jtiiyis, 191S Methods of Bacterial Analyses of Air 365 



made from the pails. Frequently, the colonies that appeared on the 

 plates which had been exposed showed by their very nature that they 

 were of composite origin. 



It seems strange that this fundamental weakness of the plate-exposure 

 method has not been properly appreciated by investigators, for it was 

 recognized by Hueppe (14) as long ago as 1891. 



This work shows that the figures obtained in all of the air investiga- 

 tions where conclusions are based upon results obtained by the plate- 

 exposure method are not nearly so large as they should have been. 

 The same criticism applies to some of the methods that have been sug- 

 gested as a means of counting bacteria in air. One of the first to use 

 this faulty principle of regarding bacteria-laden dust particles as equiva- 

 lent to individual bacteria was Hesse (13). He counted the colonies 

 developing on gelatin after a measured volume of air had been drawn 

 over it in such a way as to catch the dust particles on the gelatin. The 

 idea that the number of colonies developing on the surface of a solid 

 medium after exposure to the air really represents the number of single 

 bacteria deposited seems also to have been held by some of the later 

 investigators, among them being Harrison (12), Russell (23), and Wins- 

 low (33). 



CONCLUSIONS 



It seems reasonable to conclude that the nature of the filters tested 

 had little influence on the results secured in duplicate analyses — that is, 

 those obtained where a sand and a liquid filter were used side by side 

 agreed just as well as those where either two sand filters or two liquid 

 filters were used side by side. 



It was found that the particular form of sand-filter aeroscope recom- 

 mended by the committee on standard methods of bacterial air analysis 

 appointed by the American Public Health Association varied in its 

 filtering efficiency from 50 to 100 per cent, with the average efficiency 

 for two series of tests of 90 and 91.6 per cent. It is believed that the 

 chief cause of error with this form of aeroscope arises from the fact that 

 it is so constructed that it must be sterilized with steam, which causes 

 caking of the sand-filtering layer. 



A description is given of a modification of this form of aeroscope, so 

 constructed that it may be sterilized with dry heat. The modified 

 standard aeroscope was found to retain nearly 100 per cent of the bacteria, 

 with little chance of error. It was also found to be cheaper, less breakable, 

 easier to operate, and more adaptable to field work than either the 

 standard sand aeroscope or the aeroscope recommended by Rettger. 



The latter can be made to yield excellent results, provided sufficient 



care is exercised in handling it. Its use, however, is attended with a 



number of difficulties, among which may be mentioned its tendency to 



leakage about the rubber stoppers after being sterilized, the foaming of 



92315°— 15 7 



