IN his presidental address to the Anthropological Section of the British Association, Prof. Sayce dealt with dress as indicating certain racial facts. As he spoke of the high antiquity of the Highland dress, we quote this portion of his address, for it is both of interest and importance.

There are few things about which a population—more especially in an early stage of society—is so conservative as the matter of dress. When we find the Egyptian sculptor representing the Hittites of the warm plains of Palestine clad in the snowshoes of the mountaineer, we are justified in concluding that they must have descended from the ranges of the Taurus, where the bulk of their brethren continued to live, just as the similar shoes with turned-up ends which the Turks have introduced among the upper classes of Syria, Egypt, and northern Africa, point to the northern origin of the Turks themselves. Such shoes are utterly unsuited for walking in over a country covered with grass, brushwood, or even stones; they are, on the contrary, admirably adapted. for walking on snow. Now, the dress of Celtic Gaul and of southern Britain, also, when the Romans first became acquainted with it, was the same as the dress which "linguistic palaeontology" teaches us had been worn by the primitive Aryans' in their first home. One of its chief constituents were the braccae, or trousers, which accordingly became to the Roman the symbol of the barbarian. We learn, however, from sculptures and other works of art that before the retirement of the Romans from the northern part of Europe, they had adopted this article of clothing, at all events during the winter months. That the natives of southern Britain continued to wear it after their separation from Rome is clear from a statement of Gildas ("Hist., 19 "), in which he refers in no flattering terms to the kilt of the Pict and the Scot. Yet from within a century after the time of Gildas, there are indications that the northern kilt, which he regards as so strange and curious, had become the common garb of Wales. When we come to the 12th century, we find that it is the National costume. Giraldus Cambrensis gives us a description of the Welsh dress in his own time, from which we learn that it consisted simply of a tunic and plaid. it was not until the age of the Tudors, according to Lluyd, the Welsh historian of the reign of Elizabeth, that the Welsh exchanged their own for the English dress. The Welsh, who served in the army of Edward II. at Bannockburn, were remarked even by the Lowland Scotch, for the scantiness of their attire, and we have evidence that it was the same a century later. If we turn to Ireland we find that in the days of Spenser, and later, the National costume of the Irish was the same as that of the Welsh and the Highland Scotch. The knee-breeches and sword-coat, which characterise the typical Irishman in the comic papers, are survivals of the dress worn by the English at the time when it was adopted in Ireland. The Highland dress, therefore, was once worn not only in the Scotch Highlands and in Ireland, but also in Wales. It characterised the Celtic parts of Britain, with the exception of Cornwall and Devonshire. Yet we have seen that up to the middle of the 6th century, at the period when Latin was still the language of the fellow-countrymen of Gildas, and when "Cunedda's men" had not as yet imposed their domination upon Wales, the old Celtic dress with trousers must have been the one in common use. Now, we can easily understand how a dress of the kind could have been replaced by the kilt in warm countries like Italy and Greece; what is not easily conceivable is that such a dress could have been replaced by the kilt in the cold regions of the north. In warm climates a lighter form of clothing is readily adopted; in cold climates the converse is the case. I see, consequently, but one solution of the problem before us. On the one hand, there was the distinctive Celtic dress of the Roman age, which was the same as the dress of the primitive Aryan, and was worn alike by the Celts of Gaul and Britain and the Teutons of Germany; on the other hand there was the scantier and colder dress which originally characterized the coldest part of Britain and subsequently mediaeval Wales also. Must we not infer, in the first place, that the aboriginal population of Caledonia and Ireland was not Celtic—or at least not Aryan Celtic; and, secondly, that the dominant class in Wales after the 6th century came from that northern portion of the island where the kilt was worn? Both inferences, at all events, agree with the conclusions which ethnologists and historians have arrived at upon other grounds.