136 OXIDATION-REDUCTION POTENTIALS 



the study is extended beyoiid the range in which the flocculation method of toxin 

 estimation, used by Pappenheimer, is possible and the intradermal method of toxin 

 testing is used, then there is immediate departure from the Unear relationship. 

 Furthermore, although several hundred different strains have been studied no other 

 has been found in which there is a linear relationship between toxin production 

 and iron content. None of these other strains showed the abnormally high toxin 

 production of the Park-WiUiams 8-strain which has long been habituated to labora- 

 tory sub-culture. Recently isolated strains do not behave in the same way. 



A further interesting point arises from the effect of iron salts on diphtheria toxin 

 production. Although the lethal effect of C. diphtherice has been traced to the effect 

 of the soluble toxin (Hewitt, 1947, 1948), and full protection of animals against hve 

 cultures is given by administration of the specific antitoxin, yet the clinical severity 

 of the disease has not been correlated with the amount of toxin produced by the 

 strain when isolated in laboratory culture medium. Some of the strains (hke Park- 

 Wilhams 8) are most innocuous clinically, yet produce most toxin in laboratory 

 culture media. It has been suggested that this might be because laboratory culture 

 media for diphtheria toxin production are carefully adjusted to have a low iron content, 

 which favours toxin production, whilst the iron content of the tissues is fairly high 

 and toxin production by different strains may be differently ordered. Although 

 Mitis strains (clinically mild) produce more toxin generally than Gravis strains 

 (generally considered clinically serious) in iron-deficient medium the reverse might be 

 the case in iron-rich media like tissues. This point has been tested by the author in 

 the case ofseveral hundred recently isolated strains oiC. diphtherice, but no correlation 

 was found between the effect of iron and the Mitis-Gravis classification. Excess iron 

 diminished the toxin production of all strains, but Mitis and Gravis strains were not 

 distinguishable. 



Summary of Chapter VI 



The study of the oxidation-reduction conditions established in bacterial cultures 

 has already led to a number of interesting observations. The behaviour of the 

 peroxide-forming bacteria is entirely different from that of the catalase-containing 

 organisms. Some of the former type are, however, deficient in other enzymes than 

 catalase. The relation of peroxide to the obligatory-anaerobic bacteria is discussed 

 and problems of anaerobiosis are reviewed. The different levels of oxidation-reduction 

 potential at which different organisms multiply and the different effects they produce 

 are a reflection of their metabohc effects and the enzyme systems they possess. Some 

 differences in the behaviour of different bacteria, the effects of differences in culture 

 media and cultural conditions and some effects of lytic agents are described. The 

 inter-dependence of oxidation-reduction conditions and bacterial variation in viru- 

 lence and morphology, is discussed. The reducing conditions observed in bacterial 

 cultures are regarded as a necessary result of the metabolic activities of the bacteria, 

 rather than being due to the explosive liberation of reducing substances from the cells. 

 Bacterial cultures are to be regarded as dynamic constantly altering systems and not 

 as static stagnant pools. 



