340 On Certain Bacteria from the Air of Neio York City. 



cated that there was some spontaneous variation in this respect 

 in my experimental cultures. The variations toward a white 

 form probably tend to become eliminated under natural condi- 

 tions where the 3^ellow form doubtless grows best,* but under 

 the eft'ect of higher temperature there comes I'oom for the action 

 of the selection which I have just described. It may be that this 

 is, at least in great part, the true explanation of the production 

 of the colorless races of chromogenic bacteria, which has been 

 quoted as proof of the "transmission of acquired characters." 



Effects of Differences in the Composition of the Media. 



I need not enter into a long discussion of this subject. It is 

 well known that slight variations in the composition of the media 

 make marked changes in the bacterial growth. I have shown 

 elsewheref the effect of A^aiying degrees of acidity in gelatin ; 

 the difference in the action of m}' selected cultures on the two 

 samples of milk from different sources and with different degrees 

 of acidity, which is tabulated aboAx, is a case in point ; again it 

 was noticed that the chromogenic power of some species was less 

 on my media made with meat extract than it was in the same ones 

 grown on media in which the meat itself had been emplo^^ed. 



These changes are not true variations, as they are not inherited; 

 but I have given the exact composition of all the media used in 

 this investigation, to eliminate this source of error. The receipts 

 will be found at tlie end of the article. 



Conclusions as to Variability. 



From the above set of experiments I infer that it is premature 

 to assume that races of bacteria are produced b^^ the direct action 

 of the environment. Rather the species possess, first a power of 

 continuous variation, producing intergrading varieties, and which 

 under a long process of natural selection is capable of adapting 

 them to various situations or functions; second, a power of con- 

 siderable discontinuous variation, producing "sports," dimorphic 

 or polymorphic forms or races (which may revert to type spon- 

 taneously; but which are distinguishable from true species onlj' 



* In some old cultures of both the white and j^ellow forms, which liad 

 become partially dried, a few lumps of vigorous growth started out, and these 

 were of the yellow form. 



t See Transactions of the New York Academy of Sciences, 1895. 



