ZINSSER AND MUELLER 725 



The earliest work dealing with substances of this nature in which their immunolo- 

 gical relations were considered was published by Pick in 191 2.' In cultures of typhoid 

 bacilli Pick found a substance not destroyed by pepsin, by trypsin, or by boiling, 

 which failed to exhibit the usual chemical protein reactions but which, nevertheless, 

 was specifically precipitated by typhoid-immune serum. Since this material was de- 

 scribed by him as not possessing the power of inducing antibody formation in animals, 

 we feel justified in assuming that Pick's substances were in principle identical with 

 those later described by others. 



In 191 7, Dochez and Avery^ observed in the blood stream and urine of pneumonia 

 patients a substance which reacted specifically with immune serum but was not de- 

 stroyed by boiling or by tryptic digestion. It was precipitable by alcohol, acetone, and 

 colloidal iron and was not dialyzable. They concluded from nitrogen determinations 

 that it was of protein nature or associated with proteins, undoubtedly an error sim- 

 ilar to the one we ourselves made later with tubercle-bacillus residue, and due to the 

 difficulties of purifying the crude carbohydrate extract. They found the same sub- 

 stances in young and rapidly growing broth cultures of pneumococci and some other 

 bacteria, and concluded, therefore, that they were products of cell growth and not of 

 cell decomposition. 



In 1 92 1, one of us,^ in connection with experiments upon the tuberculin reaction, 

 reported the presence of a similar substance in tubercle-bacillus extracts prepared in a 

 manner calculated to exclude a considerable proportion of the bacterial proteins. 

 These materials, like those previously described by Pick and by Dochez and Avery 

 with other organisms, were left in alkaline extracts after removal of the acid precipi- 

 table nucleoproteins, were alcohol precipitable, heat stable, and reacted in vitro with 

 specific sera, both in giving precipitation and alexin fixation reactions. Owing to an 

 insufficient purification of these substances at that time, these "residue materials," 

 as we call them, also gave skin reactions in tuberculous guinea pigs, and it was only 

 by our subsequent studies that it was found that the skin-reactive constituent was an 

 impurity. It was then shown that the specific precipitable substance and the skin- 

 reactive fraction in these original residue materials could be separated. 



In 1923, similar substances were found by one of us with Parker in extracts of 

 pneumococci, staphylococci, influenza and typhoid bacilli, and since then the ex- 

 istence in many other bacteria of these non-protein specific residue materials has been 

 demonstrated. In spite of the earlier difficulties in defining these "residue antigens," 

 as we call them, from other bacteria, they were held to be analogous to the non-pro- 

 tein substances of Pick and to the soluble substances of Dochez and Avery. Like these, 

 they were precipitable from aqueous solution by alcohol; the precipitates were easily 

 soluble in water, were heat stable, and retained their specific properties. Also, in 

 spite of energetic attempts, these residue substances never induced antibody forma- 

 tion upon injection into animals. 



Further developments of this work by many investigators have shown that materi- 

 als similar to those described are present in all bacteria in which suitable efforts have 

 been made to find them. As greater familiarity with these substances was acquired 



' Pick, E. P., in Kolle and Wassermann: Hatidb. der path. Mikroorg. (2d ed.), i, 781. Jena, 1921. 



'Dochez, A. R., and Avery, O. T.: /. Exper. Med., 26, 477. 1917. 



3 Zinsser, H.: loc. cit.; Zinsser, H., and Parker, J. T. : loc. cit. 



