Fishes of the JVestern North Atlantic 93 



peoidea, Stomiatoidea (including its subdivisions Gymnophotodermi, Heterophoto- 

 dermi, and Lepidophotodermi, named by Parr [jj: 15-17; 24'- 136]), Salmonoidea, 

 Argentinoidea,^! Esocoidea, and Bathylaconoidea. 



These Suborders correspond to: the Suborders Clupeoidei, Salmonoidei, Stomia- 

 toidei, Opisthoproctoidei, and Esocoidei of Berg (^); the Suborders Clupeoidei, Sal- 

 moniformes, Opisthoproctoidea, Stomiatoidea, Gymnophotodermoidea, and Esoci- 

 formes of Schultz and Stern (^7); the Suborders Clupeoldea, Stomiatoidea, and Salmo- 

 noides, plus the Order Haplomi, of Norman (j/); the Suborders Elopoidei, Clupeoi- 

 dei, Salmonoidei, Opisthoproctoidei, Alepocephaloidei, Stomiatoidei, Bathylaconoidei, 

 and Esocoidei of Bertin and Arambourg {18: 221 1); and the Suborders Elopina, 

 Albulina, Esocina, Clupeina, Chanina, Gonorhynchina, Salmonina, Opisthoproctina, 

 and Stomiatina of Matsubara (27: 178—227). 



One major area of uncertainty as to the relationships among the groups of iso- 

 spondylous fishes centers around the position of the Elopidae, Albulidae, and Ptero- 

 thrissidae. Indeed, within recent years it has been questioned whether Elops is even of 

 the same lineage as other living teleosts. Thus Saint Seine has derived it from the 

 Halecostomi (^6: 297), a group known otherwise from fossils only {18: 2 195-2201), 

 and Nybelin has emphasized the holostean nature of certain characteristics of Elops^''' 

 embodying a similar point of view. 



Both the Elopidae and the Albulidae, to be sure, have characteristics that are con- 

 sidered archaic by both palaeontologists and neozoologists alike; among these are: the 

 retention of the gular plate and of the numerous branchiostegal rays in Elops and Tar- 

 pon-^ the arrangement of the mucous canals on the top of the snout in Elops (j2: 454, 

 fig. i); the presence of close-set teeth on the parasphenoid bone in the roof of the 

 mouth (Elopidae, Albulidae, and Pterothrissidae); the presence of roofed post-temporal 

 fossae; and the persistence in Albula, Pterothrissus, and Megalops (but not in Elops) of 

 two rows of valves in the arterial cone of the heart.^* 



On the other hand, the elopids and the albulids differ widely and conspicuously 

 from living members of Lepisosteus and Amia in the narrow rod-like nature of their bran- 

 chiostegal rays and in the nature of the skeleton of their caudal fin; this latter difference 

 is a conspicuous one, externally, for while the rear boundary of the fleshy base of the 

 caudal is symmetrical relative to the main longitudinal axis of the trunk in the elopids 

 and albulids, it slopes upward-rearward in Lepisosteus (Fig. 3, p. 16) and Amia. Further 

 features worth mention are: (i) the invariable presence in the elopoids of a median 

 supraoccipital bone, which, to a greater or lesser degree, is involved in the roof-complex 

 of the skull ; (2) the absence of coronoid teeth in the mandible ; (3) a one-to-one rela- 

 tionship between the principal caudal rays and hypurals of Amia and Lepisosteus com- 

 pared with the considerable consolidation of the hypural elements in the elopids; (4) 

 the absence in elopids of the rigid attachment of the premaxillaries to the skull, which 



21. The reason for using this name rather than Opisthoproctoidea is given in Suborder Argentinoidea, Part 4. 



22. "Les Elopidae m'apparaissent etre des Holosteens aussi bien qu' Amia ou Lepidosteus" [32: 458). 



23. See Senior for account of the cone for Tarpon [48: 146-151), for Elops {48: 150), and for a summary (^9: 83-84). 



