In considering this subject and other subjects allied to it, we must carefully distinguish between chemical composition and chemical constitution; between the mere elements of which a compound is formed and the manner in which these elements are put together. Thus the cyanides, or nitriles, and the isonitriles, or carbamines, both contain carbon and nitrogen, and contain them in equal proportions; but the manner in which the carbon is united with the nitrogen probably differs in the two classes, and their physiological action is different. Their chemical composition is the same, but their chemical constitution is different.

It was pointed out by Blake in 1841 that a close connection exists between the chemical constitution and physiological action of salts; their physiological action on animal organisms appearing to depend chiefly on the base. Yet the physiological action of any salt is not dependent entirely upon the base. It may be, and sometimes is, modified to a very great extent by the acid; moreover, we find that the salts which the same inorganic base may form with different acids may present very different physiological actions, as in the case of the carbonate, bromide, and cyanide of potassium. The same is the case with organic bases and Richardson, in 1865, drew attention to an example of the relation between the action of the base and acid in the amyl compounds. He found that amyl-hydride had an anaesthetic effect; the introduction of oxygen, as in amyl-alcohol or amyl-acetate, added spasm to this action; amyl-iodide produced a large excretion of fluid from the body, while amyl-nitrite had a great effect on the circulation. Thus, the base remaining the same, different acid radicals modified the action of the compound.1

1 Nature, June 22, 1882, p. 187.

2 Ibid., Oct. 8, 1885, p. 562.

The fact is that sometimes the action is determined chiefly by the base (whether it be inorganic or organic), and sometimes chiefly by the acid. The action of the whole salt may differ to a great extent from that of the substances composing it, and it may agree to some extent with other salts, which differ from it both in regard to the base and acid composing them; thus - the sulphate of magnesium and the sulphate of sodium are both purgative, and in this property they agree not only with the sulphate of potassium, in which the base is different although the acid is the same, but with the bitartrate of potassium, in which both the base and the acid are different. This fact confirms what has already been said regarding the necessity for taking into consideration crystalline form and physical aggregation, as well as chemical composition (p. 15).