170 THE TOXINS PRODUCED BY BACTERIA 



ferment, a toxin causing systolic standstill of the excised heart. Any 

 particular venom contains a mixture in varying proportions of such 

 toxins, and the different effects produced by the bites of different snakes 

 largely depend on this variability of composition. The neurotoxic, the 

 thrombotic, and the hsemolytic toxins are very important constituents 

 of any venom. The toxicity of different venoms varies much, and no 

 general statement can be made with regard to the toxicity of different 

 poisons towards man. Lamb has calculated that the fatal dose of crude 

 cobra venom for man is probably about -015 of a gramme, and that if 

 such a snake bites with full glands many times this dose would 

 probably be injected, but, of course, the amount emitted depends largely 

 on the period which has elapsed since the animal last emptied its glands. 

 When a dose of a venom not sufficient to cause immediate death from 

 general effects be given, very rapid and widespread necrosis often may 

 occur in a few hours round the site of inoculation. 



An extremely important fact was discovered by Flexner and Noguchi, 

 namely, that the hsemolytic toxin of cobra venom in certain cases has no 

 action by itself, but produces rapid solution of red corpuscles when some 

 normal serum is added, the latter containing a labile complement-like 

 body, which activates the venom. In this there is a close analogy to 

 what holds in the case of a hsemolytic serum deprived of complement by 

 heat at 55 C. (p. 479). Kyes and Sachs further showed that in addition 

 to serum-complement a substance with definitely known constitution, 

 namely lecithin, had the property of activating the hsemoly tic substance 

 in cobra venom, the two apparently uniting to form an actively toxic 

 substance. Later still, Kyes succeeded in demonstrating the union of 

 the two substances to form a cobra-lecithid, and in separating the 

 latter as a practically pure compound, which is, unlike lecithin, 

 insoluble in ether, but soluble in chloroform. So far no example of 

 activating a bacterial toxin is known, but the results mentioned point to 

 the possibility of this occurring in some cases in the tissues of the body. 



The Theory of Toxic Action. While we know little of the 

 chemical nature of any toxins we may, from our knowledge of 

 their properties, group together the tetanus and diphtheria 

 poisons, ricin, abrin, snake poisons, and scorpion poisons. 

 Besides the points of agreement already noted, all possess the 

 further property that, as will be afterwards described, when 

 introduced into the bodies of susceptible animals they stimulate 

 the production of substances called antitoxins. The nature of 

 the antagonism between toxin and antitoxin will be discussed 

 later. Here, to explain what follows it may be stated (1) that the 

 molecule of toxin most probably forms a chemical combination 

 with the molecule of antitoxin, and (2) that it has been shown 

 that toxin molecules may lose much of their toxic power and 

 still be capable of uniting with exactly the same proportion of 

 antitoxin molecules. From these and other circumstances Ehrlich 

 has advanced the view that the toxin molecule has a very com- 

 plicated structure, and contains two atom groups. One of these, 

 the haptophorous (ctTrretv, to bind to), is that by which com- 



