1903.] Specific Heats, especially at Low Temperatures. 

 Table VI. Constant Immersion of Five Minutes. 



189 



individual experiments are given. The material in the case of 

 Table V is pure copper; the three objects described were cut from 

 the same casting. In the case of Table VI the material is commercial 

 sheet lead ; all the objects described were made from the same sample. 

 The numbers in Table V might at first sight be supposed to 

 indicate that the process of formation of the ice coating is a very slow 

 one, and is not complete in less than half-an-hour. The author 

 was led to reject this hypothesis after weighing the object with its 

 ice coating during immersion. It will be sufficient to describe a single 

 experiment on the copper sphere of 62J grammes. The first observa- 

 tion was made two minutes after plunging the sphere in the calori- 

 meter. The apparent weight rapidly decreased at first, indicating 

 additional formation of ice. After a very short time the apparent 

 weight began to increase, and continued to increase during the 

 remainder of the hour during which the experiment was continued. 

 The observed increase in apparent weight divided itself into two 

 parts : (1) There was a very slow and nearly uniform increase due to 

 the gradual melting of the coating of ice ; (2) there was an occasional 

 relatively large increase due to the escape of a bubble of air.* From 

 this and similar experiments the author concluded that the formation 

 of ice might, in the case of the objects used by him, be considered 

 as being practically complete in 5 minutes or less. The rate of 

 melting in the experiment quoted was 3 per cent, of the total mass 

 per hour ; the calorimeter, however, was of a less efficient form than 

 that finally adopted. 



* The presence of the enclosed air made it useless to attempt to deduce the 

 weight of ice from the apparent weight in water. 



