TESTINXJ OP ANTITOXIN. 481 



■may yield a toxin of which an amount representing 20 or 30 

 ■times the minimum fatal dose is sufficient to neutralize one 

 immunity unit. Hence in testing antitoxins, the toxin 

 obtained from cultures which are but about a week old 

 should be employed. 



The method described recently. by Ehrlich is to be considered as 

 a distinct improvement upon the ordinary method of determining the 

 strength of an antitoxin. It is based upon the fact, mentioned above, 

 that the neutralizing power of a toxin with reference to a serum does 

 not depend upon the number of minimum fatal doses present. Al- 

 though, theoretically, 1 I. U. should neutralize 100 minimum f ataj 

 doses yet it has actually been found to neutralize anywhere from 

 16 to 108 doses. This great variation is due to differences in the com- 

 position of various toxins. According to Ehrlich the pure toxin be- 

 comes converted, to a greater or less extent, into " toxoids " which, 

 while they are no longer poisonous, yet possess the same neutralizing 

 power as the original toxin. 



Hence, the amount of toxin employed, when testing an antitoxin, 

 should depend not upon the minimum fatal dose, but upon its neutral- 

 izing power with reference to 1 1. U. of a standard test-serum. Such 

 a standard serum, prepared according to Ehrlich's directions, if 

 available, would be invaluable for comparisons and for the determin- 

 ation of the actual strervgth of an antitoxin. This test-serum is 

 dried in vacuo in the presence qf phosphoric anhydride, and is then 

 preserved in vacuum tubes. The strength of the serum is said to 

 remain unaltered, owing to the absence of air and of moisture. 



The contents (2 g.) of one of these tubes of test-serum is dis- 

 solved in 200 c.c. of a mixture of equal parts of glycerin and 10 per 

 cent. NaCl solution. Inasmuch as the original serum had a strength 

 of 1700 I. U. it follows that 1 c.c. of this solution represents 17 I. U. 



The test-dose of toxin is to be ascertained by means of 1 I. U. of 

 this standard serum. By the test-dose of toxin is understood the 

 amount which, mixed with 1 I. U. and injected subcutanfeously into a 

 250 g. guinea-pig, will cause death on about the 4th day. The test- 

 dose can be readily and accurately ascertained by adding to portions 

 of the dilute serum, each containing II. U., variable amounts of the 

 toxin (30, 50, 70, 100 times the minimum fatal dose) and then injecting 

 these mixtures into guinea-pigs. 



To determine whether a given serum has an advertised strength 

 a portion of it should be diluted so that presumably 1 1. U. is contained 



31 



