246 DEPARTMENT OF THE XAYAL SERVICE 



second, and by 1 o'clock 1-26 cm.'' per second. After this it decreased, chiefly owing to 

 the fact that the surface of the ice block had diminished in size, but also on account of 

 the lower temperature of the water in the tank. At 2 o'clock the rate of melting was 

 0-97 cm." peir second, at 3 o'clock 0-70, at 4 p.m. 0-60, at 5 p.m. 0-53, at 6 i>.m. O'-50,. 

 at 7 p.m. 0-50, and at 8 o'clock 0-50 cm." per second. Fig 30 shows the changes which 

 took place in the shape of the ice during the process of melting. The depression A on 

 the front of the block was caused by the warm water pressing against the ice at this 

 point, and giving off its heat there, whereafter the water now cooled sinks down past 

 B, where the ice melts at a slower rate, owing to the lower temperatui-e of the water 

 coming in contact with it there. 



In order to measure the movement of the water in the tank, a solution of 

 fuchsin in alcohol was prepared, in such proportion as to get a specific gravity as 

 nearly as possible equal to that of the water. A portion of this solution was dis- 

 charged into the water through a capillary glass tube at a point some 20 cm. in front 

 of the depression A in the ice (fig. 30). The solution was carried at an even rate of 

 speed into the depression, then sinking downwards, following the outline of the ice. 

 At B, the rate of movement was diminished, and a small cloud of fuchsin was formed 

 off the projecting point there. After this, the fuchsin particles were gradually driven 

 in under the ice. The lower portions of the block was passed by at a great rate of 

 speed, until the fuchsin reached the point C, where it slowly sank, and was caught 

 by the bottom current which carries the water away from the ice. 



