TECHNIC BACTERIOLOGICAL EXAMINATION OF SOILS. 285 



the samplers need not be sterile if they are clean and that the contamination 

 in the lower dilutions from clean dilution bottles would be negligible. 



As long as clean does not mean the same to all workers and just as long 

 as we will admit that unless apparatus is sterile we do not know just how great 

 or of what kind the contamination is — sterile utensils and glassware should 

 be used for the crudest of tests and are absolutely necessary for investigational 

 and research work. 



Sampler 



The reasons for using the sampler have already been given. The facts 

 (1) that this apparatus samples as accurately under sod as under clean 

 cultivation; (2) that it does not destroy the cultural practices, and (3) that 

 it may be used with as much accuracy and safety near a tree, slu'ub, or bush, 

 as in the open field are emphasized. 



Field Versus Air Dry Samples 



[f aliquots for bacteriological analysis are taken from air dry samples a dis- 

 cussion of methods and a technic for sieving or grinding the air dry samples 

 would be in place. Air dry samples cannot properly be used to determine the 

 bacteria present in field soils unless ic is proven by careful investigation that 

 changes in moisture, in aeration, and in temperature have no effect on the 

 bacterial content of the soil. 



Size of Aliquot of Soil 



A small sample of soil is not representative. Soils are not composed of 

 equal sized particles or of particles of the same material. Granting that a 

 given soil contained particles all of the same composition and which were 

 non-porous, the variation in the area of the soil particles of two different 

 chance aliquots might be as much as the area of the particles of one of the 

 aliquots. Add the factor of different kinds of particles to that of variation 

 in size and it is safe to say that no two samples of soil are exactly alike. If 

 the size of the aliquot of soil taken for analysis is decreased beyond a certain 

 point, a small proportion of the larger, rock particles must necessarily account 

 for a larger per cent of the weight and cut down the area of the soil particles 

 in the aliquot. 



If air dry samples could be used the aliquot required to be representative 

 would be smaller than that required from a field sample. It has not been 

 proven advisable to use au- dry samples and so aliquots of the moist field 

 soil are used. Dr. P. E. Brown (4) takes on one hundred gram aliquots of field 

 soil from which to make bacterial dilutions, while Dr. H. Joel C'onn (.5) uses 

 one-half gram aliquots of field soil from which to make bacterial dilutions. 



The authors of this paper have investigated the amount of soil to use to 

 get representative counts and summarize their results as follows : 



8432 — 19 



