arête and look for a more promising route on our right. Here we proceeded cautiously, crawling through a narrow niche in some overhanging rocks with a precipice of some hundreds of feet below. Then the climbing improved till our view upwards was bounded by an indefinite saddle in the rocks, which might have led to anywhere, but which did lead, as we subsequently found out, to the easy snow slopes above.
As the day advanced small falls of stone occurred, which caused some annoyance and danger, but we managed to avoid being struck by any. Then followed another stretch of rotten rock which Fox absolutely declined to tackle, and as it could not be turned by a detour we were brought up on this route.
Fox suggested descending again to cross a large glacier coming down from the ridge on our right, and trying the rocks on its opposite side. This plan we eventually carried out, but it was a fatal mistake as far as climbing Aorangi was concerned. Descending for about 1,000 feet we stepped on to the ice of what we then thought was the lower part of the Linda Glacier—owing to a strange error in Von Lendenfeld's map—but which in reality was the Freshfield Glacier. We put on the rope and our goggles, both indispensable in crossing such a snow-covered ice stream.
On taking to the rocks on the other side we soon gained the lowest ice slopes, covered with six or eight inches of snow in splendid order, and adhering well to the ice; now and then we took to the rocks, but climbed mostly by the snow slopes till we reached the
