1212 



\dTjsWJ ^Vsuu {l+r—s)RT 

 in which the volumes of solid and liquid are neglected by the side 

 of f?as, and Boylr's law has been applied to the gas phase. 

 Transformation of (12) finally yields the recjuired lieat Q, 

 dlnP I dlnP' 



dT~ ~" " f* df'^\ 



fd In P\ 

 Q^^-rQr+sQ,-{\+r-s)R[-—~] . . (13) 



\dl-^JSLjG 



As now Q„ (heat of evaporation of one molecule of water) and 



Qo (heat of solution of one molecule H^S) are known, r and s can 



be calculated from \()a and b (see also (M)), if an arbitrarily chosen 



value is substituted for 7i, Q^ can be calculated if the temperatures 



are chosen close to each other, so that the differential quotient in 



(13) can be replaced by the quotient of differences'). Thus every 



time from two observations at temperatures differing little a value 



of Q, is found for that small range of tempeiature. This value Q,, 



therefore, represents the heat of transformation on the three-phase 



line SLJt, corrected for the phenomenon of solution and evaporation 



at the conversion It is the heat belonging to the conversion: 



//, 5 . « //, ^ /y., S+nfI,0 - E, . . . . (6) 



{solid} (fjas) (lii/uid) 



The heat Q^ (and the change of energy ^,) will again be functions 

 of the teujperature. The algebraic sum of the specific heats is greater 

 here than on the three-phase line SSb^t (§ 5) ; it may not be neglected. 

 Hence the heat at transformation on SL./t in the immediate neigh- 

 bourhood of I he (piadruple point must be found by extrapolation. 

 The correction required for this is, however, small enough to allow 

 linear extrapolation, in other words to enable us to consider the 

 specific heats as independent of the temperature. 



Now the value of n follows simply from the eqnaiions (5) and (6) 

 in a way analogous to that in ^ 3. 



{To he continued). 



1) As r and s themselves represent corrections, a change of n brings about a 

 modification in llie correction which is already small. Whether 5 or 6 is chosen 

 for n gives only a slight variation in the result of the calculations. We shall 

 come back to this point later on. 



