GLACIOLOGY OF THE SOUTH ORKNEYS. 855 



glacier ice ("intermediate") ; and the third 4 feet into the ice ("deep"). The depth 

 below the actual surface, of course, varied with the accumulation of snow, increasing 

 up to a maximum about the month of October, and diminishing rapidly in summer (see 

 text-fig. 5, p. 842). Thus the shallow thermometer varied from 18 to 30 inches below 

 the surface of the snow ; the intermediate from 2 feet 8 inches to 4 feet 6 inches ; and 

 the deep one from 4 feet 8 inches to 6 feet 6 inches. The thermometers used were 

 of the ordinary British Meteorological Office pattern of earth thermometer, i.e. hollow 

 metal tubes with thermometers slung from the caps. For this sort of work these are 

 not very satisfactory. When the cap was taken off to read the temperature it was 

 impossible at times to prevent the drifting snow from blowing into the tubes, and this 

 occurred to such an extent as finally to block some of them, and render readings for 

 a time impossible, so that there is a hiatus in the observations. Moreover, as the 

 tubes had necessarily to project some way above the surface to allow for increase of 

 snow accumulation, the air column in the interior, with such a variable external air 

 temperature^ must have varied more than the surrounding ice, so that the readings 

 of the thermometers probably did not give accurately the temperature of the ice at 

 that particular level. Had the thermometers been protected by a small wooden box 

 or house, it might have been better ; but the ideal method would be platinum resistance 

 thermometers, with the registering apparatus under cover some little distance off". 

 Another drawback to the pattern employed is that during warm spells melting went 

 on at the sides of the tubes, and the water trickled down, so that temperatures actually 

 above 32° F. were sometimes registered by the shallow thermometer even in winter — 

 a state of affairs which could not, of course, correspond with the true temperature 

 of the snow or ice at that level. The actual figures obtained are shown plotted on 

 the chart (Plates II. A and II. B), along with the mean daily air temperature in the 

 shade for the same periods. 



The great irregularity of the air temperature (in winter) makes it rather difficult to 

 trace the progress of a temperature wave through the ice, but it would appear as if it 

 reached the shallow thermometer at the junction of the snow and ice within twenty-four 

 hours ; the intermediate or 2 -foot thermometer in about forty-eight hours, and the deep 

 or 4-foot one in about five days. 



From May to September the mean daily shade temperature of the air varied from 

 — 21°"8 F. to 34°'2 F., whilst the temperature at the snow and ice junction varied 

 from 2°*9 F. to 34°*0 F. ! Two feet below the surface of the ice the variation was 

 from 12°'6 F. to 27°-0 F., and four feet into the ice from 16°-9 F. to 25°-9 F. 



In October there was a steady rise in the air temperature and in that recorded by 

 the shallow thermometer (the other two were not at this time in working order). 



Between November 1903 and February 1904 (both inclusive), the mean daily air 

 temperature only varied between 22° '2 F. and 37°*7 F., and the shallow thermometer 

 between 27° "9 F. and 32° '2 F. A temperature of 32° F. was permanently recorded 

 in the shallow thermometer after 15th December until 1st January 1904, when the tube 



TRANS. ROY. SOC. EDIN., VOL. XLIX., PART IV. (NO. 15). 116 



