July 15, 1915 Methods of Bacterial Analyses of Air 363 



COMPARISON OF VARIOUS MEDIA FOR BACTERIAL ANAIvYSIS OP AIR 



In the course of this work several comparative series of tests were 

 made to determine whether or not a more suitable medium than ordinary 

 agar might be found. The media so compared with nutrient agar were 

 as follows: (a) Nutrient gelatin, containing the same ingredients as the 

 agar except for the substitution of 10 per cent of gelatin for 1.5 per cent 

 of agar; (b) a gelatin medium with 20 per cent of gelatin; (c) a gelatin 

 medium made with soil extract (4) ; (d) lactose agar, containing i per 

 cent of lactose in addition to the usual ingredients; and (e) asparaginate 

 agar (4) , an agar to which only compounds of known chemical composition 

 have been added. 



The problem was not studied exhaustively enough to prove that the 

 agar medium used in this work was the best possible medium for air 

 work, but the data gathered did warrant the conclusion that it was better 

 than any medium compared with it. The last two media mentioned 

 above gave only slightly lower results than the nutrient agar, but the 

 three gelatin media gave decidedly lower results, probably due in part 

 to the lower incubation temperature necessary when using them and in 

 part to liquefaction. The 20 per cent gelatin gave the highest results 

 of the gelatin media, owing to diminished liquefaction of the medium. 



Comparison Between Two Methods of Measuring Bacterial Precipitation 



In the study of the main problem already mentioned, to which this 

 study of technique was preliminary, it was necessary to determine the 

 number of bacteria precipitated upon a given area in a given time. 

 No standard procedure is given for this by the Committee on Standard 

 Methods of Air Analysis, although it seems evident that there should be 

 such a recognized procedure. This determination is usually made by 

 exposing a Petri plate containing solidified agar or gelatin for a given 

 period of time and counting the colonies that develop on the plates 

 after incubation. 



It was felt, however, that this method was entirely inadequate, as it 

 does not give a true measure of the number of bacteria falling on the 

 plate. This comes about because of the possibility that a dust particle 

 may carry more than one bacterium. Ordinarily but one colony devel- 

 opes from a dust particle, so that the number of colonies measures the 

 number of bacteria-laden dust particles falling on the plate rather than 

 the number of bacteria falling on the plate. 



The method for determining this bacterial precipitation which was 

 tried was to place 500 c. c. of sterile water in a sterilized pail covered 

 with a metal lid. Check samples were then taken and the lid removed 

 for a given length of time. Samples of the water were then taken after 

 thorough agitation and plated as soon as possible. Later it was dis- 

 covered that Koning (16, p. 251 et seq.) had used milk as the medium 



