Fossil focus: Giraffidae — where we’ve been and where we’re going

by Chris Basu1

Giraffes (Giraffa camelopardalis) are charismatic and iconic animals. Together with their closest living relatives, okapis (Okapia johnstonii), they are remnants of an otherwise diverse group of even-toed ungulates – Giraffidae. Giraffids are ruminants (they have a specialized four-chambered stomach), and are related to other ruminant groups such as bovids (including cattle and antelopes), cervids (deer) and antilocaprids (pronghorns).  Ruminants use microbes in their stomachs to ferment and break down vegetation that would otherwise be impossible to digest.

The origins of Giraffidae are hazy. DNA analysis confirms that they are a valid group, and that they diverged from other ruminants approximately 25 million years ago. This agrees with what is generally understood from the available fossil record, which is that giraffids and their closely related (now extinct) sister groups evolved in northern Africa or southern Eurasia towards the beginning of the Miocene epoch  (which lasted from 23 million to 5 million years ago).

What is a giraffid?

Even-toed ungulates are classified as giraffids if they have lower canine teeth with two obvious lobes, and have skin-covered, unbranched bony horns called ossicones (Fig. 1 ).  These start to form early in life as discs of cartilage under the skin. As the animal grows, the cartilage turns into bone and becomes permanently fused to the underlying frontal bone of the skull. Unlike in deer, these cranial appendages are never shed; and unlike in cattle, they are not covered by a sheath of keratin (the material that makes hair and fingernails).

Strictly speaking, bilobed canine teeth and ossicones are not found only in Giraffidae, because the extinct , closely related potential sister-group Climacoceratidae also had them. The difference is that climacoceratids had branched ossicones, and the secondary lobe of their canines was not well developed (accounting for less than one-third of the width of the tooth). In recognition of this, climacoceratids and giraffids are grouped together as Giraffoidea.

To complicate matters even further, ossicones are also seen in another closely aligned group, Palaeomerycidae. However, palaeomerycids had an additional appendage at the base of the skull. Some of these are highly ornate, as in Xenokeryx amidala (named after the Star Wars character Padmé Amidala). A recent analysis combined the three groups Giraffidae, Climaceracidae and Palaeomerycidae into the super-group Giraffomorpha. This may be sensible, because the characteristics of these three groups overla p. For simplicity, we will talk only about giraffids from this point onwards.

Diagnostic features of Giraffidae. Left image shows a highlighted bilobed canine tooth during a giraffe post-mortem examination. Right image shows a highlighted unbranched ossicone, from Samotherium boissieri. This giraffid possessed had one pair of conical supraorbital ossicones on the skull above its eyes.
Figure 1 – Diagnostic features of Giraffidae. Left image shows a highlighted bilobed canine tooth during a giraffe post-mortem examination. Right image shows a highlighted unbranched ossicone, from Samotherium boissieri. This giraffid had one pair of conical skull above its eyes.

Are okapis really a ‘living fossil’?

Okapis are often touted as ‘living fossils’ — a term coined by Charles Darwin, and first applied to okapis by early-twentieth-century US palaeontologist Edwin Colbert. These relatively short-necked, short-legged giraffids were known to zoologists first by their enigmatic skins; a (shot) specimen was later presented to the Zoological Society of London in 1901. Their affiliation with living giraffes was quickly realized, and their evolutionary significance became apparent. Some aspects of okapi anatomy are similar to the traits seen early in the giraffid evolutionary tree, but others are more derived — such as closed tear ducts , well-developed sinuses and ossicones which are positioned behind the level of the eyes. So although they can offer an insightful glimpse into the structure and function of the early giraffids, these animals are not living fossils.

Key giraffids

The relationships between various giraffid groups are debated in the literature, usually on the basis of a small number of characteristics. A major revision of the phylogeny, or evolutionary relationships, using contemporary statistical methods is needed. Over the past century, several hypotheses have been put forward. In more recent years researchers have organized the known genera  into subgroups, offering a sensible framework from which to start. The subgroups are:

  1. Canthumerycidae: This contains the earliest known giraffid, Canthumeryx sirtensis, known from the early Miocene of Libya and Kenya.
  2. Okapiinae: Extant okapis, and Afrikanokeryx leakeyi (formerly known as Palaeotragus primaevus from the mid-Miocene of Ngorora, Kenya). These animals retain relatively ancestral body proportions.
  3. Giraffokerycinae: This mid-Miocene group contains the Asian species Giraffokeryx punjabiensis and the African Giraffokeryx primaevus.
  4. Sivatheriinae: A group of large-bodied girafffids from the Pliocene epoch (5 million to 2.6 million years ago) and the Pleistocene epoch (2.6 million to 11,700 year ago), including the genera Sivatherium, Bramatherium and Helladotherium. These animals have stout and robust limbs. Many taxa have two pairs of ossicones (with the front pair partially  fused in Bramatherium).
  5. Palaeotraginae: This contains the key ‘intermediate’ species Samotherium major from Samos in Greece, among others such as Palaeotragus and Shansitherium.
  6. Bohlininae: Giraffids with markedly elongated metapodial leg bones (and probably an elongated neck). Represented by Bohlinia attica and Honanotherium schlosseri. The position of Bohlinia on the evolutionary tree is crucial, as its descendants may represent the origin of the genus Giraffa, which includes modern giraffes.
  7. Giraffinae: Giraffa spp. feature skulls containing extensive sinuses, long metapodials and a suite of unique characteristics in the cervical vertebrae, or bones of the neck. The ossicones are simple, and both sexes have them. Several extinct species are known from the Pleistocene of Asia and Africa. There are nine sub-species of G. camelopardalis, but DNA analysis suggests that some of these are actually genetically separate species.



Within Giraffidae, different species can have varying numbers, shapes, and locations of ossicones. In most taxa, females have no ossicones — modern giraffes are a notable exception. The earliest giraffids are thought to have had one pair of ossicones over the eyes. In general, more derived ossicones have a tendency to be more caudally and medially  placed (compare Samotherium in Fig. 1 with Giraffa in Fig. 2A). Some genera had two pairs (Fig. 2). In Sivatherium, the back pair are branched. Because unbranched ossicones are a trait shared by all giraffids, there has been some debate as to whether sivatheres should be included in Giraffidae or should be put in a sister group. These days, the branched ossicones are seen as an autapomorphy (a derived peculiarity, unique to one taxon) of Sivatherium, validating its inclusion with the giraffids.

Figure 2 – Diversity of ossicones in giraffids. Images not to scale. (A) Giraffa camelopardalis — both sexes have one pair of frontal ossicones, medial and caudal to the orbital rims. Some individuals also have a single median ossicone, as seen here. (B) Composite reconstruction of Sivatherium giganteum — the large, elaborate posterior ossicones are from a different individual to the skull. (C) Giraffokeryx punjabiensis. (D) Bramatherium megacephalum.

Long necks

Modern giraffes are well known for their long necks, but this is not a feature found in all giraffids. In fact, various types of neck are represented by fossils from different taxa. One group of researchers used the length–width ratios of the cervical vertebrae , and a collection of 20 characteristics of the vertebrae to map changes in vertebral shape, using a hypothesized (but reasonable) cladogram (Fig. 3). This demonstrated that necks were already slightly elongated in the earliest giraffids. We then see a key intermediate stage, represented in Samotherium major, before the marked neck elongation in Giraffa sivalensis (an extinct Asian species), and even further elongation in the living G. camelopardalis. The study also shows contrasting vertebral changes in Sivatherium giganteum, where the cervical vertebrae are instead stubby and robust.

Cladogram with geological age and back view of C3 vertebrae of taxa evaluated. Reproduced from Danowitz et al. (2015) under Creative Commons CC-BY-4.0. Pe, Prodremotherium elongatum; Cs, Canthumeryx sirtensis; Oj, Okapia johnstoni; Gp, Giraffokeryx punjabiensis; Sg, Sivatherium giganteum; Bm, Bramatherium megacephalum; Sm, Samotherium major; Pr, Palaeotragus rouenii; Ba, Bohlinia attica; Gs, Giraffa sivalensis; Gc, Giraffa camelopardalis.
Figure 3 – Cladogram with geological age and dorsal view of C3 vertebrae. Reproduced from Danowitz et al. (2015) under Creative Commons CC-BY-4.0. Pe, Prodremotherium elongatum; Cs, Canthumeryx sirtensis; Oj, Okapia johnstoni; Gp, Giraffokeryx punjabiensis; Sg, Sivatherium giganteum; Bm, Bramatherium megacephalum; Sm, Samotherium major; Pr, Palaeotragus rouenii; Ba, Bohlinia attica; Gs, Giraffa sivalensis; Gc, Giraffa Camelopardalis.

The selection pressures that drove the evolution of the neck are unclear, and highly debated. The two main hypotheses based on observations in giraffes are:

  1. ‘Necks for food’ — individuals with long necks have better access to food from tall trees. These animals are likely to be healthier, and will tend to have more offspring.
  2. ‘Necks for sex’ — males use their necks in mating-associated sparring contests with other males. The individuals with longer necks are then more likely to breed.

Both ideas are supported by evidence, but it is difficult to imagine these selection pressures working in isolation. Some researchers have concluded that both mechanisms may have contributed to neck elongation.

Body size

Another character present in (but not confined to) Giraffidae is large body size. The earliest giraffid, Canthumeryx, was not large by today’s standards, but was probably larger than the other Miocene ruminants of its time. Modern giraffes are the largest living ruminants, with an average mass of 1,200 kilograms in males. But it is the sivatheres that are traditionally seen as the true giants of the giraffids.

Both historical and modern studies often suggest that the species Sivatherium giganteum had a body mass similar to  that of the African elephant (Loxodonta africana) — which can be more than 6,000 kilograms. In a study published this year, some colleagues and I instead estimated the body mass of S. giganteum to be between 857 and 1,812 kilograms, probably about 1,246 kilograms. This was calculated using a composite computer model of the skeleton, and an estimate of mass based on the animal’s volume  (Fig. 4). We considered this estimate to be at the lower end of the real range, as some individuals would have been considerably larger. Although the estimated body mass does not even approach the size of an elephant, we confirm that Sivatherium was indeed a huge animal, and it may not be possible for ruminants to get any bigger.

Skeletal reconstruction of Sivatherium giganteum, with minimum convex hull (purple) used to estimate body mass. Black bar is 1 metre. Taken from Basu et al. (2016) under Creative Commons CC-BY-4.0.
Figure 4 – Skeletal reconstruction of Sivatherium giganteum, with minimum convex hull (purple) surrounding the bones. The estimate of body mass was calculated using the volume of the convex hull. Black bar is 1 metre. Taken from Basu et al. (2016) under Creative Commons CC-BY-4.0.

Future of Giraffidae

It is common to find more than three giraffid species per Miocene fossil site. What was once a diverse group is now reduced to two species. In a sombre end, I would like to mention the conservation status of the living giraffids. Okapis are listed on the International Union for Conservation of Nature’s Red List as ‘Endangered’ and in decline. They are unfortunately only found in the dense forests of the Democratic Republic of the Congo — an area fraught with massive social and economic problems. No one fully knows how many individuals are left, because the presence of armed militia groups in this area makes research highly dangerous. Giraffes are in a similar state of decline, with a 40% reduction in total numbers over the past decade. Worryingly, as a species they are still listed as ‘Least concern’ on the Red List. With an estimated total wild population of 80,000 (down from 140,000 in 1998), we could be witnessing their terminal decline.

The way to prevent these extinctions is through more research (in basic science and conservation), and through social and economic change — both in the animals’ home countries and overseas. A world without giraffids seems unthinkable. But the near future could bring a time when school pupils learn about okapis and giraffes at the same time as they learn about dinosaurs and mammoths.

Reading list

Basu, C., Falkingham, P. L. & Hutchinson, J. R. The extinct, giant giraffid Sivatherium giganteum: skeletal reconstruction and body mass estimation. Biology Letters 12, 20150940 (2016). DOI: 10.1098/rsbl.2015.0940

Colbert, E. H. The relationships of the okapi. Journal of Mammalogy 19, 47–64 (1938 ). DOI 10.2307/1374281

Danowitz, M., Vasilyev, A., Kortlandt, V. & Solounias, N. Fossil evidence and stages of elongation of the Giraffa camelopardalis neck. Royal Society Open Science 2, 150393 (2015). DOI: 10.1098/rsos.150393

Ganey, T., Ogden, J. & Olsen, J. Development of the giraffe horn and its blood supply. The Anatomical Record 227, 497–507 (1990). DOI: 10.1002/ar.1092270413

Hamilton, W. R. The lower Miocene ruminants of Gebel Zelten, Libya. Bulletin of the British Museum of Natural History (Geology), London 21, 75–150 (1973 ).

Harris, J., Solounias, N. & Geraads, D. Giraffoidea. In Cenozoic Mammals of Africa (eds Werdelin, L. & Sanders, W. J.) 797–811 (Univ. California Press, 2010 ).

Hassanin, A., Delsuc, F., Ropiquet, A., Hammer, C., Jansen van Vuuren, B., Matthee, C., Ruiz-Garcia, M., Catzeflis, F., Areskoug, V. & Nguyen, T. T. Pattern and timing of diversification of Cetartiodactyla (Mammalia, Laurasiatheria), as revealed by a comprehensive analysis of mitochondrial genomes. Comptes Rendus Biologies 335, 32–50 (2012 ). DOI: 10.1016/j.crvi.2011.11.002

Janis, C. & Scott, K. The origin of the higher ruminant families with special reference to the origin of Cervoidea and relationships within the Cervoidea. American Museum Novitates 2893, 1–85 (1987 ).

Sánchez, I. M., Cantalapiedra, J. L., Ríos, M., Quiralte, V. & Morales, J. Systematics and evolution of the Miocene three-horned palaeomerycid ruminants (Mammalia, Cetartiodactyla). PLoS ONE 10, e0143034 (2015). DOI: 10.1371/journal.pone.0143034

Simmons, R. E. & Altwegg, R. Necks-for-sex or competing browsers? A critique of ideas on the evolution of giraffe. Journal of Zoology 282, 6–12 (2010 ).

Solounias, N. Family Giraffidae. In The Evolution of Artiodactyls (eds Prothero, D. R. & Foss, S. E.) 257–277 (Johns Hopkins Univ. Press, 2007 ).

Okapi – the endangered forest giraffe

Spot the extinction

1Structure and Motion Laboratory, Royal Veterinary College, London