FROM KULJA TO KUMUL 461 



travellers, I noticed that the dunes showed the prevalence 

 of a west, and not an east, wind. The moving dunes were 

 all steep-sided towards the east, and long-backed towards 

 the west, and during the three days we were on them 

 the wind came from the south-west or west. The early 

 mornings were often characterized by strong south winds, 

 which veered round to the west during the day. 



Although the nature of the stationary dunes was the 

 same throughout, that of the moving sands varied con- 

 siderably. Some dunes presented the phenomena of 

 enclosed, crescent-shaped pits, the sandy hollows of 

 which lay 50 ft. below the average level of the sands. 

 There was rarely any trough between the lines of the 

 dunes, for the arms they had thrown out had joined up 

 with other dunes, and I was thus enabled to travel "up- 

 stream," as it were, or transversely to the lines of the 

 sand-hills, without any effort. It seemed to me that the 

 formation was often due to varying winds. I noticed 

 repeatedly that the hollows between the dunes had been 

 banked up by miniature dunes formed from the south ; 

 and on one occasion I found a high dune formed by 

 a south wind, situated on the top of an ordinary dune 

 formed by a west wind. 



A considerable movement of sand was in progress — 

 for while the stationary dunes were snow-covered, these 

 moving dunes were practically bare ; a careful examina- 

 tion showing that the snow had been blown as the sand 

 was blown, and that in certain places where it had 

 collected were successive layers of snow and sand. On 

 the crest of some of the highest dunes were several 

 strata of snow and sand, showing how each fall of snow 

 had been covered up by a succeeding movement of sand. 

 The stationary dunes were all of an insignificant size. 



