VIABLE BACTERIA IN YOUNG CULTURES 



429 



The error of comparable dilulioiis 



In order to form an estimate of the aetual error of the complete 

 process, 8 experiments were performed in each of which two sepa- 

 rate counts of the same emulsion were made under exactly com- 

 parable conditions. In each count three tubes were put up and 

 the arithmetic means of these were compared, and the percentage 

 deviation of each set from the arithmetic mean of the two was 

 calculated. The results are shown in table 6. 



The results are sufficient to show that the error involved in the 

 viable count performed by the method described is probably 



TABLE 6 



somewhere in the neighborhood of 5 per cent. For practical pur- 

 poses this seemed to be sufficiently small to render the method 

 applicable to the study for which it was designed. 



Before passing on, however, it is interesting to note that the 

 actual counts obtained of the number of living organisms in a 

 broth culture of Bad. suipesiifer have been found to be consider- 

 ably higher than those of many previous observers. Penfold, for 

 instance, working with Bad. coli in peptone water, records a 

 maximum of not more than 9,500,000 organisms per cc, while 

 by the method employed in this work, counts of over 700,000,000 

 organisms per cc, have been met with. Wliether such a differ- 

 ence can be ascribed merely to the nature of the medium and to 

 the method employed in counting must be left an open point. 



