Tol. 65.] GLACIAL EROSION IN NORTH WALES. 295 



graceful slopes of the soil-covered hills which rise between the 

 streams. These delicately carved forms assure us that the graded 

 courses aud accordant junctions, which obtain throughout the 

 branching systems of nimble water-streams, have been established 

 no less perfectly in the innumerable down-slope lines of what may 

 ■be called ' slow-creeping soil-streams.' 



XII. Texture op Dissection. 



In a given stage of a cycle of erosion, land forms may vary as to 

 the texture of their dissection, or number of stream-lines crossed in 

 a given distance. A stream-line is here understood to be a sloping 

 line to which the converging slopes on each side contribute drain- 

 age. Such lines are very numerous if the dissected mass is of an 

 impervious structure, yielding a fine-grained waste on a barren 

 surface ; for then every little rill will carve its minute valley, as 

 in the Bad Lands of Nebraska, where 1000 or 2000 stream-lines 

 may be counted in a mile. Stream-lines are, on the other hand, 

 relatively rare, if the dissected mass is covered with a coarse and 

 pervious sheet of creeping waste overgrown with vegetation ; for 

 here most of the rain percolates beneath the surface, instead of 

 running off in surface-rills. The subdued mountains of North 

 •Carolina offer good illustration of a thoroughly dissected region of 

 relatively coarse texture ; here many a mile of surface has not 

 more than ten or twenty stream-lines. Pre-Glacial Wales should 

 be pictured as of the latter coarse-textured kind of dissection ; and 

 this picture is warranted, because the Welsh moels to-day still have 

 large sweeping contours, seldom indented by stream-lines. 



XIII. Interdependence of Parts. 



Dr. G. K. Gilbert called attention thirty years ago, in his classic 

 report on the Henry Mountains, to the well organized interdepen- 

 dence that gradually comes to be established among all the elements 

 of a normal drainage-system ; an interdependence so delicately 

 adjusted that a change in one part calls for some slight change in 

 every other part. The later study of river-systems and of mountain- 

 sculpture has only served to confirm Gilbert's fine generalization. 

 When one is persuaded of the truth and pertinence of this physio- 

 graphic principle, its application in such a problem as the one 

 here in hand is highly serviceable. It is surely a practical aid, in 

 studying the actual forms of a glaciated district like that of 

 Snowdon, to construct as carefully and definitely as possible a 

 mental picture of the forms that should have occurred there before 

 the district was glaciated ; and in this mental construction, it is 

 reasonable to appeal to the forms of Dartmoor, the Ce'vennes, and 

 the North Carolina mountains, as homologues of the normal forms 

 of North Wales in its pre-Glacial stage. It is, moreover, a great 

 encouragement to feel that the normal development of land forms 



x2 



