have, it would seem a hopeless task at present to attempt 

 to trace successfully the genealogy of this class. One line, 

 however, appears to be direct, from our modern Gar-pike, 

 through the lower Eocene Le-pidosteus to the Lepidohis of the 

 Cretaceous, and perhaps on through the Triassic Ischypterus 

 and Carboniferous Palceoniscus ; but beyond this, in our rocks, 

 it is lost. The living Chimsera of our Pacific coast has nearly 

 allied forms in the Tertiary and Cretaceous, more distant rela- 

 tives in the Carboniferous, and a possible ancestor in the 

 Devonian Rhynchodus. Our Sharks likewise can be traced 

 with some certainty back to the Palaeozoic ; and even the 

 Lepidos'iren, of South America, although its immediate pre- 

 decessors are unknown, has some peculiar characters which 

 strongly point to a Devonian ancestry. These suggestive lines 

 indicate a rich field for investigation in the ancient life-history 

 of American fishes. 



The Amphibians, the next higher class of vertebrates, are so 

 closely related to the fishes in structure, that some })eculiar 

 forms of the latter have been c(msidered by anatomists as 

 belonging to this group. The earliest evidence of Amphibian 

 existence, on this continent, is in the Sub-Carboniferous, where 

 foot-prints have been found which were probably made by 

 Labyrinthodonts, the most ancient representatives of the class. 

 Well preserved remains are abundant in the Coal Measures, 

 and show that the Labyrinthodonts differed in important par- 

 ticulars from all modern Amphibians, the group which includes 

 our frogs and salamanders. Some of these ancient animals 

 resembled a salamander in shape, while others were serpent-like 

 in form. None of those yet discovered were frog-like, or without 

 a tail, although the restored Labyrinthodont of the text books 

 is thus represented. All were protected by large pectoral 

 bony plates, and an armor of small scutes on the ventral surface 

 of the body. The walls of their teeth were more or less folded^ 

 wdience the name Labyrinthodont. The American Amphib- 

 ians known from osseous remains are all of moderate size, 

 but the foot-prints attributed to this group indicate animals 

 2 



