648 PROTEUS 



Strains of lower virulence cause chronic inflammatory processes, either of the 

 suppurative or of the infective granuloma type (Larson and Bell 1913, Wenner 

 and Rettger 1919) ; the latter are best seen after intraperitoneal injection. 



The inoculation intraperitoneally of 0-5-1-0 ml. of a 24-hour broth culture of 

 a virulent strain generally proves fatal to rats and mice in 18-48 hours, and to 

 guinea-pigs and rabbits in 1-7 days. Severe infections in rabbits are said to be 

 characterized by extreme emaciation (Larson and Bell 1913). Though invasion 

 of the tissues may occur on parenteral inoculation, Jensen (1913) states that in 

 naturally infected calves, even when the organisms are seething in the gut, the blood 

 and tissues remain sterile. 



Classification.— Hauser (1885) divided the Proteus group into three species : 

 (1) Proteus vulgaris ; Gram-negative, liquefies gelatin, peptonizes fibxin, produces 

 indole, and has a variable action on glucose and sucrose. (2) Proteus mirahilis ; 

 Gram-negative, more highly pleomorphic, and liquefies gelatin more slowly. (3) 

 Proteus zenkeri ; Gram-positive, does not liquefy gelatin, does not form indole, 

 and fails to attack sugars. Proteus zenkeri was found to be very similar to an 

 organism described two years previously by Kurth (1883) under the name of B. 

 zopfii ; as both of these organisms are Gram-positive, they had to be transferred to 

 a separate genus, which is known as Zopfius. 



Hauser's subdivision of the Proteus group on the basis of morphology, rate of 

 liquefaction of gelatin, and indole production, has been found impracticable. The 

 morphology is variable, depending especially on the medium and the age of the 

 culture. The rate of liquefaction of gelatin is likewise variable ; it is rapid with 

 newly isolated strains, and is often much slower, or even absent, with strains that 

 have been long under artificial cultivation. Indole production used to be tested 

 by the nitroso-indole reaction ; but as Berthelot (1914) has shown, this reaction 

 is untrustworthy, and is given by indolacetic acid as well as by indole ; when tested 

 by Ehrlich's reagent, using the ether extraction method, it is found that the results 

 are different, and that indole production is not nearly so constant a feature of 

 Proteus as it was originally considered to be. 



The most striking characteristics of the members of this group are their ability 

 to swarm on solid media, their production of HjS, their decomposition of urea, their 

 liquefaction of gelatin, and their failure to ferment lactose or any of the polyhydric 

 alcohols. It should, however, be added that variants occur which have lost their 

 power of swarming, and that old strains may no longer liquefy gelatin. 



The recent observations of Rauss (1936) suggest very strongly that Morgan's 

 bacillus, which has hitherto occupied an invidious position in the Salmonella group, 

 is closely related to Proteus. Its ability to swarm under suitable conditions, its 

 frequent fermentation of xylose and its occasional fermentation of sucrose, its 

 production of indole and HjS, its formation of alkali in litmus milk, the greater 

 group specificity of its H and the greater type specificity of its antigens, the group 

 relationship of at least one of its H antigens to Proteus, its growth only under 

 favourable conditions in the intestine of human beings, and its general pathogenicity 

 for experimental animals — all bring it closely into line with organisms of the Proteus 

 group. Its failure to liquefy gelatin or constantly to ferment sucrose must be 

 considered in relation to the negative reactions obtained in these two respects with 

 known Proteus strains, particularly those that have been long cultivated in the 

 laboratory. Intermediate types that peptonize milk, and the organisms described 

 by Magath (1928), which were isolated from cystitis and which liquefied gelatin but 



