Brush Discharge in Weak Acids and Solutions. 823 



the spectra are strongest in the positive brush. In solutions, 

 the intensity of the spectra relative to the rest o£ the brush 

 spectrum (water-vapour bands, platinum and oxygen lines, 

 &c.) is considerably diminished. Neither the nature of the 

 solution nor the sign of the brush has any perceptible effect. 

 There is no certain indication that electrolysis plays any 

 part in the appearance of the spectra. 



The spectrum of the metallic lines of the solute appear in 

 the negative brush in solutions. For constant current, the 

 spectrum grows in intensity with the concentration till a 

 certain value of the concentration is reached when no further 

 increase occurs. The spectrum appears in the case of the 

 positive brush in the strongest solutions, and then only very 

 faintly. The behaviour of the spectrum in all cases can 

 rendily be explained from the point of view of electrolysis. 



The elementary line-spectrum of oxygen appears in dis- 

 tilled water in both the positive and negative brush. In 

 dilute acids, solutions of magnesium and zinc sulphates, and 

 in tap-water, it appears in strength in the positive brush, but 

 is generally absent or very weak in the negative. In solu- 

 tions of the salts of lithium, sodium, and potassium, its 

 appearance in the positive brush is the same as for other 

 solutions ; in the negative brush the spectrum appears with 

 the metallic lines of the salt, increases in intensity with them 

 and reaches its maximum with the same concentration of 

 solution. The negative bands of nitrogen behave in the 

 same manner. In the negative brush, both the positive and 

 negative bands appear in general remote from the point, 

 while the oxygen lines are always produced close to the 

 point. 



The arc spectrum of platinum appears in the negative 

 uncondensed brush. In the condensed brush, the spark lines 

 appear. When copper wire is used in the point electrode, 

 the copper lines appear in each brush but are strongest in 

 the negative. 



The water-vapour bands appear in each brush equally 

 well. 



In conclusion, I wish to thank Prof. Sir J. J. Thomson 

 for the kind interest shown during the prosecution of this 

 research. 



Cavendish Laboratory, 

 Cambridge, 1914. 



3 I 2 



