If to this we add that cyanide of potassium crystallizes in cubes, as do the chlorides, bromides, and iodides of this base, I think we exhaust the evidence which can as yet be alleged in favour of the analogy of cyanogen to the elements referred to.
Such are the reasons for classifying cyanogen with these radicals, and I will now go over them seriatim.
In the first place it is true that cyanogen combines directly with the least oxidizable metals, but so does oxygen when in the allotropic state, also sulphur at a slight elevation of temperature, and further its hydride (sulphuretted hydrogen) imitates hydrocyanic acid in presence of the metals instanced; oxygen and sulphur should therefore on this principle be admitted along with cyanogen into the group of radicals, which would be absurd, as they are not admitted as radicals at all. Therefore these tests are unreliable.
In the second place, besides cyanogen, sulphur and a number of other elements combine readily with hydrogen, the bulk of which are not halogens, while in reference to the supposed analogy existing between hydrocyanic and hydrochloric acids, I really fail to see any grounds for this.
Hydrochloric acid is a very strong one, intensely acid, and forms salts with the alkaline metals which are quite neutral. Hydrocyanic acid on the other hand, if acid at all (which I doubt), is so feebly so that "it scarcely affects the blue litmus paper"; indeed I believe it to be neutral, as any minute acid reaction which has been obtained in respect to it may be due to carbonic acid, hydrocyanic acid being very prone to decompose with evolution of carbonic acid.
Further, in accord with this, the salts of cyanogen with the alkaline metals are not neutral, as are the corresponding salts of the chlorine group, but strongly alkaline.
In reference now to the third supposed joint characteristic of cyanogen and the radicals to which it is compared, we can parallel this in the case of sulphur and phosphorus; thus either of these elements, when warmed with a solution of any caustic alkali, forms oxygenated and haloidal salts, a part of them being oxidized at the expense of the oxygen of the alkali, as in the case of cyanogen, chlorine, etc., under these circumstances.
Lastly, as to the polar affinities of cyanogen and the crystalline form of its potassium salt. Sulphur and oxygen when liberated by voltaic action also
