PROCEEDINGS OF THE CHEMICAL DEPARTMENT. 411 



speak for themselves, but a few observations may be made re- 

 garding some of the prominent lessons to be observed from them. 



It is unnecessary to do more than refer to the care which has 

 been taken by all the experimenters to secure the utmost prac- 

 ticable degree of accuracy, for this is obvious to every one who 

 examines them, and is still more so to myself, from my having 

 been in frequent correspondence with these gentlemen during 

 their progress. Neither is it necessary to enlarge on the very 

 unfavourable nature of the season, unfavourable to the turnip 

 crop, and singularly so to agricultural experiments of all kinds. 

 On this account the work of no less than four experimenters 

 proved entirely abortive, and those of several others were at one 

 time in jeopardy. The severe drought of the early, and the 

 excessive rains of the later part of the season were most unusual, 

 and, though the latter undoubtedly produced a much better crop 

 than was at one time expected, it cannot be considered as having 

 been otherwise than disadvantageous. For an ordinary season 

 the time of sowing was too late to secure a large crop ; but, on 

 the other hand, the experimental plots on that account escaped 

 risks which they might have incurred had they been put in 

 during the dry weather of the first week of June, and it must be 

 borne in mind that in such experiments it is not so much the 

 absolute amount of the crop as that of the relative produce on 

 the different plots, which is of importance. 



One of the most noticeable points observed in the experiments 

 was the tendency in some cases on the part of the plots to rob 

 one another, and at its first observation I greatly feared that the 

 whole results would be invalidated ; but in the hopes that some- 

 thing might nevertheless be made of them, I requested that the 

 outer and inner drills of each plot might be weighed separately, 

 if it appeared necessary. In some cases the experimenters saw 

 no appearance of any of the plots borrowing from one another, 

 and did not think it advisable to make this separate weighing, 

 and in others, where it was done, the result was to show that there 

 had been no such effect. It appears, indeed, only to have oc- 

 curred in the very light soils, in which the roots of the plants 

 are able to travel some distance in search of food. The possibi- 

 lity of one plot drawing sustenance from an adjoining one is a 

 matter to which I directed attention some time since ; and in an 

 address on " Experimental Agriculture," given at the Stirling 

 Show, in 1864 (Transactions, Vol. XVI., page 362), I advocated 

 the plan of surrounding each plot with a sort of frame, as it 

 were, manured in a similar manner to itself, and weighing only 

 the inner part. Having subsequently seen a large number of 

 small experiments in Ayrshire, in none of which could the 

 smallest indication of borrowing be detected, I was led to the 

 conclusion that this precaution was unnecessary. It now ap- 



