24 The Cytotoxins of Blood Serimi 



to haemolysin produced in his treated animals through injecting blood contained 

 in the testicular emulsion he used. He argues that if the action of these antibodies 

 is not specific, then an artificial haemolytic serum should destroy the spermatozoa, 

 whereas it does not. He found that a rabbit serum which was haemolytic for sheep 

 corpuscles did not exert any more action upon sheep spermatozoa than did normal 

 rabbit serum, he claims in fact that they lived longer in the haemolytic serum. He 

 concludes therefrom that the antisera may be specific for certain cells as distinct 

 from others belonging to the same animal. To further test this point, Metalnikoif 

 (IX. 1900, p. 581) treated a rabbit with a mixture of blood and spermatozoa from 

 sheep, obtaining an antiserum which haemolysed corpuscles and immobilised 

 spermatozoa. In confirmation of Metchnikoft''s results, he found that he could 

 I'emove the haemolysin by adding corpuscles to the serum, the remaining fluid 

 remaining spermotoxic. On the other hand spermatozoa fixed both the intermediary 

 bodies of spermotoxin and haemolysin, and he concludes that this must have been 

 what misled Moxter. Metalnikoff" (p. 578) obtained spermotoxin from rabbits 

 treated with guinea-pig spermatozoa. He noted that normal rabbit serum was 

 scarcely less spermotoxic than that of treated rabbits, but there was an essential 

 dift'erence, namely, that the normal spermotoxin heated to 56° C. could not be 

 reactivated, whereas the artificial spei'motoxin could be reactivated by adding fresh 

 normal guinea-pig serum thereto, the latter having no influence of itself. Con- 

 sequently artificial spermotoxins, as is the case with haemolysins, are composed 

 of an immune or intermediary body and of complement. 



The existence of normal spermotoxins, as observed by Metalnikoft", has also been 

 proved by Weichardt (25, xi. 1901, p. 835). Two out of three normal rat sera 

 were markedly spermotoxic, the third feebly .so, for rabbit spermatozoa. Pigeon 

 serum was found to be spermotoxic for rabbit, dog, and goose spermatozoa. He 

 observed the normal spermotoxic action of a goose's serum to vary widely in one 

 and the same animal at various times. 



Following along the lines of Ehrlich and Morgenroth's work with haemolysins, 

 Metalnikoff (p. 585) succeeded in producing auto- spermotoxin by injecting guinea- 

 pig spermatozoa into guinea-pigs. The auto-spermotoxin, which was active in vitro, 

 could be reactivated by fresh normal guinea-pig serum after having been inactivated 

 by heat. 



Halban and Landsteiner (25, iii. 1902, p. 475) observed that spermotoxic serum 

 obtained by injecting rabbits with ox spermatozoa agglutinated ox corpuscles much 

 more powerfully than did normal rabbit serum. 



The reaction of diflferent guinea-pigs to foreign spermatozoa injections was 

 found to vary considerably by Weichardt (p. 833), some i)roducing a highly 

 spermotoxic serum, others little or no spermotoxin. 



Anti-spermotoxin, comparable to anti-haemolysin, etc., was produced by Metal- 

 nikoff (IX. 1900, p. 583) by injecting guinea-pig spermotoxin (from rabbit) into 

 normal guinea-pigs. Serum, containing anti-spermotoxin, exerted but a slight 

 effect when fresh, but its effect was marked when it was inactivated, this being due 

 to the fresh serum containing a prejionderance of anti-complement. Consequently 

 auti-sijermotoxic serum consists of anti-intermediary body and anti-complement. 

 Weichardt (25, xi. 1901) reaches the same conclusion, adding that anti-spcrmotoxiu 

 also contains an anti-agglutiuiu. 



