258 DR JAMES GEIKTIE ON . 
by the mechanical action of the springs, but they are of course abundantly 
subject to the action of frost, and immense quantities of débris are thus 
showered down the precipices. One can see that the denudation is going on 
rapidly in many places, and thus gradually destroying the glaciated aspect of 
the ground. 
The low dip of the bedding greatly aids the frost and springs in their work 
of destruction, and now and again considerable landslips take place in 
consequence. At Tidrnevig in Stromde we saw the cultivated grounds to a 
large extent buried under masses of débris and large blocks which had fallen 
suddenly only a year or so before our visit, and the evidence of similar Jandslips 
having occurred in many of the higher valleys of the islands was abundant and 
conspicuous. So rapid, indeed, is the destruction of the mountains that one is 
apt to wonder that the valleys are not more plentifully covered with débris ; for 
the streams are insignificant, and hardly able to carry seaward any large propor- 
tion of the débris showered down the slopes by springs and frosts. Nevertheless, 
some of the streams during floods would seem to overflow wide areas, and 
probably carry seaward no inconsiderable amount of material. They are fast 
silting up the lakes, and many of these have already been obliterated. And 
the same silting-up process is going on in the fiords. Thus Saxen Fiord, which 
was once a good harbour, and could be entered by sloops and other vessels, 
will now hardly admit a small boat. It was not quite low tide when we entered, 
and yet our boat grounded several times, and was only brought to shore by dint 
of vigorous pushing. I have little doubt that could the quantity of material 
carried down by the streams be fairly estimated, we should find it considerably 
in excess of what might have been supposed. 
It is evident that springs and frosts have been among the chief agents in 
widening the valleys. When the streams first cut down into the basalt-rocks, 
they doubtless flowed in more or less deep trenches—the walls of which, under- 
mined by springs and riven by frost, would gradually recede. Moreover, owing 
to the regularity of the bedding, and the low dip of the strata, the retreat of 
the opposing cliffs would be tolerably uniform, so that each valley would tend 
to retain a somewhat equable breadth throughout. But as the streams in their 
course traversed a series of beds—some of which would yield to denudation 
much more readily than others—the waters would descend in a succession of 
runs and leaps. Each hard bed of trap, which happened to be underlaid by a 
more or less thick band of soft tuff or decomposing earthy amygdaloid, would 
give rise to a waterfall, the crest of which would gradually retreat up the valley 
as the trap continued to be undermined by the rapid wasting-away of its pave- 
ment. It would rarely or never happen, however, that all the waterfalls in a 
valley would retreat at the same rate, and thus one would in the course of time 
overtake another, with which it would coalesce, as it were, to form a higher fall. 
