6 A. W. Witkowski on the 



It is known tbat the determination of the temperature of 

 a gas is attended by some difficulty. Owing to the great 

 diathermancy of these bodies, a thermometer introduced in a 

 gas tends to show rather the temperature of the surrounding 

 bodies than of the gas itself. To get over this difficulty I 

 resolved to apply a mode of temperature-measurement which 

 has proved successful in meteorological observations, and is 

 perhaps best known in its application in Assmann's aspirating 

 thermometer. Tbe thermometer is placed inside a polished 

 metal tube, through which a brisk current of air is made to 

 pass. It has been found that under such conditions the tempe- 

 rature indicated is nearly independent of the temperature of the 

 tube. After several trials I constructed upon this principle 

 the connexion between the heater and ihe calorimeter ; it will 

 be described fully in § 7. The only thermometric apparatus 

 which could be applied under the given conditions was the 

 thermo-electric couple. The mode of experimenting was 

 such that, instead of measuring the initial temperature of the 

 gas, I determined with the aid of the couple only the small 

 loss of temperature experienced by the gas during its passage 

 from the heater into the calorimeter. For this purpose one 

 of the solderings of the couple was placed in the gas-stream, 

 just inside the calorimeter, the other in the heater, or in 

 a separate thermostat of known temperature. This second 

 soldering was always in contact with the bulb of a hydrogen 

 thermometer, in order to reduce finally the observed tempe- 

 ratures to a definite scale. 



§ 6. Another modification concerned the determination of 

 the mass of air used in every calorimetric experiment. 

 Usually the quantity of gas has been determined by previous 

 experimenters in an indirect way. E. Wiedemann, following 

 Delaroche and Berard, measured the volume of water which 

 replaced the gas, contained initially in a bladder or india- 

 rubber bag. Regnault used a spacious receiver filled with 

 compressed gas, and observed the fall of pressure occasioned 

 by the using up of gas in every experiment. The volume of 

 the receiver and its temperature being known, the mass could 

 be calculated on the basis of the law of compressibility, which 

 Regnault determined himself expressly for that purpose. I 

 have found it more convenient and accurate to measure the 

 mass of air by direct weighing. Accordingly a large 

 reservoir (about ten litres capacity) w r as filled with pure 

 compressed air at a pressure of 80-100 atmospheres : this 

 was ample enough to provide the gas for some dozen of 

 calorimetric determinations. From this store I filled small 

 metal flasks (capacities 136, 220, 208 cubic centim.), as 



