Evolution in the Order ELEPHANTIDAE

It can now be conclusively asserted that the African forest elephant is its own species (Loxodonta cyclotis), distinct from savannah elephants. Ancient DNA analysis published in September 2016 now shows that forest elephants are more closely related to the now extinct 'streight-tusked' elephant of European forests than they are to the living savannah elephant.

However, these relationships would have been controversial several years ago. Forest elephants have long been considered a subspecies of African elephant, a taxonomic unit they shared with the better known savannah elephant (L. africana). In 2003, the International Union for the Conservation of Nature (IUCN) released a statement citing the ŮuncertainÓ taxonomy of elephants, and calling for further evidence before determining the relationships between existing elephant populations (1). However, sufficient evidence is now available. Both morphological and genetic studies have demonstrated that two distinct species, African savannah (or bush) elephants and forest elephants, comprise the genus Loxodonta.

© Peter Wrege

African Forest Elephant - the 'new' species
Loxodonta cyclotis

© Katy Payne

African Savannah Elephant
Loxodonta africana

©Ton Schat

Asian Elephant - Elephas maximus

Multiple studies had presented morphological evidence supporting a species distinction by the early 2000s. For example, an analysis of elephant skulls collected across Africa proposed that forest elephants “deserve to be ranked as full species”. This study, done by Peter Grubb, Colin Groves and others (2), concluded that “living bush and forest African elephants are evolutionarily and ecologically distinct forms.” What the discussion lacked at the time was genetic evidence. With the appearance of elephant research that included genetic sequences we have now learned “that little or no nuclear gene flow occurs between forest elephant and savanna elephant populations” (3). Not only are forest elephants and savannah elephants genetically distinct, they “are as or more divergent in the nuclear genome as mammoths and Asian elephants, which are considered to be distinct genera” (4).

A mitochondrial DNA (mtDNA) study was published in 2012, conducted by Adam Brandt, Yasuko Ishida, Nicholas Georgiadis and Alfred Loca (3). It proved to be the final piece of the puzzle. Previous studies had shown that evolutionary trees based on nuclear DNA samples did not match trees based on mtDNA. mtDNA from savannah elephants had a significant overlap with mtDNA of forest elephants. The same overlap was not there in the comparison of nuclear DNA samples. In the 2012 study, specific genetic markers taken from different elephantid  lineages were analyzed. The results supported the hypothesis that the abovementioned incongruence between mtDNA and nuclear DNA studies was caused by a combination of the elephant social system and differences in the variance in reproductive success between males and females. Females of both savannah and forest elephant species remain close to their places of origin, while males disperse. Additionally, where populations of the two species mix, male savannah elephants, because of their larger size, are reproductively more successful than forest elephant males because they can dominate the latter. Incongruence between mtDNA, which is passed down only through the maternal lineage, and nuclear DNA is attributed to these differences in male reproductive success.

The evolutionary tree of proboscidea, a taxonomic group that unites all elephantid lineages as well as mammoth and mastodon species,  can now be constructed with higher certainty than ever before. Two distinct clades seem to have formed 6 million years ago. The first clade included the hypothetical ancestor of savanna and forest elephants. The two seem to have diverged sometime during the miocene-pliocene transition 5 million years ago. The second clade includes Asian elephant and the now extinct woolly mammoth. It is important to note that these two diverged from their common clade later than savanna and forest elephants diverged from theirs, making them genetically closer to each other than the two species of Loxodonta. Africa has been shown to be the cradle of all elephantid species (3). Asian elephants then migrated to Asia and the African savannah elephant began to dominate the expanding grasslands of East Africa. Forest elephants followed their independent evolutionary path in the dense Central African forests. Important differences between the forest and savannah habitats, in particular the abundance of grasses and relatively lower abundance of fruits and trees in the savannah, resulted in very different diets for the two species. In addition, savannah elephants now faced large, group-hunting predators (lions and hyenas), as well as seasonal shortages of open water, which together might explain some of the differences in the social system between savannah and forest elephants.

These taxonomic issues are more than an evolutionary debate within the scientific community. Classification of the species is important for the conservation of forest elephants. It might have been easier to ignore the decline of forest elephants if they were seen as populations within the African elephant species (in fact the relatively poor data on forest elephant populations before 2013 resulted in conservation decisions by the IUCN and CITES basically ignoring forest elephants). However, once it is acknowledged that the forest elephant is a unique species, the importance of their conservation rises greatly. The threat of the loss of biodiversity is increased even further because forest elephants have the highest within-species genetic diversity of all elephantid taxa (4). Hopefully, as the new taxonomic order enters the mainstream scientific thinking as well as the imagination of the public and the policymakers, it will facilitate the effort to study and conserve forest elephants.


  1. IUCN African Elephant Specialist Group (2003). Elephant Genetics: Statement on the taxonomy of extant Loxodonta.
  2. Grubb, P., et al. (2000). Living African elephants belong to two species: Loxodonta africana (Bumenbach, 1797) and Loxodonta cyclotis (Matschie, 1900). Elephant 2(4):1-4.
  3. Brandt, A. L., et al. (2012). Forest elephant mitochondrial genomes reveal that elephantid evolution in Africa tracked climate transitions. Molecular Ecology 21(5):1175-1189.
  4. Rohland, N. et al. (2010). Genomic DNA sequences from mastodon and wooly mammoth reveal deep speciation of forest and savannah elephants. PLOS Biology 8(12):e1000564