196 IMMUNOLOGY 



tionation of the globulins of Types 1 and 2 antipneumoeoc- 

 cal horse serum. They describe a new fraction of water-in- 

 soluble globulin which contains most of the antibody. They 

 named the fraction Piy. It seems to be made at the expense of 

 Pii since the latter had almost entirely disappeared when 

 Piv appeared. The authors suggest that the new antibody globu- 

 lin may be a modified form of the normal globulin fraction Pn- 



Ando, Takeda and Hamano (1937, 1938) report isolating two 

 fractions (A and B) of immune rabbit serum that contain anti- 

 bodies. Fraction A is a water-soluble globulin and contains anti- 

 bodies sucli as the antitoxins and the sliiga-dysentery and per- 

 haps Paratyphosus B. antipolysaccharide antibodies. Fraction B 

 is water insoluble and contains antibodies for pneumococci, E. 

 tifphosa, P. pestis and many other bacteria. They call attention 

 to the fact that these antibodies characterizing one fraction may 

 be found to some extent in the other. We have noted, in unpub- 

 lished work, that Avhile antibodies were present in either the 

 euglobulin or pseudoglol)ulin I fraction they were found pre- 

 dominating first in one and then in the other. Perhaps imperfect 

 methods of separation are partly responsible for these results. 



Horsfall and Goodner (1935, 1936) liave reported interesting 

 differences in antipneumococcal sera ol)tained from different an- 

 imal species. The.y were able to divide the ten Type I antipneu- 

 mococcal sera from the ten animal species studied by them into two 

 groups. In group one, of which the horse-antiserum is a proto- 

 type, the antibody contains the lipoid lecithin, the presence of 

 which seems to be necessary for agglutination or precipitation 

 while the antibody of the rabbit, the prototype of group two, con- 

 tains the lipoid cephalin which is essential for agglutination. They 

 found the antibodies from which the essential lipid was removed 

 Avould nevertheless combine with its antigen but neither agglutina- 

 tion nor precipitation would follow. Furthermore they found tluit 

 while the antibodies of group one combined with the correspond- 

 ing pneumococcal polysaccharide, the resulting complex fails to 

 fix complement. On the other hand, the complex resulting from 

 the union of antibodies of the second grou]i with pneumococeus 

 polysaccharide fixes complement. 



Another interesting difference is in the immunological response 

 the two groups of animals give to the injection of Type I pneumo- 



