imii"riMi SA.\D. 9 



same coarseness, the layers are more irrci^ular and liiin- 

 ner. Pebbly layers as much as an incli in thickness are 

 extremely uncommon. On the whole gravels of this kind 

 form vevy insigniticant deposits, where they have been 

 seen. This circumstance does not i-euder it unlikely that 

 lag gravels may have a greater development in regions 

 long exposed to the actions of the stronger winds. 



DRIFTING SAND. 



Lag gravels graduate imperceptibly into coarse drift- 

 ing sand, which in the field always lies in front of the 

 gravels, following the direction of the prevailing winds. 

 Farther in this direction the coarse sand becomes in turn 

 finer and finer, until the main deposit is reached, where it 

 always consists of grains of a more uniform size. In fact 

 the main bulk of all sand drifts, large enough to be called 

 dunes, have been found to contain only subordinate pro- 

 portions of sand grains measuring more than one fourth 

 or less than one eigth of a millimeter in diameter. 



Sand coarser than this is present as a maximum in- 

 gredient onl}^ in superficial layers of no very great thick- 

 ness, which lie on the rear slopes of dunes. It forms an 

 intermediate series between typical dune sand and lag 

 gravels, and it is capable of being rolled rather than 

 lifted by the winds. This is indicated bv the circumstance 

 that it is often the main ingredient on the crests of wind 

 ripples, being heavy enough to remain resting in this 

 exposed position, while the finer dune sand is lifted to the 

 upward slope of the next ripple. It differs from the lag 



