432 
Our standard-liquid was pure tapwater, which yielded a mean 
deflection of 50 scale-marks. The Utrecht tapwater is very pure 
and contains few salts. In addition the temperature is pretty 
constant, which is of great importance, since temperature very much 
influences water-electricity. 
A control-test with water was inserted between every set of two 
experiments, in order to ascertain the accuracy of the apparatus. 
Furthermore the cylinder was washed out with tapwater after 
each experiment to remove small quanta of lingering electrifying 
substances, which might render the results of the following experi- 
ments less reliable. 
In the case of water-electricity we found that odorous substances 
did not all act in the same way. Most of them reinforeed the 
positive charge of the water; others hardly modified it or did not 
do so at all; a few again even weakened it so as to excite a 
negative charge. All this occurred seemingly without any special 
method. It is true, stronger concentrations (which are insignificant 
with - the almost always slightly soluble odorous substances) 
generally give a greater increase or decrease. Besides, in the 
homologous series we found an augmentation of the deflection 
according as we passed from the lower to the higher terms. 
For the present it seems utterly impossible to draw a hard-and- 
fast line, separating the reinforcing from the weakening odorous 
substances. 
That waterfall-electricity is not identical with spray-electricity, 
may appear e.g. from the behaviour of indol, which markedly in- 
creases the charge in the former, but is almost inoperative with 
the second. 
Conversely thymol e.g. gives a strong nebula-charge and hardly 
any waterfall-electricity. 
Another instance is that of fresh-distilled water, which very 
distinctly intensified waterfall-electricity, whereas it remains inactive 
in spraying. 
We subjoin a list of some odorous substances with their charges 
in the numerator of the fraction and the deflection of tapwater in 
the denominator, as we noted them down directly after the reading. 
The sign of the charge is also given: 
Phenol) eeN leden 
Cressol linn N: ‘sol. : --.77°/,5 
Xylenol'/,.,. N. sol.: + '?°/,, 
