Blue Compounds of Cyanogen and Iron. 157 



tenaciously that the latter cannot by mere washing be re- 

 moved from it ; for as soon as the precipitate had been freed 

 from all potassium, it was found to consist no longer of prus- 

 sian blue, but hydrated oxide of iron. From this fact he 

 concludes that the purest prussian blue contains potassium. 

 Berzelius, on the contrary, by washing continuously for three 

 weeks a precipitate of ferrocyanide of potassium with a persalt 

 of iron, succeeded in obtaining a substance, which, though 

 retaining the colour of the original compound, was perfectly 

 free from potassium. The water passing through in the later 

 periods of the operation contained this potassium in the form 

 of ferridcyanide, the formation of which salt he attributes to 

 a process of oxidation resulting from the action of the air ; 

 thus admitting that a decomposition of the original precipi- 

 tate accompanied the removal of the potassium. Without the 

 proof afforded by an analysis, or some characteristic reaction, 

 we cannot venture to conclude that the residue after this de- 

 composing reaction retained the composition of prussian blue. 

 Berzelius thus agrees with Gay-Lussac in the fact that a 

 decomposition of the prussian blue is a necessary condition 

 for the removal of the potassium. The difference consisted 

 only in the kind of decomposition which took place : in Ber- 

 zelius's experiments the residue after decomposition still main- 

 tained a blue colour, while in Gay-Lussac' s, in which by the 

 action of a water which he himself concluded must have been 

 alkaline, the residue consisted of peroxide of iron. Both lead 

 however to the same conclusion. 



Berzelius analysed two kinds of prussian blue containing 

 potassium. The first, which was soluble in water, he found, 

 after separation by means of alcohol from the red and yellow 

 cyanides, which were simultaneously formed, to contain for 

 every 12 atoms of iron 1 of potassium. The other, which was 

 insoluble in water as it remained on the filter, from which 

 the soluble salt had been washed away, contained iron and po- 

 tassium in the proportion of 8 equivalents to 1. The former 

 of these substances he considers as a combination of 2 atoms 

 of the yellow prussiate with 3 atoms of prussian blue, — 



2(FeK 2 Cy 3 )+3(Fe 7 Cy 9 ). 

 The second is a combination of 1 atom yellow prussiate with 

 2 prussian blue, — 



FeK 2 Cy 3 +2(Fe 7 Cy 9 ). 



This great chemist admits however, what indeed the com- 

 plex nature of these formulae sufficiently indicates, that fur- 

 ther light is needed on the nature of these compounds than 

 these analyses afford. 



I shall now describe the results of some experiments which 

 I have performed in the laboratory of Professor Liebig, for 



