325 



THE ACCURACY OF THE PLATING METHOD 



OF ESTIMATING THE DENSITY OF 



BACTERIAL POPULATIONS 



WITH PARTICULAR REFERENCE TO THE USE OF 

 THORNTON'S AGAR MEDIUM WITH SOIL SAMPLES 



By R. A. FISHER, M.A., H. G. THORNTON, B.A., 

 AND W. A. MACKENZIE, B.Sc. 



{Rothamsted Experiment Station) 

 (With 2 Text-figures) 



1. Introduction 



The accuracy of the estimates of bacterial density, in samples of soil, 

 water, or other material, obtained by the plating method, is only one 

 of many points which arise in the interpretation of bacterial counts. 

 The full interpretation of such data would include a consideration 

 of the divers species that occur on the culture media, and of the 

 forms in which they exist in the soil. The partial or total exclusion of 

 certain forms, such as anaerobes, that require special cultural con- 

 ditions, must also be considered in a full examination of such data, for 

 a single medium supplies, necessarily, but a single aspect, however 

 comprehensive, of the bacterial flora of the soil. Questions too, as to 

 what is to be considered as the unit of enumeration — the individual 

 organism as it exists in the soil, or possibly groups of such organisms 

 adhering to single particles of soil, and undetached by the processes of 

 sampling and dilution — whatever their importance may be, are not the 

 object of the present investigation. 



For if all these inquiries could be answered with certainty and pre- 

 cision it would still remain to be discovered with what accuracy the 

 numerical estimate of bacterial density, obtained from a single set of 

 plates, represented the actual bacterial density in the sample, and in the 

 material from which the sample was drawn. 



The question of accuracy, therefore, unlike the other elements in the 

 interpretation of bacterial count data, is primarily a statistical question 



