[barnes-hayward- EXPANSIVE FORCE OF ICE 33 



MCLEOD] 



in the arc of a circle which was divided into one-hundredths of a foot. 

 The arm was 16 inches long and at a distance of 1 7/8 inches from the 

 pivoted end was attached one end of a pine rod 1 inch square; the 

 opposite end of the pine rod being attached to the opposite end of 

 the ice column. As the column of ice varied in length, the pine rod 

 remained at nearly a constant length. The temperature of the ice 

 was found by reading a thermometer placed in the hole near the middle 

 of the column. The true expansion of the ice between any two tem- 

 peratures equaled the difference in scale reading at these two tempera- 

 tures XI 7/8 -T- 16, plus the linear coefficient of expansion of the pine 

 rod X 100 X 14 1/8 4- 16. The coefficient of expansion of the ice 

 equals the total expansion divided by the variation in temperature, 

 multiplied by 100. In order to ascertain the coefficient of expansion 

 of the ice per degree Fahrenheit, the total expansion between any 

 two temperatures was divided by the total change in temperature 

 assuming that the variation of expansion was uniform, which is not 

 strictly true, as we find that the coefficient of expansion varies at 

 different temperatures. The coefficient at the higher temperatures 

 is much greater than at the lower temperatures. In computing the 

 length of the pine rod, its temperature was taken as that of the air. 

 The thermometers were only calibrated for even degrees, so that small 

 variations in temperature gave but little information, six to eight 

 degrees being the smallest change in temperature which did give 

 consistent results. 



During the first part of the experiments, Mr. Sawyer found that 

 the ice would expand during the warm hours of the day, and contract 

 during the cooler hours of the night, and that cracks were formed in 

 the column which did not close during expansion due to rising tem- 

 perature. To counteract this tendency to form cracks, weights were 

 attached to the end of the ice column, and this device worked per- 

 fectly, the column remaining intact during the subsequent obser- 

 vations. 



Sec. Ill, 1914—3 



