STRUCTURE OF FISHES. 



293 



in all fishes; and they consist of a scries of osseous pieces, 

 in which we may often recognise with tolerable precision 

 the analogous bones composing the anterior extremities of 

 a quadruped; such as the scapula, clavicle, humerus, ulna, 

 and radius.*' These two latter bones are very distinctly 

 marked in the Lophhts jmcatorius, or Ajigler, as may be 

 seen in Fig. 191, where b is the scapula; c, the clavicle; u, 

 the ulna; and r, the radius. The carpus may also be recog- 



nised in a chain of small bones, w^, interposed between the 

 radius and the Phalanges, z. In the Ray these phalanges 

 are very numerous, and each is divided into several pieces 

 by regular articulations: these are shown in Fig. 192: they 

 are arranged close to one another in one plane, and form an 

 effectual base of support to the integument which covers 

 them. The scapula, according to Cuvier, is sometimes de- 

 tached from the rest of the skeleton, and at other times con- 

 nected with the spine: in most cases, however, it is sus- 

 pended from the cranium; a fact which may be cited in 



• Those anatomists who are fond of pursuing" the theory of analogies, 

 maintain that all these bones are merely developments of certain ribs, pro- 

 ceeding- from the spine in its anterior parts. A similar orig-in has been as- 

 signed to the pieces of bone to which the ventral fins are attached: but it is 

 difficult to reconcile this tlieory with the fact that these bones do not pro- 

 ceed from the spine, and are quite detached from the rest of the skeleton. 

 It is evident, therefore, that if they are to be considered as analog-ous to the 

 bones of the hinder extremities in the mammalia, they arc in a condition of 

 very imperfect development. 



