32 The Cytotoxins of Blood Serum 



contrast to the serum of rabbits treated either with active or inactive 

 goat serum. Schattenfroh suggests that the method of urine injection 

 may be of use in the production of specific haemolysins for blood-testing 

 by the method recommended by Deutsch. 



Ruffer and Crendiropoulo (24, i. 1908) have confirmed this observa- 

 tion, finding that the injection of human urine two or three times into 

 rabbits produced a haemolytic serum for human corpuscles, the 

 liaemolysin acting however slightly also upon the corpuscles of the 

 guinea-pig. 



As has been found for the precipitins, Metalnikoff (18, IV. 1901, 

 pp. 532, 533) has observed the formation of haemolysins in the serum 

 of rats fed for 1 to 8 weeks on horse blood. An exposure to 55° C. 

 inactivated their serum, it being reactivatable through fresh normal 

 serum ; it consequently contained immune-bcxly and complement. 

 Similarly rats fed on rabbit blood, and i-abbits fed on horse blood, 

 jdelded specific haemolysins. 



Again, as has been found for the precipitins, Calmette and Breton 

 (1, XII. 1902) observed a decrease in the amount of haemolysin in 

 animals treated for longer periods. They found, however, that when 

 such over-treated animals are allowed to rest for some months they 

 produce a much more active antiserum after but two injections. 



