193 



years of the series; any decline that is taking place is hardly 

 outside the limits of seasonal variation and can easily be accounted 

 for by the difficulties of tillage and the increase of one or two 

 troublesome weeds. Mangolds again show no falling off in yield, 

 though they have now been grown upon the same land for thirty-two 

 years; but with the barley crop, despite the application of fertil- 

 isers, there is a distinct secular decline in the yield." 



In discussing, in Chapter XI, the application of fertilisers to 

 crops, the author says, that " instead of applying a kind of average 

 manure, the farmer ought to have such an appreciation of manu- 

 rial principles that he can adopt his fertilisers as economically as 

 possible to his own soil and conditions'of farming. " " The mixtures 

 sold as "Turnip Manures," "Potato Manures," and so forth, must 

 be in the majority of cases more or less wasteful if they are to be 

 effective everywhere." 



In the concluding Chapter, on " The conduct of experiments 

 with fertilisers," the author insists that it is of particular importance 

 that the degree of accuracy which may be expected from a series of 

 field experiments should be realised before any scheme of experi- 

 mentation is embarked upon. " Nor must it be supposed that by 

 any amount of care the experimental error can be got rid of; there 

 are various ways by which it may be diminished, but in some form, 

 or other it must exist in all work involving measurements, and the 

 only scientific method of dealing with it is to estimate its magnitude 

 and to draw no conclusions from results which are not well outside 

 that magnitude." For example, applying to results drawn from 

 the Rothamsted experiments the method of least squares, the mean 

 error of a single year's result is calculated db 10 per cent, and the 

 probable error of the fifty years'mean is only it 1.9 per cent. 



As to the size and distribution of experimental plots those 

 adopted in the Danish experiments conducted by Dr Sonne upon 

 the relative value of different varieties and management of barley 

 are reported. The plots are about 1/80 acre each (50 sq. metres), 

 and at any one station there are always four plots receiving the 

 same treatment, arranged about the field as follows: 



