METHODS OF PALEOGEOGKAPHY 441 



blage is American in type, while the western has the widely distributed 

 Euro- Asiatic aspect. What is it that keeps these faunas distinct, when at 

 times the formations in which they occur approach one another in type 

 .to within a distance of 50 miles, or even less? At present there are no 

 visible lands separating them, nor even easily discernible geologic struc- 

 tures indicating former land barriers. Those who have studied their 

 relationships hold that a land barrier did exist — the Kankakee axis — 

 which kept apart these distinct provincial faunas. Many similar cases 

 are indicated on the maps here presented. It is known that the forma- 

 tions thin out in places and are petrologically different on the two sides of 

 such barriers. As long as these physical conditions are maintained the 

 faunas remain distinct, but when the deposits spread far and wide the 

 faunas, through blending, lose their individuality. 



For the benefit of those not believing in land barriers unless they can 

 see decided unconformities against which two given seas deposited their 

 similar or dissimilar sediments, a few photographs will be here introduced, 

 showing conformable strata with wholly unrelated superposed faunas. In 

 the Bennett quarries at Buffalo, ~Ne\v York (plate 46, figure 1), in a 

 little cliff less than 20 feet high, may be seen the Middle Devonic Onon- 

 daga coral limestone reposing upon a slightly irregular surface of the 

 Cobleskill-Siluric. Elsewhere, between these two deposits, the Manlius- 

 Siluric and all the Lower Devonic were laid down. Such sections are by 

 no means rare. In western Tennessee, at Kewsom, is a quarry face about 

 75 feet high (plate 47), at the top of which occurs the widespread Ohio 

 black shale, here representing the Lower Mississippic. With sharp petro- 

 logie change, but otherwise without apparent break, this stratum rests 

 upon a zone ?bout six feet thick, bearing the impress of Onondaga time, 

 for in this horizon has been found the well known pentremite Nucleo- 

 crinus verneuiU. Therefore here is absent all the Hamilton, which 

 at Louisville is a limestone with a thickness of 28 feet, and in central 

 New York is 500 feet in depth (shales). In the Tennessee region under 

 •discussion, below this thin limestone is another break, for the Onondaga 

 rests on the Louisville- Siluric. Between these two beds, therefore, all the 

 Salina, Manlius, and the entire Lower Devonic are missing. Fifty 

 miles to the west, however, most of the Lower Devonic has appeared 

 between these Siluric and Middle Devonic formations. The disconform- 

 ity last mentioned is one of vast extent, and may again be well seen at 

 Louisville, Kentucky (plate 46, figure 2), in the large quarries along 

 Bear Grass creek, where the same relation exists as at Newsom, Tennes- 

 see; in both instances the Onondaga Middle Devonic coral reef reposes 

 upon the Louisville-Siluric coral reef. This extensive land interval, 



