Dana estimated the length of the whole Tertiary period at one-fifteenth of the Mesozoic and Palæozoic combined. If we take the duration of the Tertiary period, as before, as 3,000,000 to 4,000,000 years, the total will amount to from 45,000,000 to 60,000,000 years.

Lastly, Walcott[28] has estimated the duration of the Palæozoic, Mesozoic, and Cænozoic or Tertiary epochs at about 17,000,000, 7,000,000 and 3,000,000 years respectively, giving 27,700,000 years from the beginning of the Cambrian; and Williams[29] has calculated the relative duration of the smaller epochs. See the table on p. 149.

The Last Link — Part 50

The results of all these calculations fall surprisingly well within the limits of Lord Kelvin's allowance. Of course they are based upon assumptions, but none of them is inherently unreasonable; and it was my purpose to draw attention to the surprising coincidence in the closeness of these results, perhaps too good to be true. Such calculations are considered close enough if they range within a few multiples of each other.

The Last Link — Part 50

[28] 'Geological Time as indicated by the Sedimentary Rocks of North America.' _Proc. Amer. Assoc. Adv. Sci._, xlii., 1893, pp. 129-169.

The Last Link — Part 50

[29] Henry Shaler Williams, 'Geological Biology.' New York, 1895.

Zoologists have fallen into the habit of requiring enormous lengths of time for the evolution of the animal kingdom. We know that Evolution is at best a slow process, and the conception of the changes necessary to evolve man from monkey-like creatures, these from the lowest imaginary mammals, these from some reptilian stock, thence descending to Dipnoan fish-like creatures, and so on back into Invertebrata, down to the simple Monera--this conception is indeed gigantic. Innumerable, almost endless, slow changes require seemingly unlimited time, and as time is endless, why not draw upon it _ad libitum_?

Huxley pointed out that it took nearly the whole of the Tertiary epoch to produce the horse out of the four-toed Eohippos, and that, if we apply this rate to the rest of its pedigree, enormous times would be required. This is, however, a very misleading statement, which necessitates considerable reduction, in conformity with our increased palæontological knowledge. Animals of the genus Equus--namely, Ungulata, with one toe, and with a certain tooth pattern--from the Upper Miocene of India are now known. Moreover, it is not simply a question of the gradual loss of the side-toes. The change from the fox-sized little Eohippos and Hyracotherium, so far as skull, teeth, vertebral column, and limbs are concerned (about the soft parts we know next to nothing), is a very great one indeed.