Says Nicholson, ‘The teeth are replaced not merely when accident has broken off the old ones, but they are all shed at more or less regular intervals, coinciding with the casting of the epidermis.’ Not on _each occasion_ of sloughing, as we may, I think, understand this, but, like the casting of cuticle, contingently, according to the condition of the individual. Not altogether, either, or at certain periods of life, as a child loses his first teeth and gets a second crop, or as an adult cuts his wisdom teeth, but ‘a crop of young teeth work their way into the intervals of the old teeth, and gradually expel these latter.’ All the spaces and depressions between the maxillary and palatine rows are occupied by the matrix of tooth germs. Not a cut can be made in this part of the palate without the knife turning up a number of young teeth in every stage of development.[92] Independently of this accidental number, the maxillary presents certain phases which characterize families. For instance, a true viperine snake has in the upper jaw fangs _only_: non-venomous snakes have a whole row of from fifteen to twenty-five maxillary teeth, and in intermediate species their normal numbers vary considerably. Some of the highly poisonous families, notably the cobras and the sea snakes, have a few simple teeth in addition to fangs. The length of the jaw, therefore, diminishes in proportion to the number of teeth it bears. Only the viperine snakes are limited to the poison fang in the upper jaw; but fangs, like the simple teeth, are shed, broken, or lost, and renewed continually. Behind the one in use—the functional fang—others in various stages of development are found—‘a perfect storehouse of new fangs,’ as Mr. F. Buckland in his facetious style called them; ‘lying one behind another like a row of pandean pipes.’ In the skeletons of viperine snakes these may readily be observed. In the living example they are enclosed in a capsule, hidden by the loose gum sheath, called a gingeval envelope. So when the functional fang meets with an accident, or falls out in the order of things, the supplementary fangs in turn supply its place, each becoming in time firmly fixed to the jaw-bone, and ready to perform the office of its predecessor. Poison fangs succeed each other from behind, _forwards_; the simple teeth from the inner side, _outwards_. Before proceeding further, it may be well to explain that what is meant by the _true_ snakes in the foregoing rules, are those which do not possess the lizard features; _Anguis fragilis_, and some of the burrowing snakes which approach the lizards, not having the palate teeth. But here again we are tripped up with exceptions, since we are told that in dentition the boas are allied to the lizards; yet they have palate teeth.