1214 M. Berzelius on the [March, 



the quantity of oxide of cobalt thus carried down is too inconsi- 

 derable to alter in any notable degree the result of an analysis ; 

 especially as the two oxides possess the same saturating power 

 and consequently the chemical constitution of the compound is 

 easily discovered. If the oxide of nickel precipitated by the 

 potash contains any cobalt, it becomes brown when any very 

 dilute acid is poured upon it ; for the peroxide of cobalt dissolves 

 much more slowly than the oxide of nickel. It may be disco- 

 vered also by means of the blowpipe, if the oxide of nickel be 

 treated with borax until it is reduced, and the red colour which 

 it gives to glass disappears. If it contains cobalt, it is then 

 discovered by a more or less perceptible blue colour. I am of 

 opinion that by Mr. Phillips's method, we may perfectly succeed 

 in separating these two oxides, especially if the abovementioned 

 precautions are observed. 



The oxide of cobalt remaining in the ammoniacal liquor gives 

 it a rose colour. By evaporating the solution, the oxide is 

 deposited in brown flocculi, and may be collected by the filter. 

 If the ore under examination contains silica in such a state that 

 it may be dissolved, it now precipitates with the oxide of cobalt. 



Oxides of Nickel and Copper. — 1 have not been able to deter- 

 mine whether oxide of copper is soluble in ammonia or not. It 

 is certain that all those solutions which are generally regarded as 

 oxide of copper in ammonia are double salts with excess of base. 

 I digested oxide of copper in concentrated ammonia for eight 

 days in a stopped bottle. The solution became of a light-blue 

 colour in 48 hours, and it did not afterwards increase. A drop 

 of carbonate of ammonia let fall into the liquor immediately dis- 

 solved a part of the oxide, and made the lower stratum of the 

 liquid of a deep-blue colour. 



When an ammoniacal solution of oxide of copper is mixed 

 with caustic potash, the oxide of copper is precipitated in a few 

 seconds, and if the quantity of the potash is sufficient, it is 

 entirely deposited in the form of a blue hydrate, which it is very 

 easy to wash. When well washed, it yielded blue hydrate of 

 copper, combined with two atoms of water ; it does not retain 

 any trace of potash. In order that the copper may be perfectly 

 precipitated from the ammoniacal solution, much more caustic 

 potash must be employed than to separate oxide of nickel — a 

 circumstance which is probably derived from cnprate of potash 

 forming only in a liquid which is saturated to a certain degree 

 with hydrate of potash ; it is more easily decomposed than the 

 niccolate. 



I endeavoured to take advantage of these properties of oxide 

 of copper to determine the quantity of copper in these analytical 

 experiments. The method which has been always used to pre- 

 cipitate metallic copper by iron is extremely bad, and always 

 ^ives incorrect results ^ for, on one hand, the copper is always 



