BACTERIAL ANTIGENS AND SPECIFICITY 353 



gently with baetorial antigens if more were known about their 

 chemistry. This sul),ieet lias been dealt with recently hy Baiim- 

 gartel (1928), Branham (1928) and Zinsser, Enders and Fother- 

 gill (1939). Some of the more recent literature has been reviewed 

 by Heldelberger (1932, 1933) and Marraek (1938). 



Errors Due to Presence of Extraneous Material in Antigens. — 

 In the i)reparation of bacterial antigens little attention has been 

 paid to such factors although several have called attention to errors 

 m immunological studies due to the fact that the bacterial antigens 

 contained substances from the culture media that led to erroneous 

 conclusions. 



This possibility of error in immunological research is indi- 

 cated by the observation of Sordelli and Mayer (1931) reported 

 by Heldelberger,* "that a 1:400,000 solution of agar (gelose) is 

 enough to yield precipitates with antityphoid and antianthrax 

 sera." Heidelberger suggests that this may account for cross- 

 reactions reported by many w^orkers. 



Early Work on Antigenic Fractions. — Perhaps one of the earli- 

 est studies on antigenic fractions of bacteria was that of Koch 

 (1891, 1901) who found that extracts of the tubercle bacillus pro- 

 duced specific reactions when injected into tuberculous animals. 

 He regarded the extracts as containing an active principle of pro- 

 tein nature which he called tuberculin. 



Subsequent Investigations. — This has led to extensive investi- 

 gations by Long and Seibert (1926), Johnson and Coghill (1926, 

 1931, 1935), Anderson (1927, 1931, 1933), Zinsser and Rice and 

 Reed (1932), Dienes (1929, 1930) and Dienes and Freund (1926), 

 as well as many others, on the chemical and antigenic components 

 of the acid-fast group, especially the various strains of the tubercle 

 bacillus. 



Properties of Tuberculin. — Seibert (1926, 1933, 1934) suc- 

 ceeded in crystallizing tuberculin and studying some of its prop- 

 erties. The crystals, according to Long (1928), are water-soluble, 

 take a methylene blue stain, give the biuret and Millon tests and 

 are heat coagulable. The properties of a new Purified Protein 

 Derivative of Tuberculin prepared by Seibert (1932, 1934) are 

 discussed in Chapter XXVI. 



♦Heidelberger: Ann. Rev. Biochem. 1: (j!j~>, 19.32. 



