44 THE ROYAL SOCIETY OF CANADA 



than a very small fraction of a degree, from the freezing point. Hence 

 we cannot have changes of length taking place at this surface.by pure 

 expansion or contraction. The temperature falls upwards through 

 the ice to the surface, the gradient depending on the thickness of 

 the ice and the period of the cold or warm wave affecting the upper 

 surface. Contraction takes place in the upper surface, which causes 

 the sheet to bend downwards as we have represented by the dotted 

 contour. Under the tension the ice cracks, as represented in the dia- 

 gram. No force of a harmful nature can be exerted at the edges of 

 the ice sheet, under these circumstances. As long as these cracks 

 remain open, they act as pressure cracks, to take up the thrust of the 

 ice, when the temperature rises. In general we may assume, however, 

 that these cracks will become filled up either by snow or by water 

 coming over the ice surface. The ice being cold will cause the cracks 

 to freeze solid and cement the whole sheet into a firm mass. 



When a rise of temperature results, the top surface of the ice 

 will expand while the under face will remain as before, free from 

 strain. This will cause the ice sheet to bend upwards as represented 

 by the dotted figure. Relief cracks will start on the underside, and 

 the ice sheet will remain as before, level on the surface of the water. 

 There is no question but that the under surface cracks will at once 

 fill with water and freeze, so that a permanent increase in the length 

 of the sheet will result. 



There seems every reason to believe that an ice sheet may gradu- 

 ally expand by small increments, where the cracks are filled and frozen 

 and hence a permanent thrust will result. 



Pressure ridges on large lakes are seldom seen to disappear again, 

 but we believe these are permanently formed by this gradual increase 

 in length of the ice. Working back and forth in these ridges will 

 take place to some extent as was observed by Mr. Sawyer, but we 

 believe these ridges are gradually increased with every great change 

 of air temperature. 



Crushing and Tensile Strength of Ice. 



Crushing Strength. 



One of the authors (H.T.B.) has recently determined the crushing 

 strength of ice and this forms the subject of a separate paper presented 

 to this Society. In that paper it is shown that the crushing strength 

 of clear blocks of St. Lawrence river ice at or near the freezing point 

 is very near to 363 pounds per square inch. This agrees very well 

 with the values assumed by many Engineers, which is in round numbers 



