388 BABCOCK. 



temperature of the calorimeter at the same instant. In the case of 

 the cahbration measurements, it was 0°.12 C. hotter than the equahzed 

 temperature. 



12. Data on Accuracy of Chronograph. 



The chronograph was controlled by a laboratory pendulum clock 

 which was compared with standard time for several days before the 

 final measurements were made, and was found to be correct to within 

 a few seconds a day. 



The distance between second marks made by the pen on the chrono- 

 graph sheet, were approximately 1 cm., and the record marks could 

 be located to within | mm., giving an error at each end of the time 

 interval of not more than l/20th sec. The time record was made 

 automaticalh' by throwing the heater switch on and off. 



13. Data on Regulation of Constant Temperature Bath. 



The thermostat maintained the bath temperature constant within 

 a range of 0°.02 C. and in most measurements, within a range of 0°.01 C. 

 In Ammonia Measurement 9, the bath got out of control and the 

 temperature shows a variation of 0°.07 C. This variation is in a 



direction to give too high a value of . 



VI. Experimental Results. 



1. Computed and Faired Values of Heat Capacity of Ap- 

 paratus: 



(a) When Filled with Ammonia, (b) When Empty. 



Equations (6) and (18) are the laboratory equations by which 

 the values of 8H and 86 are computed. Equation (13) gives the value 

 of k/2. 



These experimental results are summarized in the following table, 

 whose explanation is as follows: 



Column 1. Number of the measurement, for reference to the tables 



from which the values were computed. 

 Column 2. Temperature of the measurement, in °C., taken as the 



temperature equivalent to R^ for that measurement. 



Column 3. Experimental values of and — -, in Joules /°C., 



86 86 



computed by equations (6) and (18) from the tabulated 



data. 



