346 Method of estimating Bacterial Density 



but could not be separated from other colonies in counting the plate. 

 The counts on these plates are therefore excessive. The presence of this 

 organism on the bottom of the plates has thus produced an abnormal 

 variation in the whole series. It will be seen that, if plates on which it 

 occurs are ignored, the y^ index for the remaining 22 plates falls within 

 the expectation of random samphng. 



A pure culture of this organism was obtained and a plating from a 

 sample of Kothamsted soil was made, a small loopful of suspension of 

 the organism being added to the first dilution flask. Table XIV, Series A, 

 shows the colonies developing on six parallel plates of the soil thus 

 treated, compared with a control series of plates of the same soil not 

 inoculated, Series B, which were made at the same time. 



Table XIV 



Effect of Leeds soil organism on colony development from suspension of 



Rothamsted soil 



In this case the organism formed a spreading growth over the bottom. 

 The area of this spreading growth, where it occurred, was measured and 

 is shown in Table XIV. It will be seen that the reduction in colony 

 development is clearly related to the amount of spreading growth. In 

 this series of plates it is also evident that the variation is greatly increased 

 by the occurrence of the organism on certain of the plates. 



From an abnormally variable series of plates of Rothamsted soil a 

 second organism has been isolated, whose frequent habit it is to spread 

 on the under surface of the agar, and which has a similar inhibitory 

 action on the development of other colonies. Table XV shows two sets 



