Fossil apes from the vallès-penedès basin
Abstract
Currently restricted to Southeast Asia and Africa, extant hominoids are the remnants of a group that was much more diverse during the Miocene. Apes initially diversified in Africa during the early Miocene, but by the middle Miocene they extended their geographical range into Eurasia, where they experienced an impressive evolutionary radiation. Understanding the role of Eurasian hominoids in the origin and evolution of the great-ape-and-human clade (Hominidae) is partly hampered by phylogenetic uncertainties, the scarcity and incompleteness of fossil remains, the current restricted diversity of the group, and pervasive homoplasy. Nevertheless, scientific knowledge of the Eurasian hominoid radiation has significantly improved during the last decade. In the case of Western Europe, this has been due to the discovery of new remains from the Vallès-Penedès Basin (Catalonia, Spain). Here, I review the fossil record of Vallès-Penedès apes and consider its implications. Although significant disagreements persist among scholars, some important lessons can be learned regarding the evolutionary history of the closest living relatives of humans. © 2012 Wiley Periodicals, Inc.
Box 1. Current Excavations on the Vallès-Penedès BasinA large-scale paleontological intervention, devised for controlling the removal of Miocene fossiliferous sediments by excavator machines during the construction of a landfill, is being carried out at Abocador de Can Mata (ACM) in els Hostalets de Pierola.12 These works (Figs. 2A–C) have proceeded almost without interruption since 2002, leading to the recovery of tens of thousands of vertebrate remains. The deposition of the Miocene sediments within an alluvial context favored the preservation of vertebrate remains in this area.12, 14 More than 200 localities have been recognized within the 250 m-thick ACM local stratigraphic series, which spans from ca. 12.5 to 11.4 Ma (late middle Miocene), on the basis of litho-, bio-, and magnetostratigraphic correlation.12, 14, 18 Most of the lower part of the ACM series is correlated to MN7 (ca. 12.4–11.8 Ma), whereas the upper part of the series is correlated to MN8 (ca. 11.8–11.2 Ma).14, 18 The recorded mammalian fauna14, 19 includes several rare to uncommon taxa, such as pliopithecoid20, 21 and hominoid13-15 primates. The associated fauna indicates that ACM paleoenvironments were more similar to those of French and Central European localities than to the more arid and less forested habitats of inner Iberian basins,19, 22 thus explaining why Miocene primates are not recorded outside Catalonia.
The site of Can Llobateres (CLL) was discovered in 1926 during the construction of a road. Subsequently exploited as a clay quarry, it has been intermittently excavated by paleontologists since 195816 (Fig. 2D–E). Two different localities must be distinguished: the older, CLL1, which delivered most of the dentognathic hominoid remains;5, 6, 17 and the slightly younger CLL2, which yielded a partial skeleton of Hispanopithecus laietanus.7-9 The lithological and sedimentological features of CLL1 are indicative of sedimentation in a poorly drained area, with development of ponds and very shallow small lakes.6, 16, 23 Biostratigraphically, the CLL series records the early/late Vallesian boundary,23 with estimated ages of 9.72 Ma (MN9) for CLL1 and 9.64 Ma (MN10) for CLL2, on the basis of litho-, bio- and magnetostratigraphic data.16, 18, 23 After years of inactivity, excavations at CLL1 were resumed in 2010 by a team from the Institut Català de Paleontologia Miquel Crusafont16 (Fig. 2E). This work has yielded plant macroremains that provide a more accurate reconstruction of the paleoenvironment of H. Laietanus,16, 24 as well as new dental remains17 that significantly increase the hypodigm of this taxon.
Schematic geological map of the Vallès-Penedès Basin, indicating the situation of the hominoid fossil localities. Modified from an original kindly provided by Isaac Casanovas-Vilar. Abbreviations for fossil localities as in Table 1.
(A) Aerial view of Abocador de Can Mata (ACM) in 2010, by courtesy of CESPA Gestión de Residuos, S.A.U. (B) Paleontological control at ACM in 2011, © FOSSILIA Serveis Paleontològics i Geològics, S.L. (C) Systematic excavations at ACM/BCV1 in 2003, by courtesy of Salvador Moyà-Solà. (D) Miquel Crusafont and other paleontologists visiting the Can Llobateres outcrop during the 1950s, © Arxiu Miquel Crusafont – ICP. (E) Excavations at Can Llobateres 1 in 2010. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
THE VALLÈS-PENEDÈS HOMINOIDS
Box 2. Systematics of the Hominoidea
The infraorder Catarrhini (Primates: Anthropoidea) includes the extant superfamilies Cercopithecoidea (Old World monkeys) and Hominoidea (apes and humans). The hominoid crown group is composed of the extant Hylobatidae (gibbons and siamangs) and Hominidae (humans and great apes). The hominoid stem lineage is composed of all those taxa that are more closely related to crown hominoids than to its living sister taxon, the Cercopithecoidea.15, 18 A distinction between the superfamilies Proconsuloidea and Hominoidea, respectively referring to stem and crown hominoids within the magnafamily Hominidea,25, 26 is unnecessary.18 Unfortunately, disagreement persists as to what extinct taxa follow, rather than precede the cercopithecoid-hominoid splitting, and hence should be interpreted as stem hominoids instead of stem catarrhines. Some authors consider that the Dendropithecidae, Proconsulidae, and Afropithecidae precede the monkey-ape divergence.27, 28 That view, however, contrasts with the most widely held opinion, that at least proconsulids and afropithecids postdate that split15, 18, 25, 29-32 and thus are members of the Hominoidea total group.
A broad concept of the Hominidae, including both humans and great apes, is employed here, following the use currently favored by most paleoprimatologists.15, 18, 25, 29, 30, 33, 34 This contrasts with the traditional distinction between Hominidae s.s., for humans and their closest extinct relatives, and “Pongidae,” for great apes, which is still employed by some paleoanthropologists, even though the latter is clearly a paraphyletic assemblage. Two extant subfamilies are distinguished, the Ponginae (orangutans) and Homininae (African apes and humans). However, difficulties arise when trying to classify extinct taxa as either stem hominoids, stem hominids, or crown hominids, largely due to the uncertain phylogenetic position of hylobatids. Thus, within the Hominoidea total group, the Proconsulidae and Afropithecidae s.s. may be either successive stem hominoids or hominids, depending on their particular branching order with regard to hylobatids.18 Two distinct afropithecid subfamilies have been sometimes recognized,15, 18 the basal Afropithecinae and the more derived Kenyapithecinae (tribes Equatorini and Kenyapithecini). However, given the apparently closer relationship between kenyapithecins and dryopithecins,15, 18 the Kenyapithecinae are formally transferred here into the Hominidae. The status of the Dryopithecinae, which other authors26, 29, 33-35 recognize only at the tribe rank, as a paraphyletic or strictly monophyletic assemblage of stem or crown hominids, is uncertain. Three distinct tribes are here provisionally distinguished among the Dryopithecinae,18 with the Oreopithecini being left as incertae sedis at the subfamily level due to phylogenetic uncertainties. Several tribes, the Pongini, Sugrivapithecini, and Lufengpithecini new subtribe (type genus, Lufengpithecus Wu, 1987) are distinguished here within the Ponginae,29, 34 even though their phylogenetic relationships are not well resolved. Some authors have employed Sivapithecini instead of Sugrivapithecini,29, 33, 35, 36 but the latter takes priority. Lufengpithecini was apparently first employed in 2002,33 but its erection failed to fulfill several requirements of the International Code of Zoological Nomenclature (Articles 16.1, 16.2 and arguably also 14), so that it is made available here.
Range chart for Miocene hominoids from Western Eurasia, reprinted from Casanovas-Vilar and coworkers21 (© by the authors). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
| Localitya | Age (Ma) | MN | Description | Taxon |
|---|---|---|---|---|
| HP | 12.5-11.218 | MN7-MN9 | M2 | Hominidae indet.2, 18 |
| TEF | 10.4-10.018 | MN9 | Partial mandible | Hispanopithecus crusafonti1, 14, 18 |
| PO2 | 11.1-9.718 | MN9 | Female C1 | Hispanopithecus laietanus5 |
| CM1 | 11.218 | MN8 | Female C1 | Hominidae indet.5 |
| CV | 12.5-11.213, 18 | MN7-MN8 | M2-M3 | “Sivapithecus” occidentalis n. dub.3, 13 |
| LT1 | 9.518 | MN10 | Lower series and partial mandible | Hispanopithecus laietanus3, 5 |
| CP1 | 10.4-10.018 | MN9 | Dentognathic remains and postcranials | Hispanopithecus crusafonti1, 14 |
| CB | 11.2-10.537 | MN8-MN9 | A few postcranials | cf. Dryopithecus fontani38, 47 |
| CLL1 | 9.7218 | MN9 | Dentognathic remains and postcranials | Hispanopithecus laietanus5, 6, 17, 18 |
| CLL2 | 9.6518 | MN10 | Partial face and skeleton | Hispanopithecus laietanus7-9, 18 |
| EDAR6 | 10.4-10.018 | MN9 | Phalanges | Hispanopithecus laietanus10, 18 |
| EDAR13 | 10.4-10.018 | MN9 | Unpublished remains | Hispanopithecus laietanus10, 18 |
| CFE | 10.0-9.711 | MN9 | Unpublished partial skeleton | Hispanopithecus laietanus11 |
| SQ | 11.8-11.218 | MN8 | Lower molar fragment | Hominidae indet.5, 18 |
| ACM/BCV1 | 11.9318, 21 | MN7 | Partial cranium and skeleton | Pierolapithecus catalaunicus13 |
| ACM/BCV4 | 11.9118, 21 | MN7-MN8 | Mandibular fragment | Hominidae indet.18 |
| ACM/C1-E* | 12.3-12.218, 21 | MN7 | Upper partial postcanine series | Hominidae indet.18 |
| ACM/C3-Ae | 11.8518, 21 | MN7-MN8 | Partial face with palate | Dryopithecus fontani14 |
| ACM/C3-Az | 11.9118, 21 | MN7-MN8 | Partial femur | cf. Dryopithecus fontani14, 18 |
| ACM/C3-Aj | 11.9418, 21 | MN7 | Partial face | Anoiapithecus brevirostris15 |
| ACM/C4-Ap | 11.8518, 21 | MN7-MN8 | Unpublished remains | Hominidae indet.18 |
| ACM/C4-Cp | 11.9218, 21 | MN7 | Unpublished remains | Hominidae indet.18 |
- a Locality acronyms (type localities indicated by an asterisk): ACM, Abocador de Can Mata local stratigraphic series; BCV1, Barranc de Can Vila 1; BCV4, Barranc de Can Vila 4; C1, ACM Cell 1; C3, ACM Cell 3; C4, ACM Cell 4; CB, Castell de Barberà; CFE, Can Feu; CLL1, Can Llobateres 1; CLL2, Can Llobateres 2; CM1, Can Mata 1 (= Bretxa de Can Mata); CP1, Can Poncic 1; CV, Can Vila; EDAR6, Estació Depuradora d'Aigües Residuals 6; EDAR8, Estació Depuradora d'Aigües Residuals 8; HP, Hostalets de Pierola indeterminate (Aragonian or Vallesian levels); LT1, La Tarumba 1; PO2, Polinyà 2; SQ, Sant Quirze – Trinxera del Ferrocarril; TEF, Teuleria del Firal.
| Superfamily incertae sedis |
| Family incertae sedis [Limnopithecus†b, Lomorupithecus†, Kalepithecus†, Kogolepithecus†] |
| Family Dendropithecidae† [Dendropithecus†, Micropithecus†, Simiolus†] |
| Superfamily Hominoidea |
| Family incertae sedis [Kamoyapithecus†] |
| Family Proconsulidae† |
| Subfamily Proconsulinae† [Proconsul†] |
| Subfamily Nyanzapithecinae† [Nyanzapithecus†, Mabokopithecus†, Rangwapithecus†, Turkanapithecus†, Xenopithecus†] |
| Subfamily incertae sedis [Samburupithecus†] |
| Family Afropithecidae† |
| Subfamily Afropithecinae† [Afropithecus†, Heliopithecus†] |
| Subfamily incertae sedis [Otavipithecus†] |
| Family Hylobatidae [Hylobates] |
| Family Hominidae |
| Subfamily Kenyapithecinae† |
| Tribe Equatorini† [Equatorius†, Nacholapithecus†] |
| Tribe Kenyapithecini† [Kenyapithecus†, Griphopithecus†] |
| Subfamily Dryopithecinae† |
| Tribe Dryopithecini† |
| Genus Dryopithecus† |
| Dryopithecus fontani† |
| Genus Pierolapithecus† |
| Pierolapithecus catalaunicus† |
| Genus Anoiapithecus† |
| Anoiapithecus brevirostris† |
| Genus incertae sedis |
| “Sivapithecus” occidentalis† nomen dubium |
| Tribe Hispanopithecini† |
| Genus Hispanopithecus† |
| Hispanopithecus (Hispanopithecus) laietanus† |
| Hispanopithecus (Hispanopithecus) crusafonti† |
| Tribe Ouranopithecini† [Ouranopithecus†] |
| Tribe incertae sedis [?Udabnopithecus†] |
| Subfamily Ponginae |
| Tribe Sugrivapithecini† [Sivapithecus†, Ankarapithecus†] |
| Tribe Lufengpithecini† [Lufengpithecus†, Khoratpithecus†] |
| Tribe Pongini [Pongo, Indopithecus†, Gigantopithecus†] |
| Subfamily Homininae |
| Tribe Gorillini [Gorilla] |
| Tribe Panini [Pan] |
| Tribe Hominini [Homo, Australopithecus†, Paranthropus†, Ardipithecus†, Orrorin†] |
| Tribe incertae sedis [Nakalipithecus†, Chororapithecus†, Sahelanthropus†] |
| Subfamily incertae sedis |
| Tribe Oreopithecini† [Oreopithecus†] |
- a The exception is the Vallès-Penedès taxa, shown to the species level (bold type).
- b The dagger symbol (†) denotes extinct taxa.
Pierolapithecus catalaunicus
Pierolapithecus is known from the holotype: a facial and postcranial partial skeleton of an adult male from ACM/BCV1 (Figs. 4A and H; 5A, 6A, and 7A), with an estimated body mass around 30-35 kg.19 Although an age of ca. 12.5-13.0 Ma was initially proposed,13 ACM/BCV1 has since been dated to 11.9 Ma on the basis of magnetostratigraphic data.14, 18, 21 Some authors remain skeptical about the taxonomic distinctiveness of Pierolapithecus and Dryopithecus,25, 26, 29, 35 but here they are considered different taxa. Besides the splanchnocranium (premaxilla, maxillae, nasals, zygomatics, lacrimals, partial frontal bone, and most of the upper dentition), the P. catalaunicus holotype comprises about 80 postcranial specimens from the hands (carpals, metacarpals, and manual phalanges), feet (tarsals, metatarsals, and pedal phalanges), girdles (clavicular and pelvic fragments), vertebrae, ribs, patella, and limb bones.
Dental remains of Vallès-Penedès hominoids (see Table 1 for locality abbreviations). (A) Right I1 of P. catalaunicus from ACM/BCV1 (IPS21350, holotype, reversed); (B) Left I1 of H. crusafonti from CP1 (IPS1808); (C) Right I1 of H. crusafonti from CP1 (IPS1807, reversed); (D) Left I2 of H. crusafonti from CP1 (IPS1809); (E) Right I1 of H. laietanus from CLL1 (IPS1770, reversed); (F) Left I1 of H. laietanus from CLL2 (IPS18800.9); (G) Right I1 of H. laietanus from CLL1 (IPS1778, reversed); (H) Right P3-M3 of P. catalaunicus from ACM/BCV1 (IPS21350, holotype, reversed); (I) Left P3-M3 of D. fontani from ACM/C3-Ae (IPS35026); (J) Left P3-M2 and right M3 (reversed) of A. brevirostris from C3-Aj (IPS43000, holotype); (K) Right M1-M3 of H. laietanus from CLL2 (IPS18000.5, reversed); (L) Left M3 of H. laietanus from CLL1 (IPS1772); (M) Left M1-M2 of H. crusafonti from CP1 (IPS1798, holotype); (N) Right M3 of H. crusafonti from CP1 (IPS1812, reversed); (O) Left m2-m3 of “Sivapithecus” occidentalis nomen dubium from CV (IPS1826+1827, holotype); (P) Left p3-m2 of A. brevirostris from ACM/C3-Aj (IPS43000, holotype); (Q) Right p3-m2 (reversed) and left m3 of H. laietanus from LT1 (IPS1803, holotype); (R) Right m1-m3 of H. laietanus from CLL1 (IPS1802, reversed); (S) Left m1-m2 of H. laietanus from CLL1 (IPS1796); (T) Right m1-m2 of H. laietanus from CLL1 (IPS1797, reversed); (U) Left m1-m2 of H. laietanus from CLL1 (IPS9001); (V) Left m3 of H. laietanus from CLL1 (IPS1800); (W) Left m3 of H. laietanus from CLL1 (IPS1822); (X) Right m2 of H. crusafonti from CP1 (IPS1816, reversed). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
(A) Face of Pierolapithecus catalaunicus from ACM/BCV1 (IPS21350, holotype, modified from Moyá-Solá and colleagues16). (B) Partial face of Dryopithecus fontani from ACM/C3-Ae (IPS35026, modified from Moyá-Solá and coworkers14). (C) Reconstruction of the face of Anoiapithecus brevirostris from ACM/C3-Aj (IPS43000, holotype, modified from Moyá-Solá and colleagues18, © by the authors). (D) Reconstruction of the face of Hispanopithecus laietanus from CLL2 (IPS18000). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
(A) Partial skeleton of Pierolapithecus catalaunicus (IPS21350, holotype, modified from Moyà-Solà and coworkers13). (B) Partial skeleton of Hispanopithecus laietanus from CLL2 (IPS18800, courtesy of Salvador Moyà-Solà). (C-D) Right partial femora of cf. Dryopithecus fontani from ACM/C3-Az (C, IPS41724, modified from Moyà-Solà and coworkers17) and Hispanopithecus laietanus from CLL2 (D, IPS18800, courtesy of Marta Pina), in posterior and anterior views. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
(A) Life reconstruction of the Pierolapithecus catalaunicus cranium (courtesy of Meike Köhler). (B) Life reconstruction of the Anoiapithecus brevirostris cranium (by Marta Palmero, © ICP). (C) Life reconstruction of Hispanopithecus laietanus. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Pierolapithecus displays a low face with a posteriorly situated glabella and a prognathous profile, coupled with a modern, great-ape-like configuration (frontal processes of the maxillae, nasals, and orbits on the same plane; flat nasals projecting anteriorly beneath the level of the lower orbital rim; high zygomatic root and long nasoalveolar clivus; wide and deep palate; and wide nasal aperture widest close to the base).13 Postcranial remains indicate that Pierolapithecus had a modern-hominoid-like, orthograde body plan.13, 40 Although orthograde-related features were present in the early Miocene afropithecid Morotopithecus,39, 40 the Pierolapithecus skeleton represents the oldest unequivocal evidence of an orthograde body plan in a fossil hominid,13, 40 by displaying clear orthograde-related characteritics associated to cranial features more derived than those of hylobatids.13
Pierolapithecus orthograde features include short lumbar vertebrae with transverse processes arising from the base of the pedicle, a relatively marked rib curvature, and a long, stout clavicle, indicating a broad and shallow thorax with dorsally situated scapulae. The wrist (triquetrum) morphology indicates a lack of ulnocarpal articulation.13, 40Pierolapithecus, however, still retains manual adaptations for powerful grasping and above-branch palmigrady: short metacarpals; stout, short, and only moderately curved proximal phalanges, with a dorsally oriented proximal articulation; large and laterally protruding palmar tubercles; and a long, palmarly unbent trochlea.16, 41-43 These features, together with relative phalangeal length and curvature, have led to conflicting views on whether this taxon lacked41-43 or possessed35, 44-46 suspensory adaptations.
Dryopithecus fontani
A partial face of an adult male from ACM/C3-Ae (Figs. 4I, 5B), with an estimated age of 11.9 Ma, but somewhat younger than ACM/BCV1,21 has been attributed to Dryopithecus fontani,14, 18 based on several craniodental features that do not fit with other Vallès-Penedès hominoids.14, 15 In particular, this specimen differs from Pierolapithecus in its higher and less posteriorly inclined face (as shown by the nasal aperture and zygomatics), more anteriorly situated anterior nasomaxillary contact, and more domed muzzle (as shown by the laterally facing frontal processes of the maxilla, which are not situated on the same plane as the nasals), as well as in several occlusal details and the overall larger size of its dentition.14 If the attribution to D. fontani is correct, this is the only specimen representing the facial morphology of the type species of Dryopithecus, previously known from mandibles and a few isolated upper teeth from several middle Miocene European localities.18
The D. fontani facial specimen displays a mosaic of derived great-ape features, including a large nasal aperture that is widest at the base, with the edges of the nasal aperture formed by the maxillae; a wide palate; a moderately high zygomatic root; and a distinctly vertical nasomaxillary suture. These features are coupled with several primitive hominoid retentions, among them a restricted maxillary sinus that does not penetrate into the zygomatic root; an upwardly inclined zygomatic arch; a stepped nasal floor with an open palatine fenestra; and proconsulid-like dental proportions.14 At the same time, the face of D. fontani displays several derived gorilla-like features: a very anteriorly situated anterior nasomaxillary contact, with a long nasomaxillary suture that is less vertically oriented than are the lateral rims of the nasal aperture, so that the latter closely approaches the posterior pole of the nasoalveolar clivus, and a lower orbital rim situated high above the tooth row and anteriorly from the zygomatic root. These characteristics might be either homologous or homoplastic with regard to hominines.14
A partial (proximal) femur from ACM/C3-Az (11.9 Ma),21 with an estimated body mass of about 44 kg, has been tentatively attributed to D. fontani, mainly on the basis of size (Fig. 6C).14 Among this specimen's modern hominoid-like characteristics are an absence of crista trochanterica, an anteroposteriorly compressed shaft, and a large femoral head situated over the greater trochanter, which are coupled with pronograde-related features such as a low position of the femoral head relative to the greater trochanter, a short and not markedly inclined femoral neck, and posteriorly situated lesser trochanter.14
D. fontani has also been recognized at Castell de Barberà. This site is generally attributed to MN8,4, 18, 24, 38, 47 although it is, in fact, best attributed either to latest MN8 or earliest MN9 (11.2 to 10.5 Ma).37 The presence of hominoids at this locality was originally reported based on a purported female upper canine5 that actually belongs to a male pliopithecid.6, 37, 38 However, the presence of hominoids at Castell de Barberà is substantiated by two pollical phalanges, proximal and distal, and a partial humeral diaphysis,4, 38, 47 which, on the basis of size (estimated body mass around 50 kg), has been also tentatively attributed to D. fontani.38 This humeral diaphysis displays several modern hominoid-like features (a conspicuous lateral supracondylar ridge; a circular proximal shaft cross section that becomes distally flattened, being only moderately convave medially and somewhat concave anteriorly; a deep olecranon fossa surrounded by a broad, flat lateral shaft portion; an apparently shallow radial fossa and well-developed coronoid fossa; and a presumably medially oriented medial epicondyle), most closely resembling, in spite of differences in overall size and robusticity, the more complete humerus of D. fontani from Saint Gaudens, France.38 The phalanges more closely resemble those of other Miocene apes than those of extant great apes, indicating a significant degree of thumb-assisted powerful grasping during above-branch quadrupedalism and cautious climbing.47
Anoiapithecus brevirostris
A. brevirostris has been described on the basis of a face with partially preserved maxillae, orbits, frontal, and zygomatics, and an associated mandible, including the symphysis and a large portion of the two corpora (Figs. 4J,P, 5C, 7B) from ACM/C3-Aj.15 Unpublished dental remains from a second individual have been recovered from the same locality, which is situated several dozens of meters away and stratigraphically positioned 21 m below ACM/C3-Ae, resulting in an estimated age almost 0.1 million years older.21 As with Pierolapithecus, doubts have been raised regarding the distinct taxonomic status of Anoiapithecus relative to Dryopithecus.25, 26, 29, 35 Although Anoiapithecus resembles Pierolapithecus and Dryopithecus in several features, such as its restricted maxillary sinus, Anoiapithecus can be distinguished from both Pierolapithecus and Dryopithecus by its strikingly orthognathous facial profile.15 Although some authors have dismissed this on the basis of purported distortion,35 there is no evidence of the extreme deformation required to transform the orthognathous face of Anoiapithecus into the prognathous face of Pierolapithecus. Moreover, Anoiapithecus differs from Pierolapithecus in the presence of a frontal sinus, its downwardly inclined zygomatic root, and several dental details.15, 48 In turn, Anoiapithecus differs from Dryopithecus in the lower degree of nasoalveolar prognathism, the more posteriorly situated anterior zygomatic root, the more buccolingually compressed male upper canine, and its thicker molar enamel.14, 15, 49Anoiapithecus shares several derived features with kenyapithecins, including the anterior position of the zygomatic root, a short mandible with a marked inferior torus, and a deep canine fossa.15 It also shares several hominid synapomorphies with other dryopithecines, including a high face with high zygomatic root, a wide nasal aperture that is widest at the base, a deep palate, and nasals that project slightly anteriorly beneath the level of the lower orbital rims.15
Hispanopithecus laietanus
Hispanopithecus laietanus was originally defined on the basis of lower jaw and dental remains from La Tarumba 1 (Fig. 4Q).3 Subsequently, most (but not all5) authors treated Hispanopithecus as a junior subjective synonym of Dryopithecus.4, 5, 33, 50, 51 Recently, however, Hispanopithecus, including the type species, H. laietanus, as well as other species previously lumped into Dryopithecus, was resurrected.14 This proposal has been followed by most researchers, at least with respect to H. laietanus.16, 26, 29, 38, 41, 52 Several Vallesian localities, including Polinyà 2,4, 5, 18 EDAR,10, 18 and Can Feu,18 have yielded fragmentary remains attributable to this taxon.11 However, this species is best known from the samples of Can Llobateres 1 and 2 (Box 1).5-9, 17, 52, 50-52 The few remains initially recovered from Can Llobateres 1 were attributed to H. laietanus, but subsequent finds (Figs. 4E,G,L,R–W) resulted in the naming of two additional taxa (Dryopithecus piveteaui and Rahonapithecus sabadellensis), currently considered nomina nuda.4, 6 Since the early 1990s, the sample from Can Llobateres 1 and 2 has been attributed to a single species, Dryopithecus laietanus1, 4, 6-9, 25, 33, 50, 51, 53 or now Hispanopithecus laietanus,14, 16, 18, 36, 37, 43, 52 with some exceptions.5, 54
The cranial anatomy of H. laietanus is known from the facial fragments and temporal bone with petrosal from Can Llobateres 2 (Figs. 4F,K, 5D),7, 8, 55, 56 with an estimated age of 9.6 Ma.18 This taxon displays several hominid-like features, including the lack of a subarcuate fossa, a shallow malar region, a high zygomatic root, and a deep glenoid fossa with a prominent entoglenoid process. At the same time, several craniodental features, including a more extensive maxillary sinus, reduced midfacial prognathism, higher zygomatic roots, and upper canine morphology, indicate that H. laietanus is more derived than the middle Miocene dryopithecins toward the modern great-ape condition. Hispanopithecus further differs from the dryopithecines in other features, such as a more orthognathous face than in Pierolapithecus and Dryopithecus, and the small frontal sinus, which is absent in Pierolapithecus and more extensive in Anoiapithecus.15, 48 Given these morphologic differences, there is current agreement to restrict Dryopithecus to middle Miocene species.16, 15, 26, 29, 35
No definitive consensus exists, however, regarding whether Hispanopithecus must be restricted to the Iberian species.26, 29, 35, 36 The Hungarian remains from Rudabánya, formerly referred to as Dryopithecus brancoi25, 33, 53 or D. carinthiacus,51 are currently designated as either Hispanopithecus hungaricus14, 18 or Rudapithecus hungaricus.26, 29, 35H. laietanus differs from the Rudabánya hominoid in several cranial features, namely, in having a more concave frontal squama, a biconvex premaxilla, and more anteriorly directed zygomatic.,26, 35 These features, coupled with some postcranial differences,35 may justify a taxonomic distinction above the species level, either as a different genus26, 29, 35 or subgenus,18 the latter being the option I provisionally favor.
The postcranial anatomy of H. laietanus is mostly known from the partial skeleton (comprising about 60 specimens) from Can Llobateres 2 (Figs. 6B,D, 7C),9, 52, 56 which is associated with the above-mentioned face. Several features of the thoracic and lumbar vertebrae indicate a wide and shallow thorax associated with an orthograde body plan.9, 52, 56 In turn, inferred limb proportions,9, 56 femoral morphology,9, 57, 58 and phalangeal features9, 43, 52 indicate adaptations for forelimb-dominated, below-branch suspensory behaviors, including a high intermembral index and a long hand, particularly at the phalangeal region. At the same time, the metacarpal and proximal phalangeal morphology indicates the retention of features functionally related to above-branch quadrupedalism,9, 43, 52, 56 although to a lesser extent than in Pierolapithecus. Such conclusions are confirmed by a female partial skeleton of H. laietanus from Can Feu, which, of the available Hispanopithecus remains, most completely preserves the proximal ulna.11
Hispanopithecus crusafonti
Most of the dentognathic remains attributed to H. crusafonti come from Can Poncic 1 (Figs. 4B–D, M–N,X). Initial finds were assigned to Hispanopithecus laietanus but, after the recovery of more material, Dryopithecus crusafonti was ultimately erected on its basis.1 After its description, this taxon was accepted,25, 33, 54 questioned,51 or formally synonymized with D. laietanus,53 depending on the authors, although most recently it has been transferred to the genus Hispanopithecus.14, 18, 26, 36, 59 On the basis of dental differences between the Can Poncic 1 sample and those from later Vallesian localities attributed to H. laietanus,1, 72 a distinct species status is favored here. The attribution to Hispanopithecus is based on several derived dental features shared with the other species of this genus, although the lack of cranial material for H. crusafonti makes this generic attribution somewhat provisional. The partial mandible from Teuleria del Firal, traditionally attributed to Dryopithecus fontani,50, 51, 52 is currently attributed to Hispanopithecus crusafonti,14, 18 based on dental differences with regard to D. fontani, as well as similarities with the few lower teeth available from the type locality of the Hispanopithecus crusafonti.1
Remains of Uncertain Taxonomic Attribution
Fragmentary hominoid dentognathic remains of uncertain taxonomic attribution from the Vallès-Penedès Basin include specimens from els Hostalets de Pierola, Can Mata 1, and Sant Quirze,2, 4, 5, 50, 51 as well as unpublished remains from various ACM localities.18 Of particular relevance are the upper teeth from the ACM/C1-E* (ca. 12.3-12.2 Ma), which provide the oldest record of hominoids in the Vallès-Penedès Basin.18, 21 The two lower molars from Can Vila (Fig. 4O)3, 5, 54 are also worth mentioning, since they constitute the holotype of “Sivapithecus” occidentalis. This nominal taxon, currently considered a nomen dubium,14, 14, 18 is potentially a senior subjective synonym of other Vallès-Penedès hominoid species, although this seems difficult to resolve with the currently available material.
SYSTEMATICS AND PHYLOGENY
Systematics of the Miocene Hominoids from Spain
In this paper, a broad concept of the Hominidae is employed, including not only humans but also great apes (Table 2, Box 2). Besides the two subfamilies of crown hominids, Ponginae and Homininae, two subfamilies of putative stem hominids, Kenyapithecinae and Dryopithecinae, are also included. The middle Miocene kenyapithecines include the African equatorins Nacholapithecus61 and Equatorius,62 as well as the kenyapithecins Kenyapithecus63, 64 and Griphopithecus33, 51, 63 from Africa, Turkey, and Central Europe. In turn, all the fossil hominoids from the Vallès-Penedès Basin are here included in the Dryopithecinae, which consists of the tribes Dryopithecini, Hispanopithecini, and Ouranopithecini. The two former tribes are recorded at the Vallès-Penedès Basin, the latter including the late Miocene genus Ouranopithecus from Eastern Europe and Turkey.26, 33, 35 This systematics follows the recognition that dryopithecines are fossil great apes instead of stem hominoids13, 15, 18, 25, 26, 29, 35, 49 (Fig. 8).
Evolutionary tree and schematic cladogram of the Hominoidea depicting the phylogenetic relationships favored in this paper. The evolutionary tree incorporates chronological and paleobiogeographic information, and further provides notes on locomotor evolution; continuous lines represent known stratigraphic ranges and discontinuous lines represent hypothetical phylogenetic relationships. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
The middle Miocene dryopithecines Pierolapithecus, Anoiapithecus, and Dryopithecus are included in the Dryopithecini, whereas the late Miocene Hispanopithecus is included in the Hispanopithecini.18 Previous diagnoses of this group at the tribe level26, 29, 35 were based on several features, such as thin enamel, large maxillary sinuses, and well-developed frontal sinuses,26 that have been contradicted by recent finds. Thus, middle Miocene dryopithecins actually have restricted maxillary sinuses,14, 15 unlike the derived, more enlarged maxilary sinus of the late Miocene hispanopithecins.8, 48 Furthermore, it is now known that not all dryopithecines display a frontal sinus (it is absent in Pierolapitheus14, 48) or thin molar enamel (Pierolapithecus and Anoiapithecus do not15, 49).
Phylogenetic Hypotheses
The recent finds of fossil hominoids from the Vallès-Penedès Basin have revealed a previously unsuspected paleodiversity of European middle Miocene hominoids,18, 35, 54, 60 although some authors remain skeptical about their generic diversity.26, 35 These finds have provided a new wealth of information that can be used to evaluate previous phylogenetic hypotheses. In particular, given the intermediate chronology of the middle Miocene finds between kenyapithecines andlate Miocene dryopithecines, the data provided by Pierolapithecus, Anoiapithecus and Dryopithecus are of the utmost importance for assessing the role of kenyapithecines in the origin of the great-ape-and-human clade.
Dryopithecini
In spite of its prognathous and relatively low face, Pierolapithecus displays a modern, great-ape-like facial configuration.13 This led to the conclusion that this taxon is a stem hominid, in agreement with the crown-hominoid status suggested by its orthograde body plan.13Dryopithecus and Anoiapithecus similarly display several hominid facial synapomorphies,14, 15 although the cranial morphology of each is distinct: whereas Anoiapithecus displays a very orthognathous facial profile interpreted as autapomorphic,15Dryopithecus has several gorilla-like lower-facial features.14 All dryopithecin genera retain several kenyapithecine-like features, such as a restricted maxillary sinus situated well above the molar roots,15, 49 which contrast with the more derived condition of hispanopithecins and crown hominids. At the same time, kenyapithecins, and especially dryopithecines, share several synapomorphies with crown hominids, such as having the zygomatic root situated anteriorly and relatively high above the alveolar plane. These synapomorphies suggest that these taxa are more derived than either hylobatids or putative stem hominoids from Africa (proconsulids and afropithecids).32, 64-66 On this basis, a sister-taxon relationship between kenyapithecines and dryopithecines plus crown hominids can be inferred, being formally recognized here by including kenyapithecines in the Hominidae (Table 2, Box 2).
The gorilla-like facial morphology of Dryopithecus, coupled with its apparently African-ape-like, thin dental enamel,49 might be interpreted as suggesting a hominine status. The Dryopithecus facial anatomy, however, is more incompletely known than that of other dryopithecins, and its gorilla-like features might be merely homoplastic.14 Moreover, enamel thickness is very unreliable, given its evolutionary plasticity.49 More detailed studies would be required to confirm that the thinner enamel of Dryopithecus does not merely result from a higher degree of wear. Cranial evidence for Pierolapithecus and Anoiapithecus, in any case, is at odds with the claim that all dryopithecins are best interpreted as stem-hominines,25, 26, 29, 31, 35, 44 since these taxa lack clear-cut hominine synapomorphies.13, 15, 42 Unlike Anoiapithecus and Hispanopithecus, Pierolapithecus lacks a frontal sinus,15, 48 and, among hominids, lack of this sinus is generally considered to be a synapomorphy of the Sivapithecus-Pongo clade,67-69 the sinus being present in stem catarrhines and stem hominoids,8, 69, 70 but absent in cercopithecoids and hylobatids.69, 70 The lack of a frontal sinus in Pierolapithecus might therefore be interpreted as indicating a stem-pongine (instead of hominine) status.48 However, given the repeated loss of this feature during catarrhine evolution, the retention of a frontal sinus in the probable stem pongine Ankarapithecus,67 and the presence of a well-developed frontal sinus in the dryopithecin Anoiapithecus,15 the Pierolapithecus condition in thisregard might be merely homoplastic.48
Hispanopithecini
Several cranial features indicate that Hispanopithecus belongs to the great-ape-and-human clade.7, 8, 56 Its inclusion in the Hominidae is therefore accepted by most researchers,18, 25, 26, 29-31, 33, 35, 36, 49 although uncertainties persist regarding its phylogenetic relationships within hominids. Some authors have favored an expanded concept of the Ponginae,56, 68 including both Hispanopithecus and Ouranopithecus. The pongine status of Hispanopithecus is supported by several putative synapomorphies with Pongo, including a low frontozygomatic suture; a flat, robust and anteriorly oriented zygomatic; and a low-positioned glenoid fossa relative to the external auditory meatus.7, 8, 56 However, the phylogenetic significance of these features requires re-evaluation in light of the recently recognized higher diversity of Miocene dryopithecines. For example, H. hungaricus displays a somewhat more convex and laterally facing zygomatic,8, 26 thus casting doubts on the interpretation of the H. laietanus morphology as an unequivocal pongine synapomorphy.
Some authors have alternatively supported the view that Hispanopithecus is a stem hominid71 or, more frequently, a stem hominine.25, 26, 29, 31, 33, 35 Its purported hominine status stands on the claim that it displays an “incipient” supraorbital torus, an ethmofrontal sinus, and a gorilla-like subnasal morphology,25, 26, 29, 31, 33, 35 but others have questioned the interpretation of these characters.7, 8, 48, 56 First, the slightly thickned supraorbital rims of H. laietanus and H. hungaricus, discontinuous at the glabellar region, are best interpreted as stem-hominoid-like supraorbital costae,7, 8 thus differing from the derived, African-ape-like condition of a continuous supraorbital torus with a prominent glabella.7, 8, 56 Second, the frontal sinus of H. laietanus is small and restricted, not penetrating into the glabella, frontal squama, or orbital rims, and is separated from the ethmoidal recess.7, 8, 48, 56 Even in extant taxa, it is difficult to determine whether the frontal sinus has an ethmoidal origin, but the H. laietanus morphology suggests, contrary to other authors' assessments,26, 35 that it lacks a true, African-ape-like ethmofrontal sinus. The restricted frontal sinus of Hispanopithecus might represent an intermediate condition toward the derived pongine condition. However, given the variable development of this structure in dryopithecines, the polarity of change is very unclear. Finally, Hispanopithecus displays a stepped subnasal fossa with the nasoalveolar clivus not overlapping the hard palate,7, 8 as in D. fontani.14 This condition is somewhat intermediate between the stem-hominoid condition of an unstepped nasal floor with a short premaxilla and the more derived, crown-hominid morphology, with a longer clivus with some degree of overlap.8, 72 However, given the apparently more derived condition displayed by the kenyapithecine Nacholapithecus, which has an elongated and procumbent premaxilla slightly overlapping the maxillary palatine process,61 interpreting the dryopithecine condition is far from unambiguous. Provisionally, hispanopithecins are here included in the Dryopithecinae, in recognition of their probable dryopithecin ancestry, as supported by its geographical distribution and the lack of clear-cut hominine synapomorphies.
APE EVOLUTION IN THE LIGHT OF THE VALLÈS-PENEDÈS HOMINOIDS
Early Hominoid Evolution in Africa and Hominid Origins
Hominoids, as conceived in this paper (Table 2), originated in Africa around the Oligocene/Miocene transition14, 25, 27 and experienced a considerable evolutionary radiation in Africa during the early and middle Miocene.25, 27 From the beginning of the middle Miocene, hominoids were also present in Eurasia,18, 25, 33, 34 where they diversified into a plethora of genera, most of them cladistically considered great apes.18, 25, 33, 34 The Eurasian diversity of Miocene hominids contrasts with the lack of unquestioned great apes in Africa before the appearance of the earliest hominins ca. 7-6 Ma.27 On this basis, several authors have favored a Eurasian origin of hominids and their early divergence during the middle Miocene into pongines and hominines, followed by a later hominine dispersal event back into Africa by the late Miocene.25, 26, 29, 31, 33, 35, 73 Besides proconsulids, the hominoid record from the middle and early late Miocene of Africa includes kenyapithecines (Nacholapithecus, Equatorius, and Kenyapithecus)61, 62, 64; the hominoids of uncertain affinities Otavipithecus and Samburupithecus27; and Chororapithecus and Nakalipithecus, which have been suggested by their discoverers to be related, respectively, to gorillas and Ouranopithecus.74, 75 Although the role of these taxa in hominid origins remains to be conclusively ascertained, they clearly indicate that there is no gap in the African hominoid record.60
Both cranially and postcranially, kenyapithecines (equatorins + kenyapithecins) appear, in most instances, to be less derived than are dryopithecines toward the crown-hominid condition. Kenyapithecus, however, shares one synapomorphy, a high zygomatic root,27, 64, 66 with dryopithecines and crown hominids, suggesting that at least kenyapithecins can be considered basal members of the Hominidae. Equatorines, in contrast, display a mosaic of primitive and derived features that is currently difficult to interpret. Some of their cranial features, such as the narrow nasal aperture (widest above midheight) and the low zygomatic root,61 are primitive relative to the derived condition shared by dryopithecines and crown hominids; on the other hand, the subnasal morphology of Nacholapithecus, unknown in other kenyapithecines, is more derived than that of dryopithecines toward the crown-hominid condition.61 On this basis, all kenyapithecines are here provisionally considered basal hominids. Other authors emphasize the primitive features of equatorins and include only kenyapithecins in the Hominidae.27 In fact, the kenyapithecine mosaic of primitive and derived features currently precludes deciphering whether the Nacholapithecus condition might indicate a closer relationship between equatorins and hominines or crown hominids, to the exclusion of dryopithecines.
With regard to the other above-mentioned African genera, the scarcity and fragmentary nature of their remains precludes an unambiguous phylogenetic assessment,27, 29 so that no unquestioned hominines are recorded during the middle and early late Miocene of Africa.27, 32 This fact could be merely an artifact due to insufficient sampling or taphonomic biases.76 In any case, hominoid dispersal and range extension events during this time interval are likely to have been multidirectional.35, 60 Given the instantaneous appearance of such events in geologic time, in light of current dating uncertainties and the fragmentary nature of the hominoid record, the place of origin of the Hominidae would remain unresolved even if the purported hominine status of some European dryopithecines or African late Miocene hominoids were more conclusively determined.
The Eurasian Hominoid Radiation
The oldest record of Eurasian hominoids corresponds to the latest early Miocene of Europe,18 consisting of a partial tooth from Engelswiess attributed to cf. Griphopithecus.77 This attribution has led some authors to favor the oldest possible age (>16 Ma) for the Turkish undoubted kenyapithecins Griphopithecus and Kenyapithecus,78, 79 and to argue that kenyapithecins dispersed into Europe before the Langhian transgression (ca. 16.3 Ma), with Kenyapithecus later dispersing back into Africa.33, 35, 79 On the basis of preserved morphology, however, the Engelswiess tooth can be attributed merely to a thick-enameled hominoid of uncertain affinities,18 whereas biostratigraphic data compatible with magnetostratigraphic evidence actually favor a younger chronology (ca. 14 Ma) for the Turkish localities, where undoubted kenyapithecins are recorded.18 Hominoid dispersals between Africa and Eurasia by the early middle Miocene cannot be further resolved18 except for concluding that kenyapithecins, previously restricted to Africa, expanded their range into Eastern Europe and Turkey around the Mid-Miocene Climatic Optimum. This warm phase peaked during the middle Miocene and was followed by gradual cooling during the remaining middle and late Miocene.80 The shared-derived craniodental features of kenyapithecins and Iberian dryopithecins, together with the more derived cranial and postcranial morphology of the latter, suggest that the Eurasian hominoid radiation might have originated from some kenyapithecin ancestor.15, 18
Like afropithecids, kenyapithecines had thick molar enamel and an associated dentognathic complex suitable for hard-object feeding, enabling them to exploit food resources previously unavailable to other catarrhines.81 This could have favored their survival in the more seasonal woodlands and forests that they would have encountered on their way from Africa into Eurasia and their subsequent diversification in the latter continent.49, 81 This hypothesis is supported by the retention of thick-enameled molars in the dryopithecins Pierolapithecus and Anoiapithecus and, to some extent, in most pongines,15, 49 even though thinner enamel appears to have independently evolved during the Miocene among several hominoid lineages, such as hispanopithecins.49
Further understanding of the initial Eurasian hominoid radiation is partly hampered by the lack of agreement about the phylogenetic relationships among dryopithecines, pongines, and hominines. Miocene hominoids from Asia have most often been regarded as pongines,25, 34, 82 with Sivapithecus representing their earliest record at ca. 13.0–12.5 Ma.82 A hominoid of uncertain affinities is recorded only slightly later (ca. 12.3–12.2 Ma) at the other end of Eurasia in the Vallès-Penedès Basin,18 being followed there by several dryopithecin genera at 11.9–11.8 Ma.13-15, 18 This suggests that the initial diversification of Eurasian hominoids took place somewhat earlier, probably between 14–13 Ma. However, with the exception of Ankarapithecus from the late Miocene of Turkey,33, 67 considered a stem pongine by most authors,15, 18, 25, 33, 56, 67 it has proven very difficult to place European fossil great apes into a coherent phylogenetic scheme. This is partly attributable to the incompleteness of the hominoid record, but also to the apparent high levels of homoplasy documented in hominoid evolution.40, 83, 84
The phylogenetic relationships of Eurasion hominoids notwithstanding, the progressive cooling and increased seasonality initiated during the middle Miocene apparently favored their adaptive radiation,18 as reflected by their widened geographic range, as well as increased taxonomic and ecological (dietary, locomotor) diversity. Such diversification likely resulted from the new selection pressures posed by changing ecological conditions toward more heterogeneous biotopes with increased seasonality.35 However, soon after the early/late Vallesian transition (ca. 9.6 Ma), hominoids are no longer recorded in Western and Central Europe, with the exception of Oreopithecus,71 which survived until ca. 7 Ma in the insular Tusco-Sardinian paleobioprovince.18 The extinction of European hominoids has been related to the vegetation changes that took place during the late Vallesian,85 which further resulted in the extinction of many other mammalian taxa (the Vallesian Crisis).22 The climatic trend towards cooling and increased seasonality initiated after the Mid-Miocene Climatic Optimum80 apparently led to the crossing of some paleoenvironmental threshold, most likely related to food unavailability,60 to which hominoids and several other mammals did not adapt.18
H. laietanus is the latest hominoid from mainland Western Europe,18 with Can Llobateres 1 (9.6 Ma) recording one of its latest occurrences, shortly before its last appearance datum at 9.5 Ma.18 The paleoenvironmental reconstruction of this locality is therefore highly significant for understanding the demise of Eurasian hominoids. In agreement with the mammalian fauna, which suggest humid and forested environments, the plant assemblage from CLL1 is indicative of marshy areas with abundant reeds and palms, with evergreen laurels and fig trees growing in the nearby plant communities.16, 24 Overall, this vegetation is consistent with a subtropical to warm-temperate climate, although evidence from other Vallès-Penedès sites indicates that such habitats would have been restricted to lowland humid areas, with warm-temperate mixed forests having a significant proportion of deciduous elements growing elsewhere.16, 24 Following climatic changes,80 the progressive disappearance of tropical and subtropical plants — providing a continuous supply of ripe fruit all year long — and the concomitant increase in dominance of deciduous trees, may have resulted in a lack of adequate trophic resources for Hispanopithecus during the unfavorable season, ultimately leading to its extinction.16 Such an explanation probably cannot be extrapolated to the eastern Mediterranean, where ouranopithecins survived until ca. 8.0-7.5 Ma.18 However, paleoenvironmental changes throughout the late Miocene ultimately led to the complete extinction of hominoids from Western Eurasia during the later Miocene (Fig. 3), as well as to their progressive decline and geographic range restriction in Asia throughout the Plio-Pleistocene.18
Dryopithecine Positional Behavior and Homoplasy in Hominoid Evolution
The view that homoplasy played a pervasive role in hominoid evolution is best exemplified by the locomotor apparatus.84 Orthogrady is a functional complex shared by crown hominoids, suitable for both vertical climbing and suspensory behaviors. As shown by Pierolapithecus and Hispanopithecus,9, 13, 32 at least some dryopithecines were characterized by an orthograde body plan, thus contrasting with the more primitive, pronograde body plan retained by African and Eurasian kenyapithecins.25, 27, 40, 86 Only the afropithecid Morotopithecus,39, 40 from the early Miocene of Africa, records orthograde-related features predating their appearance in the middle Miocene of Europe. An unambiguous interpretation of the former is complicated by the lack of comparable postcranial elements in Afropithecus,39 generally thought to display a pronograde body plan similar to that of proconsulids.40 However, given the afropithecid-like facial morphology of Morotopithecus, which lacks hominid synapomorphies,27 an independent development of orthogrady in this taxon seems the most likely interpretation.27, 40 In fact, given the hominid-like cranial morphology of Pierolapithecus, its mosaic of primitive and derived postcranial features further implies a significant degree of homoplasy in the locomotor apparatus, at least between hylobatids and crown hominids.13, 41, 42
Irrespective of whether Pierolapithecus is interpreted as having35, 44-46 or lacking41-43 suspensory adaptations, it clearly retained metacarpal and phalangeal features that were functionally related to powerful grasping and above-branch palmigrady.13, 41-43 Similarly, despite clear suspensory adaptations and the possession of a modern-hominoid-like elbow,9, 38, 40, 52, 86Hispanopithecus still retained, albeit to a lesser extent than Pierolapithecus, palmigrady-related manual features9, 11, 43, 47 and a well-developed grasping pollex unlike that of extant apes.87 A mosaic of primitive and derived postcranial features is even more clearly shown by the stem pongine Sivapithecus, which, like Pierolapithecus, lacked suspensory-related features and displayed adaptations to above-branch quadrupedalism,88 in spite of having a modern elbow configuration with a spool-shaped trochlea.38 Unless the great-ape-like cranial features of Pierolapithecus, Sivapithecus, and Hispanopithecus are homoplastic, their postcranial evidence indicates that the locomotor repertoire of the last crown hominoid ancestor was more primitive than inferred on the basis of extant taxa. Under such view, specific adaptations to suspensory behaviors would have repeatedly evolved in parallel from orthograde ancestors, at least between hylobatids and hominids.15, 41-43
Given the significant quadrupedal component inferred from the ACM Dryopithecus femur,14 and especially, the Sivapithecus postcranials,88 the possibility cannot be dismissed that orthogrady also might have independently evolved between great and lesser apes, or even between pongines and hominines.18 Deciphering whether orthogrady is homologous among crown hominoids, however, largely depends on whether hylobatids or kenyapithecins are more closely related to crown hominids. The second hypothesis is favored here on cranial grounds, although this conundrum cannot be unambiguously resolved, given the virtual lack of Miocene hylobatid remains.
Given their orthograde body plan, hylobatids have been hypothesized to be descended from great-ape-like ancestors through phyletic dwarfing.89 Alternatively, they might represent an independent off-shoot from small-bodied, putative stem hominoids from Africa, which several authors interpreted as stem catarrhines.27 In that case, at least afropithecids should be considered stem hominids. Such early branching of hylobatids remains to be demonstrated, although it is tentatively suggested by the dental similarities between African dendropithecids and the putative late Miocene hylobatid Yuanmoupithecus from China,90, 91 still largely unpublished. In light of pervasive homoplasy in hominoid evolution40, 83, 84, 92 and the convergent evolution of orthograde features between hylobatids and atelids due to similar adaptive requirements,84 an independent origin between hylobatids and crown hominids through parallelism seems even more likely, given their close phylogenetic relationships.84
CONCLUSIONS AND FUTURE PROSPECTS
In Eurasia, hominoids experienced an impressive adaptive radiation that includes dryopithecines and pongines. The fossil evidence from the Vallès-Penedès hominoids indicates that the European dryopithecines are great apes, probably evolved from the kenyapithecins that dispersed or extended their range from Africa into Eurasia by the middle Miocene. Whether hominines originated in Eurasia and later dispersed back into Africa cannot be currently resolved. Such a scenario, however, is not favored here, given the interpretation of dryopithecines as stem hominids and/or the sister taxon of pongines, instead of stem hominines. In fact, there is no strong evidence arguing against an independent evolution of dryopithecines + pongines in Eurasia, and of hominines in Africa.15, 18 This view makes sense from a paleobiogeographic viewpoint, with both groups having originated from similar kenyapithecin(e) taxa after the initial range extension of this group thoughout the Old World and subsequent vicariant events.18, 23, 92 An independent evolution of pongines in Eurasia and hominines in Africa is tentatively supported by the pongine-like cranial features of Pierolapithecus,48 the derived subnasal morphology of Nacholapithecus,61 and the presence of numerous late middle and early late Miocene hominoids in Africa.60 This hypothesis, of course, would require a higher degree of homoplasy between hominines and pongines than is generally accepted.18 Nevertheless, it cannot be dismissed a priori, given that homoplasy through parallelism, suspected to be pervasive in hominoid evolution,84 is a frequent byproduct of vicariance.92
Pervasive homoplasy, together with a fragmentary and biased fossil record, represent a serious difficulty for phylogenetic reconstruction. This can be imperfectly addressed only by applying a cladistic methodology to all available data; that is, by assuming morphological homology a priori. Most parsimonious cladistic hypothesis can be then evaluated a posteriori, based on independent sources of evidence, including chronology, paleobiogeography, and morphofunctional considerations.83 Chronostratigraphic information can also be successfully incorporated a priori (stratocladistics).65, 93 In the past, cladistic analyses based on a large data matrix of cranial and postcranial features have favored the hominine status of dryopithecines,31, 44, 65 although support for this hypothesis probably is too dependent on a particular selection and scoring of characters. Moreover, these analyses were based on a broad concept of Dryopithecus, including D. fontani as well as Hispanopithecus spp. New (strato)cladistic analyses, incorporating the recent hominoid finds from the Vallès-Penedès Basin and carefully re-evaluating character statements, will be required once detailed anatomical descriptions of all these taxa are published. Despite some recent contributions,11, 17, 38, 48, 58 most of this work is still under way. These analyses, hopefully coupled with new discoveries in the Vallès-Penedès Basin, may shed new light on the Eurasian hominoid radiation and its implications for understanding the origins and evolutionary history of the Hominidae.
Acknowledgements
This work has been supported by the Spanish Ministerio de Economía y Competitividad (CGL2011-28681 and RYC-2009-04533 to D.M.A.) and the Generalitat de Catalunya (2009 SGR 754 GRC). The author also thanks the National Geographic Society for funding field work at Can Llobateres; John Fleagle for the invitation to write this review; the several people that gave permission for reproducing illustrations; David R. Begun and four anonymous reviewers for helpful comments and suggestions on a previous version of this paper; and Sergio Almécija, Isaac Casanovas-Vilar and, most especially, Salvador Moyà-Solà, for discussion on hominoid evolution over the years.
