Vol. 50.] ME. CHAS. DAVISON- ON SNOWDRIFT DEPOSITS. 477 



The amount of snow that settles in valleys and ravines, especi- 

 ally in those whose direction is at right angles to that of the pre- 

 vailing wind, must be very considerable. I have not succeeded in 

 finding any reliable estimates. The common expression that after 

 a storm the ravines are filled with drift-snow cannot be taken to 

 mean much. At the foot of a mountain in Grinnell Land, Greely 

 met with a line of almost vertical snow-banks and drifts, the front 

 of which ranged from 100 to 150 feet in height. 1 This is probably 

 an unusual amount, and it may have been the accumulation of 

 many seasons. But if we remember that 10 or 15 feet is not an 

 excessive depth to collect in railway-cuttings during one of our 

 severer English storms, and then compare the latter with the far 

 more violent and more frequent Arctic gales, we shall be prepared 

 to accept a high estimate. On the other hand, the total snowfall 

 in Arctic regions is by no means great, but the snow is much 

 denser than that which falls in warmer climates. The average 

 weight of a cubic foot of snow at Port Bowen was found by Rowland 

 to be 30 pounds. 2 



The time of year at which snowdrifting is most frequent and 

 considerable is a subject on which information is still required; 5 

 In Russia, where snowdrifts seriously disturb the railway-system, 

 Mr. Sresnewskij has made some interesting enquiries. He finds 

 that the drifting is at a maximum in mid-winter, but there is more 

 in the second half of winter than in the first, there being then more 

 snow to drift. There are also most drifts in the months when the 

 snowfall is least and in which there are fewest days of snow. 1 



2. Origin and Transportal of the Dust. 



In Arctic countries, hills and exposed places are soon denuded of 

 their snow, a few hours being sufficient for a strong wind to remove 

 the fall of many days. Even when the snow has become so packed 

 as to provide blocks for snow-buildings, it' can in these situations 

 offer no permanent resistance to a violent wind. Plains are some- 

 times cleared so that sledging is impossible. 5 It is from such 

 places bared of snow that the material of snowdrift deposits is 

 derived. The hard frozen ground may be converted into dust in 

 two ways : (1) by evaporation of the interstitial ice, and (2), as 



1 Greely, vol. i. pp. 402-3 ; Nares, vol. i. pp. 323-24. As snow in lar»e 

 masses is granulated at a small depth, may not extensive accumulations of 

 drift-snow be sometimes confused with snow-covered glaciers ? On this point 

 see Hooker, vol. ii. pp. 90, 116; Nares, vol. ii. pp. 17-18, 79-80. 



2 Parry C, p. 77, footnote. 



3 It can hardly be determined from the casual records in Arctic narratives, 

 for the drifts would naturally attract more attention in the sledging season. 



4 ' Nature,' vol. xliv. (Aug. 20th, 1891) p. 389. 



5 Belcher, vol. i. pp. 156, 161, 238, 318; Greely, vol. i. p. 183; Lansdell, 

 vol. i. p. 228 (footnote) ; M'Olintock, pp. 210-11 ; Markham, pp. 153-54; Nares, 

 vol. i. pp. 163-64, 188 ; Parry A, p. 169 ; Ross, pp. 171, 177, 208, 618; Wran- 

 gell, pp. 36, 276. 



Q. J. G. S. No. 199. 2 k 



