Electron paramagnetic resonance (EPR) is a well-established technique for investigating the properties of metal ions in crystals, chemical complexes and biological molecules. Its major uses have been restricted to metal ions in sites of fairly high symmetry and for which the Abragam and Pryce spin Hamiltonian formalism has provided an adequate framework for the interpretation of the data. EPR provides information about electronic structure and the site symmetry or nature of the environment of metal ions. While EPR is now a mature discipline, its application to studies of metal ions in sites of low symmetry has received only spasmodic attention. There is, however, a considerable body of evidence in the literature going back, in some cases, to the early days of EPR and it is the purpose of this article to review both the historical development of low-symmetry ideas and also their theoretical basis. Major experimental results are tabulated and the discussion includes some of the attempts to interpret low-symmetry EPR data using crystal field and molecular orbital models. It is concluded that the role of excited-state orbitals, under the influence of low-symmetry fields, is largely responsible for the different orientations of the principal directions of the various interactions experienced by a paramagnetic ion.