1887.] for the Change of State from Liquid to Gas. 219 



and wjw c obtained by interpolation from Clansius's tables. From 

 these P, W, and w are calculated and compared with observation. 

 It is from this comparison of w with the volume of the liquid at 

 various temperatures and pressures that it is evident that a, is not 

 constant, for the difference between w calculated and a as observed is 

 by no means so. 



The following table exhibits some of my results : — 





o°c. 



100° C. 



200° C. 



240° C. 



242 -5° C. 



243 • 5° C. 





12-24 



1695 



22434 



46339 











12-24 



1695 



22106 



46172 

 5-43 





49000 







30 -210 



284-5 



19-09 



4-26 



3 56 







29 -046 



284-4 



19-85 



5-83 



4-62 









0-177 



0-389 



1-003 



2 43 



2-99 



3-56 







1-264 



1-390 



1-796 



2-614 



2-925 







<t — w = a 



1-037 



1-001 



0-793 



0-184 



-0-065 





In some of these cases, as for example for W at 0° C, it is very 

 difficult to interpolate accurately into Clausius's tables, and similarly 

 for the values of w near the critical point, and I consequently do not 

 attribute much accuracy to these values. On the whole, however, I 

 think that, considering the enormous range of values to be represented 

 by the formula, it is most remarkably accurate. When we compare 

 the calculated and observed volumes of the liquid, in which case a. is 

 of importance, we find that no constant value for a can make them 

 agree, for a obviously diminishes with increased temperature, and 

 near the critical point the value of a. for the liquid and gaseous 

 states is not the same. All this means of course that Clausius's 

 formula does not apply accurately to the case of alcohol. Clausius 

 has not, as far as I can find, applied his formula to calculate the 

 volumes of liquids, and without doing so the want of constancy in «. 

 would not appreciably affect the result. Messrs. Ramsay and Young 

 have made observations of the volume of the liquid at various 

 temperatures and pressures, and I have compared some of their 

 results with the formula. In this way it can be seen that a. must be 

 made a function of the pressure as well as of the temperature. I have 

 calculated the values of v — a at certain temperatures and pressures, 

 and find at 110° C— 



