ice-work he had encountered. We spent about half an hour more among the broken ice, which gave Graham an opportunity of displaying his splendid ice-craft, for which he is well noted. At last we emerged in the snow at the head of the glacier, and climbed slowly but steadily up till we reached the foot of the saddle. Here we called a halt, and indulged in a second breakfast while picking out a route. The wall of the saddle rises at an angle of 70 degrees, and is only comparable with Harper's Saddle at the head of the Hooker Glacier. Like the latter, it is seamed across with a great crevasse. We found it impossible to continue straight up for the lowest point of the saddle, the crevasse being uncrossable except at the extreme left. Even if we could find a bridge there we must then traverse back the whole length of the slope. We looked in vain for any other route, so decided to attempt it there. We quickly climbed the soft slopes to the bottom lip of the crevasse; then followed this along till at last we found a nasty but possible crossing-place. The edge of the crevasse dipped sharply down, making a bridge that ended in a pillar of snow about 2 feet broad and 5 feet high. From the top of this it was possible to reach the upper lip of the crevasse. Peter cut his way up it, and with my ice-axe well driven into the snow I played the rope round it. Suddenly I heard a sharp crack. Peter heard it too, and he stood very still on top of the pinnacle. Nothing more happened, so with a laugh he looked down to me and said, "False alarm," and proceeded to cross the crevasse. It was a long stretch, but with the aid of his axe dug in the wall above he managed it, then kicked his way up a snow slope for about 20 feet, and anchored himself safely until Milne and I arrived successfully.
About 50 feet higher up we had another crevasse to cross, and at last found ourselves clinging like files to a window-pane, on a wall of ice, at an angle of 70 degrees.
