140 Professor T)ewar [March 27, 



to —79° in the vessel C, a liquid jet is just visible. It is interest- 

 ing to note, in passing, that Pictet could get no liquid oxygen jet 

 below 270 atmos. This was due to his stopcock being massive and 

 outside the refrigerator. If the oxygen is replaced by air, no liquid 

 jet can be seen until the pressure is 180 atmos., but on raising the 

 pressure to 300 atmos. the liquid air collected well from the simple 

 nozzle. If the carbon dioxide is cooled by exhaustion (to about 1 inch 

 pressure) or — 115°, then liquid air can easily be collected in the small 

 vacuum vessel D, or if the air pressure is raised above 200 atmos., 

 keeping the cooling at —79° as before.* The chief difficulty is in 

 collecting the liquid, owing to the rapid current of gas. The amount 

 of liquid in the gas jet is small, and its collection is greatly facilitated 

 by directing the spray on a part of the metallic tube above the little 

 hole, or by increasing the resistance to the escaping gas by placing 

 some few turns of the tube, like B in the figure, in the upper portion 

 of the vacuum tube, or generally by pushing in more tube in any form. 

 A vacuum vessel shaped like an egg-glass also works well. This prac- 

 tically economises the cool gas which is escaping to reduce the tem- 

 perature of the gas before expansion, or, in other words, it is the cold 

 regenerative principle. Coleman pointed out long ago that his air 

 machine could be adapted to deliver air at as low a temperature as 

 has yet been produced in physical research. Both Solvay and Linde 

 have taken patents for the production of liquid air by the application 

 of cold regeneration, but the latter has the credit of having succeeded 

 in constructing an industrial apparatus that is lowered in tempera- 

 ture to —140°, or to the critical point of air, in about 15 hours, 

 and from which liquid air containing 70 per cent, oxygen is collected 

 after that time. 



For better isolation, the pipe can be rolled between two vacuum 

 tubes, the outer one being about 9 inches long and IJ inch diameter, 

 as shown in Fig. 3. The aperture in the metal pipe has a little piece 

 of glass tube over it, which helps the collection of the liquid. With 

 such a simple apparatus, and an air supply at 200 atmos. with no 

 previous cooling, liquid air begins to collect in about five minutes, but 

 the liquid jet can be seen in between two and three minutes. It is 

 not advisable to work below 100 atmos. 



In Fig. 4 the metallic tube in the vacuum vessel is placed in 

 horizontal rings, leaving a central tube to allow the glass tube C to 

 pass, which is used to cool bodies or examine gases under compression. 

 The inner tube can be filled for an inch with liquid air under a 

 pressure of 60 atmos. in about three minutes. Generally, in the 

 experiments, about ^ to 4 cubic feet of air passes through the dif- 

 ferent sized needle holes per minute when the pressure is about 

 200 atmos. As the small hole is apt to get stopped, for general 



* The liquefaction is takin*? place in this condition at 1| times the critical 

 temperature. Hydrogen similarly expanded at the melting point of air 

 (— 2H° G.) behaves exactly in the same way. 



