( 275 ) 



increase of pressure was slight near tlic temperature of (be lowei' 

 critical eud-poiut, we could nevertheless not conclude that the con- 

 centration of the latter must lie in the immediate neighbourhood, 

 both because we knew nothing about the shape of the loops 2.3, 

 and because we thought that the liquid-liquid plaitpoint would, no 

 doubt, move to the right with increase of tempei-ature, and we 

 found it still lying on the left side at our highest temperature of 

 nnmixing. 



Hence that our following filling has an ^' which is still considerably 

 smallei-, viz. .v = 0,090. Now however, we were undoubtedly on the 

 left hand of the lower end-point, and also of all the plaitpoints. It 

 appeared from the increase of the pressure with the condensatioJi 

 at which phase 3 disappeared now, that we were now again further 

 from this end-p'jint tlian for cr=:0,121. 



In order to find the accurate concentration we have a few times 

 allowed some mg. of carbonic acid to escape from our tube, and 

 then only examined if we were still on the left side of the lower 

 end-point. In this way we have closely approached it in three steps, 

 and we think we have got the right one in our mixture .i' = 0,114. 



If we then compute the displacement of the point 2 with the 

 temperatuie, we find that it takes plsice very regularly. In the 

 temperature range of ±10° between the critical end-points it is 



displaced from x = 0,008 to .v = 0,1 14, w^hich vields ( —^ | = — — = 



a little more than 0,01, which we also found over the temperature 

 range of 39°.l — 39°. 8. It ap[)eared from the further observation of 

 these mixtures that the liquid-liquid plaitpoint continues to lie about 

 at constant .x^). We could raise our temperature to a little more than 

 46°, and we always observed plaitpoint phenomena on compression. 

 At temperatures above 35°, however, the meniscus disappeared 

 somewdiat lower down than halfway the total volume; so this pointed 

 to retrograde condensation of the phase 3, accortlingly a slight dis- 

 placement of the plaitpoint to (he lefthand side took place. 



If we now suppose the liquid-liquid plait to be about symmetrical, 

 point 3 must move just as far to the right as 2 towaids the left, 

 and we should find for the extreme poijit of the region of non 

 miscibility a concentration .v = db 0,22. We may finally remark that 

 we have tried as much as possible to observe the different mixtures 

 at the same temperatures. As the three-phase pressures rise pretty 



1) So this is the same thing as Timmermans found for llie mixtures examined 

 by him. These Proc. XIII p. 5U7. 



