PYGOPTERUS GREENOCKII, ETC. 711 



tooth-bearing part, and of those two one also coincides Avith the edge of the entire 

 bone ; the third passes along midway between them, and carries the largest teeth. 



The lower jaw (fig. 8, L, J) is moderately stout, and between its two halves 

 are seen branchiostegal plates exactly resembling those of A mblypter us. There 

 is, namely, a median lozenge-shaped plate behind the symphysis of the jaw, and 

 the first lateral one on each side is nearly regularly rhomboidal, being very much 

 broader than those which succeed it posteriorly. The number of these plates I 

 have not ascertained ; in one example I counted at least 10, but there must be 

 many more. 



Of the pectoral arch, the only bones to be seen with certainty are the coracoid 

 and the plate succeeding it below and in front, which latter is proportionally larger 

 than in Amblypterus. The coracoid resembles that of Amblypterus in general 

 shape ; but its upper part is apparently rather slender, whereas the lower 

 reflected part is very broad. Traces of a scapula are seen in one specimen, and 

 also between and behind the scapula and coracoid a semilunar ossicle resembling 

 one of the plates found in a similar situation in Polypterus. 



On comparing these few facts regarding the head of Eurynotus with those 

 more completely elucidated in Pygopterus and Amblypterus, we see that the first- 

 named genus decidedly differs in the form of the opercular apparatus, and of the 

 superior maxillary bone, in the shape and arrangement of the teeth, and in the 

 smaller extent of the gape, so that the suspensory and opercular bones do not require 

 to be directed so very much backwards towards the articulation of the lower jaw. 



On the other hand, Eurynotus agrees with Amblypterus in the form and 

 arrangement of the branchiostegal rays, in the form of the scales, and in the 

 structure of the fins* and tail, the latter being typically heterocercal, defended 

 along its upper border by a row of V-shaped scales, while the sides of the vertebral 

 prolongation are covered by elongated ones of a lozenge shape. The fins were all 

 furnished with large fulcral scales along their anterior margins, which, in the 

 dorsal fin of a specimen in the museum at St Andrews, may be seen to be arranged 

 in a double series. This double arrangement of the fin-fulcra is also recorded of 

 Palceoniscus and Acrolepis by Muller (" Ganoiden," p. 152). 



Agreeing with Muller that the division between the " Lepidoids " and 

 " Sauj'oids^ the two families in which Amblypterus and Pygopterus have been 

 placed respectively by Agassiz, is artificial, we must class those two genera and 

 their immediate allies {Palceoniscus, Catopterus, Acrolepis, &c.) in one family of 

 Palceoniscidce, as has been done by Vogt (" Zoologische Briefe," s. 133). Then, 

 accepting meanwhile the great divisions of the Ganoid order proposed by Huxley 

 (" Memoirs of Geol. Survey, Decade X.," 1863), these fishes must come in under 

 the sub-order Lepidosteidce, characterised by the possession of non-lobate paired 

 fins, rhomboidal scales, and branchiostegal rays ; and under the family Lepidotini, 



* Except in the large size and peculiar form of the dorsal fin in Eurynotus. 



