1896.] on New Besearches on Liquid Air. 145 



no liquid jet could be seen. If the hydrogen contained a few per 

 cent, of oxygen the gas jet was visible, and the liquid collected, which 

 was chiefly oxygen, contained hydrogen in solution, the gas given off 

 for some time being explosive. 



If, however, hydrogen, cooled by a bath of boiling air, is allowed 

 to expand at 200 atmos. over a regenerative coil previously cooled to 

 the same temperature, and similar in construction to that shown in 

 Fig. 8,* a liquid jet can be seen after the circulation has continued 

 for a few minutes, along with a liquid which is in rapid rotation in 

 the lower part of the vacuum vessel. The liquid did not accumulate, 

 owing to its low specific gravity and the rapid current of gas. 

 These difficulties will be overcome by the use of a differently shaped 

 vacuum vessel, and by better isolation. That liquid hydrogen can be 

 collected and manipulated in vacuum vessels of proper construc- 

 tion cannot be doubted. The liquid jet can be used in the meantime 

 (until special apj)aratus is completed for its collection) as a cooling 

 agent, like the spray of liquid air obtained under similar circum- 

 stances, and this being practicable, the only difficulty is one of 

 expense. In order to test, in the first instance, what the hydrogen 

 jet could do in the production of lower temperatures, liquid air and 

 oxygen were placed in the lower part of the vacuum tube just 

 covering the jet. The result was that in a few minutes about 

 50 c.c. of the respective liquids were transformed into hard white 

 solids resembling avalanche snow, quite different in appearance from 

 the jelly-like mass of solid air got by the use of the air pump. 

 The solid oxygen had a pale, bluish colour, showing by reflection 

 all the absorption bands of the liquid. The temperatures reached, 

 and other matters, will be dealt with in a separate communication. 

 When the hydrogen jet was produced under the surface of liquid air, 

 the upper part of the fluid seemed to become specifically lighter, as 

 a well marked line of separation could be seen travelling downwards. 

 This appearance is no doubt due in part to the greater volatility of 

 the nitrogen and the considerable diiference in density between liquid 

 oxygen and nitrogen. In a short time solid pieces of air floated about, 

 and the liquid subsequently falling below the level of the jet, hydrogen 

 now issued into a gaseous atmosphere containing air, which froze solid 

 all round the jet. There is no reason why a spray of liquid hydrogen 

 at its boiling point in an open vacuum vessel should not be used as a 

 cooling agent, in order to study the properties of matter at some 20*^ 

 or 30° above the absolute zero. 



Fluorine. — This is the only widely distributed element that has 

 not been liquefied. Some years ago Wallach and Hensler pointed out 

 that an examination of the boiling points of substituted halogen organic 

 compounds led to the conclusion that, although the atomic weight 

 of fluorine is nineteen times that of hydrogen, yet it must in the 



* In the figure, A represents one of the hydros^en cylinders; B and C, 

 vacuum vessels containing carbonic acid under exhaustion and liquid air re- 

 spectively; D, regenerating coil; G, pin-hole nozzle ; F, valve. 



Vol. XV. (No. 90.) l 



