On Certain Bacteria from the Air of New York City. 341 



bv the occurrence of such revision) and ultimateh' species, by 

 the lapsing of the capability of reversion. 



In the following systematic account, I have provisionally 

 adopted the more conservative course of giving all diftering forms 

 specific rank until their exact relations be determined. Many of 

 them will doubtless come to be regarded as races or varieties ; 

 some might properly be so placed now,* Those forms which do 

 not differ except in degree and to no great extent are considered 

 conspecific. 



Unfortunately, we do not seem to have arrived at the point 

 where it is possible to differentiaie readih^ species from races and 

 varieties. This results from the the fact that there is no standard 

 b}^ which to determine which characters are more reliable. If the 

 several forms be divided on a single character ; for example, if we 

 select all yellow species, or all which reduce nitrate or liquefy 

 gelatin, etc., the grouping is different in every instance. There 

 is no indication of the formation of reliable groups by a converg- 

 ence of the the several characters, so that the classification is en- 

 tirely arbitrary-, depending upon which character is given prom- 

 inence. Consequently, I have as 3'et found no characters to form 

 a series of natural groups of species under the genus which has 

 been urged on me b}' Dr. Prudden to be very desirable. 



In this paper, the term " groups " has been used in a specific 

 rather than in a subgeneric sense ; for instance, the "anthrax 

 group " means a series of forms which are probably varieties of 

 one species ; but, as I have given all varietal forms the specific 

 position, I have preferred to use the term as I have, Avith this ex- 

 planation. 



A List of the Common Bacteria of the Air of New York City. 

 The following list embraces all species found more than once 

 in the air in the course of this investigation. Those occurring 

 but a single time will be referred to in the systematic part. The 

 cultures were obtained by exposing gelatin plates for from one to 

 five minutes in various situations. Of the resulting colonies, a 

 portion only were transferred to tube cultures for investigation. 

 The yeasts, Cladothrices, and moulds were disregarded. In num- 

 bers the Micrococci were considerably predominant, but of com- 

 paratively' few species. The Bacilli, though less common, were 



* For the sake of uniformitv this has not been done. 



