The metal =mercury= or =quicksilver=, for instance, may be supposed to be made up of distinct particles of mercury of extreme minuteness, and according to the temperature, these associate themselves in the solid (frozen mercury), liquid (ordinary quicksilver), or gaseous form (vapour of mercury). To whatever treatment pure mercury may be subjected, we cannot get anything but mercury out of it. The particles of mercury have never been broken up. Hence they are generally termed =atoms=, or particles that cannot be divided; and mercury is said to be an =element=, or a substance which is not compounded of any other substances.
Here is a case in which it is very useful to distinguish between fact and hypothesis. The matter of fact is that, up to the present time, no one has been able to get out of pure mercury anything but pure mercury. The statement that mercury is a simple substance, and therefore never can be broken up into any other substances, is a hypothesis which future observation and experiment may or may not confirm.
A hundred and fifty years ago it was universally believed that water was as much an element as mercury. But water is now well known to be a compound. In fact, as has already been said, the particles of water may be very readily broken up or =decomposed= (in what way, you will learn when you study chemistry) into two totally distinct substances, =oxygen= and =hydrogen=, which are gaseous at all known temperatures, though by combining vast pressure with extreme cold they have recently been liquefied. Each of these gases, according to our hypothesis, consists of particles, and since these can by no known means be further broken up, they are considered to be =atoms= like those of mercury.
Nine parts by weight of pure water always yield eight of oxygen and one of hydrogen. The hypothetical particle, or molecule of water, therefore, must be composed of atoms of oxygen and hydrogen having this relative weight; and chemists have grounds for believing that one atom of oxygen and two atoms of hydrogen exist in each molecule of water. If this be so, the structure of water must be more complicated than we thought at first; and each particle of water (the molecule) must be a system composed of three separate atoms.