Elephants and mammoths seem to have developed very rapidly. None are known from Eocene strata; but towards the end of the Miocene they had spread over Asia, Europe, and North America, and that in great numbers. The Eocene Amblypoda are still so different that we hesitate to connect them ancestrally with the elephants.
The Pinnipedia (seals and walruses) are strongly modified fissiped Carnivora, and have existed since at least the Upper Miocene; the transformation must have been accomplished within the Miocene period.
We cannot shut our eyes to the fact that various groups have from the time of their first appearance burst out into an exuberant growth of modifications in form, size, and numbers, into all possible--and one might almost say impossible--shapes; and they have done this within comparatively short periods, after which they have died out not less rapidly. It seems almost as if these go-ahead creatures had, by accepting every possible modification and carrying the same to the extreme, too quickly exhausted their plasticity--which, after all, must have limits--thereby becoming unable to meet successfully the requirements of further changes in their surroundings. The slowly developing groups, keeping within main lines of Evolution, and not being tempted into aberrant side-issues, had, after all, a much better chance of onward evolution.
A good example of the former are the Dinosaurs. We do not know their ancestors; but we have here to deal only with their range of transformation. The oldest known forms occur in the Upper Trias; they attain their most stupendous development in the Upper Jurassic and in the Wealden; and they have died out with the Cretaceous epoch. But already some of their earliest forms had assumed bipedal gait, and the Oolitic Compsognathus had developed almost bird-like hind-limbs.
On the other hand, there are many instances of extremely slow development--facts which raise the difficult question of 'persistent types.' Are these due to a state of perfection which cannot be improved upon? Or are they due to a kind of morphological consolidation (not necessarily specialization) which can no longer yield easily, so that therefore through changes in their surroundings they may come to an end sooner than more plastic groups?