THE GEOLOGY OF THE F4ROE ISLANDS. 255 
such as Qvanna Sund, Svinéefiord, Harald Sund, Kalscdefiord, Leervigsfiord, 
Westmannsfiord, &c. An elevation of 200 or 300 feet would probably suffice 
to run the sea out of all these fiords and sounds and convert them into valleys, 
communicating with each other across low level cols. They would, in short, 
resemble the long hollows that have already been described as crossing the 
islands from shore to shore. Conversely, were the islands to be submerged for 
200 or 300 feet, new sounds and fiords would make their appearance, and 
Stromée would be cut up into three separate islands, Osterde into two, Borde 
into two, and Suderoe into three, while the present fiords would then cover all 
the low grounds at their origin, stretching back into those broad amphitheatric 
cirques which are so prominent a feature in the configuration of the Feerée 
Islands. 
Although the fiords are never very deep, they yet, as we have seen, resemble 
those of Scotland and Norway in this respect, that they are shallower at or 
near their mouths than somewhat further up. Unfortunately, the soundings 
indicated upon the chart are not numerous, and reliable details are thus want- 
ing. But the fishermen assured us that it was certainly true that the fiords 
were deeper above than below their entrances. The soundings between 
Osterde and Stromée show that Sundene at least has this character ; and it is 
interesting to observe that the deepest portion of that fiord occurs just where 
it should if the depression owed its origin to the grinding action of the ice 
that flowed towards the north. There appears also to be a deep excavation in 
the sea-bottom to the north of NaalsGe, comparable to the deep hollows that 
are met with along the inner margin of the Outer Hebrides and many other 
islands off the Scottish coast—hollows which I have elsewhere shown must be 
attributed to the erosive action of glacier-ice.* 
3. Trend of Valleys and Fiords: Main Water-parting.—One sees at a glance 
that the Ferde Islands are only the more mountainous parts of a region which 
has been submerged within comparatively recent geological times. And it is 
not difficult to account for the north-west and south-east trend of so many of 
the valleys and fiords. If we draw a somewhat undulating line from north-east 
to south-west between Svinde and Waagoe, we traverse in so doing the main 
water-shed of the islands, and what must likewise have been the chief water- 
shed when the land stood several hundred feet higher. Now, if we suppose 
the original surface of the land to have been gently undulating, there can be 
little difficulty in accounting for the trend of the valleys and fiords. What is 
now the main water-parting will represent the low undulating water-shed of 
the old table-land—the ground to north or north-west of the line having a gentle 
fall in that direction, while to the south of the line the inclination would be to 
* The Great Ice Age, p. 289; Quart. Journ. Geol. Soc., vol. xxxiv. p. 861; Trans. Geol. Soc. Glasg., 
vol. vi. p. 161. 
