Under these conditions it is obvious that the magnetic needle is quite useless for steering purposes. The sun compass proved itself a more than efficient substitute. On a snowfield there is usually a total absence of landmarks of any kind, so the direction of wind, sastrugi, or perhaps a low cloud is found with the sun-compass, frequently checked, and the course kept accordingly. On camping we would generally carefully note the direction in which the sledge was left, in case the next day proved overcast. Thus we would march in the morning by the wind's direction till the sun, gleaming through the clouds for a few moments, enabled us to use the compass again. Sastrugi, only six inches high, seen on the 26th, showed the effects of wind-erosion exquisitely. In an individual case the windward end of a sastruga might be completely undercut for six or nine inches, leaving a hard crust, sometimes only one-eighth of an inch in thickness and a couple of inches wide. This would sag downwards under its own weight in a fine curve till the tip rested on the snow beneath. It is marvellous how such a delicate structure can withstand the heavy wind. November 27 proved a very hard day. The wind kept up sixty miles per hour all the time, so that, after taking four hours to do four and three-quarter miles, we were all thoroughly exhausted. It was not a great run, but the century was hoisted--one hundred and three-quarter miles by sledge-meter; altitude two thousand nine hundred feet. There was a mild celebration that night over a square of butter-scotch and half an ounce of chocolate, besides the regular hoosh and cocoa. Next day the light was very bad and the wind fifty miles per hour. Observations were therefore made inside the tent. Webb, Hurley and the instrument occupied all available space, while I spent three hours digging a shaft eight feet deep in the snow, taking temperatures every foot. It appeared that the mean annual temperature of the snow was approximately -16 degrees F. The dip was 88 degrees 54'; certainly rather too large a rise from 88 degrees 20' of twenty miles back. The declination had actually changed about 80 degrees in the last ten miles. This one-hundred-mile station was badly disturbed. From the evidence, it is possible that a subsidiary "pole" or area of almost vertical dip may exist close by this spot to the west or south-west. Going straight up wind into a "blow" which varied from forty to fifty miles per hour, we were able to make eight miles after the previous day's rest. At lunch a hole was dug five feet square and two feet deep. It served three purposes. First, it gave a good shelter for a longitude observation; secondly, with the mast, yard and floor-cloth we converted it into a shelter snug enough to house the primus and to lunch comfortably; and thirdly, a mound was left as a back-mark which was picked up on the return journey.