GREGORY: FISH SKULLS 203 



As to the branchiostegal rays, Hubbs (1919, p. 65) notes that in the Apodes (eels), 

 Heteromi and Lyopomi, the rays are all slender, usually numerous and long and frequently 

 curved upward posteriorly about the free margin of the opercular bones. 



The increased size of the respiratory muscles have apparently conditioned the powerful 

 development of the hyomandibular and of the whole roof of the cranium; the latter has to 

 resist the wrenching action of these muscles and afford a firm lodging for the brain, as well 

 as a strong base for the heavily muscled jaws. 



The eel represents a predaceous adaptation in an early stage, in which the suspensorium 

 is directed forward as in primitive teleosts. The moderate length of the jaw is obtained by 

 a moderate lengthening of the postorbital region. The moray represents an advanced 

 stage in which the very strong hyomandibular has become directed backward, thus lengthen- 

 ing the jaws posteriorly. 



The main upper jaw-bone of the morays appears to be the maxillary and not the 

 pterygoid, as Boulenger (1904, p. 599) held, for the reasons that, as seen in a fresh moray 

 head, this bone lies definitely outside of the powerful adductor mandibuls muscles and that 

 its posterior tip passes to the outer side of the quadrate. Obviously the pterygoid bone 

 could not pass through the jaw muscles, which always lie lateral to it. We may also suppose 

 that the great increase in cross-section of the adductor mandibulae muscle could not have 

 taken place if the pterygo-quadrate bar had remained in the position which it has in the eel, 

 and that it was no longer needed for the bracing of the suspensorium on account of the 

 great increase in size of the hyomandibular. The backward growth of the latter would 

 also tend to pull the posterior end of the palatopterygoid out into the thin thread of bone 

 which Tate Regan (1912^;, p. 378) records as present in the Muraenidae. 



Leighton Kesteven (1926a, p. 133) adopts the identification by Owen and Richardson 

 of the main dentigerous upper jaw bones of the muraenids with the palatines, chiefly on the 

 ground that these bones in the morays are not "lip-bones," like those of ordinary fishes, 

 but belong to the inner row. But if so, how is it that the posterior ends of these bones in 

 the morays pass outside of the jaw muscles.' 



As to the derivation of the eels, Smith Woodward (1901, p. x) noted that relatively 

 primitive representatives of the order were already in existence in the Cretaceous period, 

 and that they cannot be regarded as degenerate members of any group of Cretaceous 

 "Teleostei" hitherto discovered but have been derived directly from some of the Mesozoic 

 fishes which would be termed "Ganoidei" by some authors. On the other hand, the "lepto- 

 cephalus" larvae of the eels is very similar to that of the isospondyl Albula, while the skull 

 of eels seems to be merely a highly specialized derivative of some large-mouthed Cretaceous 

 isospondyl type. Thrissopater (A. S. Woodward, 1909) might be such a form, except that 

 the supraoccipital is in contact with the frontals, while in the eels it is separated from 

 them by the parietals. The recent Engraulis among the clupeoids shows that the hyo- 

 mandibular may easily become secondarily directed backward. 



The otoliths (sagittae) of Apodes, as studied by Frost (1926i, p. 99), fall into three types : 

 the "Anguillid," the "Congrid" and the "Heterenchelid," of which the first resembles 

 those of the Clupeoidea with certain added features. The lapillus of Anguilla resembles 

 the conchoidal lapilli of the Ariidae; the asteriscus is slight and upright as in the Elopidae 

 but is slightly different in form (1926, p. 1(X)). On the other hand, it seems not impossible 

 that the Apodes may also stand as a specialized offshoot from near the base of the Iniomi. 



