

isjlf).] On the Electrolysis of Gases. i' l.~> 



applicable when the discharges through the constituent gases of the 

 compound are of distinct and different colours; this is eminently 

 the ca.se with the hydrochloric acid gas, as the discharge through 

 hydrogen in a capillary tube is red, through chlorine green. 



Take a capillary tube of very fine bore, the finer the better (the 

 tube I used was thermometer tubing of the finest bore I could 

 procure), and insert platinum wires for electrodes in two small 

 bulbs blown on the ends of the tube ; then fill the tube with HC1 

 gas, allowing it to run through the tube for a considerable time so 

 as to get rid of any extraneous gas, and exhaust the tube so that 

 the gas in it is at a very low pressure. Then when the discharge 

 from a large induction coil passes through the tube, the following 

 phenomena are observed. When first the discharge passes through 

 the tube the colour is uniform throughout and of a greenish-grey ; 

 after the discharge has been passing for a little time the end of tho 

 tube next the cathode gets distinctly red, whilst that next the anode 

 gets green ; this difference in the colour at the ends of the tube 

 goes on increasing until the tube presents a most striking appear- 

 ance, the part near the cathode being bright red, while that near 

 the anode is a' bright green. The difference in colour attains ;i 

 maximum value, and if the discharge is allowed to run for several 

 hours the contrast between the two ends disappears to a very great 

 extent ; the discharge throughout the whole of the tube being pinkish 

 and apparently passing mainly through hydrogen. This is doubtless 

 due to the diffusion through the tube of the hydrogen which in the 

 earlier stages of the discharge had accumulated about the cathode ; 

 one advantage of using very narrow tubes is that with them this 

 diffusion is slow. When the tube is in this condition the colour of 

 the discharge sometimes changes suddenly, and for a second or two is 

 green instead of pink, showing that though in the main the discharge 

 passes through hydrogen, it occasionally leaves the hydrogen and 

 passes through the chlorine. This transference of the discharge from 

 one constituent to another of a mixture of gases is not infrequently 

 observed when the gases are mixed in certain proportions. 



Some of these capillary tubes showed after the discharge had been 

 passing through them for some time a peculiar patchy appearance, some 

 portions of the tube being a much brighter red than the others, while 

 other portions were green. In some tubes this occurred to such an 

 extent that the discharge showed an irregularly striated appearance. 

 This effect is due, I believe, to gases or moisture condensed on the 

 walls of the capillary tube, and in some cases to irregularities in the 

 chemical composition of the glass. I found that it did not occur if 

 the tube before being used was heated for some time along its whole 

 length to as high a temperature as it would stand without collapsing ; 

 this heating would tend to cleanse the walls of the tube. That dif- 



