1 877.] 
On Scientific Method. 
481 
paper of the same weight, from a height, we find that the 
lead reaches the ground long before the paper does. We 
should naturally conclude — as all, or almost all men did 
before the time of Galileo — that the nature of the two bodies 
influences the velocity of their descent ; whereas it is actually 
found, by carefully conduced experiments, that no property 
of bodies except their absolute mass has any influence upon 
their gravitating powers. In experiment we must therefore 
seek to eliminate, one by one, those circumstances which are 
really not of importance as influencing the phenomenon in 
question ; we must simplify the experiment as far as possible, 
taking care, however, that in our attempts at simplification 
we do not overlook the circumstance really governing the 
phenomenon. After all, however, some overlooked condition 
may be present, the non-observance of which entirely vitiates 
our results. Experiments were long ago very carefully 
carried out with a view to prove that earth can be formed 
from water : water was again and again distilled from a per- 
fectly clean glass vessel, yet there remained a small quantity 
of a white earth in the vessel in which the water was dis- 
tilled. If the water has not been converted into the white 
earth, where has this substance come from ? The conclusion 
seemed inevitable, and the conclusion was therefore adopted 
— water may be converted into earth. But the overlooked 
circumstance was this : — Water ads on glass, especially at 
high temperatures, so as to dissolve part of it, and the white 
earth is really a portion of the glass vessel dissolved by the 
boiling water, and left in the vessel when the water has been 
entirely boiled away. 
An exceedingly instructive example of the process of 
elimination of non-important conditions of experiment is 
afforded by Sir Humphrey Davy’s researches upon the 
electrolysis of water. When water was decomposed by the 
eleCtric current, an acid and an alkali invariably made their 
appearance at the poles along with the oxygen and hydrogen. 
Electricity, some people supposed, caused the production of 
the acid and the alkali ; others imagined that water always 
contains acid and alkali. By using agate or gold vessels in 
place of glass, to contain the water, Davy showed that less 
acid and less alkali was produced. Finally, by carrying out 
the decomposition in gold vessels, in the exhausted bell-jar of 
an air-pump, he was able to obtain from pure water, oxygen 
and hydrogen free from both acid and alkali, thus showing 
that the presence of air was in some way the cause of the 
production of the acid and of the alkali. 
If, therefore, an experiment seems to point irresistibly to 
VOL, VII. (N.S.) 2 L 
