834 COMPLEMENT FIXATION 



of the acetone insoluble lipoids, and of Calmette and Massol' on the quantitative rela- 

 tion between antigen, serum, and complement, were confirmed and extended. The 

 method of securing certain antigenic substances from the tubercle bacillus culture 

 by adsorption on globulin, precipitation of the globulin with carbon dioxide, and 

 subsequent extraction of the antigenic substances from the globulin by alcohol and 

 dispersion in salt solution of the concentrated extract has been recorded. The com- 

 parison of the activity of such purified antigens with the various fractions obtained 

 by extraction of cultures with different aqueous and lipoidal solvents is described 

 fully in a recent paper.' 



The result of these analyses and separations by various methods of extraction 

 demonstrated the distribution, so to speak, of the antigenic activity of the tubercle 

 bacillus in certain fractions containing lipoid and protein. It has not been possible 

 to separate the lipoid from the protein and thus determine the activities separately. 

 From this work, it is quite evident that when the acetone-insoluble lipoids are sepa- 

 rated from the culture material, concentrated, and suspended, a specific and very 

 active antigen is obtained. Similarly, when methyl- and ethyl-alcoholic extracts of the 

 culture material are prepared after removing the acetone-soluble lipoids with the 

 various solvents, an active, specific antigen is obtained. Moreover, simple aqueous 

 extraction also yields an antigen which apparently is quite as specific and active as 

 these other refined preparations. It would seem that the separation of the antigen 

 components has not been carried far enough to isolate the different elements, but the 

 refined product must be considered, from this standpoint, still crude, and it would 

 appear that the specific antigenic activity depends not upon one of these elements, but 

 upon a mixture or combination in very definite relations of a physico-chemical nature, 

 possibly a protein, on the one hand, and lipoid, on the other. 



The most important of recent contributions is that of Avery and Heidelberger.^ 

 The specific soluble substance first isolated by them from pneumococcus cultures and 

 now known to be produced by other bacteria, ^ if a carbohydrate, which, however, may 

 still be questioned, must also be considered. To this so-called "polysaccharide" the 

 type specificity is attributed, whereas the protein exhibits species specificity only.^ 

 Hitherto, specific antigenic activity has been linked up with protein by the great 

 majority of observers.^ 



Apparently some of the simple aqueous or alcoholic extractions yield antigens 

 with cjuite as satisfactory and specific activity as some of the more highly refined or 

 purified preparations. Yet it is also evident that with the cruder antigens their activ- 



' Calmette, A., and Massol, L.: ibid., 67, 528. 1909. 



^ Wadsworth, A., Maltaner, F., and Maltaner, E.: loc. cit. 



3 Avery, O. T., and Heidelberger, M.: /. Exper. Med., 38, 73. 1923; ibid., 40, 301. 1924; Heidel- 

 berger, M., Goebel, W. F., and Avery, O. T.: ibid., 42, 727. 1925. 



1 Avery, O. T., Heidelberger, M., and Goebel, W. F.: ibid., p. 709. 1925; Lancefield, R. C: ibid., 

 p. 377. 1925. 



5 Lancefield, R. C: ibid., p. 397. 1925; Avery, O. T., and Heidelberger, M.: ibid., 38, Si. 1923; 

 Zinsser, H., and Tamiya, T.: ibid., 42, 311. 1925. 



*' Wells, H. ().: The Chemical Aspects of immunity, p. 63. New York: Chemical Catalog Conipan)', 

 1925. 



