H. T. Fervar — The Antarctic Ice-cap. 



531 



showed that the density is not uniform. Our experience upon the 

 Boss Ice-sheet proved that normal glacier ice could be capped by any 

 thickness oi firn or loose snow; and therefore of two bergs with 

 equal volumes immersed, if the visible part of one consisted wholly 

 of snow and that of the other wholly of ice, the ratios of the visible 

 to the submerged portions would be widely different. Again, the 

 temperature and salinity of tbe sea- water have some effect in buoying 

 up the ice, but this may be neglected. However, the aeration of the 

 ice varies greatly, but for normal glacier ice Helland gives 0-886 as 

 the specific gravity, while my results show a specific gravity of 0-83, 

 Therefore, in Antarctic sea- water with an average density of 1-025 

 grammes per cubic centimetre, glacier ice would float with four parts 

 immersed and one visible. This figure agrees closely with the 

 results of Von Drygalski and Captain Scott obtained by actual 

 measurements at glacier snouts. The greatest ice-cliff observed by 



Fig. 2. — The edge of the Eoss Ice-sheet near the point -where it is thickest. 



the staff of the " Discover}'^ " was the edge of the Ross Ice-sheet 

 floating in the vicinity of Cape Crozier. The height here was 240 

 feet and the depth of water 460 fatboms. As the surface here was 

 unbroken and the cliff-face uncrevassed (see Fig. 2), it is hardly 

 likely that this extreme thickness was produced by pressure as 

 Professor Schwarz ' suggests. 



The question as to the limit of 1,600 feet being exceeded depends, 

 as Professor Schwarz rightly points out, upon the temperature of the 

 lower surface of the ice being below its melting-point. As far as 

 I remember, the figure 1,600 feet is the height of a column of ice 



1 Schwarz : Geol. M.^g., Dec. Y, Vol. Ill, March, 1906, p. 122. 



