This classification, and the three which follow it, I quote (abridging some of them) from Prof. Agassiz's "Essay on Classification."
For explanations, see "Illogical Geology," Essays, Vol. I. How much we may be misled by assuming that because the remains of creatures of high types have not been found in early strata, such creatures did not exist when those strata were formed, has recently (1897) been shown by the discovery of a fossil Sea-cow in the lower Miocene of Hesse-Darmstadt. The skeleton of this creature proves that it differed from such Sirenian mammals as the existing Manatee only in very small particulars: further dwindling of disused parts being an evident cause. The same is true as regards, now, we consider that since the beginning of Miocene days this aberrant type of mammal has not much increased its divergence from the ordinary mammalian type; if we then consider how long it must have taken for this large aquatic mammal (some eight or ten feet long) to be derived by modification from a land-mammal; and if then we contemplate the probable length of the period required for the evolution of that land-mammal out of a pre-mammalian type; we seem carried back in thought to a time preceding any of our geologic records. We are shown that the process of organic evolution has most likely been far slower than is commonly supposed.
Since this passage was written, in 1863, there has come to light much more striking evidence of change from a more generalized to a less generalized type during geologic time. In a lecture delivered by him in 1876, Prof. Huxley gave an account of the successive modifications of skeletal structure in animals allied to the horse. Beginning with the Orohippus of the Eocene formation, which had four complete toes on the front limb and three toes on the hind limb, he pointed out the successive steps by which in the Mesohippus, Miohippus, Protohippus, and Pliohippus, there was a gradual approach to the existing horse.
Several of the arguments used in this chapter and in that which follows it, formed parts of an essay on "The Development Hypothesis," originally published in 1852.
Studies from the Morphological Laboratory in the University of Cambridge, vol. vi, p. 84.