380 



TRANSACTIONS OF THE AMERICAN PHILOSOPHICAL SOCIETY 



logical resemblances between brotulids and macrurlds. The skull of a small ophidlld is 

 strongly suggestive of the gadoid type. It is rather hard to believe that all this is mere 

 parallelism. Nevertheless, according to Tate Regan (1903, p. 460) and Boulenger (1910, 

 pp. 646, 702, 703), the codfishes and their allies have not been derived from degraded 

 blennies allied to the Ophidiidae, and the striking resemblances between certain members 

 of the two groups is probably to be ascribed to convergence, brought about in terminal lines 

 of the descendants of some very early group. 



pmx.. 



B G adus 



Fig. 259. A. Lota. B. Gadus. Top views. 



Regan considers (1910a, p. 11) that "the absence of spinous fin-rays, the large number 

 of rays in the pelvic fins and the indirect attachment of the pelvic bones to the clavicles are 

 evidences that the Anacanthlni are much jnore generalized than the ophidloids, near which 

 they have been placed by some authors. They are perhaps derived from generalized 

 scopeloids, such as the Aulopidse." But the skull pattern of Aulopus, so far as I can see, 

 does not bear any special resemblance to the anacanthine types. On the other hand, many 

 features of the latter suggest relationship to various percomorphs. The opercular region of 

 Gadus and Lota (Fig. 258) differs most widely from the Aulopus type. The antero- 

 posteriorly wide hyomandibular recalls those of many higher percomorphs such as the 

 scombroids, gobies, jugulares, batrachoids; the texture of the skull suggests that of Lophius, 

 the enlarged opisthotic suggests gobioids, scorpaenoids, ophidioids. There are also note- 

 worthy resemblances in the skull to those of sciaenids. 



