( 537 ) 



deviate from tlie statically determined ones. A full discussion and 

 explanation of this behaviour is found in van Rossem's thesis for the 

 doctorate '). In the evaporation of solid substance we have in this 

 case> a superposition of two velocities, the velocity of evaporation of 

 the molecule as such, and the velocity of dissociation in the vapour. 

 If (he first velocil}' is small with respect to the second, stable eciuili- 

 brium will exist at any moment between the solid substance and the 

 vapour surrounding it, and the same values will be found according 

 to the two methods of determination. If, however, the first velocity 

 is great with regard to the second, the solid substance will be sur- 

 rounded by a vapour which is little dissociated, at least less than 

 is the case for the stable coexistence. Then the substance can be 

 heated to a higher temperature than in the former case. Hence the 

 dynamically determined P-T'-Iine, whose situation will depend on the 

 rapidity of the heating, will lie at higher temperature, resp. at lower 

 pressure, than the statical one. 



A similar deviation between P- 7 -lines determined statically and 

 dynamically, is however not known for liquids. We disregard here 

 the generally very small differences which may occur in consequence 

 of a certain excess of pressure which the boiling (or sublimating) 

 substance must exert to emit its vapour to the surrounding space, 

 which must cause the substance to assume a somewhat higher tempe- 

 rature corresponding to the somewhat higher pressure in the immediate 

 neighbourhood. The difierences described above are of an entirely 

 different order of magnitude. Nitrogen tetroxide, which Ramsay and 

 Young also included in their above described investigation, gives 

 concordant results in the determinations, made according to the two 

 different methods; this conclusion is based on the observations col- 

 lected in the following table, (p. 538). 



The statical values given in the above table were obtained by 

 making use of the compressibility of a large glass reservoir, filled 

 with mercury and provided with a narrow capillary ; the displace- 

 ment of the mercury meniscus in the capillary tube furnishes a 

 measure for the pressure exerted by the nitrogentetroxide against 

 the reservoir. The two observations marked with an asterisk, have 

 not been measured directly, but have been found by determining 

 the difference of pressure between 0° and the temperatures 13,9 and 

 19,6, in which it was assumed that the pressure determined stati- 

 cally and dynamically would agree at 0°. A difieren.ce which might 

 eventually exist between the values at 0° obtained by the two 



1) Van Rossem. Thesis for the doctorate and Z. ph. Ch. 62, 257 (1908). 

 See also Z. ph. Ch. 72. 458 (1910). 



