Nature of the so-called Elements. 99 
hypothesis. Let us suppose, for instance, that manganese is a 
compound of the form of iron represented in furnace B, with 
something else; and suppose again that the photograph of iron 
Impurity of iron in manganese, as I have eliminated it, we 
ganese taken under the same conditions. I | t fine 
lines say, therefore, that there is no impurity of iron in man- 
ganese, but although the longest iron lines are not there, some 
of the faintest basic ones are. This I hold to be the explana- 
tion of the apparent confusion in which we are landed on t 
supposition that the elements are elementary. 
