20 W. M. DAVIS 



the normal cycle of erosion is founded. This principle was regarded 

 as a dangerous extravagance less than forty years ago ; it is usually 

 taken as a matter of course today. But as regards cycles of erosion 

 in which various other processes than "normal" erosion by weather 

 and streams are involved, there is still so much to be learned that 

 they have not yet become commonplace matters. They are still 

 in need of fuller discussion; indeed various refinements are undoubt- 

 edly still to be made regarding certain aspects of the ordinary 

 cycle of normal erosion; and the gradual advances by which its 

 present development has been reached ought to prove that even 

 this simplest of all the special forms of the cycle scheme has not 

 yet reached the perfection of an infallible finahty. Campbell's 

 discussion of the effects of a moderate tilting on th6 shifting of 

 divides and the rearrangement of stream courses^ is a good example 

 of the kind of investigation that different elements and possibihties 

 in the normal cycle need. Careful study should still be given to 

 the manner in which differences of climate affect the work of weather 

 and water and the extent to which surface forms are thereby 

 influenced: the good beginning in such a study made by Passarge 

 for cold, temperate-humid, temperate-subhumid, temperate-arid, 

 and torrid zones in the third volume of his Landschaftskunde 

 (Hamburg, 1920) would have a larger value if it had not been hmited 

 by his rejection of soil-creep on forested slopes as an effective 

 agency of degradation, as I propose to show in a review to be 

 pubHshed elsewhere. A somewhat similar study by Sapper, 

 Geologischer Bau und Landschaftshild (Brunswick, 191 7), is less 

 satisfactory because no effort was there made to select really 

 comparable land forms — that is, land forms of similar structure 

 and in a similar stage of erosion — from the different zones. 



Among the land forms that vary with differences of climate, 

 to which my own attention was directed during a Pacific voyage 

 in 1914, are ridge crests in relatively homogeneous rocks. In 

 temperate climates of moderate rainfall, ridges are usually forest- 

 covered and their crests acquire a well-rounded or convex cross- 

 profile when the valleys between them are well opened; and even 



' "Drainage Modifications and Their Interpretation," Jour. Geol., Vol. IV (1896), 

 pp. 567-81, 651-78. 



