January 16, 1914] 



SCIENCE 



79 



like those with the paratyphoid strains 

 which have just been discussed, but lead- 

 ing to the ultimate separation of two 

 strains of widely different fermentative 

 power. Rettger and Sherrick (1911) re- 

 port similar results in regard to the pig- 



T 



\JY 



r~" — - 



Fig. 1. Acid Production by Paeatyphoid 

 Bacilli. (I.) Distribution of 100 original cul- 

 tures, Type A. (II.) Distribution of 100 subcul- 

 tures of descendants of maximum strain, Type A. 

 (III.) Distribution of 100 subcultures of descend- 

 ants of minimum strain, Type A. (IV.) Distribu- 

 tion of 100 original cultures. Type B. (V.) Distri- 

 bution of 100 subcultures of descendants of maxi- 

 mum strain. Type B. (VI.) Distribution of 100 

 subcultures of descendants of minimum strain. 

 Type B. 



ment-producing power of some of the red 

 chromogens and the resistance to mercuric 

 chlorid of the Aurococcus; and in their 

 paper the similar results of earlier obser- 

 vers are discussed. Such variations as 

 these strongly suggest the pure lines of 

 Johannsen and may perhaps for conveni- 

 ence be designated as pure line variations. 

 A distinct type of variation is the mu- 

 tation, a definite sharp variation sepa- 

 rated by a wide gap from the normal char- 

 acter of the type, arising spontaneously in 



a certain regular proportion of the indi- 

 viduals of a race. Barber (1907) has 

 shown for example that peculiar filamen- 

 tous chains of cells occur, rarely, but with 

 considerable regularity, in cultures of 

 B. coll, and that by isolating these filaments 

 and breeding from them he could get a new 

 race, constantly showing the filamentous 

 arrangement and possessing definite 

 cultural characters and a fermentative 

 power considerably higher than the nor- 

 mal. Similar, though less conclusive re- 

 sults were obtained with B. typhi, and in 

 one case an apparently non-spore-forming 

 race was derived from B. megatherium. 

 A particularly interesting case is that of 

 the fermenting mutants of the typhoid 

 group first observed by Neisser and Massini 

 and recently thoroughly worked out by 

 Penfold (1912). With a number of differ- 

 ent strains of typhoid and paratyphoid ba- 

 cilli and a number of different sugar media 

 it has been shown that an organism which 

 does not ferment the carbohydrate in ques- 

 tion may produce on media containing it 

 colonies which after several days bear curi- 

 ous raised papillae. Subcultures from these 

 papilla yield a pure culture of a strain 

 which resembles the parent stock in every 

 respect, except that it actively ferments the 

 specific carbohydrate and forms no papiUae ; 

 and this mutant breeds true. On the other 

 hand subcultures from other parts of the 

 parent colony produce strains which, like 

 the original stock, do not ferment en masse, 

 but do possess the property of throwing off 

 fermenting mutants. Clark (1913) has re- 

 cently shown that some at least of these 

 modifications may have been quantitative 

 only rather than qualitative, but the sharp- 

 ness of the difference involved seems to 

 warrant their recognition as true muta- 

 tions. 



Either fluctuations or mutations may 

 originate as a result of protoplasmic in- 



