132 APPLIED BA.CTEB,IOLOGY 



recorded in the tables by simply dividing the fraction of a 

 C.C. that protects by 10 — e.g., when T-roirth of a c.c. of a 

 serum protects, the serum contains 100 normal units per 

 c.c. ; when ^^th c.c. protects, the serum contains 60 normal 

 units per c.c, and so on. 



The lethal dose of toxin is determined by estimating the 

 amount of toxin required to kill per 1,000 grammes weight 

 of guinea-pig. Of this toxin, ten times the amount required 

 to kill a guinea-pig of from 200 to 300 grammes weight is 

 injected, together with the antitoxin to be tested, and the 

 effect observed. By noting the absence or presence of local 

 reaction and the increase or loss of weight, it is stated that 

 an opinion may often be formed after twenty-four hours, 

 but that after the lapse of forty-eight hours a decisive 

 conclusion can always- be arrived at. When the toxin is 

 completely neutralised, as it should be, the animal should 

 not only live, but there should be no trace of local 

 reaction (cedematous swelling). This swelling may not 

 be apparent in the first twenty-four hours, but a rapid 

 fall in weight will at that time frequently indicate its 

 probable occurrence within the next twenty-four hours. In 

 this connection it should be borne in mind that guinea- 

 pigs, taken from stock and put into small cages, usually 

 rise in weight when not injured by the action of the 

 toxin. 



Koux's method of standardising anti-diphtheric serum 

 consists in defining the proportion of serum in relation to 

 the weight of the animal which would protect a guinea-pig 

 against a lethal dose of culture or toxin of the diphtheria 

 bacillus. This method is still made use of at the Pasteur 

 Institute and by other makers, the strength of serum which 

 they consider suitable for use in the treatment of diphtheria 

 cases being one which contains sufiicient antitoxin to protect 

 an animal of from 100,000 to 200,000 times the weight of the 



