BRITISH ASSOCIATION. 



259 



downwards to the Silurian, nor upwards to the carbonife- 

 rous series. The genera which had no representatives in 

 other strata, were those which contained the greatest number 

 of species. Pterichthys, Coccosteus, Cephalaspis, Ostrole- 

 pis, Dipterus, Glyptolepis, Platygnathus, Dendrodus, Di- 

 placanthus, Cheiracanthus, and Cheirolepis. Those genera 

 which occurred also in the Silurian system and coal 

 measures, such as Onchus, Ctenacanthus, Ctenoptychius, 

 Ptychacanthus, Acanthodes, Diplopterus, and Holopty- 

 chius, did not contain a single species identical in the sepa- 

 rate formations. This result agreed precisely with that 

 which M. Agassiz stated he had obtained in the upper for- 

 mations, in which the fish, echinodermata, and molluscs of 

 separate formations never extended from one system, or 

 even subdivisions of the strata, to another. In this view, 

 which differed from that entertained by most palaeontolo- 

 gists, ne was supported by M. d^Orbigny, and he attributed 

 the similarity of the results, obtained by them from the 

 examination of remains of fish and molluscous animals, to 

 the employment of the same principles in both cases. 



M. Agassiz remarked, that the fish found in those forma- 

 tions, and, indeed, in all the older rocks, as compared to 

 those of modern times, were small and even insignificant in 

 size ; and he wished particularly to insist on this circumstance, 

 because the idea of colossal dimensions, which we were ac- 

 customed to attach to the fossils of all geological epochs, 

 was untrue both as regarded the fish and all other classes of 

 animals, with the exception of a few particular types. The 

 principal exception, was the number of gigantic saurians in 

 the secondary rocks, but their existence was the less re- 

 markable, as at that period mammifers scarcely existed, and 

 the cretaceans and gigantic pachyderms were absent. In 

 speaking of the fish of the old red sandstone, small or 

 middling sized, he meant that generally they did not exceed 



VOL. I. NO. VIII. T 



