446 SCIENCE PROGRESS 



The point of greatest importance is — can the formula be 

 extrapolated with any confidence over more than a few degrees 

 below ioo° and above 2oo D ? The critical temperature of benzene 

 from the above equation is 



6 C = ioo -f a 287*3, 



■00534 



and the value as determinedfrom direct experiment is 288 0- 5 — so 

 that the agreement is very close. Extrapolating in the other 

 direction, we have for the surface-tension of benzene at o°, 



T = 18 [1 + '00534 x ioo] 1 '" 3 = 30*3 dynes per cm. 



Unfortunately T is not given in the above table. The experi- 

 ments of Renard and Guye, extrapolated from temperatures in 

 the neighbourhood of 20 by means of the linear relation, give 

 T =3O"i,and those of Ramsay and Aston 1 similarly treated give 

 T =307. 



Again, in the case of ethyl acetate, taking 20 C. as an arbitrary 

 zero, we find from the observed values between ioo° and 180°, 



T = 23-6 [1 - -00423(0 - 20)]-% 



giving 256° as the critical temperature. The directly observed 

 value is 250 . 



In the case of ether, calculating the constants from values 

 given between o° and 130 , we find that 



T= 18-9(1 --005130)"* 



giving 195° as the critical temperature, whilst the observed 

 value is 194 — 197 . 



It seems, therefore, that this formula in the cases examined 

 gives the critical temperature very closely indeed even when 

 extrapolated over a range approaching ioo°. If this should 

 prove to be generally true, it would be possible to calculate the 

 critical temperature of a liquid with some closeness from 

 capillary observations alone, and the problem of obtaining the 

 "corresponding temperatures" at which to make comparisons 

 would be greatly simplified. 



One word in conclusion. In the comparison and collation 



1 It should be noted that the results of Ramsay and Shields and of Ramsay 

 and Aston are for benzene in contact with its own vapour, those of Renard and 

 Guye for benzene in contact with air. 



