STKATIGRAPHIC CLASSIFKJATION PALEONTOLOGIC CRITERIA 49l 



tion between the Pacific and the Gulf of Mexico is suggested in Nevada 

 during the early and middle ages of this period. The north Pacific seems 

 also to have been connected with Bohemia (in the middle Devonian), as 

 is indicated by the presence of Hercynella in southeastern Alaska. Ac- 

 cording to present probably somewhat synthetic maps of the late De- 

 vonian, it appears that the Nevada basin connected with an Arctic in- 

 vasion of northwestern North America that extended completely across 

 the continent to the Atlantic at New York city. The communication 

 with the Grulf of Mexico, however, seems to have been, if not entirely 

 abrogated, at least much more limited at this time than it had been in 

 preceding Devonian epochs. 



During the Waverlyan and Tennessean the Pacific appears to have 

 been completely separated from the Atlantic except in the early part of 

 the Osage epoch and in the closing Moorefield shale stage of the Merame- 

 cian. In the early Pennsylvanian, however, old communications seem to 

 have been reestablished, but toward the close of this period — that is, in 

 the Permian — effective separation again prevailed. About the close of 

 the St. Louis age a Pacific fauna extended by some as yet unknown 

 pathway from Nevada to the northern part of the Oklahoma basin, and 

 thence farther eastward across the Tahlequah axis into northern Arkan- 

 sas, where it is found in the Spring Creek limestone. Its further prog- 

 ress seems to have been stopped by the Saint Francis axis. However, 

 before the beginning of the Sainte Genevieve this axis was submerged 

 permitting the entrance of the Atlantic fauna and apparently causing a 

 reversal of the trend of migration. This is indicated by the occurrence 

 of closely allied and identical cephalopods in Belgium, in shaly beds just 

 over the Saint Louis in Kentucky, in the Moorefield shale in northern 

 Arkansas, and in Nevada. 



DEDUCTIONS BASED ON FAUNAL DISTRIBUTION 



The fact that several closely allied or identical species are found in 

 southeastern North America only in Devonian rocks, while in California 

 and Nevada they occur in post-Devonian formations, has long been cited 

 as proving "that the reappearances of older faunas in younger rocks have 

 been due to migrations consequent upon the shifting of physical bar- 

 riers."®^ 



The logical propriety of this deduction is manifest and its truth is 

 established beyond question by many similar occurrences in American 



81 J. p. Smith : Journal Geology, vol. li, p. 198, and vol. ii, 1804, p. 598. 

 H. S. Williams : American Journal Science, II Ser., vol. xlix, pp. 94-101. 



