CHAPTER XXIV 



THE EVOLUTION OF FISHES 



HE Geological Distribution of Fishes. The oldest un- 

 questioned remains of fishes have been very recently 

 made known by Mr. Charles D. Walcott, from rocks 

 of the Trenton period in the Ordovician or Lower Silurian. 

 These are from Canon City in Colorado. Among these is cer- 

 tainly a small Ostracophore (Asteraspis desideratus). With it 

 are fragments (Dictyorhabdus] thought to be the backbone of a 



FIG. 246. Fragment of Sandstone from Ordovician deposits, Canon City, Colo., 

 showing fragments of scales, etc., the earliest known traces of vertebrates. 

 (From nature.) 



Chimaera, but more likely, in Dean's view, the axis of a cephalo- 

 pod, besides bony, wrinkled scales, referred with doubt to a sup- 

 posed Crossopterygian genus called Eriptychius. This renders 

 certain the existence of Ostracophores at this early period, but 

 their association with Chim&ras and Crossopterygians is question- 

 able. Primitive sharks may have existed in Ordovician times, 

 but thus far no trace of them has been found. 



The fish-remains next in age in America are from the Bloom- 

 field sandstone in Pennsylvania of the Onondaga period in the 



435 



