Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Buronius manfredschmidi—A new small hominid from the early late Miocene of Hammerschmiede (Bavaria, Germany)

  • M. Böhme ,

    Contributed equally to this work with: M. Böhme, D. R. Begun

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Geosciences, Section Terrestrial Palaeoclimatology, Eberhard-Karls-Universität Tübingen, Tübingen, Germany, Section Palaeontology, Senckenberg Centre for Human Evolution and Palaeoenvironment, Tübingen, Germany

  • D. R. Begun ,

    Contributed equally to this work with: M. Böhme, D. R. Begun

    Roles Conceptualization, Formal analysis, Investigation, Writing – original draft, Writing – review & editing

    Affiliation Department of Anthropology, University of Toronto, Toronto, Ontario, Canada

  • A. C. Holmes,

    Roles Formal analysis, Investigation, Methodology

    Affiliation Department of Anthropology, University of Toronto, Toronto, Ontario, Canada

  • T. Lechner,

    Roles Data curation, Formal analysis, Investigation, Writing – original draft

    Affiliations Department of Geosciences, Section Terrestrial Palaeoclimatology, Eberhard-Karls-Universität Tübingen, Tübingen, Germany, Section Palaeontology, Senckenberg Centre for Human Evolution and Palaeoenvironment, Tübingen, Germany

  • G. Ferreira

    Roles Formal analysis, Investigation, Methodology, Visualization, Writing – original draft

    Affiliations Department of Geosciences, Section Terrestrial Palaeoclimatology, Eberhard-Karls-Universität Tübingen, Tübingen, Germany, Section Palaeontology, Senckenberg Centre for Human Evolution and Palaeoenvironment, Tübingen, Germany


The known diversity of European middle and late Miocene hominids has increased significantly during the last decades. Most of these great apes were frugivores in the broadest sense, ranging from soft fruit frugivores most like chimpanzees to hard/tough object feeders like orangutans, varying in size from larger than siamangs (over 17 kg) to larger than most chimpanzees (~60–70 kg). In contrast to the frequent sympatry of hominoids in the early-to-middle Miocene of Africa, in no European Miocene locality more than one hominid taxon has been identified. Here we describe the first case of hominid sympatry in Europe from the 11.62 Ma old Hammerschmiede HAM 5 level, best known from its excellent record of Danuvius guggenmosi. The new fossils are consistent in size with larger pliopithecoids but differ morphologically from any pliopithecoid and from Danuvius. They are also distinguished from early and middle Miocene apes, share affinities with late Miocene apes, and represent a small hitherto unknown late Miocene ape Buronius manfredschmidi. With an estimated body mass of about 10 kg it represents the smallest known hominid taxon. The relative enamel thickness of Buronius is thin and contrasts with Danuvius, whose enamel is twice as thick. The differences between Buronius and Danuvius in tooth and patellar morphology, enamel thickness and body mass are indicative of differing adaptations in each, permitting resource partitioning, in which Buronius was a more folivorous climber.


Miocene hominoid localities become increasingly common in Europe from the late middle Miocene onwards, shortly after they become rare in Africa. Despite their frequency, richness, and in three cases an exceptional abundance of well-preserved hominoid fossils (Can Llobateres, Hammerschmiede and Rudabánya), no European locality has yielded more than one hominoid taxon. In a few cases the hominoid from a site is accompanied by a pliopithecoid, though in only one case, Rudabánya, are the two catarrhines found co-mingled in the same stratigraphic level [1, 2]. In contrast, all comparably rich early and middle Miocene hominoid sites in Africa contain at least two catarrhines and often more [3, 4].

The Hammerschmiede fossil site is best known as the Danuvius locality. It preserves multiple individuals of Danuvius guggenmosi including well-preserved postcranial bones [5]. The Danuvius fossils come from the 11.62 Ma old HAM 5 level at Hammerschmiede [6], which is highly constrained stratigraphically and taphonomically. The HAM 5 level contains a huge taxonomic diversity, from plants to molluscs to mammals, including many partial skeletons and well-preserved crania. From this level alone, 112 vertebrate species are known (147 from all levels together), including 73 species of mammals (84 mammals in total from the outcrop). So far, only a fraction of the enormous vertebrate fauna has been studied in detail, including carnivores [710], artiodactyles [1113], beavers [14], small mammals [1520] and birds [2123].

HAM 5 also yielded two primate teeth and one patella that are too small and morphologically different to be attributed to Danuvius guggenmosi or any known European Miocene catarrhine. These specimens represent a small hitherto unknown European late Miocene ape, which we describe in this contribution. The presence of a second hominoid at HAM 5 is more consistent with the level of primate diversity found at many early Miocene localities and hence we discuss the new taxon in the light of sympatry in fossil apes.

Geologic and taphonomic setting

The Hammerschmiede outcrop is an active clay-pit in the Upper Series lithostratigraphic unit of the Upper Freshwater Molasse in the North Alpine Foreland Basin [6, 24] (Fig 1A and 1B). It exposes a more than 25 meters thick fluvial sequence, composed of clayey to silty overbank sediments, incised by sandy channel-fills, in addition to two lignite seams, representing a swamp facies [6]. Fossils are mainly known from fluvial channels. The fossil-bearing level HAM 5 represents a riffle pool sequence of a small meandering rivulet (Fig 1C) [5]. The fossiliferous, one-metre thick channel-fill is composed of three fining-upward beds with reworked pedogenic carbonates at their bases [11]. Skeletal elements of vertebrates are commonly disarticulated and show no abrasion, except for some specimens from large bodied taxa such as rhinos and proboscideans. Associated elements of medium-sized mammals like Danuvius guggenmosi, are found frequently within and especially next to the channel (Fig 1C), suggesting rapid deposition and only minor downstream transport of carcasses.

Fig 1.

Geographical position of the Hammerschmiede locality (A, B) and excavation plan (C) of the channel structure HAM 5 (grey areas excavated from 2011 to 2019). Dashed line represents the channel structure. Red stars represent the specimens of Buronius manfredschmidi and grey symbols represent Danuvius individuals; stars–GPIT/MA/10000 (male holotype), diamonds–GPIT/MA/10001 (female paratype), circles–GPIT/MA/10002 (juvenile paratype), triangles–GPIT/MA/10003 (female paratype). Red encircled areas have no tachymeter measurements. Coordinates correspond to Gauss-Krüger Zone 4 grid with easting (R) and northing (H) in metres. The topographic maps have been created using the Generic Mapping Tools program [25].

The material described here was recovered from the HAM 5 channel structure in close proximity to the hypodigm of Danuvius guggenmosi (Fig 1C). Specimens GPIT/MA/10007 and 13005 were found in 2011 next to each other during the first test excavations in Hammerschmiede. Therefore, we have no tachymeter measurements for those specimens. GPIT/MA/13004 has been excavated in 2017, about 25 m downstream from the first two specimens (Fig 1C).

Materials and methods

Comparative material

GPIT/MA/13005, 13004, and 10007 were compared to fossil catarrhines (propliopithecoids, pliopithecoids, cercopithecoids, hominoids) and to extant catarrhines. Morphometric data on extant catarrhine patellae have been taken from zoological collections of the Royal Museum for Central Africa (Tervuren) and the Bavarian State Collection (Munich). Dimensions were measured with dial calipers and recorded to the nearest 0.1 mm.

ROPA (S1 Fig)

Relative occlusal polygon area (ROPA) is the ratio of the area of the polygon defined by the tips of the four principal upper molar cusps over the total crown base area [26]. Measurements were taken from high-resolution images of original fossil and extant specimens with the maximal occlusal surface view oriented normal to the camera focal plane.

Protocone and paracone angles (S1 Fig)

The protocone angle, reflecting the positions of the paracone and hypocone relative to the protocone, is measured as the occlusal polygon angle with the protocone at its apex. The paracone angle, reflecting the positions of the protocone and metacone relative to the paracone, is measured as the occlusal polygon angle with the paracone at its apex.

Micro-CT scanning

The teeth were scanned with an X-ray tube containing a multi-metal reflection target with a maximum acceleration voltage of 225 kV in the Nikon X TH 320 μCT scanner of the 3D imaging lab of the University of Tübingen. All specimens (except GPIT/MA/10000-03, for scan data see [5]), were scanned using a 0.1 mm copper filter, but with different settings: GPIT/MA/10001-01 was imaged using 4476 projections, at 210 kV and 45 μA with a voxel size of 0.011504 mm; GPIT/MA/10002-07 with 4476 projections, 180 kV and 50 μA, with a voxel size of 0.011942 mm; and GPIT/MA/13005 with 3500 projections, 200 kV and 27 μA, with a voxel size of 0.006741.

Calculations of 2D enamel thickness

Following [27], virtual buccolingual sections of the teeth were performed using Dragonfly software, Version 2022.1.0.1259 for Windows ( Mesial and distal virtual sections in upper molars were defined by the tips of the protocone-paracone and metacone–hypocone. The following variables were measured two-dimensionally in both planes: dentine area (b), enamel cap area (c), length of the enamel–dentine junction (e) and the bi-cervical diameter. According to [28] the average enamel thickness is calculated as c/e and the relative enamel thickness (RET) is calculated as .

The investigated molars show different stages of wear [29], ranging between unworn (wear stage 1) and full cusp removal with some to large dentine exposure (wear stage 3–4). For worn molars (M1, M2 from Danuvius guggenmosi holotype maxilla and the paratype M1, see [5]) we performed measurements on digitally reconstructed enamel surfaces.

Nomenclatural acts

The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix """". The LSID for this publication is: The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories: PubMed Central, LOCKSS.


Systematic paleontology

Order Primates Linnaeus, 1758

Infraorder Catarrhini Geoffroy, 1812

Family Hominidae Gray 1825

Buronius manfredschmidi nov. gen. et sp.

Derivation nominis.

Genus name after Buron, the medieval name for the city of Kaufbeuren, which is located 5 km to the south of the Hammerschmiede clay-pit. The specific epithet is in honour of Dr. med. dent. Manfred Schmid (Marktoberdorf), a private collector who joined Sigulf Guggenmoos in collecting fossils from Hammerschmiede since the late 1970’s.


GPIT/MA/13005: An unworn left upper M2 germ, crown complete with no root formation (Figs 2 and 3).

Fig 2. Buronius manfredschmidi nov. gen. et sp. photographs.

Upper panel: holotype left upper M2 (GPIT/MA/13005), A–occlusal, B–buccal, C–lingual, D–mesial, E–distal. Lower panel: paratype left lower P4 (GPIT/MA/13004), F–occlusal, G–buccal, H–lingual, I–mesial, J–distal. Scale bars equal 10 mm.

Fig 3. Buronius manfredschmidi nov. gen. et sp. surface renderings.

Upper panel: 3D rendering of the holotype left upper M2 (GPIT/MA/13005), A–occlusal, B–buccal, C–lingual, D–mesial, E–distal. Lower panel: 3D rendering of the paratype left lower P4 (GPIT/MA/13004), F–occlusal, G–buccal, H–lingual, I–mesial, J–distal. Scale bars equal 10 mm.


GPIT/MA/13004: A lightly worn left lower P4 fragment preserving the buccal portions including the mesial fovea, an intact protoconid, and the buccal half of the talonid and metaconid, and missing the lingual third of the crown (Figs 2 and 3). GPIT/MA/10007: A left patella slightly damaged proximally, with a porous articular surface suggestive of a juvenile (Fig 4 and S2 Fig).

Fig 4. Buronius manfredschmidi nov. gen. et sp. paratype left patella (GPIT/MA/10007).

Anterior (A), posterior (B) and distal (C) views. Scale bar equals 10 mm.

Measurements of the holotype and paratypes are provided in Table 1.

Table 1. Buronius manfredschmidi nov. gen. et sp., tooth and patella metrics (mm).

Locality and horizon.

Hammerschmiede Clay pit near Pforzen (Allgäu region, Bavaria, Germany; 47.927° N, 10.592° E); level Hammerschmiede (HAM) 5 at stratigraphic metre 12 in the local section, which has been dated magnetostratigraphically to 11.62 million years ago [6].


Small hominid in the low end of the body size range of Symphalangus, suggesting an average body mass of about 10 kg. M2 with prominent cusps positioned near the periphery of the crown, with protocone-paracone and hypocone-metacone nearly aligned transversely, a shallow lingual cingulum confined to the protocone, broad mesial fovea, mesiodistally long trigon, lingually placed metacone and hypocone, well developed and continuous crista obliqua, slit-like distal fovea, hypocrista well elevated in relation to the distal marginal ridge, thinly enamelled with prominent, widely spaced dentin horns, and sloped, concave (with respect to the dentine) lingual enamodentin junction. P4 with protoconid and metaconid of equal prominence, moderate protoconid lingual flare, well developed protocristid elevated relative to the thick mesial marginal ridge, broad, deep mesial fovea, elongated trigonid with a strongly inclined distolingual margin, modestly developed hypoconid, sharp distal marginal ridge with multiple conulids. Patella oblique ovoid in shape, relatively thick and broad compared to length, well developed keel separating relatively concave condylar surfaces resulting in a saddle shaped joint surface, prominent and medially shifted distal apex.

Differential diagnosis.

Primitive catarrhines, early Miocene stem hominoids, and middle Miocene stem hominids. GPIT/MA/13005 (Figs 2, 3, 5, 8 and S3, S4, S7, S8 Figs and Table 1) differs from Griphopithecus in being much smaller and in having thin enamel. It differs from early Miocene hominoids (e.g. Ekembo) in being smaller than all but the smallest specimens, reduced lingual cingulum, longer relative to breadth, having expanded mesial fovea and elevated protocrista and hypocrista relative to the mesial and distal marginal crista. Differs from Pliobates in being larger, elongated relative to breadth, reduced lingual cingulum, larger mesial and distal fovea, and reduced buccal style. Differs from Anapithecus and other pliopithecoids in being longer relative to breadth, reduced lingual cingulum and buccal style, broader more peripheralized cusps, less compressed and elevated cristae, larger mesial fovea, larger, elongated trigon, lower mesial and distal marginal ridges. Differs specifically from Anapithecus in much less strongly developed cingulum, which is better described as a rounded rim rather than a true cingulum (lacking a sharp margin, the shelf between the margin and crown wall, and the cristae between the crown wall and the cingulum edge), greater cusp peripheralization, more lingually inclined postparacrista, postparacrista-premetacrista not aligned and positioned closer to the buccal margin of the crown, shorter talon, slit-like distal fovea lacking the distal expansion seen in Anapithecus, hypocone more cone-shaped (in comparison with the pointed hypocone in Anapithecus), positioned at the distolingual corner of the crown, mesially oriented prehypocone crista connected to the postprotocone crista (which is absent in Anapithecus) rather than in crista obliqua. The preparacone cristae also differ in being sharper and more directly aligned with the other buccal cristae in Anapithecus, while strongly curved mesiolingually and less sharp in Buronius.

Fig 5. Comparison of Buronius (left, GPIT/MA/13005) and Danuvius (right, GPIT/MA/10002-07) upper M2 occlusal morphology with the nomenclature labelled on the Buronius tooth.

Pr—protocone; Prl—protoconule; MMR—Mesial marginal ridge; Pac—paracrista; Prpac—preparacone crista; Pa—paracone; Popac—postparacone crista; BS—buccal style; Prmc–premetacone crista; Me—metacone; Pomc—postmetacone crista; DMR–Distal marginal ridge; H-Mc–hypocone-metacone crista; Hyc—hypocrista; Pohc—posthypocone crista; Hy—hypocone; Prhc—prehypocone crista; CO—crista obliqua; Poprc—postprotocone crista. Note the unusual configuration of the crista obliqua (yellow line) and postprotocone crista (orange line). In Buronius the crista obliqua meets the concave postprotocone crista distal to the protocone, in contrast to Danuvius, in which the two cristae meet at the protocone tip (For a description of GPIT/MA/10002-07 see S2 File).

GPIT/MA/13004 (Figs 2 and 3 and Table 1) differs from Ekembo, Proconsul, Afropithecus, Griphopithecus, Equatorius, Kenyapithecus, and Nacholapithecus in the height of the talonid relative to the trigonid.

GPIT/MA/10007 (Fig 4 and S2 Fig and Table 1) differs from stem hominoids (Ekembo, Equatorius, Nacholapithecus) and Epipliopithecus in being proximo-distal shorter and antero-posteriorly thicker.

Middle/late Miocene hominids/hominines. GPIT/MA/13005 (Figs 2, 3, 5, 8 and S3, S4, S7, S8 Figs and Table 1) differs from Danuvius in being much smaller and in having a more strongly developed protocone lingual cingulum, relatively larger metacone, more sloped lingual and buccal crown surface, larger mesial fovea, more strongly developed paracrista connected to the protoconule, elevated hypocrista relative to the distal marginal ridge, more elongated trigon, less truncated crown shape distobuccally, a more strongly developed buccal style conule, more peripheralized cusps and thinner enamel (possibly except Dryopithecus, see discussion). Differs from Pierolapithecus, Anoiapithecus and Dryopithecus in being smaller and in the development of the mesiolingual cingulum, metacone size, crista development, and enamel thickness. Differs from Rudapithecus and Hispanopithecus in being smaller and in having more a strongly pronounced mesiolingual cingulum, a larger mesial fovea, a more prominent hypocrista and more sloped lingual and buccal surfaces. Quantitatively, Buronius is much smaller than any dryopithecin M1 and M2 (S5 Fig), has more peripheralized cusps (higher ROPA) than all dryopithecins (Fig 10), a smaller protocone angle than Danuvius and a larger protocone angle than most Hispanopithecus (Fig 9C), a smaller paracone angle than Anoiapithecus, Dryopithecus and Rudapithecus, and a larger paracone angle than Danuvius (Fig 9D). GPIT/MA/13005 is much longer relative to width than the M1 in Anoiapithecus, Dryopithecus and at the high ends of the ranges of variation in other dryopithecins (S5C Fig). It is longer relative to width than the M2 in Danuvius and Dryopithecus and at the low ends of the ranges in other dryopithecins (S5D Fig). A PCA also distinguishes Buronius from all dryopithecins on component 1, most strongly influenced by protocone angle (Fig 9C). S3 and S4 Figs includes visual comparisons of the occlusal morphology of Buronius and dryopithecins and compares GPIT/MA/13005 with associated M1-M2 in other dryopithecins.

GPIT/MA/13004 (Figs 2 and 3; Table 1) differs from dryopithecins (unknown for Danuvius) in the elongated talonid and thick mesial marginal ridge.

GPIT/MA/10007 (Figs 4, 11 and S2 Fig and Table 1) differs from Danuvius in being smaller, thicker and narrower, with a more strongly developed saddle-shaped joint surface and a prominent distal apex; Differs from Rudapithecus and Pierolapithecus in its smaller size and prominent distal apex.


GPIT/MA/13005 (Figs 2, 3, 5 and S3, S4 Figs) is a perfectly preserved upper M2 crown with no wear and no root formation. Crown formation is practically completed and only a very narrow strip on the lingual cervix remains incomplete. The enamel surface is comparatively smooth and few crenulations appear only in the talon basin. The enamodentin junction is well preserved and visible on the underside of the enamel cap. The dentine horns are well separated and compressed at their tips, penetrating deeply into the enamel cap, as confirmed by μct scan (Fig 6). The 2D relative enamel thickness (RET) is quite thin compared with that of Danuvius permanent and deciduous dentition (Table 2 and S6 Fig). The protocone is the largest cusp, followed by the paracone and slightly smaller metacone and hypocone, the latter only minimally smaller than the former. The hypocone is lingually positioned relative to the protocone. The metacone has a sloped distobuccal corner. The size and position of the hypocone and the morphology of the metacone are consistent with an M2. The trigon is large, especially mesiodistally, with a well-defined paracrista close in height to the mesial marginal crista, separating the trigon from a broad mesial fovea and connected to a well-defined protoconule lingually. The crista obliqua is sharply defined and continuous, with only a shallow notch separating the protocone and metacone portions. The lingual cingulum is shallow and confined to the mesiolingual corner, lacking any development of a shelf or accessory crista, as commonly developed on pliopithecoid upper molars. The hypocrista is strongly developed and elevated in relation to the distal marginal crista. The postmetacrista is equal in length to the premetacrista, while in Danuvius the premetacrista is shorter. Buccally there is a mild shelf (style) between the paracone and metacone with a small conule. A peculiarity of GPIT/MA/13005 is the generally concave shape of the cristae, especially the postprotocone crista and the postparacone-premetacone cristae. In most middle and late Miocene European apes these cristae are straight or convex. The concavity of the postprotocone crista is most noticeable (Fig 5). The crista obliqua joins with the postprotocone crista buccal and distal to the protocone, while in Danuvius and most dryopithecins it connects directly to the protocone (Fig 5).

Fig 6. Enamel thickness on unworn upper molars.

Distal virtual section through the tips of metacone and hypocone for the left upper M2 of Buronius manfredschmidi nov. gen. et sp. (A, GPIT/MA/13005) and the left upper M2 of Danuvius guggenmosi (B, GPIT-MA-10002-07). Note the substantial difference in enamel thickness between Buronius and Danuvius. Scale bar equals 2 mm.

Table 2. 2D relative enamel thickness of the holotype molar of Buronius manfredschmidi nov. gen. et sp. and the molars of the hypodigm of Danuvius guggenmosi.

GPIT/MA/13004 (Figs 2 and 3) is a fragmentary left lower P4 with light wear along the buccal edge, the buccal and mesiobuccal aspects of the protoconid, and the cusp tip, but without dentin exposure. The roots are not preserved except for a small portion mesially. The metaconid is broken a bit buccal to the apex. The break reveals a thick mesial marginal ridge, also visible in cross section as a prominent enamel fold, with the mesial part thicker than the distal part (Figs 2 and 3). The protocristid is sharp, prominent and well above the mesial marginal ridge. It appears to connect directly to the metaconid. The mesial fovea is deep and probably longer than wide, though it is damaged. The talonid is high in relation to the mesial cusps and the buccal edge from the protoconid to the buccal edge of the talonid (postprotoconid cristid) slopes gradually to the distobuccal margin of the talonid. The cusps are positioned mesially, resulting in an elongated talonid. The long postprotoconid cristid constitutes the buccal and distobuccal edge of the talonid. It curves distolingually to close to the distal margin of the tooth. Unfortunately, there is no P4 for Danuvius to compare with GPIT/MA/13004.

GPIT/MA/10007 (Fig 4 and S2 Fig) is a small left patella close in size to that of Symphalangus and Ekembo heseloni. The posteroproximal edge is damaged, more so medially, but this does not significantly affect the measurements. The cortical bone is thin and poorly mineralized, with a porous articular surface and prominent nutrient foramina on the joint surface distally. Cancellous bone is visible on the damaged edge. The fragile preservation suggests that this patella may derive from a subadult individual, while the well-defined edges of the joint surface suggest that adult size had been attained. The patella has an oblique ovoid shape (major axis proximolateral to distomedial), with the joint surface for the lateral condyle superiorly placed relative to the medial surface (Fig 4). However, the maximum proximodistal and mediolateral dimensions are the same (16.6 mm). The lateral border is more convex than the medial border, which is nearly flat around the midline. The articular surface for the femoral patellar groove is well-developed, mediolaterally convex and proximodistally concave (saddle-shaped). The lateral articular facet is wider than the medial one and encompasses about two thirds of the mediolateral breath (Fig 4C). The anterior surface is convex with the same asymmetry as the posterior surface. The highest point of the anterior surface convexity is medially positioned and continuous with the well-developed distal apex, which protrudes distally medial the midline. The apex is a prominent, drop-shaped point (mediolaterally narrow and proximodistally short). The distal apex merges with the distal edges on either side asymmetrically, with the edge being deeper and more strongly notched laterally.

Taxonomic results

Based on tooth size alone (both Buronius teeth are consistent in size with a single species, Fig 7A), Buronius is unlikely to be a small Danuvius as the difference in tooth size with the latter exceeds that documented for any catarrhine taxon analyzed here (Fig 7 and S5 Fig). Morphologically, Buronius is also quite distinct from Danuvius (Fig 5) and differs in numerous characters, as in a more sloped lingual and buccal crown surface, in a better development of protocone lingual cingulum and paracrista, the latter connected to the protoconule, by both larger mesial fovea and metacone, by a more elongated trigon, and an elevated hypocrista relative to the distal marginal ridge, by a less truncated crown shape distobuccally, a more strongly developed buccal style conule, by more peripheralized cusps and finally by thinner enamel.

Fig 7. The P4/M2 ratio.

A–Comparison of lower P4 (orange bars) and upper M2 (blue bars) lengths in selected catarrhines. The relationship between these two teeth in Buronius is consistent with other catarrhines. Numbers at top are percentages of premolar to molar length. Note the much lower ratio of the Buronius P4 and the smallest Danuvius M2. B–P4/M2 within individual ratios in great apes compared with the same ratio between Buronius P4 and the smallest Danuvius M2. The box plot shows the centre line (median), box limits (upper and lower quartiles), crosses (arithmetic mean), whiskers (range) and individual values (circles).

GPIT/MA/13005 is also close in size to GPIT/MA/10002-04, a right deciduous dP4 of Danuvius guggenmosi (Fig 8B and S7, S8 Figs). However, GPIT/MA/10002-04 is clearly a deciduous molar (low, flared crown, simple occlusal morphology) with thick enamel similar to permanent molars (Table 2 and S6 Fig), and as such is easily distinguished from GPIT/MA/13005. As in permanent molars of Danuvius, the preprotocone crista is short and the crista obliqua connects directly to the protocone, but a postprotocone crista is absent in the deciduous molar of this species (Fig 8). We describe and document the differences between upper deciduous last molars and GPIT/MA/13005 in the Supporting Information file (S7S14 Figs and S1 Table). GPIT/MA/13005 is narrower than pliopithecoids M2s, and closest to the means of Pan, Rudapithecus and Hispanopithecus (Fig 9A). The protocone angle in GPIT/MA/13005 is lower than in pliopithecoids, propliopithecoids and Pliobates, and within the 25–75% quartiles of most extant hominoids and dryopithecins (Fig 9C). This reflects in large part the position of the hypocone relative to the protocone, with a more lingually displaced hypocone associated with larger angles. Paracone angles, which reflect the position of the metacone relative to the paracone, distinguish less between primitive catarrhines and hominoids, but do separate Buronius and Danuvius (Fig 9D).

Fig 8.

The Buronius M2 (A) compared with a dP4 (B), and M2 (C) of Danuvius and a M2 of Anapithecus (D). (A–GPIT/MA/13005; B–GPIT/MA/10002-04, inverted; C–GPIT/MA/10002-07; D–RUD 90). Scale bars equal 5 mm. Mesial is to the left.

Fig 9. Comparative analysis of the Buronius upper M2.

A-D–Box plots of length/breadth ratios, molar size and protocone and paracone angles. Buronius falls among fossil and living apes in length/breadth and among larger primitive catarrhines and siamangs in overall size. The protocone angle (see Methods) clearly distinguishes Buronius from primitive catarrhines including Pliobates. Curiously, in paracone angle living apes are distinct from both primitive catarrhines and fossil great apes including Buronius. All box plots show the centre line (median), box limits (upper and lower quartiles), crosses (arithmetic mean), whiskers (range) and individual values (circles). C—Principal component analysis based on mesiodistal length, buccolingual breadth, ROPA, protocone and paracone angles in Buronius and dryopithecins. Hispanopithecus (green dots) and Rudapithecus (blue crosses) are mostly distinguished, with a small area of overlap. Buronius is isolated from these larger samples. It is also quite distinct from Anoiapithecus (red squares), Pierolapithecus, (brown inverted triangle), Dryopithecus (black circle), and the dryopithecin indet molars from Melchingen (magenta triangle). With larger samples from Buronius there may be less of a distinction from Danuvius. PC 1 is most strongly influenced by protocone angle and PC 2 by ROPA and paracone angle.

GPIT/MA/13005 differs in occlusal morphology from all early and middle Miocene catarrhines and most closely resembles that of late Miocene apes. The M2 is unlike pliopithecoids and Pliobates, in having a reduced, smooth cingulum and buccal style and being narrower relative to length (Figs 8 and 9A). Crista are less strongly developed, cusps less compressed and basins broader and shallower. The cingulum and style distinguish it from Danuvius and all European hominines (several recent comprehensive phylogenetic analyses differ in their placement of some European late Miocene apes as stem hominines or stem hominids; for a more complete discussion of middle and late Miocene ape phylogeny see [3033] Cusp peripheralization, as measured by the relationship of total crown basal area (TCBA) compared with occlusal polygon area (Fig 10) groups Buronius with extant hominoids and far from primitive catarrhines. The P4 fragment lacks the typical pliopithecoid cristodonty and deep basins.

Fig 10. Relative Occlusal Polygone Area (ROPA).

The ratio of M2s mesiodistal/buccolingual dimensions (X axis) relative to ROPA (see Methods section and S1 Fig). Primitive catarrhines (including Pliobates) separate well from extant hominoids, with many fossil great apes falling between the two groups. Buronius falls in the overlap of the hylobatid and Pan polygons.

To summarize, while GPIT/MA/13005 is in the size range of larger, stratigraphically younger pliopithecoids, it is easily distinguished from them in morphology. GPIT/MA/13005 clusters with hominoids but without a specific similarity to Danuvius or other late Miocene apes. The distinction from Danuvius is especially pronounced in overall size (Fig 9B and S5 Fig) and in the comparison of enamel thickness, which is among the thickest for late Miocene apes in Danuvius and the thinnest for Buronius (Table 3). GPIT/MA/13005 is unique in its combination of size and morphology, which justifies a new genus. We acknowledge that the presently known hypodigm is small. Larger samples are needed for a more complete characterization of Buronius.

Table 3. 2D relative enamel thickness (RET) in Buronius and fossil and extant hominoids.

The GPIT/MA/10007 patella is morphologically distinct from the patella attributed to Danuvius (GPIT/MA/10000-12) and within the size range consistent with the size of the teeth attributed to Buronius (Figs 11 and 12 and S2 Fig). The simplest explanation is that the distinctive small patella belongs to the same taxon as the distinctive small teeth. Extant catarrhine families are easily distinguished in patellar shape (Fig 11). By its roundish overall morphology and anteroposterior thickness GPIT/MA/10007 compares well with fossil and living hominids (Fig 11A and 11B). Stem-hominoids (e.g. Ekembo, Equatorius, Nacholapithecus) and Epipliopithecus (the only pliopithecid patella known [35]) are proximodistally slightly longer and hylobatids and cercopithecids (especially colobines) significantly longer (Fig 11A and 11C). The elongation of the patellae of both latter groups results from an extremely pronounced distal apex, which represents, however, an anteriorly flat, very broad (base of apex is as wide as the patella) and tongue-like projection (see e.g. [36] and S2 Fig), unlike the apex found in GPIT/MA/10007. The patella of Epipliopithecus is furthermore anteroposteriorly thin, similar to hylobatids. This elongate and thin pliopithecid patella falls between the convex hulls for hominids and hylobatids and within the area defined by the stem-hominoids Ekembo, Equatorius and Nacholapithecus (Fig 11; [37]).

Fig 11. Size-adjusted morphometrics of the patella of extant and fossil catarrhines.

GM–geometric mean, ML–mediolateral breath, PD–proximodistal length, AP–anteroposterior thickness. Measurements of fossil stem-hominoids and Epipliopithecus are from [37, 38]. Note that the patella morphospace separates the families Cercopithecidae, Hylobatidae and Hominidae. Stem-hominoids (Ekembo, Equatorius, Nacholapithecus) overlap with hominids in the morphospace direction of hylobatids, e.g. they tend to be more elongate and thinner than most hominids.

Fig 12. Body mass and patella size.

A–Body mass data for extant hominoids, Buronius, Danuvius and Rudapithecus from regressions of M2 size [42] and observed body masses (open boxes [43, 44]. Note the close correspondence in hylobatids of values from molar size and observed ranges, in contrast to Pan, in which dentally derived estimates of body mass significantly underestimate observed body mass (arrow). The same pattern characterizes Danuvius body mass estimates from dental vs postcranial dimensions [5] and Grabowski pers. comm. B–Patellar mediolateral breadth of living catarrhines and selected fossil hominoids (fossil data from [3638]). Ekembo heseloni specimens are from KPS at Rusinga. Larger patella from Rusinga, tentatively attributed to Ekembo nyanzae (not shown here) fall among Danuvius and Rudapithecus. Whatever the actual body mass of Danuvius was, it was much larger than Buronius, as are all known fossil and extant hominids.

GPIT/MA/10007 differs from the Danuvius male patella (GPIT/MA/10000-12) by several features. The Danuvius patella is mediolaterally 35% larger, wider than long, the posterior articulation facet is comparatively flat, the anterior surface less convex and the distal apex is practically absent. Furthermore, the size-adjusted anteroposterior thickness is slightly higher in GPIT/MA/10007. The saddle-shaped condylar surface of GPIT/MA/10007 more closely resembles Rudapithecus. The distal apex, at which the patellar ligament inserts, is not well-developed in any extant and fossil hominoid, except in one individual of Ekembo nyanzae (KNM-RU 18384). However, this patella (S2G Fig) differs from Buronius by a stronger symmetry, marked elongation and much larger size. The asymmetric, oblique ovoid shape with a medially shifted apex may represent unique features of the new genus. A pointed patellar apex is otherwise known from the genus Homo, which however is longer (25% of the proximodistal patellar length) with a wider proximal base [39].

Based on patellar and tooth morphology Buronius differs from stem-hominoids and hylobatids, having its closest affinities with hominids. We conclude that Buronius is most probably a crown hominid (including all living great apes, humans and their fossil relatives).

Enamel thickness

The unworn upper M2 of Buronius manfredschmidi exhibits a 2D relative enamel thickness (RET) of 10.87 (Table 2). This value contrasts with the thickly enamelled permanent and deciduous molars of Danuvius guggenmosi, which range for six molars between 16.15 and 21.66 (m2 19.36; dP4 20.48; M1 mean 17.9, n = 2; M2 mean 20.69, n = 2; Tables 2 and 3). Table 3 shows the range of RET values in a number of Miocene apes and crown hominids.


The occlusal morphology of GPIT/MA/13005 (M2) is perfectly preserved, which makes comprehensive comparisons possible despite the limited sample. Its uniqueness, both in size and morphology, warrants the recognition of a new genus. The size and morphology of GPIT/MA/13004 (P4) and GPIT/MA/110007 (patella) are consistent with this conclusion. While the small sample size is less than ideal, what is known of Buronius cannot be accommodated within any known hominid taxon. While clearly different from pliopithecoids, the small hominid from HAM 5 is also distinguished from most Miocene apes. Regarding the holotype (GPIT/MA/13005), there is no close match with early and middle Miocene apes in quantitative morphology or in the morphology of the cusps, cristae and basins. The closest morphological match is with late Miocene European apes and extant hominoids. Like late Miocene hominines (or hominids) the cusps in GPIT-MA 13005 are widely spaced, the trigon is spacious and the crista obliqua is sharply defined. Pliopithecoid upper molars are distinguished from hominids in crown shape, cingulum development, and cusp peripheralization. The lower P4 is also distinguished from pliopithecoids in cusp and basin morphology. Interestingly, Buronius is more modern in cusp peripheralization than most middle and late Miocene hominids, which fall between extant hominids and primitive catarrhines (Fig 10). One specimen of Hispanopithecus also falls within the extant polygons. If the distributions of Rudapithecus and Hispanopithecus are an indication of typical within-genus variation, a larger sample of Buronius may help to clarify the distinction from other late Miocene apes.

Relative enamel thickness

Buronius has the lowest RET of any Miocene ape for which data are known [5, 34]. The relative enamel thickness (RET) of the five dryopithecin genera Anoiapithecus, Pierolapithecus, Dryopithecus, Rudapithecus, Hispanopithecus is higher than in Buronius (Tables 3 and 4). RET in Buronius is much smaller than in Danuvius, with Danuvius falling among the most thickly enamelled taxa in our sample. However, sample-sizes for RET of fossil hominids is small. Larger samples of extant great apes, particularly Pongo, can show substantial variation in enamel thickness (Table 3). The lowest value measured so far from a European ape is an M2 of Dryopithecus fontani [40], which is about 20% thicker than in Buronius (Table 3), but both values may be included in the variance of the species if larger samples are available. Therefore, it is possible that Buronius cannot be differentiated from Dryopithecus based on RET alone.

Table 4. Relative enamel thickness in extant hominoid upper second M2 and their thickness categories and dietary affinities.

The 2D RET value of Buronius is closest to the means of gorillas and siamangs while Danuvius is in the upper end of the range of variation in Pongo and within the range of modern humans, Sivapithecus, and Griphopithecus (Table 3). The pronounced disparity in enamel thickness between Buronius and Danuvius strongly supports the genus-level distinction. The difference in enamel thickness between Danuvius and Buronius is not attributable to the difference often observed between permanent and deciduous teeth, as we have demonstrated that the type of Buronius is a permanent upper molar (see above and Supporting Information file, S6S14 Figs).

Patella function

The accentuated apex of the GPIT/MA/10007 patella is probably related to the development and orientation of the patellar ligament and quadriceps femoris force transmission. Furthermore, the better expressed convexity of the articular facets implies a deeper patellar groove of the femoral trochlear surface in GPIT/MA/10007 than in Danuvius. This suggests a greater degree of constraint on the direction of forces resulting from quadriceps femoris contraction in the former than the latter [36], A relatively short, thick patella, as found in Buronius, has been related to the mechanical advantage of the quadriceps, which is apes has been related to climbing [36]. The asymmetry of the patella may reflect differences in the development of the vasti muscles, with a larger vastus lateralis compared with the vastus medialis. The combination of asymmetry, patellar shape and the development of the apex may all be related to a specific pattern of quadriceps femoris function in Buronius [36, 41].

The greater extent of the vastus lateralis insertion and its greater distance from the patellar ligament attachment on the patella suggests an emphasis on a response to adduction moments, to maintain the knee in a neutral or abducted position. The asymmetric position of the patellar apex, which influences the position of the patellar ligament, may also reflect quadriceps function. The development of the patellar apex may be related to the force being transmitted thought the patellar ligament or it may serve to maintain the orientation of the ligament relative to the quadriceps and the long axis of the tibia. It, along with the more strongly developed saddle-shaped joint surface suggests some degree of directional constraint to the forces crossing the knee joint. Without additional postcranial elements for Buronius it is difficult to reconstruct the precise behavioural implications of this morphology except to say that it probably excludes stereotypical pronograde quadrupedalism. Similarity with Rudapithecus, which is known from many postcranial elements with a strong arboreal and suspensory signal, suggests the same for Buronius, with perhaps a greater degree of constraint in knee extension.

Body weight

The teeth and patella of Buronius are close in size to siamangs, suggesting a body mass of about 10 kg (Fig 12). In contrast, Danuvius guggemosi has a calculated body mass ranging from 17 to 31 kg using regressions for several measurements from the femur and tibia [5] or, by using a different methodology of reconstruction, from 14.5 to 46.3 kg (Mark Grabowski personal communication 27. February 2023).

Fig 12A depicts body mass ranges for a variety of hominoids based on both estimates from M2 size and observed body masses [4244]. Hylobatid body mass ranges are very similar between estimates from dental size and observed masses. This suggests that in this range body mass estimates from M2 size in hominoids are reliable. However, in Pan body mass estimates from M2 size are much lower than actual body masses collected from different individuals (arrow). In Gingerich et al. [42], from which the formula to estimate body mass was taken, the body masses for male and female Pan used by the authors to produce the formula (43900g and 31500g respectively) are higher than those predicted by M2 size by 53 and 40% respectively. Body mass estimates from dental data for Pan underestimate actual mass, but it is not clear if this is specific to Pan or an effect of size. A size effect would explain the discrepancy in dental vs postcranial derived estimates for similarly sized Danuvius. Either way, the estimate for Buronius appears to be reliable as far as body mass estimates from teeth go, and clearly shows that it is much too small to be Danuvius. Similarly, the mediolateral breadth of the patella, which scales with body mass [45], falls within the range of siamangs and is near to Ekembo heseloni. Until more postcranial specimens of Buronius are found, we think that our conservative estimate of a body mass around 10 kg is reliable.

Hominid sympatry

The taphonomic context of fossils assigned to Buronius manfredschmidi and Danuvius guggenmosi support the direct sympatry (sensu [46], in contrast to broad sympatry over wider geographic areas), of both species in the immediate habitat of the ecosystem of Hammerschmiede level HAM 5 (Fig 1). To avoid competition, directly sympatric primates have to rely on different resources [46]. The differences between Buronius and Danuvius in functional tooth morphology, enamel thickness, patellar morphology, and body mass strongly support resource partitioning between both species. The thin tooth enamel with fewer crenulations and the relatively more accentuated shearing crests of the small-sized Buronius suggests a more fibrous diet that may have included both soft fruits and more fibrous vegetation. However, there is little indication of specialized folivory, as in gorillas or siamangs, although in enamel thickness Buronius is more similar to folivores (Table 4). In contrast, the larger Danuvius guggenmosi shows blunter molar cusps and crenulated tooth enamel that is, even in its deciduous dentition, near twice as thick as in Buronius (Figs 5 and 6), pointing to higher bite forces and a diet including harder/tougher food items (cf. [47, 48]). The differences in body size and patellar morphology suggest differences in canopy use, with Buronius possibly spending more time higher in the canopy, possibly in the terminal branches. However, it is likely that most if not all European hominoids were opportunistic feeders, given the latitudinal and climatic position of their habitats. Even if the winter season was warm, it had a short day-length, implying reduced photosynthesis and more limited availability of fresh leaves and other potential fallback foods. They likely experienced selection to exploit a wide range of food resources of varying seasonal availability, hence, opportunistic.

Present-day instances of hominoid direct sympatry usually occur between more folivorous vs more frugivorous taxa (gibbons and siamangs in Asia [49] and chimpanzees and gorillas in Africa [46]). In both extant cases the more folivorous species is twice as large as the more frugivorous one. At Hammerschmiede the sympatric hominids differ in size to a comparable degree without evidence of a strong frugivore-folivore dichotomy, but instead a divergence of dietary preferences that may have been more on the soft fruit to hard/tough object spectrum. This may be more analogous with orangutan/gibbon sympatry in Borneo and Sumatra [50]. The smaller taxon (Hylobates) with a preference for soft fruits feeds higher in the canopy than the much larger taxon (Pongo), which, with the differences in diet would have aided in avoiding competition [50]. This is consistent with our observation of a possible preference for higher canopy/terminal branch foraging in Buronius. Hypotheses of feeding ecology in Buronius will need to be tested with more specimens. More research is also needed both in reconstructing vegetation and fruit availability at Hammerschmiede. This is all the more necessary as the determining factors for primate species richness are the primary productivity [51], the forest structure [52] and its structural diversity [50].


Two teeth and a patella from the HAM 5 level at Hammerschmiede are pliopithecoid in size but easily distinguished from that taxon in morphology. The unworn M2 crown is unlike pliopithecoids and Pliobates, having a reduced, smooth cingulum and buccal style and being narrower relative to length. Cristae are less strongly developed, cusps less compressed and basins broader and shallower. However, the cingulum, style and other occlusal details distinguish it from Danuvius and all other European hominines (or hominids, according to some [32]). The P4 fragment also lacks the typical pliopithecoid cristodonty and deep basins and is aligned more closely with hominids. Similarly, the patella differs from pliopithecoids and non-hominids by its roundish overall morphology. A pointed distal patellar apex and an asymmetric, saddle-shaped condylar surface is possibly suggestive of climbing locomotion with more abducted hindlimb postures. The morphology and size of this sample requires recognition of a new genus from the stratigraphic level at Hammerschmiede in which the Danuvius hypodigm was recovered. Dental and patellar morphology differ between Buronius and Danuvius, suggestive of differences in feeding ecology. The Buronius M2 is suggestive of a soft diet that may have also included a significant fibrous component, while in Danuvius the dentition is more consistent with a preference or ability to exploit hard/tough objects. Differences in patella morphology suggest more climbing with abducted hindlimb postures in Buronius, as opposed to the probably more cautious extended limb clambering of Danuvius. Buronius may have typically foraged higher in the canopy than Danuvius. With a body size of about 10 kg Buronius is the smallest crown-hominid known so far.

Two hominid genera at Hammerschmiede is unique among European middle and late Miocene localities (Paşalar in Anatolia also has two hominid genera, but not in direct sympatry [53]), but more like taxon diversity in African early and middle Miocene hominoid localities. A re-examination of morphological variation at other rich European Miocene localities may reveal heretofore unrecognized diversity in these samples as well (e.g. [2, 54]).

Fossil repository

All described Hammerschmiede fossils are stored in the palaeontological collection of the University of Tübingen (acronym GPIT), a research infrastructure of the Senckenberg Institute for Human Evolution and Palaeoenvironment (SHEP) Tübingen.

Supporting information

S1 Fig. Measurements used to calculate and display the relative occlusal polygon area (ROPA), and protocone, and paracone angles.


S2 Fig. Anterior and posterior views of left patella of living and fossil catarrhines.

A–Nasalis larvatus (ZSM 1907/3048), B–Papio hamadryas (RMCA A3.40.M.14), C–Symphalangus syndactylus (ZSM 1905/60), D–Pongo pygmaeus (ZSM 1909/801), E–Pan paniscus (RMCA 15293), F–Ekembo heseloni (KPS PT4), G–Ekembo nyanzae (KNM-RU 18384), H–Buronius manfredschmidi (GPIT/MA/10007), I–Danuvius guggenmosi (GPIT/MA/10000-12). All the patellas are from left side, except the anterior view in D, C and E. Anterior view is left, posterior view is right (except the anterior view only in D). Scale bar is 10 mm.


S3 Fig. Occlusal views of upper molars of late Middle and late Miocene European hominines (or stem hominids).

A: GPIT/MA/13005 (Buronius); B: GPIT/MA/10002-07 (Danuvius); C: Alsótelekes (Rudapithecus) M1-M2; D: GPIT/MA/2122 (Melchingen dryopithecin indet.); E: IPS 35026 (Dryopithecus); F: IPS 21350 (Pierolapithecus); G: IPS 43000 (Anoiapithecus), reversed; H: IPS 1815 (Hispanopithecus), modified from Alba et al. (2012). Note the large difference in size between Buronius and all other dryopithcins. The Buronius specimen has tall, pointed cusps with sharp principal crista and few assessory cristae (all other taxa have secondary cristae in the trigon, usually directed between the paracone and the lingual third of the crista obliqua), a lingually concave, notched, sharp postprotocone crista, a low and deeply notched hypocone-metacone crista, a rounded, shallow cingulum remnant, a shallow buccal shelf (style), no mesial fovea, and a strongly lingually positioned hypocone. Other differences include a shorter talon, lacking the distolingual expansion of Danuvius, Dryopithecus, Pierolapithecus, Anoiapithecus and Hispanopithecus; more obliquely oriented, continuous postparacone-prematacone cristae compared with all taxa except Danuvius (in Danuvius the cristae are separated by a fissure); continuous postprotocone-prehypocone cristae in contrast to the deep notch separating the crista in all taxa except the tooth from Melchingen.


S4 Fig. Comparison of Buronius manfredschmidi with associated upper first and second molars of dryopithecins.

All teeth from the left side except Anoiapithecus, which is photographically reversed. All specimens scaled to the same M2 size (scales = 10 mm). Danuvius guggenmosi (GPIT/MA/10000-01), Rudapithecus hungaricus (Alsótelekes), Dryopithecus fontani (IPS 35026), Pierolapithecus catalaunicus (IPS 21350), Anoiapithecus brevirostris (IPS 43000), Hispanopithecus laietanus (IPS 1798).


S5 Fig. Metrical comparison of Buronius manfredschmidi to other dryopithecins.

A–M1 Size: M1 mesiodistal length times buccolingual breadth (MD)*(BL). Buronius is much smaller than any known dryopithecin M1. B–M2 size: M2 mesiodistal length times buccolingual breadth (MD)*(BL). Buronius is much smaller than any known dryopithecin M2. C–M1 length/breadth (MD/BL) ratios in Buronius and dryopithecins. The Buronius M2 is relatively broad compared with Anoiapithecus and Dryopithecus, and within the 25–75 confidence intervals in other dryopithecins. D–M2 length/breadth (MD/BL) ratios in Buronius and dryopithecins. The Buronius M2 is relatively broad compared with Danuvius and Dryopithecus, and within the 25–75 confidence intervals in other dryopithecins.


S6 Fig.

Enamel thickness on unworn upper molars (A, B) compared to worn deciduous molar (C). Distal virtual section through the tips of metacone and hypocone for the left M2 of Buronius manfredschmidi (A, GPIT/MA/13005), the left M2 of Danuvius guggenmosi (B, GPIT-MA-10002-07), the right DP4 (reversed) of Danuvius guggenmosi (C, GPIT-MA-10002-04) and the left M2 of Rudapithecus (RUD 200; modified from Smith et al., 2019). Note the conservative reconstruction of worn enamel on the tips of metacone and hypocone in the deciduous molar (B). Scale bar is 2 mm.


S7 Fig. Deciduous and permanent upper molar comparisons.

Comparisons between the dP4 and M2 of selected catarrhines. Top row, left to right: A—Buronius (M2), B—Danuvius (dP4, reversed), C—Danuvius (M2), D—Anapithecus (M2), E—Rudapithecus (dP4, RUD 124). Box: dP4 of Griphopithecus (four teeth modified from Mortzou and Andrews (2008). Two teeth in the lower right corner are casts of Griphopithecus from Pasalar (left) and from Devinska Nova Nes (Slovakia) (right). Note that the dP4s are nearly always worn. The metacones are displaced lingually, contributing to a tapered crown distobuccally. The crowns are always flared, especially lingually. Trigons are usually short and crest poorly developed.


S8 Fig. Lingual views of fossil hominoid dP4 compared with the M2 of Buronius.

Upper left: Danuvius dP4 (GPIT-MA-10002-04); Upper right, Buronius M2 (GPIT/MA/13005); Lower left, Griphopithecus dP4 from Devinska Nova Ves; Middle: Griphopithecus dP4 from Pasalar; Lower right, Rudapithecus dP4. Note the much lower crowns of the dP4s.


S9 Fig. Ratios of buccolingual breadth to mesiodistal length in selected hominoids.

The dP4 tends to be broader relative to length but there is much variation and overlap with the permanent molars. Pan paniscus has relatively broad upper molars, like Buronius.


S10 Fig. Lateral enamel thickness diphyodontic index (Zanolli et al. (2017).

This is a measure of the ratio of enamel thickness between the lower dp4 and lower m1. Since enamel thickness does not vary much between M1 and M2 and since we are comparing the dP4 and upper M2 of Buronius and Danuvius this comparison is relevant. There is considerable variability in this ratio within the samples included in Zanolli et al. (2017). The ratio resulting from a theoretical pairing of the Buronius M2 and the Danuvius dP4 (B/D) in this plot) falls outside the range of variation of all samples included here, indicating that the difference in lateral enamel thickness is too great between Buronius and Danuvius to be accommodated within a single genus. Note as well that some thickly enameled hominins (modern Homo, Australopithecus) can have values close to one. The ratio in Danuvius falls closest to the mean for modern Homo).


S11 Fig. Comparisons of the development of the protoconule in Danuvius and Buronius.

The protoconule in the dP4 of Danuvius is worn (note the large pit at its apex). See S1 File for description.


S12 Fig. Oblique views from distally showing the difference in development of the crista obliqua (yellow arrow) in Danuvius and Buronius.

See S1 File for description.


S13 Fig. The mesial fovea in relation to the protoconule.

The mesial fovea does not reach the protoconule in Danuvius while it does in Buronius, as its mesial and distal borders converge lingually on to the tip of the protoconule. Despite damage, there is a distinct ridge mesiobuccal to the protoconule on the deciduous tooth corresponding to a similar ridge on the permanent tooth of Danuvius. See S1 File for description.


S14 Fig. Enamel-dentine junction of the Buronius M2 (A, GPIT/MA/13005), compared to the dP4 (B, GPIT/MA/10002-04) and M2 (C, GPIT/MA/10002-07) of Danuvius guggenmosi.

Not to scale. Images aligned along dentine horns of protocone and metacone. See S1 File for description.


S1 Table. Summary of comparisons among the permanent molars of Danuvius and Buronius and the dP4 of Danuvius.



S2 File. Description and comparison of the left upper M2 from Danuvius guggenmosi (GPIT-MA-10002-7).



We thank two anonymous reviewers for their very thorough reading of our manuscript. Their suggestions helped to improve this publication. Furthermore, we are indebted to the following curators for granting access to collections under their care: E. Gilissen (Tervuren), A. van Heteren (Munich), I. Werneburg (Tübingen). We thank A. Fatz (Tübingen) for providing photographs of the fossil material and help with figure preparation, and C. Kyriakouli (Tübingen) for assistance with μCT scanning. Last, but not least, we acknowledge all participants of the numerous excavations in the Hammerschmiede fossil site since 2011, who helped to detect and collect the studied material. Since 2020, the excavations and associated research were supported by the Bavarian State Ministry of Research and the Arts and by the Bavarian Natural History Collections (SNSB). DRB acknowledges support from the Natural Sciences and Engineering Research Council of Canada (Grant File Number RGPIN-2016-06761).


  1. 1. Deane AS, Nargolwalla MC, Kordos L, Begun DR. New evidence for diet and niche partitioning in Rudapithecus and Anapithecus from Rudabánya, Hungary. Journal of human evolution. 2013;65(6):704–14.
  2. 2. Kordos L, Begun DR. Primates from Rudabánya: allocation of specimens to individuals, sex and age categories. Journal of human evolution. 2001;40(1):17–39.
  3. 3. McNulty KP, Begun DR, Kelley J, Manthi FK, Mbua EN. A systematic revision of Proconsul with the description of a new genus of early Miocene hominoid. Journal of Human Evolution. 2015;84:42–61. pmid:25962549
  4. 4. Harrison T. Dendropithecoidea, proconsuloidea, and hominoidea. Cenozoic mammals of Africa. 2010:429–69.
  5. 5. Böhme M, Spassov N, Fuss J, Tröscher A, Deane AS, Prieto J, et al. A new Miocene ape and locomotion in the ancestor of great apes and humans. Nature. 2019;575(7783):489–93. pmid:31695194
  6. 6. Kirscher U, Prieto J, Bachtadse V, Aziz HA, Doppler G, Hagmaier M, et al. A biochronologic tie-point for the base of the Tortonian stage in European terrestrial settings: Magnetostratigraphy of the topmost Upper Freshwater Molasse sediments of the North Alpine Foreland Basin in Bavaria (Germany). Newsletters on Stratigraphy. 2016;49(3):445–67.
  7. 7. Kargopoulos N, Kampouridis P, Lechner T, Böhme M. A review of Semigenetta (Viverridae, Carnivora) from the Miocene of Eurasia based on material from the hominid locality of Hammerschmiede (Germany). Geobios. 2021;69:25–36.
  8. 8. Kargopoulos N, Valenciano A, Kampouridis P, Lechner T, Böhme M. New early late Miocene species of Vishnuonyx (Carnivora, Lutrinae) from the hominid locality of Hammerschmiede, Bavaria, Germany. Journal of Vertebrate Paleontology. 2021;41(3):e1948858.
  9. 9. Kargopoulos N, Kampouridis P, Lechner T, Böhme M. Hyaenidae (Carnivora) from the Late Miocene hominid locality of Hammerschmiede (Bavaria, Germany). Historical Biology. 2022;34(11):2249–58.
  10. 10. Kargopoulos N, Valenciano A, Abella J, Kampouridis P, Lechner T, Böhme M. The exceptionally high diversity of small carnivorans from the Late Miocene hominid locality of Hammerschmiede (Bavaria, Germany). PloS one. 2022;17(7):e0268968. pmid:35830447
  11. 11. Fuss J, Prieto J, Böhme M. Revision of the boselaphin bovid Miotragocerus monacensis Stromer, 1928 (Mammalia, Bovidae) at the Middle to Late Miocene transition in Central Europe. Neues Jahrbuch für Geologie und Paläontologie Abh. 2015;276:229–65.
  12. 12. Hartung J, Lechner T, Böhme M. New cranial material of Miotragocerus monacensis (Mammalia: Bovidae) from the late Miocene hominid locality Hammerschmiede (Germany). Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen. 2020;298(3):269–84.
  13. 13. Hartung J, Böhme M. Unexpected cranial sexual dimorphism in the tragulid Dorcatherium naui based on material from the middle to late Miocene localities of Eppelsheim and Hammerschmiede (Germany). Plos one. 2022;17(5):e0267951. pmid:35584185
  14. 14. Lechner T, Bohme M. The beaver Steneofiber depereti from the lower Upper Miocene hominid locality Hammerschmiede and remarks on its ecology. Acta Palaeontologica Polonica. 2022;67(4).
  15. 15. Mayr H, Fahlbusch V. Eine unterpliozäne Kleinsäugerfauna aus der Oberen Süßwasser-Molasse Bayerns. Mitteilungen der Bayerischen Staatssammlung für Paläontologie und hist Geologie. 1975;15:91–111.
  16. 16. Prieto J. The genus Eomyops Engesser, 1979 (Rodentia, Eomyidae) from the youngest deposits of the German part of the North Alpine Foreland Basin. Swiss Journal of Palaeontology. 2012;131(1):95–106.
  17. 17. Fahlbusch V, Mayr H. Microtoide Cricetiden (Mammalia, Rodentia) aus der Oberen Süßwasser-Molasse Bayerns. Paläontologische Zeitschrift. 1975;49:78–93.
  18. 18. Prieto J, van den Hoek Ostende LW, Böhme M, Braze M. Reappearance of Galerix (Erinaceomorpha, Mammalia) at the Middle to Late Miocene transition in South Germany: biostratigraphic and palaeoecologic implications. Contributions to Zoology. 2011;80(3):179–89.
  19. 19. Prieto J, van Dam JA. Primitive Anourosoricini and Allosoricinae from the Miocene of Germany. Geobios. 2012;45(6):581–9.
  20. 20. Prieto J, Rummel M. Evolution of the genus Collimys DAXNER-HOCK, 1972 (Rodentia, Cricetidae)-a key to Middle to Late Miocene biostratigraphy in Central Europe. Neues Jahrbuch fur Geologie und Palaontologie-Abhandlungen. 2009;252(2):237.
  21. 21. Mayr G, Lechner T, Böhme M. The large-sized darter Anhinga pannonica (Aves, Anhingidae) from the late Miocene hominid Hammerschmiede locality in Southern Germany. PLoS One. 2020;15(5):e0232179. pmid:32374733
  22. 22. Mayr G, Lechner T, Böhme M. A skull of a very large crane from the late Miocene of Southern Germany, with notes on the phylogenetic interrelationships of extant Gruinae. Journal of Ornithology. 2020;161:923–33.
  23. 23. Mayr G, Lechner T, Böhme M. Nearly complete leg of an unusual, shelduck-sized anseriform bird from the earliest late Miocene hominid locality Hammerschmiede (Germany). Historical Biology. 2022:1–10.
  24. 24. Doppler G, Heissig K, Reichenbacher B. Die Gliederung des Tertiärs im süddeutschen Molassebecken. Newsletters on Stratigraphy. 2006:359–75.
  25. 25. Wessel P, Luis J, Uieda L, Scharroo R, Wobbe F, Smith WH, et al. The generic mapping tools version 6. Geochemistry, Geophysics, Geosystems. 2019;20(11):5556–64.
  26. 26. Bailey SE. A morphometric analysis of maxillary molar crowns of Middle-Late Pleistocene hominins. Journal of human evolution. 2004;47(3):183–98. pmid:15337415
  27. 27. Smith TM, Olejniczak AJ, Martin LB, Reid DJ. Variation in hominoid molar enamel thickness. Journal of Human Evolution. 2005;48(6):575–92. pmid:15927661
  28. 28. Martin L. Significance of enamel thickness in hominoid evolution. Nature. 1985;314(6008):260–3. pmid:3920525
  29. 29. Smith BH. Patterns of molar wear in hunter–gatherers and agriculturalists. American journal of physical anthropology. 1984;63(1):39–56.
  30. 30. Sevim-Erol A, Begun D, Sözer ÇS, Mayda S, van den Hoek Ostende L, Martin R, et al. A new ape from Türkiye and the radiation of late Miocene hominines. Communications Biology. 2023;6(1):842.
  31. 31. Nengo I, Tafforeau P, Gilbert CC, Fleagle JG, Miller ER, Feibel C, et al. New infant cranium from the African Miocene sheds light on ape evolution. Nature. 2017;548(7666):169–74. pmid:28796200
  32. 32. Pugh KD. Phylogenetic analysis of Middle-Late Miocene apes. Journal of Human Evolution. 2022;165:103140. pmid:35272113
  33. 33. Urciuoli A, Zanolli C, Almécija S, Beaudet A, Dumoncel J, Morimoto N, et al. Reassessment of the phylogenetic relationships of the late Miocene apes Hispanopithecus and Rudapithecus based on vestibular morphology. Proceedings of the National Academy of Sciences. 2021;118(5):e2015215118. pmid:33495351
  34. 34. Smith TM, Tafforeau P, Pouech J, Begun DR. Enamel thickness and dental development in Rudapithecus hungaricus. Journal of Human Evolution. 2019;136:102649. pmid:31542560
  35. 35. Zapfe H. A new fossil anthropoid from the Miocene of Austria. Current Anthropology. 1960;1(5/6):428–9.
  36. 36. Pina M, Almécija S, Alba DM, O’Neill MC, Moya-Sola S. The Middle Miocene ape Pierolapithecus catalaunicus exhibits extant great ape-like morphometric affinities on its patella: Inferences on knee function and evolution. PLoS One. 2014;9(3):e91944. pmid:24637777
  37. 37. Ward CV, Ruff CB, Walker A, Teaford MF, Rose MD, Nengo IO. Functional morphology of Proconsul patellas from Rusinga Island, Kenya, with implications for other Miocene-Pliocene catarrhines. Journal of Human Evolution. 1995;29(1):1–19.
  38. 38. Rose MD, Nakano Y, Ishida H. Kenyapithecus postcranial specimens from Nachola, Kenya. African Study Monographs, Supplement. 1996;24(3).
  39. 39. Fox AJ, Wanivenhaus F, Rodeo SA. The basic science of the patella: structure, composition, and function. The journal of knee surgery. 2012;25(02):127–42. pmid:22928430
  40. 40. Alba DM, Fortuny J, de Los Rios MP, Zanolli C, Almécija S, Casanovas-Vilar I, et al. New dental remains of Anoiapithecus and the first appearance datum of hominoids in the Iberian Peninsula. Journal of Human Evolution. 2013;65(5):573–84. pmid:23932194
  41. 41. Cazenave M, Oettlé A, Thackeray JF, Nakatsukasa M, De Beer F, Hoffman J, et al. The SKX 1084 hominin patella from Swartkrans Member 2, South Africa: An integrated analysis of its outer morphology and inner structure. Comptes Rendus Palevol. 2019;18(2):223–35.
  42. 42. Gingerich PD, Smith BH, Rosenberg K. Allometric scaling in the dentition of primates and prediction of body weight from tooth size in fossils. American Journal of Physical Anthropology. 1982;58(1):81–100. pmid:7124918
  43. 43. Smith RJ, Jungers WL. Body mass in comparative primatology. Journal of Human evolution. 1997;32(6):523–59. pmid:9210017
  44. 44. Zihlman AL. Body build and tissue composition in Pan paniscus and Pan troglodytes, with comparisons to other hominoids. The Pygmy Chimpanzee: Evolutionary Biology and Behavior. 1984:179–200.
  45. 45. Jungers WL. Scaling of postcranial joint size in hominoid primates. Human Evolution. 1991;6(5–6):391–9.
  46. 46. Stanford CB. The behavioral ecology of sympatric African apes: implications for understanding fossil hominoid ecology. Primates. 2006;47:91–101. pmid:16283423
  47. 47. Vogel ER, van Woerden JT, Lucas PW, Atmoko SSU, van Schaik CP, Dominy NJ. Functional ecology and evolution of hominoid molar enamel thickness: Pan troglodytes schweinfurthii and Pongo pygmaeus wurmbii. Journal of Human Evolution. 2008;55(1):60–74. pmid:18243275
  48. 48. Kay RF. Dental evidence for the diet of Australopithecus. Annual Review of Anthropology. 1985;14(1):315–41.
  49. 49. Raemaekers J. Ecology of sympatric gibbons. Folia Primatologica. 1979;31(3):227–45. pmid:511046
  50. 50. LaRue EA, Fahey RT, Alveshere BC, Atkins JW, Bhatt P, Buma B, et al. A theoretical framework for the ecological role of three‐dimensional structural diversity. Frontiers in Ecology and the Environment. 2023;21(1):4–13.
  51. 51. Kay RF, Madden RH, Van Schaik C, Higdon D. Primate species richness is determined by plant productivity: implications for conservation. Proceedings of the National Academy of Sciences. 1997;94(24):13023–7. pmid:9371793
  52. 52. Gouveia SF, Villalobos F, Dobrovolski R, Beltrão‐Mendes R, Ferrari SF. Forest structure drives global diversity of primates. Journal of Animal Ecology. 2014;83(6):1523–30. pmid:24773500
  53. 53. Kelley J. Identification of a single birth cohort in Kenyapithecus kizili and the nature of sympatry between K. kizili and Griphopithecus alpani at Paşalar. Journal of human evolution. 2008;54(4):530–7.
  54. 54. Kordos L, Begun DR. Rudabánya: a Late Miocene subtropical swamp deposit with evidence of the origin of the African apes and humans. Evolutionary Anthropology: Issues, News, and Reviews: Issues, News, and Reviews. 2002;11(2):45–57.