284 THE CARBONIFEROUS SYSTEM. 



(2.) The changes and subdivisions of this fauna are related not 

 merely to lapse of time, but to vicissitudes of physical conditions. At 

 Windsor, for example, the fauna of the Aviculopeeten bed is manifestly 

 that of a shallower and more sandy sea than that of the Productus bed ; 

 and further, the change from the fauna of the Lower series to that of 

 the Upper series coincides with the deposition of the great gypsums 

 and gypseous marls. It is the same in the Shubenacadie section. 



(3.) It follows that, if the peculiar Permian conditions indicated by 

 the rocks came in earlier in Nova Scotia than in Europe, the character 

 of the fauna might also be changed earlier. In other words, we have 

 both rocks and shells with Permian aspect in the Lower Carboniferous 

 period. 



(4.) In accordance with this, it is the Upper series of limestones, and 

 those most nearly related to the gypsums and marls, that have the 

 most Permian aspect. The lower Windsor limestones and those of 

 Economy and Pictoii have much more the ordinary Lower Carbon- 

 iferous character and fossils. 



(5.) In the little bed of marine limestone at M'Kenzie's Mill, 

 Wallace, we have an example of the existence of some members of 

 this fauna in the period of the Upper Coal formation, where we have 

 also a greater number of the fossil plants that extend upward from the 

 Coal formation into the Permian ; and there is nothing to preclude 

 the supposition, already stated in the preceding chapter, that some of 

 the upper limestones of Colchester and Hants may have been deposited 

 contemporaneously with the Middle Coal formation. At the same 

 time, it must be admitted that this last supposition is not proved, and 

 that the appearances in those places where the Coal measures occur 

 are not in its favour. 



(6.) It is evident that the marine fauna of the Lower Carboniferous 

 in Nova Scotia more nearly resembles that of Europe than that of the 

 Western States. This is no doubt connected with the fact that the 

 Atlantic was probably an unobstructed sea basin as now, while the 

 Appalachians already, in part, separated the deep sea faunae of the 

 Carboniferous seas east and west of them. In the Permo- carboniferous 

 period the connexion may have been more complete, or perhaps the 

 shallow-water species may have at all times been able to migrate. 

 Perhaps, however, there was no migration in the case, but only the 

 recurrence of similar and representative species under similar conditions 

 of existence. 



(7.) It must not be overlooked that, as a set-oflF to the Permian 

 appearance of the fossils of the Lower Carboniferous in Nova Scotia, 

 we have the occurrence of such old forms as Phillipsia, Centronella, 



