60 Mr Arthur Schuster, On the passage of [Apr. 23, 



electrode. If, after the absorption, the electrodes are heated from 

 outside, the gas is given off again and not reabsorbed until 

 discharges of electricity are sent thi-ough the tube. These dis- 

 charges cause the gradual reabsorption of the gas. 



The following observation is of interest in connexion with this 

 subject. One of the electrodes of a Geissler tube containing 

 oxygen was accidentally covered with minute specks of carboni- 

 ferous matter. The negative glow in oxygen has a purple colour, 

 in carbonic oxide it is blue. In the tube in question the purple 

 glow of oxygen was traversed by blue lines of flow, extending out- 

 wards from the negative electrode into the surrounding space. 

 Along these lines of flow the carbon therefore was carried away 

 from the electrode. The sharp outline of the line of flow shew^ed 

 that the carbonic oxide, giving rise to the blue colour, did not diffuse 

 into the oxygen, and vice versa, so that the fresh supply of oxygen 

 can only have come from the other side along the stream-line. 



A tube containing either a hydrocarbon or carbonic oxide will 

 always shew the spectrum of carbon, in addition to the other 

 spectra seen, though the compound may not be decomposed. The 

 spectrum of carbon however is different in the two cases, being a 

 line-spectrum in one case, and a band-spectrum in the other. 

 This admits of an easy explanation, but only on the supposition 

 that the different spectra of one element are due to different 

 molecular combinations. According to this view it is only natural 

 that the carbon atom dissociated out of carbonic oxide or carbonic 

 acid should give a simpler spectrum, than the carbon separated out 

 of hydrocarbons w^hich contain more than one atom of carbon, 

 bound together. 



In conclusion, I may mention an observation which, though not 

 directly connected with this subject, is of interest. Stratifications 

 are sometimes seen in one tube, which shew different colours on 

 different sides. I have projected the image of such a stratification 

 on the slit of a spectroscope. The part of the stratification which 

 was turned towards the positive electrode was pink, and shewed a 

 spectrum the origin of which is not quite settled. It is the so- 

 called band-spectrum of hydrogen, which however, according to 

 Salet, is due to a hydrocarbon. The part of the stratification which 

 was turned towards the negative metallic electrode was blue aud its 

 spectrum is that of carbonic oxide. The tube some time before 

 did not shew any stratifications. These came in owing to the 

 gradual absorption of the gas by the electrode. The spectrum seen 

 in the pink part of the stratification was then seen in the capillary 

 part, while the spectrum of the blue parts was seen in the wider 

 part of the tube. A mere difference in temperature in different 



