328 Dr. Ramsay. On the Critical State of Gases. [Apr. 22, 



base of its capillary neck. It is improbable that the liquids ex- 

 perimented on dissolve air, for when heated, no air-babbles can be 

 seen to escape. 



2. The liquids may not have been quite pure ; they may not have 

 been absolutely free from water. 



The methyl formate used in the experiments presents no special 

 advantage but that of its meniscus disappearing at a comparatively 

 low temperature t It was procured from Kahlbaum, and previous to 

 use had been distilled over metallic sodium. Its boiling-point was 

 absolutely constant. Still, as treatment with sodium might have led 

 to the formation of some other compound of nearly the same boiling- 

 point, it was redistilled, in accordance with a suggestion of Professor 

 Stokes, and the first and last portions of the distillate were selected 

 for experiment. Two tubes were filled to as nearly the same height 

 as could be judged by eye, one with the first portion of the distillate 

 and one with the last. Both were heated at the same time in the 

 copper heater, and were exactly similarly placed with regard to the 

 thermometer. The temperature of the heater having been raised till 

 the meniscus had disappeared in both tubes, the source of heat was 

 removed and the heater was allowed to cool. The tube containing 

 the first fraction gave the first symptoms of a cloud, or mist, at 

 218*4 (corr.), and that containing the last fraction at 218'1 (corr.). 

 The very slight difference is doubtless owing to the two tubes not 

 containing exactly the same amount of liquid. 



To eliminate any source of error dependent on the use of methyl 

 formate, two other substances were employed, viz., carbon disulphide, 

 CS 2 , and carbon tetrachloride, CC1 4 . The former of these liquids was 

 rectified five times over sodium, and then boiled at 48' 7 (corr.). The 

 latter was rectified four times over phosphoric anhydride, and boiled 

 constantly at 77*5 (corr.). 



They yielded the following results : — 



Tube more Tube less 



than half full. , than half full. 



CS 2 2827 286-4 



CC1 4 283-3 288-4 



These readings are given for the first appearance of a cloud in the 

 tube on cooling, and differ from the temperature at which the meniscus 

 disappears by being about half a degree lower. They also do not 

 represent extreme instances, for in the first cases the liquids do not 

 entirely fill the tube, and in the second about half an inch of liquids 

 remains in the tubes before it becomes impossible to distinguish liquid 

 from gas. 



3. Owing to too rapid rise of temperature, the tubes and the heater 

 may not have been at the same temperature. 



