164 Prof. W. Ramsay and Miss E. Aston. [June 14, 



naturally depending on tlie barometric pressure of the day. As none 

 of the esters boiled much above 100 at atmospheric pressure, it was 

 possible by lowering pressure to cause them to boil at 132, especially 

 as in filling the tube a trace of air was purposely left in the liquid. 

 It is not necessary that this air should be visible as a bubble, but it is 

 sufficient if the liquid is not thoroughly boiled in vacuo. It may 

 appear strange that such a course was followed, but repeated expe- 

 rience has shown that if a liquid is wholly deprived of dissolved gas 

 by boiling it in vacuo, it is impossible to cause it to boil, even at 

 atmospheric pressure, although heated to 100 U above its normal 

 boiling point. 



Having determined the capillary rise at 132, the pressure in the 

 jacket was lowered, so as to cause the chlorobenzene to boil at 78 or 

 80, care being taken not to allow the gas present in the upper part of 

 the tube to condense wholly. The rise was again noted. The tube 

 was then jacketed with quinoline vapour at about 185, as well as at 

 higher temperatures, and readings were again taken. Some six or 

 seven points on the curve were thus determined, a sufficient 'number 

 to characterise it. 



It will conduce to clearness to give the essential data at this stage, 

 reserving details of experiment to the Appendix, where they are 

 tabulated. For completeness' sake, the results previously published 

 in the ' Philosophical Transactions ' for methyl formate and for ethyl 

 acetate are here included. 



As the molecular surf ace- energy of a liquid, provided it does not 

 dissociate with rise of temperature, may be calculated by means of the 

 equation 



7 (Mt>) = k(-r-d) 



(where Jc is a constant characteristic of each liquid but varying only 

 slightly from 2'1, ? is the temperature measured from the critical 

 point downwards, and d is a constant), the liquid is sufficiently 

 characterised by giving the values of &, the critical temperature, and 

 d. They are as follows : 



Table I. 



Critical 

 temperature, 

 Ester. C. Jc. d. 



Methyl formate 214'0 2'042 5'9 



Methyl acetate 233'7 2'109 4'5 



Methyl propionate 257'4 2'182 5'3 



Methyl butyrate 281'25 2'220 3'75 



Methyl isobutyrate 267-55 2'248 5'25 



