140 



which, considering the uncertainties in the values of b]^ and i^ajc 

 calculated from the halogen compounds, is alreadj' very fair agree- 

 ment. With \/ah=\0,S we should have found 306 = 305. 



With x=0,i we should find 393 = 305 with Ka)r. = 9,7 + 3,6 = 13,3, 

 which therefore does not agree at all. And for still higher values 

 of X the difference would have become greater and greater. 



The least dissociation of Hg, to Hgi for Tk is accordingly quite 

 excluded, so that we must assume that the mercury is perfectly 

 bimolecular tliere. The liquid mercury is a fortiori bimolecular, the 

 saturated mercury vapour on the other hand will be Hgi only at 

 lincer temperatures (where the large volume prevails); at higher 

 temperatures, however, in consequence of the decreasing volume, it 

 will become Hg^ in ever increasing degree (cf. also loc. cit. p. 3 — 4). 



3. With the somewhat lower value for )., viz. A := 0,936, and 

 with èi- = 150.10-^ |/ra = 10,8. 10-2 we now find: 



71- = 10,10 X 116,6 = 1178° abs. = 905° C. 



It is therefore seen that at 880° Cailletkt c. s. have been exceed- 

 ingly near the critical temperature of mercury, viz. ± 900° C. ; 

 it lies possibly scarcely %f higher than the highest temperature at 

 which they have still carried out a vapour pressure determination 

 (880° C). ' 



Would it not be desirable in virtue of this fact to determine the 

 critical temperature of mercury at last experimentally? As it was 

 possible to Rotinjanz to determine those of the mercury halides, 

 which lie resp. at 976°, 1011° and 1072° abs*, it may also be 

 possible to come to ± 1200° abs. (at the melting point of silver, 

 viz. 960° C, we are already far above the critical temperature of 

 mercury). 



The only scientist that calculated a fairly plausible value for T]c 

 for Hg, was Happel (Ann. der Ph. (4j 13, 351 (1904)), who gave 

 1370° abs., i. e. only 100° higher than was calculated by me in 

 1916 (1260° abs.). But the values of Königsberger (1912), viz. 

 1270° C. (1543 abs.) and of Bender (1915), viz. > 1500° C. (> 1773° 

 abs.) are very certainly too high. As /7\- not far below 7^ amounts 

 only to =r 3080 (this follows with absolute certainty from the 

 vapour pressure observations, see above), the value of the factor ƒ 

 would at high temperatures be only =^ 2, or even 1,7, instead of 

 approaching 2,8 (see below), whereas this factor is already about 

 2,7 or 2,6 between 0' and 100° C. — still apart from the fact that 



