Skip to content

What if, anything, is a hyeanodont?

September 28, 2015
The hyaenodont Apterodon macrognathus and possible relatives: the spotted hyena (if carnivoramorph), Cimolestes (if non-placental eutherian; feat Dinosaur Train) and the aardvark (if afrothere)

The hyaenodont Apterodon macrognathus and possible relatives: the spotted hyena (if carnivoramorph), Cimolestes (if non-placental eutherian; feat Dinosaur Train) and the aardvark (if afrothere)

“Creodont” is a complicated term. First coined in 1875 by Edward Drinker Cope, Creodonta was an order of mammals erected to classify several lineages of non-carnivoran carnivore mammals: first Oxyaenidae and Didymictis, Arctocyonidae, Leptictidae, Ambloctonidae, and Mesonychidae. Latter, William Diller Matthew expanded the term in 1909 to include yet another lineage of carnivorous mammals, Hyaenodontidae. Both authors considered creodonts to be either “insectivores” or a link between them and Carnivora, a first primitive attempt by mammals to be successful predators, replaced by their more “advanced” carnivoran cousins, in some schemes directly descended from them.

And so did many posterior authors.

But, with time, it became clear that “Creodonta” was nothing more than an artificial group, few if any of its members actually closely related at all. One by one, the lineages forming “Creodonta” were reassigned to other branches of the mammal phylogenetic tree – Didymictis became known as a stem-carnivoran, Arctocyonidae and Mesonychidae became understood as ungulate relatives, and Leptictidae drifts about -, or absorbed into other clades – Ambloctonidae became part of Oxyaenidae -, until, by 1969, only Oxyaenidae and Hyaenodontidae remained. These are to this day still grouped together, but only out of convenience/tradition, and it’s now widely agreed that they are not at all closely related, leaving “Creodonta” as an invalid wastebasket taxon (Rose & Archibald 2005, Morlo, M., Gunnell G., and P.D. Polly. 2009, Katharina Anna Bastl 2013).

So, with “Creodonta” not being valid clade, this leaves the question as to what exactly were oxyaenids and hyaenodontids.

Traditionally, most authors place them as part of Ferae, the same group that includes pangolins and modern carnivorans. However, not only is this interpretation simply a holdover from the “insectivore-carnivoran missing link” days, it is also becoming increasingly problematic, in the light of dental and post-cranial characteristics (Morlo, M., Gunnell G., and P.D. Polly. 2009). We will discuss this connection in relation to hyaenodontids later on.

For oxyaenids, little discussion has ever taken place. Generally, they are assumed -or rather, dismissed – to be somewhere in Laurasiatheria, either basally or as part of Ferae. Part of the reason why little discussion has been had is because their dentition is rather derived, making it rather hard to immediately connect them to other mammal groups known only from teeth and jaws.

For hyaenodontids, however, a wealth of discusion has taken place. Their diversity – spanning over 70 genera and covered a variety of ecological niches from cursorial predators to bone-crushers to aquatic piscivores to arboreal hunters and even fossorial forms -, as well as the fact that their eponymous genus, Hyaenodon, is a rather well studied prehistoric megafaunal icon, makes them relatively more accessible than oxyaenids. They are frequently included as outgroups in most phylogenetic Cenozoic mammal studies, and most discussion in regards to “creodont” affinities is focused on them.

So, over time, several alternative positions in the mammalian tree of life have emerged. Each bears significant implications not only to hyaenodont biology, but also to the diversity patterns of eutherian mammals as a whole.

Here, we will examine these numerous possibilities, their plausibility, and the implications that they have on our understanding of mammal diversity and evolutionary tendencies.

Hyaenodonts as part of Ferae


The most conservative of the proposed hyaenodont classifications. The interpretation of “creodonts” as part of Ferae first dates to McKenna et al 1975, who in fact coined the term “Ferae” to describe the clade containing Creodonta, Carnivora and Cimolesta. However, the “spirit” of this idea first dates to William Diller Matthew’s review of “Creodonta”, suggesting it to be a clade closely related to or leading to carnivoramorphs. To this day, most papers default to this classification, even as hyaenodonts and oxyaenids are no longer considered to be closely related.

The focal reason for an assumed relationship between hyaenodonts and carnivoramorphs are almost entirely based on dentition. As cited in McKenna 1975:

– The presence of carnassial shears;

– A large metastylar blade on the first molar;

– Large conical canines;

– A last premolar as a primary nonmolariform.

However, not only have these features evolved multiple times among mammals, they are also not basal to “creodonts” nor cimolestans nor carnivoramorphs (Rose & Archibald 2005, Morlo, M., Gunnell G., and P.D. Polly. 2009), and the elements involved are completly different within each group. For example, the carnassial teeth in carnivoramorphs are the fourth upper premolar and the first lower molar, while the hyaenodontids they are the second upper molar and the third lower molar. To add insult to injury, not only other dental synapomorphies have been found among carnivoramorphs and pholidotans that so far considered to be absent in hyaenodontids and oxyaenids (M.S. Springer et all 2003), but as well between laurasiatheres as a whole that are absent in “creodonts” (Morlo, M., Gunnell G., and P.D. Polly. 2009).

So, at least currently, the basis for a Hyaenodontidae + Carnivoramorpha clade seem to be null, and adherence to it is mostly out of conservative caution. In fact, considering how another clade contained in McKenna’s Ferae, Cimolesta, is now considered to not even be within Placentalia, it’s very likely that the current placement of hyaenodonts here is similar to the adherence to “Creodonta” as a whole, simply a game of convenience.

In the likelihood that these assessments are wrong and hyaenodonts are in fact part of Ferae, it would seem that this lineage of placental mammals dominated carnivorous niches in Eurasia, Africa and North America for most of the Cenozoic, contested only by carnivorous ungulates, crocodilians and possibly oxyaenids (if the latter aren’t part of Ferae themselves). It would seem that at least two independent waves of these mammals acquired specialised carnivorous habits, since the differences in hyaenodontid and carnivoran dentitions suggest that their adpatations for carnivory evolved idependently. Given the accepted african origin for hyaenodonts (Grohé 2012), it would be possible that they evolved from an early branch of Ferrae that became isolated in Africa during the Palaeocene; with this in mind, it possible that hyaenodontids diverged first from the rest of Ferrae, instead of a status as sister taxa to Carnivoromorpha.

Currently, there is no solid evidence for a Hyaenodontida + Carnivoramorpha clade

Currently, there is no solid evidence for a Hyaenodontida + Carnivoramorpha clade

Hyaenodonts as cimolestids/non-placental eutherians

Hyaenodonts as cimolestans. Note that cimolestid affinities in and of themselves are unclear, and that the cladogram is highly simplified.

Hyaenodonts as cimolestans. Note that cimolestid affinities in and of themselves are unclear, and that the cladogram is highly simplified.

A more radical proposition is that hyaenodonts not only aren’t closely related to carnivorans and oxyaenids, but also not placental mammals at all. Rather, that they are part of Cimolesta, a group of eutherian mammals outside of crown group Placentalia.

This idea was first proposed in McKenna 1975, where a connection between cimolestans and “creodonts” was proposed in Ferae, under the same synapomorphies. Later studies have firmly identified cimolestans as non-placental eutherians, and a connection between them and hyaenodonts (and oxyaenids) has largely been ignored.

However, two studies have independently posited that hyaenodonts are actually derived cimolestans:

– Polly 1996 proposes that hyaenodontids are part of Palaeoryctidae, which in their definition includes Cimolestes itself, on the basis of a similar non-molariform last premolar, a small facial expansion of the lacrimal, a present medial wall of auditory bulla and transversely wide upper molars with wide stylar shelves and large parastylar areas.

– Ausgustí 2002 makes a similar assessment in regards to “Creodonta” and Palaeoryctidae.

– Gheerbrant et all 2006 proposes a close relationship between hyaenodontids and cimolestids on the part of the paraconid and paracristid devleopment in molars 1-3 and anterior premolar morphology. The inclusion of ptolemaiidans, then considered pantolestids, also offers the possibility of a pantolestid identity for hyaenodonts.

Several posterior studies cite Gheerbrant, and indeed a cimolestid hyaenodont identity has been taken for granted almost as frequently as one as part of Ferae.

In light of the african origins of hyaenodonts, the presence of a high diversity of cimolestids in the early Cenozoic of North Africa is certainly conspicuous. Africa in the early Cenozoic was an island continent, with only occasional interchanges with Europe, so in isolation a lineage of these mammals could have evolved to explore carnivorous niches, and these succeded in not only becoming the dominant mammalian predators of Africa, but indeed most landmasses aside of South America, Antarctica and Australia. If hyaenodonts are cimolestids, some points of interest are to be considered:

– Hyaenodonts would have been the most diverse lineage of unambiguous cimolestan mammals. They would also be the longest spanning lineage of Cenozoic non-placental eutherians, as the most recent species became extinct in the late Miocene, some 22 million years after the youngest pantolestids died out. This would make them also among the youngest non-placental mammals in landmasses outside of South America and Australia, alongside herpetotheriid metatherians.

– Because the possible cimolestid Maelestes is known to have epipubic bones, a cimolestid identity for hyaenodonts could have drastic implications for their reproductive biology. Like most non-therian mammals, the incapacity of expanding the torso would mean that hyaenodonts would give birth to miniscule, undeveloped young, possibly carried about in pouches as in modern marsupials. Alternatively, given the lack of epipubic bones in known hyaenodonts, it would mean that these animals either lost their epipubic bones and developed full placentas independently from placental mammals, or that the cimolestid identity for Maelestes is in question, and that it is a more basal eutherian mammal.

– As we see below, the fact that hyaenodonts display a late tooth eruption could mean that this is an eutherian synapomorphy and not a trait exclusive to atlantogenatans.

However, a cimolestan identity is not without problems. For one thing, Cimolesta may be paraphyletic or even polyphyletic, and this is most evident and problematic in relation to the studies uniting Hyaenodontidae to them:

– Palaeoryctidae, the clade to which Polly 1996 and Ausgustí 2002 reffered hyaenodontids to, is now known to be polyphyletic, its individual members scattered across Eutheria outside of Placentalia and Eulipotyphla (Manz 2015)

– Many of the north african “cimolestids” mentioned in Gheerbrant 2006, such as ptolemaiidans, are now thought to be afrotheres (Cote et all 2007, Erik R Seiffert 2007).

With this in mind, the connection of hyaenodontids to cimolestids is uncertain since the synapomorphies connecting them cannot be taken for granted, and so they are pulled in different directions. Once again, it is an instance where there’d be an artificial taxon based on attributes present on several distinct mammalian clades.

Therefore, a cimolestid identity for Hyaenodontidae is currently difficult to discern.

Currently, a cimolestid identity for Hyaenodontidae seems invalid as well.

Currently, a cimolestid identity for Hyaenodontidae seems invalid as well.

Hyaenodonts as afrotheres

Hyaenodonts as afrotheres.

Hyaenodonts as afrotheres.

A final possibility is that of hyaenodonts as part of Afrotheria. This hypothesis has been briefly alluded to in several papers (Katharina Anna Bastl 2013, Grohé 2012, Cote et all 2007), but never formally discussed in literature. Informal discussions are however common, and this has been the subject of speculation on the part of researchers since Afrotheria has been recognised as a clade in the 2000’s.

The reasons for this hypothesis are as it follows:

– Hyaenodonts are of african origin. This is a current consensus by most researchers on the clade, as seen above: the earliest known hyeanodonts are found in Africa, and it appears to be there where the focus of hyaenodont diversity is located. Considering the immense diversity of afrotheres in several ecological niches from aquatic piscivores to massive terrestrial herbivores – including a lineage of fairly large animalivorous afrotheres, the ptolemaiidans -, it is feasible that hyaenodonts are part of the afrothere assemblage.

– Hyaenodonts display a late tooth eruption (Katharina Anna Bastl 2013). This is one of the signature synapomorphies of Atlantogenata (Asher & Lehmann 2008, Asher et al. 2009).

– Gheerbrant et all 2006 connects hyaenodonts to ptolemaiidans, which are now thought to be afrotheres. Likewise, many “cimolestids” in that study may be afrotheres as well. Likewise, both Ausgustí 2002 and Polly 1996 connect them to Eoryctes, sometimes recovered as a relative of modern otter-shrews (Wible 2007), though given latter assessments as a solenodontid (Manz 2015) this may be questionable.

Compared to the Ferrae and cimolestan hypotheses, an afrotherian identity seems rather unexplored, but seems far more solidly built, taking into consideration both the biogeographical and anatomical characteristics of the clade. In particular, examining the patterns of tooth eruption in placental mammals may be extremely relevant in figuring out their phylogenetic positions (Katharina Anna Bastl 2013, Asher et al. 2009), in which case it is the most convincing argument in regards to the connection of hyaenodonts to afrotheres.

Within Afrotheria, hyaenodonts could possibly fit in a number of positions:

– As basal representatives (outside of core-Afrotheria or even core-Atlantogenata). This is supported by possibly primitive traits such as arboreal tendencies, thought to be basal to placental mammals (Floréal Solé 2015).

– As afrosoricids. Even if the similarities to Eoryctes are possibly a red-herring, a relation to mainland african afrosoricids like the otter-shrews is a possibility. This is particularly since one of the most basal hyaenodont lineages, Apterodontinae, bears similar semi-aquatic habits (Grohé 2012). An afrosoricid identity for hyaenodonts would in several ways be a parallel to the diversity of dasyuromorphs in Australia: in both Paleocene Africa and Neogene Australia, the dominant carnivorous mammals are closely related to the insectivore analogues.

– As part of the Macroscelida + Tubulidentata clade. Hyaenodonts are compared strongly to the ptolemaiidans (Gheerbrant et all 2006), now thought to be stem-aardvarks (Cote et all 2007, Erik R Seiffert 2007). It is tempting to suggest that hyaenodonts and ptolemaiidans formed the afrothere analogue of Carnivora, being the only known afrothere lineage to uniquely specilise in macropredatory habits. In this case, the insectivorous and carnivorous mammal guilds are represented by distinct clades, a similar arrangement to that seen not only in most of the Cenozoic ecosystems in Laurasia and North America, but also in South America. In any circumstance, the only living representative sof this hypothetical clade would be the specialised insectivorous aardvarks, nested among the more diverse ptolemaiidans, a specialised progression which is consistent with the hypothesised competitive displacement on the part of carnivorans.

Much as with a cimolestan identity, an afrothere relation also posits interesting questions on hyaenodont reproductive biology. Modern atlantogenatans do not have external testicles, and tenrecs are unique among therian mammals in having a true cloaca. Given the selective pressures that lead to the externalisation of genitals in therian mammals, whereas hyaenodonts have an internal or external male genital condition could impact our understanding of how and why this occured in placental mammals. In particular, some hyaenodont species had cursorial habits, which goes hand-in-hand with the galloping hypothesis (Chance 1996).


In summary, while currently underexplored, an afrotherian identity for Hyaenodontidae seems to be the most strongly supported by availiable evidence. This in turn has multiple implications on the biology of hyaenodonts and the nature of their diversity.

Hyaenodonts ar emost likely afrotheres.

Hyaenodonts ar emost likely afrotheres.


Rose, Kenneth David; Archibald, J. David (2005). The Rise of Placental Mammals: Origins and Relationships of the Major Extant Clades. Baltimore: Johns Hopkins University Press.

Morlo, M., Gunnell G., and P.D. Polly. 2009. What, if not nothing, is a creodont? Phylogeny and classification of Hyaenodontida and other former creodonts. Journal of Vertebrate Paleontology 29(Supplement 3): 152A.

Katharina Anna Bastl, First evidence of the tooth eruption sequence of the upper jaw in Hyaenodon (Hyaenodontidae, Mammalia) and new information on the ontogenetic development of its dentition, Paläontologische Zeitschrift (Impact Factor: 1.1). 10/2013; 88:481-494. DOI: 10.1007/s12542-013-0207-z

McKenna, M. C. (1975). “Toward a phylogenetic classification of the Mammalia”. In Luckett, W. P.; Szalay, F. S. Phylogeny of the Primates. New York: Plenum. pp. 21–46.

Camille Grohé, Michael Morlo, Yaowalak Chaimanee, Cécile Blondel, Pauline Coster, Xavier Valentin, Mustapha Salem, Awad A. Bilal, Jean-Jacques Jaeger, and Michel Brunet, New Apterodontinae (Hyaenodontida) from the Eocene Locality of Dur At-Talah (Libya): Systematic, Paleoecological and Phylogenetical Implications, Published online 2012 Nov 21. doi: 10.1371/journal.pone.0049054

Amrine-madsen, H.; Koepfli, K.P.; Wayne, R.K.; Springer, M.S. (2003). “A new phylogenetic marker, apolipoprotein B, provides compelling evidence for eutherian relationships”. Molecular Phylogenetics and Evolution 28 (2): 225–240. doi:10.1016/S1055-7903(03)00118-0. PMID 12878460.

Polly, P. D. 1996. The skeleton of Gazinocyon vulpeculus gen. et comb. nov. and the cladistic relationships of Hyaenodontidae (Eutheria, Mammalia). Journal of Vertebrate Paleontology 16 (2): 303-319.

Agustí, Jordi; Antón, Mauricio (2002). Mammoths, Sabertooths, and Hominids: 65 Million Years of Mammalian Evolution in Europe. New York: Columbia University Press. ISBN 0-231-11640-3.

Gheerbrant, E., M. Iarochene, M. Amaghzaz & B. Bouya. 2006. Early African hyaenodontid mammals and their bearing on the origin of the Creodonta. Geological Magazine 143 (4): 475-489.

Cote S, Werdelin L, Seiffert ER, Barry JC (March 2007). “Additional material of the enigmatic Early Miocene mammal Kelba and its relationship to the order Ptolemaiida”. Proc Natl Acad Sci USA. 104 (13): 5510–5. Bibcode:2007PNAS..104.5510C. doi:10.1073/pnas.0700441104. PMC 1838468. PMID 17372202.

Seiffert, Erik R (2007). “A new estimate of afrotherian phylogeny based on simultaneous analysis of genomic, morphological, and fossil evidence”. BMC Evolutionary Biology 7 (1): 224. doi:10.1186/1471-2148-7-224. PMC 2248600. PMID 17999766.

Carly Manz, New partial skeletons of Palaeocene Nyctitheriidae and evaluation of proposed euarchontan affinities, Biology letters (Impact Factor: 3.25). 01/2015; 11(1):20140911. DOI: 10.1098/rsbl.2014.0911

Asher, R. J., Bennett, N. & Lehmann, T. 2009. The new framework for understanding placental mammal evolution. BioEssays 31, 853-864.

Asher, R. J. & Lehmann, T. 2008. Dental eruption in afrotherian mammals. BMC Biology 2008, 6: 14.

Floréal Solé, Eli Amson, Matthew Borths, Dominique Vidalenc, Michael Morlo, Katharina Bastl, A New Large Hyainailourine from the Bartonian of Europe and Its Bearings on the Evolution and Ecology of Massive Hyaenodonts (Mammalia) Published: September 23, 2015DOI: 10.1371/journal.pone.0135698

M.R.A. Chance, Reason for externalization of the testis of mammals, Journal of Zoology
Volume 239, Issue 4, pages 691–695, August 1996

2 Comments leave one →
  1. Gray Stanback permalink
    October 10, 2015 6:41 pm

    Gotta love the picture you used for Cimolestes.

  2. November 1, 2015 2:48 am

    This is a really fascinating hypothesis that only just now came onto my radar. I would really like to see some sort of analysis performed. Hyaenodonts were weird animals. I don’t think we can get our heads wrapped around how strange they were.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: