SPECIFIC HEAT OF AMMONIA. 387 



the vahe, and then closed. The Hne was pumped down to a pressure 

 of about 2 mm. of mercury, and then ammonia was admitted to the 

 system and allowed to stand for about twenty minutes. The line 

 was again pumped down to about 2 mm. of mercury and the glass 

 tube leading to the pump sealed off. A thermos bottle had been 

 previously filled with solid carbon dioxide and the ammonia container 

 was now immersed in this CO2 snow and allowed to cool for fifteen 

 minutes. The ammonia valve was again opened and ammonia 

 admitted to the system until no more flowed. The pipe line was then 

 unscrewed and the filling plug screwed in place, there being no appreci- 

 able escape of ammonia from the container during this operation. 



After the container had warmed up to room temperature it was 

 weighed, and the ammonia allowed to escape until the mass of 

 ammonia in the container had been reduced to slightly over 100 grams. 

 The exact mass of ammonia has been given above. 



It was desired to use as large a mass of ammonia as possible in the 

 container but to so load it that all the vapor would not be condensed 

 at the highest temperature contemplated for these measurements. 

 Using Keyes and Brownlee's data on the specific volume of the liquid, 

 it was determined that with a filling of 100 grams the vapor would be 

 condensed at a temperature of about 123° C. which put an upper limit 

 on the measurements, for that particular filling. It was originally 

 intended to let out about fifty grams of ammonia, after these measure- 

 ments had been made up to 123° C. and go on up to the critical point, 

 132.9° C. but time was not available to do this. 



11. Data on Calorimetric Lag. 



In Part IV, Section 4, the various calorimetric lags are discussed. 

 With the assumptions and methods there outlined, the following data 

 was obtained from the observed calorimeter curves. 



The length of time taken for the mercury in the calorimeter to 

 equalize its differences of temperature after the heater was switched 

 off, was, on the average, about 12 sees., varying from 10 to 14 sees. 

 The rise of temperature of the mercury after the heater was switched 

 off was, on the a\erage, about 0.00030 ohms on the observing thermom- 

 eter, or about 0°.0044 C. The time taken to equalize the differences 

 of temperature in the entire calorimeter after the heater was switched 

 off, was different for the ammonia measurements and for the calibra- 

 tion measurements, averaging 1 1 minutes for the former, and 15 

 minutes for the latter. In the case of the ammonia measurements, 

 the temperature of the mercury at the instant the heater was switched 

 off, was, on the average, about 0°.046 C. hotter than the equalized 



