Since the early days of immunology, it has been considered that high virulence and recency of isolation from pathological material were essential attributes of strains of bacteria to be used in production of superior immunizing agents. In spite of a surprising paucity of supporting factual information this concept has pesisted.Perhaps in no field of investigative immunology has the theory of an essential relationship of virulence to immunogenicity been accepted more widely than it has been in relation to the development of immunizing agents against typhoid. As early as 1910, Leishman (1) pointed out that employment of highly virulent cultures of Salmonella typhosa appeared to have certain theoretical advantages in the preparation of typhoid vaccines. Perry, Findlay, and Bensted (2), who employed active-immunity protection tests on mice, concluded that the more highly virulent the strains of S.typhosa employed in preparing vaccines, the greater was the immunizing activity of the resultant products. Similar conclusions were reached by Arkwright (3), Grinnell (4), and Brown (5). In many of these earlier investigations, however, virulence and avimlence was not dearly distinguished from smooth-rough variation of strains; in fact, virulence and smoothness (also avirulence and roughness) generally were considered essentially synonymous.One of the most comprehensive studies of the relationship of virulence to antigenicity of S. typhosa cultures was conducted by Siler and his associates at the Army Medical School (6). This investigation included a study of seven strains of varied virulence and it was found that, as measured by active-immunity mouse-protection tests and by "0" agglutinin and mouse-protective antibody responses in human volunteers, vaccines prepared from virulent strains were more highly immunogenic than were those prepared from avirulent strains.Felix (7) likewise reported that highly virulent strains of S. typhosa exhibited the greatest immunizing activity. He placed major emphasis on the r61e played by Vi antigen both in virulence and in immunogenicity. There is, however, no unanimity of opinion in regard to any essential relationship between virulence and immunogenicity. This is evidenced more by the development of certain immunizing agents and methods than by any specific investigations of the phenomenon itself. It is not difficult to understand why there is no unanimity of opinion in regard to the supposed mutual interdependent relationship of virulence and immunogenicity since there has been a marked lack of uniformity in the experimental techniques employed in the various investigations and most of the published