New Outlines of Chemical Philosophy. 35 1 



untanglble elements in matter, which are only known to us by 

 the effects which they produce. 



There are two modes or ways of investigating physical pro- 

 blems. Some are investigated I)y experiment only; but in others 

 experiment and observation are the data we have to reason from. 

 As examples of the first kind of investigation, let two experi- 

 ments be taken from electricity. 



It has long been a disputed point among electricians, whether 

 glass be permeable by the electric fluid, and whether electricity 

 by induction be permanent. 



To investigate these problems by a single experiment, J took 

 the Florence flask described in a former paper*, and suspended 

 it by a short brass chain from the ceiling of a room. When the 

 electricity of the earth was verv strong, I took a glass tube about 

 3-4ths of an inch in diameter and two feet long ; and after having 

 excited it by a silk rubber, I held it under the flask at the tli- 

 stance of about two feet : the gold-leaves within the flask in- 

 stantly began to show signs of electricity; and on gradually car- 

 rying the tube up towaids the flask, the divergency of the leaves 

 increased. 



When the tube was brought to the distance of about nine 

 inches from the flask, the leaves diverged to an angle of sixty 

 degrees, and continued electrified for fourteen days. The tube 

 was not in this experiment brought nearer the flask than nine 

 inches. This experiment shows by mere inspection, that the 

 electrical element met with very little obstruction in its passage 

 through thin glass ; and that electricity by position or induction 

 is permanent. 



In making delicate experiments on electricity, it is necessary 

 to attend to the electricity of the earth ; for, if that state of the 

 earth had been very weak at the time when the above experi- 

 ment vvaa made, a very different conclusion might have been 

 drawn. 



But in reasoning on experiments and observations, in deducing 

 general conclusions from them, it is of the utmost importance 

 to observe : 



1. That all the circumstances or natural things concerned in 

 the experiment be accurately known, and taken into the investi- 

 gation ; and, 



2. That no hypothetical opinion be taken for granted. 



In the experiment to show the decomposition of water, there 

 arc three things concerned in it ; that is to say, the two electri- 

 cal elements, — thermogen the element of heat, photogen the ele- 

 ment of light and water. These being the only circumstances 

 • Pliil. Mag. vol. xivi. p. 210. Exp. 10, 



concerned 



