. All these appear to be of the Tupi stock, with dialects akin to the Cocama and Omagua.[449]

The Panos. When the missionaries first crossed the Cordillera and explored the upper Ucayali river, they found a number of related tribes, the principal of whom were the Panos. By their traditions they had moved from near the equator at the north. They differed little in culture from their neighbors, and are now nearly extinct. By the earlier writers they were placed in relation to the Omaguas as members of the Tupi stock,[450] but the researches of M. Raoul de la Grasserie have vindicated for them an independent position.[451] They are said to have possessed a form of hieroglyphic writing, which they painted on a sort of paper manufactured from vegetable fibre.

Some of the Mayorunas are reported as having thick beards and white skins (Martius), but these peculiarities are probably attributable to early admixtures with the white race.

The largest of these tribes at present is that of the Conibos, who constitute now the greater part of the natives the traveler encounters on the Ucayali. In appearance they have some resemblance to the Peruvians. The nose is aquiline and prominent, the forehead broad, the eye large, and the cheek bones not prominent. In intelligence they are superior to their neighbors, learning the Spanish language readily, and proving themselves valuable house-servants. They are apathetic, however, and none of the Panos have shown any earnest desire to adopt a civilized life.[452]

The Cashibos are the most savage tribe on the Ucayali or its affluents, and are said to have the ugly custom of eating their relations when they die, and if this event is long delayed, the old men are killed. But such is the power of ideas, that one of the obstacles to their conversion is that they so much prefer their bodies to become food for their relatives than a feast for worms![453]

The Pacaguaras or Pacavaras, on the rivers Beni and Mamore, classed by D’Orbigny as a separate stock, belong among the Panos, as is clearly seen by the vocabulary furnished by that traveler, and later that by Mr. Heath.[454] The easternmost branch of the stock (not noted by M. de la Grasserie), are the Canawarys (Canamarys), who live on the banks of the Purus. Mr. Chandless heard that they were related to the Conibos, and the few words he obtained of their language prove the statement correct.[455]

PANO.PACAGUARA.CANAWARY.
Sun,bari,uari,wari.
Fire,chi,chi-i,chi-i.
Water,uaca,waka.

Mr. Chandless also says, “The Conibos are of the same tribe as the Manitenerys of the river Purus,” which would bring these latter also into the Pano stock. The short vocabulary of their language which he supplies does not bear out this assertion. Mr. Richard Spruce considered that it proved them to be of the Carib stock;[456] but to me it seems unmistakably a member of the Arawak family, as will be seen from the following analysis:

MANITENERY.ARAWAK STOCK.
Sun,cashi,catche.
Moon,siri,casiri.
Fire,chi-chi,chichi.
Water,huni,uni.

From the above considerations I offer the following names as comprising the

PANO LINGUISTIC STOCK.

Mr. Chandless[460] met on the rivers Purus and Jurua tribes of a stock whose tongue I have not been able to connect with any other. They are represented on the former stream by the Pammanas or Pammarys (pama-ouiri, eaters of the pama, a kind of berry), or Puru-purus (piru-poru, name of a skin disease which prevails there), whose name has been transferred to the river. These are believed by Martius to be the same or allied to the Pamas, a tribe who formerly lived on the Madeira, but were driven thence by the Caripunas.[461] On the Jurua are the apparently related Arauas and Araicus. All these depend on hunting and fishing, and are of migratory habits. Some of the Pammanas are reported as light in hue, with blue eyes and brown hair.[462]

Many tribes with names differing from the above are recorded by the older writers as resident on these rivers, but owing to the absence of linguistic material, no identification is possible.

The close relationship of the Pammarys of the Purus and the Arauas of the Jurua is shown by the following comparison:

PAMMARY.ARAUA.
Moon,massicu,massicu.
Fire,si ju,sihu.
Water,paha,paha.
Dog,djuimahi,jumayhi.

So far as known, I would place the following tribes in the

ARAUA LINGUISTIC STOCK.

The jargon of the Yaguas, on the Amazon between Nauta and Pebas, seems to have borrowed from this stock; as:

YAGUA.PAMMARY.
Sun,ini,saf-iny.
Water,haha,paha.

The neighbors of the Arauas on the river Purus are the Hypurinas (better Jupurinas) of whose language Mr. Chandless also supplies a short vocabulary. It contains a few words in common with the Pammary, but probably only borrowed by both from the Arawak. The following will illustrate the two tongues:

PAMMARY.JUPURINA.
Sun,safiny,atocanti.
Moon,massicu,cassiri.
Fire,siju,chamina.
Water,paha,iborahai.
River,wainy,weni.
Dog,djuimahi,anguity.
Tortoise,ú-jurú,chetuyu.
Tapir,dama,chama.

The Hypurinas on the Rio Acre (or Aquiri) belong to the same tribe. They are said to be related to the Chacobos and the Piros of the Ucayali. They are without civilization. The women go naked, but the men wear long purple robes, and both sexes pierce the lips and nose. Some agriculture is carried on, but hunting and fishing are the main sources of the food supply.[463]

The total number of natives on the Purus and its affluents was estimated by Colonel Labré, in 1885, at 40,000, “speaking forty or more different languages;” but this last assertion we may take with large allowance. Probably not over four or five stocks are represented. The same explorer names nine tribes visited by him on the river Ituxy. They are the: 1, Caccharari; 2, Canamary; 3, Catauxi; 4, Guarayo; 5, Huatanary; 6, Hypurina; 7, Hyuma; 8, Pamana; and 9, Pammary tribes.[464]

In this list, as elsewhere, the term Guarayos has no ethnic significance. It is a Tupi word applied in this Spanish form to various wild, uncivilized tribes.

9. The Bolivian Highlands: the Chiquitos, Yurucares, Mosetenas, Tacanas, Samucus, Canichanas and others.

On the Atlantic face of the Cordillera, in the easternmost portion of Bolivia, where the head-waters of the Madeira are known by the names of the Mamore, the Guapai and the Beni, there is an astonishing variety of linguistic stocks. It would seem that the broken remnants of many diverse nations had sought refuge in the deep vales and dense forests of this region.

We have already seen that the Caribs were represented here by the Palmellas, and the Arawaks by the Moxos and Baures. South of the Moxos was the extensive region of the Chiquitos, stretching between south latitude 16° and 18°, and from the upper affluents of the Paraguay river to the summit of the Cordillera. On the south it adjoined the Gran Chaco, and on the west the territory of the Kechuas. They were a medium-sized, mild-mannered people, mostly of little culture, depending on the chase for food, but willingly adopting the agricultural life recommended to them by the missionaries. They were divided into a vast number of small roving bands, the most important group of which were the Manacicas, whose homes were near Lake Xaray, about the head-waters of the Paraguay. Their myths relating to a male and female deity and their son reminded the Jesuits of the Christian Trinity.[465] The Manacicas were agriculturists and remarkably skilful potters. The villages they constructed were surrounded with palisades and divided by broad streets. The corpses of the dead were deposited in underground vaults, and both property and rank passed in the male line to the sons of the deceased.

The Chiquito language is interesting for its scope and flexibility, being chiefly made up of generic particles capable of indefinite combination.[466] It is singular in having no numerals, not even as far as three. Its four principal dialects were those of the Taos, the Piñocos, the Manacicas and the Penoquies.[467] It was selected by the missionaries as the medium of instruction for a number of the neighboring tribes.

Of such tribes there were many, widely different in speech, manners and appearance from the Chiquitos. Some of them are particularly noteworthy for their un-Indian type. Thus, to the west of the Chiquitos, on the banks of the rivers Mamore and Chavari, were the Yurucares, the Tacanas and the Mosetenas, all neighbors, and though not of one tongue, yet alike in possessing a singularly white skin and fine features. Their color is as light and as really white as many southern Europeans, the face is oval, the nose straight, fine, and often aquiline, the lips thin, the cheek-bones not prominent, the eyes small, dark and horizontal, the expression free and noble. They are of pure blood, and the most important tribe of them derived their name, Yurucares, white men, from their Kechua neighbors before the conquest. They are usually uncommonly tall (1.75), bold warriors, lovers of freedom and given to a hunting life. The women are often even taller and handsomer than the men.

The traveler D’Orbigny suggested that this light color arose from their residence under the shade of dense forests in a hot and humid atmosphere. He observed that many of them had large patches of albinism on their persons.[468]

The branches of these stocks may be classed as follows:

YURUCARI LINGUISTIC STOCK.
MOSETENA LINGUISTIC STOCK.

The Toromonas occupy the tract between the Madre de Dios and the Madidi, from 12° to 13° south latitude. According to D’Orbigny they are, together with the Atenes, Cavinas, Tumupasas and Isuiamas, members of one stock, speaking dialects of the Tacana language. He was unable to procure a vocabulary of it, and only learned that it was exceedingly guttural and harsh.[469] From their position and their Kechua name (tuyu), low or swamp land, I am inclined to identify the Toromonas with the Tuyumiris or Pukapakaris, who are stated formerly to have dwelt on the Madre de Dios and east of the Rio Urubamba, and to have been driven thence by the Sirineris (Tschudi).

According to recent authorities the Cavinas speak the same tongue as the Araunas on the Madre de Dios, which are separated from the Pacaguaras by the small river Genichiquia;[470] and as the language of the Toromonas is called in the earlier accounts of the missions Macarani, I may make out the following list of the members of the

TACANA LINGUISTIC STOCK.

The Araunas are savage, and according to Heath “cannibals beyond a doubt.” He describes them as “gaunt, ugly, and ill formed,” wearing the hair long and going naked.[471] Colonel Labré, however, who visited several of their villages in 1885, found them sedentary and agricultural, with temples and idols, the latter being geometrical figures of polished wood and stone. Women were considered impure, were not allowed to know even the names of the gods, and were excluded from religious rites.[472] The Cavinas, on the other hand, are described by early writers as constructing houses of stone.[473] The Maropas, on the east side of the river Beni near the little town of Reyes, speak a dialect of Tacana as close to it as Portuguese to Spanish. They are erroneously classed as a distinct nation by D’Orbigny, who obtained only a few words of their tongue. The Sapiboconas, who lived at the Moxos Mission, and of whose dialect Hervas supplies a vocabulary, are also a near branch of the stock. We now have sufficient material to bring these tribes into relation. With them I locate the Lecos, the tribe who occupied the mission of Aten, and are therefore called also Atenianos.[474] At present some civilized Lecos live at the mission of Guanay, between the Beni and Titicaca; but we have nothing of their language.[475]

The Tacana dialects present a number of verbal analogies to Kechua and Aymara; so many in fact that they testify to long inter-communion between the stocks, though I think not to a radical identity. I present a few:

TACANA.KECHUA.
Man,reanci,runa.
Water,jene,una.
Hand,ma,maqui.
Foot,quatri,chaqui.
House,etai,uta (Aymara).
Stone,tumu,rumi.
Star,emata,matti.
Lightning,ilapa,illapa.
Year,mara,mara.
Three,quimisha,quimsa.
Four,puschi,pusi (Aymara).
Five,pischica,pichka.

The numerals above “two” have clearly been borrowed from the Kechua-Aymara.

There are also a large number of verbal coincidences between the Tacana and the Pano groups, but not enough to allow us to suppose an original unity.

The Samucus (Zamucas) embraced a number of sub-tribes dwelling on the northern border of the Chaco, between 18° and 20° south latitude, and about the river Oxuquis. They did not resemble the Chaco stocks, as they were not vagrant hunters, but dwelt in fixed villages, and pursued an agricultural life.[476] Their language was singularly sweet in sound, and was called by D’Orbigny “the Italian of the forest.” They included the following members:

SAMUCU LINGUISTIC STOCK.

Among these the Morotocos are said to have offered the rare spectacle of a primitive gynocracy. The women ruled the tribe, and obliged the men to perform the drudgery of house-work. The latter were by no means weaklings, but tall and robust, and daring tiger-hunters. The married women refused to have more than two children, and did others come they were strangled.

On the river Mamore, between 13° and 14° of south latitude, were the numerous villages of the Canichanas or Canisianas. They were unusually dark in complexion and ugly of features; nor did this unprepossessing exterior belie their habits or temperament. They were morose, quarrelsome, tricky and brutal cannibals, preferring theft to agriculture, and prone to drunkenness; but ingenious and not deficient in warlike arts, constructing strong fortifications around their villages, from which they would sally forth to harass and plunder their peaceable neighbors. By a singular anomaly, this unpromising tribe became willing converts to the teachings of the Jesuits, and of their own accord gathered into large villages in order to secure the presence of a missionary.[477] Their language has no known affinities. It is musical, with strong consonantal sounds, and like some of the northern tongues, makes a distinction between animate and inanimate objects, or those so considered.[478]

Between 13° and 14° of south latitude, on the west bank of the Rio Mamore, were the Cayubabas or Cayuvavas, speaking a language without known affinities, though containing words from a number of contiguous tongues.[479] The men are tall and robust, with regular features and a pleasant expression. The missionaries found no difficulty in bringing them into the fold, but they obstinately retained some of their curious ancient superstitions, as, for instance, that a man should do no kind of work while his wife had her monthly illness; and should she die, he would undertake no enterprise of importance so long as he remained a widower.[480]

Brief notices will suffice of the various other tribes, many of them now extinct, who centered around the missions of the Chiquitos and Moxos early in this century.

The Apolistas took their name from the river Apolo, an affluent of the Beni, about south latitude 15°. They were contiguous to the Aymaras, and had some physical resemblance to them. From their position, I suspect they belong in the Tacana group.

The Chapacuras, or more properly Tapacuras, were on the Rio Blanco or Baures in the province of Moxos. They called themselves Huachis, and the Quitemocas are mentioned as one of their sub-tribes. Von Martius thinks they were connected with the Guaches of Paraguay, a mixed tribe allied to the Guaycuru stock of the Chaco. The resemblance is very slight.

The Covarecas were a small band at the mission of Santa Anna, about south latitude 17°. Their language was practically extinct in 1831.

The Curaves and the Curuminacas, the former on the Rio Tucubaca and the latter north of them near the Brazil line, were said to have independent languages; but both were extinct at the time of D’Orbigny’s visit in 1831. The same was true of the Corabecas and Curucanecas.

The Ites or Itenes were upon the river Iten, an affluent of the Mamore about 12° south latitude. They were sometimes improperly called Guarayos, a term which, like Guaycurus, Aucas, Yumbos and others, was frequently applied in a generic sense by the Spanish Americans to any native tribe who continued to live in a savage condition.

The Movimas (Mobimas) occupied the shores of the Rio Yacuma, and Rio Mamore about 14° south latitude. In character and appearance they were similar to the Moxos, but of finer physique, “seldom ever under six feet,” says Mr. Heath. They are now civilized, and very cleanly in their habits. The vocabularies of their language show but faint resemblances with any other.

The Otuquis, who in 1831 did not number over 150 persons, lived in the northeast part of the province of Chiquitos near the Brazilian line. Their language was nearly extinct at that time. The short vocabulary of it preserved by D’Orbigny does not disclose connections with other stocks, unless it be a distant affinity with the Tacana group. This may be illustrated by the following words:

OTUQUI.TACANA DIALECTS.
Man,vuani,reanci.
Woman,vuaneti,anu.
Sun,neri,ireti.
Moon,ari,bari.
Water,uru,yuvi.
Head,ikitao,ekuya.

It was the policy of the Jesuits in their missions in this district to gather the tribes from the forest and mountain into permanent settlements, and reduce as far as possible the number of languages and dialects, so as to facilitate instruction in religious teaching. Shortly after this Order was expelled from their missions (1767), an official report on their “reductions” was printed in Peru, giving a list of the tribes at each station, and the languages in use for instruction.[481] From this scarce work I extract a few interesting particulars.

The province of Apolobamba is described as extending about eighty leagues northeast-southwest, east of the Cordillera, and west of the Rio Beni. The languages adopted in it were the Leca, spoken by the Lecos Indians at the mission of Aten, and the Maracani, at the mission of Tumupasa, on the Rio Beni. Forty-nine nations are named as belonging to the mission of the Chiquitos, each of whom is stated to have spoken a different language or dialect, though all were instructed in their religious duties in Chiquito. At the mission of Moxos twenty-nine tribes are named as in attendance, but it had not been found possible, such was the difference of their speech, to manage with less than nine languages, to wit, the Moxa, the Baure, the Mure, the Mobima, the Ocorona, the Cayubaba, the Itonama and the Maracani.[482]

Of these tongues I have classed the Leca and Maracani as dialects of the Takana, not from comparison of vocabularies, for I have seen none of either, but from the locations of the tribes speaking them. The Moxa and Baure are dialects of the Arawak stock. The Mura is a branch of the Tupi, spoken by the powerful tribe of the Muras on the Medeira and Amazon, who distinctly recalled in tradition their ancestral home in the west.[483] The Chiquito, the Mobima, the Caniciana (Canichana), the Cayubaba, the Itonama and the Ocorona remain so far irreducible stocks. Vocabularies of the first five have been preserved, but nothing of the Ocorona. It is probably identical with the Rocorona, in which Professor Teza has published some texts.[484] I have not been able to identify it with any other tongue. Hervas unites both with the Herisebocona as a single stock.[485]

2. THE PAMPEAN REGION.

South of the dividing upland which separates the waters of the Amazon from those which find their way to the Rio de la Plata, the continent extends in broad level tracts, watered by numerous navigable streams and rich in game and fish. Its chief physical features are the wooded and rolling Chaco in the north, the treeless and grassy Pampas to the south, and the sterile rocky plains of Patagonia still further toward the region of cold. In the west the chain of the Cordilleras continues to lift its summits to an inaccessible height until they enter Patagonia, when they gradually diminish to a range of hills.

The tribes of all this territory, both east and west of the Andes, belong ethnographically together, and not with the Peruvian stocks. What affinities they present to others to the north are with those of the Amazonian regions.

1. The Gran Chaco and its Stocks. The Guaycurus, Lules, Matacos and Payaguas. The Charruas, Guatos, Calchaquis, etc.

The great streams of the Parana and Paraguay offer a natural boundary between the mountainous country of southern Brazil and the vast plains of the Pampas formation. In their upper course these rivers form extensive marshes, which in the wet season are transformed into lakes on which tangled masses of reeds and brushwood, knitted together by a lush growth of vines, swim in the lazy currents as floating islands. These were the homes of some wild tribes who there found a secure refuge, the principal of whom were the Caracaras, who came from the lower Parana, and were one of the southernmost offshoots of the Tupi family.[486]

For five hundred miles west of the Parana and extending nearly as far from north to south, is a wide, rolling country, well watered, and usually covered with dense forests, called El Gran Chaco.[487] Three noble rivers, the Pilcomayo, the Vermejo and the Salado, intersect it in almost parallel courses from northwest to southeast.

Abounding in fish and game and with a mild climate, the Chaco has always been densely peopled, and even to-day its native population is estimated at over twenty thousand. But the ethnology of these numerous tribes is most obscure. The Jesuit missionaries asserted that they found eight totally different languages on the Rio Vermejo alone,[488] and the names of the tribes run up into the hundreds.

As is generally the case with such statements, distant dialects of the same stock were doubtless mistaken for radically distinct tongues. From all the material which is accessible, I do not think that the Chaco tribes number more than five stocks, even including those who spoke idioms related to the Guarani or Tupi. The remainder are the Guaycuru, the Mataco, the Lule and the Payagua. This conclusion is identical with that reached by the Argentine writer, Don Luis J. Fontana, except that he considers the Chunipi independent, while I consider that it is a member of the Mataco stock.

One of the best known members of the Guaycuru stock was the tribe of the Abipones, whose manners and customs were rendered familiar in the last century through the genial work of the Styrian missionary, Martin Dobrizhoffer.[489] They were an equestrian people, proud of their horsemanship and their herds, and at that time dwelt on the Paraguay river, but by tradition had migrated from the north.

The Guaycurus proper were divided into three gentes (parcialidades) located with reference to the cardinal points. On the north were the Epicua-yiqui; on the west the Napin-yiqui, and on the south the Taqui-yiqui. Their original home was on the Rio Paraguay, two hundred leagues from its mouth, but later they removed to the banks of the Pilcomayo. Their system was patriarchal, the sons inheriting direct from the father, and they were divided into hereditary castes, from which it was difficult to emerge. These were distinguished by different colors employed in painting the skin. The highest caste, the nabbidigan, were distinguished by black.[490]

The Abipones were almost entirely destroyed early in this century by the Tobas and Mbocobis,[491] and probably at present they are quite extinct. The Tobas are now the most numerous tribe in the Chaco, and their language the most extended.[492] They remain savage and untamable, and it was to their ferocity that Dr. Crévaux, the eminent French geographer and anthropologist, fell a victim in recent years. The dialects of the Abipones, Mbocobis and Tobas were “as much alike as Spanish and Portuguese” (Dobrizhoffer).

The Guachis speak a rather remote dialect of the stock, but undoubtedly connected with the main stem. According to the analogy of many of their words and the tenor of tradition, they at one time lived in the Bolivian highlands, in the vicinity of the Moxos and Chiquitos. It is probable that they are now nearly extinct, as for several generations infanticide has been much in vogue among them, prompted, it is said, by superstitious motives. Forty years ago an inconspicuous remnant of them were seen by Castelnau and Natterer in the vicinity of Miranda.[493]

The Malbalas, who were a sub-tribe of the Mbocobis, dwelling on the Rio Vermejo, are described as light in color, with symmetrical figures and of kindly and faithful disposition. Like most of the Chaco tribes, they were monogamous, and true to their wives.[494]

The Terenos and the Cadioéos still survive on the upper Paraguay, and are in a comparatively civilized condition. The latter manufacture a pottery of unusually excellent quality.[495]

On the authority of Father Lozano I include in this stock the Chichas-Orejones, the Churumatas, that branch of the Mataguayos called Mataguayos Churumatas (from the frequent repetition of the syllable chu in their dialect), the Mbocobis and Yapitalaguas, whose tongues were all closely related to the Toba;[496] while Dr. Joao Severiano da Fonseca has recently shown that the Quiniquinaux is also a branch of this stock.[497]

The Lules are a nation which has been a puzzle for students of the ethnography of the Chaco. They were partly converted by the celebrated Jesuit missionary and eminent linguist, Father Alonso de Barcena, in 1690, who wrote a grammar of their language, which he called the Tonicote. The Jesuit historian of Paraguay, Del Techo, states that three languages were spoken among them, the Tonicote, the Kechua and the Cacana, which last is a Kechua term from caca, mountain, and in this connection means the dialect of the mountaineers. Barcena’s converts soon became discontented and fled to the forests, where they disappeared for thirty years or more. About 1730, a number of them reappeared near the Jesuit mission of the Chaco, and settled several towns on the rivers Valbuena and Salado. There their language was studied by the missionaries. A grammar of it was composed by Machoni,[498] and a vocabulary collected by the Abbé Ferragut.[499] Meanwhile the work of Barcena had disappeared, and the Abbé Hervas expressed a doubt whether the Lule of Machoni was the same as that of his predecessor. He advanced the opinion that the ancient Lule was the Cacana; that the modern were not the descendants of the ancient Lules, and that the Mataras of the Chaco were the Tonicotes to whom Barcena was apostle.[500]

The missionary Lozano to some extent clears up this difficulty. He states that the Lules or Tonicotes were divided into the greater and lesser Lules, and it is only the latter to which the name properly belonged. The former were divided into three bands, the Isistines, the Oristines, and the Toquistines.[501] None of these latter existed under these names at the close of the last century, and at present no tribe speaking the Lule of Machoni is known in the Chaco. The language has evident affinities both with the Vilela and the Mataco,[502] but also presents many independent elements. The statement of Hervas, copied by various subsequent writers,[503] that the ancient or greater Lules spoke the Cacana, and that this was a different stock from the Lule of Machoni, lacks proof, as we have no specimen of the Cacana, and not even indirect knowledge of its character. Indeed, Del Techo says definitely that the missionaries of the earliest period, who were familiar with the Lule of that time, had to employ interpreters in ministering to the Cacanas.[504]

The modern Vilelas live on the Rio Salado, between 25° and 26° south latitude. I find in it so many words of such character that I am inclined to take it as the modern representative of the Lule of Machoni, though corrupted by much borrowing. When we have a grammar of it, the obscurity will be cleared up.

LULE.VILELA.
Tongue,lequy,lequip.
Tooth,llu,lupe.
Hand,ys,ysip.
House,enú,quané.

A comparison of the Vilela with the Chunipi, (Chumipy, Sinipi or Ciulipi,) proves that they are rather closely related, and that the Chunipi is not an independent tongue as has often been stated. In view of this, I include it in the Lule dialects.

The third important stock is that of the Matacos. It is still in extensive use on the Rio Vermejo, and we have a recent and genial description of these people and their language from the pen of the Italian traveler, Giovanni Pelleschi.[505] They are somewhat small in size, differing from the Guaycurus in this respect, who are tall. Their homes are low huts made of bushes, but they are possessed of many small arts, are industrious, and soon become conversant with the use of tools. Their hair is occasionally wavy, and in children under twelve, it is often reddish. The eyes are slightly oblique, the nose large, straight and low. Like all the Chaco Indians, they do not care for agriculture, preferring a subsistence from hunting and fishing, and from the product of their horses and cattle. What few traditions they have indicate a migration from the east.

The term Mataguayos was applied to some of this stock as well as to some of the Guaycurus. The former included the Agoyas, the Inimacas or Imacos, and the Palomos, to whom the Jesuit Joseph Araoz went as missionary, and composed a grammar and dictionary of their dialect. He describes them as exceedingly barbarous and intractable.[506] The Tayunis had at one time 188 towns, and the Teutas 46 towns. This was in the palmy days of the Jesuit reductions.[507] Both these extensive tribes are classed by D’Orbigny with the Matacos.

According to the older writers the Payaguas lived on the river Paraguay, and spoke their tongue in two dialects, the Payagua and the Sarigue. Von Martius, however, denies there ever was such a distinct people. The word payagua, he remarks, was a generic term for “enemies,” and was applied indiscriminately to roving hordes of Guaycurus, Mbayas, etc.[508]

The Payaguas, however, are mentioned distinctly by the early missionaries as a nation with peculiar language and habits. They differed from their neighbors as being aquatic, not equestrian. They were singularly skilful boatmen and had a mythology apart from the other tribes, “worshipping the devil under the figure of a great bird.”[509] There is also a manuscript in the Library of the American Philosophical Society, written in the middle of the last century, describing the visit of a missionary to the Payaguas, at that time resident near Santa Fé in Paraguay. He accuses them as given to revolting vices and utterly barbarous.[510]

The statement of Von Martius that the nation has entirely disappeared is incorrect, as quite recently a vocabulary of it has been obtained by Don Luis de Fontana, which shows it to be distinct both from the Guaycuru and any other known stock.[511]

LINGUISTIC STOCKS OF THE GRAN CHACO.

Guaycuru Stock:

Lule Stock:

Mataco Stock:

Payagua Stock:

Among the independent Chaco stocks, D’Orbigny classes the Lenguas, who in 1828 lived, about 300 in number, near Corrientes.[512] Von Martius believed they were a branch of the Guaycurus.[513] There is ample evidence, however, that they were a wandering branch of the Chiquitos of Bolivia. The missionary, J. P. Fernandez, who visited them about a century before D’Orbigny, says expressly that they spoke the same tongue as the Chiquitos;[514] and the statement of Hervas that the similarities of their words to the Chiquito arose from borrowed expressions is not well founded.[515]

The Charruas were a barbarous nation living in the extensive plains which stretch from the banks of the Parana to the sea coast. They were savage and courageous, without fixed homes, and skilled in the use of the bola. One of their customs was to cut off a joint of a finger on the death of a relative, and there were few of the adults that were not thus maimed.[516] In appearance they were usually large in size, heavily built, with big heads and broad faces, narrow noses, small eyes and large mouths. Their color was dark.[517]

The members of this family as recorded by the early writers, especially Hervas, are as follows:

CHARRUA LINGUISTIC STOCK.

Dr. Paul Ehrenreich describes them as they are to-day, splendid riders and daring soldiers, but faithless and tricky;[518] so they have not much improved since Father Chomé in 1730 stigmatized them as francs voleurs de grand chemin.[519]

The Guatos or Vuatos were accolents of the upper Paraguay and Araguay, and had fixed settlements near Albuquerque. Travelers report them as an unusually handsome people. They are well-built, light in hue, with Roman noses and regular features, and the men with a well-developed beard on lip and chin. This appearance does not belie their intelligence, which is above the average. Polygamy prevails to an uncommon extent. Von Martius thought that they were of a northeastern origin, connected perhaps with the Malalis of Bahia, who are a Tapuya people.[520] There may have been some admixture, as from a small vocabulary I quote the following resemblances:

GUATO.TAPUYA.
Water,maguen,magnan.
Head,dōken,dicran.
Hand,ida,danicra.
Foot,apoo,po, ipaa.
Tooth,maqua,aiqua.
Tongue,chagi,dageuto.

A recent writer does not give so favorable an opinion of this people. He found them living about the junction of the Rio San Lorenzo with the Rio Paraguay, and in a depraved condition. Girls who were not more than five or six years old were used by the men as wives. Sterility and premature decrepitude were the natural consequences.[521]

On the western border of the Chaco, in the provinces of Tucuman and Catamarca, resided the Calchaquis, a tribe interesting as the only one in the South Atlantic Group who constructed walls of cut stone. At least, such are found in their country, as for instance, one about thirty miles from Andalgala, where there is a well-constructed dry wall about ten feet high, enclosing a space nearly a mile in diameter, evidently once a walled city. Stone built tombs are also frequent, from which the rifler is rewarded with mummies, ornaments of impure gold, and small idols of copper. But I doubt if the Calchaquis developed any such ripe arts as these. History tells us that they voluntarily accepted the rule of the Incas about the middle of the fifteenth century, and that their land became part of the Collasuyu or southern district of the empire. All these remains have a distinct impress of Kechua art, and we may be sure that their inspiration was throughout Peruvian.[522]

The earliest missionaries depict the Calchaquis with curious usages and with a certain barbaric splendor. A widow became the wife of her husband’s brother, as of old in Israel. So long as she was a virgin, a girl could dress in the gaudiest colors, but once prostrato pudore, as the monk delicately puts it, she must change to sober weeds. Their ornaments were of silver and copper, and the nobles wore a circlet of gold and brilliant feathers. Their seasons of mourning were accompanied with the most violent orgies. Over the dead they raised heaps of stones, and held that the souls became stars.[523]

We have no specimen of the language of the Calchaquis, although a grammar of it was written by the Jesuit, Alonso de Barcena, and perhaps published. It is called the Katamareño or Cacana tongue, terms derived from the Kechua. The proper names, however, which have been preserved in it indicate that it was different from the Kechua.[524] I have already referred (page 227) to Von Tschudi’s suggestion that it survives in the modern Atacameño.

From the few specimens of skulls which have been examined, the Calchaquis appear allied to the Aucanian stock,[525] and it may be that further research will prove them a branch of the Araucanians.

The following tribes are mentioned by old writers as members of the

CATAMAREÑA LINGUISTIC STOCK.

The learned Barcena also prepared a grammar of the Natixana or Mogana language, spoken by the Naticas, whom we find mentioned by later authorities as neighbors of the Calchaquis in the government of Santa Fé.[526] They apparently belonged among the Chaco tribes. Barcena adds that nine different tongues were spoken in the district of Cordova, among which were the Sanavirona and Indama, which had not been learned by the missionaries.[527]

2. The Pampeans and Araucanians.

South of the Gran Chaco, say from south latitude 35°, begins the true Pampas formation. This, according to the geologist Burmeister, is not a marine deposit, but the result of fluvial overflows and dust storms. It is diluvial and quaternary, and overlies the Patagonian formation, which is marine and early Pleistocene. The pampas are in parts wide grassy plains, like the prairies of the upper Mississippi valley; in parts they are salt deserts, in parts more or less wooded. With little variety, this scenery reaches from the Chaco to the Rio Negro, S. lat. 40°. Nearly the whole of this territory was occupied by one linguistic stock. It is the same which is found in Chili, where its most prominent members are the Araucanians.

Which was the course of migration, whether from the Pacific coast to the Pampean plains or the reverse, is not positively decided, but I am inclined to believe it was the latter. The ancestors of the Araucanians would not willingly have crossed the barren wastes of the desert of Atacama; there are evidences of a different people inhabiting Chili before they possessed it, and we have traces that they had not obtained full possession of that country at its discovery. This view does not deny subsequent migrations of the Araucanians into the Pampas under the pressure of the Spanish invasion.[528] In such moving they were simply returning to the traditional homes of their ancestors. As the name of the whole stock, I adopt the word Aucanian, from the Araucanian verb aucani, to be wild, indomitable, from which are derived the tribal names Aucanos and Aucas, occurring on both sides of the Andes.[529]

The Pampeans are principally nomadic hordes wandering from pasture to pasture with their horses, cattle and sheep. Their transitory encampments, called tolderias, are pitched by the side of some pond or stream. There their low tents made of dried horse skins are grouped confusedly, one to each family. Their food is chiefly horse flesh and mutton, often eaten raw. They raise no vegetables, and dislike agriculture. They carry on, however, many small industries, tan and dye leather, which they work up into boots and horse furniture, and forge with skill iron heads for their long lances, and knives for the chase, while the women trim the ostrich skins into rugs, and weave wool into blankets and ponchos, highly prized for their serviceable qualities.[530] These products are bought up by the merchants in the cities, and thus the tribe is supplied with what it most prizes from European markets.

These roving hordes have no particular names. They are referred to as the northern, eastern or western peoples by the Aucanian terms having these significations, Puelches, Moluches, Huiliches. Besides these, there are the Ranqueles on the Rio Quinto, directly west of Buenos Ayres, who are said to have immigrated from Chili,[531] and the Querandies, now probably extinct, who once dwelt near that city.

Those living on the eastern slopes of the Andes, about the city of Mendoza, and in the ancient province of Cuyo, are described as taller and stronger than the Araucanians of Chili, and as claiming descent from the Pampean tribes.[532] They were locally known as Guarpes, and spoke dialects called the Allentiac and the Milcocayac, not distant from the Pampean proper, concerning which some grammatical description has been preserved.[533]

Few of the Pampean tribes have been induced to accept civilization or Christianity. They still believe in their good spirit, Chachoa, and in one of evil or misfortune, Gualicho; they continue to obey their priests or medicine men; and the resting places of the dead are regarded with superstitious awe. Marriage among them, while it has the appearance of violence, is really carried out with the consent of the girl and her parents, for a sum agreed upon.

The Molu-Che or Manzaneros are said to be the best of the Pampeans. They are sedentary and have extensive orchards of apples and flocks of sheep to the north of the Rio Limay. They have well-cut features, fresh light complexion, black fine hair, and their women are considered really handsome.[534]

The Araucanians of Chili, known as singularly bold warriors who defied successfully the Incas, and gave the Spaniards the greatest trouble, occupy the Pacific coast from south latitude 25° to about 43°, and number about 20,000. In physical appearance they resemble the Pampeans, and present marked differences from both the Kechuas of Peru and the Tapuyas of Brazil, having high, brachycephalic skulls,[535] and a clear copper color of skin. They are of moderate stature, but muscular, with black hair, round faces, small eyes, and small feet and hands. They are divided locally into northern and southern tribes, but there is little difference in dialects. Their tongue, the Chilidungu, has been extravagantly lauded by some who have studied it, and one worthy missionary was so enamored with it that he published a grammar and dictionary of it in Europe, that it might be introduced as the learned language there, to supersede the Latin:[536] it certainly is harmonious and flexible.

The Araucanians did not at any time rise in culture above the level of the Iroquois and Algonquins in the northern continent. It is true that in the tombs in their country we discover fine specimens of pottery, some good work in bronze, gold, copper and silver, and beautiful specimens of polished stone implements.[537] But if one examines closely the art-forms of these relics, he can not fail to recognize in them the potent inspiration of the Inca civilization; and we may be sure that if they were not directly booty from that nation, they were the products of its trained workmen, and are not to be put to the credit of Aucanian industry.

AUCANIAN LINGUISTIC STOCK.

The Pacific coast of Patagonia, gashed by ancient glaciers into deep fiords and rocky islands, harbors various tribes whose affinities are uncertain. The most curious of them would seem to be the Chonos or Chunos, or Cuncones. They lived south of the archipelago of Chiloe, and are described as having red hair, a light olive complexion, and of mild and friendly manners. They raised a breed of dogs (perhaps guanacos), and wove their clothing from its coarse long hair.

This account comes to us from as far back as 1619, when the first missionaries visited them,[538] and these traits cannot therefore be attributed to intermixture with Europeans. They are not peculiar in these respects. Similar traits are reported of the Boroas, a tribe in one of the valleys of central Chili;[539] and I have already referred to the red hair of the boys among the Matacos of the Gran Chaco. Perhaps it was not unusual among these nations, as I can in no other way explain the strange idea of the poet Ercilla the Homer of the Araucanian Conquest, that these people were descendants of the Frisians of North Holland![540]

The language of the Chonos is said to be quite different from that of the Araucanians. Pöppig believed it to be a distant dialect of the same stock. Some recent travelers assert that they are now extinct, but Dr. C. Martin informs us that the original inhabitants of the Chonos Islands, who were the “Huaihuenes” Indians, were transported in 1765 to the island of Chaulañec, where their posterity still survive.[541]

3. The Patagonians and Fuegians.

The Patagonians call themselves Chonek or Tzoneca, or Inaken (men, people), and by their Pampean neighbors are referred to as Tehuel-Che, southerners. They do not, however, belong to the Aucanian stock, nor do they resemble the Pampeans physically. They are celebrated for their stature, many of them reaching from six to six feet four inches in height, and built in proportion.[542] In color they are a reddish brown, and have aquiline noses and good foreheads. They care little for a sedentary life, and roam the coast as far north as the Rio Negro. They are not without some religious rites, and are accustomed to salute the new moon, and at the beginning of any solemn undertaking to puff the smoke of their pipes to the four cardinal points, just as did the Algonquins and Iroquois.[543]

Their language differs wholly from the Araucanian, though it has borrowed many words from it. An interesting fact illustrating its stability in spite of their roving life has been brought out by Ramon Lista. He has compared its present form with the vocabulary of it given by Pigafetta in his voyage in 1520, and shows that in the intervening generations it has undergone scarcely any change.[544]

Von Martius believed that a connection between the Patagonian and the Tapuya stocks could be shown, and gives a tabular comparison of the two.[545] I have extended this by means of Ramon Lista’s vocabulary of the former and Dr. Ehrenreich’s corrected forms of the Tapuya, and conclude that the resemblances are illusory, depending on incorrect orthography of the sounds.

About the beginning of the last century the tribes known as Poyas (Pey-yuy) and Reyes (Rey-yuy) were collected at a Mission established on Lake Nahuelhuapi, about south latitude 42°. Hervas reports them as speaking a language radically different from the Araucanian, and probably they should be classed with the Tzonecas.[546]

On the inhospitable shores of Tierra del Fuego there dwell three nations of diverse stock, but on about the same plane of culture. One of these is the Yahgans or Yapoos, on the Beagle canal; the second is the Onas or Aonik, to the north and east of these; and the third the Alikulufs, to the north and west.

Of these the Yahgans are the best known, through the efforts of the English missionaries who have reduced their language to writing. It is a polysyllabic, agglutinative tongue, with both pre- and suffixes, and is extremely rich in expressions for the ordinary needs of their life. The verb has four numbers, a singular, dual, trial and plural. It does not seem in any way related to the Aucanian stock.[547]

The tongue of the Onas, who are known as the Yakanna-Cunni, is apparently connected with the Tsoneca or Patagonian, which people they also resemble in stature and physical traits.[548]

The Fuegians are generally quoted as a people on the lowest round of the ladder of culture; and so they are painted by many observers. They have no government, they can count only to three, ordinary family affection is not observable, and even mothers manifest a lack of love for their offspring. Their shelters are wretched, and they go almost naked in a climate which is both cold and damp.

On the other hand, they display singular ingenuity in their utensils for hunting and fishing; they use the sling, the club, the bow, the bola and the lance; the women weave reed baskets so firmly that they will hold water, and their bark canoes are light and seaworthy.

In hunting they have the service of a native dog which they have trained, and whose welfare they look after with sedulous attention. Though devoid of idols and external rites of worship, they manifest in many ways a sense of religion. Thus the relations of the sexes are surrounded with ceremonies of fasting and bathing, to neglect which would entail misfortunes, and the name of the dead is not pronounced out of superstitious awe. The songs and legends of the Yahgans show some imaginative power. Many of them relate to the marvelous achievements of the national hero, Umoara, who appears to be a wholly mythical individual. Their strongest passion would seem to be for personal adornment, and for this purpose shells, vegetable beads, bright pebbles and variegated feathers, are called into requisition.[549]

These traits are not those of an enfeebled intellect, and an examination of their physical powers supports a favorable opinion of their capacities. Some of them are unusually tall and strong, especially those on the east coast. Their skulls are mesocephalic and prognathic, and their brains, which have been examined most carefully by a German anatomist, show not a single point of inferiority to the average European brain.[550]

From examinations which have been carried on in the numerous shell-heaps which line the shores, there is no evidence that any other people ever occupied the islands. Skulls and relics are such as those of the present inhabitants.[551] The total number of these is about 8000, nearly equally divided between the tribes named.

The classification of the smaller tribes under the above stocks is not yet complete. So far as I can make it out, it is as follows:

ALIKULUF LINGUISTIC STOCK.
ONA LINGUISTIC STOCK.

YAHGAN LINGUISTIC STOCK.

The opinion has been advanced by Dr. Deniker of Paris,[552] that the Fuegians represent the oldest type or variety of the American race. He believes that at one time this type occupied the whole of South America south of the Amazon, and that the Tapuyas of Brazil and the Fuegians are its surviving members. This interesting theory demands still further evidence before it can be accepted. It is not confirmed by such linguistic comparisons as I have been able to institute.