26 The Cytotoxins of Blood Serum 



reactivated, consequently it contained an intermediary body and complement. 

 Goutscharnkow (1901) also observed toxic effects in dogs injected with a similar 

 antiserum obtained from sheep. He however found that the thyroids of sheep 

 were affected by injecting these animals with dogs' thyroids, general systemic 

 disturbance being produced. 



Von Dungeru (1902, p. 51) treated rabbits with emulsions of the eggs of starhsh 

 and sea-urchins, and found the antisera to agglutinate the corresponding spermatozoa 

 of these animals. 



From the work of Metchnikoft" it would appear as if the antibodies for special 

 cells of one animal exerted no influence on others. Delezenne (x. 1900, p. 704) 

 also found that his artificial neurotoxin for the dog did not haemolyse dog 

 corpuscles, and that his artificial haemolysin exerted no neurotoxic action. We 

 have seen that other workers have noted special effects on the cells, emulsions 

 of which had been used for the production of antibodies, but that they acted 

 besides on other elements. I would add that this has been the case in the investiga- 

 tions of Boeri (28, x. 1902) who found a neurotoxic serum to be also haemolytic, 

 and a haemolytic serum to be slightly neurotoxic. The evidence is certainly 

 unanimous with regard to the different antibodies exerting a special action. 



Regardinij the Source of Haemolysins and Bacteriolysins. 



Gengou (1901, Ann. de I'lnst. Pasteur, xv.) concluded that the haemolysins are 

 derived from leucocytes, for the reason that plasma separated from fresh blood, kept 

 cold throughout, by centrifugalization, was less haemolytic than serum. G. Ascoli 

 (9, X. 1902, p. 736) came to the opposite conclusion. Ascoli obtained plasma in the 

 same naanner as Gengou, from a dog which had been immiuiified with rabbit 

 corpuscles. When 2 — 5 c.c. of this dog's plasma were injected into a normal rabbit 

 the latter developed haemoglobinuria lasting 24 — 48 hoiu's. Ascoli concludes 

 that plasma possesses the same haemolytic properties as serum both iji corpore and 

 in vitro. Whereas Nolf claimed that serum injections did not lead to the formation 

 of haemolysins, the contrary has been found to be the case by von Dungern (1899) 

 and Morgeuroth (1902). Schattenfroh (1901) was unable to obtain haemolytic sera 

 by goat serum injections, but did so by injecting goat and human urine, the latter 

 observation being confirmed for human urine by Rufter and Crendiropoulo (24, i. 

 1903). Ide (27, vii. 1902, p. 269) considers that such results are to be explained 

 by the presence of haemoglobin in serum and urine. 



Ide produced anti-pseudoglobulin and anti-serum-albumin and found the first 

 not to haemolyse corresponding corpuscles, whereas the second did so, although 

 its action was but one-tenth as strong as that of anti-haemoglobulin. He concludes 

 therefrom that intact serum and serum albumin possess receptors in common with 

 those of the red blood corpuscles, and these receptors, Ide thinks, are to be found 

 in the haemoglobulin which remains present in serum and serum albumin. 



Shibayama (5, xii. 1901) found haemolysins for dog corpuscles in the spleen 

 and lymphatic glands of normal guinea-pigs, not elsewhere. He considers that 

 when dog blood is injected into such animals these organs are stimulated to 

 increased production of haemolysin in the sense of Ehrlich's theory. 



