(ἐθνος, nation, ἄρχω, command), in Antiquity, a governor or ruler of a country. The title of ethnarch, like those of tetrarch and phylarch, appears to have been used by the Romans to designate such tributary princes as were not of sufficient importance to be styled kings. See TETRARCH. Ethnology. Ethnology is a word of Greek derivation, belonging to the same class, and formed on the same principles, as Geology, Astrology, Biology, Physiology, &c. The initial element, Ethno-, means, in its primary sense, nation; in its secondary, the Varieties of the Human Species. These are the true objects of ethnological study. Ethnology has been defined to be the Science of Races, but to this definition there is an objection. What is a race, if it be neither a species nor a permanent variety? Then, if it be neither of these, nor yet anything different from them, what is the use of the term? The answer to this has been suggested by Dr Prichard. Suppose an inquirer into the natural history of mankind to be doubtful whether a certain division constitute a species or only a variety, his argument may take such a form as to preclude him from the use of either term. His judgment is in suspense, and, as a consequence thereof, his phraseology is undefined. In such a case race is a useful word, being one which can be used when there are doubts as to whether we are dealing with a separate species or a variety of some species already recognised. Hence the term is subjective, i.e., it applies to the opinion of the investigator rather than to the object of his investigation; its power being that of the symbol of an unknown quantity in algebra. Useful, however, as it is, it is more so in the investigation of a doubtful problem than in the exposition of a known series of facts.

Another synonym to the word Ethnology is The Natural History of Man. The Physical History of Man is another. Both these convey the idea of ethnology, and something more, i.e., of the closely allied science of anthropology. Between these the Natural History of Man is divided; anthropology determining the relations of man to the other members of the animal kingdom, ethnology the relations of the different divisions of mankind to each other. There existed the materials for anthropology when the first pair of human beings stood alone on the face of the earth, and there would exist the same materials for anthropology if the world were reduced to its last human family, if it had no inhabitants but Englishmen, or none but Chinese; none but red men of America, or none but blacks of Africa. Were the uniformity of feature, the identity of colour, the equality of stature, the rivalry of mental capacity, ever so great, there would still be an anthropology. This is because anthropology deals with man as compared with the lower animals. As anthropologists we consider the structure of the human extremities, and enlarge upon the flatness of the foot and the flexibility of the hand. The one is subservient to the erect posture, the other to the innumerable manipulations which human industry demands. We compare them with the fins of fishes, or the wings of birds; in doing which we take the most extreme contrasts we can find. But we may also take nearer approximations, e.g., the hands of the higher apes. Here we find likeness as well as difference; difference as well as likeness. We investigate both, and record the result either in detail or by some general expression. Perhaps we pronounce that the one side gives the conditions of an arboreal life, the other those of a social state; the ape being the denizen of the woods, the man of towns and cities; the one a climber, the other a walker. Or we compare the skull of the man and the chimpanzee; noticing that the ridges and prominences of the external surface, which in the former are merely rudimentary, become strongly marked crests in the latter. We then remember that the one is the framework for the muscles of the face, the other the case for the brain. All this is anthropology as opposed to ethnology, the latter being a study which has no existence where there is no variety. The more manifold this variety the greater the scope of the ethnologist, and the wider his field. No matter how we classify our varieties. Whether the individuals and groups of individuals which exhibit them form different species of a genus, different genera of an order, or merely different varieties of a species, is indifferent.

The word, like the department of knowledge that it expresses, is new; so new that it may almost be said to be unfixed both in power and in form. Instead of ethnology many writers say ethnography. Some use the two words indifferently. Others use both, but distinguish between them; the latter meaning the descriptive, the former the speculative, portion of the subject.

No science has its relations to the other branches of study more accurately defined than ethnology. It is connected with zoology proper through anthropology; anthropology differing from zoology in its greater simplicity in one respect, and its greater complexity in another. So far as it deals with a single order, genus, or species, anthropology is simpler than zoology. So far as the characteristics of this order, genus, or species are peculiar, zoology is simpler than anthropology. The chief criteria of the animals below man are physical rather than moral; of man they are moral rather than physical. Anthropology gives us the naturalist view of our species.

Ethnology, on the other hand, gives us the historic view of it. Yet ethnology is different from ordinary history. In ordinary history we trace the effects of human actions upon humanity, the influences being moral. In ethnology we investigate the influences of soil, climate, nutrition, and similar agencies, for the most part physical. There is a difference between the two studies in all this. There are further points of difference. The facts with which history proper (civil history) deals are of later date than those of the ethnologist, the arena of the ethnologist being in the earlier period of the world's history, a period for the most part anterior to the existence of written records and the other forms of historic testimony. In this way ethnology is the general archaeology of man, to which the special branches of that subject are, more or less, subordinate. The methods of the ethnologist and the civil historian are not less different than the fields in which they work. History collects its facts from testimony; ethnology infers them by means of its own proper induction, arguing backwards, from the known to the unknown, from the effect to the cause. It was not until the publication of one of his latest works that Dr Prichard recognised the inductive character of ethnological research, and the extent to which his science was thereby separated from that of the civil historian. However, in his anniversary address to the Ethnological Society in 1847, he distinctly recognised it. "Geology, as every one knows, is not an account of what nature produces in the present day, but of what it has long ago produced. It is an investigation of the changes which the surface of our planet has undergone in ages long since past. The facts on which the inferences of geology are founded are collected from various parts of natural history. The student of geology inquires into the processes of nature which are at present going on, but this is for the purpose of applying the knowledge so acquired to an investigation of what happened in past times, and of tracing, in the different layers of the earth's crust, displaying, as they do, relics of various forms of organic life—the series of the repeated creations which have taken place. This investigation evidently belongs to history or archaeology, rather than to what is termed natural history. By a learned writer, whose name will ever be connected with the annals of the British Association, the term palaeontology has been aptly applied to sciences of this department, for which physical archaeology may be used as a synonym. Palaeontology includes both geology and ethnology; geo- Ethnology is the archaeology of the globe; ethnology that of its human inhabitants." For the important term Palaeontology we are indebted to Dr Whewell; for the best illustrations of the palaeontological method, to Sir C. Lyell. Each of these writers, though specially engaged upon other subjects, has largely contributed to impress upon English ethnology, at least, a scientific character—having done more than any proper ethnologist towards defining the method of ethnological investigations.

A great portion of the subject-matter to which this method applies has already been suggested by the relations of ethnology to anthropology on the one hand, and to civil history on the other. In the former, we give prominence to the phenomena of physical conformation—bodily and mental structure. In the latter, we investigate the phenomena and development of the social state. Subordinate to these studies are the ordinary preliminaries to a naturalist's and to a historian's education. To these, however, ethnology superadds much that is peculiar and proper to itself—just as the study of the varieties of such species as the horse, the domestic fowl, the sheep, or the dog, introduces a multitude of questions that the study of simpler and less varied species dispenses with. Let a naturalist employ himself on say the dog. He will assuredly find that the study of the variety far surpasses in complexity that of the species. But what if the dog-tribe had the use of language? What if the language differed with each variety? In such a case the study of canine ethnology would be doubly and trebly complex, though at the same time the data for conducting it would be both increased and improved. A distant, yet a very distant, approach to this exists. The wild dog howls; the companion of man alone barks. This is a difference of language as far as it goes, which is far enough to foreshadow the importance of the study of language as an instrument of ethnological investigation. Again, what if the dog tribe were possessed of the practice of certain human arts, and if these varied with the variety? What if they buried their dead, and their tombs varied with the variety? If those of one generation lasted for years, decadums, or centuries? The ethnology would again increase in complexity, and the data would again be increased. The graves of an earlier generation would serve as unwritten records of their habits of sepulture, and these differences in the mode of sepulture would be measures of some difference in the way of ethnology—indicative, perhaps, of some moral peculiarity; or, perhaps, indicative only of certain physical conditions of soil, climate, animal or vegetable products. The nidification of birds is a real example of this kind.

Hence the domain of the ethnologist touches those of the archaeologist and the philologue. But by far the most important of the accessories to ethnology is physical geography, studied with a special view to the relations between the bodily conformation of the occupants of a given area, and its climatological and other conditions. That any amount of intertropical influences can convert a white man into a negro is what many deny, and that on reasonable grounds. That any degree of arctic cold can convert a negro into an Eskimo is in like manner doubted. Neither is the possibility of two such extreme forms being developed out of some intermediate one at all freely admitted. In other words, the effects of climate, &c., upon the human frame are by no means held to be indefinite. That climate, &c., however, have some influence no one denies, though many limit it to a minimum.

So much for the method. As to the results of ethnological investigation, they are either so many points of classification, or so many points of history. They are so many points of classification, if we suppose that the differences between the different divisions of one kind have always been what they are at present; and they are so many points of history (physical history), if, supposing the whole species to have once been the same throughout, we deduce the present distinctions from the influences of climate and other causes acting during a long or short space of time.

It has been stated that the science is a new one; so new as for its method to have been but recently explained, and so new as for its nomenclature to be, even at the present moment, more or less unfixed. The materials, however, for a science are always older than the science itself. Hence, we find in more writers than one (some of considerable antiquity), ethnological notices, and even ethnological trains of reasoning, though no system of ethnology. In Greek literature not a little of this sort of information is found in Herodotus, an unconscious and instinctive ethnologist. He has described manners, and he has given glosses from several barbarous forms of speech. In Hippocrates we find an approach to a theory as to the effect of the external physical conditions of climate and the like on the human frame. Something, too, we find in Aristotle, and something in Plato; nothing, however, by which the study of man as an animal is recognised as a separate substantive branch of science. More than this, in works where the description of new populations was especially called for, and where the evidence of the writer would have been of the most unexceptionable kind, we find infinitely less than there ought to be. How little we learn of Persia from the Cyropaedia, or of Armenia from the Anabasis; yet how easily might Xenophon have told us much!

The opportunities of the Romans were greater than those of the Greeks, and they were better used. This we see in Sallust's sketch of Northern Africa, and in Caesar's of Gaul. The nearest approach, however, to a proper ethnological monograph is the Germania of Tacitus. It is far, however, from either giving us the facts which are of the most importance, or exhibiting the method of investigation by which ethnology is most especially contrasted with history. But the true measure of the carelessness of the Romans upon these points is to be taken by the same rule which applied to that of the Greeks, i.e., the contrast between their opportunities and their inquiry. How much they might have told us of such vast areas as Asia Minor and Persia—as the Danubian Principalities (Dacia), as Pannonia, Dalmatia, and Rhactia. We now ask, with interest and uncertainty, such questions as who were the Getæ? who the Thracians, &c.? Some answer Germans, some Slavonians, some an extinct division of our species. The commonest slave-dealer of Byzantium or Olbiopolis could have told us more than all the learned men since employed on such subjects.

We approach our own times, and the field of observation enlarges. Africa is circumnavigated, the parts beyond India visited, America discovered. The world becomes known in its extremities as well as in its centre. Nevertheless, the human naturalists anterior to Buffon and Linnæus, are like the great men who lived before Agamemnon.

Buffon made a general history of man, as well as a theory of the earth, important portions of his great work, prominent and full. Buffon gives us description rather than classification. Linnæus classification rather than description. How thoroughly zoological is the following table from the first edition of the Systema Naturæ.

| Quadrupedalia | |---------------| | Corpus hirsutum, pedes quattuor, fossam viviparos, lactifex. | | Anthrophomorphæa | | Diantæ primæs in utriquetque vel nulli. | | Homo..........Nosce teipsum.........II. | | Anteriores, Posterioræ, Simia........Digitæ 5. Digitæ 6....Simia canæ caræns. Papio satyrus. | | Posteriores anterioribus similes. | | Bradypus.....Digitæ 3, vol. 2. Digitæ 3. Al ignavus. Tardigradus |

1 For the ethnological views of Linnæus, see article MAN. In 1790, the first decal of the anatomical descriptions of Blumenbach was published, the special part investigated being the cranium; and that with a view to ethnology rather than to anthropology.

Little was said about the *differentiae* between the human skull and the skull of the higher apes; much about the distinctions between crania from America, crania from Asia, and crania from Africa. The last pentad of these researches was published in 1820. It is Blumenbach to whom we owe the division of mankind into the following five classes: (1) the Caucasian; (2) the Mongolian; (3) the *Ethiopic*; (4) the American; and (5) the Malay; the nomenclature also being Blumenbach's. Of the five terms before us, the second, the fourth, and the fifth are current without being inconvenient. The third (*Ethiopic*) is rarely used. The first (Caucasian) is unfortunately as current as it is both incorrect and inconvenient.

The *Regne Animal* gives the anthropological characters of man fully, placing him as the only species of the genus *Homo*, the only genus of the order *Bimana*. The ethnology is that of Blumenbach modified, the Malay and American divisions being subordinated to the Mongolian. Meanwhile, our improved knowledge of New Holland and New Guinea had given prominence to the Australian and Papuan varieties—varieties which somewhat complicated the previous classifications, but which were too slightly investigated to create an entire revision of it.

It was the anatomists and zoologists that gave to ethnology its naturalist aspect. The philologues, on the other hand, viewed it as historians.

Philological ethnology has advanced, at least, as rapidly as physiological. We may also add that it began earlier. As early as the voyage of Magalhaens, vocabularies of languages, which could only be valuable as ethnological materials, were collected—Pigafetta being the collector. Reiland knew of the existence of Malay words in the Island of Madagascar. Leibnitz had speculated on the Basque language. The Abbé Hervas established a voluminous correspondence with the missionaries of the Propaganda wherever they were found, and his *Saggio del Universo* gives us the results. In 1801, the *Mithridates* of Adelung appeared, giving specimens, more or less imperfectly classified and analyzed, of all the known languages of the world. Notwithstanding all this the world had yet to see a special and proper ethnologist. No philologue had known much of either anatomy or zoology; no anatomist or zoologist of philology. Nor yet had any one seen the subject in its full dimensions, or treated it otherwise than as an accessory to some other department of knowledge. With the naturalist it was an adjunct to zoology, with the scholar to comparative philology. The first who combined the two methods of investigation with a full perception of the magnitude of the subject was Dr Prichard—not a special naturalist, not a special anatomist, not a special philologue, but enough of all to make him the first and last of ethnologists. Such is his position, and in it we get a measure of the extent to which his subject is a new one.

The present author will now lay before his readers an exposition of the primary divisions and main sub-divisions of Mankind, reserving his criticism until the classification has been completed, promising that it is only where it differs from that of Dr Prichard that he feels that any real opposition of doctrine is encountered. The groups of all other writers must be looked upon as groups founded upon partial and incomplete examinations, groups formed (so to say) under no ethnological responsibility, but merely as extraneous speculations—so much zoology, so much anatomy, so much history, or so much philology having been diverted into a new and near channel.

This is only, however, to the writers who have delivered opinions upon general and systematic ethnology that this applies. The authorities that, upon any special division of Ethnology, the subject, demand respect and attention are numerous. To go no further than the limits of the English language, we have for American ethnology the names of Morton, Galatin, Hale, Schoolcraft, and others—for Malay and Indo-Chinese, Leyden, Crawfurd, Logan, Earle, &c.—for Indian and Himalayan, Hodgson, and many others besides.

What portion of the earth's surface is it best to start with? It is as certain that this is an inquiry of considerable practical importance as that it is one to which a variety of answers may be given. It is important, because the question as to the dispersion of mankind over the earth implies the existence of some special locality as a starting-point; and the starting-point being given, a tendency arises to attach to it what we have called the typical standard or average sample of our species. However much we may guard against these views, they continually obtrude themselves on our attention. On the other hand, the answers concerning the point of the earth's surface, and the division of the earth's inhabitants, with which it is most convenient to begin, are numerous. They differ with the point of view favoured by the inquirer. The zoologist, the philologist, answer differently. The starting-point that purely zoological considerations more especially suggest are the countries of the anthropoid (or anthropomorphic) apes, countries which agree with each other in being intertropical (indeed equatorial), but which lie in distant meridians—Western Africa and the southern extremity of Asia—the banks of the Gaboon and Borneo. Purely naturalist in our view, we may argue not only from the general phenomena of the geographical distribution of animals, but also from the natural conditions of human life as compared with the artificial. In the protoplasts of his species the zoologist sees but so many naked bipeds, with the capabilities, indeed, of working out for their future behoof the essentials of clothing, the use of fire, and the like, but, in the first instance, intolerant of any climate but the mildest, and incapable of sustenance on any soil but the most luxuriant. Hence, from the purely zoological point of view, the tropics are the cradle of our kind; and of the intertropical points, the habitats, or the parts about them, of the anthropoid apes. The philologist, on the other hand, looks towards China, Tibet, and the Trans-Ganggetic Peninsula; inasmuch as the peculiar character of the languages of these parts arrests his attention. They are monosyllabic, and destitute of inflexions. Such being their character, they are reasonably supposed to give us the simplest existing forms of human speech. Whether this view be right or wrong, it evidently favours the area of the so-called monosyllabic tongues being taken at the beginning of any ethnological exposition.

Another centre is what we call the logical one—logical because it is formed upon general rather than special grounds, and because it is based upon the principle that *forces must not be multiplied unnecessarily*. If the earth were one large circular island; if its populations were admitted to have been diffused over its surface from some single point, and if that single point were at one and the same time unascertained and requiring investigation, what would be the method of our inquiries? I suppose that both history and tradition are silent, and that the absence of other data of the same kind force us upon the general probabilities of the case, and a large amount of *a priori* argument. We should ask what point would give us the existing phenomena with the least amount of migration, and we should ask this upon the simple principle of not multiplying causes unnecessarily. The answer would be—the centre. From the centre we can people the parts about the circumference without making any line of migration longer than half a diameter, and without supposing any one out of such numerous lines to be longer than the other. This last is the chief point, the point which more especially Ethnology fixes us to the centre as a hypothetical birth-place, since the moment we say that any part of the circumference was reached by a shorter or longer line than any other, we make a specific assertion requiring specific arguments to support it. These may or may not exist. Until, however, they have been brought forward, we apply the rule de non apparentibus, &c., and keep to our conventional and provisional point in the centre—remembering, of course, its provisional and conventional character, and recognising its existence only so long as the search for something more real and definite continues.

Under any or all of these principles some portion of Southern Asia becomes our starting-point. Let the particular area be the Trans-Gangetic or Indo-Chinese peninsula, an area which coincides with none of our views exactly, but approximates them all. Of course it is, more or less, conventional, hypothetical, and provisional. It is by no means said to give us the cradle of our species. It is only said to be a convenient centre from which we may follow out so many lines of migration in so many different directions.

A.

ASIATICS AND NORTHERN EUROPEANS—POLYNESIANS, AMERICANS.

CLASS I.

Contains those populations whose physical conformation is, either typically or sub-typically, Mongol, having been generally recognised as such.

Area.—Northern, Central, and South-Eastern Asia; Northern Europe.

DIVISION I.—Languages monosyllabic.

Populations.—The Burmese, Siamese, Tibetans, Peguans, Cambodians, Cochinchinese, Chinese, various mountain tribes, Nepalese, Lepchas, Limbus, Chepang, Kocch, Bodo, Dhimal, Assam hill tribes, Islanders of the Indian Archipelago, the Nicobar and Carnicobar Isles, the Andaman Islanders.

The language of all the members of this group is not only monosyllabic, but destitute (or nearly destitute) of inflections. We may add, that the absence of inflections arises from the fact of their non-development—not from development followed by subsequent loss. The colour of the skin varies from a light yellow to almost black, the colour of the hair being more uniform, as also its texture. So is the stature—most of the populations under notice being undersized rather than oversize. The civilizational forms are extreme, i.e. there is the proverbial industry of the Chinese at one end of the chain, at the other the barbarism of the Andaman Islanders. The Chinese section of this group is far greater than all the others put together, hence (if we look only to the number of individuals) the religion of Fo is the dominant creed. The smaller and obscurer tribes of the mountain ranges between Cochin-China and Cambodia, Cambogia and Siam, Siam and the Burmese Empire, China and Assam, the Burmese Empire and Arakhan, claim, on the part of the ethnologist, an amount of attention that the civil historian has never yet given them. The Andaman Islanders are placed in this group on the strength of the language. Their colour is against their belonging to it.

DIVISION II.—Turanians—Languages other than monosyllabic.

Turanian area (either now or originally).—Mongolia, Manchuria, Siberia, Tartary (Independent and Chinese), Turkestan, Anatolia, Rumelia, parts of Persia, Armenia, Syria, the Crimea, Lapland, Finland, Esthonia, Livonia, Courland, governments of Archangel, Olonetz, Novgorod, Twer, St Petersburg, Yaroslav, Vologda, Permia, Viatka, Kazan, Astrakhan, Simbirsk, Saratov, Nizhgorod, Penza, Ethnology, Tambov—Hungary—the Kurile Isles, Kamtschatka, Japan.

Groups:—1. The Mongolian; 2. the Turk; 3. the Ugrian; 4. the Tungus; 5. the Peninsular.

1. The Mongolians.—Mongolia is part of the largest steppe in the world, extreme in climate, and poor in soil, and the Mongolians are the most nomadic of populations, their physiognomy and habits being uniform over a vast extent of country. The skull is broad, the face flat, the eyes oblique, a crescentic fold at the inner angle separating them in appearance more than they are in nature. The extent to which the Mongolian physiognomy has been considered typical has already been noticed. That the calves of the legs are undeveloped is no more than the natural effect of their equestrian habits. A tent is the house, milk and flesh the food of the Mongolian. Of the milk, when fermented, he makes his intoxicating drink, kumiss.

Temudzhin (Jenjiz-khan) is the hero and conqueror who has given the Mongolians their chief historical importance, China being the point towards which the edge of the Mongol sword has most turned. But the spirit of conquest has long been dead. The Mongolians under China are peaceable Buddhists, eminent for their amenability to priestly influence. Some few of them are independent, others subjects to Russia, where they are known as Kalmyks. Of these some have been imperfectly Christianized. The Kalmyks of the governments of Caucasus and Astrakhan are a colony. So are some small Mongol settlements in Cabul.

2. The Turks.—Turk, in ethnology, is a word of a very wide import. It denotes all the populations whose language is allied to that of the Turks of the Ottoman Empire. These, indeed, are best called Osmanli, while Turk means the Turcomans, both of Asia Minor and the Persian frontier, and a vast number of important tribes besides. No division of our species surpasses this in interest and importance. The original Turk area is of the largest. Conquest and encroachment have made it larger still—conquest and encroachment in every direction. How far this has moved westward we know, for we know that the Osmanli Turks of Turkey are the latest invaders of Europe. In Asia Minor they are older occupants, but even there they were originally invaders. They are to be found, mixed with Arabs, in Syria and Palestine, even to the frontier of Arabia. Their great centre, however, is Independent Tartary. From this they have spread as far east as Chinese Tartary; for, not only are the Uzbeks of Bokhara Turks, but the tribes of Khoten and Yarkend are Turks also. Bounded on the east by the Mongols, the Turk area prolongs itself northwards; and that to the very verge of the Arctic Sea. The Lena is a Turk river, and the Yakut Pagans who occupy it speak a language closely akin to the Osmanli of Constantinople. Add to this that all the Tartars (so-called) of the Russian Empire belong to the same stock. In Kazan, Astrakhan, and the Crimea, they have founded empires, and attained a civilization like that of the Ottomans. In Orenburg they are, perhaps, as much Ugrian as Turk—this being the view taken of the Bashkirs, whose language, however, is undoubtedly Turk. The Barabinski steppe is Turk. Nogays (wherever found) are Turk. The numerous tribes, more or less isolated, on the Upper Obi, and Yenesey, are Turk. So are some as far eastwards as the Lake Baikal. In Caucasus the Karatschai and Basiani are Turks. With an area so large, the Turk family is subjected to almost every physical influence. Again, it occupies some of the highest inhabited levels, e.g. the table-land of Paner, at the head of the Oxus. As far as any general statement can be made in respect to its distribution, it may be said that it is an overland one. Few Turks are in any of the maritime localities, or on any of the deeply indented sea-boards. Again, the Turk area is fitted for a pastoral, rather than an agricultural life. Still, there are excep- Ethnology. Along the Jurgan and Lower Oxus, the Turcomans, who are elsewhere so pastoral, are industrial tillers of the soil. What applies to the Turk habits applies also to the Turk creed. As a general rule they are Mohammedans, just as (as a general rule) they are pastoral. Yet many of the Turks under Russia are Christians, whilst the Yakut are Pagans. A few perhaps, on the Mongol frontier are Buddhists. Part of the Turk area is covered with archaeological remains,—which yet remain to be accurately investigated, e.g. the Russian governments of Kazan, Astrakhan, Saratov, as well as Independent Tartary. So is the Crimea. Neither the Turks nor the Mongols have been eminent as builders of cities. Mohammedan as is the Turk religion in general, it was, nevertheless, preceded, in some parts at least, by Christianity. Hence, the earliest Turk alphabet—which was of Syrian origin, introduced by the Nestorian missionaries, the Uighur form of speech being the one to which it was applied. From the Turks the Mongols took it, from the Mongols the Mantshu. In physiognomy the Turk is Mongolian; the Osmanli of Europe and, in a less degree, the Turkoman of Asia Minor, being exceptional sections modified by intermixture. The family is pre-eminently a family of conquerors. Tamerlane, the Seljukians, and the Osmanli, were Turk. The obscure conquest of the valley of the Lena by the ancestors of the present Yakuts was Turk. There is evidence—whether good or bad is doubtful—that one of the early conquest of China was Turk. A case may be made in favour of Asia Minor and Syria having been overrun by Turks long before the Seljukian period, and, indeed, the very dawn of history. A case can be made out for the Parthians (and possibly the Persians) having been Turk. That the Comanians, Uzes, Petschings, and Avars, were Turk, is certain. It is almost certain that the Huns of Attila were the same—probably the Bulgarians also (though now Slavonized). Lastly, the Scythians of Europe were Turk, rather than ought else.

The chief divisions are—the Kirghiz and less important tribes in Independent Tartary, the Uzbeks of Bokhara, the Turks of Chinese Tartary (of Uighur origin), the Turcomans, the Basiani and Karatschai of Caucasus, the Osmanli, the Crim Tartars, the Nogays, the Bashkirs (of mixed blood), the Barabinski, &c., of Siberia.

3. The Ugrians.—The Ugrian family is common to both Asia and Europe, being the most western member of the former, the most eastern (north-eastern) of the latter, continent. The Uralian range lies well-nigh in the centre of its area. As compared with a Mongolian, a Ugrian has a European—as compared with a German, a Mongol, physiognomy. The hair is often red; indeed, the Ugrian is the first great section of our species where a light complexion becomes anything like general. Placed between two great powers—the Turk and the Russian—the Ugrian family has been greatly encroached upon, greatly broken up; so that at present it presents but fragments of its former greatness. Originally, I think it extended from Caucasus to the Icy Sea, from the Dnieper to Behring’s Straits—an area not less, but rather greater, than that of the Turk. Nine-tenths of what is now Russian was once Ugrian.

The primary divisions into which this stock falls are—(1), the Eastern; (2), the Western, Ugrians.

1. The Eastern are the Samoyeds, the Yeniseians, and the Yukahiri, all populations of Asia, the Yukahiri being the most eastern. None of these are in contact with each other, inasmuch as northern offsets of the Tungus and Turks separate them. They are all nomades—all in the tundra rather than in the forest.

2. The Western Ugrians consist of the Laplanders, the Finlayders, the Permans, Siranians, and Votials of the Russian governments of Perm, Vologda, and Viatka; the Tseremiss, the Mordvins, the Tshuvash, on the middle Volga; the Voguls and Ostiaks on the ridge of the Ural mountains, and along the rivers Obi and Yenesey; and, finally, the Majars of Hungary. Between the extreme types there are broad differences, e.g. between the Laps and Majars. So there are in respect to their social and intellectual histories.

The two great representatives of Ugrian stock are, (1), the Finlanders of Finland, and (2), the Majars of Hungary, the latter being recent conquerors and intruders, the date of their conquest and intrusion being the tenth century. That the difference between the Finlander and the Lap is so great as it is, must be accounted for by the displacement of transitional forms. All Norway and Sweden was once Lap. So was Courland, and, perhaps, certain districts still farther west. The Ugrian creeds are either Christian or pagan, there being little or no Mohammedanism, except, of course, where there has been Turk intermixture. At present the most southern Ugrians (laying the Majars out of the account) are the Mordvins of the governments of Saratov and Astrakhan.

Few suggestions as to the early history of the more obscure nations are more important than that of Mr Norris respecting the third language of the Arrowheaded Inscriptions. It is to the effect that it was Ugrian. If so, the Ugrians from the southern part of the Uralian range may have overrun Persia several centuries B.C., just as certain Majar Ugrians overrun Hungary and Transylvania several centuries A.D.

The Samoyeds have been added to the Ugrian family by Castren, the Yeniseians and Yukahiri by the present writer. It is doubtful whether the group has yet attained its due dimensions. Possibly the Eskimo may have to be added to it, as well as some American tribes. Possibly, also, the Kamiskadale and Koriaks; an addition affecting the value of the Peninsular group.

4. The Tungus.—The Tungus area lies to the north and east of the Mongol. On the drainage of the Amur the older geographical names are Tungus, the newer Chinese. This shows that the latter nation is the encroaching one. On the other hand, it is Tungus dynasty (the Mantshu) that is dominant in the Celestial Empire. Divided in their political relations between Russia and China, the Tungus differ from each other in creed. In China they are Buddhist, in Russia imperfect Christians and pagans. The Mantshus have adopted the Mongol alphabet. Upon some of the tribes to the north of them the Tungus have encroached and effected displacements, the probable direction having been from south, or south-east, to north. Their appearance in history is late, inasmuch as their area lay beyond the pale of Greek, Roman, and Arab intercourse. It is the Chinese who most mention them; a tribe called Niuju more especially. This term (Niuju), Daurian, Tshapodzhir, Mantshu, and Lamut, are the most currently known names of the Tungus family. The Tshapodzhir tattoo their faces. The different zoological regions of the Tungus area give us the divisions of the Horse Tungus, the Reindeer Tungus, and the Dog Tungus. The Tungus has encroached upon the Ugrian area, isolating certain members of it; e.g. the Yukahiri.

The word Tungus = donki = man, a root that reappears in the Yeniseian, and more than one American language, e.g. the north-western dialects of the Athabaskan (vid. infra).

5. The Coreans, Japanese, Kurile Islanders, Kamskaldes, and Kortaks.—I throw all these into a single class of value as yet undetermined. I suggest for it the name Peninsular, from the extent to which its area is affected by the neighbourhood of the sea. In the way of language, the affinities of these populations are Tungus and Ugrian rather than monosyllabic, their physiognomy being unequivocally Mongolian. The civilization varies; the proximity of China having told favourably on Japan, less favourably on Corea, and little at all on the Kurile Islanders. Of the Kamska- Ethnology. The peninsula the aborigines are nearly extinct. Not so the Tshukshi of the coast of the Arctic Ocean. They hold their own bravely; bold in temper, strong in body.

Ermann has remarked that the Kamskadales and Ostiak languages appear liker each other than either are to the intermediate tongues, e.g., the Tungus and Mongol. If so, we see our way to the possibility of the Peninsular group (either wholly or in part) becoming subordinated to the Ugrian. Probably, too, the Kamskadales and Koriaks belong to one and the same division of this group. The Peninsular, Ugrian, and Tungus forms of speech are the most American of the Asiatic languages, the Tshukshi being also American in physiognomy.

**CLASS II.—IRANIANS.**

**Groups.—1. The Persians; 2. The Paropamissans; 3. the Armenians; 4. the Dioscurians (i.e., the Caucasians of Caucasus, or Caucasians in the limited, proper, and convenient sense of the term).**

**Physiognomy.—Caucasian (in the general, Blumenbachian, and inconvenient sense of the term) rather than Mongolian; in some cases Mongolian.**

**Languages.—Pauro-syllabic (i.e., of so few syllables as to approach the monosyllabic character). Inflections few; probably, in all cases from non-development rather than loss; certainly so in the Dioscurian division.**

**Area.—Kurdistan, Persia, Afghanistan, the Kohistan of Kabul, Bokhara, Kafiristan, Armenia, Caucasus.**

1. **Persians.**—Divided into Persians Proper, Kurds, Bilchik, Afghans. Shiite Mohammedans, and fire-worshippers in religion. Christianity rare.

The Persians of history must be looked upon as a mixed race, large infusions of Turanian blood having been introduced from the northern frontier. In the comparative compression of the cheek-bones, the oval outline of the face, and the prominence of the nasal bones, we find wide departures from the Mongol type. Aquiline noses first appear in this group, just as red hair first appeared with the Ugrians. The philology of Persia is closely connected with that of India.

2. **Paropamissans.**—Occupants of the ancient Paropamissa, i.e., the Kohistan of Kabul.

The so-called Kafres of Kafiristan, to the north of Pesha-ware, give us the type of this division—which, perhaps, should be subordinated to the proper Persian. The Kafres still retain the paganism of the times anterior to Mohammadanism. Their country is inaccessible, so that they are amongst the obscurest tribes of Asia. Their physical appearance is fine. The Chitrali, Dardoh, probably the Cashmirians, along with certain fragmentary populations of Kabul, are Paropamissan.

3. **Armenians.**—The features more massive than those of the Persian; the creed Christian rather than Mohammadan; the language, as also the alphabet, peculiar; the civilization old. Like the Jews, the Armenians are scattered beyond the limits of their own area.

4. **Dioscurians.**—This is a term suggested by the author. It is taken from the town Dioscurias, wherein Pliny says that 130 interpreters were wanted; so numerous were the dialects of Caucasus. They are numerous now. Hence the suggestion of the word instead of the awkward term Caucasian in the limited sense of the word, a circumlocution otherwise necessary from the Blumenbachian use of the adjective in another sense.

In the western Caucasus the physiognomy is Persian rather than Mongolian; in the eastern it approaches the Mongolian. The language has its nearest affinities with the Tibetan and other monosyllabic tongues; a statement which, in the case of the Irón, many philologues are indisposed to admit. The sections and chief subsections of the Dioscurians are as follows:

1. **Circassians.**—Adighé, Abassians, Kabardinians, &c. Ethnology. 2. (a) Tsesheth or Mzhdzhedzhi.—Central and on the watershed between the Terek and Kuban rivers; probably to be subordinated to the— (b) Lesgians, or Eastern Caucasians.—The Avars, the Anzokh, the Andi, &c. 3. **Irón, or Osset.**—Central and on the watershed between the Kuban and the Kur. 4. **Georgians.**—Mingrelians, Imeretians, Suan, Lazians, and Kartulinians.

**CLASS III.—INDIANS.**

In India we have a physiognomy referable to two types, one Persian rather than Mongol, the other Mongoloïd. In each case the skin is dark, i.e., brunette, or black. The chief creeds are Buddhist and Brahminic. The languages for the south are undoubtedly of Tamul origin. For the north, see Language Notes. Among many of the hill-tribes neither of the leading religions have yet struck root. Hence the original paganism of the peninsula still remains.

The minute ethnology of India is as interesting as it is complicated. It is the ethnology of a country of castes; of a teeming, ingenious, and industrious, but rarely independent population; of an ancient literature and an ancient architecture. It is a country which, whatever may have been the origin of its own civilization, helped to civilize the majority of the countries of the monosyllabic languages—Ava, Tibet, Siam, and (more than is generally believed) China. To the Brahminic and Buddhist religions India stands in the same relation as Arabia does to Mohammadanism.

In all Indian investigation we must bear in mind that there is one native and at least two foreign elements. Of the latter, one is that of the populations akin to the Nepalese and Tibetans whose area at one time reached the Ganges. The other is that of the speakers of the Sanskrit language, whoever they were and whenssoever they came. According to some, they were originally strange to the north as well as to the south of India, whilst from the south every one excludes them.

The languages wherein the Tamul character is undoubted are the Tamul Proper, the Kanara, the Telenga, the Tulava, the Malayalam, and the Coorgi forms of speech (all in the south); along with most of the dialects of the hill-tribes, one of which, the Rajmahal mountaineers, lies as far north as the Ganges, and within 100 miles of the most southern of the monosyllabic languages, i.e., the Garo.

The languages wherein the Sanskrit elements make the original Tamul character doubtful, are the Punjab, the Multán, the Sind, the Gujarati, the Rajputana, the Hindi, the Bengali, and the Udiya forms of speech.

**CLASS IV.—THE OCEANIC STOCK.**

**Area.—The Peninsula of Malacca; the Islands of the Indian Archipelago, Chinese Sea, and South Sea; Madagascar, New Guinea, Australia, Van Diemen's Land; i.e., Islands or Peninsula; whence the term Oceanic.**

**Primary Divisions.—(1.) Amphinesia; (2.) Kelonensesia.**

**Physiognomy.—When Amphinesian, more Mongol than African; when Kelonensesian, more African than Mongol.**

**Language.—When Amphinesian, with patent and recognised affinities to the Malay; when Kelonensesian, with Malay affinities fewer and only partially recognised.**

The Turk group illustrates the diffusion of man over a continent; the group before us the distribution of man over the ocean. The spread of a population over a continent is continuous; its spread over an ocean interrupted, or, at any rate, liable to interruption. Thus Gaul, most probably, began to receive a population as soon as the parts in contact with it were full. Britain, on the other hand, may have remained a solitude for centuries and millenniums Ethnology, after Gaul had been over-peopled. Unless we imagine the first canoe to have been built simultaneously with the first demand for water transport, it is as easy to allow that a long period intervened between that time and the first effort of seamanship as a short one. Hence the date of the original populations of islands is not in the same category with that of the dispersion of men and women over continents.

Now, great signs of the interruption of migration present themselves in the ethnology of the Oceanic Stock; and it is only by the admission of these that the group becomes natural—these and the evidence (often obscure) of language, along with the phenomenon of transitional, or intermediate, forms.

Amphinesians.—These fall into

1. Protonesians, or occupants of the Indian Archipelago and Chinese Sea, Sumatra, Borneo, Java, Moluccas, Philippines, &c.; the names being derived from πρώτος, first, and νῆσος, island, these being the islands first occupied from the Continent. The chief populations here are Malays, Dyaks, Javanese, Bugis, and Philippine Islanders.

2. Micronesians of the Caroline and Marianne Islands.

3. Polynesians, of the South Sea Isles, in general, from the Sandwich Islands to New Zealand, from the Fijis to Easter Island.

4. Malagasi of Madagascar.

Kelenonesians.—1. Papuans of New Guinea; Louisiade, New Hebrides, Tanna, Mallicollo, and New Caledonia; 2. Tasmanians; 3. Australians.

Kela, black; ἀπό, about, around.—A reference to the map will now explain the nomenclature. Polynesia is considered to have been peopled from Southern Micronesia, Micronesia from the north-eastern part of Protonesia (i.e., from the Philippines, via Lord North's Island, the Pelews, &c.) If so, the Amphinesian migration laps round the Kelenonesian, which is deduced, for Australia, from Timor, for New Guinea, from the Aru Isles. Tasmania has such remarkable affinities with the Papuan Islanders, that even there the population seems to have come round Australia rather than across it. The Kelenonesian civilization is far lower than that of the Amphinesian. Thereof much of the Protonesian part is Mohammedan, a portion Christian, a portion Brahminic, a portion still Pagan. There is also more than one Protonesian alphabet, e.g., the Batta, the Rejang, the Lampong, the Javanese, the Tagala. In most of the Protonesian islands, as well as in the Malaccan Peninsula, there are certain wild tribes, many of which are blacker in skin than the ordinary Malay, Javanese, Dyak, &c. These may be called the Negritos of the Malay, or Amphinesian area. In no case where their language has been examined has it been found to be Kelenonesian; nor yet has it been found to differ from the Protonesian dialects of the parts around it more than the Protonesian dialects differ from each other. The generality of writers have separated these from the Malays; some connecting them with the Papuans or Australians. The present writer does not do this. He sees in them only so many members of the ordinary Protonesian group in a ruder and more primitive state. Nevertheless, it was when this state was general, and the ordinary colour of the Protonesians darker, that he believes the migration by which New Guinea and Australia were peopled to have taken place.

In the Fiji islands, the Papuan and Polynesian characters are intermixed, just as if the stream of population that went round Kelenonesia and the extreme Kelenonesian met. The Papuan is one of the least known sections of mankind.

Class V.—Americans.

As long as the north-western parts of America were so imperfectly known as they were till within these last few years, the origin of the American population was uncertain; Ethnology, inasmuch as the Red Indians and the Eskimo on the eastern side of the Continent stood in strong contrast to each other; the transitional or intermediate forms having been obliterated. We no sooner, however, study the tribes of Russian America and the Oregon, than all such contrasts disappear. The Eskimo of the west graduates into the Indian of the west. Neither are there wanting direct proofs of philological affinity between the languages of the parts in question and those of the Tungus, Peninsulars, and Ugrians.

The classification of the divisions and sub-divisions of the American group (a group of value yet undetermined) has yet to take its final form. Provisionally, however, we have the following arrangement:

Eskimos in Greenland, Labrador, the coasts of the Arctic Sea, Russian America, the Alaskan Peninsula, the Aleutian Isles, and the extreme north-east of Asia. Kolch in the parts about Sitka, with affinities to the Eskimo on the one hand, and, on the other, to the Athabaskans, whose area subtends that of the Eskimo, and reaches from the Pacific to Hudson's Bay. The chief divisions of the Athabaskan group are the Chipewyans, the Beaver, Strong-bow, and other northern Indians, the Takulli, the Sikamni, the Suscees, the Loucheux, the Kenays, to which we may add outlying members as far south as Mexico, viz.: the Kwa-liokwa and Tlatskanai on the Californian frontier, the Navahos, and Jecriillas, and Apatches of California and Mexico. No group has its members lying further north and south of each other than the Athabaskan, some being within the Arctic Circle, some within the Tropics, the southern members isolated. Of equal magnitude with the Athabaskan group is—

The Algonquin, to which the majority of the Indians of the United States are referable, some as far north as Labrador, some as far south as Tennessee. The Delawares, Ojibways (different from the Athabaskan Chepewyans), and more than 30 other tribes are Algonkin.

The Iroquois group contains the Oneidas, Hurons, Wyandots, Senecas, Cayugas, Mohawks, &c.

The Sioux, the Ioways, Miniatees, Osages, and numerous tribes of the parts between the Mississippi and Rocky Mountains.

The Woccens, Catawbas, Cherokees, Choctas, Creeks and Caddo, have all Sioux and Iroquois affinities, and may probably have, under some future classification, to form with those two groups one large class.

Paducaes.—This means the tribes that speak languages akin to the Comanch, numerous in Oregon, California, Texas, and Mexico.

Texas and the south-western parts of the older states give us a vast mass of small tribes still requiring classification, Uche, Coosasas, Alibamas, Taenzas, Pascagoulas, Colapissas, Biluxis, Chetimachas, Humas, Tuncas, Pacanas, Natchitoches, Adahi, Opelusas, Attacapas, Natchez. Neither are the Riccarces, Pawnees, and Ahnenin (Fall Indians), further north classed.

In New Caledonia, Oregon, and California, we have

On the coast—The Haida, Chemmesyan, Billechula, Haitse, Chinuk, and Nutka tribes.

In the interior—The Atmas (Shushwah), Cutinis, and Sahaptins.

Add to these Jakons, Molele, Cayús, Kalapuyas, Lutuami, Santsila, Shastis, Palaks, &c.

In California a large portion of the Indians is either Athabaskan or Paduca. Still, there are several unclassed tribes. In the direction of New Mexico, and in New Mexico itself, we have the so-called Puebla Indians, partially unclassed—Moqui, Zuni, Taos, &c. In Old California, we have the Cochimi and Pericu tongues. Sonora gives us the Pimo and Coco-maricopas, the Opata, Ciri, and Jaqui languages. Further inland lie the Cora and Tarahumara, leading to Ethnology—Mexico, where the chief form of speech is the Aztek, not without many others (the Tancas, the Otomi, &c.) around it. In Central America the Maya is preponderant—also with numerous minor languages around it. In Nicaragua the Mexican reappears with the Dirian, Chontal, and other forms of speech, lately illustrated by the researches of Squier in its neighbourhood. The Moskito, the Bayano, &c., lead us to the South American, where the Muyasca section takes the same prominence in New Grenada that the Aztek did in Mexico, the Maya in Guatemala, and the Quichua, will take in Peru.

These give us the maximum amount of American civilization; a civilization which we either infer from the architectural and other remains spread over the Mexican and Peruvian areas, or else find described in the writings of the early Spanish conquerors. Undoubtedly, these supply us with some strong contrasts to the rudeness of the ordinary American Indian of the lake districts and prairies. They are differences, however, of degree rather than kind, as may be seen by taking the chief characteristics of the Mexican and Peruvian culture, and asking how far they appear in a rudimentary form elsewhere. Assuredly, some of the most important will be found amongst tribes of far less historical prominence than the ones in question. The existence of such an empire as Montezuma's is, undoubtedly, a measure of the Aztek civilization. Yet the obscurer one of Powhatan in Virginia exhibited an equal amount of organization. The phrase picture-writing applies to some of the imperfect historical records of the same country. Yet it is but the dawning of the Pawnee, or Algonkin in an improved form. The agricultural and metallurgic industry of the Anahuac and Cordillera has its origin quite as much in the physical conditions of the soil and climate as in any original intellectual superiority on the part of its occupants; besides which, it is approached by that of the Puebla Indians. The Mexican arithmetic has long commanded attention. It not only gives us a simple term for 20, but also one for 400 and 8000. Much stress has been laid upon this so-called vigesimal system. There is reason, however, for referring it to a very common-place origin. Twenty = man (i.e. ten fingers and ten toes). Twenty men made the first division of the Mexican army—a company; 20 of these made (say) a regiment; 20 regiments an army. This we infer from the symbols for 20, 800, and 4000, which are also those of the three military divisions just enumerated. We cannot, with such a fact as this before us, credit the Mexicans with having an extensive multiplication table. They may have been unable to count the intermediate numbers between 40 and 400, or even between 20 and 40. Another line of argument has helped to isolate these civilizations. Wherever similar characteristics appear they have been called, off-hand, Mexican or Peruvian, as the case might be. Nevertheless, the truer inference from the fact of certain architectural remains, &c., being found over a vast tract of country is, not that there has been a certain amount of conquest or colonization, but that more divisions of the American population than one have worked in the same way on similar materials.

Language, too, has helped to isolate. The Otomi tongue has long been known to be eminently monosyllabic. But there are others in the same category, e.g. certain Athabaskan dialects, the Attacapa, &c.

Again, many American nations flatten the skull artificially, and are known to do so; others are found with it flattened, but are not known to flatten it. Hence, we get the possibility of such a phenomenon as naturally flat crania, a phenomenon on which many authors insist. If their view be accurate, it gives us a physical distinction of great value, amounting, perhaps, to a specific difference. The natural character, however, of the skulls in questions is, as yet, unproved. Such are some of the differentiae—physical and moral—between the more extreme American populations; and it must be admitted that they cannot all be explained Ethnology away without a considerable amount of assumed influences. The more, however, these are studied, the more probable they appear. The more, too, we go into the details of the division itself, the more we find transitions and intermediate forms.

The great South American families of which the limits are the most defined are—1. The Quichuas. 2. The Caribs. 3. The Guarani; the Chaco, the Pampa, and the Chileno groups being of somewhat less, though still of considerable magnitude.

1. The Quichua section follows the line of the Andes from the equator to S. Lat. 28°, extending, at one point, as far east as Tucuman.

2. The Caribs have the same prominence in Venezuela that the Quichuas have in Peru and Bolivia. They extend from New Grenada to French Guiana. In Trinidad, part at least of the early population was Carib, as it was in the Antilles.

The Quichuas are mountaineers. The Caribs seem to have followed the courses of the rivers—being fluviatile so to say. More fluviatile, however, than the Caribs are.

3. The Guarani. Guarani or Tupi forms of speech are found at the mouth of the Amazonas, also at the mouth of the river Plata, also on the intermediate coast. In Entre Rios, Corrientes, and Paraguay, it spreads inland. On the watershed of the Amazonas and Plata, in the province of Santa Cruz de la Sierra, the Chiriguans, Sirionos, and Guayaros, are Guarani. Again, on the Napo and Putumayu, the Omaguas are Guarani also. Hence, the best provisional view that we can take as to the diffusion of so important a stock, may be to consider the Siriono and Guarayo districts as the original localities. These are common to two river-systems; so that, starting from these, the Omagua branch may have reached the Amazonas, whilst the Guarani reached the Parana and Uruguay rivers. Still the view is only provisional.

Next to the Quichuas, the Guarani gives us some of the most civilized tribes of South America. On the other hand the Mundruco, of the middle Amazonas, can be shown by their language to be Guarani; the Mundruco serving as the very type and standard of savage wildness. When a Mundruco has slain an enemy he cuts off his head, extracts the brain through the occipital foramen, washes the blood away, fills the skull with cotton, and then converts the whole into a kind of mummy, by drying it before the fire. The eyes he gouges out, and he fills up the orbits with colouring matter. Thus prepared, the head is placed outside his hut. On festive occasions it is placed at the top of a spear. Such is the history of the head of an enemy. Those, however, of friends and relations are preserved and kept, though with certain differences of detail. Thus on certain days dedicated to the obsequies and memory of the dead, the widow of the deceased takes a skull, seats herself before the cabin, and indulges either in melancholy lamentations, or in fierce encomium; the assembled friends meanwhile dancing round her.

The Chileno-Patagonians. The name is more expressive than convenient. It indicates, however, by its very composition, the magnitude of the group to which it applies. When we get into Chili, we arrive beyond the limits of the Quichuas and a new family makes its appearance, extending over Chili; over the whole of the country S. of the river Plata, over the islands of the Chloe Archipelago, and Terra del Fuego. Its divisions comprise, (a), the Chileno (or Araucanian) Indians; (b), the Pampa Indians; (c), the Patagonians; (d), the Fuegians.

The range of differences in respect to physical form is wide in this group; the range of differences in respect to the geographical conditions under which they are found, being also wide; e.g., there are the Andes of Chili, the level Ethnology, plains of the Pampas, and the insular character of the parts about Cape Horn; not to mention the fact of South America extending farther in the direction of the Antarctic Circle than any other part of the world.

In contrast to the large areas occupied by single families, we have, in South America, as elsewhere, small districts with a multiplicity of distinct divisions and subdivisions. One of these lies in the north, on the Orinoco and Amazons, in contact with the Carib area. Hence, we find the Maypuri, the Saliva, the Achagmas, and the Ottomakas, &c. On the Uapes only, Mr Wallace enumerates the Queianas, the Tarianas, Ananas, Coben, Pirinara, Pisa, Carapana, Tapura, Uaracu, Cohidias, Tacundera, Tacami, Miriti, and Omañas. These will probably turn out to be Carib, the import of the word as a class-name being enlarged.

Surrounded more or less by the Guaranis and occupants of Brazil, come the Botocudos, Canarins, Goitacás, Machacari, Patachos, Camacans, Matali, Cacriabas, &c., &c., falling into divisions and sub-divisions. Add to these the Indians of the missions of (a) Moxos and (b) Chiquitos; the Indian of the Eastern side of the Andes (Juracares, Mocetenes, Tacanas, and Apolistas); the Indians of the Chaco, chiefly Abiponan. To this class belongs, with others, the Abiponians Proper, the Mbocobi and Toba, the Lenguas, the Payaguas, the Matajeyes, the Guaycurus, perhaps the Charruas, known at present only in fragments, whose sections of it being either extinct or incorporated. The original divisions, however, were as follows:

1. The Charruas Proper; 2, the Chagas; 3, the Chanás; 4, the Guenonas; 5, the Martedunas; 6, Niboanes; 7, the Yaros; 8, the Minones; 9, the Casiguas; 10, the Bagaces; 11, the Tapes. Of these the Chanás and Niboanes inhabited, at the arrival of the Spaniards, the islands of the Uruguay, at the junction of the Rio Negro.

One of the most remarkable, and, at the same time, isolated, populations of South America are the Warawes.—Their occupancy is the Delta of the Orinoco, a swamp; as is a considerable portion of the sea-coast to the south of it. If it were not for the straightness of his hair, the Waraw (writes Sir Robert Schomburgk) might be taken for a negro. Doubtless he is dark-skinned, but I do not imagine that he has the negro lip. His skin is dark, and dirt gives intensity to its natural darkness; for the Waraw is uncleanly, even for an Indian. His language is certainly unintelligible to all his neighbours; neither has it been placed in the great Carib class, wide and capacious as that class is; nevertheless, it is far from being wholly isolate. It has miscellaneous affinities, and plenty of them; but the most notable characteristic of the Waraw is his industrial activity as a boat-builder. This furnishes nearly the whole of the Demerara with canoes. They are made either of the Cedrela odorata, or of a tree called Bisi, and are sometimes fifty feet long and six feet broad.

When a suitable tree has been found, the Waraw builds a hut in his neighbourhood, which he occupies as long as the boat is being built. The floor of the hut must be some feet above the level of the ground, and this is effected by selecting a spot where the tia-palm grows in thick clusters. This is docked to the requisite height, the root and a part of the trunk being left standing. The trunk of the manaca-tree is then cut into planks and made into a floor. Clay is laid on the floor, and a fire kept burning on the clay. The Manicaria saccharifera supplies the thatch.

Raleigh came in contact with the Waraws, whom he describes under the name of Tivitivas, adding that they fall into two divisions—the Cinecani and the Arumcate; that "they are a goodly people, and very valiant;" that, "in summer, they have houses on the ground as in other places. In winter they dwell in trees, where they build very artificial towns and houses; for between May and September the river of Orinoco riseth 30 feet upright, and then these Ethnology, islands are overflowed 20 foot high above the level of the ground, saving some few raised grounds in the middle of them; and for this cause they are enforced to live in this manner." The undoubted peculiarities of the Waraws have been exaggerated; and they have been described as men who live in trees,—as arboreal varieties of the human species,—even as arboreal species of the genus Homo.

For South America I give D'Orbigny's classification, promising that I do not find it coincide with the divisions deduced from the comparison of the South American languages.

South American Indians.—Colour yellow, brown, or copper-red; height variable; hair thick, coarse, black, smooth and long; beard thin, coarse, black, never wavy, late in making its appearance; chin short; eyes small, deep-set; jaws prominent; teeth nearly vertical; eyebrows prominent.

1. Primary divisions, or races (so-called)— A. Ando-Peruvian.—Colour olive-brown; stature low; forehead either depressed or but slightly vaulted; eyes horizontal, never bridés at their outer angle. B. Pampa.—Stature often considerable; forehead vaulted; eyes sometimes bridés at the outer angle. C. Brazilio-Guarani.—Colour yellowish; forehead not retreating; eyes oblique.

A. Ando-Peruvians— a. Peruvian branch.—Colour deep olive-brown; form massive; trunk long in proportion to the limbs; forehead retreating; nose aquiline; mouth large; physiognomy sombre.—Aymara and Quichua Peruvians. b. Antisian branch.—Colour varying from a deep olive to nearly white; form not massive; forehead not retreating; physiognomy lively, mild.—Yuracares, Moceténés, Tacanas, Maropas, and Apolistas. c. Araucanian branch.—Colour light-olive; form massive; trunk somewhat disproportionally long; face nearly circular; nose short and flat; lips thin; physiognomy sombre, cold.—Indians of Chili and the Chonos Archipelago. The Fuegians.

B. Pampas— a. Pampa branch.—Colour deep olive-brown, or maroon; form herculean; forehead vaulted; face large, flat, oblong; nose short; nostrils large; mouth wide; lips large; eyes horizontal; physiognomy cold, often savage.—Indians of the Chaco and Patagonia. b. Chiquito branch.—Colour light olive; form moderately robust; mouth moderate; lips thin; features delicate; physiognomy lively.—Indians of the Mission of Chiquitos. c. Moxos branch.—Form robust; lips thickish; eyes not bridés; physiognomy mild.—The Indians of the Mission of Moxos.

C. Brazilio-Guarani.—A simple branch.—Colour yellowish, with a slight tinge of red; form massive; height moderate; face circular; nose short and straight; nostrils narrow; mouth moderate; lips thin; eyes oblique; eyebrows prominent; features delicate (effeminés); physiognomy mild.—Guarani, Caribs (?), and all the unplaced tribes of Paraguay, Brazil, the Guianas, and Venezuela (?).

From the parts about Cape Horn, we return to the western frontier of the Ugrian area, this being the frontier where European and Asiatic ethnology join.

B.

CENTRAL AND SOUTHERN EUROPEANS.

European ethnology is peculiar,—we may call it minute ethnology. The main divisions are known from history. So are their chief characteristics. Hence the points that command attention are certain of the finer shades of difference, Ethnology. The general and leading facts that, in the case of such tribes as the Polynesian, the American, or the African, have to be collected as so many fresh points of information, are already known. As compared with Polynesia, America, or Africa, Europe is but a portion of Asia. Nevertheless its physical conditions claim consideration. No part of Europe lies between the tropics; so that the luxuriance of a spontaneous and varied vegetation, with its pernicious tendencies to incline the habits of its population to idleness, is wanting. The rank and rapid growth of the plants which serve as food to men and animals, and which dispense with labour, nowhere occurs. Few parts come under the class of steppes, or at most but imperfectly approach their character. In Asia, the vast table-lands of the centre, occupied by the Turks and Mongols, have ever been the cradle of an active, locomotive, hungry, and aggressive population. And these have ever had a strong desire to possess the more favoured areas of the south, and have conquered them accordingly. The Luneburg Heath, and parts of Russia, are the nearest resemblances to the great steppes of Mongolia and Independent Tartary; but they are on a small scale. In Russia, where the land is flat and level, the ground is also fertile, so that agriculture has been practicable, and (being practicable) has bound the occupant to the soil, instead of mounting him on fleet horses to wander with his flocks and herds from spot to spot, to become a shepherd by habit and a warrior by profession.

Europe is narrowest in its northern parts. This has had the effect of limiting those populations of the colder climes, whose scanty means of subsistence at home incline them to turn their faces southwards with the view of conquest, and supply them with numbers to effect their purpose.

Its diameter from north to south is less than its diameter from east to west. This has kept the mass of its population within a similar climate, or, if not within a similar climate, within a range of temperature far less wide than that which separates the African, the American, or the Asiatic of the northern parts of their respective continents, from the Hottentot of the Cape, the Fuegian of Cape Horn, and the Malay of the Malayan peninsula.

In no country are the great levels more broken by mountains, or the great mountains more in contiguity to considerable tracts of level country. The effect of this is to give the different characters of the mountaineer and the lowlander more opportunity of acting and re-acting on each other.

In no country are the coasts more indented. We may look in vain for such a sea-board as that of Greece elsewhere. The effect of this is to give the different characters of the sailor and landsman, the producer and the trader, more opportunity of acting and re-acting on each other.

The greatest rivers fall into seas navigable throughout the year. Contrast with this the great rivers of Asia—the Obi, the Lena, the Yenesei, and others, which for the purposes of navigation are useless, falling as they do into an Arctic sea.

Our greatest river, the Danube, runs from east to west. This ensures a homogeneous character for the population along its banks. Contrast with this the Nile, the Mississippi, and the Yenesei, in all of which the simple effect of climate creates a difference between the populations of the source and the embouchure. The great rivers of China do the same as the Danube; but the Danube differs from them, and from all other rivers running in a like direction, by emptying itself into an inland sea; a sea which gives the opportunity of communication not only with the parts north and south of the rivers which fall into it, but with those to the east of it also. The Hoang-ho and Kiang-Ku empty themselves into an ocean that, in these days of easy communication, leads to America, but which, in the infancy of the world, led to a coasting-trade only, or at most to a large island—Japan. The Baltic and Mediterranean have,—the Ethnology, one Africa, the other Scandinavia, to ensure their being put to the uses of trade.

Add to these the relations of south-eastern Europe to the countries watered by the Nile, the Tigris, and the Euphrates; in other words, in investigating the conditions under which European civilization evolved itself, remember those of the eastern extremity of the Mediterranean.

There is another point which stands out more prominently in European ethnology than elsewhere. European history goes far back. Doing this, it tells us of great changes—changes which we may assume elsewhere, but which we know to have existed in Europe. What if these go to the extinction of some stock, group, or family? A great complication is introduced. The extinction has more meanings than one. It often means absolute obliteration. It often, however, means no more than the abolition of the outward and visible signs of ethnological difference. A negro marries a white. In the fourth, fifth, sixth, or seventh generation, as the case may be, his descendant is, to all intents and purposes, a white man. Yet the negro blood is not extinguished. It exists, though in a small proportion. Again, a Cornishman loses his native language and speaks English as his mother tongue. Many generations before he did this he differed from the Englishman in speech only. Is his British blood extinguished? No. The chief sign of it has been lost; that is all—that stocks may be obliterated, may intermix, or may lose their characteristics. To determine what existing populations are descended from the populations of the classical times, and to give to existing populations their ancestors during the same period, is a business which continually devolves upon the European ethnologist; more rarely upon the ethnologist for Asia, never upon the ethnologist for Polynesia or America.

Physically, the European populations are alike, the differences lying within narrow limits. In language the lines of demarcation are broader.

The European physiognomy is generally considered to be Caucasian as opposed to Mongolian, i.e., Caucasian, in the wide, Blumenbachian, and inconvenient sense of the term. And this it is in the southern and in the more cultivated parts of Central Europe. But it is by no means so generally. I do not say that truly Mongolian faces are common anywhere. I do say, however, that the Slavonic physiognomy approaches the Ugrian, and the Lithuanian does so still more; also that the Malay features (sub-typically Mongul) are common in the Alps, the Pyrenees, the ruder parts of Germany, Ireland, Scotland, Wales, France, and England.

The languages fall into four divisions. The first two contain two unplaced forms of speech. The third contains a form of speech which has lately been incorporated with the fourth, but only lately and only partially. The fourth contains a group of languages long and universally admitted not only to be allied, but to be closely allied:

1. The Basks.—Their language isolates the Basks. Their locality is the western Pyrenees, the north of Spain, and the south of France. Here it is where, although in the towns, like Bayonne, Pampeluna, and Bilboa, the population is French or Spanish, the country people are Basks or Biscayans—Basks or Biscayans not only in the provinces of Biscay, but in Alava, Upper Navarre, and the French districts of Labourd and Soule. Their name is Spanish (the word having originated in that of the ancient Vascones), and it is not the one by which they designate themselves; though, possibly, it is indirectly connected with it. The native name is derived from the root Eusk, which becomes Eusk-ara when the language, Eusk-herria when the country, and Eusk-alduae when the people are spoken of; so that the Bask language of the Biscayans of Biscay Ethnology. Is, in the vernacular tongue, the Euskara of the Euskaldunae of Euskerrria. That the Euskara is no new tongue may be inferred from the fact of its falling into dialects, which Humboldt limits to three, whilst others extend them to five or six. The Biscayan proper is spoken in the country of the ancient Autrigones and Caristii, and it has been proposed to call it the Atrigonian. It has less correctly been called Cantabrian, and this is the name which the national taste best likes; for a descent from the indomitable Cantabrian, who so long and so successfully spurned the yoke of Rome, and who transmitted the same spirit and the same independence to the Asturian, is creditable enough to be claimed. Nor is the claim unfounded, since, in all probability, the ancient Catabria included some of the ancestors of the Euskaldunae. The Guipuscoan is the western Biscayan. The Laburitanian is the Euskarian of France, spoken in the parts about St. Jean de Luz; and which, in the district of Soule, is supposed to fall into a sub-dialect.

Even as the mother-tongue of the present Welsh was, originally, the language of the whole of Britain, so was the mother-tongue of the present Basque the language of the whole of the Spanish peninsula. The generic name for this is Iberian, so that, assuming a certain amount of intermixture between the invading Romans and the aboriginal inhabitants, the present Spaniard, though Latin in speech, is more or less, Iberian in blood. So are the south-western French, inasmuch as the old Iberian reached the Garonne, perhaps the Loire.

2. The Skipetar or Albanians.—Their language isolates them, though many hold that it can be connected with those of the fourth class. Albania is their occupancy. The testimony of travellers to their belonging to the fair-complexioned and grey-eyed populations is pretty general, although Skene gives the Mirdite tribe a swarthy skin and black eyes. The evidence, too, as to their bulk and stature varies; some writers giving them spare, light, and tall forms, others making them shorter and more square-built than the Greek. That the eye has less animation, and the countenance less vivacity (in other words, that the Albanian is heavy featured as compared with his quick-witted neighbours) is certain. Both the men and women are hardy, and expose their bodies freely to the atmosphere, accustoming themselves to an out-door life amongst their flocks and herds, and dwelling, when in-doors in rude huts. Like the Swiss, they willingly let out their valour and hardihood in military service; and the best and most unscrupulous soldiers of the Sultan are those recruits, who, partly by force, partly by pay, are brought from Albania. Hence we find Albanians far beyond the pale of Albania; in Greece, in Thrace, in Asiatic Turkey, in Egypt, and even in Persia. The tribes too, amongst themselves, indulge in the right of private quarrel, rarely rising to the dignity of warfare, but more like the old border feuds between England and Scotland. Some of the Skipetar are Mohammedans, some Roman Catholics, some Christians of the Greek church. These are the modern representatives of the ancient Illyrians. Skipetar blood must be found far south of the present Skipetar area, even as Basque (Iberic) blood is to be found in the non-Iberic parts of Spain and Portugal and France.

3. The Kelts.—By raising the value of the class called Indo-European, a class based upon philological rather than physical grounds (for which see Notes on Language), the Kelts (= the Welsh, Bretons, Gaels, and Manxmen) can be placed in the same division with the

4. Sarmatians, Germans, Latins, and Greeks, who are Indo-European (q.v.) in the strictest sense of the term. Sarmatian means Slavonic and Lithuanic collectively. The great historical nations of Europe belong to this class. For the relations of the speakers of the Sanskrit language see Language (Notes).

AFRICANS AND SOUTH-WESTERN ASIATICS.

A return to the south-western parts of Asia leads us to the ethnology of Africa, which begins in Syria and Mesopotamia; the populations akin to the Arab being (in the opinion of the present writer at least) African rather than Asiatic; at any rate, transitional or intermediate.

The populations speaking languages akin to those of the Bible and Koran (the Hebrew and the Arabic) are termed A. Semitic.—This group contains,

In Asia, the Syrians, Assyrians, Babylonians, Phoenicians, Ammonites, Moabites, Ishmaelites, Edomites, Samaritans, Jews, along with their colonies, &c., of antiquity; of modern populations, the Arabs, the Jews, the Kaldani (of Kurdistan), the Arabs, &c.

In Africa, the Abyssinians of the province of Tigré (Christian), the Abyssinians speaking the Amharic language (Christian), the Gafat Abyssinians (pagan).

Sufficiently akin to these to have become called Sub-Semitic are the Amazigh or Berber, and Copt or Egyptian, tribes. With the modern Copts the hair is black, crisp, or curled; the cheek-bones prominent; the lips thick; the nose somewhat depressed, nostrils wide; complexion brown or yellowish; eyes oblique (?); frame fleshy; physiognomy heavy.

With some of the Berbers—e.g., the Tuariks of Wadreag—the hair and skin are nearly the hair and skin of the negro. With others—the Amazigh of the ancient Mons Aurasius—the complexion is so light as to have engendered the doctrine that they are descended from the Europeans who, under Geneseric, sacked Carthage.

We approach the negro types in the division for which the name

B. Nilotic is suggested, its area being the drainage of the Upper and Middle Nile. Of the four chief Nilotic groups (1 and 2) the Nubians and Bishari approach the Copts, (3) the Agows, the Abyssinians, (4) the Gallas having both Semitic and Kafre characteristics. The colour is often brown rather than black, the frame spare, the nose straight or aquiline, lips moderate, hair long. In respect to language the Agow dialects are Sub-Semitic. Hence, whilst the Semitic populations are extreme—i.e., Asiatic or European rather than negro—the Nilotic are transitional. Several of the Nubian tribes are truly negro, at least the languages of Kordofan and Darfur belong to this class. For the philological import of the term Semitic see Language (Notes); also for that of

C. Kafre, the name of the next group.

The Kafre area extends from the parts north of the equator to the Hottentot frontier, and that on both sides of the continent; the tribes of the higher levels and the more southern areas departing furthest from the negro type; those of the coast, the alluvial tracts, and the parts nearest the equator approaching it the closest. In many cases the Kafre physiognomy is absolutely negro. The eastern Kafres, of Mozambique, Zanzibar, &c., are conterminous with the southern Gallas; the southern—those of the Cape—with the Hottentots and European colonists; the western—of Benguela, Angola, Loango, &c.—with the negroes of the parts between the Niger and the Gaboon, members of a class pre-eminently, but not exclusively, nor yet to the exclusion of other classes.

D. Negro.—Skin black, hair woolly, lips thick, nose depressed, jaw protruded, forehead retiring, proportions of the extremities abnormal. Such is the extreme type. Add to this that the civilization is low, and the area chiefly consists of lowlands, coasts, and the deltas of rivers, rather than of elevated plateaux or mountain ranges. When the occupancies are of this latter kind the physiognomy approaches that of the Semitidse, the language in many cases being negro, or like that of the nearest tribes. With the negro branch of the African division it is convenient to begin with the part where the black tribes touch the south-western Amazighs and Moors. This means the parts about the Senegal.

The Wolofs.—Skin black, conformation negro, religion pagan rather than Mohammedan. Large portion of their area the valley of the Senegal.

The Mandingos.—Less dark than the Wolofs, negro, original paganism largely replaced by Mohammedanism. Large portion of their area the valley of the Gambia.

The Fulahs.—Skin brown rather than black, form less negro and more Arab than that of the Mandingos and Wolofs, creed largely Mohammedan. Area elevated, the watershed between the Gambia and Senegal.

With all these populations the civilization is comparatively high.

The languages unequivocally related to the Mandingo extend as far south and east as the Ivory Coast; for the strip which extends along the sea from Cape Mount to Cape Palmas, and which is occupied by tribes akin to the Krumen, we (after several doubts) subordinate to the Mandingo area, or rather we make both the Kru and Mandingo group members of the same large class. On the other hand, a portion at least of the Kong range is Mandingo; indeed, the extremity of this area reaches the Ashanti frontier inland, and the Fanti frontier on the Gold Coast; these two names belonging to one and the same class, a class conveniently called Inta.—To this belong the Booroom, Aowin, Amanaha, Affutu, Ahanta, and Odzhi forms of speech. Mohammedanism decreasing, fetishism predominant.

The Gha or Akra, occupants of the parts about Cape Coast Castle, speak a language said to be unintelligible to the Fanti, but with numerous Fanti, as well as other affinities.

The Dahomey tribes occupy the Slave Coast;

The Benin, or Moho, the delta of the Niger along with

The Ibos.—These are all subdivisions of some single group of unascertained but no very high value. The influences of Mohammedanism are here at the minimum; snake worship and human sacrifices being common; circumcision being common also. The Bonny tribes form a division of this class.

The Mahas, who reach as far inland as the Kong range, are Dabomoy tribes.

At the back of the Benin and Ibo countries inland, lie three allied groups—allied also to the groups just enumerated—viz.,—

The Yoruba (or Aku), the Kouri, and the Tapua (or Nyi).

The Yoruba and Tapua populations lead to those of the interior.

In the parts about the Old Calabar river the type (of language at least) is considered to change. Such being the case, we may return to the Wolof area, and go over some portions of the ground again for the sake of certain minutiae which the general sketch hitherto given has prevented us from noticing.

The three great classes of the Wolofs, Fulahs, and Mandingos, by no means exhaust the populations between the Senegal and Cape Mount. About Cape Verde, surrounded by Wolofs, lie the Sereres. Further inland, separated from the Sereres by Wolofs, and in contact with both Fulahs and Mandingos, lie the Serawolli. The languages of each of these divisions have miscellaneous affinities with each other, and the three leading tongues just named. Perhaps, they are most like each other, next like the Fulah and Wolof, least like the Mandingo.

From the Casamance to the Sherbro, the languages of the coast are all other than Mandingo, Fulah, or Wolof Proper. They differ, too, from each other. Some have been subordinated to other groups, though only by certain writers. The general phenomenon is the appearance of Ethnology—several mutually unintelligible and by no means visibly alike tongues within a comparatively small area; these being—The Felip, the Banyon, the Papel, the Balantes, the Biagures, the Bissago Islanders, the Nalu and the Sapi, the languages akin to the Timmani and Ballom, south of which the Mandingo group shows itself on the coast.

Again,—east of Cape Palmas we have the imperfectly known languages of the Ivory Coast, with miscellaneous affinities to the Mandingo, Fanti, &c.

Amongst the tribes thus enumerated the Fulahs stand most by themselves in physical form, being brown rather than black. In civilization the Mohammedan Mandingos equal them. In natural vigour and spirit the Wolofs and Krumen are honourably distinguished. Fetishism and slave-dealing take their worst forms in Dahomey. For simple rudeness of habits the Felipps and the smaller sections like them are most conspicuous. Laying language (which will be considered elsewhere) out of the question, the chief objection to throwing all these into one large group lies in the physical peculiarities of the Fulahs. We must note, however, the extent to which it coincides with certain physical conditions, and remember that we have met it under similar conditions elsewhere, i.e. in Abyssinia. The reasons for not carrying this class further south, lie in certain points of language also considered elsewhere. Arguing from the direction in which the leading populations seem to have spread themselves, we may, in a somewhat bold attempt to reconstruct the original situs of the northern section of this great western group, not unreasonably hold—that the Sereres, Serawolli, and the minor populations, like the Felip, &c., are in situ; the Wolofs, Mandingos, and Fulahs, that most indent their areas, being intrusive; the first having spread from the Lower Senegal, the second from the Upper or Middle Senegal, the third from the Upper Gambia and the parts south of it. The Timmani seem to have encroached from the north, the Gha from the interior; the Ashanti having also effected large displacements. The Mohammedan Africans use, though sparingly, the Arabic alphabet—the Mandingos being, perhaps, the best scribes and scholars. The Vei (a division of the stock) have a syllabic alphabet of their own, invented by a native lately dead, Doala Bukara. He had seen both Arabic Korans and English Bibles, so that the idea of writing sounds was one with which he was acquainted.

South of the Berber and east of the Fulah, Mandingo, Jorruba, and Tapua areas, come the inland populations of the northern tropic, some of the extreme negro type, rude, and wild, and others only subtypically negro, and modified by Mohammedanism. The chief divisions here are philological; the physical differences lying within narrow limits. Neither does anything but language separate them from the western tribes. On the other hand, there are great gaps in the geography.

The Haussa, Timbutcu, Adamouan, Bornu, Begharmi, Mandara, Mobba, Furian (Dorfurian), and Kodogi (Kordofean) groups, of unascertained value, and chiefly philological, lead us across the drainage of the Lake Tshad to that of the Nile, where the phenomena of the Gambia repeat themselves in a very marked manner. There are the analogues of the Fulahs in the mountain districts and plateaux of Abyssinia, those of the Wolofs, Mandingos, and smaller tribes like the Felip on the alluvial soils. The negroes are chiefly those of the Kordofan frontier, of Sennar, and of the lower level of Abyssinia—Shelluk, Denka, Tumali, Shabun, Fertit, Qamamyl, Dalla, Doba, Gonga, &c. In Abyssinia some of the languages are decidedly Semitic, e.g. Gheez, others Sub-Semitic, as the Agow, with transitional or intermediate forms. In the elevated table-lands Ethnology, the skin is brown rather than black, and the features Arab or Persian rather than negro. There are also features comparatively Arab or Persian, with black complexions. An early Christianity, as well as Mohammedanism, has taken root in Abyssinia. The Galla, Nubian, and Bishori divisions (of unascertained value) give us notable deviations from the extreme negro type—the hair being long, the nose often curved, the face elongated.

On the east coast there is a certain number of tribes yet to be accurately distributed between the Kafre and the Abyssinian stocks, just as on the west there are (in the parts about the Old Calabar river) some similar tribes whose relations to the populations north and south of them are equivocal.

The last division of the African stock are—

E. The Hottentot.—With the extreme varieties of this division the physical conformation departs widely from even the negro standard. The stature is low, limbs slight, cheek bones prominent, zygomatic arches bent outwards (giving a Mongoloid form to the face), hair in tufts, eyes oblique. In the female organs of generation the nymphæ are sometimes enormously developed—the buttocks being steatomatous. The pelvis exhibits a maximum difference according as it is male or female—the male being strong and dense, the female light. In both cases the diploe between the bony plates is small in amount, the ossa ili vertical, the sacrum narrow, the conjugate diameter short. The neck of the thigh bone is short, with an oblique direction. Such, at least, are the results of the researches of Vrolik. The cranium is dolichocephalic, i.e., according to the terminology of Retzius, more remarkable for its fore-and-aft than for its side-to-side diameter. In this it is African rather than Mongol—it being Africa and Mongolia that supply the two extreme forms of the dolichocephalic and brachycephalic cranium. The language contains more than one articulate sound, described as the click.

The general habits of the Hottentot family are pastoral; the inferior members of it, the Saabs or Bushmen of the Karroo country being hunters—in many cases in the still lower condition of the Australian. The Hottentot has a better claim to be considered as a separate species of the genus Homo than any other section of our kind. Further remarks on this statement will be found in Language (Notes). All that will be done at present will consist in a short notice of the two extremes of the group. The Saabs are the lowest in civilization and the smallest in size, their country being the district between the Roggeveld and the middle part of the Orange River—the most unfavoured spot, perhaps, in Africa. Their skin is tawny rather than black, their language either wholly or nearly unintelligible to the Hottentots. The Koran Hottentots continually encroach on them, and hunt them down like beasts. Of the Hottentots of the Cape frontier the Koranas are the best-looking. Their claims, however, to stand as the models of the family are modified by the recent researches of Galton in the parts east of Walvisch Bay, and the explorers of Lake Ngami. The Hottentot tribes reach thus far north, where they are the equals, and, in some cases, the superiors of the Kafre populations in frame and daring. On the other hand, where the external conditions of aliment, security, and freedom are unfavourable, there is a visible degeneracy. The same has been remarked at the Cape. According to Mr Thompson, a division of the Koranas, who lost their herds, rapidly approached the Bushman type. The Griquas are a mixed breed—Dutch and Hottentot.

Such is the classification which, in the opinion of the present writer, best represents the divisions of the human species. He is well aware that the value of the classes is unequal, and that the nomenclature is imperfect. On the other hand, he believes that both the arrangement and the phraseology are sufficient for all questions likely to arise out of our present data, and also that they give a good groundwork for future investigations. That they are based upon a certain amount of assumption is clear—in other words, they are more or less hypothetical. In the eyes of several respectable authorities, they assume the chief point in question, viz., the specific unity of man, and the division of the species constituted by the men of present and past times into varieties and sub-varieties. What if all these varieties be species, and the species a genus? To this objection the writer will forthwith address himself—premising his remarks by an explanation of the point of view from which he has taken the preceding outline.

1. That all the languages of the earth's surface have had either one common origin, or else been, one and all, largely modified by some single language, he is convinced; due allowance having been made for a large amount of similarity independent of any ethnological connection. This means that a large number of words in different languages will be like each other, not because two or more tongues lent or borrowed certain words, nor yet because they all came from a common mother tongue, but because the human organism, under certain conditions, acts according to certain laws.

2. The similarity between languages, which is not thus explained, is held to be primâ facie of a common descent on the part of those members of the human kind that speak them—primâ facie evidence, and that of a strong sort.

3. That this primâ facie evidence will not be over-ridden by the conflicting phenomena of physical and moral differences is the writer's impression, an impression that has been forced upon him less by the small amount of such differences (for he freely admits that, with the ordinary interpretation of the term, they are in some cases specific), than by the extent to which they may be referred to certain physical and other influences. That these effects something is admitted by all who have written on the subject. That they effect more than any single supporter of the multiplicity of the species of the genus Homo has recognised, is certain—at any rate we may seek in vain for the naturalist who has taken in half the complications of the question. Thus—

That the physical characters of our kind are permanent, is argued from such a fact as the mummies of ancient Egypt giving us the same organization that is given by the modern Copts. Why should they not? The country in which they occur is as truly the valley of a certain river, between a certain degree of latitude, now as ever. The only facts on this question that are other than irrelevant are those that show that 2000 years under a change of climates go for nothing.

Again—that comparatively light complexions are found in some of the hottest parts of the world, whereas dark ones are found in comparatively cool ones, passes for an argument against the effects of heat on the skin. It only shows that heat is not the only cause of an increased secretion of the epidermic colouring matter.

Moisture—light—altitude—what has been done, worthy of the name of science, in the investigation of these influences—even singly? Next to nothing. Let it be granted that a little is known about the acclimatization of new-comers in fresh habitats. The process by which a world is peopled by the gradual diffusion of a given population from the circumference of an ever-increasing circle, has never been imitated, and never will be.

Then there is the question of descent. That two localities, one in Africa, and the other in South America, may so closely resemble each in their physical conditions of heat, light, altitude, moisture, land-and-water relations, as, for the purposes connected with the modification of the human organism, to be considered as identical, is highly probable. What, however, if the human organisms thereof notably differ? Not the inference that physical conditions either act irregularly, or not at all, but that the objects on which Ethnology; they acted in cases under notice were different. The negro that from (say) Central Asia reaches the Lower Niger is the descendant of ancestors whose organizations were acted upon by the physical influences of a line drawn through southwestern Asia and north-eastern Africa, whilst the Indian of the Lower Amazons is the descendant of ancestors whose organizations were acted upon by the physical influences of a line drawn through Siberia, the Arctic Circle, North America, Central America, and the north-western parts of South America.

The phenomenon of transitional and intermediate forms has again been greatly neglected by the advocates of the doctrines under notice.

However, as all this and a great deal more may be worked out with greater care than has hitherto been applied, the doctrine that the species of mankind are numerous is still an open question. Even if what we call the dynamic question (the question as to the power of external influences) be settled in favour of the unity of man, there will still be many plausible objections that a little ingenuity may discover. To say nothing about the extent to which the ordinary naturalist views of species may be modified, the value of language as an instrument of classification would be materially altered by any facts which might lead us to believe in the existence of a species of our kind, originally destitute of language, but not destitute of the capability of learning it when heard from others. Again, the value of transitional or intermediate forms as connecting links between extreme types depends greatly upon the character of the facts connected with hybridism and intermixture.

At present, however, we may be satisfied with the doctrines lately laid down, viz.:

1. That, as a matter of fact, the languages of the earth's surface are referable to one common origin.

2. That, as a matter of logic, this common origin of language is prima facie evidence of a common origin for those who speak it.

For the purposes of exposition, and even for those of investigation, the speculative question as to the unity or non-unity of our species is of far less importance than it seems to be. With ethnology, as with other studies, the bulk of its ordinary results are but slightly affected by the hypotheses that apply to the remoter and more abstruse questions of its obscurest departments. Even if we multiply species to the utmost, the unity classification gives us something. It tells us how language spread, even if it fails to tell us that the populations who spoke that language were specifically the same. Or, if we account for the phenomena of transitional and intermediate forms by the doctrine of hybridism and intermixture, the elements of this hybridism and intermixture are indicated. On the other hand, however, there are grave differences between the classes of the two systems. As long as the species is single our questions are questions of descent. Where, however, we have several species, descent is out of the question, and the liklest groups are those that have the most points of resemblance. Upon the whole, however, the two views give us like results than we expect a priori. This is because the arrangement of varieties closely approaches the arrangement of species.

To understand this, lay out of the question the islands of the earth's surface, and ask how an ethnologist upholding the unity of our species would people the world. He deduces his population from some point more or less central. His area, being continental, is continuous, and he supposes the stream of population by which its several portions were occupied to have been continuous also. The spread is that of circles on a still piece of water. Now, if so, all changes must have been gradual, and all extreme forms must have passed into each other by means of a series of transitional ones.

It is clear that such forms, when submitted to arrange-

ment and classification, will not come out in any definite Ethnology and well-marked groups like the groups that constitute what is currently called species. On the contrary, they will run into each other, with equivocal points of contact, and indistinct lines of demarcation; so that discrimination will be difficult, if not impracticable. If practicable, however, it will be effected by having recourse to certain typical forms, around which such as approximate most closely can most accurately and conveniently be grouped. When this is done, the more distant outliers will be distributed over the debatable ground of an equivocal frontier. In short, varieties as opposed to species imply transitional forms, whilst transitional forms preclude definite lines of demarcation.

Yet what is the actual classification of the varieties of mankind, and what is the current nomenclature? To say the least, it is very like that of a collection of species. Blumenbach's Mongolian, Blumenbach's Caucasian, Blumenbach's Ethiopian, are all terms that suit the nomenclature of the naturalist of genera. Nevertheless, however much it may give us of broad and trenchant lines of demarcation between varieties which (ex vi termini) ought to graduate into each other, it is far from being indefensible.

Man conquers man, and occupant displaces occupant on the earth's surface. By this means forms and varieties which once existed become extinct. The more this extinction takes place, the greater is the obliteration of those transitional and intermediate forms which connect extreme types; and the greater this obliteration, the stronger the lines of demarcation between geographically contiguous families. Hence a variational modification of a group of individuals simulates a difference of species; forms which were once wide apart being brought into juxtaposition by means of the annihilation of the intervening transitions. Hence what we of the nineteenth century—ethnologists, politicians, naturalists, and the like—behold in the way of groups, classes, tribes, families, &c., is beheld to a great extent under the guise of species; although they may not be so in reality, and although they might not have been so had we been witnesses to that earlier condition of things when one variety graduated into another and the integrity of the chain of likeness was intact.

A group is sharply defined simply because we know it in its state of definitude; a state of definitude which has been brought about by a displacement and obliteration of transitional forms. An ethnologist, then, may think with advocates of the unity of mankind, and employ the nomenclature of their opponents; in other words, the chief groups into which mankind is distributed are much the same in their relations to each other, whatever may be the opinion as to their absolute value and importance.

The last point to which the attention of the advocate of the plurality of the human species is directed is that of time, it being clear that in any estimation of the effect of physical and other causes upon the human organism, the duration of their action is an all-important element. The changes that ten generations fail to effect may be effected by ten times ten. Now, it is too often assumed, that because an investigator supports the origin of mankind from a single pair, he also supports the recent origin of that pair, and by so doing, limits the duration of the forces by which he supposes the existing differences to have been effected to a historical instead of a geological period. The two doctrines, however, by no means necessarily go together.

The consideration of the amount of change that external and other causes have effected, leads to the consideration of the changes that are likely to be effected hereafter, to questions as to the acclimatization of colonies in foreign parts, the fitness of the whiter varieties of our kind for the tropics, the fitness of the darker varieties of our kind for the colder portions of the earth's surface, and a long range of similar investigations. Whether, and under what conditions, an Englishman can become a permanent occupant of such countries as Brazil or India, is a truly ethnological problem, and also one of no mean importance. Akin to this is the valuation of the effects of civilization, and the moral influences that it engenders.

Hence, the chief ethnological problems, generally expressed, are those of (1.) the unity (or non-unity); (2.) the geographical origin (or origins); (3.) the antiquity; and (4.) the futurity of mankind—questions all closely connected with each other, and all mutually illustrative of each other—questions to which the phenomena of classification and the several ethnological methods are subordinate. In the extent to which the whole subject is new, we get the measure of the amount of thought and research requisite for even an approximation to any legitimate ethnological hypothesis upon these large and complicated questions.