30 THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMERICA. 



terraces corresponding to the gravel terraces lower down, but exactly the reverse in phase, 

 for the horizontal portion of the upper terraces would be synchronous with the vertical 

 portion of the lower. Probably such terraces exist somewhere in the world. They are 

 apparently scarce, however, and in the regions under discussion none has as yet been 

 pointed out, although those of the other type are found in scores of valleys. The two 

 kinds are scarcely associated in any such way as they would be if the ordinary terraces 

 were due to uplift of portions of the earth's crust. It seems reasonably certain that, so 

 far as most of the terraces of arid regions are concerned, we must give up the theory that 

 they are due to intermittent uphft of the mountains, for this not only would involve an 

 incredible degree of agreement between hues of drainage and lines of earth movement, but 

 also would demand that terraces of rock should be a characteristic feature of scores of valleys 

 where none are to be found. 



If uplift of the mountains is incapable of explaining the terraces, can they be explained 

 as the result of crustal movements which would diminish the grade of the lower portions 

 of the streams? The first step, according to this form of the hypothesis, would be a tilting 

 of the surface of the earth in such fashion as to lessen the grade of the rivers and thereby 

 cause them to deposit part of the load of detritus which they were bringing from the 

 mountains. The process of deposition would continue until the streams had filled up the 

 low parts of their valleys to the point where conditions of perfect grade were obtained, 

 that is, conditions of perfect adjustment of load to velocity and volume. If the process 

 of deposition were sufficiently rapid it would keep pace with the tilting, and would come 

 to an end as soon as the tilting ceased. In that case the cessation of tilting would cause 

 an immediate change in the mode of activity of the streams. They would cease to deposit 

 their loads in large quantities and would tend once more to cut down their channels. 

 Thus terraces would be formed which would possess the character of those which we are 

 attempting to explain. Further tilting would cause renewed deposition, and cessation of 

 tilting would permit the cutting of a second terrace along each stream. If the deposits 

 due to the second tilting did not reach the level of the first deposits, the first terrace would 

 persist with no changes except those arising from ordinary weathering and erosion. Further 

 repetition of the tilting process would of course cause further terraces, provided always 

 that the later deposits did not cover the earlier, and that the times of erosion were not so 

 prolonged as to cause the complete removal of the older deposits. 



At the very outset this form of the hypothesis of crustal movements is confronted with 

 the same difficulty as the other form. The terraces are in one sense universal in arid 

 regions, but in another sense they are very local. As has already been said, they are limited 

 to the mountainous regions, and more specifically to those mountainous regions where the 

 topography is still rugged and the elevation considerable. The terraces varj' in size in 

 proportion to the height and steepness of the slopes immediately adjacent. They grow 

 higher where the valleys become narrow, as a rule, and lower where the valleys broaden. 

 Sometimes, however, if the narrow parts of the valley happen to be so located that they 

 have a steep grade and are not plentifully supplied with rock waste from tributaries, the 

 case is reversed. Often the terraces die out entirely in the gorges because the grade is 

 such that no deposition took place even in times of the heaviest load, or else because the 

 small size of the valley has allowed all the deposits to be washed out since the last main 

 epoch of deposition. From what has just been said it is clear that the terraces may be a 

 widespread phenomenon in one sense, but in another they are very local, for their size and 

 character depend largely upon the condition of the mountains in the immediate vicinity. 

 This means that the cause, whatever it may be, acts upon each individual river and mountain 

 independently. In other words, general, regional warping of the earth's crust can not be 

 appealed to in explanation of the phenomena. Such warping would cause the streams to 

 be terraced continuously throughout the whole district which was warped, but would not 



