44 ROYAL SOCIETY OF CANADA 



of the atmosphere, the temperature ean liave only a secondaiy action in 

 keeping the water at the freezing point. 



In concluding this paper I should like to refer to the disagreement 

 among observers in regard to the density of subaqueous ice. It seems 

 improbable that ice of a lower specific gravity than normal could be 

 formed under any condition. There are certainly no experimental data 

 to support such a supposition. 



The fact that a spong}- mass of frazil floats low in the water, does not 

 furnish anv proof, especially when it is considered into how small a mass 

 of solid ice quite a large quantity of these fine needle crystals could be 

 pressed. 



It is quite conceivable that the earthy matter contained amongst the 

 ciystals might become suflScient to counteract the force tending to float 

 them. Anchor-ice does not rise, because in general, it is formed iti situ. 

 When, however, the Av^armer layers of water sink to the bottom on a mild 

 day so as to loosen the hold it has upon the stones or boulders, the force 

 tending to raise the ice becomes sutficient to tear it away from its found- 

 ation and it rises to the surface. This force is often sufficient to lift stones- 

 of considerable size. 



