484 THE ANTARCTIC MANUAL. 
the snow lying on the flat shelves contrasted with the dark colour 
of the steep slopes and thus emphasised the structure. The same 
thing was seen in other places where similar conditions prevailed, 
and it seemed that these steps were old levels of the snow-field, 
and their existence proves a former greater extension of the névé. 
While the névé remained at one level, the exposed portion of the 
rock wall crumbled under the influence of atmospheric agencies, while 
the part covered by snow was protected, hence the cutting of the 
step. 
At noon, Lecointe, Danco, Racovitza, Amundsen and I landed with 
the instruments necessary for astronomical and magnetic observations ; 
FIG. 8.—MOUNT WILLIAM. 
but, unfortunately, we were too late to observe the meridian altitude, 
a misfortune the more regrettable because Lecointe had urged the com- 
mandant to allow us to land sooner. We remained some time on the 
strip of bare rock which was exposed between the field of névé and the 
sea. It was the same black granitoid rock traversed by thick veins and 
narrow threads of quartz; and there was a great variety of erratic 
blocks, including specimens of basalt, breccia, several blocks of con- 
glomerate, and some fragments of quartzite. A cave was found in the 
large-grained, porous ice-wall, along the uncovered bed of which a little 
stream flowed, the first glacier stream 1 had seen. It came from the 
direction of a nunatak, and consequently could not have pursued its 
course to a long distance under the ice; in its bed there were rolled 
pebbles of eruptive rocks. We were only two hours ashore, so it was 
impossible to get as far as the nunatak. At our landing-place, and for 
some distance out from the shore, we saw the bottom of the sea very 
distinctly, and in some places could even touch it with the oars; it was 
composed of pebbles and boulders. 
At 5p.m. the Belgica resumed her voyage southward, and we entered 
