Stegodon: Does this ancient elephant have origins in Asia?

So much has been said in recent years about the wealth of fossils in China. Almost all of it about dinosaurs: exciting new species, feathered fossils, nest upon nest of dinosaur eggs.  There is no doubt that China holds exciting clues to the history of our planet; one has only to wait to hear of the next discovery.

Within the past few months, yet another exciting find was revealed, but this time about a little known mammalian ancestor: Stegodon.


Stegodon by artist Hannah Stephens

Painting of a Stegodon by artist Hannah


The name Stegodon, to me, evokes ‘dinosaur’, not ‘mammal,’ but this was, indeed, an ancient animal.  Its fossils resemble those of other similar mammals, from mastodons to mammoths to today’s elephants.



Alexandra van der Geer - shrinking elephants

Figure 1: Reconstruction of four insular dwarf proboscideans with their respective mainland ancestors. Mainland proboscideans: 1, Palaeoloxodon antiquus; 2, Mammuthus columbi; 3, Stegodon zdanskyi [stegodon found in China]. Insular proboscideans: 4, Palaeoloxodon ‘mnaidriensis’; 5, Palaeoloxodon falconeri; 6, Mammuthus exilis; 7, Stegodon aurorae [a type of dwarf stegodon found in Japan]. Based on skeletons at Museo di Paleontología, University of Rome, Italy (1), American Museum of Natural History, New York (2), Taylor Made Fossils, U.S. (3), Museo di Paleontología e Geología G.G. Gemmellaro, Palermo, Italy (4), Forschungsinstitut und Naturmuseum Senckenberg, Frankfurt, Germany (5), Santa Barbara Museum of Natural History, Santa Barbara, U.S. (6), Taga Town Museum, Honshu, Japan (7). Photos 1–2, 4–7 George Lyras, photo 3 courtesy of, reproduced here with permission.

From The effect of area and isolation on insular dwarf proboscidea by Alexandra A. E. van der Geer et al; photo and caption courtesy of Dr. Alexandra van der Geer.


When Stegodon skulls with tusks attached have been found, many (but not all) of the tusks are close together–preventing the trunk (the ‘proboscis,’ from which this group gets its name; proboscis —> proboscidea) from hanging between them.

They lived in what is now Africa and Asia, causing continued debate over its place of origin. Until recently, the oldest known Stegodon fossil, a 6.5+ million-year-old partial molar from Kenya, was described by William J. Sanders in 1999.  That record changed this past December when Dr. Hong Ao and his colleagues published their results dating the sediment in the Lanzhou Basin, China, from which a number of fossils–including that of a Stegodon–were found.

And that Stegodon was found to be between 8 – 11 million years old.


GSA Geologic Time Scale - Neogene

Detail of the Geologic Time Scale, created by the Geological Society of America.  Stegodon is believed to have existed between the Miocene to the Pleistocene, a relatively small segment of time in Earth’s overall history, but still considerably longer than that of our own species!  (You can view the time scale in much better detail here.) 


The fossils of the Stegodon, along with at least 5 other species, were actually found in the 1980s by Professor Xing Zhang of Northwest University in China.  Given the length of time between the fossil excavations and the recent dating of these fossils, one might wonder why determining the fossil age took so long.

Dr. Ao, a scientist at the State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, and his colleagues state that western China does not have suitable material for radiometric dating, an oft-used method for this purpose.

When I asked why this was so, Dr. Ao replied by email, “Because volcanic activities are rare in western China during the late Cenozoic, it is difficult to find  in situ tephros or tuffs for radiometric dating (e.g., 40Ar/39Ar dating).”

Instead, they conducted a magnetostratigraphic study, one in which they determined the age of the rock through the polar reversal record.  Combining this with analysis of the fossils provided evidence that this Stegodon is 2 – 5 million years older than that of the Kenyan partial molar.

“We are indeed surprised by our dating results,” Dr. Ao continued, “which document that the Lanzhou Stegodon is the oldest Stegodon worldwide, although the Stegodon  fossil [was] not discovered by us. However, our dating results document it to be the oldest known Stegodon fossil.”

Dr. Hong Ao 1


Dr. Hong Ao 2

Dr. Hong Ao 3

Dr. Hong Ao 4

Images of Professor Yongxiang Li (from Northwest University) and his master student as well as several employed workers who helped to excavate mammal fossils, Lanzhou Basin, China; photos courtesy of Dr. Hong Ao.


indricotherine fossil

indricotherine fossil2

indricotherine fossil3

Images of indricotherine fossils found in Lanzhou Basin, China; photos courtesy of Dr. Hong Ao.


Dr. Ao himself has been working with fossils in the Lanzhou Basin for 5 years.  When asked how  he and his co-authors chose to work together on this recent paper, he wrote, “I have collaborated with them on broad subjects before, thus I [invited] them to join [in] this research.”

Finding information about this extinct species is difficult.  Unlike mammoths or mastodons, Stegodon does not appear to be a popular ancient animal.

Fortunately, Dr. Alexandra van der Geer–paleontologist, indologist, ethno-zoologist and author –has not only studied this species, she very generously made herself available for questions.  Currently, Dr. van der Geer is an Associate Researcher at both the National and Kapodistrian University of Athens and the Naturalis Biodiversity Center in the Netherlands. She is part of the Isolario project, a project that studies biodiversity and cultural evolution within an island context.

When I asked why there was such a dearth of information on Stegodon, she wrote, “We can think of several reasons for this.”

“First of all, stegodons typically are the elephants of Southeast and East Asia, where most countries did not have the resources and opportunities most Western countries had when it comes to scientific research and excavations. These activities are costly, and since they don’t have a direct use in the sense that they don’t advance the medical, technical or economic levels of a country, they have, understandably, a lower priority. Furthermore, because of this region, half the publications (especially before the 1990s or so) you can find are in Chinese or Japanese, which is not very helpful to the [English-speaking] world.

“Secondly, stegodons are forest elephants. Forest areas are very unlikely places for the long-term preservation of organic materials: everything is eaten, digested or otherwise broken down into smaller components in no time. The tropical (warm, humid) climate of these forests is not helpful either, as decomposition is much faster here. Stegodon remains only have a chance to be preserved when (1) they are covered fast, such as with river sediments, volcanic ashes etc., (2) or are in an oxygen-free environment, such as sunken deeply into a swamp, (3) or were deposited in a natural fridge such as limestone caves where they are gradually covered in clayish sediment or [travertines]. The same is valid for Palaeoloxodon, the Old World fossil elephant, but Europe has many limestone caves, which are excellent for preservation (for a nice [travertine]-preserved negative skull, see Stuttgart museum: skull cast SMNS nr. 32888 from Bad Cannstatt).

“As you can deduct from these preservation issues, it is more likely to find molars and tusks than skeletal material, which is much softer. The vast majority of proboscidean findings all over the planet consists of molars and tusks, and that is not for nothing. Inherently this means that there is much more information about their dentition and diet than about their bodies.”

I was interested in understanding why Stegodons are portrayed as hairless animals, so very similar to contemporary elephants.  Was this just an artistic guess?

“The hairlessness of stegodons is not an artistic guess but a scientific guess instead,” Dr. van der Geer answered.  “Very large animals with thick skins (pachyderms) in a (sub)tropical environment are unlikely to have a significant hair covering. Elephants lost their hairs secondarily. The information for hair growth is not lost, and baby elephants still have a thin, woolly coat. Woolly mammoths lived in the cold, temperate zones, and needed hair, so they were covered in a thick layer of hairs, and for this is evidence (mummies preserved in the permafrost), but the other mammoths (M. meridionalis, M. columbi, M. exilis, etc.), [did] not, and it’s generally assumed that they had a light coat fitting to the temperate zones.

“Tropical and subtropical stegodons almost certainly did not have any coat that’s worth mentioning. Stegodons of temperate zones, however, may have been more hairy. Indeed, the lack of hairs makes them look more like today’s elephants.”


Alexandra van der Geer - Stegodon ganesa-model-I.Vjdchauhan-SiwalikHills

Photo of the two life-size models of Stegodon ganesa;photo courtesy of Dr. Gerrit van den Bergh (University of Woolongong, Australia); special thanks to Dr. Alexandra van der Geer.


“Note, however, that the proboscis is carried very differently. Their tusks are set very close to each other, so the proboscis doesn’t fit in between as in modern elephants, Asian and African alike. This means that the mobility of their proboscis was more restricted, relative to their living relatives.”


Alexandra van der Geer - Flores-excavation-31-stegodon-florensis

Fossils of Stegodon florensis insularis, from Flores, Indonesia; photo courtesy of Dr. John de Vos (Naturalis, the Netherlands); special thanks to Dr. Alexandra van der Geer.


Alexandra van der Geer - stegodon-timorensis-mandible

Mandible (and holotype!) of  Stegodon timorensis; photo courtesy of Eelco Kruidenier (Naturalis, the Netherlands); special thanks to Dr. Alexandra van der Geer.  Anyone familiar with proboscidean teeth and jaws will recognize the similarities instantly.


But how do we know that Stegodon–a rather enormous animal–evolved into something smaller?

“[D]warfs and giants are relative. Something can be a dwarf, yet have a considerable size. When we speak of dwarf stegodons, we mean stegodons that are much smaller than their ancestors. For this, you have of course to have identical or otherwise similar elements from both the descendant and the ancestor in order to compare reliably,” she continued.

“The expectation is that dwarf stegodons must have existed on the islands, following the so-called island rule, according to which large animals get smaller in isolation. There is sound evidence that this rule still stands, and is even more pronounced for fossil species (see Lomolino et al., 2013, in Journal of Biogeography).

“Indeed, the many fossil molars from the Southeast Asian islands (‘Wallacea’) are all much and much smaller than the same molars from their mainland ancestors (see Van den Bergh, 1999). True, you first have to know what is the ancestor, and for this you need information about morphology, or how the molars, tusks, skulls and postcranial elements look like. After that, you compare the sizes.

“Note that if a molar is, for example, half the length of the same molar of its ancestral species, the body weight of that animal must have been a quarter of that of its ancestor! (the cubic law: linear reduction 50% means volume reduction 50% of 50%).”

Alexandra van der Geer - Flores-stegodon-florensis

Molar of Stegodon florensis; photo courtesy of Dr. Gerrit van den Bergh (University of Woolongong, Australia); special thanks to Dr. Alexandra van der Geer.  


“The most interesting dwarf stegodon is Stegodon sondaari, named after the Dutch palaeontologist Paul Yves Sondaar (1934-2003), expert in fossil insular mammals. This stegodon lived on the island of Flores about a million years ago, and weighted only about 15% of the weight of its ancestral species, S. elephantoides (see Van der Geer et al., 2016, in Journal of Biogeography, doi:10.1111/jbi.12743).

“Sondaar’s dwarf stegodon is not the smallest stegodon, that honour goes to the Sumba stegodon (S. sumbaensis), of only 8% of the original weight. Sondaar’s stegodon is interesting because it may have witnessed the arrival of early humans, possibly the ancestors of the Hobbit, or Homo floresiensis. Its fossils are contemporaneous with primitive lithic artefacts, dated to about a million years ago (see Brumm et al., 2010, in Nature 464, pp. 748–752).”


Alexandra van der Geer - Sumba-stegodon-sompoensis-holotype-in-Naturalis-Leiden-2

Molar (and holotype!) of Stegodon sompoenisphoto courtesy of Dr. Gerrit van den Bergh (University of Woolongong, Australia); special thanks to Dr. Alexandra van der Geer.  


“[R]ecently,” she concluded, “one of the island dwarf stegodons (S. timorensis of Timor) has been dated to about 130 thousand years ago (see Louys et al., 2016, in PeerJ 4:e1788). This excludes, according to the authors, an anthropogenic cause for its extinction, because humans had not yet arrived at the island.”



So many people helped with this blog post!  (But please remember that any errors are my own.)

Many, many thanks to Dr. Hong Ao (Dr. Ao Hong) from the State Key Laboratory of Loess and Quaternary Geology (Chinese Academy of Sciences) for his fascinating responses and the great images of fossil excavations in the Lanzhou Basin.  I am thrilled that he was willing to answer questions about his research and that of his colleagues! It was a great honor and a pleasure connecting with him!

I am indebted to Dr. Alexandra Van der Geer, who very kindly (and so very quickly–despite everything else she has going on!!) answered specific questions about Stegodon that I could not find anywhere else and who provided pictures of dwarf Stegodon fossils.  It was an equally great honor and pleasure connecting with her!

A mastodon-sized thank you to the amazing Dr. Katy Smith for providing needed and hard-to-find material on Stegodon fossils!

And an enormous thank you to artist Hannah Stephens for her depiction of a Stegodon as it may have appeared in life.  I am particularly moved by the warmth of its intelligent-looking eyes, and I love the tones within its skin.  I adore this picture.  I am grateful to have it in this post;  I am thrilled to have the actual painting hanging on my wall!  Please be sure to check out her artwork at: or


References from Dr. Alexandra Van der Geer:

  1. Brumm A, Jensen GM, van den Bergh GD, Morwood MJ, Kurniawan I, Aziz F, Storey M (2010) Hominins on Flores, Indonesia, by one million years ago. Nature 464, 748–752.
  2. Lomolino MV, van der Geer AAE, Lyras GA, Palombo MR, Sax DF, Rozzi R (2013) Of mice and mammoths: generality and antiquity of the island rule. Journal of Biogeography 40, 1427–1439.
  3. Louys J, Price GJ, O’Connor S. (2016) Direct dating of Pleistocene stegodon from Timor Island, East Nusa Tenggara. PeerJ 4:e1788
  4. van den Bergh GD (1999) The Late Neogene elephantoidbearing faunas of Indonesia and their palaeozoogeographic implications; a study of the terrestrial faunal succession of Sulawesi, Flores and Java, including evidence for early hominid dispersal east of Wallace’s line. Scripta Geologica 117, 1–419.
  5. van der Geer AAE, van den Bergh GD, Lyras GA, Prasetyo UW, Due RA, Setiyabudi E, Drinia H (2016) The effect of area and isolation on insular dwarf proboscideans. Journal of Biogeography, doi: 10.111/jbi.12743

References used in this blog post:

  1. New magnetochronology of Late Miocene mammal fauna, NE Tibetan Plateau, China: Mammal migration and paleoenvironments; by Hong Ao, Peng Zhang, Mark J. Dekkers, Andrew P. Roberts, Zhisheng An, Yongxiang Li, Fengyan Lu, Shan Lin, Xingwen Li; Earth and Planetary Science Letters; 1o December 2015
  2. Oldest record of Stegodon (Mammalia: Proboscidea); by William J. Sanders; Journal of Vertebrate Paleontology; Vol. 19, No. 4, Dec. 13, 1999, pp. 793 – 797
  3. Fossil elephantoids, Awash paleolake basins, and the Afar triple junction, Ethiopia; by Jon E. Kalb; Palaeogeography, Palaeoclimatology, Palaeoecology; 1995, pp. 357 – 368
  4. The effect of area and isolation on insular dwarf proboscidea; by Alexandra A. E. van der Geer, Gerrit D. van den Bergh, George A. Lyras, Unggul W. Prasetyo, Rokus Awe Due, Erick Setiyabudi, and Hara Drinia; Journal of Biogeography; 11 March 2016.
  5. Magnetostratigraphy – concepts, definitions, and applications, by Cor G. Langereis, Wout Krijgsman, Giovanni Muttoni, and Manfred Menning; Newsletter on Stratigraphy, Vol. 43/3: 207–233, April 2010
  6. Mammoths and Mastodons of the Ice Age, by Adrian Lister, Firefly Books, 2014
  7. Mammoths, by Adrian Lister and Paul Bahn, University of California Press, 2007
  8. The Proboscidea: Evolution and Palaeoecology of Elephants and Their Relatives, Edited by Jeheskiel Shoshani and Pascal Tassy, Oxford Science Publications, 1996
  • Stegodontidae: evolutionary relationships by Haruo Saegusa, pp. 178 – 190, The Proboscidea: Evolution and Palaeoecology of Elephants and Their Relatives
  • Palaeobiogeography of late Neogene African and Eurasian Elephantoidea by Jon E. Kalb, David J. Froehlich, and Gordon L. Bell, pp. 117 – 123, The Proboscidea: Evolution and Palaeoecology of Elephants and Their Relatives