=Peculiarities of Jaws and Teeth.=--The jaws of fishes are subject to a great variety of modifications. In some the bones are joined by distensible ligaments and the fish can swallow other fishes larger than itself. In other cases the jaws are excessively small and toothless, at the end of a long tube, so ineffective in appearance that it is a marvel that the fish can swallow anything at all. In the thread-eels (_Nemichthys_) the jaws are so recurved that they cannot possibly meet, and in their great length seem worse than useless. In some species the knife-like canines of the lower jaw pierce through the substance of the upper. In four different and wholly unrelated groups of fishes the teeth are grown fast together, forming a horny beak like that of the parrot. These are the Chimæras, the globefishes (_Tetraodon_), and their relatives, the parrot-fishes (_Scarus_, etc.), and the stone-wall perch (_Oplegnathus_). The structure of the beak varies considerably in these four cases, in accord with the difference in the origin of its structures. In the globefishes the jaw-bones are fused together, and in the Chimæras they are solidly joined to the cranium itself. =The Angler-fishes.=--In the large group of angler-fishes the first spine of the dorsal fin is modified into a sort of bait to attract smaller fishes into the capacious mouth below. This structure is typical in the fishing-frog (_Lophius_), where the fleshy tip of this spine hangs over the great mouth, the huge fish lying on the bottom apparently inanimate as a stone. In other related fishes this spine has different forms, being often reduced to a vestige, of little value as a lure, but retained in accordance with the law of heredity. In a deep-sea angler the bait is enlarged, provided with fleshy streamers and a luminous body which serves to attract small fishes in the depths. The forms and uses of this spine in this group constitute a very suggestive chapter in the study of specialization and ultimate degradation, when the special function is not needed or becomes ineffective. Similar phases of excessive development and final degradation may be found in almost every group in which abnormal stress has been laid on a particular organ. Thus the ventral fins, made into a large sucking-disk in _Liparis_, are lost altogether in _Paraliparis_. The very large poisoned spines of _Pterois_ become very short in _Aploactis_, the high dorsal spines of _Citula_ are lost in _Alectis_, and sometimes a very large organ dwindles to a very small one within the limits of the same genus. An example of this is seen in the poisoned pectoral spines of _Schilbeodes_. =Relation of Number of Vertebræ to Temperature and the Struggle for Existence.=--One of the most remarkable modifications of the skeleton of fishes is the progressive increase of the number of vertebræ as the forms become less specialized, and that this particular form of specialization is greatest at the equator.[21]