348 MANUAL OF BACTERIOLOGY. 



Small quantities of blood may be withdrawn from time to time, and the serum 

 tested for its antitoxic strength. When a satisfactory serum has been attained, 

 the animal may be bled and the serum saved for therapeutic purposes. 

 Through an incision in the skin a trocar is inserted into the jugular vein. The 

 blood is drawn into sterilized flasks with every precaution to insure sterility. 

 The blood is allowed to coagulate and is placed for a time in the ice -chest. 

 The serum is then withdrawn with sterilized pipettes. Small amounts of 

 chemical germicides, as carbolic acid or chloroform, are sometimes added to 

 assist in preserving it. This serum is the so-called antitoxin used in medical 

 practice. 



Many methods have been recommended for concentrating antitoxin. Gibson* 

 recommends a modification of the method devised by Pick. Briefly stating 

 and omitting many details, Pick's method consists of precipitating the anti- 

 toxic serum with a saturated solution of c.p. ammonium sulphate. Gibson 

 redissolves this precipitate with a saturated solution of sodium chloride, repre- 

 cipitates with ammonium sulphate or better with dilute acetic acid and dializes 

 in running water over night, neutralizes the acetic acid if this has been used, 

 and dializes in running water for two days longer. This preparation is said 

 to possess many advantages. For one thing there is said to be no rash fol 

 lowing its use as with so many preparations of antitoxin. 



Since antitoxin is not obtained as a pure chemical substance, and conse- 

 quently cannot be weighed and measured as other theraputic preparations, 

 an arbitrary standard to express the potency of the serum, called an immunity 

 unit, has been devised by Behring and modified by Ehrlich.f Formerly this 

 unit was taken to be 10 times that amount of antitoxic serum which just neu- 

 tralized 10 fatal doses of toxin for guinea-pigs weighing 250 grams. In other 

 words, the exact amount of a certain toxin required to kill a guinea-pig weigh- 

 ing 250 grams in four days having been determined by inoculating a number 

 of guinea-pig?., ten times this amount was put into each of a number of test- 

 tubes, and the antitoxin to be tested was added, a slightly different amount to 

 each tube of toxin. The contents of each tube was then injected into a sepa- 

 rate guinea-pig. 



If any of the animals survived, the amounts of antitoxin in the tubes with 

 which they had been inoculated having been noted, the smallest of these amounts 

 i. e., the smallest amount found necessary to neutralize the toxin was re- 

 gared as one-tenth of an antitoxic unii. It was naturally assumed that 10 

 times this amount of antitoxin would neutralize 100 fatal doses. This, how- 

 ever, was found not to be the case (see Immunity, page 223). So the revised 

 standard now employed in Germany, France, America and other countries is 

 the unit recommended by Ehrlich. This consists of comparing the antitoxin 



*Journ. Biol. Chem. Vol. I. Jan., 1908. pp. 161-170. Abstr. in Bun. 

 de I' lust. Past. Vol. IV., No. 9. 1906. p. 426. 



fRosenau. Immunity Unit for Diphtheria Antitoxin. Public Health and 

 Marine Hospital Service. Bulletin 21. 1905. 



