328 On Certaiii Bacteria ft^om the Air of New York City. 



" spontaneous variations," occurring in different individuals in'the 

 same stock under essentially indentical conditions, such as we see 

 amons: the hisfher oro-anisms and the mode of occurrence of which 



O O i—? 



has been explained by Weismann as the effect of amphimixis, or, 

 as would apply more exactly to this case, of bud variation b}- an 

 ^'abnormal differential nuclear division '' (Germ-plasm, p. 442). 



Bacillus lactis erythrogenes was selected as the subject of vari- 

 ation experiment. I have shown in a short paper read before the 

 New York Academy of Sciences* that many closel}^ allied and 

 variable forms are to be met with in nature, some of them clearl}- 

 to be referred to this species, others doubtful. A knowledge of 

 the degree of spontaneous variation of this germ was very desir- 

 able in the special relation of determining the standing of these 

 forms, as well as in the general one of the nature of variation 

 in bacteria. It was thought that light might be thrown on both 

 subjects by a study of this species. 



Spontaneous Variation. 



1 — Slight Continuous Variations. 



The variations were tested as to the liquefaction of gelatin, the 

 coagulation of milk, the reduction of nitrate to nitrite, and the 

 amount of pigment produced in an agar culture. 



B. lactis erythrogenes commonl)' produces a quick liquefaction 

 of gelatin : it forms a soft flakv coagulum in milk which on boil- 

 ing has no consistence and which is gradually dissolved ; it re- 

 duces the nitrate solution f partialh' so that the test gives a faint 

 red color, about half wa^- between the deepest tint that can be 

 produced and no color ; in other cases completely ; in an agar cul- 

 ture, the mass of growth is 3'ellow and a pink tint is seen in the 

 medium. 



A culture was selected for experiment, taken from the air, in 

 which the liquefaction of gelatin was rapid (normal), action on 

 milk mormal, the pink tint good, the yellow paler than usual, 

 somewhat whitish, and nitrate scarcely' reduced at all, in twenty- 

 eight days only showing a faint trace of color with the test. 

 From the culture a series of gelatin plates were made by dilutions. 

 From one of the plates, separate cultures were made in ten tubes, 



* Published in the " Transactions " for 1895. 



t The formulas for all media used are given at the end of this article. 



