THE GEOLOGY OF THE FAROE ISLANDS. 257 
of the land have been hollowed out along the course of large dykes. Many of 
the small veins which are now exposed to the action of the weather certainly 
crumble rapidly away, and so give rise to more or less deep gullies. We have 
no reason to believe, however, that any dykes actually reached to the original 
surface of the old plateau. And if they did break through that surface they 
must have overflowed, and cooled under circumstances which would necessarily 
produce a rock differing considerably in character from that of the dykes which 
we now see, but probably approaching to that of the dolerites which they 
intersect. 
5. Atmospheric Erosion.—From the appearances presented by the land- 
valleys there can be no doubt that these hollows owe their origin to the action 
of the usual atmospheric agents, aided by the subsequent erosive powers of 
glacier-ice. To make this clear, a few notes on the nature of the subaerial 
denudation now going on seem desirable, and these I shall supplement with 
some remarks on the effects that have been produced by former glacial 
action. 
Although the basalt-rocks of the Frde Islands, when freshly exposed, are 
hard and tough under the hammer, yet their composition and structure render 
them peculiarly liable to more or less rapid denudation. Not only are they 
frequently decomposed by the chemical action of the atmospheric forces, but 
their abundant jointing enables frost to act most effectively upon them, while 
the work of demolition is still further aided by the horizontality of their bedding. 
I have already mentioned the fact that some of the basalts weather more readily 
than others, and that even in one and the same bed there are often great 
differences in this respect. Thus certain basalts show rough irregular surfaces 
—the wacké-like portions crumbling to earth and sand—while the harder parts 
weather less rapidly, and thus amorphous hollows, and ruts, and shapeless humps, 
knobs, and ridges come to diversify the faces of the rocks in many localities. 
These appearances, however, are best seen in the sea-cliffs. In the inland 
valleys the hollows in the rock are often masked by the fall of débris from the 
ledges above, and are only conspicuous upon very steep cliffs where the loose 
material finds no angle of repose. As some basalts weather more rapidly than 
others, their demolition often leads to the destruction of the harder masses 
that overlie them. The latter are undercut, and by and by large segments 
split off and fall away. But the undermining of the rocks is carried on in the 
most marked manner by springs which frequently issue in abundance all along 
the outcrops of certain beds, particularly when more or less impermeable tufts 
alternate with the basalts. This action results in a more or less steep cliff, 
broken by little sloping ledges which are often thickly carpeted with bright 
green mosses, in striking contrast to the bare walls of brown rock above and 
below them (see Plate XV. fig. 23). Not only are the basalt-rocks undermined 
WOl, XXK. PART I. 2Q 
