OF ARTS AND SCIENCES. 



125 



3d. The top of the calorimeter not being in contact with the water, 

 its temperature may be uncertain. To eliminate this, the calorimeter 

 was often at the temperature of the air to commence with. Also the 

 water was sometimes violently agitated just before taking the final 

 reading, previous to letting in the cold water. Even if the tempera- 

 ture of this part was taken as that of the air, the error would scarcely 

 ever be of sufficient importance to vitiate the conclusions. 



4th. The specific heat of copper changes with the temperature. 

 Unimportant. 



5th. Some water might remain in the spout whose temperature 

 might be different from the rest. This was guarded against. 



6th. Evaporation. Impossible, as the calorimeter was closed. 



7th. The introduction of cold water may cause dew to be depos- 

 ited on the calorimeter. The experiments were rejected where this 

 occurred. 



The corrections for the protruding thermometer stem, for radiation, 

 &c, were made as usual, the radiation being estimated by a series of 

 observations before and after the experiment, as is usual in determin- 

 ing the specific heat of solids. 



296.8 = 26°.597 

 Correction for stem -|~ .019 



Initial temp, of calorimeter 26°. 61 6 



Water 1654.7 



Capacity of calorim. 35.4 



" thermom. 1.1 



Total capacity 1691.2 



