DIPHTHERIA TOXIN. 477 



which is the amount just sufficient to kill, lies between 

 these two limits. Another series of guinea-pigs is then 

 taken and injected with ire, A, *V and T V c.c. respectively odL 

 the filtered toxin. If fa of a c.c. still kills, whereas -fa of a 

 c.c., although it sickens the animal, does not kill, it fol- 

 lows that the former is to be considered as the minimum 

 fatal dose.. 



The minimum fatal dose of a filtered diphtheria culture 

 will vary with the duration of culture, the medium used, 

 the temperature of incubation and the virulence of the ba- 

 cillus employed. It has been obtained as low as 0.001 c.c., 

 but as a rule it is above 0.01 c.c. 



In accurate work it is desirable to inoculate three or six guinea- 

 pigs with the established minimum fatal dose. The individual ani- 

 mals vary more or less in resistance and it is, therefore, necessary ta 

 make certain that the minimum dose is surely fatal, not merely to 

 one but to a number of animals. In a physiological experiment of 

 this nature it must not be expected to obtain an accuracy compara- 

 ble to that of a quantitative chemical analysis. Sometimes a guinea- 

 pig will not die when inoculated with what is presumed to be several 

 times the minimum fatal dose. Hence, it may happen that out of a 

 dozen animals, inoculated with the minimum fatal dose, perhaps one 

 or two -may eventually recover. The resistance of the animal body 

 is not a constant factor and for that reason the minimum fatal dose 

 cannot be considered as an absolute result. Moreover, since the 

 strength of an antitoxin is determined by testing against a known 

 amount of the toxin it follows that the former result, expressed in 

 immunity units, is approximate and not exact. 



Recognizing the difficulties mentioned above, it is cus- 

 tomary to define a ' * minimal fatal dose " as that amount of 

 toxin which will usually kill guinea-pigs, weighing 250 g., 

 on the 4th day, or at most on the 5th. This amount may 

 kill a very susceptible animal in 1^-2 days. If, however, 

 a number of animals die in less than two days it is evident 

 that the quantity employed contains more than one mini- 

 mum fatal dose. The animal should be weighed and the 

 fatal dose per 250 g. of body weight calculated. Thus, if 



