1880.] State of Fluids at their Critical Temper atures. 481 



There is another mode of experimenting which illustrates this even 

 more strikingly, A quantity of alcohol with hydrogen over it was 

 raised to a temperature of 228° to 230° under a pressure of 120 atmo- 

 spheres, and the pressure and temperature maintained for five hours, 

 when it was seen that the line of demarcation was in the same place, 

 and as sharp as before, and on lowering the pressure, the liquid boiled, 

 showing that it had not mixed with the gas. The temperature was 

 now raised to 240°, and the pressure as before, at 120 atmospheres, and 

 at the end of a quarter of an hour, instead of a sharp line of demarca- 

 tion there was only a faint, broad indication of a change of density near 

 the middle of the tube, and at the end of an hour all mark of a 

 difference had disappeared, the two gases having mixed. Doubtless, 

 the mixture was not yet uniform, but it had proceeded sufficiently far 

 to show that the two were miscible. Here we see plainly, that a few 

 degrees below the critical point, the substance was undoubtedly liquid, 

 and a few degrees above, undoubtedly gaseous. The experiment was 

 repeated at 200 atmospheres' pressure, the temperature being 220° and 

 240° , with precisely the same result, and I have since repeated it 

 several times. This shows that the limit of the liquid state depends 

 upon temperature, and not upon pressure. 



The method of examining the loss of cohesion of a liquid by means 

 of a capillary tube was also applied at any pressure by the introduction 

 of a permanent gas over the liquid, as in the foregoing experiments. 

 This was done as follows. 



In the working tube of the pressure apparatus A, a small piece of 

 capillary tubing B, was cemented by fusing a minute piece of silicate 

 of soda between it and the side of the tube. The liquid C, which was 

 in this case carbon disulphide, was placed in the tube over the mer- 

 cury, and readings taken of the height of the liquid in the capillary 

 tube. In order that the same liquid should be used in both cases, 

 and the experiments thus rendered strictly comparable, the point of 

 A was broken off, and a quantity of pure dry nitrogen introduced. 

 Pressure was then applied, and the capillarity noted, the temperature 

 raised, and the result again noted. The following table illustrates the 

 results obtained. The capillary heights are arbitrary numbers of the 

 cathetometer : ten experiments were done in each case. 



Liquid alone. Liquid with nitrogen. 



Temp. Pressure. Height. Temp. Pressure. Height. 



51° . . 1 atmos. . . 53*6 . . 51° . . 81 atmos. . . 532 

 220° ..63 „ .. 12-2 .. 220° ..180 „ .. 119 

 Probable error 



of mean 07 „ .. 0*2 11 „ .. 0*25 



The experiments were then pushed up to the critical point when, in 

 both cases, the one at 80 atmospheres and the other at 194, the capil- 



