AND HORTICULTURE. 33 
knowledge is at present very inadequate. Most probably 
the intensity in question is very variable. Hence it seems 
that this relation between the increase per cent. and the 
period during which the electrical air-current exercises its 
effects is also a variable quantity, and that the effect of the 
electric air-current is different under different exterior con- 
ditions. This must be taken into consideration in judging 
the amount of the increase percentage. 
As mentioned above, the greatest difficulty in determining 
this percentage of increase lies in the want of uniformity 
in the soil. An augmentation of the number of squares 
might perhaps lead to an equalising of the difference .in 
the average fertility of the soil; but this will lead to 
another difficulty—namely, that which consists in making 
out the many squares. Much skill and foresight are neces- 
sary to avoid confusion. 
As will be seen from the experiments previously des- 
cribed, there were for rye and wheat 12 squares (six for each). 
It would have been necessary to prepare each square 
separately in order to attain positive results. That mis- 
takes might happen here is easy to comprehend, and 
some have really happened; but they were so apparent 
that they could be immediately corrected. To prepare 
the whole soil beforehand would obviously have been too 
costly. 
A relatively satisfactory result could certainly be obtained 
in the following way :—Two squares of 3 m. (10°8 sq. ft.) 
are prepared with the soil from the field in which the elec- 
trical treatrnent is to be applied. They must lie at least 
5 m. from this field and from each other. The soil on both 
the squares must be well mixed and put out in equal quanti- 
ties in every square. One is put under the same electric 
treatment as the field. If care is taken that the sunshine 
and the rain will reach both the squares in the same pro- 
portions, comparable results can be obtained. The increase 
per cent. for the whole field can then be calculated. 
D 
