which the lemmings and their arctic congeners had formerly prevailed. Throughout both tundra and steppe epochs Palæolithic man was an occupant of middle Europe. To the steppe epoch succeeded a forest epoch, with its characteristic fauna, by which time Palæolithic man had vanished, his place being taken by the so-called Neolithic race, or races, for there were several of these.
We must now ask what relation the tundra and steppes deposits bear to other well-known superficial accumulations of Europe. To what particular stage of the geological history of our continent do they belong? When we remember that an arctic-alpine flora formerly flourished on the low grounds of central Europe, it seems extremely probable that the tundra epoch must fall within the glacial period. But the glacial period embraced a complex series of geographical and climatic changes, and it is necessary, therefore, to come somewhat closer to the question. Among the most conspicuous deposits of the Ice age are moraines of all kinds and fluvio-glacial gravels, while the löss, as we have seen, is the most prominent accumulation of the tundra and steppe epochs. How, then, does the latter behave with regard to the typical glacial and fluvio glacial formations? Is it older or younger than these, or are the two sets of accumulations contemporaneous? The answer we get to that question is, at the first blush, disconcerting, for we learn that it is each in turn—sometimes underlying, sometimes overlying, and in other places occurring intercalated among glacial deposits. This only means, however, that löss appears to have been formed during different stages of the Ice age. It will be remembered that while we discussed the wind-blown character of the loss, we left untouched the question of the origin of its materials. Whence were those materials derived which the wind worked over, and largely rearranged, and redistributed in the low grounds of central Europe? To answer this question we must examine more closely the relation borne by the löss to the fluvio-glacial deposits and morainic accumulations. We note, in the first place, that in its horizontal distribution it follows closely that of the valley gravels of glacial times. Where the latter are well developed, the löss appears in full force; where they are wanting there is a like absence of löss. In all the valleys leading down from the Alps to the low grounds of middle Europe the loss puts in a prominent appearance. It obviously bears a close relation to the main lines of drainage, and may be said to be confined to valleys that head in formerly glaciated areas. So, again, in north Germany and southern Russia it spreads over all the low-lying tracts that lay in front of the vast mers de glace of glacial times. These facts alone, taken in connection with the occasional well-stratified character of the löss, the intercalation in it now and again of beds of sand, and the presence ever and anon of fresh-water shells, seem strongly suggestive of a fluviatile origin. And that such was really the origin of the materials of the löss will appear clear enough when we consider the conditions that obtained during a glacial epoch. (See Map A.)
