AND HORTICULTURE. 21 
sensitive plants, such as sugar beets, carrots, beans and 
cabbage, the soil during 1899 was not sufficiently cultivated, 
and so a result from them could not be expected, particu- 
larly when we remember that the greatness of the percent- 
age of increase depends on the vigour of vegetation. 
There remained this year, therefore, only barley, oats, peas 
and potatoes. 
One might think that all inequalities in fertility would be 
eliminated when the experimental squares and the corre- 
sponding control squares are situated in the manner shown 
in Fig. 4, where the squares under C are control fields and 
under E experimental fields; but that was not the case, for 
the soil represented in Fig. 3 under the last column, C,,, 
showed such a superiority over the soil in the columns 
marked C,, and E,,, that the given results would, without 
observing this fact, be erroneous. To obtain a reliable 
result, it will, therefore, be necessary to graduate the soil 
according to its fertility. 
If to all the squares under E,, the value 1 is given, the 
squares under C,,, must be put =1°4, and the squares 
under C,, =0'8. The harvest from the squares under C,,, 
must be divided by 1°4, and under C,, by o°8. In this 
way we shall attain, not the reality, but a minimum of 
increase. 
By this, graduation we had a good proportion of oats 
growth in the fields which lay between C,, and E,, and 
between E,, and C,,, ; the growth in the latter was at least 
twice the former. 
The new influence machine was used on these experiments 
(see Fig. 5), with cylinders of about 30 cm. in diameter 
and 40 cm. in length. It was kept in motion about eight 
hours daily, from 7 a.m.—11 a.m. and from 4 p.m.—8 p.m. 
This rule could not, however, be regularly followed, especially 
in humid weather. In case of rain the machine ceased to 
work. On the other hand, the machine was at work during 
a longer period in cloudy days, when it was not to be 
