DEFLATION IN DRY VALLEY 59 



tion of the sandstone and the removal of debris by wind. To this must 

 be added occasional corrasion by sand-laden blasts. When an alcove has 

 been formed its growth may be greatly accelerated through the scaling o3 

 of rock flakes from roof and side walls. 



Aside from the larger alcoves just discussed, the faces of La Plata 

 . sandstone are in numerous localities characterized by small pits, which 

 are in some places most abundant in certain strata and are hence arranged 

 with some regularity; but in other localities they are abimdant and very 

 irregularly arranged in the massive cross-bedded rock. These cavities are 

 common features of crumbling rocks in desert lands and have been abun- 

 dantly illustrated by Walther and others. In figure 2, plate 3, and figures 

 1 and 2, plate 4, may be seen areas where such small pits are plentiful. 



As processes now in operation, the decay of rocks in an arid region and 

 the action of the wind in the removal of waste are more strikingly illus- 

 trated in Dry valley than in any other district I have examined, but the 

 extent of the work thus performed in the past is most plainly indicated by 

 other facts than those thus far recited. The floor of Dry valley, like that 

 shown in figure 1, plate 3, is strewn with subangular or partially rounded 

 fragments of impure chert, indurated sandstone, conglomerate, and bluish 

 gray sandy limestone. Many fragments are from 1 to 3 feet in diameter 

 and a few are much larger. They are not by any means sufficiently abun- 

 dant to form a connected layer, but several lie close together in many 

 places. It is r^are to find an interval of as much as 200 yards between 

 fragments. No part of the valley floor traversed by various members of 

 our party was found to be free from such large fragments, nor were they 

 seen in stream beds, as if accumulated by torrents. The hard cherty or 

 dense siliceous rock fragments were most abundant. ]\Iany blocks are 

 polished and grooved by sand-laden wind. Pew are rounded sufficiently 

 to be compared to water-transported boulders. 



As to the origin of these masses, it is evident that they are not derived 

 from the La Sal or Abajo moxintains by recent glacial transport. The 

 former small valley glaciers of both groups ended at or very near the 

 debouchure of the mountain gulches on the surrounding plains, as was 

 ascertained by morainal and other evidence at several localities. The 

 debris can not, indeed, be plausibly referred to a mountain source by any 

 means or at any time, for the reason that the hard intrusive porphyries 

 of the numerous laccoliths and sills of these mountains are not repre- 

 sented in this debris. Gravel plains coated by small, well rounded peb- 

 bles, among which are many of porphyrj', do extend a short distance into 

 the soutliwest head of Dry valley, and these are probably outwash plains 

 of the recent glaciers or interglacial terraces comparable to some of those 



