Il6 MAINE AGRICULTURAL EXPERIMENT STATION. I916. 



soil. The details of this method have been published 11 in another 

 place and will not be repeated here. Briefly the principle upon 

 which this method is based is to determine first the most proba- 

 ble yield of each plot in the field on the assumption that there is 

 no intrinsic difference in the yield of the different varieties. In 

 other words, a "calculated" yield is obtained which is the proba- 

 ble yield of each plot on the assumption that all plots had been 

 planted with a hypothetical variety whose mean yield is the same 

 as the observed average of all the plots. 



The differences between the "calculated" yield of any plot 

 and the average of the field may be taken as a measure of the 

 goodness of the soil in that plot. Thus if a certain plot has a 

 "calculated" yield of 10 bushels above the average of the 

 field, it means that the soil on this plot is capable of producing 

 about 10 bushels per acre more grain than the field as a whole. 

 Therefore, the observed yield of the variety on this plot is 

 higher than it ought to have been for comparative purposes. 

 In order to make the yield of this plot comparable with the 

 field as a whole it would be necessary to deduct 10 bushels (or 

 at least a percentage figure based on this amount) from the ob- 

 served yield. In case the "calculated" yield is below the average 

 of the field a corresponding amount must be added to the ob- 

 served yield. For further details of this method with examples 

 the reader is referred to the above mentioned paper. 



This method has been tested out under a variety of conditions 

 and appears to give very satisfactory results. We have applied 

 this method of correction to the 1914 and 1915 plots and the 

 discussion of the results will for the most part be based on 

 these corrected yields. The 1913 plots were on a somewhat 

 irregularly shaped piece of ground and it was not possible to 

 use this method of correction satisfactorily. 



There is no doubt but that small plots of oats like these 

 give a somewhat higher absolute yield than ^an be expected 

 when an entire field is planted. The reason for this is that 

 the necessary pathways allow more air and sunlight and also 

 afford more plant food for the marginal plants. However this 

 may be, there is positive evidence that much better results are 



"Surface, F. M. and Pearl, R. On a Method of Correcting for Soil 

 Heterogeneity in Variety Tests. Jour. Agric. Research, Vol. V, pp. 

 1039 to 1050, 1916. 



