Graphiurus has been divided into as many as four separate genera (e.g., G. M. Allen, 1939; Holden, 1996b; Pavlinov and Potapova, 2003): Aethoglis, containing the largest African dormouse, G. nagtglasii (sometimes erroneously including G. monardi); Graphiurus, comprised of G. ocularis, with its reduced, simple premolar; Gliriscus, consisting of the rupicolous G. platyops and G. rupicola, with their flattened skulls; and Claviglis, the so-called "tree dormice", to which the remaining species of Graphiurus were assigned. Two of these, Graphiurus and Claviglis, have often been retained as subgenera (e.g., Ellerman et al., 1953; Rosevear, 1969). No published studies based upon a broad sample of species have addressed the validity of these subgeneric boundaries as used by past authorities or presented hypotheses of relationships among species. However, a phylogenetic study based on cranial and middle ear morphology, supports the recognition of three traditionally recognized subgenera of Graphiurus: Aethoglis containing G. nagtglasii, Claviglis comprised of G. crassicaudatus, and Graphiurus which would include all other graphiurines that were sampled (Pavlinov and Potapova, 2003). Their analyses showed that first nagtglasii, then crassicaudatus, diverged early in the evolution of African Dormice, and that the rest of the graphiurines sampled (angolensis, christyi, kelleni, lorraineus, murinus, ocularis, parvus and surdus) form a monophyletic group. Based on Pavlinov and Potapova’s results, three subgenera are recognized below. For synonyms see Ellerman (1940), Ellerman et al. (1953), McKenna and Bell (1997), Rosevear (1969).

The revision of Graphiurus by Genest-Villard (1978), based mostly on size grades, underestimated species diversity, particularly in the G. murinus group. Subsequently, species limits were defined in reports covering different African regions (e.g., Ansell and Dowsett, 1988; Holden, 1996b; Robbins and Schlitter, 1981) The species recognized below reflect information in the literature, as well as myexamination of museum specimens and preliminary, mostly unpublished multivariate analyses of cranial and dental measurements.