344 On Certain Bacteinafrom the Air of New York City. 



the most part when the media used are taken into account. 

 Gelatin is to be regarded as liquefied even when the fluid evap- 

 orates as fast as formed and only a dry hollow results. Conse. 

 quentl}', if an}- hollow is formed bv the growth mass, the gelatin 

 is to be considered liquefied. 



In the descriptions I have not repeated the characters given in 

 the sj'n optic table except where necessary' to enlarge on them. 



The Characters for which the Several Media are used. 



It is important that the number of media should not be unne- 

 cessarily multiplied, or else the labor of determining species will 

 become too great. The following shows the characters which I 

 have obtained from the several media used, and how labor may 

 be saved in some instances : 



Broth. This medium may usuall}' be omitted. The characters 

 to be obtained from it are good ones, but can usually be detected 

 as well in the fermentation tube and water of condensation in the 

 agar cultures. 



Gelatin. Only used for the liquefaction test and the characters 

 of colonies on plate. I have found it practicable to dispense 

 with a separate planting on simple gelatin 113^ preparing a lactose- 

 litmus gelatin, which combines the characters of liquefaction and 

 acid or alkali formation in one medium. I have disregarded the 

 decoloi'izing effect on litmus so often seen, as I have not found 

 the characters reliable on solid media. 



Agar. Used for the growth forms and necessary. Glycerine 

 agar is less important, and I have not used it for diagnostic 

 purposes. 



Fermentation broth in bent tubes is necessary for the gas forma- 

 tion test and relation to air. The gas formation is often indicated 

 on lactose-litmus gelatin, and in such cases scarcely needs con- 

 firmation on this medium. 



Milk for the eftects of coagulation. It should always be boiled 

 before the test can be considered complete. 



Nitrate solution. The test for nitrates ma}' be applied bj'^ 

 adding two to five drops of napthylamine sulphate, a small cr3'S- 

 tal of sodium sulphanilate and boiling the medium to hasten the 

 reaction. When cool it is read}' to examine for the color. I do 

 not test for ammonia. 



