This is what Cuvier has called subordination of characters, distinguishing between characters that control the organization and those that are not essentially connected with it. The skill of the naturalist consists in detecting the difference between the two, so that he may not take the more superficial features as the basis of his classification, instead of those important ones which, though often less easily recognized, are more deeply rooted in the organization. It is a difference of the same nature as that between affinity and analogy, to which I have alluded before, when speaking of the ingrafting of certain features of one type upon animals of another type, thus producing a superficial resemblance, not truly characteristic. In the Reptiles, for instance, there are two groups,--those devoid of scales, with naked skin, laying numerous eggs, but hatching their young in an imperfect state, and the Scaly Reptiles, which lay comparatively few eggs, but whose young, when hatched, are completely developed, and undergo no subsequent metamorphosis. Yet, notwithstanding this difference in essential features of structure, and in the mode of reproduction and development, there is such an external resemblance between certain animals belonging to the two groups that they were associated together even by so eminent a naturalist as Linnaeus. Compare, for instance, the Serpents among the Scaly Reptiles with the Caecilians among the Naked Reptiles. They have the same elongated form, and are both destitute of limbs; the head in both is on a level with the body, without any contraction behind it, such as marks the neck in the higher Reptiles, and moves only by the action of the back-bone; they are singularly alike in their external features, but the young of the Serpent are hatched in a mature condition, while the young of the type to which the Caecilians belong undergo a succession of metamorphoses before attaining to a resemblance to the parent. Or compare the Lizard and the Salamander, in which the likeness is perhaps even more striking; for any inexperienced observer would mistake one for the other. Both are superior to the Serpents and Caecilians, for in them the head moves freely on the neck and they creep on short imperfect legs. But the Lizard is clothed with scales, while the body of the Salamander is naked, and the young of the former is complete when hatched, while the Tadpole born from the Salamander has a life of its own to live, with certain changes to pass through before it assumes its mature condition; during the early part of its life it is even destitute of legs, and has gills like the Fishes. Above the Lizards and Salamanders, highest in the class of Reptiles, stand two other collateral types,--the Turtles at the head of the Scaly Reptiles, the Toads and Frogs at the Lead of the Naked Reptiles. The external likeness between these two groups is perhaps less striking than between those mentioned above, on account of the large shield of the