480 WILLIAM K. GREGORY 



On account (i) of their having more than five basal bones in the 

 pectoral fin, and (2) of the presence of true Eels in the Cretaceous 

 period, Dr. Smith Woodward 1 is of the opinion that the Apodes 

 are not degenerate offshoots from the Isospondyli as here defined, 

 but independent derivatives from some Holostean Ganoids. The 

 ancestral Apodes were possibly pelagic, subsequently invading 

 the rivers and giving off the fresh-water eels. These still require 

 the salt water for the development of the reproductive organs, 

 and undergo a remarkable metamorphosis, passing through a 

 ' Leptocephalus ' or Glass-eel stage quite similar to that of the 

 Albulids among Isospondyles. Extremely predatory and vor- 

 acious, sometimes becoming semiparasitic (Simenchelys ; com- 

 pare similar results among Cyclostomes) . Swimming rapidly 

 by lateral undulations, hence not requiring a homocercal 

 tail for propulsion. Teeth primarily adapted for seizing and 

 holding a struggling prey. The reduction of the maxillaries 

 their functional replacement by the more advantageously placed 

 pterygopalatines, and the separation of the latter from the 

 quadrate, finally resulting (in the Morays) in a large snake-like 

 and very loose jaw-apparatus with backwardly inclined sus- 

 pensorium; these changes in turn necessitating the great reduc- 

 tion of the branchial and opercular bones, the total separation 

 of the shoulder girdle from the skull, and the development of 

 large branchial pouches for sucking in water. 



The classification here adopted is as follows: 



Order Apodes (Linn.) Kaup. 

 Suborder 1. Archencheli Jordan 



Fam. AnguillavidaeHay. Up. Cretaceous, Mount Leba 

 non, Syria. 

 Suborder 2. Enchelycephali Cope 

 Fam. Anguillidae (Eels) 



" Nemichthyidse (Thread Eels) 

 " Synaphobranchidse 

 Suborder 3. Colocephali Cope 

 Fam. Muraenidse (Morays) 



1 Op. Cit., p. x. 



