GENERAL OBSERVATIONS. 79 



ocean; but it is not improbable that many of the brackish water accumula- 

 tions, especially those of late Cretaceous and early Tertiary age, may have 

 been made in land-locked waters which had their saline character continued 

 from former times by the leachings of surface drainage, similar to that of 

 Great Salt Lake, but in a far less proportionate degree. This idea is sug- 

 gested by the fact that the final change to exclusively fresh water lacustrine 

 deposits was so gradual that all the former brackish water species, among 

 which no open-sea forms have been discovered, passed away without any 

 perceptible physical change in the accumulating strata. 



LOWER SILURIAN AGE. 



The Lower Silurian age is represented in the collections only by a very 

 few imperfect fossils from Kwagunt Valley, Grand Canon of the Colorado, 

 Arizona. The Brachiopod genera Lingulella and Obolella are recognized 

 with a good degree of certainty, and, distributed through the small masses 

 of rock which contain them, there are apparently fragments of two or three 

 other species. 



These specimens doubtless belong to the Primordial period, and possess 

 much interest as regards the geological age of the rocks which underlie and 

 overlie them. 



UPPER SILURIAN AND DEVONIAN AGES. 



The present collections contain no fossils from strata of either Upper 

 Silurian or Devonian age, but it is not improbable that rocks of these ages 

 will yet be discovered in the Plateau Province. x 



CARBONIFEROUS AGE. 



The collections contain fifty species from strata of Carboniferous age, 

 much the greater part of which are from the Lower Aubrey Group. A few 

 are from the Upper Aubrey, and a still less number from the Red Wall 

 Group, while none are reported from the Lodore Group. 



A large proportion of all these fossils are specifically identical with well- 

 known forms in the strata of the Carboniferous or Coalmeasure period in 

 the States of the Upper Mississippi Valley; and all but two of them belong 

 to such types as we might naturally expect to find in the equivalents of 



