( öfi ) 



For (liis foniiiila can very easily — as far as its foi-m is eoiiccnied 

 — be derived IVoiii the equal ion of state combined witli Maxwell's 

 theorem. But in the second place that in virtue of considei-ations on 

 'db\ fdrh 



the vahie of ( | and — ) on p. 1221 — 1229 loc. cil. he ai-rives 

 at the empii-ical foi-mula (in the neighbourhood of the critical point) : 



p. 1227, where then n = 47» is found. 



Now in virtue of considerations — which are in close connection 

 with the theory of association, developed by me in connection with 

 the solid state in the six preceding papers — 1 think we have to 

 arrive at the i-esull, that the dependence of the quantity h on /; in 

 the neighbourhood of the critical point is represented better by the 

 relation : 



;;=-(^-j '" 



and this led by the following theoretical considerations. 



In order to arrive at the form of the function />=/(r,7') in the 

 equation of state (y^ + ^tOO' — h)^lVl\ we can, namely, follow 

 two difTerent courses. 



The first conrse, which is generally followed, is this that the 

 problem is considered from a purely ^Y/ic^/c point of view. According 

 to the method of Maxwell, Boltzmann, v. d. Waals, Kokteweg, 

 LoRENTZ, Reinganum, and others the vicissitudes of every molecule 

 separately are followed, the effects of collisions etc. etc. To shorten 

 the calculations we can also make use of the theorem of the Virlal 

 (Clausius). By often laborious calculations w^e arrive iji this way at 

 the formula of appi-oximation 



b = 4bj 1 ^ -h etc. 



^ V 32 y ^ 



the coefficient ^"/s^ of which has afterwards proved to be := Vs- 

 The calculation of the following coefficients becomes practically about 

 infeasible. In this molecular forces are still left entirely out of con- 

 sideration. If we wanted to include them into the considerations, 

 the calculations become still much more complicated, and the tem- 

 perature also appears as influencing factor. (Reinganum). 



So the above formula gives the "apparent" change of b, when 

 the volume decreases. We leave aside here a "real" diminution, 

 fully discussed by van der Waals some years ago. 



