VARIATIONS IN VIRULENCE OR TOXIGENIGITY 307 



tioii is usually considered to be the result of selection of a few virulent organisms, 

 which are either present in, or develop from, the injected culture. 



Zelle (1942), working with a strain of Sahn. typhi-murimn, has recently provided 

 evidence of discontinuous variations of S and R forms, and their selection by the environ- 

 ment in the infected host. For example, an unstable S variant was shown, by a micro- 

 technique of separating cells as they divided in vitro, to throw stable R mutants. In one 

 experiment, the division of an S ceU, the unstable variant, into an S and an R daughter 

 cell, was observed directly. The R mutation was observed twice in 296 divisions, a rela- 

 tively high mutation rate which conforms to the hypothesis already noted, that instabihty 

 is a manifestation of high mutation rates of the order of 1 per cent. Mixtures of small 

 numbers of stable virulent variants with large numbers of less virulent stable variants 

 were injected into mice, the organs of which after death yielded a culture with an increased 

 proportion of the virulent variants. In other tests, virulent and avirulent variants were 

 injected into inbred strains of susceptible and highly resistant mice. It was hoped to 

 demonstrate a differential effect of resistant and susceptible host environments on the 

 variants, such as a reduction in virulence of highly virulent strains propagated in the 

 presumably less selective tissues of the susceptible mice. The only change observed, 

 however, was an enhancement of virulence of certain variants, and this occurred both 

 in susceptible and in resistant mice, showing that whatever the general susceptibility of 

 the mice, any less virulent mutant, thrown by an injected strain of given virulence, was 

 always more readily destroyed than the bacteria from which it arose. 



Non-toxigenic strains of C. diphtherice, CI. tetani and other normally toxigenic ' 

 species have been frequently described. There is no reason to suppose that this loss 

 of toxigenicity is in any way related to the S — > R variations, since there is 

 no evidence that the production of a filtrable toxin is affected by the presence or 

 absence of the smooth somatic antigen. 



REFERENCES 



Alloway, J. L. (1932) J. ezp. Med., 55, 91 ; (1933) Ibid., 57, 265. 



Arkwright, J. A. (1920) J. Path. Bad., 23, 3.58 ; (1921) Ibid., 24, 36 ; (1924) Brit. J. 



exp. Path., 5, 23. 

 Avery, 0. T. and Heidelberg er, M. (1923) J. exp. Med., 38, 81. 

 Avery, 0. T., MacLeod, C. M., and McCarty, M. (1944) J. exp. Med., 79, 137. 

 Badger, E. (1944) J. Bad., 47, 509. 



Baerthlein. (1912a) Zbl. Balct., 66, 21 ; (19126) Ibid., Ref., 54, 178. 

 Barber. (1907) Kansas Univ. Sci. Bull., 4, No. 3, 1. 

 Benecke, W. (1909) Z. indukt. Abstamm. -u. VererbLehre, 2, 215. 

 Bhatnagar, S. S. (1940) Indian J. med. Res., 28, 1, 17. 



Bigger, J. W., Boland, C. R., and O'Meara, R. A. Q. (1927) J. Path. Bad., 30, 261. 

 BiRCH-HiRSCHFELD, L. (1935) Z. Hyg. InfeklKr., 117, 626. 

 Bon^iN, A. (1939) G. R. Soc. Biol., 132, 370. 



BoiviN, A. and Mesrobeanu, L. (1936) C. R. Acad. Sci., 203, 1402. 

 BoiviN, A., Mesrobeanu, L., Magheru, G., and Magheru, A. (1936) G. R. Soc. Biol., 



121, 169. 

 Bordet, J. and Renaux, E. (1930) Ann. Inst. Pasteur, 45, 1. 

 Bordet, p. (1930) Ann. Inst. Pasteur, 45, 26. 

 Braun, H. (1918) Berl. klin. Wschr., 55, 637. 

 Bruner, D. W. and Edwards, P. R. (1941) J. Bad., 42, 467. 

 BuRK, -^A. (1908) Arch. Hyg., 65, 325. 

 BuRRi, R. (1910) Zbl. Bakt., lite Abt., 28, 321. 

 Chandler, C. A., Fothergill, L. D., and Dingle, J. H. (1937) J. exp. Med., 66, 789 



(1939) J. Bad., 37, 415. 

 Craigie, J. and Brandon, K. P. (1936) J. Path. Bad., 43, 233, 249. 

 Dawson, J. and Happold, F. C. (1943) Biochem. J., 37, 389. 

 Dawson, M. H. (1930a) J. exp. Med., 51, 99; (19306) Ibid., 51, 123. 

 Dawson, M. H., Hobby, G. L., and Olmstead, M, (1938) J. infect. Bis., 62, 138. 



