248 DR JAMES GEIKIE ON 
owing to the absence of any well-defined Lee-sette, and the disappearance of the 
striae, we could not be certain as to the direction followed by the ice. On the 
opposite side of the island, however, we found strongly marked roches 
moutonnées, and very fine examples of striation. As these are perhaps the best 
preserved specimens to be found in the Ferde Islands, it may be well to 
indicate the precise locality. We met with them at the point which forms the 
south-west limits of Sandsbugt. Close to this point there is a deep ragged cleft 
in the rocks into which the sea has access by a subterranean passage. The 
dolerite at this place shows fine strize pointing to 8. 40° W., but the best example 
occurs on the headland at the point itself. Here the roches moutonnées indicate 
very clearly the direction of the ice-flow, and the strize (S. 40° W.) are particularly 
sharp and fresh. Nearer the village of Sand, we found strize with a more 
southerly trend—S. 15° W. Of the interior part of Sandée I can say very little 
—for we traversed it in a dense drizzling fog. We could only see that voches 
moutonnées and ice-worn rocks accompanied us across the hills. 
The higher parts of Skuée, as seen from our boat, appeared to be smoothed 
off from the north or north-east ; but Store Dimon and Lille Dimon, when we 
passed them the first time, were shrouded in mist, and on our return from 
Suderée rough weather prevented us approaching them. 
Suderde has supported a considerable mass of ice, for we traced the 
glaciated outline up to a height of 1400 feet. Above that level all is rough, 
angular, and serrated. The low ground that extends from the head of Qvalbée- 
fiord to the west coast is highly moutonnée, the position of the Stoss- and Lee- 
seiten indicating an ice-flow from east to west. Here also the striz point 
E.and W. In Trangjisvaag valley, the direction of glaciation is towards south- 
east, as shown both by roches moutonnées and striz. Both sides of the fiord 
into which this valley opens are highly glaciated in the same direction. At 
Ordevig, the strize point E. 30° N., and correspond in direction with the trend of 
the valley in which they occur. The fine cirque-like valley of Howe affords 
admirable examples of glaciation. The whole broad amphitheatric space has 
been filled with ice, like a great reservoir; the flat bottom being thickly set 
with roches moutonnées, and the smoothed and rounded glacial contour rising 
on the hill-slopes to a height of 1400 feet. The upper part of the valley is 
sprinkled with many lakelets, which rest in true rock-basins. Striz are not 
abundant, but we noticed them in several places, and they all pointed to the 
east, or down the valley. Another finely glaciated cirque valley descends from 
Kyvannafield and Borgaknappen to the cliffs on the west coast, north of Fama- 
rasund. The ice that filled Howe valley must have brimmed over and become 
confluent, not only with the Trangjisvaag ice, but also with the glacier masses 
that descended the Dalbofos valley and Waagsfiord, for the hills above Porkerji 
and Naes are strongly glaciated all over. The trend of the abraded rocks on 
