Species of reptiles are, for the most part, vague quantities in paleontology; they can be determined with assurance only by the comparison of abundant material. Adult characters in mammals are apparent in the ossification of the skeleton, and size can be used within moderate limits in the determination of species; but size in reptiles means but little; no one could possibly say that the skeleton of an alligator six feet in length is not that of an adult animal if he knew nothing else about the Crocodilia. So also the compression and malformations of bones from the processes of fossilization obliterate specific characters in great part. Nor are specific characters easily distinguishable in the skeletons of living reptiles. The genus, therefore, among fossil reptiles is practically the unit, and we may be sure that for every well-defined genus we discover there existed numerous minor variations, which, had we the living animals to study, we should call species. We classify the living Crocodilia into two families, about four well-defined genera—perhaps even five or six—and about twenty-five species. Of the living lizards there are about eighteen hundred species, twenty families, and four larger groups or suborders. In all probability the lizards have never been more abundant and more varied than they are at the present time. Possibly these proportions of species, genera, families, and suborders may represent approximately the proportions that have existed at some time or other in most of the other groups which we call orders—approximately only, for we can never be quite sure that we evaluate the structural characters of different groups of organisms quite equally. The absence of a molar tooth in a mammal would ordinarily indicate a genus, the absence of a tooth in a reptile might not indicate even a variety or a race. Whence it follows that classification of organisms is not and never will be an exact science. The value of characters used in classification is very unequal, as we have seen. No two persons see these characters from the same viewpoints, and in consequence no two persons whose opinions are worth while ever wholly agree as to classification.