66 GEOLOGY OF THE NARRAGAJs'SETT BASIN. 



far more tlian that amount. When we remember that the whole drainage 

 hasin of the Mississippi— a region which is probably many times as great 

 as the field whence this detritus came — ^yields to erosion not more than 

 about a twentieth of a cubic mile each year, it becomes evident that we 

 have to bring into our conception of the causes operating in the olden day 

 some more effective agent of erosion than is found in free water. 



If all the detritus of the Carboniferous conglomerates were of the same 

 nature as that which is found in the Southern Millstone grit and the related 

 beds, we could perhaps assume that their production was due to the invasion 

 of the sea acting upon a deep decayed zone, but the fact that the thickest of 

 these deposits occur in the northern part of the Appalachian field and are 

 composed of undecayed pebbles negatives this hypothesis and requires 

 us to assume that the erosion attacked the bed rock with great intensity. 

 That this attack was by torrent action is extremely improbable; for, as 

 before stated, no torrents are now known to produce so large amounts of 

 pebbles of crystalline rocks as were formed at this time; and when such 

 fragments are formed, so far as my observations go, they always present 

 marks of decay, due to the slow manner in which they are shaped and to 

 the conditions of their storage in detrital cones. The pebbles of the 

 Narragansett and other conglomerates of the same age which I have 

 examined, even those of a compound nature, are in practically all cases as 

 fresh as those contained in the bowlder deposits which were formed during 

 the last Glacial period. This appears to negative the supposition that they 

 could have been the result of ordinary torrent action and to require a 

 method of formation which apparently can be explained on the hypothesis 

 of glacial erosion. 



It should be noted that the pebbles of the Carboniferous conglomerates, 

 especially in the Narragansett district, show no trace of glacial scratches; 

 moreover, they generally have a rather rounded form and are of less varied 

 size than those in any of the till deposits formed during the last Glacial 

 period. In some cases, however, they seem to me to retain the faceted 

 ^hape which is so characteristic of ice-made pebbles. When compared with 

 the pebbles of the last Glacial period, which, in a measure, have been sub- 

 jected to marine or stream action, they are found to correspond with them 

 in all essential features, except when, as is often the case, the old fragments 

 have been deformed by stresses which came upon them since they were 



