1912.] NATURAL SCIENCES OF PHILADELPHIA. 379 



of swamp development, attributable to a temporary modification of 

 climatic conditions, which would of course affect the whole region 

 at the same time, the fossils from the various places could hardly 

 be expected to be otherwise than equivalent. It is these that are 

 used in correlation of the "Newark" with the European Keuper. 

 The only fossil locality at any considerably lower horizon is that at 

 Holicong, Bucks County (B. 2). The forms occurring there are 

 unfortunately not specifically identifiable, and so can throw little 

 light on the age of the beds. But since both Cycadites and Sehizo- 

 neura are genera which are found in the Bunter or lower Triassic of 

 Europe, it is by no means impossible that the Norristown-Stockton 

 formation is really the approximate equivalent of that horizon, 

 which it certainly closely resembles lithologically. 



Because of the total absence of fossils of diagnostic value in the 

 upper 10,000 feet of the Brunswick formation in Pennsylvania, its 

 exact position is also indeterminate. It further does not seem 

 advisable to attempt extrapolation into other districts, where the 

 succession of formations is in general quite different from that here 

 outlined. But the absence of beds of uppermost Triassic or even 

 of Jurassic age can in no way be regarded as certain. 



It is to be concluded, then, that there is no evidence whatever 

 of the deposition of any part of the New Red or Newark group 

 during the Permian period; but since all of the fossils of diagnostic 

 value, indicating middle-upper Triassic age, have come from a 

 rather limited horizon, about the middle of the group, we are not 

 justified in concluding either that the whole group is of the same age 

 or that the Bunter sandstone below and the upper Keuper or Rhsetic 

 above are not represented in the American rocks. 



