62 EIGHTH REPORT — 1838. 



meleon. On careful analysis, however, the author has assured himself 

 that manganese does not occur as a constituent of caustic potass, and 

 that its aqueous solutions own their colour, and the change of their 

 colour, to the presence of protoxide and proto-chloride of iron in solu- 

 tion ; the former being held so by the presence of chloride of potas- 

 sium. These, by taking up oxygen (probably from the air combined 

 with the water), become further oxydized and gradually precipitate in 

 combination, leaving the solution colourless, and giving rise to a new 

 compound of sesquioxide and sesquichloride of iron, consisting, by the 

 author's analysis, of 10 atoms of the former and one of the latter body, 

 or 



(Fe, CI,) + 10 (Fe, O^). 



It is anhydrous. Having observed that these changes took place 

 at very different rates in bottles of variously coloured glass, the 

 author commenced a series of experiments on the relative effects of 

 light transmitted to the solution through various media, and has found 

 and recorded in ruled sheets, by curves, the results of observations 

 made every two hours. In these curves, the ordinates represent time, 

 and the abscissae the rate of chemical change, as marked by certain 

 changes of colour. 



Means were taken to prevent inequality of temperature, or of inten- 

 sity of light, in each solution exposed to a coloured ray, and some of 

 the results arrived at are given in the subjoined table. 



Mode of exposure to light. ^SoratiS!* 



Violet glass, exposed to air 30 hours. 



Violet glass, closed 50 



Flint glass, colourless, exposed to air 80 



closed 115 



. . yellow „ 170 



. . blue „ 185 



. . orange „ 190 



, . red „ 200 



Green glass, by oxide copper ] wholly unchanged in 



Do. Bristol bright metal j colour in 200 hours. 



The rate of progressive decoloration for different rays at different pe- 

 riods is very various. 



The result as to grcon glass agrees with Mrs. Somerville's experi- 

 ments. Ink, which Sir John Herchel found transmitted white light 

 unaltered, was used to reduce all the coloured media to the same illu- 

 minating power, and immersion in water to preserve the temperature 

 constant. 



The author conceives that this is the first attempt that has been 

 made to reduce the phenomena of the chemical action of light to nu- 

 merical registration, and suggested the importance of the subject, 

 upon which so little was known, and in which the observed cases were 

 so few. 



