NATURE OF ANTITOXIC ACTION 471 



animal no symptoms take place ; this shows that after a time 

 neutralisation is complete. Again, in cases when the toxin has 

 some definite physical effect demonstrable in vitro, e.g. lysis, 

 agglutination, coagulation, or the prevention of coagulation, its 

 action can be annulled by the antitoxin ; in such circumstances 

 manifestly no physiological action of antitoxin through the 

 medium of the cells of the body can come into play. These 

 facts are practically conclusive in favour of antitoxin action 

 depending upon a direct union of the two substances concerned. 



The evidence usually brought forward against the direct union of 

 toxin and antitoxin rests chiefly on certain observations of Calmette, 

 who found that the antitoxin to a snake venom was more easily destroyed 

 by heat than the toxin, and stated that when a neutral mixture of the 

 two was heated at a temperature sufficient to destroy free antivenin, the 

 toxic properties in part returned. Hence he concluded that the two 

 bodies existed in an uncombined condition in the mixture. Martin and 

 Cherry, however, on repeating these experiments, found that the above 

 result was not obtained if sufficient time for complete combination was 

 allowed ; but if this precaution was not taken, then the presence of the 

 free toxin was revealed when the antitoxin was destroyed by heat. 

 Even, however, if Calmette's results were quite correct, they cannot be 

 considered to constitute a proof that chemical union does not occur : 

 they would only prove that the toxin has not been destroyed. If 

 two complicated chemical compounds of unequal stability are in loose 

 chemical union, it is quite conceivable that the less stable may be 

 destroyed (e.g. by heat), whilst the more stable escapes. 



Although practically all authorities are now agreed as to the 

 direct combination of toxin and antitoxin there is still much uncer- 

 tainty as to the exact nature of this union. Controversy on this 

 subject may be said to date from the important work of Ehrlich 

 on the neutralisation of diphtheria toxin. Using an immunity 

 unit of antitoxin (the equivalent of 100 doses of toxin) he deter- 

 mined with any example of crude toxin the largest amount of 

 toxin which could be neutralised completely, so that no symptoms 

 resulted from -an injection of the mixture. This amount he 

 called the limes null dose, expressed as L . He then investigated 

 the effects of adding larger amounts of toxin to the immunity 

 unit and observed the quantity which was first sufficient to 

 produce a fatal result, that is, which contained one M.L.D. of 

 free toxin ; this amount he called the limes todtlich, fatal limit, 

 expressed as L t . Now if, as he supposed, the union of toxin and 

 antitoxin resembled that of a strong acid and base, L^ L ought 

 to be the equivalent of a minimum lethal dose of the toxin alone. 

 This, however, was never found to be the case, the difference 

 being always considerably more than one M.L.D. For example, 

 in the case of one toxin, M.L.D. = '0165 gr., L t . = 1'26 gr., L = 



