432 



GEOLOGY. 



to the south, on the whole, although in southwestern Missouri, Upper 

 Devonian beds appear to be present where the older parts of the 

 system are not known, 1 and the same relations appear to hold in 

 northern Arkansas. In some parts of Iowa, the Upper Devonian 

 is unconformable on the Middle (Fig. 197), but the separation of the 



Fig. 197. — Hypothetical section showing the probable relations of the Upper (State 

 Quarry) and Middle (Cedar Valley) Devonian, Penn Township, Johnson Co., la. 

 (Calvin, la., Geol. Surv.) 



Upper Devonian from the Middle is generally based on faunal rather 

 than on stratigraphic grounds, the division being made where the 

 distinctive Hamilton fauna gives place to another. The faunal changes 

 w]iich the upper Devonian records, seem to have resulted, in part 

 at least, from the immigration of life from the northwestern sea of 

 the previous epoch into the eastern interior. This immigration took 

 place toward the close of the Hamilton epoch, and the invading 

 species probably moved somewhat slowly from west to east, appear- 

 ing earlier in Wisconsin than in New York. A connection between 

 the Atlantic and the interior by the old route during the early part 

 of the late Devonian (Portage, see p. 433) has been suggested to account 

 for certain peculiarities of the Devonian faunas of this stage in the 

 east. 2 



The Tully, Genesee, and Portage formations, or Senecan series. — At 

 the base of the Upper Devonian in New York, there is, locally, a thin 

 bed of limestone (the Tully). At the horizon of this limestone, but 

 where it has otherwise run out, there is a discontinuous layer of iron 

 pyrites, which extends over a considerable part of western New York. 

 The waters of this region were polluted by some iron compound (prob- 

 ably carbonate). The iron is believed to have been precipitated by 



1 Shepard, Mo. Geol. Surv., Vol. 12, p. 65. 



2 Ulrich and Schuchert, op. cit., p. 656. 



