THE CLIMATIC THEORY OF TERRACES. 29 



beheading of streams, may be suggested, Ixit these are more or less accidental occurrences 

 and can not influence all the streams of a region in the same way. It must be borne in 

 mind that in 99 out of 100 of the terraces which we are trying to explain the first process 

 is the deposition of a considerable thickness of alluvium, generally of a gravelly nature. 

 We may therefore disregard all possibilities except the two just mentioned, which seem 

 to be the only ways in which earth movements can give rise to abundant deposition in 

 the immediate valleys of the streams. 



To take the first possibility, we find that an uplift of the mountains would satisfy the 

 requirements imposed by the terraces of Arizona and elsewhere in one important respect. 

 It would steepen the grade of the upper parts of the streams and thus increase the amount 

 of erosion. This in turn would overload the streams in their unsteepened portions, and 

 would cause deposition at the immediate base of the mountains, which is the place where 

 the heaviest deposits actually occur in the regions under discussion. The deposits, how- 

 ever, are not confined to the base of the mountains as they ought to be according to our 

 present supposition. In regions of mountainous uplift the main valleys within the dis- 

 turbed area are preeminently the scene of active erosion, as may be seen in the case of 

 the Colorado Canyon or the gorge of the Rhine. Such valleys ought to be free from 

 accumulations of gravel; yet we find that the upper Santa Cruz and Magdalena, for 

 example, which according to the theory of mountain uplift should have been steepened and 

 caused to cut gorges, are actually burdened with enormous deposits of gravel which con- 

 tinue far back into the interior of the mountains. The same conditions seem to prevail 

 in all regions where terraces of the kind here described are found. 



It is possible to obviate the difficulty just suggested by assuming that the mountainous 

 region has not been uplifted as a whole, but that individual ridges have been uplifted, while 

 the intervening main valleys have remained unaffected. This would certainly explain the 

 occurrence of the gravel in the main valleys. It would not, however, explain the occurrence 

 of similar gravel deposits in practically every one of the side valleys, such as the upper part 

 of the Canada del Oro north of the Santa CataUna Range and directly among the supposedly 

 uplifted mountains. It would also involve an assumption contrary to some of the chief con- 

 clusions of geology as to the nature of crustal movements. It would be necessary to assume 

 in the first place that important changes in the relative attitude of the neighboring parts of 

 the earth's surface have taken place in the very latest geological times without leaving any 

 visible sign of movement, such as fault scarps or gorges due to uplift. It would also involve 

 the assumption that movements of the interior of the earth's crust take place in such a way 

 that all the ridges and elevated parts of scores of mountain systems have been raised so as to 

 steepen the grade of the minor tributaries, while the main valleys remain unchanged. In 

 other words, the assumption is that the internal movements of the earth adjust them- 

 selves most delicately to the minor features of the surface, an assumption the exact reverse 

 of the truth. 



There is another objection to the theory that the terraces are due to the intermittent 

 uplift of the mountains and the consequent alternation of periods of rapid and slow erosion. 

 The uplift would, of course, occasion active deposition at the base of the mountains, while 

 active erosion was in progress higher up. The cessation of upUft and hence of active ero- 

 sion among the mountains would leave the streams with comparatively light loads and thus 

 cause erosion to begin in the piedmont region of previous deposition. This would doubtless 

 give rise to terraces resembling those whose origin we are discussing. Directly among 

 the mountains, however, it is evident that if erosion were first active and then slow, and if 

 this process were repeated several times, terraces of rock would necessarily be formed 

 at the same time that terraces of gravel were being foimed lower down. While erosion 

 was active the vallej'S would be deepened; when it became slow they would be broadened. 

 The continual repetition of these processes could scarcely fail to produce a series of rock 



