70 A. Smith Woodward — The English Wcalden Fish-Fauna. 



that this shark had a persistent notochord and other characters 

 like the typical species of the genus in the Lower Lias of Lyme 

 Eegis. They show only one difference from the Liassic type, 

 namely in the cephalic spines; the Wealden specimens not being 

 barbed and never observed in more than one pair. In Cretaceous 

 strata, on the other hand, there is no certain evidence of Hybodtis. 

 Most of the teeth commonly ascribed to this fish probably belong 

 to another shark (Synechodus), which has well-developed vertebral 

 centra and smooth dorsal fin-spines ; while the ribbed dorsal fin- 

 spines occasionally found in the English Gault, Cambridge Green- 

 sand, and Upper Greensand, may be referable either to Ifybodus 

 or Acrodus. Typical teeth of the latter occur rarely in the Gault 

 of Folkestone, but are not known elsewhere. The tuberculated 

 fin-spines of Asteracanthus are scarce in the Wealden, and have 

 never been recorded from a higher horizon, unless a fragment 

 described by Pictet from the Lower Neocomian of Switzerland ' 

 happens to be of a little later date. 



Lepidotus Mantelli is most closely related to the species from the 

 Lithographic Stone of Bavaria. Only few and fragmentary remains 

 of the same genus are known from the Neocomian, and all frag- 

 ments of later date are very doubtfully placed here. Caturus is 

 also typically, if not exclusively, Jurassic; and the abraded skull 

 in the British Museum from the Wealden of Hastings (No. P. 6360) 

 cannot be generically distinguished from that of this fish. Ol'ujo- 

 pleurus is also known only from the Jurassic, occurring in the 

 Lithographic Stone (Lower Kimmeridgian) of Aiu, France, and 

 in the Purbeck Beds of Dorsetshire. 



Belonostomus ranges throughout the Upper Jm-assic and to the 

 Upper Cretaceous, with a very wide geographical distribution ; but 

 the only known specimens from the Wealden are too imperfect 

 to determine whether they are most closely related to the earlier 

 or to the later species. Neorhombolepis, however, although very 

 rare, is an essentially Cretaceous fish, the typical species occurring 

 in the Lower Chalk of Kent; and it is noteworthy that the most 

 closely related genus, Otomitla, is found in the Neocomian of 

 Mexico. 2 The Pycnodont Ccelodus is also typically Cretaceous, 

 though represented occasionally in the Purbeck Beds and other 

 formations on the horizon dividing the Jurassic and Cretaceous 

 series. 



The result is, therefore, that all the known English Wealden 

 fishes are survivors of typically Jurassic genera, except Neorhom- 

 bolepis and Ccelodus, and these are their little-modified repre- 

 sentatives. None but Belonostomus appear to range throughout the 

 Cretaceous. In fact, the Wealden estuary seems to have been the 

 last refuge of the Jurassic marine fish-fauna in this part of 

 the world, not invaded even by stragglers from the dominant race 



1 Asteracanthus granulosus (Agassiz), Pictet and Canipiche, " Foss. Terr. 

 Cretace St. Croix," p. 98, pi. xii, fig. ii (1859). 



3 J. Felix, Pakeontogr., vol. xxxvii. p. 189, pi. xxix, fisr. iii, pi. xxx, fi>s. 3-5 

 (1891). • l l ' 5 ' * 



