Yol. 54.] ON THE GLACIAL GEOLOGY OF SPITSBEEGEN. 215 



fell to freezing-point, this water, being practically quite fresh., was 

 frozen. Owing to the resistance of the ice-belt, the newly-formed 

 ice could expand only on the landward side ; as the young ice 

 presses against the shore, it pushes the beach-material before it 

 into ridges. These are often somewhat irregular in their course, 

 for the water freezes into hexagonal masses, and the ridges neces- 

 sarily acquire the angular course of the margin of the ice that 

 forms them. 



A second characteristic feature of a beach moulded by shore-ice 

 is the occurrence of lines of large, rounded, or subangular stones, in 

 form strikingly like those of the East Anglian Boulder Clay. They 

 are pressed together so closely and regularly as in some cases to 

 resemble a terrace of masonry or a pavement of cobbles. The first 

 beaches of this type that we met with were at Cape Starashchin. 

 The shore is steep, and is broken into a series of platforms ; the 

 faces of the platforms are almost entirely free from boulders, while 

 the terraces between are often paved with closely-packed boulders. 

 Subsequent observation and dredging between tide-lines showed 

 that these peculiar congregations of boulders were due to stranding 

 ice, which had pushed the boulders before it to just above sea-level, 

 and left the beach between tide-lines quite clear. 



A good illustration of similar boulder-terraces on Rolfsoe, near 

 Hammerfest, is given in the atlas to the voyage of the B-echercIie} 

 Baron von JSTordenskiold ^ has described similar boulder- accumula- 

 tions in Nova Zemlya ; he calls them ' stone ramparts,' and gives 

 the same explanation as that at which we had independently 

 arrived. The ' boulder-pavements ' on the shores of the American 

 lakes, which have been admirably described and figured by Prof. 

 J. W. Spencer,^ may be mentioned in this connexion ; but they are 

 different both in aspect and origin. 



These peculiar lines of boulders are probably not easily destroyed, 

 and thus they may sometimes be useful as a test of shore-action. 

 Campbell* has described a beach near Dunrobin in Sutherland, 

 composed of ' terraced piles of boulders which do not seem to be 

 m.oraines,' but which may be an example of this form of ice-action. 

 Campbell himself suggested that it was probably due to the ice- 

 foot. 



(4) The Striation_, Hounding, and Furrowing of Kocks. 



We had few opportunities of observing the action of shore-ice on 

 rocky shores, as our camps during the early period of our visit were 

 on wide raised beaches. But that shore-ice can both scratch and 

 polish rocks we had abundance of indirect proof. The same results 



^ * Voyages en Scandinavia .... sur la corvette la Recherche,^ Atlas geo- 

 logique, pi. vii. 



2 A. E, Nordenskiold, ' The Voyage of the Vega round Asia & Europe,' 

 vol. i (1881) p. 188. 



^ J. W. Spencer, ' Ancient Shores, Boulder-pavements, & High-level Gravel- 

 deposits in the Region of the Great Lakes,' Bull. Geol. Soc. Amer. vol. i 

 (1890) pp. 71-86 & pi. i. 



^ J. F. Campbell, ' Frost and Fire,' vol. ii (1865) p. 155. 



q2 



