184 QN THE TOTAL AND PARTIAL 



been said, that as the modern northern animals migrate, the Siberian 

 elephant may also have shifted his place during the inclemency of the 

 season ; but this conjecture seems forced, even in regard to the elephant, 

 and still more so when applied to the Siberian rhinoceros, found in the 

 frozen gravel of that country, as animals of this genus are heavy and 

 slow in their motions, and can hardly be supposed to have accomplished 

 great periodical migrations to southern latitudes. That the mammoth, 

 however, continued for a long time to exist in Siberia after the winter 

 had become extremely cold, is demonstrable, since their bones are found 

 in icebergs, and in the frozen gravel, in such abundance as could only 

 have belonged to herds which lived at one time in the district, even if 

 those northern countries had once been clothed with vegetation as luxu? 

 riant as that of an Indian jungle. But if we suppose the change to 

 have been extremely slow, and to have consisted not so much in a 

 diminution of the mean annual temperature, as in an alteration from 

 what has been termed an "insular" to an "excessive" climate, — from 

 one in which the temperature of winter and summer were nearly equal- 

 ised, to one wherein the seasons were violently contrasted, — we may, 

 perhaps, explain the phenomenon. Siberia, and other arctic regions, 

 after having possessed for ages a more uniform temperature, may, after 

 certain changes in the form of the arctic land, have become occasionally 

 exposed to extremely severe winters. When they first occurred at 

 distant intervals, the drift snow would fill the valleys, and herds of 

 herbivorous quadrupeds would be surprised and buried in a frozen mass, 

 as often happens to cattle and human beings, overwhelmed in the alpine 

 valleys of Switzerland by avalanches. When valleys have become filled 

 with ice, as those of Spitzbergen, the contraction of the mass causes 

 innumerable deep rents, such as are seen in the Mer de Glace on 

 Mont Blanc. These deep crevices usually become filled with loose 

 snow; but sometimes a thin covering is drifted across the mouth of the 

 chasm, capable of sustaining a certain weight. Such treacherous bridges 

 are liable to give way when heavy animals are crossing, which are then 

 precipitated at once into the body of a glacier, which slowly descends 

 to the sea, and becomes a floating iceberg. As bears, foxes, and deer 

 now abound in Spitzbergen, we may confidently assume that the im- 

 pedding of animal remains in the glaciers of that island must be an 

 event of almost annual occurrence. The conversion of drift snow into 

 permanent glaciers and icebergs, when it happens to become covered 

 over with alluvial matter, transported by torrents and floods, is by no 

 means a rare phenomenon in the arctic regions. During a series of 



