HOW DISCOVERIES ARE MADE 121 



that Lavoisier, a celebrated French chemist, gave the 

 correct explanation of combustion namely, that it is 

 caused by the union of oxygen with the substance burning. 

 Lavoisier, however, cannot be ranked as a great discoverer, 

 though he shone as an interpreter of the discoveries of 

 others. 



Henry Cavendish, who did his best work between 1770 

 and 1790, discovered the composition of water ; that it is 

 produced when oxygen and hydrogen unite; and he 

 determined with great accuracy the proportions by volume 

 in which the union of the two gases is completed. He 

 also attempted to show, by passing electric sparks through 

 a mixture of the inert gas of the atmosphere, nitrogen, 

 mixed with oxygen, that nitrogen was a single substance 

 and not a mixture ; nearly all the nitrogen disappeared 

 under this treatment, only about one hundred-and- 

 twenty-fifth of the whole being left. It would hardly 

 have been possible for him, in the existing state of know- 

 ledge, with the imperfect appliances which alone were 

 available at that time, to have identified his inactive 

 residue with 'argon/ a gas discovered more than a 

 century later; for the spectroscope was then unknown, 

 and it is the chief means of identifying and characterising 

 gases, and indeed elements of every kind. This is an 

 example of how discovery has sometimes to wait on in- 

 vention ; for, until the instruments of research are invented, 

 it is almost impossible to confirm a discovery, even 

 although it may be genuine. 



The true nature of flame, which, as before remarked, 

 has been a puzzle since the remotest ages, has had to 

 wait on invention for its discovery. When a current of 

 electricity of high tension, such as is produced by an 

 induction-coil or by an electric machine, is passed through 

 any rarefied gas, it gives out a peculiar and often a very 

 beautiful coloured light : sometimes red, as in the case of 



