RELATION OF FAULTS TO CYCLE OF EROSION 191 



structures. This is a real difficulty, which at present seriously embar- 

 rasses many physiographic studies ; but it is not an insuperable difficulty. 

 One of the readiest means of lessening it is not to neglect it but to attack 

 it — that is, consciously to direct our attention toward the existing surface 

 product of past processes on underground structures — and thus avoid 

 entanglement in the maze of the past processes themselves. Our effort 

 should constantly be turned to telling what a district looks like, whether 

 we tell about it in the empirical terms of an earlier century or in the 

 explanatory terms now more in fashion. The reasons which have led 

 many to follow the explanatory method I have sufficiently set forth 

 in an address of a year ago (a, '12, 104-112). 



An ideal Series of Forms on faulted Structures 



Two methods of procedure toward our present object are in use. One 

 consists in gathering the accounts of various actual forms on faulted 

 structures, arranging them in some suitable sequence, and giving them, 

 one after the other, explanatory description in some fitting terminology ; 

 these actual examples then serve as types for the description of new ex- 

 amples. Another method begins like the first one with a collection of 

 actual examples, but it goes much farther; it establishes the general 

 principles that are exemplified in the collected cases, and then deduces 

 from these general principles a systematic series of purely ideal examples, 

 and describes them in a correspondingly systematic series of explanatory 

 terms. The second method is here adopted, but only its deductive part is 

 presented in this essay. In passing, it may be pointed out that in the 

 possibility of greatly increasing the number of deduced ideal examples 

 over the number of observed examples, and in the full understanding that 

 is gained regarding the ideal forms, inside and out, in their past, present, 

 and future conditions, lies the great advantage of a deductively developed 

 explanatory terminology. Evidently the description of a new example 

 will be much facilitated if the observer has previously made acquaintance 

 with a large number of thoroughly understood types; and no method yet 

 employed provides so large an outfit of well conceived types as the deduc- 

 tive method here adopted. But it should not be forgotten that the de- 

 ductive part of this method of establishing a terminology can not be 

 properly undertaken until after an earlier inductive and analytical stu 'v 

 of many actual examples has made it possible safely to establish the 

 general principles from which the deduced ideal examples are drawn. 



Let us therefore begin the present study at the beginning, or initial 

 stage, of an imagined new cycle of erosion, which is introduced by the 



