MIGRATION OF DUNE-SANDS. 211 



of their natural history which, so far as I know, have not been made the 

 subject of inquiry. Some of these I now propose to consider. 



RESISTAXCE TO PERCOLATION OF WATER. 



Those who may have watched the movement of dunes may have had 

 an op2)ortunity of noting that they are often found with their surfaces 

 in the state of dry sand in the course of a few hours after a heavy rain. 

 01)serving this fact and knowing how readily water is drawn by capil- 

 larity through the materials of which they are composed, I watched the 

 effect of heavy showers on these wind-built hills. To my surprise I 

 found that after a summer shower giving a rainfall of an inch the dune- 

 materials would often not be wet for more than three-fourths of an inch 

 beneath the surface ; below that line the sand remains quite dry, wet 

 sand not being found until the section is carried some feet down into 

 the mass. 



The explanation of this phenomenon appears to be as follows : When 

 rain falls on dry sand the water finds difficulty in overcoming the 

 repulsion which the dry material offers, and so works but slowly down- 

 ward ; at the same time the interstices of the outer layer permits it to 

 flow down the slopes, keej^ing near the surface until it reaches the bottom 

 of some depression such as abound on the surfaces of dunes. Here the 

 hydrostatic pressure becomes sufficient to drive the fluid dowuAvard. 

 Only in the winter, when the water in the upper part of the dune-sands 

 has been from time to time frozen, do we ordinarily find the mass wet to 

 the depth of a foot or more. 



MIORA TION. 



Little affected by Rain. — In consequence of this peculiarity of dune- 

 sands which retards their deep wett?ng in ordinary seasons they are 

 retained in marching order. In a few hours after a rain the thin, water- 

 soaked layer, not having water supplied to it from below, may become 

 perfectly dry, so that a strong wind may excavate and bear away large 

 quantities of the material. I am inclined to believe that the ready 

 movements of these sands are in the main to be attributed to these con- 

 ditions. 



Retarded by Decay of the Material. — An examination of the dunes of the 

 Atlantic coast has shown that the detritus of which they are composed 

 is generally in process of division and decay. This is indicated by the 

 fact that the materials, if taken from a site where they have evidently re- 

 mained for years in repose, show the existence of much fine dust, while the 

 recent accumulations and the beach-sands from which they are derived 



