( 587 ) 



make the corresponding isothermals of tlie different diagrams coincide 

 by shifting them only in the direction of the log v-axis. 



I found tliat it was not possible to place the diagrams in such 

 a way over each other that the isothermals of the whole system 

 coincided ; I, however, succeeded for the large volumes. According to 

 Kamerlinoh Onnes this signifies that in the mixture the coel'ficients 

 A,B,C,D,E,F of his series do not hang together with the coefficients of 

 the pure substance at corresponding temperatures in such a manner as 

 would follow from the law of corre8[)onding states, but that for the 

 volumes where the terms with D, E, F in comparison with the others 

 are still so small that we ma_y neglect the differences between these 

 terms and those given by coefficients which would obey the law 

 of corresponding states, the diagi-ams will coincide over such a large 

 range of temperature that B and (' may still be regarded as corre- 

 sponding functions of temperature, hence a critical temperature and 

 a critical pressure holding for this range of volume and temperature 

 may be deduced from the law of cori-esponding states. 



On the plate the diagrams are placed over each ' other so as 



to give the best agreement for the larger volumes. In the hori- 



pv 

 zontal direction loij v is drawn, in the vertical hg — . The complete 



lines refer to carbon dioxide, the dash lines to the lirst mixture, 

 the dash-dots lines refer to the second mixture. The points for 

 carbon dioxide are enclosed in circles, those for the iirst mixture in 

 triangles, for the second in squares. The different isothermals are 

 indicated by the letters a, h, etc., with which they are communicated 

 in IV § 2 and V §§ 2 and 3. The points at the beginning and the 

 end of the condensation are marked by h.c. and e.c. ; they are con- 

 nected by curves which thus sej^arate the homogeneous area from 

 the area where separation occurs. Finally the plaitpoints and the 

 points of contact are indicated by P and R, K is the point adopted 

 according to IV ^ 5 for the critical point of carbon dioxide. 



It appears that the agreement for volumes larger than the critical 

 is very satisfactory. In the smaller volumes, howe\'er, considerable 

 systematic deviations from the law of corresponding states occur. 

 We ha^'e already used these diagrams to determine the critical 

 temperature, pressure and volume in the sense as explained above 

 for the different mixtures. Although these results are satisfactory 

 they will not be given here, as more accurate results may be expected 

 from the same operations with diagrams projected on double the 

 scale for the part concei-ned. 



If on a j>7^-diagram we draw the point of contact- and the plait- 



