Skip to main content
NCBI home page
Journal of Anatomy logoLink to Journal of Anatomy
. 2015 Aug 17;227(3):277–285. doi: 10.1111/joa.12348

Endocasts and brain evolution in Anthracotheriidae (Artiodactyla, Hippopotamoidea)

Ghislain Thiery 1, Stéphane Ducrocq 1
PMCID: PMC4560562  PMID: 26278931

Abstract

Anthracotheres are a fossil family of ‘Suiformes’ from the Old World, North and Central America. They are known from the middle Eocene to the late Pliocene, and are suggested to be the stem group of Hippopotamidae. Yet, their soft anatomy remains poorly known. In this study we describe the virtual endocast of the late Oligocene anthracothere Microbunodon minimum, reconstructed using microtomography, as well as the natural endocast of Merycopotamus medioximus from the late Miocene. These are the first anthracothere endocasts ever described. Particular attention is given to the relative proportions of the brain, the neocortex, the cerebellum and the olfactory bulbs. The ‘backward shift’ of the pituitary of M. minimum, and the possible presence of a K lobe in M. medioximus, is discussed. Previous statements that some endocranial characters were subject to convergence among mammals are also corroborated.

Keywords: anthracothere, cerebellum, encephalization quotient, K lobe, olfactory bulbs, pituitary

Introduction

Anthracotheres are fossil ‘Suiforme’ artiodactyls. They are known from the middle Eocene to the late Pliocene, in Eurasia, Africa, North America (Lihoreau & Ducrocq, 2007) and Central America (Rincon et al. 2013). Once thought to be essentially semi-aquatic mammals (e.g. Cuvier, 1822; Falconer & Cautley, 1836), it was later revealed that they occupied a greater diversity of habitats with corresponding behaviors (e.g. Lihoreau, 2003; Lihoreau & Ducrocq, 2007).

Most of the authors (Gentry & Hooker, 1988; Van der Made, 1999; Boisserie et al. 2005, 2010; Boisserie & Lihoreau, 2006; Orliac et al. 2010) consider that the Bothriodontinae (one of the three subfamilies of Anthracotheriidae) are the stem group of hippopotamids. The recent discovery of Epirigenys lokonensis in Kenya (Lihoreau et al. 2015) supports this hypothesis. On the other hand, Pickford & Morales (1989) suggested that the sister taxon of hippopotamids belongs to Palaeochoeridae (‘Old World Tayassuidae’), whereas Pearson (1929) and Theodor & Foss (2005) supported a cebochoerid origin. In addition, a consensus has not been reached on the position of genera, families and subfamilies within Hippopotamoidea sensu Van der Made, 1999 (i.e. ‘Anthracotheriidae’ + Hippopotamidae). For instance, the Microbunodontinae subfamily has been described alternatively as the sister taxon of Anthracotheriinae (Lihoreau & Ducrocq, 2007) or occupying a basal position in Hippopotamoidea (Orliac et al. 2010).

In this context, the endocranial features of the Anthracotheriidae (Artiodactyla) have never been studied. Endocasts are reliable proxies of the shape of the brain covered by the dura mater (Radinsky, 1975, 1977; Fournier et al. 2011). They can provide indirect data on the behavior, through the relative size of some structures (e.g. Jerison, 2009, 2012). Also, the morphology of the mammalian brain is intricately linked to mammal phylogeny (Voogd et al. 1998). Thus, some authors attempted to use endocranial characters in phylogenetic reconstructions (Danilo, 2013) or to infer the taxonomic status of isolated endocasts (Macrini, 2009).

In this study, the virtual endocast of Microbunodon minimum (Microbunodontinae, Anthracotheriidae) and the partial endocast of Merycopotamus medioximus (Bothriodontinae, Anthracotheriidae) are described. Microbunodon minimum was a small, slender anthracothere that lived during the late Oligocene in Europe (Lihoreau & Ducrocq, 2007). Merycopotamus medioximus was much larger and likely had a gregarious semi-aquatic lifestyle. It lived during the late Miocene in Pakistan, Iraq and Thailand (Lihoreau & Ducrocq, 2007; Lihoreau et al. 2007). As these are the first anthracothere endocasts ever described, some of the features reported in this study might be helpful in future studies involving the brains of fossil and extant Hippopotamoidea.

Material and methods

The first endocast comes from LM1967MA300, the skull of a male anthracothere assigned to M. minimum (Brunet, 1968; Cabard, 1976). It was collected from the latest Oligocene deposits of La Milloque (Lot-et-Garonne, France; European mammalian reference level MP29; Mennecart et al. 2012) and housed at the iPHEP. In addition, the endocast of Y13310, a skull assigned to M. medioximus (Lihoreau et al. 2004, 2007), is described here. It was collected at the Dhok Mila locality (Potwar Plateau, Pakistan), and was dated from the late Miocene, between 10.4 and 8.6 Myr (Lihoreau et al. 2004). The skull belongs to the GSP collections and is housed at the Peabody Museum, Harvard University.

To access its inner part, the skull of LM1967MA300 was scanned using micro-computed tomography devices NCT – VISCOM X8050 scanner at the University of Poitiers. The scan resulted in 1004 slices with a pixel size of 146.9 × 146.9 μm. Slices are 86.9 μm thick. The endocast surface was extracted and reconstructed using avizo® fire and geomagic studio® 2013. A pdf version of the endocast surface can be consulted in supporting information (Fig. S1). The endocast of Y13310 was not virtually reconstructed, as a natural endocast made the internal part of the skull already accessible.

Following previous work (Macrini, 2009; Orliac & Gilissen, 2012) six linear measurements were asserted on LM1967MA300 (Fig. 1). The endocast was oriented so that the horizontal plane would link the frontal and occipital poles of the telencephalon (Holloway & Shapiro, 1992). Linear, surface and volumetric measurements were taken using geomagic studio. When it was possible, comparative linear measurements of other endocasts were taken as well from the literature or using a caliper (Table 1).

Fig. 1.

Fig. 1

Open in a new tab

Detail of the linear measurements on LM1967MA300, adapted from Macrini (2009). BL, brain length; CLH, cerebellum height; CLW, cerebellum width; CRH, cerebrum height; CRL, cerebrum length; CRW, cerebrum width.

Table 1.

Measurements of the virtual endocast of LM1967MA300, and of other fossil artiodactyls

Specimen Taxon BL (mm) CRL (mm) CRW (mm) CRH (mm) CLW (mm) CLH (mm) w h
MNHN 33-1967 Tapirulus 43.5 30.7 23.5 22.4 17.8 17.8 0.757 0.795
MNHN 42-1971 Mixtotherium 53.3 31.9 28.8 21.4 28.4 20.7 0.986 0.967
LM1967MA300 Microbunodon 70.6 41.8 39.9 23.9 26.2 26.3 0.657 1.100
V (Macrini, 2009) Merycoidodon NA 45 47 33 36 32 0.766 0.970
IV (Macrini, 2009) Merycoidodon NA 46 46 31 35 33 0.761 1.065
III (Macrini, 2009) Merycoidodon NA 44 46 33 35 34 0.761 1.030
II (Macrini, 2009) Merycoidodon NA 44 45 33 33 33 0.733 1.000
Cast I (Macrini, 2009) Merycoidodon NA 47 42 21 27 32 0.643 1.524
Cast III (Macrini, 2009) Merycoidodon NA 38 40 23.5 22 24 0.550 1.021
MNHN coll. Piveteau B Diplobune 70.4 44.5 39.9 31.2 36.9 28.4 0.925 0.910
MNHN coll. Piveteau without no. Diplobune 89.1 49.8 42.3 37 35.9 34.9 0.849 0.943
mMNHN coll. Piveteau without no. Dacrytherium 65.7 38 37.8 27.4 31.3 26.4 0.828 0.964
MNHN 34-1967 Cebochoerus 42.4 25.8 23.4 21.6 18.9 21 0.808 0.972
MNHN 42-1971 Cainotherium 37.8 28.5 22.6 18.9 18.7 16.9 0.827 0.894
TMM 40209-198 (Macrini, 2009) Bathygenys 52.7 28.5 29.1 21.7 21.2 22.5 0.729 1.037
MNHN coll. Piveteau without no. Anoplotherium 95.5 60.7 52.2 42.4 38.7 40.1 0.741 0.945
MNHN without no. Anoplotherium 90.2 59.5 49.5 38.2 35.5 36 0.717 0.942
Open in a new tab

BL, brain length; CLH, cerebellar height; CLW, cerebellum width; CRH, cerebrum height; CRL, cerebrum length; CRW, cerebrum width; h, cerebellar height index = CLH/CRH; w, cerebellar width index = CLW/CRW. The flexure was not studied, as the pituitary was not visible on every fossil; however, it was evaluated to 10.9° in LM1967MA300.

Results

The endocast of M. minimum

The endocast of LM1967MA300 (Fig. 2) is slightly asymmetric, and suffered from some deformation of the dorsal anterior part and the ventral part of the telencephalon. Nevertheless, it is well preserved and allows a good description of what would have been the brain of M. minimum. Nomenclature follows studies of Eocene and Oligocene taxa (Dechaseaux, 1961, 1969a; Orliac & Gilissen, 2012).

Fig. 2.

Fig. 2

Open in a new tab

Labeled endocast of Microbunodon minimum (LM1967MA300) in dorsal (A), lateral (B) and ventral views (C). Ars, anterior rhinal sulcus; Cb, cerebellar hemisphere; Crs, cruciate sulcus; Cos, coronal sulcus; Dsd, Dorsum sellae depression; Es, ectosylvia featuring a small structure, interpreted as a possible sylvian complex; Fj, Foramen jugulare (IX); Fl, Foramen lacerum; Fop, Foramen optica (II); For, Foramen orbitorotundum (III, IV, V1, V2 and VI); Fov, Foramen ovale (V3); G3, Gyrus 3; Iam, internal auditory meatus; Ls, lateral sulcus; Mo, Medulla oblongata; Nc, neocortex; Ob, olfactory bulb; Op, olfactory peduncle; Ot, olfactory tubercle; Pa, paraflocculus; Pc, piriform cortex; Pi, pituitary; Prs, posterior rhinal sulcus; Sas, sagittal sinus; Sps, superior petrosal sinus; Sss, suprasylvian sulcus; Ts, transversal sinus; Ve, vermis; II-IX refer to cranial nerves. Scale bar: 1 cm.

The olfactory bulbs could not be entirely segmented, as the cribriform plate is damaged and the right bulb is sliced in two by a fragment of the inner surface of the skull. The anterior limit of the cavity that contains the bulbs was estimated following Macrini et al. (2006), i.e. at the level of the circular fissure (sensu Rowe, 1996a,b). They are rather large, as is the rest of the rhinencephalon (i.e. the olfactory tubercles and the piriform cortex). The telencephalon of Microbunodon stands dorsally higher than the olfactory bulbs, contrasting with the lower position in Eocene genera such as Diacodexis (the most primitive artiodactyl; Orliac & Gilissen, 2012), Cebochoerus (Cebochoeridae; Dechaseaux, 1969a) and Tapirulus (Choeropotamidae; Dechaseaux, 1969a). However, olfactory peduncles are clearly visible in the dorsal view (Fig. 2A). This is not the case for the endocasts of many other Oligocene artiodactyls, such as Bathygenys (Oreodontiform; Macrini, 2009; Orliac & Gilissen, 2012) or Cainotherium (Dechaseaux, 1961, 1969a). In these genera, the telencephalon abuts or even re-covers the olfactory bulbs.

The rhinencephalon and the neocortex are separated by the anterior and posterior rhinal fissures. Both sulci meet to form a reversed V-shaped groove that we interpret as an ectosylvia (Fig. 2B). This phenomenon is referred as the operculization of the brain, i.e. the fact that the area delimited by the rhinal fissures, the suprasylvia and the presylvia becomes progressively separated in two distinct gyri by the ectosylvia (Dechaseaux, 1968). In LM1967MA300, there is a partial operculization, as the ectosylvia only crosses the first half of the gyrus. A small structure pointing toward the anterior part of the telencephalon can be observed on the upper part of the ectosylvia (Fig. 2B). This process can be seen on the slices (Fig. 3) and is present on both sides of the endocast. Consequently, it is unlikely that this structure is due to an artifact or the deformation of the skull. A similar feature on the endocast of Anoplotherium (Anoplotheriidae) has been attributed to blood vessels (Dechaseaux, 1961).

Fig. 3.

Fig. 3

Open in a new tab

Coronal CT slices of LM1967MA300 in cranial view: (A) located at the back of the telencephalon, (B) and (C) before the piriform cortex, and (D) at the beginning of the olfactory bulbs. The processes on the ectosylvia are designated with white arrows. Because the endocast is slightly asymmetric, these structures are not situated on the same slices. Scale bar: 1 cm.

The suprasylvia extends backward and forward, where a short sulcus assigned to the coronal sulcus connects to its anterior part. Such a connection is present in all the artiodactyls that have a coronal sulcus, except Mixtotherium, the anoplotheriids or the girafids, in which the coronal connects to the lateral sulcus (Dechaseaux, 1969b, 1973). There is one more longitudinal pair of fissures that we interpret as a lateral sulcus. Together with the suprasylvia, it delimits an elongated gyrus 3. This contrasts with the almond-shaped gyrus of the first artiodactyls such as Diacodexis, Tapirulus, Cebochoerus and Helohyus (Helohyidae) (Orliac & Gilissen, 2012). On the forepart of the telencephalon, there is a short, medio-laterally oriented fissure that could be interpreted as a cruciate sulcus. It is deeply grooved and seems to extend backward.

The cerebellum of LM1967MA300 is higher than the cerebrum and shows a protruding vermis, a condition found in most Eocene ungulates. On the large cerebellar hemispheres, a small paraflocculus is almost totally hidden by a protruding superior petrosal sinus. The transverse sinus is well developed, covering the area around the mesencephalon. It is therefore impossible to tell whether the midbrain is exposed in dorsal view or if the cerebral hemispheres about the cerebellum.

The cranial foramina of LM1967MA300 was already described by Cabard (1976). They are very similar to the condition found in most of the artiodactyls. However, the scan revealed the presence of a foramen lacerum hidden under the auditory bulla. In addition, the optic nerves (II) of M. minimum seem to diverge next to the optic chiasma, going through two distinct bony channels. In basal artiodactyls like Diacodexis, optic nerves usually go through the same channel (Orliac & Gilissen, 2012). In LM1967MA300, the optic nerves seem rather large and wide. However, a proper quantification of optic nerves size across artiodactyls is still needed for comparisons.

A short depression sits right in front of the optic chiasma (Fig. 2C). The pituitary should be located between this depression and the dorsum sellae depression. Actually, it is quite difficult to spot, either because it is internalized or because the venous sinuses of this region (i.e. the Willis polygon) are masking its limits. Dechaseaux (1969a) invoked the latter reason to explain visibility differences of the pituitary. Nevertheless, the position of the pituitary contrasts with that in other artiodactyls. It is somewhat further from the optic chiasma in LM1967MA300. According to Dechaseaux (1961), this ‘backward shift’ can be found in Choeropsis liberiensis and in the fossil hippopotamid Hippopotamus protamphibius, but not in Hippopotamus amphibius.

The endocast of M. medioximus

In contrast to LM1967MA300, the endocast of Y13310 (Fig. 4) mainly preserves its dorsal part. We carefully checked that the visible sulci could not be mistaken with the bone sutures. Nomenclature follows recent studies on the pig brain (Okada et al. 1999; Lind et al. 2007). While suids are not the closest extant relative of Bothriodontinae, their brain anatomy is accurately described. Conversely, ruminants like the sheep have a very different brain morphology, and there is no recent and accurate nomenclature of the hippopotamus brain anatomy.

Fig. 4.

Fig. 4

Open in a new tab

Labeled endocast of Merycopotamus medioximus (Y13310) in dorsal (A,B) and right lateral view (C,D). (E) Brain of a Choeropsis liberiensis (UPMC 1958-19, MNHN of Paris) in right lateral view. The K lobe is figured in gray. Aes, anterior ectosylvia; Cb, cerebellum; Cos, coronal sulcus; Crs, cruciate sulcus; Ds, diagonal sulcus; Ls, lateral sulcus; Ob, olfactory bulb; Pes, posterior ectosylvia; Prs, posterior rhinal sulcus; Sas, sagittal sinus; Sss, suprasylvian sulcus; Ts, transverse sinus. Scale bar: 1 cm.

Although the right, very small olfactory bulb seems to be preserved, most of the rhinencephalon is lacking. At least six pairs of sulci can be identified in dorsal view. The most anterior one is interpreted as a cruciate sulcus. Its shape is very similar to what can be seen on a Sus scrofa brain. A short, steep diagonal sulcus is visible by the cruciate side. It is perpendicular to it and seems to expand laterally on the left hemisphere. The cruciate extends very far backward, until it reaches a coronal sulcus that is parallel to the long axis of the brain and a fissure interpreted as an anterior ectosylvia. The posterior part of the coronal sulcus seems to connect with a faint suprasylvia. The fifth fissure is a short lateral sulcus partially hidden under a large sinus on the right hemisphere.

In lateral view (Fig. 4C), two more fissures can be seen. A posterior rhinal sulcus is visible on the right side. It cannot be observed on the left side because this part of the endocast is hidden under a small piece of bone. We interpret the second fissure as a posterior ectosylvia. Both the anterior and posterior ectosylvia surround an additional gyrus. In some artiodactyls (hippopotamids, suids), the upper part of the ectosylvia splits in two and forms a ‘sylvian complex’. This splitting sometimes results in the formation of an additional gyrus, surrounded by an anterior and a posterior ectosylvia: the K lobe (Dechaseaux, 1961; Fig. 4D). Both extant hippopotamuses species have an extended K lobe. Thus, the presence of this structure in a crown anthracothere is not unlikely.

The cerebellum of Y13310 seems to be preserved. A protruding vermis stands at the posterior end of the endocast. However, the limit between the cerebellum and the cerebrum is blurred by a large transverse sinus, suggesting that the cerebellum size may be underestimated (Fig. 4B).

Brain proportions in anthracotheres

To evaluate the relative brain size of M. minimum, the encephalization quotients (EQ) of Jerison (1973) and Eisenberg (1981) were calculated. Their values are 0.29 and 0.30, respectively (the estimation of the weight of M. minimum is from Lihoreau (2003). LM1967MA300 shows one of the smallest EQ for either value when compared with fossil and extant artiodactyls (Fig. 5), close to that of H. amphibius and C. liberiensis. In contrast, Tapirulus hyracinus, S. scrofa or Tayassu pecari display much higher values. However, one should remain cautious, since no other anthracothere EQ was available.

Fig. 5.

Fig. 5

Open in a new tab

EQs of Microbunodon minimum, as well as some Paleogene and extant artiodactyls. Data come from Orliac & Gilissen (2012), while the weight estimation of M. minimum comes from Lihoreau (2003).

The relative size of the brain does not seem to increase from Paleogene to extant species. This is consistent with recent works on the EQ of Artiodactyla (Shultz & Dunbar, 2010; Orliac & Gilissen, 2012). There is evidence that higher EQ might be associated with more developed social behavior among mammals, including Artiodactyla (Radinsky, 1975; Shultz & Dunbar, 2006, 2010). Despite its low EQ values, however, female H. amphibius has been considered as very social (Blowers, 1996; to our knowledge, this is the only study on this topic).

The neocortex is the part of the mammalian brain where the most advanced cerebral activities take place. Following Jerison (2009, 2012), the neocorticalization (i.e. extension of the neocortex surface over the total brain surface) of LM1967MA300 has been estimated at 34% of the brain excluding olfactory bulbs. The neocorticalization among extant ‘Ungulates’ averages 40–70% (Voogd et al. 1998) – which means that M. minimum had a less developed neocortex. However, the neocorticalization of two Paleogene artiodactyls, Anoplotherium commune and Bathygenys reevesi (Oreodontiforms, Artiodactyla), averages 30% (Jerison, 2009). The neocortex of M. minimum thus does not exceed the range observed in contemporaneous artiodactyls. This is consistent with previous studies suggesting that the neocortex folding was prone to convergence among mammals (Dechaseaux, 1968; Voogd et al. 1998). Because the endocast of M. medioximus was not complete, the relative size of its neocortex could not be estimated.

Although it is probably underestimated, the volume of the olfactory bulbs of M. minimum averages 783.5 mm3. This represents 2.8% of the overall brain volume. Microbunodon minimum had a rather large rhinencephalon compared with extant artiodactyls, and likely had a developed sense of smell. However, its olfactory bulbs are rather small compared with those of other Eocene and Oligocene taxa (Fig. 6). The ecology of M. minimum has already been studied through its dental and post-cranial remains (Lihoreau, 2003). It was probably a frugivorous browser, living in scarce to medium-dense equatorial forest. A moderately developed olfaction could have helped it to retrieve ripe fruit on the forest ground. In contrast, M. medioximus seems to have had very small olfactory bulbs (Fig. 4C,D).

Fig. 6.

Fig. 6

Open in a new tab

Proportions of the olfactory bulbs of LM1967MA300, compared with other Paleogene artiodactyls. Data come from Orliac & Gilissen (2012), and from Macrini (2009) for Bathygenys.

We also evaluated the proportions of the cerebellum of M. minimum. We used a width index (w = cerebellar width/cerebral width) and a height index (h = cerebellar height/cerebral height). We compared them with those of available Paleogene artiodactyls (Fig. 7). No length index was calculated, because data were not available for every taxon. The cerebellum of M. minimum is very narrow with a very low w, as for the Oligocene Merycoidodon. In contrast, the height index did not allow for a good segregation of specimens. There is a general mammalian trend to the decrease of the cerebellum size through time (Orliac et al. 2012), with some notable exceptions (Rilling & Insel, 1998). The cerebellum of M. medioximus is not complete, but it seems to be relatively much smaller. Thus, the cerebellum of Anthracotheriidae likely follows this trend.

Fig. 7.

Fig. 7

Open in a new tab

Proportions of the cerebellum of LM1967MA300, compared with the endocasts of Paleogene artiodactyls. Oligocene specimens (in black) show a narrower, smaller cerebellum (with the exception of Cainotherium). See Table 1 for measurements.

Discussion

Given the lack of fossils and the absence of extant representatives of anthracotheres for comparison, some of the features described here might be subject to different interpretation. Intraspecific variation, pathology, injuring, postmortem deformation, or irrelevance of the nomenclature choice is possible. Nevertheless, two characters of LM1967MA300 differ from those in other contemporaneous fossil artiodactyls. The first one is the operculization of the brain. The only Oligocene artiodactyl genera showing an ectosylvia are Microbunodon and Anoplotherium (Anoplotheriidae). Although anoplotheriids have disputed phylogenetic affinities (Heissig, 1993; McKenna & Bell, 1997; Erfurt & Métais, 2007), they differ from anthracotheres in several aspects, including cerebral ones (e.g. sulcus to which the coronal sulcus is linked). Therefore, the operculized brain is likely an independent synapomorphy of Hippopotamoidea. The fact that numerous other families develop an ectosylvia after the Oligocene is probably a parallel adaptation, as suggested by Dechaseaux (1968, 1969b, 1973).

The second feature is the ‘backward shift’ of the pituitary. Apart from M. minimum, it has only been attested in the brains of C. liberiensis and of the fossil hippopotamid H. protamphibius (Dechaseaux, 1961). However, such a condition can also be observed (although it has not been described) in some other artiodactyls, e.g. Dacrytherium (Anoplotheriidae; Dechaseaux, 1969a). Given the basal position of Microbunodon within Hippopotamoidea, this character is expected to be found in all Hippopotamidae and most of the ‘Anthracotheriidae’. The absence of a ‘backward shift’ of the pituitary in H. amphibius would mean that it has been secondarily lost. This apparent loss might be the consequence of the orbits dorsalization and a subsequent backward shift of the optic nerves in Hippopotamus (Dechaseaux, 1961). A proper methodology to quantify this ‘backward shift’ has yet to be developed and applied on a large array of taxa.

In extant Hippopotamidae, the K lobe is extremely extended and covers most of the actual ectosylvia (Dechaseaux, 1968; G Thiery, personal observation). On the endocast of LM1967MA300, there is neither a visible K lobe nor a sylvian complex. However, the brain of an S. scrofa fetus (e.g. MNHN 1930-350, G Thiery, personal observation) shows a process similar to the structure observed in M. minimum (Figs 2B and 3), which is not a blood vessel. A K lobe has been identified on the brains of some domestic pigs (Dechaseaux, 1968). This process might represent the beginning of a sylvian complex folding. On the other hand, M. medioximus shows a structure interpreted as a K lobe (Fig. 4D). While this structure is only visible on the right side, an anterior and a possible posterior ectosylvia can be seen on the left side. The presence of a K lobe on this fossil could imply that at least crown anthracotheres would have displayed a K lobe. It is unlikely that the K lobe of Suoidea and Hippopotamoidea would be the result of common ancestry. Rather, its multiple occurrences could be the consequence of convergent selective pressures. Such pressures might be related, for instance, to the lack of available space for the growing neocortex (Voogd et al. 1998).

Conclusions

The excellent preservation of LM1967MA300 allowed for the thorough description of the endocast of M. minimum. Although less preserved, the endocast of Y13310 provided useful observations on M. medioximus, a more recent Bothriodontinae. These are the first anthracothere endocasts ever described and, as such, they fill a major gap in our knowledge of mammal brain evolution. The brain of Anthracotheriidae likely follows some of the mammalian evolution trends: increase of the relative surface of the neocortex (Dechaseaux, 1968, 1969b) and reduction of the relative size of the cerebellum through time (Orliac et al. 2012).

This study could point out some endocranial characters that might be helpful for future studies on the brain of Hippopotamoidea. Among them is the presence of an ectosylvia as early as the late Oligocene. One possible synapomorphy of the superfamilly can be proposed: the ‘backward shift’ of the pituitary along the antero-posterior axis. Observation of this feature on more anthracothere skulls has yet to confirm this putative synapomorphy. On the other hand, most of the endocranial characters detailed here appear to be convergent among Artiodactyla or even between mammalian orders, which can be a problem for phylogenetic studies operating at these levels. As stated in previous works (e.g. Dechaseaux, 1968, 1969b, 1973; Voogd et al. 1998), the modalities for the neopallium folding are often the same, even in very distantly related taxa.

Several statements hypothesized in this study need to be confirmed. The first one is the secondary loss of the ‘backward shift’ of the pituitary for H. amphibius. We hypothesized that this loss was attributed to the backward shift of the optic chiasma and of the optic nerve, as the orbits moved upward while H. amphibius adapted to its aquatic environment (Dechaseaux, 1961). A morphometric study of this region of the inner skull among hippopotamids could be very instructive. Also, a proper quantification of this ‘backward shift’ has yet to be developed.

The inferred presence of a K lobe in anthracotheres also needs to be investigated. First, the structure observed on the right side of the M. medioximus endocast should be present on the left side as well. More complete braincases of Bothriodontinae might confirm it. It is also unclear whether the K lobe develops from a small process on the sylvian complex, such as the structure observed on the M. minimum endocast. One way to tackle this question may be to investigate the development of the domestic pig brain. Finally, there is not a single study on the role of the K lobe, and the relevance of this structure has yet to be assessed.

Acknowledgments

We thank A. Euriat, who segmented the skull of LM1967MA300, as well as M. Herbin (MNHN UMR 7179) and C. Argot (MNHN USM 0203) for having granted access to the material used for the comparisons. Many thanks to L. Foley-Ducrocq, M. Orliac as well as the referees for critical review of earlier versions of this study. This work has been supported by the ANR-09-BLAN-0238 Program, the CNRS UMR 7262 (IPHEP) and the University of Poitiers.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Fig. S1. 3D surface of the endocast of Microbunodon minimum (LM1967MA300).

joa0227-0277-sd1.pdf (10.9MB, pdf)

References

  1. Blowers T. 1996. Social grouping behaviors of captive female Hippopotamus amphibius. PhD Thesis, University of Central Florida.
  2. Boisserie J-R, Lihoreau F. Emergence of Hippopotamidae: new scenarios. CR Palevol. 2006;5:749–756. [Google Scholar]
  3. Boisserie J-R, Lihoreau F, Brunet M. The position of Hippopotamidae within Cetartiodactyla. Proc Natl Acad Sci U S A. 2005;102:1537–1541. doi: 10.1073/pnas.0409518102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Boisserie J-R, Lihoreau F, Orliac M, et al. Morphology and phylogenetic relationships of the earliest known hippopotamids (Cetartiodactyla, Hippopotamidae, Kenyapotaminae) Zool J Linn Soc. 2010;158:325–366. [Google Scholar]
  5. Brunet M. Découverte d'un crâne d'Anthracotheriidae, Microbunodon minimum (Cuvier), à La Milloque (Lot-et-Garonne) CR Acad Sci Paris D. 1968;267:835–838. [Google Scholar]
  6. Cabard P. 1976. Monographie du genre Microbunodon Deperet, 1908 (Mammalia, Artiodactyla, Anthracotheriidae) de l'Oligocène supérieur d'Europe de l'Ouest. PhD Thesis, University of Poitiers.
  7. Cuvier G. Recherches sur les ossements fossiles, où l'on rétablit les caractères de plusieurs animaux, dont les révolutions du globe ont détruit les espèces. Vol. 5. Paris: Edition d'Ocagne; 1822. [Google Scholar]
  8. Danilo L. 2013. Evolution des structures neurocrâniennes des Equoidea (Perissodactyla, Mammalia) européens paléogènes. PhD Thesis, University of Montpellier II.
  9. Dechaseaux C. Encéphales d'artiodactyles fossiles. Traité Paleontol. 1961;6:1085–1104. [Google Scholar]
  10. Dechaseaux C. Les débuts de l'histoire de la fissuration du néopallium chez les carnivores fissipèdes et chez les artiodactyles. CR Acad Sci Paris. 1968;266:2320–2323. [Google Scholar]
  11. Dechaseaux C. Moulages endocrâniens d'artiodactyles primitifs, essai sur l'histoire du néopallium. Ann Paleontol. 1969a;55:195–248. [Google Scholar]
  12. Dechaseaux C. Les grandes lignes de la fissuration du néopallium des artiodactyles. CR Acad Sci Paris. 1969b;268:653–655. [PubMed] [Google Scholar]
  13. Dechaseaux C. Essais de paléoneurologie. Ann Paleontol. 1973;59:115–132. [Google Scholar]
  14. Eisenberg JF. The Mammalian Radiations. Chicago: University of Chicago Press; 1981. [Google Scholar]
  15. Erfurt J, Métais G. Endemic European Paleogene artiodactyls. In: Prothero DR, Foss SE, editors. The Evolution of Artiodactyls. Baltimore: Johns Hopkins University Press; 2007. pp. 59–84. [Google Scholar]
  16. Falconer H, Cautley PT. Note on the fossil Hippopotamus of the Siwalik Hills. Asiat Res. 1836;19:39–53. [Google Scholar]
  17. Fournier M, Combès B, Roberts N, et al. Mapping the distance between the brain and the inner surface of the skull and their global asymmetries. Medical Imaging 2011: Image Process. 2011;7962:79620. [Google Scholar]
  18. Gentry AW, Hooker J. The phylogeny of the Artiodactyla. In: Benton M, editor. The Phylogeny and Classification of the Tetrapods, Tome 2: Mammals, Systematics Association Special Volume no. 35B. Oxford: Clarendon Press; 1988. pp. 235–272. [Google Scholar]
  19. Heissig K. The astragalus in anoplotheres and oreodonts, phylogenetical and paleogeographical implications. Kaupia. 1993;3:173–178. [Google Scholar]
  20. Holloway RL, Shapiro JS. Relationship of squamosal suture to asterion in pongids (Pan): relevance to early hominid brain evolution. Am J Phys Anthropol. 1992;89:275–282. doi: 10.1002/ajpa.1330890302. [DOI] [PubMed] [Google Scholar]
  21. Jerison HJ. Evolution of the Brain and Intelligence. New York: Academic Press; 1973. [Google Scholar]
  22. Jerison HJ. How can fossils tell us about the evolution of the neocortex? In: Kaas JH, editor. Evolutionary Neuroscience. Oxford: Academic Press; 2009. pp. 497–508. [Google Scholar]
  23. Jerison HJ. Digitized fossil brains: neocorticalization. Biolinguistics. 2012;6:383–392. [Google Scholar]
  24. Lihoreau F. 2003. Systématique et paléoécologie des Anthracotheriidae (Artiodactyla; Suiformes) du Mio-Pliocène de l'Ancien Monde: implications paléobiogéo-graphiques. PhD Thesis, University of Poitiers.
  25. Lihoreau F, Ducrocq S. Family Anthracotheriidae. In: Prothero DR, Foss SE, editors. The Evolution of Artiodactyls. Baltimore: Johns Hopkins University Press; 2007. pp. 89–105. [Google Scholar]
  26. Lihoreau F, Barry J, Blondel C, et al. A new species of Anthracotheriidae, Merycopotamus medioximus nov. sp. from the Late Miocene of the Potwar Plateau, Pakistan. CR Palevol. 2004;3:653–662. [Google Scholar]
  27. Lihoreau F, Barry J, Blondel C, et al. Anatomical revision of the genus Merycopotamus (Artiodactyla; Anthracotheriidae): its significance for late Miocene mammal dispersal in Asia. Palaeontology. 2007;50:503–524. [Google Scholar]
  28. Lihoreau F, Boisserie J-R, Manthi FK, et al. Hippos stem from the longest sequence of terrestrial cetartiodactyl evolution in Africa. Nat Commun. 2015;6:6264. doi: 10.1038/ncomms7264. [DOI] [PubMed] [Google Scholar]
  29. Lind NM, Moustgaard A, Jelsing J, et al. The use of pigs in neuroscience: modeling brain disorders. Neurosci Biobehav Rev. 2007;31:728–751. doi: 10.1016/j.neubiorev.2007.02.003. [DOI] [PubMed] [Google Scholar]
  30. Macrini TE. Description of a digital cranial endocast of Bathygenys reevesi (Merycoidodontidae; Oreodontoidea) and implications for apomorphy-based diagnosis of isolated, natural endocasts. J Vertebr Paleontol. 2009;29:1199–1211. [Google Scholar]
  31. Macrini TE, Rowe T, Archer M. Description of a cranial endocast from a fossil platypus, Obdurodondicksoni (Monotremata, Ornithorhynchidae), and the relevance of endocranial characters to monotreme monophyly. J Morphol. 2006;267:1000–1015. doi: 10.1002/jmor.10452. [DOI] [PubMed] [Google Scholar]
  32. Mc Kenna MC, Bell SK. Classification of Mammals Above the Species Level. New York: Columbia University Press; 1997. [Google Scholar]
  33. Mennecart B, Scherler L, Hiard F, et al. Large mammals from Rickenbach (Switzerland, reference locality MP29, Late Oligocene): biostratigraphic and palaeoenvironmental implications. Swiss J Palaeontol. 2012;131:161–181. [Google Scholar]
  34. Okada Y, Lähteenmäki A, Xu C. Comparison of MEG and EEG on the basis of somatic evoked responses elicited by stimulation of the snout in the juvenile swine. Clin Neurophysiol. 1999;110:214–229. doi: 10.1016/s0013-4694(98)00111-4. [DOI] [PubMed] [Google Scholar]
  35. Orliac MJ, Gilissen E. Virtual endocranial cast of earliest Eocene Diacodexis (Artiodactyla, Mammalia) and morphological diversity of early artiodactyl brains. Proc R Soc B. 2012;279:3670–3677. doi: 10.1098/rspb.2012.1156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Orliac M, Boisserie J-R, MacLatchy L, et al. Early Miocene hippopotamids (Cetartiodactyla) constrain the phylogenetic and spatiotemporal settings of hippopotamid origin. Proc Natl Acad Sci U S A. 2010;107:11871–11876. doi: 10.1073/pnas.1001373107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Orliac MJ, Argot C, Gilissen E. Digital cranial endocast of Hyopsodus (Mammalia, ‘Condylarthra’): a case of Paleogene terrestrial echolocation? PLoS One. 2012;7:e30000. doi: 10.1371/journal.pone.0030000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Pearson HS. The hinder end of the skull in Merycopotamus and in Hippopotamus minutus. J Anat. 1929;63:237–241. [PMC free article] [PubMed] [Google Scholar]
  39. Pickford M, Morales J. On the tayassuid affinities of Xenohyus, Ginsburg, 1980, and the description of new fossils from Spain. Estud Geol. 1989;45:233–237. [Google Scholar]
  40. Radinsky L. Primate brain evolution. Am Sci. 1975;63:656–663. [PubMed] [Google Scholar]
  41. Radinsky L. Brains of early carnivores. Paleobiology. 1977;3:333–349. [Google Scholar]
  42. Rilling JK, Insel TR. Evolution of the cerebellum in primates: differences in relative volume among monkeys, apes and humans. Brain Behav Evol. 1998;52:308–314. doi: 10.1159/000006575. [DOI] [PubMed] [Google Scholar]
  43. Rincon AF, Bloch JI, Macfadden BJ, et al. First Central American record of Anthracotheriidae (Mammalia, Bothriodontinae) from the early Miocene of Panama. J Vertebr Paleontol. 2013;33:421–433. [Google Scholar]
  44. Rowe T. Brain heterochrony and origin of the mammalian middle ear. In: Ghiselin M, Pinna G, editors. New Perspectives on the History of Life. San Francisco: California Academy of Science; 1996a. pp. 71–95. [Google Scholar]
  45. Rowe T. Coevolution of the mammalian middle ear and neocortex. Science. 1996b;273:651–654. doi: 10.1126/science.273.5275.651. [DOI] [PubMed] [Google Scholar]
  46. Shultz S, Dunbar R. Both social and ecological factors predict ungulate brain size. Proc R Soc B. 2006;273:207–215. doi: 10.1098/rspb.2005.3283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Shultz S, Dunbar R. Encephalization is not a universal macroevolutionary phenomenon in mammals but is associated with sociality. Proc Natl Acad Sci U S A. 2010;107:21582–21586. doi: 10.1073/pnas.1005246107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Theodor JM, Foss SE. Deciduous dentitions of eocene cebochoerid artiodactyls and cetartiodactyl relationships. J Mamm Evol. 2005;12:161–181. [Google Scholar]
  49. Van der Made J. Superfamily Hippopotamoidea. In: Rössner GE, Heissig K, editors. The Miocene Land Mammals of Europe. Munich: Verlag Dr Friedrich Pfeil; 1999. pp. 203–208. [Google Scholar]
  50. Voogd J, Nieuwenhuys R, van Dongen PAM. Mammals. In: Nieuwenhuys R, ten Donkelaar HJ, Nicholson C, et al., editors. The Central Nervous System of Vertebrates. Berlin: Springer; 1998. pp. 1637–2098. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1. 3D surface of the endocast of Microbunodon minimum (LM1967MA300).

joa0227-0277-sd1.pdf (10.9MB, pdf)

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES