According to this scientist, the complex molecule of albuminoid substances is constituted by a fixed central nucleus, and by a number of lateral chains or receptors, fixed to this nucleus, which possess diverse accessory functions, and which serve, particularly, for the nutrition of the cells. These receptors have a great affinity for the various substances necessary for the support of the living elements, and they seize upon the alimentary substances, in normal life, just as a leaf of the _Dionæa_ seizes a fly which serves as its food. In these special conditions the receptors may attach themselves to the complex molecules of albuminoid substances, such as the different toxins. Ehrlich supposes, as we have already seen, that a toxin contains two special groups--a _toxophore_ group, which poisons, and a _haptophore_ group, which combines with the receptor. According to this theory, the toxophore group of a toxin can act on an organism _only_ when the haptophore group of the toxin encounters a suitable attachment or receptor. The receptors attached to the living protoplasmic molecule attract the toxin, just as a lightning-rod attracts the lightning. It is hence clearly proved that the toxigenic poisons exert their noxious action on the cellular elements of sensitive organisms, by entering into combination with these. Experience has shown that they attach themselves, in a most rigorously elective manner, to the tissues, and rapidly disappear from the general circulation. Numerous facts, clearly established, attest the reality of this fixation or attachment. It is thus that von Behring and Wernicke[33] sought to ascertain the quantity of antitoxin (we will see farther on that this name is given to those substances which neutralize the activity of toxins under certain conditions) which, introduced a certain time after the introduction of the poison, will save the life of the animal. They have experimented with diphtheria toxin, which we will study later, and they have demonstrated that, if the antitoxic serum be introduced immediately after the toxin, a dose of antitoxin twice as large as that of the toxin suffices to effect a cure. [33] VON BEHRING and WERNICKE: Zeitschrift für Hygiene, XII. Eight hours after the administration of the toxin the dose must be trebled, while after thirty-six hours it is necessary to have recourse to a quantity of antitoxin eight times as great. These experiments show that the curative action of the antitoxin is so much the less the longer the period of time that has elapsed between the introduction of the toxin and the antitoxin. This is because the toxin has become so intimately attached to the tissues that the antitoxin introduced has not the power to destroy the combination. These facts have been confirmed by Donitz[34] and by the classic experiments of Decroly and Rousse.[35] [34] DONITZ: Ueber die Grenzen der Wirksamkeit des Diphtheria Heilserums. _Deutsche Med. Woch._, No. 27, 1897.