542 '^he Philippine Jownal of Science wis 



Miiller ^' describes a strain of the Flexner type which produced 

 secondary colonies capable of fermenting isodulcite. 



Bernhardt '" investigated a strain of dysentery bacilli, type Y, 

 which formed secondary colonies on maltose agar capable of 

 fermenting maltose. The new characteristic was transmissible 

 when cultures were transferred at short intervals, but, after 

 eight weeks' growth on weakly alkaline agar or after half a year 

 in sealed agar tubes, the power to acidify maltose was lost. On 

 transferring to fresh agar, new secondary maltose-fermenting 

 colonies were formed. 



Baerthlein,''' in an extensive article on the so-called mutations 

 in various bacilli, vibrios, and cocci, describes secondary colo- 

 nies in strains of dysentery bacilli of both the Shiga-Kruse and 

 of the nontoxic varieties. Secondary colonies possessing new 

 morphological as well as fermentative qualities were obtained, 

 although there was no constant correlation between the mor- 

 phological and chemical changes. As a rule, secondary colonies 

 were formed on transfer of old agar cultures to new media. 

 Some mutants of the same strain of toxin-free dysentery bacilli 

 possessed varying agglutinability to the same serum. In some 

 cases the same strain gave two or more sorts of mutation. 



Practically all of the above-mentioned writers obtained sec- 

 ondary colonies only in media containing the sugar fermented 

 by the new race, and in nearly all cases the new race transmitted 

 its characteristics indefinitely to offspring. In most cases the 

 purity of the cultures used was established by plate cultures only, 

 but in some the varying strain was started from a single bacillus 

 obtained by the Burri method. 



In working out the maltose-fermenting variety of dysentery 

 bacilli observed by me, it was proposed to examine the variations 

 carefully, not only beginning with a single cell, but testing 

 numerous individuals, both of the stock and of the new races, 

 by this method. It was recognized that dysentery bacilli of the 

 Flexner type are relatively inconstant in their characteristics, 

 and that maltose is a comparatively unreliable substance with 

 regard to fermentation ; but the characteristics of the races ob- 

 tained were so well marked and constant, that they could be 

 safely taken as a basis for a study primarily of variation. 



First, the uniformity of the individuals of the red and the blue 



" Centralbl. f. Bakt. etc., Ref. (1908), 42, Beiheft, 57. 

 '^ Zeitachr. f. Hyg. u. Infectionskrankh. (1912), 71, 229. 

 " Arh. a. d. kais. Gesnndheitsamte (1912), 40, 433. 



