ANTITOXIC SERUM 559 



it were, a group in the corresponding antigen, the two groups 

 having been compared to a lock and key. It is, however, to be 

 noted that this specificity is a chemical one rather than a 

 biological one. An anti-serum, for example, developed by the 

 injection of bacterium A may also have some effect on bacterium 

 B, and thus appear not to be specific. We have, however, 

 evidence to show that the antigens in bacterium A are not all 

 identical, and that some of them may be present though in smaller 

 proportion in bacterium B ; thus the theory of combining speci- 

 ficity is not invalidated. The number of different anti-substances, 

 as judged by their combining properties, would appear to be almost 

 unlimited, a fact which throws new light on the complexity of 

 the structure of living matter. When anti-substances are studied 

 as regards their action in vivo or in vitro on the substances with 

 which they combine, different degrees of complexity may be 

 recognised. In certain cases simple combination may occur 

 (antitoxins, antiferments), in other cases physical effects may be 

 associated with combination (agglutinins), and in a third group 

 of cases the anti-body may lead to the union of another body 

 normally present in serum, called complement or alexin. The 

 combination may or may not result in physical changes in the 

 antigen, the evidence of the latter occurrence being elicited by 

 the deviation method (p. 131). Anti-bodies of the third class 

 are known as immune-bodies or amboceptors (Ehrlich) or sensi- 

 tising substances, substances sensibilisatrices of French writers. 



After this preliminary statement in explanation, we shall con- 

 sider the actual properties of the two classes of serum, and later 

 we shall resume the theoretical consideration. 



Antitoxic Serum. In a previous chapter (p. 194) a distinction 

 has been drawn between extra- and intra-cellular toxins, and 

 with regard to these the general statement may be made that 

 while antitoxins are, as a rule, comparatively easily obtained in 

 the case of the former, the matter is quite otherwise in the case 

 of the latter. In fact some writers have gone so far as to say 

 that antitoxins to endotoxins cannot be obtained. Such an 

 extreme view is in our opinion unjustifiable in the light of the 

 recent work on antitoxins to the typhoid, cholera, and dysentery 

 endotoxins (pp. 377, 469, 401). Nevertheless we have the im- 

 portant fact that in many cases by the injection of dead cultures 

 an active anti-bacterial serum can be obtained which has no 

 neutralising action on the endotoxins, and we must conclude 

 either that a large proportion of the endotoxin does not lead to the 

 production of antitoxin or does so only with great slowness, the 

 latter alternative being on general grounds rather improbable. 



