No.583] ON SUBSTRATUM HOMOGENEITY 



453 



parisons from field to field, variety to variety or character 

 to diameter, may he directly made. It slioul'd also, if pos- 

 sible, offer no difficulties of calculation. 



The criterion proposed is the coefficient of correlation 

 between neighboring plots of the field. An exceedingly 

 simple formula for the calculation of such coefficients has 

 been deduced. 



The method of application of this coefficient is here il- 

 lustrated by four distinct series of experimental data. 



The remarkable thing about the results of these tests 

 is that in every case the coefficient of correlation has the 

 positive sign and that in some instances it is of even more 

 than a medium value. In short, in every one of these ex- 

 perimental series the irregularities of the substratum have 

 been sufficient to influence, and often profoundly, the ex- 

 perimental results. 



It might be objected that by chance, or otherwise, the 

 illustrations ;ne not typical of what ordinarily occurs in 

 plot cultures. But they have been purposely drawn from 

 the writings of those who are recognized authorities in 

 agricultural experimentation, and who have given their 

 assurance of the suitability of the fields upon which the 



For example, Mercer and Hall state the purpose of 

 their research to be, "to estimate the variations in the 

 yield of various sized plots of ordinary field crops which 

 had been subjected to no special treatment and appealed 

 to the eye sensibly uniform." Their mangolds "looked 

 a uniform and fairly heavy crop for the season and soil," 

 while in their wheat field "a very uniform area was se- 

 lected, one acre of which was harvested in separate plots, 

 each one five hundredth of an acre in area." The data 

 of Larsen were drawn from an experiment "auf einer 

 dem Auge sehr gleichmassig erscheinenden, 3 Jahre alten 

 Timotheegraswiese." Montgomery's data were secured 



