PKODUCTION OF BACTERIAL RACES 117 



into actual spirilla. This polymorphism and its extent depend 

 (as I have already shown in the case of the Bacillus pyocyaneus) 

 upon the composition of the medium. Upon returning such 

 altered microbes back to the potato there is a gradual return to 

 the original state. Yet it is possible by slight modification of 

 this method to produce permanent races. Taking beef-broth 

 to which tartaric acid has been added, Wasserzug found that 

 when the culture was a day or two old there was a great develop- 

 ment of spirilla. Later, when through the production of tri- 

 methylamin the medium became alkaline, the spirilla ceased 

 to be developed ; there was a return to the coccus or shortened 

 bacillary form. If now he did not permit the production of 

 the alkaline trimethylamin, but made new acid growths before 

 the cultures were more than forty-eight hours old, he discovered 

 that the longer he persevered with his series of cultures the 

 longer the microbe took when grown upon potato to return to 

 the coccus form, until finally by such acid growth and subsequent 

 heating to 50° C. for a few minutes, all that he could obtain upon 

 sowing on potatoes was, not a micrococcus, but a long bacillus — 

 a permanent race of such. 



Or let us return again to the Bacillus pyocyaneus, which, 

 though pathogenic, may be discussed along with other chromo- 

 genic forms. 



By placing his modifications, to which I have already referred, 

 under special conditions of heat, etc., Gessard found that he 

 could obtain permanent races upon ordinary beef-broth, one 

 giving the green fluorescent pigment alone, another the blue 

 pyocyanin, another colourless. 1 The same observer, studying 

 the bacillus of blue milk (Bacillus cyanogenus or syncyanus), 

 found that by passages through egg albumen he could obtain a 

 race giving a deep pigment (which becomes red with alkalies, 

 blue with acids), another race which is ordy fluorescent, gained 

 from cultures that are some months old, and a third colourless 

 race, this last either by taking extremely old' cultures or by 

 warming a recent culture to a temperature which does not 

 cause death. 



Similarly Laurent, 2 working with the Bacillus ruber from 

 Kiel (Bacillus rouge de Kiel), found that exposure of a growth 

 to bright sunlight for three hours gave colonies which with but 

 1 Loo. tit. * Loc. cit. 



