i88 Natural History Bulletin. 



ered material along the bluff failed to reveal a single specimen. 

 The place it occupies farther north seems to be taken by 

 Phillipsastrea gigas Owen, which is here somewhat abund- 

 ant. P. gigas was once very common a few miles northeast 

 of Troy mills, at what are known among the old settlers as 

 the Walton and McPike springs; and good specimens may 

 still be collected in the same localities. P. gigas and A. -pro- 

 funda were found weathered out in ravines near the river 

 above Quasqueton. The first is known to occur there in 

 place, the second may possibly occur in the same way. At 

 Solon, in Johnson count}^, Iowa, P. gigas is found in its nor- 

 mal position a few feet above beds that are the equivalent of 

 S. pennata beds at Independence, but no specimens of A. -pro- 

 funda are known in any part of the region. A. profunda 

 seems to have been confined to a limited area in Buchanan 

 and Blackhavv^k counties, its center of distribution being near 

 Littleton. 



The rectified section of Devonian strata in Buchanan county 

 embraces the following members, each of which may be still 

 farther sub-divided : 



1. Independence shales, with a unique fauna having a very 



close relationship with the Rockford shales along Lime 

 Creek above Rockford, Iov»^a. 



2. Brccciaied beds, embracing the beds characterized by 



Gyroceras and numerous usually detached valves of 

 Gypidnia occidentalis. 



3. Barren beds. 



4. Spirifera pennata beds, characterized by the abundance of 



easily recognized types of Spirifera, Atrypa, and 

 Strophodonta. 



5. Acervidaria profunda beds, containing Cystiphyllum and 



A. profunda in great profusion. 



6. AcervuJaria davidsoni beds beginning with the beds con- 



containing S. parryana, and including all up to the 

 sumimit of the reef of A. davidsoni and associated corals. 



