6 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 52 



material reported to be chalk is only a white clay, but he confirms 

 the existence of a thin-bedded limestone beneath the fossiliferous 

 sandstone, and "beneath this limestone is a bed of very bituminous 

 laminated shale. It is a kind of lignite changed to coal, only a few 

 inches in thickness." He says that fossil wood was found by Dr. 

 Gongalves Dias near Sao Pedro, two leagues from the Villa de Mila- 

 gres. The existence of the limestone and lignite induce one to hope 

 that a careful search may yet lead to the discovery of considerable 

 additional paleontologic material, though Gardner distinctly states 

 that no fossils could be found in the limestones.^ 



In the Rocha Collection the rock inclosing the fossil fishes contains 

 manv fish scales and the remains of microscopic shells that have not 

 been studied. One of the concretions is rather darker and more 

 marly than the others, and in this are found many small rounded 

 bodies evidently of organic origin. Some of these were submitted 

 to Dr. E. O. Ulrich, paleontologist of the U. S. Geological Survey, 

 who kindly reports as follows upon them : 



The ostracod seems to be one of the simple types of Cyfheridea, apparently 

 closely allied to the Miocene C. suhovata U. & B. It resembles an Eocene 

 species also very closely, and I know of a late Cretaceous form that is not far 

 removed. However, with specimens in rock like yours it is difficult to satis- 

 factorily determine even the genus of the host of smooth and subovate 

 ostracods. 



Referring to the broader features of the Cretaceous geology of 

 northeastern Brazil, the area covered by the Cretaceous rocks is not 

 known wath any certainty. Even where they are best known they 

 have been identified at only a few places on and near the coast. On 

 the coast, however, they form only a narrow belt approximately 

 parallel with the present shoreline, toward which they have a general 

 and gentle dip, except on the immediate shores, where the dip is 

 often landward. This coast belt of Cretaceous sediments is in places 

 from twenty-five to fifty miles or more in width, while at other places 

 the beds have been entirely removed and the old underlying rocks 

 of the interior are exposed on the seashore. On the land side of 

 the Crfetaceous belt the surface rocks are usually granites, gneisses, 

 schists, and other mctamorphics of uncertain age or ages. 



In the region drained by the Rio Paranahyba above Theresina, 

 and lying mostly in the State of Piauhy, there is a series of hori- 

 zontal sedimentary beds which appear to be the inland remnants of 

 the series exposed along the coast. But little is known of the geology 



'Trans. Brit. Assoc, 1840, itq. 



