AND HORTICULTURE, 57 
that, because only sugar beets, fodder beets and carrots 
were unripened, he did not think it worth while to let the 
machine work. There were in all 46 days of inactivity, 
among them 15 rainy days, during which it was indifferent 
whether the machine worked or not. Altogether, therefore, 
we have 38 days when the machine was out of use. There 
was a further eight days of inactivity, three being in con- 
sequence of necessary repairs and three days because of 
great conducting power of the air due to its humidity. 
During two of the eight days this humidity of the atmos- 
phere caused another difficulty, due to radzo-activity. This 
quality of the air was discovered by Elster and Geitel in 
1902-1903, and acts chiefly in such a manner that the air 
has a great conducting power for electricity. When this 
condition of the air arrives the only thing to do is to wait 
for its cessation, which happens generally after some few 
‘hours, or at most oneday. During Juneand July it occurred 
but three times. Mr. Tillberg tried to neutralise the effect 
of this radio-activity by putting new cylinders on the 
machine and by varnishing the old cylinders, &c. This 
work might have been avoided, however, for, after having 
ascertained conclusively that the air in the room is quite 
dry, and after having given the cylinders a careful drying 
by warming them, we can but wait for the cessation of the 
conducting power in the air before resuming operations. 
According to information from Atvidaberg, the weather 
during the latter part of September was not actually wet, 
but humid. It is, therefore, probable that the machine 
could have been kept in activity during 14 days more. 
Remembering that the effect of the electric air current is 
nearly the same as that of sunlight, it is very likely that a 
continuance of the electrical air current treatment during 
these two weeks would have augmented to a perceptible 
degree the increase per cent. of the roots. One consequence 
of the experiments this year is evident, namely, the necessity 
of improving the method of control. (See pages 33-34.) 
