THE TRANSFORMATION OF ELEMENTS 233 



tube j to remove hydrogen and any other gas from the electrodes. 

 g is then closed. The stopcock c is opened and the gas 

 allowed to enter the apparatus ; to prevent leakage mercury 

 is raised above the stopcock, which is then closed, e is cooled 

 in liquid air, and the charcoal adsorbs any nitrogen and argon 

 (also krypton and xenon) present. The tube h is surrounded by 

 a paper cup, moistened with water; on filling this with liquid 

 air a vessel of ice is produced, in which the liquid air evaporates 

 comparatively slowly. The object of this precaution is to 

 prevent loss of helium and neon through their contraction 

 to small volume in the cold region e. d is then closed and 

 mercury forced up through c to the point 1. The spectrum can 

 then be examined, and, if desirable, the spectrum tube sealed off. 



The stopcock g is next opened, the charcoal again heated to 

 230 , and the gas pumped off and collected in a test-tube over 

 mercury. Oxygen is added to it, and sparks from an induction 

 coil are passed through the mixture for some hours ; a small 

 quantity of potassium hydroxide absorbs all the oxides of 

 nitrogen formed. The residue, oxygen and inert gases (minus 

 helium and neon) are then admitted to an apparatus similar to 

 that just described, but without the charcoal bulb, containing 

 instead phosphorus in a bulb k to absorb the oxygen. After 

 this has taken place, mercury forces any gas that remains into 

 the small spectrum tube, and its spectrum can be examined. 



The three experiments of which mention has been made gave 

 the following results. From the " untreated " copper nitrate 

 no gas could be pumped off, as of course was expected. The 

 " treated " water gave hydrogen, oxygen, traces of the oxides of 

 carbon, and a small quantity of nitrogen. The experiment with 

 copper nitrate gave a small quantity of nitric oxide (observed by 

 the formation of red fumes during admixture of the portions of 

 gas pumped off before each addition of emanation). There were 

 also present hydrogen, oxygen, and small quantities of carbon 

 dioxide and nitrogen. 



The inert gas from the experiment with pure water consisted 

 almost entirely of neon. Initially the spectrum tube gave the 

 blazing red colour of neon, tinged slightly heliotrope owing to 

 the presence of a small quantity of helium. Only the red, 

 yellow, and green lines of helium were visible ; the blue and 

 violet lines were not observed. 



In the gases from the copper solution no neon or helium 



