where, and that large swarms issued from a central bee-hive which contained untold millions of human beings. This may or may not have been so. But first of all we ought to remember that a common language is by no means a certain proof of a common beehive. We know from history how families, clans, and whole nations were conquered and led into captivity, learning the language of their conquerors; how tribes were exterminated, women and children carried off; and how even conquerors had sometimes to learn the language of the country which they had subdued. All this does not destroy the continuity of language, but it breaks the continuity of blood. If the indigenous races of India learnt Sanskrit, and dialects derived from Sanskrit, they became representatives of Aryan speech, whatever their blood may have been. Have not the Jews forgotten Hebrew, and learnt English, German, and French? Have not the Beauchamps and St. Legers broken their tongues to Saxon idiom and Saxon grammar in England? How then shall we tell what races had to learn the language of their Aryan conquerors or their Aryan slaves? There is no Aryan race in blood, but whoever, through the imposition of hands, whether of his parents or his foreign masters, received the Aryan blessing, belongs to that unbroken spiritual succession which began with the first apostles of that noble speech, and continues to the present day in every part of the globe.

And why should there have been in the beginning a vast number of Aryan men? Let us remember that one couple, having two children, would, if every successive marriage was blest with two children only, produce a population of 274,877,906,944 human beings in about 1200 years. Now the population of the whole earth at the present moment is vaguely estimated at 1,500 millions only. We are not driven, therefore, particularly if the first Aryan separation may be placed at least 2,000 years B.C., to the admission of a vast Aryan stock which was broken up into seven or more nationalities. That may be the more natural hypothesis, but whether more natural or not, it is not the only possible hypothesis. Granted two Aryan couples, each with seven children, and everything that has to be explained may be explained quite as well with this as with the bee-hive theory. Each of the seven children, by marrying children of the other family, might become, particularly if they settled in different forests or valleys, founders of dialects; and each of these dialects might, in twenty generations, or six hun

dred years, be spoken by more than two millions of human beings. Two millions of human beings, however, are much more difficult to move from one country to another than two hundred; and it is, at all events, quite open for us to imagine that the Aryan migrations took place by hundreds instead of by millions. If one missionary is able, in twenty years, to impose his peculiar, and perhaps not quite grammatical, dialect on the population of a whole island, why should not one shepherd, with his servants and flocks, have transferred his peculiar Aryan dialect from one part of Asia or Europe to another? This may seem a very humble and modest view of what was formerly represented as the irresistible stream of mighty waves rolling forth from the Aryan centre and gradually overflowing the mountains and valleys of Asia and Europe, but it is, at all events, a possible view; nay, I should say a view far more in keeping with what we know of recent colonisation.

But the old question returns, Can we not discover the cradle of our race? I say, decidedly we cannot. We may guess, with more or less probability, but if our guesses are to be submitted to the tests of mathematical certainty, not one of them will stand the test. This ought to be understood; and is, in fact, understood among most scholars. Many opinions held with regard to periods of history which are beyond the reach of historical evidence can never be more than possible or plausible. To demand for them a different character does not show any critical sagacity, but rather ignorance of the limits of our knowledge. Thus, when we see the Celts driven to the western parts of Europe, pushed forward by Teutonic tribes, and these again pressed hard by Slavonic neighbours, we naturally conclude that the Celts were the first to arrive in Europe, the Germans the second, the Slaves the third. But there is no mathematical certainty for this. It is nothing but the result of an historical combination, and can never be more. Again, if we see Hellenic civilisation extending from Asia Minor to Greece, and from Greece to Italy, and if we find the Italians pressed by successive inroads from the north, we are inclined here too to admit a progress of Aryan speech and thought from the east to the west, and from the north to the south. If, on the contrary, we consider that the Aryan conquerors of India came clearly from the north along the rivers of the Punjab, while before that time they must have dwelt for a certain period together

The first fact that was supposed to militate against it was the absence of common Aryan names for animals, which ought to have been known to dwellers in Asia, such as the lion, the elephant, the tiger, and the camel. The dog, for instance, must have been known to the Aryans before their separation, because it has the same name in Sanskrit, svan, in Zend, spá, in Greek, kov, in Latin, canis, in Irish, cú, in Lithuanian, szü, in Gothic, hunds. These are all dialectic modifications of one typical form kuan. But there are no common Aryan names for lion, elephant, tiger, and camel, and therefore, it is concluded, the Aryans could not, before their separation, have known these animals or lived in a country where they were known. This argument is ben trovato, but no more. First of all, if some perverse critic were to say that the Aryans may have possessed common names for these animals, but lost them, we might shrug our shoulders, but we could not prove the contrary. Ever so many Aryan words exist in one or two branches of the family only, and if they disappeared in some, they might have disappeared in all.

Indic

Secondly, and this is a more powerful argument, we find that the animals which have common Aryan names are mostly such as had become familiar and useful by being domesticated. The fact that the dog and the horse have common names in all Aryan languages seems to me the best proof that they had been tamed. Tigers and lions were simply wild beasts, and there was no necessity for distinguishing and naming them beyond classing them as fera or Sýp, the objects of the chase (pa). And suppose the camel had really been known and utilised as a beast of burden by the Aryans, when living in Asia, would it not have been most natural that, when transplanted to more northern regions, their children, who had never seen a camel, should have lost the name of it? We have no longer any doits, and the word would have been altogether lost but for such familiar phrases as "I care not a doit." The Americans have no pence, and in America penny is no longer an American, but only a foreign word.

But if we proceed to ask in what exact spot the Aryan centre has to be placed geographically, the answers will vary very considerably. "Somewhere in Asia," used to be the recognised answer, and I do not mean to say that it was far wrong; only we must not expect in a subject like this our much-vaunted mathematical certainty. The reasoning which we have to adopt is one that Mill recommends for other complicated and, at first sight, confused sets of appearances.* We have to begin by making any supposition, even a false one, to see what consequences will follow from it, and by observing how these differ from the real phenomena. The simplest supposition which accords with the more obvious facts is the best to begin with, because its consequences are the most easily traced. This rude hypothesis is then rudely corrected, and the operation repeated, and the comparison of the consequences deducible from the corrected hypothesis with the observed facts suggests still further correction, until the deductive results are at last made to tally with the phenomena.

Now the first rough hypothesis is that the cradle of the Aryans was somewhere in Asia, and the question is, do we know of any facts which make that rough hypothesis untenable?

"Logic," iv. 14, 5.

The negative argument, therefore, which, from the absence of common Aryan words, tries to establish the absence of their objects during the Aryan period, breaks down. It is plausible, it appeals to scholars and historians, but it must not be subjected to a crossexamination in a court of law. Much stronger, however, is the positive argument. If the

or less plausible case, as M. Piètrement has done for Siberia, and Dr. Penka for Scandinavia, as the true officina gentium Aricarum. Dr. Penka's arguments are extremely interesting. He tries to show that the picture which

But because the Aryans, before they branched off into North-Westerns and SouthEasterns, knew the bear, it does not follow that we must push their original home to the Arctic regions. Even the north of India may be called arctic in one sense, for the Great Bear is visible there; nor need we go to Arcadia or Germany in order to meet with bears. That the Aryans did not come from a very southern climate may indeed be gathered from their possessing common names for winter, such as Sanskrit hima, Greek xeuwv, Latin hiems, Old Slav zima, Irish gam." Ice, too, is represented by is in Teutonic, by isi in Iranic; snow by nix in Latin, vípa (acc) in Greek, snaivs in Gothic, sněgu in Old Slavonic, snigis in Old Irish, and çnizh, to snow, in Zend. There being a common name for the birchtree, bharga in Sanskrit, bereza in Russian, birke in German, likewise points to a more temperate climate. But none of these climatic indications drives us as yet out of Asia. I confess I was at first inclined ("Selected Essays," I. 187) to follow Benfey, when, on the strength of this and similar linguistic facts, he proposed to fix the original home of the Aryans on the very frontiers of Asia, "north of the Black Sea, from the mouth of the Danube to the Caspian Sea." Unfortunately, however, Benfey did not live to publish the fuller account of his researches which he had promised, and the arguments which other scholars have added, and by which they have tried to push the frontiers of the Aryan home as far as Germany, Scandinavia, and Siberia, seem to me to have rather weakened than strengthened The reasons which induced Geiger to proclaim Germany as the original home of the Aryans have not stood the criticism of unprejudiced scholars, though the evidence with which we have to deal is so pliant that it is possible to make out a more

his case.

life previous to the separation, coincides with the picture which archæological palæontology constructs from the neolithic remains discovered in Scandinavia. But how would this prove that the Aryans were autochthonous in Scandinavia? Even the identity of the flora and fauna of Scandinavia with the fauna and flora attested by the common Aryan language would not decide the question, unless we could prove that no other country could put forward the same claim. And are we to suppose that the original Aryan type was Scandinavian, and that our forefathers had blue eyes and blonde hair? Blue eyes and blonde hair are in such a minority among the Aryans that Pösche, in his book, "Die Arier," 1878, felt driven to place our ancestors near the Rokitno swamps in Russia, near the river Prypet, a branch of the Dniepr (Borysthenes), because in that locality depigmentation takes place most rapidly, not only with men but with plants and animals also.

On no subject have positive assertions been made with such assurance as on the home of the Aryans. If it is difficult to prove, it is equally difficult to disprove anything with regard to such distant times, and scientific imagination has therefore free scope to roam. Scholars, however, who know how thin the ice really is on which they have to skate, are not inclined to go beyond mere conjecture, and they tremble whenever they see their own fragile arguments handled so daringly by their muscular colleagues, the paleontologists and cranioscopists.

Dr. Penka, for instance, tells us that the Aryans had a common name for the sea, but he must know that this is one of the most contested points among scholars. I hold that aλs (fem.) meant, first, sea, afterwards, as a masc., salt. Others take the opposite view. I connect äλ-s and sal with sal-ila, water, in Sanskrit. Curtius objects to that derivation. I think that Benfey was right in assigning to Sanskrit sara the meaning of salt (see his last article in Sitzungsberichte der Göttinger anthropologischen Gesellschaft, 15 July, 1876), but I doubt whether this proves that sea-salt was known to the Aryans before they separated. If aλs meant salt, because it first meant the sea, no one can say that sara in