4 Thomson, Detection of Arsenic in Beer. 



came off gradually for a long time, it was found that 

 smaller quantities of arsenic could be detected with it 

 than with the platinum kathode, but the results were 

 irregular. 



Copper also gave irregular results. 



With aluminium and plumbago as kathode, the -g-^o^th 

 of a grain per gallon when using 50 c.c. of the solution 

 gave no mirror. 



Tin was tried, but I failed to get any of this metal 

 free from arsenic, even after long continued passage of the 

 current through it. If pure it might possibly form a 

 satisfactory kathode. 



Zinc as kathode. 



A quantity of pure granulated zinc was put into the 

 porous pot, and one of the pieces attached to the platinum 

 wire which carried the negative electric current so that 

 the granulated zinc constituted the kathode. This 

 arrangement gave stronger mirrors for the different 

 quantities of arsenic than the platinum kathode apparatus, 

 but the mirror obtained with the loWth of a grain (when 

 using 50 c.c.) was not equal to the standard produced 

 from the Marsh-Berzelius method. The photograph of 

 these mirrors is shewn in Series A Fig. i, Plate VIII. The 

 first tube in each series is a blank, made with the reagents 

 alone, and they are all free from any trace of arsenic 

 mirror, and as the photographs were taken with the 

 illumination immediately above them, for the purpose of 

 preventing the reflection of the glass from interfering 

 with the photographs of the mirrors, those tubes which 

 contain no arsenic mirrors are either seen with diflficulty, 

 or are invisible in the photographic prints : by following 

 up, however, from the figures at the bottom, the positions 

 of the tubes without mirrors can be fixed by the eye in 



