DISTRIBUTION OF ARKOSES. 53 



suppose that conditions favoring the rapid erosion of the decayed material 

 were established. These conditions inay have been brought about in any 

 one of several ways: The region may have been subjected to glacial action; 

 the rainfall of the area may have been increased in such a measure that the 

 streams were made competent to waste the surface; or the area may have 

 been exposed to wave action, either by being lowered beneath the level of 

 the sea or by becoming the seat of a lake. 



The hypothesis of glacial action at the time these arkoses were formed 

 does not, at first sight, seem to be supported by the evidence which is derived 

 from the presence of well-preserved vegetable remains in the beds; but, as 

 remarked in the preceding paragraph, these remains seem not to have been 

 deposited in a fresh state in the growing accumulation, but to have been 

 washed from some antecedently formed but practically contemporaneous 

 deposit into the positions which they now occupy. Therefore the existence 

 of ice wearing in this district at the arkose-building time does not seem 

 improbable. We shall hereafter note that there is other evidence of a more 

 positive nature going to show the existence of glacial conditions in this 

 gravel period in the district about the Narragansett Basin. 



As a whole, the distribution of the arkose deposits of the Carboniferous 

 time around the margins of the Narragansett Basin seems most easily to 

 be explained by the supposition that streams of a swiftly flowing nature 

 formed torrent cones when they debouched into a fresh-water lake which 

 occupied almost the whole area now covered by the coal-bearing- rocks of 

 the district. That the deposits are in general those of torrent cones or 

 deltas appears to be shown by their irregular distribution. In all cases 

 where they have been observed they occur in rather detached patches, like 

 accumulations which have no great extension in the direction of the ancient 

 shore line. This seems to exclude the supposition that the deposits were 

 formed in the manner of ordinary shore accumulations, where the debris 

 is transported from a neighboring coast escarpment. 



There are no observations on record concerning arkoses now in process 

 of formation, though such may be accumulating in many parts of the world. 

 It is therefore not possible to ascertain with certainty the distance to 

 which the angular crystalline debris of granitic rocks can be conveved by 

 stream action without losing the peculiar features which separate it from 

 the ordinarv products of the erosive forces which have acted on much 



