Currently the African elephant is considered a single species Loxodonta africana with two extant sub-species, a forest form L. a. cyclotis and the savanna/bush elephant L. a. africana. 

Recent morphological and genetic work has produced strong evidence to support the view that there is more than one species of African elephant.
 
However, the taxonomic status of elephants from West Africa remains uncertain.  In addition, some populations of high conservation value may consist wholly or partly of interspecific hybrids under a multiple species scenario.  Premature allocation of Africa’s elephants to two or more species may result in significant populations being left in taxonomic limbo.  The AfESG strongly encourages further genetic and morphological studies to resolve this situation.  Until this is done, the AfESG will continue to refer to the single species, L. africana, but as far as possible will distinguish between forest (‘cyclotis’ form), and savanna (‘africana’ form) and the West African forms.  If this split into forest, savanna and West African forms were to become more readily adopted, the AfESG wishes to emphasize the importance of locally adapted and regionally distinctive populations.
 
AfESG recommends that the following conditions be met before formal acceptance of the division of the African elephant into new species:
  • Further analysis of the existing West African samples should be carried out.
  • Additional genetic samples from a wider range of sites should be collected and analysed (see Annex 1).
  • There should be a general consensus on the significance of the genetic and morphological data between the scientists working on this issue.
The Chair is mandated to advise the membership when they consider that these conditions have been met, and will seek a consensus amongst the AfESG membership before implementing changes in the treatment of the genus Loxodonta
 
Annex 1 Recommended sites for sample collection
  1. Northern forest-savanna boundary, including eastern CAR (Bangassou region along Oubangi River) and northern DRC.
  2. Relict forest areas outside the central African forest block (includes north-eastern CAR: Manovo, Mt Elgon).
  3. Eastern forest-savanna boundary, especially Albertine Rift.
  4. Southern forest-savanna boundary (includes Southern DRC, particularly on the left bank of the Congo River e.g Upemba, southern Salonga, Kasai, Bushimai, northern Angola).
  5. Ethiopia/Eritrea. North-western cross-border population, Babille 
  6. Pendjari/Arly/Park W., Niokolo-Koba, Yankari, (savanna populations),
  7. Gourma population in Mali
  8. Liberia/Guinea (westernmost forest populations where samples are possible)
  9. Southern Nigeria (ideal is Taylor Creek, alternative Omo FR)
  10. Additional deep forest samples: Lope, Ituri, Salonga, Kahuzi-Biega, Minkébé, Dzanga-Sangha and Noubalé-Ndoki.