RESULTS TO BE TEUSTED ? 53 



orthy and laboriously conducted experiments, the results of 

 hich have been published of late years. Thus 

 1. In a series of experiments made by Mr Gardiner at 

 Barochan, in Renfrewshire, in 1844, upon the effects of various 

 substances applied as a top-dressing in spring to oats, (sown 

 upon land trenched with the spade out of a nine-year-old lea,) 

 we find, among others, the following results : * 



Produce per Imperial acre. 

 Grain. Straw. 



No dressing average of four portions, 6 qrs. 3 bush. 30 cwt. 

 Rape dust, 5 cwt. . . . 6 ... 3 ... 39 ... 



Sulphate of magnesia, 2 cwt., . 5 ... 7^ ... 28 ... 



Now, as the results are here stated, it would appear that the 

 rape-cake added nothing to the produce of grain, though it 

 gave nearly half a ton more straw, and that the sulphate of 

 magnesia actually diminished both to a large extent. But if, 

 instead of the mean of the four undressed portions, the produce 

 of each of these four had been recorded, we might possibly have 

 found that one of them was smaller both in grain and straw 

 than that of the portion to which the sulphate of magnesia was 

 applied ; while in another, the straw might have been as heavy 

 as that which the rape-cake portion yielded. And if so, the 

 inference to be drawn from the table as it stands, that the sub- 

 stances applied were in the one case really hurtful, and in the 

 other only doubtfully beneficial, would not even have been sug- 

 gested. It is not wonderful that land trenched from a nine-year- 

 old lea should give a good crop of oats without assistance, and 

 that it should be little in want of salts of soda and magnesia ; 

 but such tables as the one from which the above results are 

 taken, do not satisfactorily prove it. 



Had there been four separate experiments made with each of 

 the substances, then the means of each set being taken and 

 compared together, a very satisfactory result might have been 

 obtained. As yet we do not possess any such system of mean 

 results, though few things would at present do more to clear up 

 our ideas as to the precise influence of this or that substance on 

 the growth of plants. 



* Transactions of the Highland Society, 1845, p. 421. 



