212 



The experiments were kept foi- hours at tlie same temperature as the 

 room. Meanwhile the colour was accurately examined; while I used the 

 following indications for the red tint, arranged according to decrea- 

 sing intensity: light-red > somewhat red > reddish > somewhat 

 reddish > colourless. In the table we find the colour-changings re- 

 gistered; the data concerning one experiment, obtained at the times 

 indicated, are given on one horizontal line. 



