A lecture given by Tony Paxton, at the Adventures in the Physical Metallurgy of Steels (APMS) conference held in Cambridge University. Electron theory in the understanding of iron, its magnetic properties and the behaviour of intersitials such as hydrogen and carbon in iron. An argument is made that the well established tight binding approximation to the electronic structure now furnishes us with a reliable theory for the description of the three phases of pure magnetic iron and its principal interstitials, including carbon and hydrogen. This will open the way in the coming five years to fully quantum mechanical atomistic simulations of steel and its embrittlement by hydrogen. Some key questions have recently been answered using density functional theory, namely: what are the equilibrium concentrations of vacancy-carbon and vacany-hydrogen point defects? Can hydrogen dissolve exothermically in ferrite through its binding to defects? Do carbon atoms form "dimer molecules" when bound to vacancies in ferrite? The tight binding theory can also answer these question quantitatively as I shall show, and moreover this is a theory that can be implemented with sufficient speed in a computer to admit molecular dynamics and statics simulations well out of the reach of density functional theory. Soon we will be able to answer questions such as, how do carbides act as traps for hydrogen? How deep are these traps? Are they located at the metal-carbide interface or in the depths of the precipitate? These are questions that are also just beginning to be answered using atom probe tomography, so the theory and experiment are at similar stages of this enquiry.