On Certain Bacteria from the Air of New York City. 339 



no further wliitenino; of the sfrowth. It was noticeable that each 

 culture grew less vigorousl}' than the preceding while in the 

 incubator, though it seemed to revive fidly when removed to the 

 room temperature. Culture No. 1 was of a distinctly more yel- 

 lowish shade than No. 6, and not whitish at the edge of the growth. 

 A culture made from it, grown two days in the incubator, and 

 transferred to the room temperature, gave a paler yellow ; one 

 from this, about the same tint ; the next generation was less 

 uniform, somewhat spotted with white and yellow, and the next 

 generation was distinctly spotted. The same debilitating effects 

 of the repeated exposure to the high temperature was noted here 

 as with No. 6. The last, most spotted (^. e., least yellow) culture 

 was examined as follows : 



(1) It was allowed to grow well for several days at the room 

 temperature. It was then seen that the yellow parts of the 

 growth were less active in extending their borders than the white 

 parts were, which resulted in an uneven outline of the growth 

 mass, the yellow fans terminating at the incisions. (2) A set of 

 agar plates was made from this growth. Of the resulting colonies, 

 some were nearly white, others distinctly 3'ellow. It was noted 

 that, on the whole, the yellow colonies were smaller. None of 

 the largest were yellow, though some of the yellow were as large 

 as some of the white ones. (3) Cultures were made from a white 

 and a 3'ellow colony- respectively, from one of these plates, and they 

 were planted on milk, nitrate solution and lactose-litmus agar. 

 On all these media the white culture advanced more rapidly' than 

 the 3'ellow one ; the gelatin was slightl3^ but decidedh' more 

 quickl3^ liquelied, the milk more quickly coagulated and the 

 nitrate more strongl3' reduced in the same time and under identi- 

 cal conditions. 



From the above I conclude that the white form is less injuriously 

 affected by the abnormal condition of increased temperature than 

 the yellow from is, and the apparent effect of the temperature in 

 producing a white form in the first instance ma3- have really 

 been due to a process of selection, the white from growing the 

 faster and tending to supplant the other. Thus the ordinary 

 course of transference of a small part of the growth on the inocu- 

 lation needle would be the agent in this selection. Now it is a 

 fact that the different cultures of B. lactis erythrogenes obtained 

 from ditterent sources varv a good deal in tint, and I have indi- 



