34 THE CLIMATIC FACTOR AS ILLUSTRATED IN ARID AMERICA. 



the Tucson Mountains, near the raikoad station of Rillito, about 17 miles northwest of 

 Tucson. According to the botanists the palo verde trees, Parkinsonia iorreyana, seen partly 

 embedded in the bank of the channel must have begun their growth at a time when the place 

 where the bottoms of the trunks now stand was the surface of the alluvial plain. While the 

 trees were getting a start and making their first growth, the plain must have remained at 

 nearly the same level. Later, however, it was built up about 5 feet by river deposits, as 

 can be plainly seen. The condition of other trees on all sides shows that the entu-e flood- 

 plain was here being built up. The length of time during which this process has been going 

 on may be judged from the age of the trees. I cut down the largest of those seen in Plate 1, a 

 (page 21), and counted its rings. They show that the tree began its growth between the 

 years 1670 and 1680 a.d. For about 200 years deposition appears to have predominated 

 over erosion, so that the plain was gradually built up by the addition of 5 feet of silt. 

 Then in a decade or two the slowly accumulated deposits were swiftly washed away, not 

 only here, but for 20 or 30 miles upstream. To turn back to figure 6, it appears as if 

 erosion suddenly began and a channel was cut at several places in the vicinity of Tucson, 

 corresponding to C" in the diagram. Then it rapidlj^ extended upstream and still more 

 downstream, until a continuous channel 30 miles long and from 2 to 15 feet deep was 

 formed. If we assume that this new channel extended from F to F', it is evident that 

 the material which would formerly have been deposited in that region began to be carried 

 below F', thus completely changing the area of maximum deposition. If the channel 

 should lengthen still more the area of deposition would be moved still farther downstream, 

 and might even disappear if the stream finally attained such size as to flow through 

 the region of bahadas into the main Gila River and the sea. This has not happened, 

 however, for great floods such as those of the late eighties or of 1905 are not common. 

 During the relatively dry years from 1906 onward there has been a shght tendency for the 

 new channel to silt up. In other words, the area of chief deposition is shifting upstream 

 once more. 



Here, on a small scale, we have an example of the entire process of terrace-making. 

 First slow deposition lasting 200 years, next a rapid cutting of a channel with a marked 

 shifting downstream of the area of deposition, and finally a slight and possibly temporary 

 resumption of the process of deposition in the old area. Man, to be sure, has played 

 some part in the matter, but he has simply served as the means, so to speak, of pulling the 

 trigger which allowed certain natural forces to come into play. If any other cause, such as 

 protracted hea^^^ rains, had gathered the water into a single channel or had increased its 

 amount so that it flowed farther than hitherto and ran down the steeper slope of the con- 

 vexity below Tucson, the same thing would have taken place. Moreover, another vital 

 point must be remembered. Man alone did not cause the terracing. The cutting out of 

 the inner channel required a number of exceptionally severe floods, and the later slight 

 refilhng of the channel demanded a period of diminished rainfall. Possibly the floods 

 would have caused the cutting of the channel even without man's intervention, and certainly 

 the refiUing has nothing to do ■with man. It is significant that this same tj'pe of channehng 

 and refilling, tlfis process of terrace-making on a small scale, has occurred at practically 

 the same time, not only on scores of streams in the arid Southwest, but on an equally large 

 number in various parts of Asia, where man's relation to nature has not been subject to 

 any change like that due to the settlement of our own regions. Therefore it would seem 

 that the process of terrace-making is now going on irrespective of man; it may be accel- 

 erated or checked by his actions, but it seems to occur primarily in response to climatic 

 variations. 



Manifestly, in the case of the little terraces now under consideration, earth movements 

 have had nothing to do with the matter. We have seen that in the past, also, earth move- 

 ments of sufficient magnitude to produce the large terraces of earher times apparently 



