432 Lisboa — Permian Geology of Northern Brazil. 



ciated with dicotyledonous woods. Silicified conifers are abun- 

 dant in the south of Brazil, not only associated with Sterioster- 

 num tumidum Cope, of the Permian, but very probably with 

 Psaronius also. 



Characteristic Geological Features. 



In the absence of paleontologic evidence, some special fea- 

 tures of the rocks and of the geology in the north permit a 

 provisional determination of the age of certain sediments. 

 Thus the nature and mode of occurrence of the eruptive rocks, 

 as well as the presence of siliceous beds or concentrations in 

 limestones and other rocks, or in the pebbles of conglomerates 

 derived from these materials, are very characteristic of certain 

 Triassic or Permian sediments. 



Eruptive rocks. — Two different eruptive rocks cut the pre- 

 Cretaceous sediments in the region explored. Both are diabasic 

 and differ from each other in their textures, which seem to 

 have a definite relation to the more or less recent age of the 

 beds which they cross. Observation here showed that the pres- 

 ence of these two types seems to have a definite relation to the 

 age of the sediments just as in the south of Brazil. 



The compact diabase. — This black, eruptive rock is a normal 

 diabase which crops out in the bottom of the valleys where ero- 

 sion has exposed beds older than the Psaronius-bearing sand- 

 stones. It occurs in two manners : 



1. In place, forming dikes or sheets which crop out above 

 the marly shales. These eruptives form the base of the Per- 

 mian sandstones with Psaronius in the channel of the Parna- 

 hyba river. In Surubi, in Coroa do Pinga, and in other points 

 along the Parnahyba, it forms the rapids of the river; in Ama- 

 rante it seems to form a sheet in the lower part of the Permian 

 sandstones. At Picos Dr. Roderic Crandall has observed this 

 same rock cutting through the argillaceous shales and outcrop- 

 ping in the lower zone of the eroded red sandstone. Dr. G. A. 

 Waring and Dr. Manoel Arrojado Lisboa have also noticed 

 this type under identical stratigraphic conditions on the upper 

 Poty, directly behind the scarps of the Serra Geral. 



2. The same rock is also exposed in great rounded blocks 

 scattered over the surface of the Permian sandstone areas. 

 These accumulations of blocks bring to mind glacial bowlders, 

 but very likely they merely result from the rapid decompo- 

 sition of the diabase. 



(b) Trap. — The other eruptive is also a pyroxene-bearing 

 rock with plagioclase, but is commonly amygdaloidal and rich 

 in zeolytes. It represents a later eruptive phase of the same 

 diabase (?). This rock occurs in Grajahii in a large flow. It 



