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



barring the possibility of some metliocl of contamination whieli I 

 have not thought to sfuard ao;ainst.* 



Summary of the Variations of Bacillus lactis erythrogenes. 



Form 1. (Normal) Soft, smooth growth in colonies and in tube. 



Form 2. Soft and smooth, but the colonies with lobed edges ; 

 growth in tube finely creased and milled before the edges. 



Form 3. Rather crusty, the colonies even, flat, granular centrally 

 with a well defined rim-like edge ; in tubes a flat wrinkled and 

 folded growth with irregular edges. 



Form 4. Slightly more skinny than form 3. Colonies and growth 

 on tube alike, very much folded, with irregular edges, tending* 

 to produce a margin of the soft form. (See tlie figure ; also the 

 description of Xo. 107, B. erythrogenes rugatus.) 



Form 5. There is also to be recorded here the granular variety 

 derived from B. helvolus. (See No. 106, B. heloolus granulatus.} 



Note — These are only variations in the form of growth. All the 

 cultures were of the shade of yellow of the particular form of 

 Bacillus lactis erythrogenes from which they were derived, and 

 produced also the pink tint. So far as tested, they reproduced 

 the biogogical characters of their parent culture, which, as pre 

 viously noted, were not in all respects normal. 



Effect of the Environment. 



As an example of the effect of the environment, the action of a 

 change of temperature was selected. It is well known that man}^ 

 chromogenic species, when grown " for several generations " at 

 the body temperature, produce white races, which are more or less 

 permanent. What is the nature of this process ? 



The same rather pale yellow culture of Bacillus lactis erythro- 

 genes was subjected to experiment. The culture Xo. 6 exhibited 

 a pale growth on agar, especially whitish along the edge. A 

 culture from this edge kept at 37^° C. for two days, and then 

 transferred to the room temperature, grew nearly white. Three 

 subsequent "generations " treated in the same manner produced 



* Two more sets of plates were made from the last obtained smooth growth, 

 without further reversion ; all the colonies were alike. But, as the number of 

 colonies in the two platings was 525 and 1685 respectively, whereas the ten- 

 dency to reversion was calculated to have diminished at least to 1 :950 and 1 :1900 

 respectively, the results are not conclusive against the variation hypothesis. 



