404 Dr. W. A. Tilden. Influence of Temperature [June 18, 



dnced into the observed result. The glass bulb is then broken, and 

 the paddle gently worked by one person whilst another reads the 

 thermometer by the telescope. In every case a definite interval 

 (usually one or two minutes) is allowed to elapse before the final 

 reading is taken, not only to allow time for solution, but for the 

 alteration of temperature in the liquid to be imparted to the calori- 

 meter. In those cases in which the calorimeter had been in the 

 bath for some time at a rather high temperature, and there was 

 reason to fear that slight loss had occurred by evaporation, the 

 whole was weighed after the experiment was concluded, and a cor- 

 rection introdaced accordingly. The amount thus lost never exceeded 

 a gram or thereabouts, and as the thermal equivalent of the calori- 

 meter and its contents amounted to 100 grams and upwards, the 

 result is not appreciably affected. 



In the accompanying tables the numbers contained in the column 

 headed " Water equivalent of the calorimeter," include the values of 

 the calorimeter, the glass bulb, the thermometer, and the solution, the 

 specific heat of which is taken generally from Thomsen's numbers. 



Three series of experiments were undertaken with sulphate of 

 sodium, and the results are embodied in the three tables, A, B, and C. 



For A the salt had been heated to redness, not to fusion, about a 

 fortnight before. 



For B the salt had been heated to redness one day before. 



For C the precipitated salt had been dried at 120°. 



These varieties were taken on account of the differences in the heat 

 of dissolution of sulphate of sodium which have been observed by 

 Mr. Pickering (loc. cit.). 



