Intensified exploration of sub-Saharan Africa during the 18th and 19th centuries led to many newly described giraffe subspecies. Several populations described at that time are now extinct, which is problematic for a full understanding of giraffe taxonomy. In this study, we provide mitochondrial sequences for 41 giraffes, including 19 museum specimens of high importance to resolve giraffe taxonomy, such as Zarafa from Sennar and two giraffes from Abyssinia (subspecies camelopardalis), three of the first southern individuals collected by Levaillant and Delalande (subspecies capensis), topotypes of the former subspecies congoensis and cottoni, and giraffes from an extinct population in Senegal. Our phylogeographic analysis shows that no representative of the nominate subspecies camelopardalis was included in previous molecular studies, as Zarafa and two other specimens assigned to this taxon are characterized by a divergent haplogroup, that the former subspecies congoensis and cottoni should be treated as synonyms of antiquorum, and that the subspecies angolensis and capensis should be synonymized with giraffa, whereas the subspecies wardi should be rehabilitated. In addition, we found evidence for the existence of a previously unknown subspecies from Senegal (newly described in this study), which is now extinct. Based on these results, we propose a new classification of giraffes recognizing three species and 10 subspecies. According to our molecular dating estimates, the divergence among these taxa has been promoted by Pleistocene climatic changes resulting in either savannah expansion or the development of hydrographical networks (Zambezi, Nile, Lake Chad, Lake Victoria). A correction has been published: Petzold A., Magnant A.-S., Edderai D., Chardonnet B., Rigoulet J., Saint-Jalme M. & Hassanin A. 2020. First insights into past biodiversity of giraffes based on mitochondrial sequences from museum specimens – Corrigendum. European Journal of Taxonomy 717: 1–2. https://doi.org/10.5852/ejt.2020.717.1093