20 PRINCIPLES OF SOIL MICROBIOLOGY 



taken from a thorough mixture of several representative borings to the 

 same depth in various parts of the field. This may be sufficient, if the 

 probable error involved in the determination is known. A study of the 

 variability of the soil in relation to the numbers of bacteria, as deter- 

 mined by the plate method, was found 34 to yield in one particular plot 

 eight to twenty-five millions per gram of soil. These results were based 

 on fifty-one samples of soil taken from one-twentieth of an acre; the 

 samples were taken at intervals of 5 by 8 feet. The numbers of bacteria 

 of practically a whole series of plots variously treated and where definite 

 bacteriological differences have been established were found to fall 

 within that range of numbers. This would make any results obtained 

 from a determination of numbers of bacteria based upon a single sample 

 practically valueless. Before any definite conclusions can be drawn 

 from a determination of bacterial activities in the soil, it is advisable to 

 study the variability of the particular soil. This is done by taking a 

 large number of samples from a particular field and working out the 

 probable error of the particular variable which should not be greater 

 than Em = 2.5 per cent. 



For practical purposes, five composite soil samples each consisting 

 of three to four borings taken from different parts of the field will give 

 good results not only in the study of numbers of microorganisms but 

 also in the study of specific physiological activities of soil bacteria. 



In taking the samples, a small amount of soil (j inch) is scraped away from 

 the surface by means of a spatula. The samples are taken by means of sampling 

 tubes or augers, carefully cleaned previously, to a depth of 6 to 6| inches, unless 

 a study is made of the distribution of microorganisms at different depths. In 

 that case a ditch may be made 3 feet long by 1 foot wide and, after scraping off the 

 surface soil, at the desired depth, by means of a spatula, the samples are taken; 

 small metallic sampling tubes (about 1 inch in diameter) made of iron or copper, 

 with sharp ends may also be used for this purpose. 



Care is taken not to contaminate the soil samples with other soil or 

 by exposing them too long to the atmosphere. However, no absolute 

 sterility in the process of sampling is required, since the numbers of 

 microorganisms in the soil are very large in comparison with any possible 

 contamination from a brief exposure. The samples are placed in 

 sterile sampling bottles and brought to the laboratory as quickly as 

 possible. A rapid change was found 35 to take place in the numbers of 



34 Waksman, S. A. Microbiological analysis of soil as an index of soil fer- 

 tility. I. The mathematical interpretation of numbers of microorganisms in the 

 soil. SoilSci. 14: 81-101. 1922. 



88 Frankel, 1887 (p. 12). 



