336 THE REGOLITH 



and magnetites show quite well-preserved crystal outlines. It 

 is noticeable that these sands contain no mica, although the 

 mineral occurs in the sea-sand, from whence the dunes are 

 derived. Eutgers accounts for this on the supposition that 

 during the transportation of the material the mica folia be- 

 come so finely shredded as to be sifted out from the heavier 

 particles of sand, and quite dissipated. It is well to note that 

 the abrasive power of wind-blown particles is greater than 

 that of those carried by water, since, as noted by Daubree, a 

 thin intervening film of water may serve to buoy up the gran- 

 ules, and keep them apart. To this fact is ascribed the angular 

 nature of many of the wind-blown grains. This same authority 

 seems to think that with wind-blown sand, as with water-worn 

 material, there is a minimum limit, beyond which reduction 

 in size of particles rarely goes. This minimum he places at 

 about .25 millimetre in diameter. It seems, however, more 

 probable that attrition may go on to an almost indefinite limit, 

 but that the finer and lighter materials are driven farther 

 away — perhaps not collecting in the form of dunes at all — 

 leaving, as one would naturally expect, the sands of any one series 

 of dunes of nearly uniform size.^ 



It was noted by Blake during the surveys of the railway 

 routes to the Pacific that the wind-blown sands of the Colorado 

 desert were sometimes in the form of almost perfect spheres, all 

 their sharp edges and asperities having been woi*n away by 

 mutual attrition. The grains were composed mainly of quartz, 

 agate, garnet, and dark granules derived from the debris of vol- 

 canic rocks. In places there is a black iron sand, and usually 

 a considerable proportion of lime carbonate, as indicated by its 

 brisk effervescence when treated with acid. The sand dunes of 

 the Bermudas, as elsewhere noted, are composed wholly of cal- 

 careous material from finely comminuted shells and corals, while 

 those of the Sevier desert region of Utah, as described by Gilbert,* 

 are of fine gypseous sand formed by the evaporation of the water 

 in the neighboring playa lakes. 



Volcanic Dust. — The finely comminuted materials ejected 



1 XJdden has shown that the atmospheric currents being for the most part 

 loaded only lo an insignificant fraction of their capacity, their sediments 

 will be more evenly assorted than those of water currents. (Journal of 

 Geology, YoL II, 1894.) 



« Monograph I, IT. S. Geol. Survey, 1890. 



