292 SCOTT'S LAST, EXPEDITION 
of parallel valleys. These are about 10 or 12 miles long, and 
are in many cases occupied by small glaciers in the western half 
of the valley.. They are identical with the ‘ finger’ valleys de- 
scribed in the reports on the glacial geology of the Rocky Moun- 
tains, U.S.A. Narrow ridges about 3000 feet high separate 
them. Some ‘hang’ a thousand feet above the Koettlitz. Char- 
acteristic hills, triangular in plan, occur where these valleys join, 
and all of them ‘head’ in beautiful cwms. Above these cwms, 
and more cwms, fret the scarp of Mt. Lister over the whole 
extent of its 10,000 foot face. There is little doubt that we have 
here an example of the way the glacial cycle commences its opera- 
tions, for this is a fault scarp of comparatively recent date. 
(c) Granite Harbour, like New Harbour, is probably a relic 
of the period of glacial maximum when the ice flood exerted tre- 
mendous erosive power on its bed, and was able to erode far 
below sea-level. We shall however never be able to witness these 
maximum forces in operation. Because a dwindling river has 
little effect on the topography it would be foolish to deny the 
action of a great river in flood; just as our observations in the 
Antarctic on a nearly stagnant or receding glaciation are not 
to be taken as descriptive of the most active periods in glacial 
history. 
The first feature that strikes the geologist is that as one pro- 
ceeds north there is less and less land exposed below the snow and 
ice mantle. This implies, I think, that the precipitation in the 
south-west corner of McMurdo Sound is extremely little, and in- 
creases both northward and southward. The most striking fea- 
ture in the harbour—the Ice Tongue—has been described in the 
narrative. The Mackay Glacier moves 3 feet a day, as already 
recorded. Mention must be made of the ridge separating the 
new glacier from the Devil’s Punch Bowl. This has certainly 
been covered quite lately by the new glacier. The harder dykes 
are striated, but the ridge is for the most part covered with gran- 
ite debris. ‘There is practically no englacial rock débris in the 
glacier, so that one is led to the important conclusion that the 
floor of the new glacier is covered with rock débris and that no 
erosion is taking place under this fairly large glacier. What was 
the floor of the Mackay Glacier at its period of greater area is 
exposed in many places 1000 feet above the sea ice in the form 
of rock-strewn plateaux. 
