64 



The term Columnar is applied by Scoresby to that form of ice which 

 la more especially the object of this paper, and is accurately descriptive of 

 it. He adds, however, the word crystallization, apparently regarding each 

 column as a crystal. It is not so, however, each column being rounded and 

 excepting that the intimate structure of all ice is of course crystalline, to call 

 these columns crystals must lead to misapprehension. I ought to add, perhaps, 

 that I provisionally retain for it Scoresby's name of ground-ice, which also well 

 describes its usual situation. I do not know whether this form of ice or that 

 which forms in the bods of livers has the prior claim to the name, but as 

 they arc very distinct phenomena, the one that has not the priority ought to 

 receive a different name. 



The usual position in which this form of ice is to be met with is an 

 inclined surface of earth, occasionally it is to be seen on level ground, but its 

 peculiar structure is then rare y so clear and distinct, principally, I believe, 

 because the simply of water necessary is apt, during frost, to fail on level 

 ground ; the beautiful instance observed by Scoresby occurred in such a 

 situation. The place where I most carefully examined it was on a steep clayey 

 bank, at an angle of 70° to 80° to the horizon. 



On this bank, after a few nights sharp frost, the thermometer being at or 

 below 20°, there would be a coating of clear ice of five or six inches in thickness, 

 not solid but full of perforations of j to v. an inch in diameter, traversing it 

 perpendicularly to the surface of the bank, so that looking along these perfora- 

 tions the clay of the bank could be seen behind the ice. Often these perforations 

 were of simple cylindrical shape, more usually of complicated outline from 

 several cylinders coalescing to form one space, so that the configuration of the 

 ice between them might be briefly described as labyrinthine. Nevertheless, the 

 ice is sufficiently strong to make it a matter of some difficulty to remove a 

 portion. On doing so it is seen that the icy columns dwindle away towards the 

 bank in some places, in others they become wider and thicker. The surfaces 

 of the columns of ice are smooth and rounded, not angular and crystalline. 

 In some parts of the structure the air spaces so preponderated over the ice that 

 the latter was reduced to a number of completely isolated columns, looking 

 like icicles growing upwards instead of downwards. Sometimes such columns 

 having, to a certain extent, fallen together by their own weight, were, never- 

 theless, tolerably perfect for several inches in length, and of diameters varying 

 up to a quarter of an inch. Usually this more strictly columnar form was 

 much marred by the ice having fallen away from the bank in patches, upon 

 which a fresh croxi appears, all usually more or less twisted and contorted. I 

 have seen many intermediate forms connecting this with the variety that I have 

 called labyrithine, showing that their mode of formation is the same. 



The surface of the ice was usually soiled by a little clay raised from the 

 bank, and the ice was divided into strata by layers of clay-soiled ice. I found 

 that the strata of ice corresponded in number with the number of nights of frost, 



