and of the selective influence of their surroundings operating on the mixture.

The assumption that, as there must have been a primitive Aryan people, in the philological sense, so that people must have constituted a race in the biological sense, is pretty generally made in modern discussions of the Aryan problem. But whether the men of the primitive Aryan race were blonds or brunets, whether they had long or round heads, were tall or were short, are hotly debated questions, into the discussion of which considerations quite foreign to science are sometimes imported. The combination of swarthiness with stature above the average and a long skull, confer upon me the serene impartiality of a mongrel; and, having given this pledge of fair dealing, I proceed to state the case for the hypothesis I am inclined to adopt. In doing so, I am aware that I deliberately take the shilling of the recruiting sergeant of the Light Brigade, and I warn all and sundry that such is the case.

Looking at the discussions which have taken place from a purely anthropological point of view, the first point which has struck me is that the problem is far more complicated and difficult than many of the disputants appear to imagine; and the second, that the data upon which we have to go are grievously insufficient in extent and in precision. Our historical records cover such an infinitesimally small extent of the past life of humanity, that we obtain little help from them. Even so late as 1500 B.C., northern Eurasia lies in historical darkness, except for such glimmer of light as may be thrown here and there by the literatures of Egypt and of Babylonia. Yet, at that time, it is probable that Sanskrit, Zend, and Greek, to say nothing of other Aryan tongues, had long been differentiated from primitive Aryan. Even a thousand years later, little enough accurate information is to be had about the racial characters of the European and Asiatic tribes known to the Greeks. We are thrown upon such resources as archæology and human palæontology have to offer, and notwithstanding the remarkable progress made of late years, they are still meagre. Nevertheless, it strikes me that, from the purely anthropological side, there is a good deal to be said in favor of the two propositions maintained by the new school of philologists; first, that the

people who spoke" primitive Aryan'' were a distinct and well-marked race of mankind; and, secondly, that the area of the distribution of this race, in primeval times, lay in Europe, rather than in Asia.

For the last two thousand years, at least, the southern half of Scandinavia and the opposite or southern shores of the Baltic have been occupied by a race of mankind possessed of very definite characters. Typical specimens have tall and massive frames, fair complexions, blue eyes, and yellow or reddish hair-that is to say, they are pronounced blonds. Their skulls are long, in the sense that the breadth is usually less, often much less, than four-fifths of the length, and they are usually tolerably high. But in this last respect they vary. Men of this blond, long-headed race abound from eastern Prussia to northern Belgium ; they are met with in northern France and are common in some parts of our own islands. The people of Teutonic speech, Goths, Saxons, Alemanni, and Franks, who poured forth out of the regions bordering the North Sea and the Baltic, to the destruction of the Roman Empire, were men of this race; and the accounts of the ar cient historians of the incursions of the Gauls into Italy and Greece, between the fifth and the second centuries B.C., leave little doubt that their hordes were largely if not wholly, composed of similar men. The contents of numerous interments in southern Scandinavia prove that, as far back as archæology takes us into the socalled neolithic age, the great majority of the inhabitants had the same stature and cranial peculiarities as at present, though their bony fabric bears marks of somewhat greater ruggedness and savagery. There is no evidence that the country was occupied by men before the advent of these tall, blond long-heads. But there is proof of the presence, along with the latter, of a small percentage of people with broad skulls; skulls, that is, the breadth of which is more, often very much more, than four-fifths of the length.

At the present day, in whatever direction we travel inland from the continental area occupied by the blond long-heads, whether south west, into central France; south, through the Walloon provinces of Belgium into eastern France; into Switzerland, South Germany, and the Tyrol; or south-east, into Poland and Russia; or north, into Finland and Lapland, broad

heads make their appearance, in force, among the long-heads. And, eventually, we find ourselves among people who are as regularly broad-headed as the Swedes and North Germans are long-headed. As a general rule, in France, Belgium, Switzerland, and South Germany, the increase in the proportion of broad skulls is accom panied by the appearance of a larger and larger proportion of men of brunet complexion and of a lower stature; until, in central France and thence eastward, through the Cevennes and the Alps of Dauphiny, Savoy, and Piedmont, to the western plains of North Italy, the tull blond long-heads* practically disappear, and are replaced by short brunet broadheads. The ordinary Savoyard may be described in terms the converse of those which apply to the ordinary Swede. He is short, swarthy, dark-eyed, dark-haired, and his skull is very broad. Between the two extreme types, the one seated on the shores of the North Sea and the Baltic, and the other on those of the Mediterranean, there are all sorts of intermediate forms, in which breadth of skull may be found in tall and in short blond men, and in tall brunet men.

There is much reason to believe that the brunet broad-heads, now met with in central France and in the west central European highlands, have inhabited the same region, not only throughout the historical period, but long before it commenced; and it is probable that their area of occupation was formerly more extensive. For, if we leave aside the comparatively late incursions of the Asiatic races, the centre

I may plead the precedent of the good English words "block-head" and "thickhead" for broad-head" and " long-head,' but I cannot say that they are elegant. might have employed the technical terms brachycephali and dolichocephali. But it cannot be said that they are much more graceful; and, moreover, they are sometimes employed in senses different from that which I have given in the definition of broad-heads and long-heads. The cephalic index is a number which expresses the relation of the breadth to the length of a skull, taking the latter as 100. Therefore broad heads" have the cephalic index above 80 aud "long-heads" have it below 80. The physiological value of the difference is unknown; its morphological value depends upon the observed fact of the constancy of the occurrence of either long skulls or broad skulls among large bodies of mankind.

of eruption of the invaders of the southern moiety of Europe has been situated in the north and west. In the case of the Teutonic inroads upon the Empire of Rome, it undoubtedly lay in the area now occupied by the blond long-heads; and, in that of the antecedent Gaulish invasions, the physical characters ascribed to the leading tribes point to the same conclusion. Whatever the causes which led to the breaking out of bounds of the blond long-heads, in mass, at particular epochs, the natural increase in numbers of a vigorous and fertile race must always have impelled them to press upon their neighbors, and thereby afford abundant occasions for intermixture. If, at any given pre-historic time, we suppose the lowlands verging on the Baltic and the North Sea to have been inhabited by pure blond long-heads, while the central highlands were occupied by pure brunet short-heads, the two would certainly meet and intermix in course of time, in spite of the vast belt of dense forest which extended, almost uninterruptedly, from the Carpathians to the Ardennes and the result would be such an irregular gradation of the one type into the other as we do, in fact, meet with.

On the south-east, east, and north-east, throughout what was once the kingdom of Poland, and in Finland, the preponderance of broad heads goes along with a wide prevalence of blond complexion and of good stature. In the extreme north, on the other hand, marked broad-headedness is combined with low stature, swarthiness, and more or less strongly mongolian features, in the Lapps. And it is to be observed that this type prevails increasingly to the eastward, among the central Asiatic populations.

The population of the British Islands, at the present time, offers the two extremes of the tall blond and the short brunet types. The tall blond long-heads resemble those of the continent; but our short brunet race is long-headed. Brunet broadheads, such as those met with in the central European highlands, do not exist among us. This absence of any considerable number of distinctly broad headed people (say with the cephalic index above 81 or 82) in the modern population of the United Kingdom is the more remarkable, since the investigations of the late Dr. Thurnam, and others, proved the existence of a large proportion of tall broad-heads among

the people interred in British tumuli of the neolithic age. It would seem that these

broad-skulled immigrants have been absorbed by an older long-skulled population; just as, in South Germany, the longheaded Alemanni have been absorbed by the older broad-heads. The short brunet long-heads are not.peculiar to our islands. On the contrary, they abound in western France and in Spain, while they predominate in Sardinia, Corsica, and South Italy, and, it may be, occupied a much larger area in ancient times.

Thus, in the area which has been under consideration, there are evidences of the existence of four races of men-(1) blond long-heads of tall stature, (2) brunet broad-heads of short stature, (3) mongoloid brunet broad-heads of short stature, (4) brunet long-heads of short stature. The regions in which these races appear with least admixture are-(1) Scandinavia, North Germany, and parts of the British Islands; (2) central France, the central European highlands, and Piedmont; (3) Arctic and eastern Europe, central Asia; (4) the western parts of the British Islands and of France; Spain, South Italy. And the inhabitants of the regions which lie between these foci present the intermediate gradations, such as short blond long-heads, and tall brunet short-heads and long-heads which might be expected to result from their intermixture. The evidence at present extant is consistent with the supposition that the blond long-heads, the brunet broad-heads, and the brunet long-heads have existed in Europe throughout historic times, and very far back into pre-historic times. There is no proof of any migration of Asiatics into Europe, west of the basin of the Dnieper, down to the time of Attila. On the contrary, the first great movements of the European population of which there is any conclusive evidence is that series of Gaulish invasions of the east and south, which ultimately extended from North Italy as far as Galatia in Asia Minor.

It is now time to consider the relations between the phenomena of racial distribution, as thus defined, and those of the distribution of languages. The blond longheads of Europe speak, or have spoken, Lithuanian, Teutonic, or Celtic dialects, and they are not known to have ever used any but these Aryan languages. A large

proportion of the brunet broad heads once spoke the Ligurian and the Rhætic dialects, which are believed to have been nonAryan. But, when the Romans made acquaintance with Transalpine Gaul, the inhabitants of that country between the Garonne and the Seine (Cæsar's Celtica) seem, at any rate for the most part, to have spoken Celtic dialects. The brunet long-heads of Spain and of France appear to have used a non Aryan language, that Euskarian which still lives on the shores of the Bay of Biscay. In Britain there is no certain knowledge of their use of any but Celtic tongues. What they spoke in the Mediterranean islands and in South Italy does not appear.

The blond broad-heads of Poland and West Russia form part of a people who, when they first made their appearance in history, occupied the marshy plains imperfectly drained by the Vistula, on the west, the Duna, on the north, and the Dnieper and Bug, on the south. They were known to their neighbors as Wends, and among themselves as Serbs and Slavs. The Slavonic languages spoken by these people are said to be most closely allied to that of the Lithuanians, who lay upon their northern border. The Slavs resemble the South Germans in the predominance of broad-heads among them, while stature and complexion vary from the, often tall, blonds who prevail in Poland and Great Russia to the, often short, brunets common elsewhere. There is certainly nothing in the history of the Slav people to interfere with the supposition that, from very early times, they have been a nixed race. For their country lies between that of the tall blond long-heads on the north, that of the short brunet broad-heads of the European type on the west, and that of the short brunet broadheads of the Asiatic type on the east: and, throughout their history, they have either thrust themselves among their neighbors, or have been overrun and trampled down by them. Gauls and Goths have traversed their country, on their way to the east and south: Finno-tataric people, on their way to the west, have not only done the like, but have held them in subjection for centuries. On the other hand, there have been times when their western frontier advanced beyond the Elbe; indeed, it is asserted that they have sent colonies to Holland and even as far as

shern England. A large part of eastern Germany; Bohemia, Moravia, Hungary; the lower valley of the Danube and the Balkan peninsula, have been largely or completely Slavonized; and the Slavonic rule and language, which once had trouble to hold their own in West Russia and Little Russia, have now extended their sway over all the Finno-tataric populations of Great Russia; while they are advancing, among those of central Asia, up to the frontiers of India on the south and to the Pacific on the extreme east. Thus it is hardly possible that fewer than three races should have contributed to the formation of the Slavonic people; namely, the blond long-heads, the European brunet broad-heads, and the Asiatic brunet broadheads. And, in the absence of evidence to the contrary, it is certainly permissible to suppose that it is the first race which has furnished the blond complexion and the stature observable in so many, especially of the northern Slavs, and that the brunet complexion and the broad skulls must be attributed to the other two. But, if that supposition is permissible, then the Aryan form and substance of the Slavonic languages may also be fairly supposed to have proceeded from the blond longheads. They could not have come from the Asiatic brunet broad-heads, who all speak non-Aryan languages; and the presumption is against their coming from the brunet broad-heads of the central European highlands, among whom an apparently nonAryan language was largely spoken, even in historical times.

In the same way, the tall blond tribes among the Fins may be accounted for as the product of admixture. The great majority of the Finno-tataric people are brunet broad-heads of the Asiatic type. But that the Fins proper have long been in contact with Aryans is evidenced by the many words borrowed from Aryan which their language contains. Hence there has been abundant opportunity for the mixture of races; and for the transference to some of the Fins of more or fewer of the physical characters of the Aryans and vice versa. On any hypothesis, the frontier between Aryan and Finno tataric people must have extended across west-central Asia for a very long period; and, at any point of this frontier, it has been possible that mixed races of blond Fins or of brunet Aryans should be formed.

So much for the European people who now speak Celtic, or Teutonic, or Slavonian, or Lithuanian tongues; or who are known to have spoken them, before the supercession of so many of the early native dialects by the Romance modifications of the language of Rome. With respect to the original speakers of Greek and Latin, the unravelling of the tangled ethnology of the Balkan peninsula and the ordering of the chaos of that of Italy are enterprises upon which I do not propose to enter. In regard to the first, however, there are a few tolerably satisfactory data. The ancient Thracians were proverbially blue-eyed and fair-haired. Tall blonds were common among the ancient Greeks, who were a long-headed people; and the Sphakiots of Crete, probably the purest representatives of the old Hellenes in existence, are tall and blond. But considering that Greek colonization was taking place on a great scale in the eighth century B. C., and that, centuries earlier and later, the restless Hellene had been fighting, trading, plundering and kidnapping, on both sides of the Agean, and perhaps as far as the shores of Syria and of Egypt, it is probable that, even at the dawn of history, the maritime Greeks were a very mixed тасе. On the other hand, the Dorians may well have preserved the original type; and their famous migration may be the earliest known example of those movements of the Aryan race which were, in later times, to change the face of Europe. Analogy perhaps justifies a guess, that those ethnological shadows, the Pelasgi, may have been an earlier mixed population, like that of Western Gaul and of Britain before the Teutonic invasion. At any rate, the tall blond long-heads are so well represented in the oldest history of the Balkan peninsula, that they may be credited with the Aryan languages spoken there. And it may be that the tradition which peopled Phrygia with Thracians represents a real movement of the Aryan race into Asia Minor, such as that which in after years carried the Gauls thither.

The difficulties in the way of a probable identification of the people among whom the various dialects of the Latin group developed themselves, with any race traceable in Italy in historical times, are very great. In addition to the Italic "aborigines" northern Italy was peopled by Ligurian brunet broad-heads; with Gauls,

probably, to a large extent, blond longheads; with Illyrians, about whom nothing is known. Besides these, there were those perplexing people the Etruscans, who seem to have been, originally, brunet long-heads. South Italy and Sicily present a contingent of "Sikels," Phoenicians and Greeks; while over all, in comparatively modern times, follows a wash of Teutonic blood. The Latin dialects arose, no one knows how, among the tribes of Central Italy, encompassed on all sides by people of the most various physical characters, who were gradually absorbed into the eternally widening maw of Rome, and there, by dint of using the same speech, became the first example of that wonderful ethnological hotch-potch miscalled the Latin race. The only trustworthy guide here is archæological investigation. great advance will have been made when the race characters of the pre-historic people of the terremare (who are identified by Helbig with the primitive Umbrians) become fully known.

A

I cannot learn that the ancient literatures of India and of Persia give any definite information about the complexion of the Indo Iranians, beyond conveying the impression that they were what we vaguely call white men. But it is important to note that tall blond people make their appearance sporadically among the Tadjiks of Persia and of Turkestan; that the Siahposh and Galtchas of the mountainous barrier between Turkestan and India are such; and that the same characters obtain largely among the Kurds on the western frontier of Persia at the present day. The Kurds and the Galtchas are generally broadheaded, the others are long-headed. These people and the ancient Alans thus form a series of stepping-stones between the blond Aryans of Europe and those of Asia, standing up amid the flood of Finno-tataric people which has inundated the rest of the interval between the sources of the Dnieper and those of the Oxus. If only more was known about the Sarmatians and the Scythians of the oldest historians, it is not improbable, I think, that we should discover that, even in historical times, the

* Die Italiker in der Poebene, 1879. See for much valuable information respecting the races of the Balkan and Italic peninsula, Zampa's essay, “Vergleichende Anthropologische Ethnographie von Apulien," Zeitschrift für Elhnologie, xviii., 1886.

area occupied by the blond long-heads of Aryan speech has been, at least temporarily, continuous from the shores of the North Sea to central Asia.

Suppose it to be admitted, as a fair working hypothesis, that the blond longheads once extended without a break over this vast area, and that all the Aryan tongues have been developed out of their original speech, the question respecting the home of the race when the various families of Aryan speech were in the condition of inceptive dialects remains open. For all that, at first, appears to the contrary, it may have been in the west, or in the east, or anywhere between the two. In seeking for a solution of this obscure problem, it is an important preliminary to grasp the truth that the Aryan race must be much older than the primitive Aryan speech. It is not to be seriously imag ined that the latter sprang suddenly into existence, by the act of a jealous Deity, apparently unaware of the strength of man's native tendency toward confusion of speech. But if all the diverse languages of men were not brought suddenly into existence, in order to frustrate the plans of the audacious bricklayers of the plain of Shinar; if this professedly historical statement is only another "type," and primitive Aryan, like all other languages, was built up by a secular process of develop. ment, the blond long heads, among whom it grew into shape, must for ages have been, philologically speaking, non Aryans, or perhaps one should say pro Aryans.' I suppose it may be safely assumed that Sanskrit and Zend and Greek were fully differentiated in the year 1500 B.C. so, how much further back must the existence of the primitive Aryan, from which these proceeded, be dated? And how much further yet, that real juventus mundi (so far as man is concerned) when primitive Aryan was in course of formation? And how much further still the differentiation of the nascent Aryan blond longhead race from the primitive stock of mankind?

If

If any one maintains that the blond long-headed people, among whom, by the hypothesis, the primitive Aryan language was generated may have formed a separate race as far back as the pleistocene epoch, when the first unquestionable records of man make their appearance, I do not see