Hols? s Studies in Glacial Geology. 711 
terminal moraine in front of the ice-rivers, and the material is 
gradually worked over by the jékel-rivers. 
The force of the jékel-rivers is greatest nearest the inland-ice and 
diminishes as they approach the sea. In consequence, the greatest 
stones are found near the terminal moraine, whilst further on their 
size is reduced more and more until all is sand, spreading out to 
wide sand-plains, as is the case, for instance, off Fredrikshaab’s ice- 
blink. The finest impalpable material is carried out into the fjords 
and open sea, where it forms deposits of clay. 
The upper drift of Greenland shows a considerable resemblance 
to that of Sweden. Both are free from boulders. The gravel is 
assorted and stratified. The stones are well ground, although more 
rounded in Sweden than in Greenland. These deposits in the latter 
country are not unfrequently of a considerable thickness. In the 
bottom of the Tasiussak tjord and in a few other places they 
measured about 100 feet. 
There are, however, no typical osar in the part of Greenland 
visited by Dr. Holst, who found only some smaller hills to a faint 
degree resembling those formations. There can be no doubt that 
the osar are formed near and in close relation to an inland-ice. 
Nothing but such ice could have transported these masses to their 
present locations, and nothing but the enormous force of the torrents 
rushing from the ice could have wrought the material so thoroughly. 
Still it may be less certain that the ose form has been caused in 
Greenland by the same agencies as those that produced glacial osar 
in Sweden. It may be a mere accidental resemblance and the 
form may depend on later excavations. Such osar were formed 
within the time of the melting of the inland-ice. The coast-land of 
Greenland presents the same character as Sweden so far that it has 
formerly been covered with inland-ice which has long ago melted 
away. Why then is it that typical osar do not exist in the said 
district of Greenland ? Dr. Holst finds the answer to this qnestion 
in the topographical differences of the two countries. He refers to 
his earlier discussion of the formation of the glacial osar in Sweden, 
4 summary of which discussion was given at the beginning of this 
article. As stated there, such osar are formed as sediment in the 
beds of rivers, having cut their channels into the surface of an 
inland-ice. The requisite conditions for the formation of large osar 
