388 Experiments with Salt upon Mangolds. 
On Plot No. 1, where no salt was used, only 510 roots were 
grown ; whilst on the 2nd unsalted plot (No. 6) 615 mangolds, 
or 105 more than on No. 1, were raised. 
The deficiency of roots on No. 1, when multiplied by 20, so 
as to obtain the produce per acre, is considerable ; and it is due, 
no doubt, to the difference of the number of roots on the Plots 
No. 1 and No. 6 that the weight of the mangolds on No. 6 was 
much heavier than those on No. 1. 
In field experiments upon root-crops, not less than l-20th of 
an acre should be reserved for each trial ; and the number of 
roots should be counted when they are taken up ; for, unless the 
number of roots on the several pieces agree pretty well with 
each other, the results of the different experiments are abnormal, 
and cannot be compared. 
With the exception of No. 1, the number of roots on the 
other plots does not vary very much, though more than I like 
to see. 
On No. 6 (unsalted), the number of roots approaches more 
nearly to that grown on the other plots, excepting in No. 1 and 
No. 2. The produce of No. 6, therefore, answers better for com- 
parison than that on No. 1. 
Notwithstanding some strange, discrepancies in the results of 
the preceding experiments, it will be observed that common salt 
has had a favourable effect upon the produce in every instance. 
The increase in the weight of roots on No. 2 and No. 3, 
is indeed, very considerable. As salt is a cheap manure, its use 
in every instance was most beneficial in an economical point of 
view. 
The unsalted Plot No. 6 passed through the centre of the 
experimental half-acre. On one side of it, salt, at the rate of 
1 cwt. per acre, was applied to Plot No. 5 ; and on the other 
side, 2 cwt. of salt on Plot No. 7. This order, it will be seen, was 
preserved up to No. 8, and we have thus, on the right side of 
No. 6, plots dressed with salt at the rate of 1, 3, 5, and 7 cwt., 
and on the left plots manured with 2, 4, 6, and 8 cwts. of salt 
per acre ; to which is added No. 11, with 9 cwts. of salt. 
In these trials, 1 cwt. and 2 cwts. of salt per acre did not 
appear much to affect the produce ; whilst a dressing of 3 to 
8 cwts. ^)er acre gave large increase in the weight of mangolds. 
On poor, sandy soils, it would appear from these experiments 
that as large a dressing as 9 cwts. of salt per acre may be used with 
advantage. I do not recommend so large a dose, but simply 
state that it gave 2J tons more clean and dressed mangolds per 
acre than the unsalted portion of the same field. Probably 4 to 
5 cwts. is a sufficient dressing on the lightest sands ; and on 
good sandy loams, and warm, friable turnip-soils 3 cwts. of 
