Page:A Treatise on Geology, volume 1.djvu/336

 320 of old or a general creation of new quadrupedal forms of life. The same appears to be true with reference to the buried forests so often associated with diluvial deposits. It is confirmed by the gradual change in the proportion of existing among extinct species of tertiary shells; so that the most recent groups of tertiary strata contain 40 to 90 per cent, of living forms, while among a dozen or twenty shells in the gravel of Holderness one extinct species is met with.

On the other hand, it must be remembered, that no palæotheria, lophiodontes, or other genera, chiefly belonging to the older tertiary genera, are mentioned as occurring among the diluvial accumulations, though in certain freshwater deposits, as at Gmünd, lophiodontes, oxen, hippopotami, &c. occur together.

Again, certain animals which lived in the diluvial period, as cervus megaceros, appear, by various evidence, not to have been extinct till later times; though we should not venture to adopt Dr. Hibbert's opinion, that they have really lived within the historic ages of Europe. However, it deserves remark, in connection with this subject, that no one has yet succeeded in showing a real and certain distinction between the common red deer and domestic ox of Europe, and the analogous bones of Kirkdale and other caverns.

Upon the whole it seems probable that the palæotherian and other tertiary races of quadrupeds died and became extinct gradually, but not by any one law of uniform progression; that the elephant, and his accompanying tribes, began to exist during tertiary eras, rose to predominance before the close of the diluvial period, and, for the most part, perished in that period, or soon after. Some modern species (stag, ox) were co-existent with the elephant and hippopotamus in northern zones; others (elephas primigenius, rhinoceros tichorhinus), which abounded in diluvial, were also living in tertiary periods; and, perhaps, a few (as the horse) may have been in existence during all these periods. This is a point, however, extremely hard to determine; since, if,