54 Proceedings of Royal Society of Edinburgh. [sess. 
and finds expression in the numerous votive crosses exposed to view 
in convenient corners along the road in the Saas Valley. 
Glacier-lakes, formed after the manner of the Mattmark See, are 
numerous, and are to be found not only in Switzerland but over 
the whole geographical area formerly occupied by glaciers. On the 
other hand, those resulting from the blocking up of the entrance to 
a tributary valley are comparatively rare, and only one or two 
examples are at the present time known to be in existence. Their 
comparative rarity is due to the fact that the obstructing dam is 
invariably composed of ice, and the eflPect of the lateral moraine, 
should any exist, may be disregarded. The continuance of the lake, 
therefore, depends entirely on the evanescent character of the glacier, 
and no lake remains after its disappearance. 
The most noted glacier-lake of this kind, and the best known in 
Europe, is the Marjelen See, situated at the foot of the JEggischhorn, 
on the left bank of the Great Aletsch glacier. It lies in a small 
lateral valley which, at its upper end, communicates by a low 
“ col,” or pass, with the adjacent valley of the Viesch ; and it is by 
this “ col,” and not by the glacier, that in ordinary circumstances 
its surplus water is discharged. If, however, in consequence of 
some accidental fissure, or other change in the ice, the water finds 
an escape in this direction, it does so with such rapidity that some- 
times in a few hours the lake disappears altogether, so that nothing 
is to be seen in its former bed but some stranded masses of ice and 
a small stream at the bottom of the basin. The lake then begins to 
refill, and after about a year the water reaches its former level, at 
which it remains stationary till some other structural change in the 
ice-dam allows it to escape, and the phenomenon of its sudden 
drainage is repeated. The devastation usually caused by these 
floods on the lower lands along the banks of the Massa and the 
Khone (into which the former falls), recently induced the interested 
landowners to deepen the outlet by the “ col ” so as to lessen the 
water capacity of the Marjelen See, and, of course, to proportionally 
diminish the damage from these periodically recurring inundations. 
The Marjelen See was visited in 1865 by Sir Charles Lyell, and its 
then condition is thus described in his Principles of Geology^ 11th 
ed., p. 373: — “The Marjelen See was about 2 miles in circum- 
ference when I visited it in August 1865, and about 40 feet below 
