250 Prof. J. J. Thomson. [June 20, 



tive. Ifc is not safe to draw any conclusions from the variations in 

 intensity of one line or one group of lines on the two sides of the 

 plate, as the total quantity of light coming from the neighbourhood 

 of the cathode often differs considerably from that coming from the 

 anode. When, however, we get an increase in the brilliancy of one 

 set of lines accompanied by a diminution in the brightness of another 

 set, when we move across the plate we eliminate this source of error. 

 The differences in the spectra at the two sides of the plate are most 

 easily observed at pressures where there is not any very great 

 difference between the luminosity of the cathode and the anode. As 

 was mentioned at the beginning of the paper, there is a range of 

 pressure within which the effects are irregular, and no decided 

 differences are observed between the spectra at the two sides of the 

 plate. It is desirable in these experiments to keep the tube on to 

 the pump as long as the experiment lasts, for the discharge always 

 decomposes the compound gas, and unless the products of decompo- 

 sition are continually pumped off and replaced by fresh supplies of 

 the compound gas, the spectra of the discharge keep changing. With 

 organic compounds this is especially necessary, as the character of 

 the spectrum often changes entirely very shortly after the com- 

 mencement of the discharge unless fresh gas is continually intro- 

 duced. 



In the following experiments the current was produced by a large 

 induction coil with a mercury slow break. 



When the tube was filled with hydrochloric acid gas at a low 

 pressure, the separation of the hydrogen and chlorine was seen very 

 distinctly, the hydrogen line being much brighter on the side of the 

 plate which acted as the cathode (which we shall call the negative 

 side of the plate) than on the positive side, while the chlorine, on 

 the other hand, was brighter on the positive than on the negative 

 side of the plate. 



When the tube was filled with ammonia gas, the hydrogen lines 

 were bright on the negative side of the plate but were absent from 

 the positive side, while on the positive side of the plate there was 

 the positive pole spectrum of nitrogen, and on the negative side of 

 the plate the negative pole spectrum of nitrogen and the hydrogen 

 spectrum. 



Sulphur Monochloride. When the tube was filled with the vapour 

 of this substance at a low pressure, the chlorine lines were brighter 

 on the negative side of the plate than at the positive, while the sulphur 

 lines were brighter at the positive side than at the negative. Thus 

 the chlorine in this substance behaves in the opposite way to the 

 chlorine in HC1 ; in the latter compound the chlorine ion has a charge 

 of negative electricity, while in the sulphur monochloride it has 

 charge of positive electricity. 



