Meanwhile Hassel and I jogged on, and the others followed.  We had to get through a good many difficulties yet before we reached that point, but, compared with all the breakneck places we had already crossed, these were of a comparatively tame description.  It was with a sigh of relief that we arrived at the plain that promised so well; its extent was not very great, but we were not very exacting either in this respect, after our last few days’ march over the broken surface.  Farther to the south we could still see great masses piled up by pressure, but the intervals between them were very great and the surface was whole.  This was, then, the first time since we tackled the Devil’s Glacier that we were able to steer true south for a few minutes.

As we progressed, it could be seen that we had really come upon another kind of ground; for once we had not been made fools of.  Not that we had an unbroken, level surface to go upon —­ it would be a long time before we came to that —­ but we were able to keep our course for long stretches at a time.  The huge crevasses became rarer, and so filled up at both ends that we were able to cross them without going a long way round.  There was new life in all of us, both dogs and men, and we went rapidly southward.  As we advanced, the conditions improved more and more.  We could see in the distance some huge dome-shaped formations, that seemed to tower high into the air:  these turned out to be the southernmost limit of the big crevasses and to form the transition to the third phase of the glacier.

It was a stiff climb to get up these domes, which were fairly high and swept smooth by the wind.  They lay straight in our course, and from their tops we had a good view.  The surface we were entering upon was quite different from that on the northern side of the domes.  Here the big crevasses were entirely filled with snow and might be crossed anywhere.  What specially attracted one’s attention here was an immense number of small formations in the shape of haycocks.  Great stretches of the surface were swept bare, exposing the smooth ice.

It was evident that these various formations or phases in the glacier were due to the underlying ground.  The first tract we had passed, where the confusion was so extreme, must be the part that lay nearest the bare land; in proportion as the glacier left the land, it became less disturbed:  In the haycock district the disturbance had not produced cracks in the surface to any extent, only upheaval here and there.  How these haycocks were formed and what they looked like inside we were soon to find out.  It was a pleasure to be able to advance all the time, instead of constantly turning and going round; only once or twice did we have to turn aside for the larger haycocks, otherwise we kept our course.  The great, clean-swept stretches of surface that we came upon from time to time were split in every direction, but the cracks were very narrow —­ about half an inch wide.