UNIVERSITY OF CALIFORNIA PUBLICATIONS 



IN 



AGRICULTURAL SCIENCES 



Vol. 3, No. 9, pp. 243-270, 2 text figures June 22, 1918 



VARIABILITY IN SOILS AND ITS SIGNIFI- 

 CANCE TO PAST AND FUTURE 

 SOIL INVESTIGATIONS 



I. A STATISTICAL STUDY OF NITRIFICATION IN SOIL 



BY 



DEAN DAVID WAYNICK 



*KY 



' 



It is very generally recognized that different soils vary widely as 

 regards their physical, chemical, and biological nature. It has also 

 been recognized among soil investigators, at least, that different samples 

 of the same soil type taken from a comparatively limited area may 

 show considerable variation among themselves if we apply quantitative 

 measurements to the various constituents of the soil mass. In any small 

 area of the size usually employed in field experiments, these variations 

 seem to have been regarded as of such limited magnitude as to be 

 worthy of but secondary consideration. In many instances, a single 

 sample from such an area has been taken and the assumption made 

 that it represented the entire soil mass of the depth to which it was 

 taken. In other words, the soil has been considered as a constant to 

 which no corrections need be applied. Most workers in the field of 

 soils have been content with taking relatively few samples and regard- 

 ing determinations made upon the composite of these samples as accur- 

 ate within the limits of error of the experiment. That the variations 

 between different samples taken from a small area may be of such 

 magnitudes as to bring experimental data obtained with one or a 

 limited number of samples into very serious question, or even to 

 invalidate such data entirely, seems not to have been considered. It 

 is the purpose of this paper to emphasize the importance of this phase 

 of soil investigation as regards both past and future endeavors. 



It is obviously impossible, within the limits of a single paper, to 

 consider the variations which characterize all the constituents of any 



