204 Prof. K. Olszewski on the Critical and 



0'025 millim., and which was not insulated, and in conse- 

 quence was able instantly to assume the temperature of the 

 surrounding gas. The wire was wound in a spiral on a very 

 delicate frame, made of ebonite or of very thin mica sheets, 

 in such a manner that each single turn of the wire was 1 or 

 0*5 millim. distant from its neighbours and did not touch 

 them anywhere. Nor was the wire coil enclosed in a copper 

 tube, but it was in direct connexion with the surrounding 

 hydrogen, and thus the rapidity of registration of the ther- 

 mometer became considerably increased thereby. 



The construction of the apparatus used to determine the 

 critical temperature of hydrogen was on the whole the same 

 as was used to determine the critical pressure of this gas, as 

 described in the Philosophical Magazine (I. c). I will there- 

 fore describe only those parts of the apparatus which were 

 altered for these experiments. 



So far as it was required to determine the critical pressure 

 of hydrogen by means of an expansion method, we could not 

 help using a glass vessel in which the cooled hydrogen was 

 submitted to expansion from a high initial pressure ; for 

 these experiments consisted in the observation of the pressure 

 at the moment of the ebullition of the hydrogen, which could 

 only be done in a transparent vessel. But after the critical 

 pressure of hydrogen had been determined, a metal vessel 

 could be substituted for the glass one. This change was 

 advantageous for several reasons. It was possible to perforin 

 the experiments very quietly without any fear of an explosion, 

 and without using masks or taking other precautions. The 

 use of a metal vessel instead of a glass one allowed me to 

 employ larger dimensions and to produce expansions from 

 much higher initial pressures, and all this had a very great 

 influence on the efficacy of the expansion and the precision 

 of the measurement of the temperatures at the moment it 

 took place. 



The section of the apparatus serving to determine the 

 temperature of the hydrogen during its expansion is repre- 

 sented in the subjoined diagram (fig. 1). In the steel vessel 

 a, tested for a pressure of 220 atm., is placed a frame, b, of 

 ebonite or mica with the platinum wire wound round it. 

 Fig. 2 represents its vertical projection. One of the ends of 

 the platinum wire is soldered to the tube c passing through 

 the cover of the vessel a, and provided above with a binding- 

 screw d' ; the other is soldered to the insulated copper wire 

 e which passes through the tube c, in which it is closely 

 cemented, and also provided with a binding-screw d. In the 

 course of my experiments I used different frames, of various 



