The first principle of chemical composition is that elements are actually present in their compounds. It is a golden thread running through the history of compositional thinking in chemistry since before the chemical revolution. Opposed to this principle, which I call Actually Present Elements (APE), is the idea that elements are merely potentially present in their compounds: although not actually present, it is possible to recover them. In this paper I follow that golden thread, and then discuss the status of (...) APE itself: is it true? What arguments were there for it, and when? I argue that APE is a metaphysical principle, albeit at a lower level of generality and abstraction than the term ‘metaphysical’ usually suggests. I critically examine a range of different views on how metaphysical principles might be involved in research programmes in empirical science, and conclude by endorsing Elie Zahar’s view that metaphysical principles such as APE are to be found at the very heart of science. Moreover, they can be recipients of empirical support just like other parts of scientific theory. (shrink)

Lavoisier defined an element as a chemicalsubstance that cannot be decomposed usingcurrent analytical methods. Mendeleev saw anelement as a substance composed of atoms of thesame atomic weight. These `definitions' doquite different things: Lavoisier'sdistinguishes the elements from the compounds,so that the elements may form the basis of acompositional nomenclature; Mendeleev's offersa criterion of sameness and difference forelemental substances, while Lavoisier's doesnot. In this paper I explore the historical andtheoretical background to each proposal.Lavoisier's and Mendeleev's explicitconceptions of elementhood differed from eachother, and (...) from the official IUPAC definitionof `element' of the 1920s. However, Lavoisierand Mendeleev both subscribed to – andemployed – a deeper notion of a chemicalelement as the component of compound substancesthat (i) can survive chemical change, and (ii)explains the chemical behaviour of itscompounds. (shrink)