ENVIRONMENTAL CONDITIONS 395 



existence during the earlier time, but were more restricted in distribution 

 and apparently did not as yet dispute the habitat of the primitive sharks. 

 The latter also, as previously noted, were, in those early stages, more' 

 lestricted than the ostracoderms, appearing less abundantly than the 

 latter in the brackish embayments of the sea, though showing themselves 

 more abundant in the first clearly fresh-water fauna. 



Each higher class of chordates is thus found to be more restricted at 

 first than the lower groups. When higher and lower are both existent, 

 the lower are found as earlier waves and farther from the original center 

 of evolution. This is a well known principle when applied to mammalian 

 evolution in the Tertiary. For mammals the record is exceptionally clear. 

 For the earlier chordates it is scanty, but, so far as it is clear, points to a 

 center of evolution within the land waters. From that original home the 

 primitive chordates which still exist are farthest removed. 



The nature of the early vertebrate record is strikingly like the record 

 of the Eurypterids and receives here the same interpretation which Grabau 

 aiid Miss O'Connell have given to their history. 10 The bulk of sands and 

 muds deposited in the seas is of river origin, and the currents are known 

 to bring with them in abundance certain kinds of land life, especially 

 fragments of vegetation. The material given by a strong and sediment- 

 laden river to the weak waves of a shallow sea is deposited near shore and 

 is subjected to very little wear, but will probably bear evidence of marine 

 deposition. With lesser sediment and stronger wave action, resistent 

 fragments, like the teeth, spines, and plates of river fishes, will be more 

 worn, but as rare fragmentary fossils will continue to occur. Besides 

 these fragments brought by the rivers, there is a tendency for the living 

 fauna to spread as far as it may, and consequently forms may live in 

 brackish waters and leave fossil evidence of that fact at a time before they 

 had spread into a truly marine habitat. 



But if it be true that fishes originated in fresh waters, why is the early 

 evidence known chiefly from scanty remains deposited in the margins of 

 the sea? This requires a digression on those conditions of continental 

 deposition which render such deposits generally unfavorable for the preser- 

 vation of fossils. 



Bivers deposit their sediment over floodplains in time of flood. During 

 the low-water stages these plains are exposed to the air and their materials 



10 A. W. Grabau : Early Paleozoic delta deposits of North America. Bull. Geol. Soc. 

 Am., vol. 24, 1913, pp. 399-512 ; Relation of the Eurypterids to their environment, pp. 

 498-499 ; Principles of stratigraphy, 1913, pp. 1029, 1030. 



Miss M. O'Connell : Distribution and occurrence of the Eurypterids. Bull. Geol. 

 Soc. Am., vol. 24, 1913, pp. 499-515. Her complete paper, entitled "The habit of the 

 Eurypterida," will soon be published as one number of the Bulletin of the Buffalo Society 

 of Natural Sciences, 1916. 



