Brusli Discharge in Weak Acids and Solutions. 817 



in the positive brush, while the hydrogen lines are con- 

 siderably weakened. As the concentration increases, the 

 lines remain of the same intensity, with slight fluctuations, 

 provided the current is kept constant. The behaviour of 

 the oxygen lines in the positive brush is the same for all 

 solutions. 



In acid solutions only the faintest traces of the lines could 

 be detected in the negative brush, and were generally very 

 short and close to the point. Generally the spectrum was 

 entirely absent. In solutions of magnesium sulphate and 

 zinc sulphate, very faint traces were discovered at concen- 

 trations less than '005 per cent, but above this none of the 

 lines were discernible. Faint traces of the spectrum occur 

 in tap- water. 



In solutions of hydrogen peroxide and of the salts of the 

 alkali metals, the behaviour is quite different. Hydrogen 

 peroxide, however, is not an independent example, as it was 

 very considerably contaminated with sodium. In weak solu- 

 tions of these salts the spectrum disappeared from the nega- 

 tive brush but reappeared with further concentration. The 

 lines at first were short and faint, being confined close to 

 the point. With further concentration, the current remaining 

 constant, they grew in intensity and length till they were 

 brighter than the hydrogen lines and much brighter than 

 the corresponding spectrum in the positive brush. With 

 further concentration, the spectrum remained approximately 

 of the same intensity for the same current. The concentra- 

 tion at which the oxygen lines reach their maximum intensity 

 lies between *005 per cent, and *01 per cent. It will be seen 

 that the behaviour of the oxygen spectrum in this respect is 

 exactly similar to the behaviour of the metallic lines of the 

 solute (see above). Indeed, the connexion between the 

 spectra is very evident on all the negatives. The connexion, 

 however, is not absolutely rigid, though it is quite definite. 

 After the maximum has been reached for both spectra, one 

 may show a slight increase in intensity in a subsequent 

 spectrogram, while the other shows a slight decrease, and 

 vice versa. 



In all cases the oxygen lines are brightest close to the 

 point in the negative brush. In this respect they are 

 different from the positive spectrum. Here the lines are 

 long and of unvarying intensity throughout their length, 

 being longer than the hydrogen lines, which in some eases 

 are brighter than they are and in all cases are brightest close 

 to the point. In a few spectra of the positive brush, the 

 oxygen lines stop short a small distance from the point. 



