168 ORDERS OF FISHES. 



extremity of the paddle ; each joint bears a pair of three- or 

 two- or one-jointed branches (fig. 489). This is the case in 

 the liind as well as fore paddles, and we are justified in sup- 

 posing that those extinct Ganoids of which impressions of 

 paddles with scaly centres have been preserved, were pro- 

 vided with a similar internal skeleton." 



Upon the whole, Dr Glinther concludes: 1. That the 

 Barramunda is not generically separable from the almost 

 exclusively Triassic genus Ccratodus, which was founded 

 simply upon detached teeth ; 2. That the Barramunda is 

 very closely allied to certain of the Crossopterygious Ganoids, 

 such as the Dijpterus of the Old Eed Sandstone, the chief 

 difference being, that the tail of the latter is heterocercal ; 

 3. That the order Dipnoi should be considered merely as 

 forming a sub-order of the Ganoidci ; 4. That the Ganoidei 

 may be united wath the Masmoh^anchii into a single group, 

 which may be termed Palceichthyes, and which is character- 

 ised by having a "heart with a contractile bulbus arteriosus, 

 intestine with a spiral valve, and optic nerves non-decus- 

 sating;" 5. That the Ganoidei are the Fresh-water Paloe,- 

 ichtliyes, and the Elasmohrancliii are the Mai^ine Paloiichthyes. 



It would be out of place to enter here into any discussion 

 of the systematic changes above mentioned, as proposed by 

 this distinguished authority. Whether these views be ulti- 

 mately adopted or not, it seems most convenient from a 

 palseontological point of view, to retain the Dipnoi as a 

 distinct order of the Fishes in the meanwhile. If this course 

 be followed, we find that the order can readily, in the light 

 of recent researches, be split into two distinct sections — the 

 Sirenoidei, comprising Lepidosiren and Ceratodus, and the 

 Ctenodipterini, comprising Dipiterus, Ctenodus, and some other 

 less important fossil forms. 



In dealing with the first of these sections, we may leave 

 Lepidosiren out of account, as it is not known to occur in a 

 fossil state at all. The genus Ceratodus, however, has con- 

 siderable geological importance, both intrinsically and from 

 the light which it throws upon the real structure of the 

 Ctenodipterines. The genus w^as originally founded by 

 Agassiz to include certain singular dental plates (fig. 535) 



