NATURE OF ANTIBODIES 197 



eoccal polysaccharide. The animals in group one respond with 

 antibody production while those of group two give no antibody- 

 response to its presence in a free state in the body. The sera com- 

 prising group one are horse, man, mouse, cat, dog and goat, while 

 those of group two are rabbit, guinea pig, rat, and sheep. 



The same authors (Horsfall and Goodner) have also shown 

 that the antibody molecule in horse antipneumococcus serum is 

 larger than the corresponding one in rabbit pneumococcus immune 

 serum. This difference in size of the antibody molecule probably 

 explains, in part at least, why the antiserum obtained from rab- 

 bits is more effective in the treatment of pneumonia than antiserum 

 fi'om the horse. Presumably the rabbit serum antibody molecule 

 can diffuse more readily into the tissues. Another difference be- 

 tween antibody from horse and that from the rabbit is observed 

 in the Quellung reaction where the antibody from the horse fails to 

 give the reaction in contrast with the positive results obtained with 

 antibody from the rabbit. According to Chase and Landsteiner 

 (1939) the molecular weights of Types I and III pneumococcal 

 antibodies have been determined in Svedberg's laboratory to be 

 930,000 for horse, cow and pig and 157,000 for the rabbit and 

 monkey. 



Goodner and Horsfall (1937) have also shown that antipneumo- 

 coccus horse serum contains at least three antibodies which 

 precipitate the specific polysaccharide but differ in their ability 

 to protect mice. They employed Heidelberger 's method of sepa- 

 rating antibodies from immune precipitates in hypertonic salt 

 solution. By employing 10 per cent sodium chloride they sepa- 

 rated two fractions of antibody. One of these antibodies was a 

 water-soluble pseudoglobulin-like substance, P, which had low 

 protective capacity and a second water-insoluble (euglobulin- 

 like) antibody, E, which exhibited more protective power but 

 formed precipitates less rapidly than P. Chase and Landsteiner 

 suggest that these antibodies may both be lecithoproteins. 



Origin of Antibodies. — Buchner originally conceived of anti- 

 bodies as being formed from the antigens injected but his theory 

 soon gave way to the ''side chain theory" of Ehrlich. Recently 

 Manwaring and others have attempted to revive Buchner 's theory 

 although modifying it to some extent. Heidelberger (1932) in 

 discussing Manwaring 's theory cites a statement by Doerr "that 



