164 Dr. Williamson on the 



the manner described in the preceding instance, and was 

 found, though not absolutely free from potassium, to contain 

 so extremely small a quantity of this element, as to render a 

 determination of it next to impossible ; so that the difference 

 produced by neglecting it fell far within the limit of the or- 

 dinary limit of analysis. After drying at the ordinary tem- 

 perature over sulphuric acid it was subjected to analysis. 



I. (f 765 grm. gave 0*383 peroxide of iron, corresponding 

 to 34*6 per cent, of iron. 



II. 0*763 grm. gave 0*379 peroxide of iron, corresponding 

 to 34*4 per cent, of iron. 



Burnt with chromate of lead, — 



I. 0*535 grm. gave 0*344 carbonic acid and 0*155 water, 

 corresponding to 17*5 per cent, carbon and 28*9 water. 



II. 0*563 grm. gave 0*359 carbonic acid and 0*157 water, 

 corresponding to 17*4 per cent, carbon and 27*9 water. 



III. 0*477 grm. gave 0*307 carbonic acid and 0*135 water, 

 corresponding to 17*5 per cent, carbon and 28*2 water. 



The proportion of 5 equivalents of iron to 6 of cyanogen 

 requires to 34*5 of iron, which is the average found, 18*2 car- 

 bon instead of 17'5, as is found. If it be considered that the 

 error of iron and carbon determinations are here added to- 

 gether, the accordance will I think be held to be sufficient. 

 My reason for not calculating any formula for the water was, 

 that owing to the readiness with which the compound is de- 

 composed, I was compelled to analyse it without drying at 

 an elevated temperature, which would have produced an evo- 

 lution of hydrocyanic acid. On decomposing this substance 

 in the fresh prepared state by potash, proto- and peroxides of 

 iron are set free and ferrocyanide of potassium formed. A 

 portion of this compound was, after carefully washing out and 

 without drying, decomposed by carbonate of potash. The 

 proto- and peroxides of iron, thus separated, after long-con- 

 tinued heating in the air, weighed 1*627 grm. of peroxide. 

 The liquid boiled in and treated with sulphuric acid gave 

 1*077 grm. peroxide of iron. The proportion of 3 to 2 re- 

 quires 1*623 and 1*082. Of 5 equivalents of iron 3 were se- 

 parated in combination with 4 atoms of oxygen of the potas- 

 sium, which takes their place, while the other two being con- 

 tained in the composition of the radical remained in solution. 



I will now pass to the consideration of the manner in which 

 we may consider these elements to be combined. It is well 

 known that ferridcyanide of potassium is reduced by the ac- 

 tion of sulphuretted hydrogen, that is to say, it takes up hy- 

 drogen, and is thus converted into a mixture of 3 atoms fer- 

 rocyanide of potassium and 1 atom hydroferrocyanic acid (or 

 ferrocyanide of hydrogen). 



1 



