TRANSACTIONS OF SECTION E, 621 
case with the strain of sudden descent, is no doubt the cause of both phenomena. 
Generally the peculiarities of the great glaciers of Kangchenjunga may be attri- 
buted to a vertical sun, which renders the frozen material less liable to crack, less 
rigid, and more plastic. 
A glacier, as a rule, involves a moraine. Now moraines are largely formed 
from the material contributed by subaerial denudation, in plain words by the 
action of heat and cold and moisture on the cliffs that border them. It is what 
falls on a glacier, not that which it falls over, that mainly makes a moraine. 
The proof is that the moraines of a glacier which flows under no impending cliffs 
are puny compared with those of one that lies beneath great rockwalls. 
Take, for example, the Norwegian glaciers of the Jostedals Brae and compare 
them with the Swiss. The former, falling from a great névé plain or snowfield, 
from which hardly a crag protrudes, are models of cleanliness. I may cite as 
examples the three fascinating glaciers of the Olden Valley. The Rosenlaui 
Glacier in Switzerland owed the cleanliness which gave it a reputation fifty years 
ago, before its retirement from tourists’ tracks, to a similar cause—a vast snow- 
plateau, the Wetterkessel. 
One peculiarity very noticeable both in the Himalaya and the Caucasus I have 
never found satisfactorily accounted for. I refer to the long grassy trenches lying 
between the lateral moraine and the hillside, which often seem to the mountain 
explorer to have been made by Providence to form grass paths for his benefit. 
They may possibly be due to the action of torrents falling from the hillside, 
which, meeting the moraine and constantly sweeping along its base, undermine it 
and keep a passage open for themselves. There are remarkable specimens of this 
formation on hoth sides of the Bezingi Glacier, in the Caucasus, and on the north 
side of the Zemu Glacier, in Sikhim. 
Water is one of the greatest features in mountain scenery. In Norway it is 
omnipresent. In this respect Scandinavia is a region apart, the streams of the 
more southern ranges are scanty compared with those of a region where the snow- 
fall of two-thirds of the year is discharged in a few weeks. Greece stands at 
the opposite pole. By what seems a strange perversity of Nature, its slender 
streams are apt to disappear underground, to reissue miles away in the great 
fountains that gave rise to so many legends. Arcadia is, for the most part, a dry 
upland, sadly wanting in the two elements of pastoral scenery, shady groves, 
and running brooks. 
The Alps are distinguished by their subalpine lakes— 
‘Anne lacus tantos ; te, Lari maxime, teque 
Fluctibus et fremitu assurgens, Benace, marino ?’ 
of Virgil. But perhaps even more interesting to the student are the lake basins 
that have been filled up, and thus suggest how similar lakes may have vanished 
at the base of other ranges. 
I know no more striking walk to anyone interested in the past doings of 
glaciers than that along the ridge of the mighty moraine of the old glacier of 
Val d’Aosta, which sweeps out, a hill 500 feet high, known as ‘ La Serra,’ from 
the base of the Alps near Ivrea into the plain of Piedmont. Enclosed in its 
folds still lies the Lago di Viverone; but the Dora has long ago cut a gap in 
the rampart and drained the rest of the enclosed space, filling it up with the 
fluvial deposit of centuries. 
It is, however, the tarns rather than the great lakes of the Alps which have 
been the chief subjects of scientific disputation. Their distribution is curious. 
They are found in great quantity in the Alps and Pyrenees, hardly at all in the 
ara and comparatively rarely in the part of the Himalaya I am acquainted 
with. 
A large-scale map will show that where tarns are most thickly dotted over the 
uplands the peaks rise to no great height above the ridges that connect them. 
This would seem to indicate that there has been comparatively little subaerial 
denudation in these districts, and consequently less material has been brought 
down to fill the hollows. Again, it isin gneiss and granitic régions that we find 
