ANALYSIS OF COMPLEX MATTER 147 



We have observed that a given volume of oxygen unite? 

 with double its volume of hydrogen. If Avogadro's hypo- 

 thesis be true, then each molecule of water must contain at 

 least twice as many particles of hydrogen as of oxygen. But 

 this observation alone tells us nothing more. When we come, 

 however, to measure the densities of hydrogen, oxygen, and 

 water, under precisely the same conditions, i.e. at a tempera- 

 ture at which water is a gas like its components, we find that 

 water-gas is nine times, and oxygen sixteen times, as heavy as 

 hydrogen. These numbers must, from Avogadro's hypothesis, 

 also stand for the densities of those particles, which are con- 

 tained in equal number, in equal volumes under the same 

 conditions. That is, these are the relative densities of mole- 

 cules of water, oxygen, and hydrogen. The molecule of 

 hydrogen being taken, since it is the lightest, as the standard, 

 and, since it is supposed to contain two atoms, being called 2, 

 then the corresponding numbers for water and oxygen will be 

 18 and 32. These numbers are called molecular masses. 

 The molecule of water must therefore contain one atom of 

 oxygen and two atoms of hydrogen, and is written shortly as 

 H 2 O, while the reaction will be expressed by the following 

 equation : 



2 H 2 + O 2 =2 H 2 O. 



Or, two molecules of hydrogen uniting with one molecule of 

 oxygen yield two molecules of water. 



It must always be remembered that the numerical values 

 called atomic masses require to be very clearly distinguished 

 from those relative quantities of different kinds of matter 

 which have been found by experiment to be chemically equi- 

 valent. Chemical equivalents are based upon the results of 

 observation. They cannot be altered, except so far as modes 

 of measurement may become more accurate, for they represent 

 facts of nature. On the other hand, atomic masses are values 

 which are based upon theories, and chiefly on the theory of 

 Avogadro, which may perhaps have to succumb to facts not 

 yet observed. It is true that atomic masses are derived 

 from chemically equivalent quantities, and are always simply 



i2 



