PROCEEDINGS OF THE CHEMICAL DEPARTMENT. 



161 



In examining these results, it is at once apparent that the 

 soil has been of a very uniform character, and extremely well 

 adapted for experiment. The difference in the produce of the 

 nothing plots is extremely trifling, and it is scarcely possible 

 that this uniformity should be exceeded except under most 

 exceptional circumstances. It is to be noted, however, that the 

 produce slightly diminishes as we pass from the upper to the 

 lower part of the plan, but this does not affect the accuracy of 

 the results, any discrepancy which exists disappearing when 

 the average of the different plots are taken. Along the middle 

 of the plan, at the line of the junction of the two sections, there 

 is, in fact, an average soil, and from the peculiar arrangement of 

 the plots, those to which the same manure has been applied are 

 always equidistant from it, so that if the produce of that in the 

 first section is above the average, that in the second is as much 

 below it; so that the mean of the two must give a perfectly 

 exact result. This becomes still more apparent by reference to 

 the next table, which gives the results calculated to the acre 

 with the average of each pair of plots. The quarters and bushels 

 of grain are calculated at the weight of 40 lbs. per bushel. The 

 sum of grain and straw is also given in cwts. per acre, which 

 admits of reference to the preceding table, because, from the 

 particular size of plot adopted, pounds of produce in the field 

 plan correspond to cwts. per acre. 



Table III. Giving the average 'produce of oats per acre, the grain 

 in quarters, bushels, and pounds ; the straw in tons and acts. 



