They are essentially humoral substances; they are found in the blood of vaccinated animals, from which may be obtained antitoxic serums with a specific but transient immunity; and they are also found in the plasmas of the lymph and exudates, in aqueous tumors, and in the milk. They are seldom found in the cells. =Mode of Action.=--Frequent attention has been paid to the mode of action of the antitoxins upon the toxins, a phenomenon of great importance in relation to the phenomenon of immunity acquired against the toxins. At the beginning of our knowledge on this subject, the idea of a destruction of the toxin immediately suggested itself, and was advanced by von Behring.[43] According to this scientist the antibody inhibits the morbigenic action of the toxin by neutralizing the toxin, combining with the latter to form a compound of a chemical nature which is devoid of toxicity and without action on the organism. According to this theory, the influence of the antitoxin on the toxin is direct, and does not require the intervention of the living cellular protoplasm. Such was also the belief of Prof. Ehrlich.[44] [43] VON BEHRING and KITASATO: _Deutsch. med. Wochenschr._, 1890, p. 1113. [44] EHRLICH: _Klin. Jahrb._ 1897, VI, p. 292. Buchner, a little later, believed that the antitoxin, instead of acting directly on the toxin, exercised a direct influence on the living elements of the organism, preserving them from intoxication.[45] [45] BUCHNER: _Münchener med. Wochenschr._, 1893, p. 480. Such was also the opinion of Roux[46]; and Calmette demonstrated that a mixture of venom and of a non-toxic antivenom recovered its toxicity on being heated to 68° C, whereby the antivenom was destroyed (Calmette: _Le Venin des Serpents_, Paris, 1897, p. 58); and Wassermann arrived at the same result.[47] [46] ROUX: _Annales de l'Institut Pasteur_, 1894, VIII, p. 724. [47] WASSERMANN: _Zeitschr. für Hygiene_. The array of proofs offered by these scientists, which we cannot here enlarge upon without uselessly extending our subject, would tend to make one believe, at first glance, that the antitoxin does not act directly on the toxin, but at the present time Buchner's theory appears untenable. Numerous researches have proved conclusively that the toxin and the antitoxin have a specific affinity for each other, by virtue of which these principles combine to form a substance free from all toxicity, but unstable, and which may be decomposed by heat or certain other factors.[48] [48] J. DANZSY: _Annales de l'Institut Pasteur_, XVI, p. 331.