THE NORWEGIAN LEMMING AND ITS MIGRATIONS. 
147 
they uttered cries of rage. I collected 500 skins, with the idea 
of making a rug, but was surprised to find that a portion of 
the rump was nearly always denuded of hair, and it was long 
before I discovered that this was caused by the habit of 
nervously backing up against a stone, of which I have just 
spoken. As this action is excited by every appearance of an 
enemy it seems surprising that a natural callosity should not 
take the place of so constant a lesion ; possibly, however, rhe 
time during which this lesion occurs is too short to cause con- 
stitutional change. 
Early in the autumn, and just a year after their arrival at 
Heimdalen, the western migration commenced anew. Every 
Fig. 2. 
Plan of Heimdalen drawn to scale, in which the course of the Lemmincs, 
indicated by the arrows, is seen to cross Lake Heimdalsyand and the swift 
river Leirungen, both of which might be avoided by a slight detour. The 
river is of glacier origin, very cold and very rapid. 
morning I found swarms of lemmings swimming the lake 
diagonally instead of diverging from their course so as to go 
round it, and mounting the steep slopes of Heimdals-ho (figs. 
2 and 3) on their way to the coast, where the harassed crowd, 
thinned by the unceasing attacks of the wolf, the fox, and the 
dog, and even the reindeer, pursued by eagle, hawk and owl 
(See PI. IV.), and never spared by man himself, yet still a vast 
multitude, plunges into the Atlantic Ocean on the first calm 
day, and perishes with its front still pointing westward. No 
iaint heart lingers on the way ; and no survivor returns to the 
mountains. 
There appears to have been a difficulty in keeping these rest- 
less creatures in captivity, both because they escape through in- 
