EFFECT ON DRAINAGE OF A FEOZEX MOSS-LAYER. L33 



to corrade. Their ability to dissolve the rocks with which they come in 

 contact is also greatly reduced by their low temperature. Moreover, the 

 banks of the streams, and even the bottoms of the smaller rivulets, in many 

 instances, are moss-covered, and the soil beneath the moss is frozen. The 

 erosive power of surface water is thus reduced to a minimum. Only the 

 larger creeks and the rivers obey the laws of erosion and of corrasion which 

 are in force in warmer and less humid regions. 



Another result of a low mean annual temperature in a humid region is 

 that dead vegetation decays slowly, and prostrate trees and obstructions to 

 drainage formed by drift-wood remain a long time, thus retarding the streams 

 and favoring sedimentation. Many of the smaller drainage valleys of Alaska 

 are impassible on account of the trees that have fallen from either bank and 

 interlaced their branches in the center. Dams are thus formed which favor 

 the increase of swamps. The growth of moss is thus promoted and the 

 difficulties of drainage still farther augmented. 



The mossy covering of Alaska decreases in thickness towards the east, and 

 at the head-waters of the Yukon in the North West Territory it is not 

 especially remarkable. In southern Alaska, at least from Juneau south- 

 ward, the mosses are wonderfully luxuriant, and although not generally 

 frozen, as in the region of the Lower Yukon, they thoroughly protect the 

 subjacent strata. 



Decay of Rocks. 



Geographical Distribution of Rock Decay. — The prevalence of residual 

 deposits resulting from the atmospheric decay of rocks in warm and humid 

 regions and their decrease in thickness and extent in the colder and more 

 arid portions of the earth's surface has been discussed by me in a previous 

 paper.* At the time the paper referred to was written but little informa- 

 tion was available concerning the condition of rock surfaces in high latitudes. 

 What is here presented in this connection may be considered as a supple- 

 ment to the paper just mentioned and as sustaining in a marked manner the 

 conclusion that rock decay is a function of existing climatic conditions, and 

 in general decreases from tropical to arctic regions. 



The conditious for noting the effects of a rigorous climate on rock surfaces 

 are especially favorable in Alaska, for the reason, as will be explained on 

 pages 137-41, that a very large part of our northern territory was not occu- 

 pied by glaciers during the Pleistocene period. Hence a comparison of the 

 amount of alteration of the rock surfaces there found with the decayed sur- 

 faces of similar outcrops in the driftless area of the upper Mississippi valley 

 and in the nonglaciated portion of the Appalachian mountains would reveal 



* U. S. Geological Survey, Bulletin No. 52, 1889. 



