( 177 ) 



in q, u\,f and }v,f would still have had the same sign as for p, but 

 it will never occur that w,f has such a large positive value there 

 that in conseiiucnce of this the numerator will become negative. 



So the second part of the melting- 

 l)oint line must always have a shape 

 as is schematically represented in tig. 3. 

 This second branch of the melting- 

 point line, which has been determined 

 for the system etker-anthraquinone, could 

 not be investigated for the system HjO — 

 NasSO discussed here for want of 

 suitable material. 



I hope to discuss the course of the 

 melting-point lines under constant pres- 

 sure in a following communication. 



I 



Amsterdam, May, 1911, 



Anorg. Chem. Laboratory 

 of the University. 



Chemistry. - — "On retrogressive vapour-lines \ (First communication). 

 By Prof A. Smits. (Communicated by Prof. J. D. van der Waat.s). 



In connection with the discussion of the 7!«;-projection of the liquid 

 line of the three-phase region of a system that belongs to the type 

 ether-anthraquinone we shall now proceed to the discussion of the 

 vapour line corresponding to it. 



We may arrive at the equation of this vapour line by combination 

 of the following two differential equations: 



Vs,, dp z=z {.r,s - Xg) I -j~^ ] d.Vg -\- 7,scj dT . . . . (I) 



and 



ff/PT 



H 



(2) 



Vl^ dp r= {.I' I A-,,) — — (IVrj + 7^/,/ dT . . . 



■ \0.v-gJpT' 



in which we come to the following expression for the said vapoui 

 line : 



rp 'I 



dT 



[il ) 



{Vs — Xg) Vlg {Xl — Xg) Vsf 



(3) 



PT 



if we, namely, write Vgf and lOsf instead of Vsg and iv 



^9 



