Available online at www.sciencedirect.

com

Journal of Human Evolution 54 (2008) 323e382

The anatomy of Dolichocebus gaimanensis, a stem platyrrhine monkey from Argentina

Richard F. Kay a,*, J.G. Fleagle b, T.R.T. Mitchell a, Matthew Colbert c, Tom Bown d, Dennis W. Powers e
a

Department of Biological Anthropology and Anatomy, Duke University, Durham, NC 27708-0383, USA b Department of Anatomical Sciences, Stony Brook University, Stony Brook, NY 11794, USA c Jackson School of Geological Sciences, University of Texas at Austin, Austin, TX 78712-0254, USA d Erathem-Vanir Geological, 2300 Arapahoe Avenue, No.236, Boulder, CO 80302, USA e 140 Hemley Road, Anthony, TX 79821, USA Received 11 January 2006; accepted 7 September 2007

Abstract Dolichocebus is known from the type skull encased in a concretion, numerous isolated teeth, parts of two mandibles, and a talus. The specimens come from the Trelew Member (early Miocene, Colhuehuapian South American Land Mammal Age) of the Sarmiento Formation near the village of Gaiman, Chubut Province, Argentina, dated to about 20 Ma. We describe all Dolichocebus fossil material using conventional surface anatomy and micro-CT data from the cranium. The new material and newly imaged internal anatomy of the skull demonstrate that anatomical characters hitherto supposed to support a phyletic link between Dolichocebus and either callitrichines (marmosets, tamarins, and Callimico) or Saimiri (squirrel monkeys) are either indeterminate or absent. To more fully explore the phyletic position of Dolichocebus, we undertook a comprehensive phylogenetic analysis. We examined 268 characters of the cranium and dentition of 16 living platyrrhine genera, some late Oligocene and early Miocene platyrrhines, Tarsius, some Eocene and Oligocene stem anthropoids, and several extant catarrhines. These analyses consistently indicate that Dolichocebus is a stem platyrrhine, as are late Oligocene Branisella and early Miocene Tremacebus, Soriacebus, and Carlocebus. Platyrrhine evolution often is conceived of as a single ancient adaptive radiation. Review of all available phyolgenetic data suggests a more layered evolutionary pattern, with several independent extinct clades lling modern platyrrhine niche space, and modern platyrrhine families and subfamilies appearing over a nine-million-year interval in the Miocene. The outcome of these analyses highlights the pervasiveness of homoplasy in dental and cranial characters. Homoplasy is a real evolutionary phenomenon that is present at all levels of biological analysis, from amino-acid sequences to aspects of adult bony morphology, behavior, and adaptation. 2007 Elsevier Ltd. All rights reserved.
Keywords: Argentina; Colhuehuapian; Dolichocebus; Gaiman; Miocene; Platyrrhine origins; Sarmiento Formation

Introduction It is now generally agreed that platyrrhines (New World monkeys) are most closely related to catarrhines (Old World monkeys and apes). Fossil evidence indicates that Africa is the most likely origin for the last common ancestor of these

* Corresponding author. Tel.: 1 919 684 2143; fax: 1 919 684 8034. E-mail address: richard.kay@duke.edu (R.F. Kay). 0047-2484/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jhevol.2007.09.002

taxa, suggesting that the ancestor of extant platyrrhines must have crossed the Atlantic Ocean (Hoffstetter, 1980; Tarling, 1980; Hartwig, 1994; Houle, 1999). The timing of this event is not well constrained. Molecular studies have dated the platyrrhine-catarrhine divergence to the middle Eocene (~43 Ma, Eizirik et al., 2004), 37.0 3.0 Ma, or 38.9 4.0 Ma (Poux et al., 2006), while the earliest catarrhines in the fossil record date to the late Eocene. Thus, this migration must have occurred sometime before the late Eocene (i.e., earlier than 37 Ma) (Kay et al., 2004c; Seiffert et al., 2004), or a stem

324

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

platyrrhine lineage could have lingered in Africa for some unknown period of time before arriving in South America. Early fossil primates from South America are uncommon and often poorly preserved. Branisella boliviana, the oldest South American primate, appears in the Deseadan fauna of Salla, Bolivia, in rocks now securely dated to about 26 Ma (latest Oligocene) (MacFadden et al., 1985; MacFadden, 1990; Kay et al., 1998b). Thus, there is a gap in the fossil record of platyrrhines of at least eight million years, when the platyrrhine stem group must have existeddeither in Africa or South America. Moreover, Branisella is known only from teeth and jaws (Takai et al., 2000; Kay et al., 2001), making phylogenetic interpretation tenuous. The succeeding Colhuehuapian Land Mammal Age of Argentina dates to the early Miocene (falling into the interval between about 21 and 18.5 Ma; (Marshall, 1985; Flynn and Swisher III, 1995; Kay et al., 1999b), and marks the appearance of at least four primate taxa, Tremacebus, Dolichocebus, Chilecebus, and an undescribed taxon from Gran Barranca, Chubut Province, Argentina (Kay et al., 1999b). Each taxon is from a different locality. Tremacebus is known by only a single deformed skull with canine and postcanine roots and broken molar crowns from Sacanana in Chubut Province, Argentina (Hershkovitz, 1974; Fleagle and Bown, 1983; Kay et al., 2004a; Kay et al., 2004b). A mandible with one premolar and one molar from Sacanana may belong to Tremacebus or possibly Soriacebus (Fleagle and Bown, 1983; Fleagle, 1990). Dolichocebus was originally described on the basis of a battered and distorted skull (Fig. 1) from Gaiman, Chubut Province, Argentina (MACN, Museo Nacional de Ciencias Naturales 14128), evidently collected by Professor Augusto L. Fistolera Mallie from the Trelew Member at Gaiman (Bordas, 1942; Kraglievich, 1951; Rosenberger, 1979a; Hershkovitz, 1982; Rosenberger, 1982; Fleagle and Rosenberger, 1983). From Gran Barranca, Argentina, comes a mandibular specimen of Colhuehuapian age that has been previously allocated to Homunculus (Hershkovitz, 1981) or Soriacebus (Fleagle, 1990). However, additional material of this monkey suggests that it is a new taxon (Kay et al., 1999b). Finally, from Chile comes a platyrrhine skull of Colhuehuapian age that is the only specimen of Chilecebus carrascoensis (Flynn et al., 1995). The maxillary teeth of this skull have been described briey, but the skull has not been described, only illustrated (Fleagle and Tejedor, 2002). Since its initial description by Bordas (1942), the phylogenetic position of Dolichocebus has been the subject of much controversy, partly due to the broken nature of the remains. Kraglievich (1951) assigned Dolichocebus to Callitrichinae based on his interpretation that only two upper molars were present.1 This view has not been accepted by later workers,
1 In the older literature on platyrrhine systematics, two groups are recognizeddCallitrichidae for platyrrhines with two molars and Cebidae for the rest. The classication scheme followed here recognizes three family-level taxa: (1) Atelidae for Alouatta, Ateles, Brachyteles, and Lagothrix; (2) Pitheciidae for Callicebus, Pithecia, Cacajao, and Chiropotes; and (3) Cebidae for Saimiri, Cebus, Aotus, and the Callitrichinae. See Fig. 6 for supporting data.

Fig. 1. Lateral (left), dorsal, ventral, and coronal views of the cranium of MACN 14128, type specimen of Dolichocebus gaimanensis.

beginning with Hershkovitz (1970, 1974), who, based on further cleaning of the specimen, concluded that Dolichocebus had three upper molars (the third lost postmortem) and placed it with Homunculus in the family Homunculidae Ameghino, 1894 based partly on the perceived overall primitiveness of both taxa compared to living platyrrhines with three molars (his Cebidae). Rosenberger (1979a: 412e416) ratied Hershkovitzs interpretation of the dental formula but pointed to many ways in which Dolichocebus and Homunculus differ. Rosenberger commissioned further preparation of the type skull and concluded that its bony interorbital septum was perforated in life by an interorbital fenestra, a derived feature of extant squirrel monkeys (Saimiri). Principally for that reason, he placed Dolichocebus with Saimiri (and Neosaimiri Laventiana, middle Miocene, Colombia) in the tribe Saimiriini, allied with Cebus in the subfamily Cebinae.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

325

However, this interpretation has been open to question because there is not universal agreement as to whether the bony interorbital fenestra is anatomical or the result of postmortem damage (the specimen was distorted and broken in other ways before and after burial) (Hershkovitz, 1982; Rosenberger, 1982; Fleagle and Rosenberger, 1983; Ford, 1986; Horovitz, 1999). Between 1983 and 1989, joint Argentine/American expeditions collected a number of isolated primate teeth and a talus from Gaiman, the type locality of Dolichocebus. The talus and some of the teeth have already been described (Fleagle and Bown, 1983; Gebo and Simons, 1987). Fleagle and Bown (1983) identied several features of isolated upper molars from Gaiman that resemble the extant platyrrhine Saimiri but also Oligocene African anthropoids Aegyptopithecus and Propliopithecus. They identied these as shared primitive features. Gebo and Simons (1987) called attention to general similarities between the Dolichocebus talus and that of generalized New World monkeys but found no certain shared derived features indicating a relationship with any platyrrhine lineage. In the most recent comprehensive phylogenetic survey of living and fossil platyrrhines, Horovitz (1999) did not consider resolvable the arguments about interorbital fenestration in Dolichocebus. In her character-taxon matrix, Horovitz scored Dolichocebus for only ve cranial characters and none pertaining to the talus. In her phylogenetic analysis using these ve cranial and 22 dental characters, she found that Dolichocebus was a stem callitrichine with three molars (see footnote 1 for information about the taxonomic terminology used here). In summary, there remains little agreement about the place of Dolichocebus within the platyrrhine clade, with the three currently held views being that (1) it is a basal callitrichine (Horovitz, 1999), (2) it is specially related to Saimiri (Rosenberger, 1979a), and (3) it is a sister group of all living platyrrhines (Fleagle and Kay, 1989) (Fig. 2). As already noted, following Fleagle and Bowns (1983) contribution, additional primate material has come to light at Gaiman. Expeditions from Stony Brook University in collaboration with a group from MACN, Buenos Aires, collected isolated teeth, an edentulous mandible, and a talus in the 1980s and early 1990s. Later, Dr. Guiomar Vucetich of Mueso de La Plata found two mandibles. Taken together, all parts of the permanent dentition are now represented. In addition, the type skull was recently CT-scanned in an effort to clarify the controversial question of whether there was an interorbital fenestra and to examine other hitherto inaccessible features of the skull. The purpose of the present contribution is threefold: (1) to summarize the available evidence concerning the stratigraphy and geologic age of Colhuehuapian fossils at Gaiman; (2) to describe and illustrate the dental specimens, as well as to offer new details about the skull based on high-resolution CT scans; and (3) to reconsider the phylogenetic position of Dolichocebus and other late Oligocene and early Mocene platyrrhines. With respect to this last goal, we note that there coexist in the current literature two radically different hypotheses about platyrrhine cladgenesis. One hypothesis, which we will call the deep-time hypothesis, is that living platyrrhines arose

Marmosets & Tamarins Saimiri Cebus Aotus Pitheciidae Atelidae Dolichocebus

Marmosets & Tamarins Dolichocebus Saimiri Cebus Aotus Pitheciidae Atelidae

Marmosets & Tamarins Saimiri Dolichocebus Cebus Aotus Pitheciidae Atelidae

Fig. 2. Contrasting views of the afnities of Dolichocebus relative to living platyrrhines. (A) Dolichocebus is a stem platyrrhine (Hershkovitz, 1982; Fleagle and Bown, 1983). (B) Dolichocebus is a relative of marmosets and tamarins (Horovitz, 1999). (C) Dolichocebus is the sister taxon of Saimiri (Rosenberger, 1979a).

as single adaptive radiation of great antiquity. Soon after they arose, platyrrhines entered adaptive niches that are similar to those occupied by todays platyrrhine niches. The occupants of those niches are the ancestors of the modern clades on South American monkeys. This view was foreshadowed by the work of George G. Simpson and Bryan Patterson (Patterson and Pascual, 1972; Simpson, 1980). Its current proponents place all the known late Oligocene-to-Recent South American fossil monkeys within the extant platyrrhine familiesdCebidae, Atelidae, and Pitheciidae (Rosenberger, 1979b, 1980, 1988; Rosenberger et al., 1990; Rosenberger, 1992, 2000;

326

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Rosenberger, 2002; Tejedor et al., 2006). A second hypothesis, which we call the layered hypothesis, envisions a more complex evolution of platyrrhines and their adaptive niches. Earliest known platyrrhines are stem taxa. Some of these taxa followed adaptive strategies similar to those alive today; others entered different niches not represented today. There was no single cladogenetic event in platyrrhine evolution. Living family clades diverged at around 20 Ma. They coexisted through the early and middle Miocene with stem taxa (Kay, 1990; Fleagle et al., 1997; Kay et al., 1998a). Our phylogenetic analysis of Dolichocebus and some late Oligocene and early Miocene platyrrhines revisits this debate.

Geology The geologic setting of the Gaiman fossils was reviewed in Fleagle and Bown (1983), to which we add some information originally mentioned in an unpublished manuscript by two of the authors of this paper (TB and DWP). The principal fossil localities with primates are situated in the Province of Chubut, Argentina, and the Dolichocebus remains come from the southern barranca along the lower Ro Chubut Valley near Gaiman (Fig. 3). Dolichocebus fossils come from the Trelew Member of the Sarmiento Formation. Roth (1899) rst reported fossils from this area. Simpson (1935) reviewed the

Sacanana Gan Gan

Gaiman

Trelew

Ro Chubut

Lago Musters

Lago Colhu Huap

Gran Barranca

50 100 Comodoro 0 Rivadavia KILOMETERS

Ro

Ch

ico

10

Trelew
Dolavon
Ro Chubut

Gaiman

Rawson
Bryn Gwyn

Angostura

Drofa Dulog

Fig. 3. Top, map of Chubut Province, Argentina, showing the geographic location of the Colhuehuapian-aged fossil-primate-bearing localities at Trelew/Gaiman, Sacanana, and Gran Barranca. Bottom, map of the lower valley of the Ro Chubut, showing barrancas (bluffs; in stipple) north and south of the river valley, pampas (lined), and the location of the Sarmiento section (aea0 ) with respect to towns and villages.

Ba

hi

a'

En

ga

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

327

geology and assigned this fauna to what is now conventionally referred to as the Colhuehuapian Land Mammal Age (see also Bordas, 1939). The primate teeth, mandibles, and talus were collected in the Trelew Member of the Sarmiento Formation just above the Bryn Gwyn Member, which forms a distinct topographic bench in the middle of the barranca (Fig. 4). Figure 5 gives a stratigraphic prole of the fossil occurrences at Gaiman. The precise stratigraphic level of the type skull of Dolichocebus is not known, but it almost certainly came from the Trelew Member at Gaiman because fossil land mammals are known from no other part of the south baranca in this region (personal experience and Bordas, 1939). The Trelew Member contains a mammalian assemblage typical of the Colhuehuapian South American Land Mammal Age (Pascual and Odreman Rivas, 1971). This similarity implies that the Trelew fossils are roughly similar in age to Colhuehuapian fossil assemblages reported elsewhere in Patagonia, notably at Sacanana, where Tremacebus harringtoni occurs, and Gran Barranca, where another early primate is found (Kay et al., 1999b). The reported absolute ages of these faunas are somewhat conicting. Bown et al. (1988) reported a ssion-track age of 15.8 2.5 Ma at 20 m above the fossil level in the Trelew Member of the Sarmiento Formation at Gaiman that contains Colhuehuapian mammals. Kay et al. (1999a,b) mentioned a series of 40Ar/39Ar dates on glass and plagioclase separates from samples at Gran Barranca in association with Colhuehuapian faunas. In the context of an associated magnetic-polarity prole, they suggested that the bulk of the Gran Barranca Colhuehuapian fauna falls within Chron 6r and 6n (19.1e20.5 Ma). Flynn et al. (1995) reported an 40 Ar/39Ar date for another apparently Colhuehuapian fauna from Chile of 20.09 0.27 Ma where Chilecebus is found, in agreement with the dates of Kay et al. (1999a). A second independent source of information about the age of the Colhuehuapian at Gaiman derives from the fact that a marine bed in the Gaiman Formation containing sharks,

whales, and penguins overlies the Colhuehuapian levels at Gaiman. These marine faunas are of Aquitanian age (Cozzuol et al., 1991; A.L. Cione, unpubl. data). If so, then the underlying Trelew Member (and its Colhuehuapian faunas) at Gaiman must be older than 20.5 Ma, because the Aquitanian interval is dated to 20.5e23.8 Ma (Berggren et al., 1995). This date agrees with the published dates for Gran Barranca but conicts with Bown et al.s (1988) suggestion of a younger age. Bown et al.s reported date, if correct, suggests that the Trelew Member is, at least in part, younger than the Santacrucian SALMA faunas, which are older than 16 Ma (Fleagle et al., 1995). This is exceedingly unlikely, and therefore we accept the preponderance of evidence that Dolichocebus and other faunas of Colhuehuapian age are ~20 Ma.

Materials and methods The type specimen of Dolichocebus gaimanensis is a cranium, the subject of our CT-imaging analysis, as described below. All remains are summarized in Table 1 and a list of dental dimensions is provided in Tables 2 and 3. Allocation of unassociated specimens to a particular taxon is always problematic. Our reasons for assigning all of the new specimens to Dolichocebus are as follows: (1) There is no evidence that more than one taxon is present among the isolated teeth available. The dental remains compose a harmonious assemblage in terms of size and morphology. (2) The isolated upper teeth are of appropriate size to go with the palate and roots of the type skull (Fleagle, 1985). (3) Dolichocebus is the only primate taxon denitely known to occur at Gaiman. The only other known Colhuehuapian Patagonian fossil primates to which these teeth can be compared are Tremacebus and the new monkey from the Colhuehuapian of Gran Barranca. The morphology of the Gaiman teeth, as described below, denitively rules out allocation to the new Gran Barranca monkey or, for that matter, to Santacrucian late early Miocene Soriacebus, Carlocebus, or Homunculus. Possible allocation to Tremacebus deserves more consideration. Both of these taxa come from the Colhuehuapian, and Sacanana and Gaiman are less than 300 km apart. Available material of Tremacebus consists only of the type cranium with most of the teeth broken off (only parts of M1e3 crowns are preserved) (see Fig. 1 in Kay et al., 2004a). The upper molars from Gaiman differ in some respects from those of Tremacebus. Although both have strong lingual cingula with welldeveloped M1e2 hypocones, those of Tremacebus are much smaller and more rounded in lingual occlusal prole than those of Dolichocebus. A mandibular fragment from Sacanana with a broken part of P4 and M1 (Fleagle, 1990) may not belong with Tremacebus, but in any event, it is quite unlike the M1 from Gaiman. Based on the foregoing, we conclude that the isolated teeth from Gaiman belong with the skull.

Fig. 4. Fossil deposits of the Sarmiento Formation, south of the Chubut River near Gaiman, Chubut Province, Argentina. The fossils come from the Trelew Member of the Sarmiento Formation, just above the bench in the middle of the gure.

328

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Gaiman Formation (marine) conformable contact Sandstone tuff Sarmiento Formation COLHUEHUAPIAN
meters 10

Terlew Member

erosional unconformity Bryn Gwyn SF Member ? erosional unconformity

Tuffaceous sandstone Tuffaceous ss. with large-scale trough cross-strata Tuffaceous ss. & sandy mudstone with epsilon cross-strata Bedded calcrete nodules Bedded & unbedded calcium carbonate concretions Calcium carbonate veinlets

Pan d'Azucar Member

Sarmiento Formation

CASAMAYORAN

Tuffaceous muddy ss.

Bedded & unbedded silicate nodules Irregular masses of opaline silicates Chalcedony geodes

Tuffaceous sandy mudstone

Tuffaceous mudstone

conformable contact

Ro Chico Fm.
Fig. 5. Idealized stratigraphic prole at Gaiman showing the Trelew Member, the principal fossil level containing a Colhuehuapian assemblage. Note the position of the ash level referred to in the text.

To determine more precisely the limits of bone and matrix and to better appreciate the structural details of the interior of the type skull of Dolichocebus, the skull was scanned at the High-Resolution X-ray Computed Tomography (CT) Facility at the University of Texas at Austin, as described by Ketcham and Carlson (2001). X-ray energies were set to 150 kV and 0.16 mA using a FeinFocus X-ray source. X-ray intensities were measured using an Image Intensier detector employing a 1024 1024 video camera. Each slice was acquired using 1000 views (angular orientations), with four samples taken per view. The specimen, mounted in a plastic cylinder, was scanned with a centered axis of rotation and a source-to-object distance of 135 mm (Ketcham and Carlson, 2001). Slice thickness and interslice spacing were 0.0466 mm (one video line). The image eld was reconstructed to 43.5 mm, based on a maximum eld of view of 44.164 mm, yielding an interpixel

spacing of 0.042 mm. Reconstruction parameters were calibrated to maximize usage of the 16-bit range of grayscales available in the output images. Twenty-seven slices were acquired for each rotation of the turntable, with a resulting acquisition time of about 11.1 seconds per slice. The data consist of 1430 slices, from the front to the back of the skull. The coronal slice-by-slice animation, representing the original CT data, may be viewed on the DigiMorph Web site, http://www.digimorph.org/index.phtml. The coronal movie begins at the tip of the rostrum; the slices are on the coronal plane. These original data were digitally resliced to provide slice-by-slice animations on the horizontal and sagittal planes. The horizontal animation starts dorsally and passes ventrally; the slices are oriented in dorsal view. The sagittal animation proceeds from left to right through the specimen; the slices are in left lateral view.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Table 1 Dolichocebus gaimanensis specimens Museum number MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN MACN 14128 CH 356 CH 362 CH 357 CH 358 CH 359 CH 361 CH 864 CH 865 CH 866 CH 868a CH 868b CH 870 CH 871 CH 872 CH 873 CH 876 CH 878 CH 896 CH 897a 2CH 897b CH 898 CH 1011 CH 1012 CH 1300 Specimen Cranium (type specimen) Left M1 (Figs. 6, 7a in Fleagle and Bown, 1983) Talus (Fig. 4 in Gebo and Simons, 1987) Right M3 (Fig. 7c in Fleagle and Bown, 1983) Lingual part of left upper molar (Fig. 7b in Fleagle and Bown, 1983) Left I1 (Fig. 8 in Fleagle and Bown, 1983) Right canine (Fig. 8 in Fleagle and Bown, 1983) Right P2 Right P2 Left M2 Right P3 Right dp3 Right lower canine Left lower canine Right I1 or I2 Left P3 Left M1 Right P4 Left M1 or M2 (crown eroded, enamel mostly gone) Upper molar fragment Left I2 (?) Left P4 Left dp4 Left I2 Mandible fragment with root socket for I2, roots for canine, and P2e3 Left canine Right canine Mandible fragment with left M1e2 Mandible fragment with right M1e3

329

MACN CH 1302 MACN CH 1303 MPEF 5146 MPEF 5147

All specimens come from the Trelew Member of the Sarmiento Formation at Gaiman, Chubut Province, Argentina.

The platyrrhine skulls that constitute the comparative database for cranial anatomical data were scanned with various protocols, depending on skull size, in order to maximize resolution for each specimen. Details of the CT sample and the resolution of the images are summarized in Table 4. All dental measurements were made using a Wild M-5 binocular microscope using 10 oculars and set at 12 or 25, with a calibrated reticle. Anatomical landmarks for the measurements are summarized in Kay (1977). Phylogenetic analysis A series of analyses was undertaken to explore the phylogenetic placement of Dolichocebus and other late Oligocene and early Miocene platyrrhines. In addition to Dolichocebus, we included data from 16 genera of extant platyrrhines, the late Oligocene Branisella from Salla, Bolivia, the Colhuehuapian (early Miocene) platyrrhine Tremacebus, and the earliest Santacrucian taxa Soriacebus and Carlocebus. The sole known Patagonian middle Miocene primate Proteropithecia, was also included. Several taxa are excluded because some or all the pertinent material is undescribed. These include Chilecebus (Colhuehuapian, Chile) (Flynn et al., 1995), Homunculus from the

Santacrucian (latest early Miocene, Santa Cruz Province)2, and an unnamed taxon from the Colhuehuapian of Gran Barranca, Chubut Province, Argentina (Kay et al., 1999b). As outgroups, we included three extant catarrhines (Miopithecus talapoin, Presbytis melalophos, and Hylobates lar), ve of the better-known taxa of Eocene/Oligocene African anthropoids (Aegyptopithecus, Proteopithecus sylviae, Apidium phiomense, Simonsius [ Parapithecus] grangeri, and Catopithecus browni), and the extant haplorhine Tarsius. An analysis of the possible relationship of platyrrhines to earlier anthropoids of Africa would require a more exhaustive sampling of the African taxa. For example, inclusion of other lesser known but older and perhaps more primitive taxa (e.g., Qatrania, Abuqatrania, Propliopithecus, Oligopithecus) has the potential to change the branching pattern among and between outgroups and platyrrhines. We used the parsimony criterion and utilized the computer program PAUP version 4.0b10 (Swofford, 2002). The list of dental and cranial traits comes mostly from earlier studies (Kay et al., 2004c). However, we have added several cranial characters and revised some cranial and dental characters. We evaluated 268 characters (85 cranial and 183 dental). Among the platyrrhines used in this study, we identied 39 characters that exhibit no variability among the anthropoid taxa studied here (but which are included in the character-taxon matrix so as to enable future comparisons with nonanthropoid taxa in which they do vary). Thirty variable characters in the data set are parsimony-uninformative because they are autapomorphies of terminal taxa. This leaves 199 parsimony-informative characters upon which to base our phylogenetic observations. A list of all cranial and dental characters and their states is given in Appendix 1. The character-taxon matrix is presented in Appendix 2. A list of specimens examined for the phylogenetic analysis is provided in Appendix 3. Missing data can negatively impact the outcome of an analysis by affecting resolution, making determination of polarity uncertain, and underrepresenting homoplasy (Nixon and Davis, 1991; Simmons, 1993; Kay and Williams, 1994). Data for the extant taxa are complete. (However, some characters cannot be scored. For example, the position of the M1 paraconid relative to protoconid and metaconid is scored as missing in any taxon where a paraconid is absent.) Data for the extinct taxa are incomplete. A list of the percentage completeness of all taxa is given in Appendix 4. We considered including postcranial characters based on the work of Ford, Dagosto, Gebo, and others (Ford, 1986;
2 Recently the rostrum and face of a monkey from the locality of Killik Aike Norte, Ro Gallegos, Santa Cruz Province, was described as Killikaike blakei (Tejedor et al., 2006). The diagnosis of this new genus includes comparisons to extant platyrrhines, but not with abundant comparative material of Homunculus patagonicus. Homunculus material comes from the same formation and same approximate stratigraphic levels as Killikaike (Tauber et al., 2004; Tejedor et al., 2006). Most of the distinctions drawn between Killikaike and extant platyrrhines are the same as those between Homunculus and extant platyrrhines (e.g., Tauber, 1991), and the published linear dimensions and anatomy are very similar between Homunculus and Killikaike (Tauber, 1991; Kay et al., 2005). Provisionally, we regard Killikaike blakei as the same genus and perhaps the same species as Homunculus patagonicus.

330

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Table 2 Dimensions (mm) of the upper teeth of Dolichocebus gaimanensis MACN CH 356 I1 MD BL C MD BL Height P2 MD BL P3 MD BL P4 MD BL dp4 MD BL M1 MD BL M2 MD BL M3 MD BL 2.75 4.73 3.32 4.48 4.03 5.43 4.22 5.50 2.88 4.41 2.56 4.09 2.62 2.81 3.96 3.75 7.03 3.20 2.94 4.35 3.45 2.94 4.16 2.98 2.19 MACN CH 357 MACN CH 359 MACN CH 361 MACN CH 864 MACN CH 876 MACN CH 878 MACN CH 1011 MACN CH 1012 MACN CH 1302 MACN CH 1303

Table 3 Dimensions (mm) of the lower teeth of Dolichocebus gaimanensis MACN CH 865 I1 MD BL C MD BL P2 MD BL P3 MD BL dp3 MD BL P4 MD BL M1 MD BL (tal) BL (trig) M2 MD BL (tal) BL (trig) 2.75 3.07 3.20 2.88 2.88 3.52 3.91 3.45 3.46 4.03 3.32 3.20 3.90 3.30 3.20 2.75 3.13 2.88 2.75 3.77 3.00 3.32 2.62 1.66 2.24 MACN CH 866 MACN CH 868a MACN CH 868b MACN CH 870 MACN CH 871 MACN CH 872 MACN CH 873 MACN CH 898 MPEF 5146 MPEF 5147

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Table 4 Skulls of living platyrrhines examined by CT-imaging and used in the comparative sample Taxon Specimen Field of reconstruction (mm) 104 52 93 66 42 36 35 36 28 79 36 37.5 49 65 Coronal interslice spacing (mm) 0.11104 0.05708 0.1017 0.0723 0.04635 0.03778 0.03378 0.04077 0.03194 0.08683 0.03778 0.0412 0.0551 0.063477

331

Interslice spacing for coronal and horizontal reslicing (mm) 0.10156 0.0508 0.0908 0.0645 0.0410 d d d d 0.07715 d 0.03662 0.0479 d

Alouatta guariba female Aotus lemurinus Ateles geoffroyi female Callicebus torquatus Callimico goeldii Callithrix (Mico) argentata female Callithrix jacchus jacchus Callithrix jacchus female Cebuella pygmaea male Cebus olivaceus Saguinus fuscicollis Saguinus fuscicollis Saimiri sciureus sciureus Pithecia sp.

USNM 518255 USNM 464844 USNM 291056 USNM 406411 USNM 303323 USNM 239463 USNM 503885 USNM 503895 USNM 337324 USNM 338960 USNM 518577 Kay collection 142 NMNM 518538 Fleagle collection (Suriname?)

Meldrum, 1993; Dagosto and Gebo, 1994; Ford, 1994; Gebo et al., 1994) but ruled it out from the present analysis because none of the taxa of late Oligocene/early Miocene platyrrhines evaluated here are known from very much postcranial material. Just a few isolated specimens are described for Carlocebus or Soriacebus (allocations uncertain) and only the talus of Dolichocebus is represented. No postcranial elements of Branisella and Tremacebus are known. In our PAUP analyses, a number of multistate characters were used (indicated in Appendix 1). There is disagreement about whether this sort of character should be ordered or unordered (Mickevich, 1982; Mabee, 1989; Mickevich and Weller, 1990; Hauser and Presch, 1991; Slowinski, 1993; Mabee, 2000). Multistate characters should be designated as ordered if changes from one state to another require passing through intermediate states also represented in the data set (e.g., to go from absent tolarge one must pass through the state small) (Slowinski, 1993). Ordering characters excludes the possibility that characters can change from one end of a morphocline to the other without an intermediate state (Hauser and Presch, 1991). To evaluate the effect of character ordering on tree topologies derived from our data set, we ran our analyses with (1) all multistate characters unordered and (2) some multistate characters ordered. Aweighting scheme was used in which some multistate characters were ordered and their weight scaled. By using scaled weighting, ordered multistate characters are set to have the same weight regardless of the number of characters states. The total breadth of each morphocline is set to a base weight of 100. For a two-state character, it takes one step to cross the morphocline (0 to 1 or 1 to 0) and each step is assigned a weight of 100. For an ordered three-state character it takes two steps to cross the morphocline (0 to 1, and 1 to 2, or the reverse), so each step is assigned a value of 50. This weighting scheme eliminates the situation where ordered multistate characters could have more weight (because they use more steps) and thus differentially inuence tree topologies. Following the recommendation of Springer et al. (2001), we established a molecular scaffold upon which to

superimpose the character distributions using the Constraints Backbone option of PAUP. Under the backbone constraint, extinct taxa are unconstrained and can move about on the molecular phylogenetic scaffold. Springer et al. argued that clades established by maximum-parsimony analysis of molecular data should be assumed to depict a clade accurately if they receive !90% bootstrap support. In this case, molecular-sequence data establishes the phylogenetic relationships among many extant platyrrhines to a high degree of probability (Fig. 6). This branching pattern has been supplemented and veried in several cases with Alu data (see Fig. 6 caption for details). For completeness, we also examined the consequences of unconstrained analyses. The distinction between polymorphic and uncertain character states was enforced. The tree-bisection-reconnection (TBR) branch-swapping algorithms of PAUP were selected. (Other algorithms were tried but never yielded shorter trees). For each set of comparisons, starting trees were obtained via stepwise addition with a random-addition sequence with one tree held at each step. The search process was replicated 1000 times. Abbreviations Specimens come from the following museums: ANS, Academy of Natural Sciences, Philadelphia; CGM, Cairo Government Museum, Cairo; DPC, Duke University Lemur Center fossil collections; FMNH, Field Museum of Natural History, mammal collections; MACN, Museo Nacional de Ciencias Naturales; MNHN Bol V, Museo Nacional de Historia Natural, La Paz, Bolivia; MSP, Museu de Zoologia da Universidade de Sao Paulo, Brazil ; YPM, Yale Peabody Museum paleontology collections; CORD-PZ, Museo de Paleon tologa, Facultad de Ciencias Exactas, Fsicas y Naturales de la Universidad Nacional de Cordoba; Tremacebus skull, Rusconi Collection at Museo de Fundacion Miguel Lillo, Tu cuman, Argentina; MPEF, Museo Paleontologico E. Feruglio de Trelew, Chubut Province, Argentina; and USNM, Smithsonian Institution, National Museum of Natural History, vertebrate zoology collections.

332

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Tarsius Hylobates Miopithecus talapoin Presbytis melalophos Callicebus Pithecia Cacajao Chiropotes Saimiri Cebus Aotus Saguinus Leontopithecus Callimico Callithrix Cebuella Alouatta Ateles Brachyteles Lagothrix
Fig. 6. Cladogram of extant platyrrhine primates based on molecular-sequence and Alu data. Molecular-sequence data provide powerful evidence for resolving most of the nodes within the extant platyrrhine clade; Alu elementsdshort interspersed nuclear elements (SINEs)dprovide additional evidence for resolving platyrrhine cladogenesis because their mode of evolution is predominantly unidirectional and homoplasy-free (Hillis, 1999; Bashir et al., 2005). Molecular-sequence data strongly support the monophyly of the Platyrrhini and recognition of three clades within itdAtelidae, Cebidae, and Pitheciidae (Harada et al., 1995; Schneider et al., 1996; Barroso et al., 1997; Schneider et al., 2001; Singer et al., 2002). The rooting of Callimico with the Callithrix/Cebuella clade is very strongly supported by sequence data despite an overall lack of morphological support (Hill, 1957). Eighty-seven Alu elements support platyrrhine monophyly (Singer et al., 2002; Ray et al., 2005). The tree gured here is resolved at those nodes where bootstrap support of molecular-sequence data is equal to or greater than 90% and/or when one or more Alu supports it. Alu data indicate that Atelidae and Cebidae are sister taxa to the exclusion of Pitheciidae (Ray et al., 2005). Linkage of callitrichines with Aotus, and a Saimiri/Cebus group is strongly supported by sequence data and three Alus. Ray et al. (2005) reported that one Alu links cebines with Aotus (also supported by recent data from the laboratory of T. Disotell, pers. comm.). That Callicebus is a basal pitheciine is strong supported by many sequence studies and three Alus. Aotus consistently is excluded from the Callicebus-pitheciine clade (contra Rosenberger, 1981). Atelidae has strong molecular support (including six Alus), with Alouatta as the sister to a clade consisting of Ateles, Brachyteles, and Lagothrix. Meireles et al. (1999a, b) reported a fully resolved tree for extant Atelidae using a combination of g-globin, e-globin, RBP, G6PD nuclear genomic sequences, and mitochondrial COII sequences.

is preserved bilaterally between the foramen magnum and the anteriormost extent of the petrosals. The area between the anterior apex of the petrosal and the posterior extent of the M2s, including the entire pterygoid region, is poorly preserved or missing altogether. A part of the caudal margin of the palatal process of the palatine appears to be intact. The incisors and premaxilla are not preserved. The maxilla preserves the roots of the canine through M2 bilaterally, but the crowns have been lost. The zygomatic arches are broken away and lost. Although the face has also been badly distorted and damaged, the interorbital and orbital regions are preserved in part. Skull length (prosthioneinion) is estimated to be approximately 68.5 mm. Thus, the skull length is similar to that of extant Callicebus, the titi monkey (Hershkovitz, 1990). An exhaustive summary of anatomical comparisons between Dolichocebus, extant platyrrhines, Tarsius, representative basal anthropoids, and crown catarrhines is embodied in the list of characters and their states, and the character-taxon matrix (Appendices 1 and 2, respectively). The Nexus-formatted

A
Matrix

Bone

Lacrimal crest

Interorbital hole Lacrimal crest

Results Descriptive anatomy of the cranium CT-imaging of the type skull clearly distinguishes fossil bone from matrix (Fig. 7A, B). The cranium is cemented within a soil-derived concretion, the parent material of which was volcanic ash. While its overall integrity was maintained during diagenesis and subsequent exposure on the surface, the neurocranium was compressed bilaterally, badly crushed in some areas, and some parts were decayed or broken off before and after fossilization. The rostal part of the basicranium
postorbital flange of the zygomatic
Fig. 7. (A) Left anterolateral CT reconstruction of Dolichocebus gaimanensis showing the clear separation of gray values between fossil bone and matrix. Note especially the good preservation of the ventral margin of the left orbit but absence of bone on its lateral margin and over glabella. (B) Right anterolateral CT reconstruction of the same specimen showing better preservation of the right lateral margin of the orbit, including the postorbital ange of the zygomatic, but poor preservation of the ventral orbital margin. Bone is absent from the margins of an interorbital hole on the right and left sides. The anterior crest of the lacrimal is noted on the right and left sides.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

333

character-taxon matrix is available upon request from the senior author. Some phylogenetically or adaptively signicant comparisons are elaborated in the text. Where mentioned in the text, comparisons with Tremacebus harringtoni are based on the type specimen (for more details see Kay et al., 2004a). Except as noted, cranial observations of Homunculus are based on personal examination of the Cordoba skull CORD-PZ 1130, collected by A. Tauber (Tauber, 1991). Orbital and interorbital region. The medial and ventral margin and a small part of the dorsomedial wall of the left orbit are preserved (Fig. 7). Parts of the left postorbital septum also remain. A larger portion of the right orbit has been preserved, although the right ventral orbital rim is missing. Towards its apex, the right orbit is preserved dorsally but the optic canal and superior orbital ssure are not present (Fig. 7). From anatomical landmarks preserved on the left side (ventrally) and right side (superiorly, laterally, and medially), the orbit diameter is estimated to be about 14.5 mm in a dorsoventral plane and about 14.2 mm mediolaterally. Relative to prosthioneinion length, these orbital dimensions are comparable to those of similarly sized platyrrhines, such as Callicebus moloch. Aotus trivirgatus stands apart in having very large orbits (data in Kay and Kirk, 2000). Relative orbit size in early Miocene Homunculus resembles that of most extant platyrrhines (Tauber, 1991). The orbits of Tremacebus are somewhat larger than those of most platyrrhines but not nearly as large as in extant Aotus (Hershkovitz, 1974; Fleagle and Rosenberger, 1983; Martin, 1990; Kay et al., 2004a; Kay et al., 2004b). On supercial inspection, the frontal bone appears to be depressed at glabella above weak brow ridges. However, the CT images reveal this region to be completely devoid of bonedwhat appears to be glabella is actually the natural cast of the inner table of bone over the frontal lobes. Thus, glabellar shape is indeterminate (Fig. 7A). The interorbital region is too poorly preserved to determine the size and position of the frontal and nasal bones. The maxillary bone contributes to the ventral orbital rim on the left side of the skull and would therefore have separated the lacrimal from the zygomatic. Fragments of the anterior crest delimiting the lacrimal fossa and a small part of the medial wall of the lacrimal fossa are preserved bilaterally (Fig. 7B). CT cross sections show parts of the nasolacrimal ducts bilaterally as well. From this, it appears that the lacrimal fossa is situated in the margin of the orbit. However, it is not clear whether the anterior lacrimal crest and anterior part of the lacrimal fossa are composed of the maxillary bone or are part of the lacrimal bone extending onto the face. In Tremacebus, the anterior crest of the lacrimal fossa is composed of the lacrimal bone. In Homunculus (Cordoba skull), the lacrimal bone extends onto the face anterior to the orbital margin and contacts the frontal, thereby separating the maxilla from the frontal. Extension of the lacrimal onto the face anterior to the orbit is seen also to varying degrees in Callicebus and atelids. Postorbital closure is extensive on the right side of the Dolichocebus skull (Figs. 7B and 8A), but it is impossible to discern the precise contributions made by individual bones. Nevertheless, the closure was similar to that of most platyrrhines (e.g.,

Callimico; Fig. 8B) and much more than in Aotus, where a lateral orbital ssure extends laterally to the root of the zygomatic arch (Fig. 8C). Homunculus has an extensive postorbital septum (Tauber, 1991). Although Hershkovitz (1974) claimed that Tremacebus possessed a large infraorbital ssure, our observations on the original specimen in Tucuman indicate that the size of the opening in the orbital apex is largely due to postmortem breakage (see also Fleagle and Rosenberger, 1983). As in all anthropoids, the orbits are convergent and frontated and closely approximated beneath the olfactory bulbs. The exact amount of convergence is uncertain because the orbital margins are incompletely preserved. However, comparison of the rostra of Dolichocebus and Saimiri, when distortion is accounted for, show that the former is considerably more prognathic and that its right ventral orbital margin is oriented less transversely (i.e., less convergently) than in Saimiri (Fig. 1, dorsal view, and Fig. 9B). Orbital convergence is notably less in extant callitrichines (range 58 e63 ) than in other living platyrrhines (range 66 e75 ) or catarrhines (range 70 e82 ) (Ross, 1993), and convergence in Dolichocebus was primitive, and that of extant callitrichines is likely to have evolved independently. Tremacebus also is comparable to callitrichines and has less orbital convergence than other extant platyrrhines of similar size (see Fig. 7 in Hershkovitz, 1974). Callitrichinelike less convergent orbits also characterize late Eocene and early Oligocene African anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus (Simons, 2004), suggesting that the extent of orbital convergence in Dolichocebus and Tremacebus is primitive, and that of extant callitrichines is likely to have evolved independently. Aegyptopithecus, at 72 or greater (Ross, 1995; Simons, 2004), had a catarrhine-like level of orbital convergence. The interorbital region of Dolichocebus appears to be narrow (interorbital breadth, 5.7 mmdalthough we caution that this is a minimum estimate because the skull has been crushed mediolaterally) (Rosenberger, 1979a). In Dolichocebus, the interorbital region is most notable for a large opening in the matrix connecting the orbits (Figs. 7, 9B). Coronal interorbital sections of Saimiri (Fig. 10A) show that the edges of the interorbital fenestra are bounded by a single lamina of bone both dorsally and ventrally. In Saguinus, an interorbital septum is composed of two laminae of bone; further ventrally in Saguinus, the vomer is sandwiched between the right and left ventromedial surfaces of the orbit (Fig. 10B). The CT images show that the area immediately surrounding the fenestra of Dolichocebus is composed mostly of matrix. Preserved interorbital bone ventral to the interorbital opening in Dolichocebus (Fig. 11) shows a right and left paranasal sinus, most probably the sphenoidal (cupular) sinus above the nasal cavity, and below the interorbital septum. However, the right and left orbital walls on either side of the sinus converge dorsally, suggesting that the interorbital septum in Dolichocebus was probably not wide as in Callicebus, but narrow as in Cebus or Saguinus (Fig. 20b,c in Rossie, 2006). Thus, while it seems likely that Dolichocebus had a narrow interorbital septum, the CT scans neither conrm nor disprove the presence of an interorbital fenestra. Nuchal region. As in all anthropoids, Dolichocebus has a pneumatized mastoid bone. The portion of the mastoid

334

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Optic canal

Superior orbital fissure Inferior orbital fissure

C
Optic canal

Superior orbital fissure

Inferior orbital fissure

Fig. 8. Views of the right orbits of platyrrhines showing the optic canal, superior orbital ssure, and inferior orbital ssure: (A) Dolichocebus gaimanensis, (B) Callimico goeldii, and (C) Aotus trivirgatus. Note that the postorbital ange of the zygomatic (indicated by the arrows) is more extensive in Callimico than in Aotus. The postorbital ange of the zygomatic in Dolichocebus is partially preserved. Its extent is also documented in Fig. 7B. The ange of Dolichocebus more closely resembles that of Callimico in its preserved parts.

well as Tremacebus and Homunculus, but differs from the condition in Saimiri, in which the nuchal plane forms a much more acute angle with the Frankfurt horizontal (Fig. 12C) (Hershkovitz, 1977). Zygomatic region. The zygomatic arches are broken away bilaterally, with only a sliver of the posterior root preserved on the right side. It is impossible to tell from this remnant whether the entire arch was lightly constructed, as in Saimiri and many callitrichines, or robust, as in, for example, Callicebus (Fig. 12B, C). Lateral wall of the braincase. The sutures in the region of pterion are obscured by breakage, and it is impossible to tell whether the frontal and alisphenoid are in contact (as in living catarrhines) or if the zygomatic and parietal are in contact (as is typical of platyrrhines, with the exception of Alouatta and some other atelids, in which it is variable) (Ashley-Montague, 1933; Mouri, 1988). Notably, the pterionic region in both Tremacebus and Homunculus (Tauber, 1991) shows the catarrhine condition. Although the lateral wall of the braincase is also broken in many places, the remnants on the right side suggest that Dolichocebus did not possess a temporal emissary foramen, as reported for atelids (Conroy, 1981; Horovitz and MacPhee, 1999). Neither Tremacebus nor Homunculus has a temporal emissary foramen. On the lateral wall of the braincase at the anteroposterior level of the external auditory meatus, a faint right temporal line is present (Fig. 13). This line does not reach the midline of the skull, so there is no sagittal crest, but, as also noted by Rosenberger (1979a), the temporal lines are more closely approximated to the midline of the cranial vault than in Saimiri. Rosenberger (1979a) suggested the closer approximation of the temporal lines in Dolichocebus to be a consequence of its larger, more prognathic face and larger teeth than Saimiri. An alternative possibility that needs further investigation is that Dolichocebus had a smaller brain and braincase, leaving less bony surface for the temporalis muscle to attach. Tremacebus has the same morphology as Dolichocebus. Temporomandibular region. The surface of the mandibular fossa, preserved on the right side of the skull, is broad and at, with no indication of an articular eminence (Fig. 14). The right postglenoid process is slightly damaged, but its length can be reliably estimated as ~1.65 mm. This is relatively very small, within the range of extant Saimiri and callitrichines.3 Tremacebus likewise has a very small postglenoid process, whereas the process is large in Homunculus. The right postglenoid foramen of Dolichocebus is situated posterolateral to the postglenoid process and is quite large (Fig. 14)dapproximately 1.8 mm in diameterdindicating

closest to the occipital protuberances is attened as in Tremacebus, Homunculus, and cebids. In contrast, pitheciids (Rosenberger, 1979b) and atelids have small paraoccipital processes. Inion of Dolichocebus is placed dorsally relative to the foramen magnum. Thus, the nuchal plane forms a relatively obtuse angle with the Frankfort horizontal (Fig. 12A). This is typical of most living platyrrhines (Fig. 12B) (Hershkovitz, 1977), as

3 We measured the length of the postglenoid process in small samples of 15 genera of extant platyrrhines. Relative size of the process is quantied by the ratio 100 (postglenoid length/prosthioneinion length). The ratio for Dolichocebus was estimated to be 242. Four genera and 16 specimens of callitrichines range from 67 to 243, with Leontopithecus at the upper end of the range. Three specimens of Saimiri range from 195 to 282. Atelids have much larger postglenoid processes: nine specimens representing four genera range from 525 to 1240.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

335

A
incompletely ossified interorbital septum

more prognathic facial region

Fig. 9. Anterolateral views of the rostra of (A) Saimiri and (B) Dolichocebus. The incompletely ossied interorbital septum of Saimiri is mimicked in matrix by Dolichocebus. Note also the more prognathic face of Dolichocebus.

that a large intracranial venous drainage channel (the petrosquamous sinus) emerged at this point (Saban, 1963; Conroy, 1980; Kay et al., 2008). The foramen is smaller in similarly sized Tremacebus (~0.74 mm). Homunculus resembles Dolichocebus in having a relatively large foramen. Pterygoid and palatal region. The pterygoid fossa, pterygoid plates, and the pyramidal processes are broken away. Distally, the tooth rows of Dolichocebus diverge only slightly: the ratio of the distance between the lingual sides of the upper canines (9.49 mm) to the distance between the lingual sides of the M2s (11.44 mm) is 83%. These proportions of Dolichocebus are similar to those of Tremacebus (Hershkovitz, 1974). Eocene/Oligocene anthropoids and many extant catarrhines have similarly nondivergent tooth rows, whereas many crown platyrrhines have postcanine teeth that diverge more markedly. The remnants of the posterior parts of the palatal processes of the palatine bone are slightly thickened to form a weak

Fig. 10. Coronal CT sections of the interorbital region: (A) Saimiri sciureus and (B) Saguinus fuscicollis.

posterior palatine torus. We score this feature as present in Dolichocebus in our character-taxon matrix while recognizing that its development is far less robust than in some nonanthropoid primates like Adapis (Ross, 1994). Facial region. The depth of the maxilla of Dolichocebus is comparable to that in most anthropoids. The distance from the left orbital margin to the anterior edge of the canine alveolus is 10.8 mm (preorbital rostrum length; [Ross, 1994]), and the

336

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

olfactory fossa and medial orbital wall

maxillary sinus

anterior cranial fossa

nasal cavity

medial orbital walls and vomer

Anterior ethmoid or 'sphenoid' sinus

dorsal roof of nasal cavity

nasal septum
Fig. 11. Coronal CT sections of the interorbital region of Dolichocebus illustrating the maxillary sinus, the nasal cavity, and a large sinus, probably the cupular sinus, dorsal to the nasal cavity.

distance from the anterior edge of the canine alveolus to the nasomaxillary suture (maxillary depth [Ross, 1994]) is >9.5 mm. Interpolating these data onto a log-log bivariate plot of preorbital rostrum length versus maxillary depth for 11 extant anthropoids and 19 extant strepsirrhines (Fig. 17 in Ross, 1994), Dolichocebus ts within the anthropoid cluster (short rostrum and deep maxilla compared with strepsirrhines). As in other anthropoids, the snout of Dolichocebus is short. With the skull oriented in the Frankfurt horizontal, the left infraorbital foramen is located quite far anteriorly on the face, above a point between P2 and P3 (Fig. 12A). Ear region and braincase. The petrosals are well preserved on both the right and left sides. The right middle ear cavity and its anterior accessory chamber are exposed (Fig. 14). As in most platyrrhines, excepting atelids and Cacajao (Horovitz, 1997), Dolichocebus possesses two prominences on the lateral surface of the promontorium (as does Tremacebus). These

prominences are the external manifestations of the cochlea on the promontory surface of the middle ear. The dual prominences may be a derived feature of crown platyrrhines subsequently lost in atelid platyrrhines. This is supported by the observation that Tarsius, Aegyptopithecus, and Simonsius have a singular prominence (Kay et al., 2008). However, there are paired prominences on the cochlear housing in the middle ear in Apidium (see Fig. 1 in Cartmill et al., 1981). Moreover we have observed variation in this feature in extant Saimiri, Cebus, and Aotus. As in all haplorhines, a transverse septum separates the tympanic cavity proper from a well-developed anterior accessory chamber (AAC) (Cartmill and Kay, 1978; MacPhee and Cartmill, 1986). The latter develops as a diverticulum from the auditory tube (Ross, 1994). The AAC extends more medially and has trabeculae within it, as in other anthropoids (Ross, 1994). On the right side of the cranium, a transverse septum is

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

337

postorbital closure

angle of nuchal plane infraorbital foramen

Fig. 13. Dorsal view of skull of Dolichocebus gaimanensis showing the position of the temporal lines.

postorbital closure

infraorbital foramen

angle of nuchal plane

postorbital closure

angle of nuchal plane infraorbital foramen

Fig. 12. Views of skulls showing angle of the nuchal plane, root of the zygomatic arch, and position of the infraorbital foramen: (A) Dolichocebus gaimanensis, (B) Callicebus torquatus, and (C) Saimiri sciureus.

present, as is an anterior accessory chamber. The right petrosal of the specimen is broken ventrally, exposing the middle ear cavity, the auditory tube, and the AAC. The AAC extends medially and posteriorly onto the promontorium. It appears to be largely devoid of trabeculation dorsally, but is trabeculated laterally and ventrally. Medial extension of the anterior accessory chamber is a derived feature of crown and (as far as is known) stem anthropoids. It is a feature of Aegyptopithecus, Apidium,

and Simonsius ( Parapithecus), as well as Proteopithecus and Catopithecus (Kay et al., 2008). The posterior carotid foramen of Dolichocebus (marking the entrance into the bulla of a large internal carotid artery) is identiable on the petrosal. Inside the bulla, a ridge on the ventrolateral surface of the promontorium marks the presence of a bony tube for the internal carotid (promontory) artery (Fig. 14), the presence of which is conrmed in CT scans (Fig. 15B). As in other platyrrhines, the posterior carotid foramen is located posterior to a line joining the midpoints of the ectotympanic elements, and medial to the midline of the auditory bulla, here established as a line joining the stylomastoid foramen and the anteromedial-most point on the petrosal, but far forward of the stylomastoid foramen. The same position appears to hold for Proteopithecus, Catopithecus, and parapithecids. In catarrhines and Aegyptopithecus, it is also positioned posteriorly but more nearly in the middle of this line (Kay et al., 2008). The ectotympanic element is completely visible on the right (Fig. 14), while only a dorsal remnant is preserved on the left. As is the case among all known Fayum anthropoids and platyrrhines (Kay et al., 2008), the ectotympanic is extrabullar, forming a bony ring at the entrance to the tympanic cavity. The ring is ossied outward to form a tube in catarrhines (only partially so in Pliopithecus) (Szalay and Delson, 1979; Fleagle and Kay, 1987) and Tarsius. On CT images of the skull of Dolichocebus, a narrow canal can be seen connecting the subarcuate fossa with the sulcus for the sigmoid sinus (Fig. 16). This is Cartmills canal, a venous channel (Cartmill et al., 1981; Kay et al., 2008) present in all platyrrhines, including Tremacebus. It is absent in Tarsius and catarrhines and absent in the Oligocene parapithecids Apidium (Cartmill et al., 1981) and Simonsius (Kay et al., 2008). The canal is partially obliterated in the stem catarrhine Aegyptopithecus, but is present and well developed in early stem anthropoids such as Proteopithecus and Catopithecus (Kay et al., 2008). On coronal CT slices through the braincase, the remnants of an ossied tentorium cerebelli are seen bilaterally above the subarcuate fossa (Fig. 15). Horovitz and MacPhee (MacPhee et al., 1995; Horovitz, 1999; Horovitz and MacPhee,

338

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

for. magnum

jugular for. carotid for.

stylomastoid for. dual promomtorial processes

postglenoid for. & process transpromontorial carotid canal occipital condyle

basioccipital stem

ant. accessory chamber

Fig. 14. Stereopair of right basicranial region of Dolichocebus gaimanensis.

1999) stated that the tentorium is absent in Tarsius and (sometimes absent in Saimiri) but present in all other platyrrhines. An ossied tentorium is also absent in Oligocene Fayum anthropoids Aegyptopithecus and Apidium (Kay et al., 2008). In contrast, Hershkovitz (1977) noted that it is most extensive in Ateles, Lagothrix, and Brachyteles; peripheral in Callicebus, Aotus, and pitheciines; variable in Alouatta; minimal in Saimiri, Cebus, and Callimico; and absent or rudimentary in other callitrichines. Our observations on additional specimens support Hershkovitzs observations for the most part. However, CT images show the tentorium to be quite well developed in Callimico. Descriptive anatomy of the dentition Lower incisors (MACN CH 872, a right I1 or I2 [not gured]). The crown of MACN CH 872 is broken a short distance above the cervix. The tooth is spatulate. The root has a broadly oval cross section, not mesiodistally compressed as in Cebuella and Callithrix. The lingual aspect of the crown has enamel (in Callithrix, the lingual enamel is extremely thin or absent (Rosenberger, 1978)) and a well-developed lingual heel (unlike in Cebuella and Callithrix or pitheciines, in which the lingual face is attened). Mesial and distal lingual crests are present, but they do not join lingually in the midline. Compared to the lower rst molar described below, it is a very small toothdthe area ratio is 0.27, comparable to ratios in Alouatta, Brachyteles, Callimico, and Callicebus. Among early Miocene platyrrhines, Homunculus has similarly small incisors (Hershkovitz, 1970), but those of Soriacebus are enlarged (Fleagle et al., 1987; Fleagle, 1990). In all observable features of structure and proportions enumerated in our character-taxon matrix, the structure of the lower incisor of Dolichocebus is most similar to that of Callimico. Lower canine (MACN CH 870, a right canine; MACN CH 871, a left canine [Fig. 17A,B]). Both of these teeth have

well-preserved crowns, although the distal heel of MACN CH 870 is partly broken away. Specimen MACN CH 870 is a larger tooth with a broadly oval cross section, whereas MACH CH 871 is smaller and slightly more compressed. However, neither tooth is as mesiodistally compressed as in

petrosquamous sinus

subarcuate fossa tentorium cerebelli middle ear carotid canal cochlea carotid canal

B
trabeculated anterior acessory chamber

inferior petrosal sinus

Fig. 15. Two coronal CT sections of the skull of Dolichocebus gaimanensis; the upper image (A) is more posterior than the lower image (B). Features identied include the carotid canal crossing the promontorium, trabeculation in the anterior accessory chamber, the petrosquamous sinus (venous drainage through postglenoid foramen [not shown]), the subarcuate fossa, and a partially ossied tentorium cerebelli.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

339

mastoid air sinus subarcuate fossa Cartmill's canal

1122

sigmoid sinus semicircular canal

Cartmill's canal

1141

sigmoid sinus

C
Cartmill's canal

1156

jugular foramen

Fig. 16. Three coronal CT sections of the skull of Dolichocebus gaimanensis; the upper image (A) is more posterior than the middle image (B), which is more posterior than the lower image (C). Features illustrated include mastoid air sinuses and the posterior end of the subarcuate fossa. Cartmills canal, a vascular channel, leads transversely to open into the sigmoid sinus near the junction with the jugular foramen.

Cebuella or Callithrix. The size difference between the two specimens could be due to sexual dimorphism. If so, the degree of dimorphism would have been considerable, within the range seen in Saimiri. Canine dimorphism is either unknown or has not been examined among other early Miocene platyrrhines. On both teeth, the lingual crest is rounded, not sharp, like pitheciines. A raised and unbroken lingual cingulum runs from the large distal heel to the mesial terminus of the paracristid. The ratios of canine size (area) to the size of the only M1 in the sample are 0.64 and 0.83 for the two specimens. Although the specimens come from different individuals and the ratio is not necessarily reliable, lower canine size seems to be comparable to that of Aotus or Callicebus

in our sample of extant taxa and smaller than in Saimiri, in which the full range of ratios for a sample of 44 specimens is 0.79 to 1.39. Among early Miocene platyrrhines, the canine size of Dolichocebus is similar to that of Carlocebus and Homunculus, but smaller than in Soriacebus. Lower premolars (MACN CH 865, P2; MACN CH 868a, right P3 [Fig. 17C]; MACN CH 868b, right dp3; MACN CH 873, left P3 [Fig. 17D]; and MACN CH 898, left P4). Specimen MACN CH 865 is tentatively identied as a P2 because of the elongate paracristid, the buccal swelling at the base of the protoconid, and the simplicity and narrowness of the talonid. If Dolichocebus were sexually dimorphic, as seems likely from the canines, this tooth would be that of a female. The tooth has a single

340

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

root, as in all living and extinct platyrrhines (including Carlocebus, Homunculus, and Soriacebus) and EoceneeOligocene Proteopithecus and parapithecids. Catarrhines, propliopithecids, and oligopithecids have lost the P2. It has an oval occlusal outline with a slight buccal are at the base of the protoconid. A buccal cingulum is absent, as in Proteopithecus (late Eocene, Africa), catarrhines, Homunculus, Soriacebus, and most extant platyrrhines. The trigonid has a protoconid but lacks a metaconid and paraconid. A P2 metaconid occurs variably in Eocene/Oligocene Fayum anthropoids. A small metaconid appears variably in Homunculus but is not present in either Carlocebus or Soriacebus. The protoconid appears not to have projected above the protoconids of either P3 or P4, as it does in Saguinus and Callimico. As in all known platyrrhines, the talonid is small and lacks a hypoconid and entoconid. Two teeth in our sample, MACN CH 873 and MACN CH 868a represent the P3. Each tooth has a single root (as in the P3 and P4 of all other living and fossil platyrrhines but not in any catarrhine or stem anthropoid). The crown is mesiodistally short and buccolingually broad. There is a prominent protoconid and a smaller metaconid distolingually. The small size of the metaconid and its close proximity to the protoconid resemble the condition in most living callitrichines (but not Callimico). Other extant platyrrhines have a larger metaconid with greater separation from the protoconid. A small P3 metaconid also characterizes late Oligocene Branisella and early Miocene Carlocebus, Homunculus, and Carlocebus. The distolingual position of the cusp is also seen in other early Miocene platyrrhines. No paraconid is present (see comments below concerning paraconid loss on P4). The trigonid is open lingually, unlike that of any extant platyrrhine. The talonid has a short cristid obliqua and a very small hypoconid positioned distolingual to the protoconid. No hypocristid is present. The lingual marginal crest of the talonid is weak and lacks a discrete entoconid. The buccal surface of the crown of MACN CH 873 has a poorly developed cingulum, whereas there is no buccal cingulum on MACN CH 868a. The crown is slightly swollen buccal to the protoconid, although less so than in P2. Specimen MACN CH 898 is identied as a P4 because of the greater structural complexity of the root and crown compared with the aforementioned premolars. This specimen is singlerooted, but the root is deeply grooved. The crown is mesiodistally short and buccolingually broad. The trigonid and talonid are approximately equal in mesiodistal dimensions. The trigonid supports a widely spaced protoconid and metaconid; the latter is smaller than the protoconid and placed slightly distal to it. There is no paraconid, which is a resemblance to all living and extinct platyrrhines (except Carlocebus) and catarrhines.4 The trigonid is widely open lingually, as in some Homunculus, whereas it is closed in Soriacebus, Carlocebus, and Branisella; P4 trigonids in extant platyrrhines are closed. The talonid is broad buccolingually with a small hypoconid buccally. No

Catopithecus has a paraconid, and thus if oligopithecids are stem catarrhines, as advocated by many, then paraconid reduction would have occurred in parallel in platyrrhines and catarrhines.

hypocristid is present. A lingual marginal crest supports a small entoconid. There are several obvious distinctions from Soriacebus, in which P4s are much narrower buccolingually, with the metaconid much closer to the protoconid. Specimen MACN CH 868b is probably a dp3, a conclusion we reach for two reasons: (1) this tooth has similar proportions and shape to the dp3 of Aotus (although it is very different from that taxon in the details of the talonid); and (2) the color of the enamel on the tooth is distinctly lighter than in the other specimens, a phenomenon frequently, but not always, seen in the deciduous teeth of fossil mammals (Kay and Simons, 1983). This tooth is single-rooted and has an elongate oval occlusal outline; it is not buccolingually broad, as in MACN CH 898, nor is there any buccal swelling mesiobuccally. The trigonid is relatively short mesiodistally. A paraconid is absent. The metaconid is larger and is spaced further from, and more directly lingual to, the protoconid. The trigonid is open lingually. The talonid is buccolingually broad with a welldeveloped hypoconid and a small but distinct entoconid. There is no buccal cingulum. There is very little similarity in shape between this tooth and the P4 of Soriacebus. The basins of MACN CH 868b are more squared off, the talonids are broader, the metaconid is bigger and further from the protoconid, and the talonid is more cuspidate. Both P3 and P4 have transversely (buccolingually) constricted crowns with broadly sloping buccal faces. This is characteristic of early Miocene Carlocebus, Homunculus, and Soriacebus. Branisella and most living platyrrhines have more marginally placed cusps (excepting some callitrichines, Cebus, and Ateles). Lower molars (MACN CH 866, left M2 [Fig. 17E]; MPEF 5147 mandible fragment with right M1e3; MPEF 5146 mandible fragment with left M1e2 [Fig. 17F,G]). In the collections made before 1990, only one poorly preserved and heavily worn lower molar (MACN CH 866) was identied. The tooth has two roots but these are fused over two-thirds of their length. The mesial root is larger than the distal. The distal root projects posteroinferiorly from the crown. The distal crown margin is deeply notched by interproximal wear. The form and structure of this wear facet shows that the posteriorly adjacent M3 was displaced laterally. A very weak buccal cingulum is visible in the hypoexid. The trigonid and talonid are approximately equal in breadth. The cristid obliqua is obliquely oriented so that the hypoexid is moderately incised. Two mandibular fragments contain molars. Specimen MPEF 5147 preserves the right M1e3, and MPEF 5146 has left M1e2. Neither specimen preserves sufcient parts of the mandible to gain an impression as to its depth or morphology. The occlusal surfaces of MPEF 5147 were broken postmortem, with only the distolingual corner of M1, the entire M2, and the buccal half of M3 remaining. Occlusal wear in life obliterated the details of the crown morphology. However, the crown proportions can be estimated and root structure discerned. On MPEF 5146, M1 and M2, each with two roots, are well preserved and the dentin is exposed only at the cusp tips. An interproximal wear facet at the distal margin of M2 indicates that three molars were present. The M1 and M2 of

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

341

Fig. 17. Lower teeth of Dolichocebus gaimanensis: (A) MACN CH 870, a right canine, in lingual view (mesial is to the left); (B) MACN CH 871, a left canine, in lingual view (mesial is to the right); (C) MACN CH 868a, a right P3, in occlusolingual view (mesial is to the left); (D) MACN CH 873, a left P3, photographically reversed for comparison, in occlusolingual view (mesial is to the left); (E) MACN CH 866, a left M2, in occlusobuccal view (mesial is to the left); (F) MPEF 5146, a mandible fragment of Dolichocebus with left M1e2, in occlusolateral view; (G) MPEF 5146 in occlusal view (mesial is to the left). Scale bars 3 mm. AeE are at the same scale; F and G are at the same scale.

both specimens possess two widely separated roots; M3 of MPEF 5147 has two widely separated roots. The two-rooted condition is characteristic also of Branisella, Soriacebus, Carlocebus, the unpublished Colhuehuapian species from Gran Barranca, and Homunculus, as well as all taxa of African Eocene/Oligocene anthropoids. In contrast, all living platyrrhines except Alouatta (and with a small number of exceptional specimens of some other species) have single-rooted M3s. For details of the crown anatomy of the lower rst molar, we must rely entirely on MPEF 5146. The M1 crown is squared, having a low ratio of mesiodistal length to buccolingual breadth (Table 3). Low cusp relief and smooth enamel characterize the crown. The trigonid is mesiodistally short, constituting about 35% of the mesiodistal length. The overall breadth of the trigonid exceeds that of the taloniddthe crown bulges buccally and lingually (to a lesser degree) such that the

ratio of protoconid-metaconid breadth to trigonid breadth is only 50%. Thus, the actual occlusal surface is quite compressed and the sides of the marginal cusps bulge, a feature Dolichocebus shares with Soriacebus and a few extant platyrrhines (Cebus and Ateles). Likewise, the occlusal surface of the talonid crown is compressed and the sides of the tooth bulge. The trigonid is three-cusped, with a small paraconid on the mesiolingual edge of the trigonid. The metaconid is positioned distolingual to the protoconid. A lower rst molar paraconid is unusual for a platyrrhine. We note its presence only in some specimens of Homunculus and in an undescribed taxon of Colhuehuapian monkey from Gran Barranca (Kay et al., 1999b). The paracristid is short and mesiodistally oriented. A premetacristid encloses the trigonid lingually. Lingual and buccal protocristids form a complete distal trigonid wall. Wear surface Xdfound on the distal trigonid of

342

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

some parapithecids and (homoplastically) in catarrhines (Kay, 1977)dis absent. The trigonid and talonid are essentially identical in height and breadth. This is typical of platyrrhines generally, with the exception of Soriacebus, in which the trigonid is extremely elongate. The entoconid is large and positioned distolingual to the hypoconid. In contrast, Soriacebus has a very reduced entoconid. The reduced condition is also seen in many callitrichines and some advanced pitheciines (but not reduced in Pithecia). The distal edge of the talonid supports a raised ridge conuent with the hypocristid and adorned with two bulges, one of which is in the midline of the tooth and the other at its lingual terminus. The latter is interpreted to be a hypoconulid, placed lingual to the midline of the tooth. A larger, more lingually placed hypoconulid is present in Branisella. The small, medially or lingually placed cusp also appears in Carlocebus, Soriacebus, and, variably, in Homunculus, whereas it is an unusual feature in living platyrrhines. Distolingually, the hypoconulid is separated from the entoconid by a deep sulcus (groove) and there is a small posterolingual fovea. The presence of a sulcus is unusual in extant platyrrhines, occurring in some specimens of Callicebus and Alouatta. It is prominently or weakly displayed in Branisella, Homunculus, Carlocebus, and Soriacebus. The posterolingual basin is variably present in Callicebus alone among extant platyrrhines, and also occurs in early Miocene Carlocebus. The cristid obliqua connects with the trigonid wall at a point distolingual to the protoconid and does not extend up the trigonid wall to the protocristid. The lingual margin of the talonid basin has a distinct notch where the postmetacristid abuts the preentocristid. The presence of a lingual talonid notch is unusual in living platyrrhines, appearing only in Callicebus and some specimens of Alouatta. The notch is present in Branisella and also in the undescribed Gran Barranca platyrrhine. A strong buccal cingulum wraps around the protoconid and spans the hypoexid, but disappears lingual to the hypoconid. Similar cingulum development is found in the Gran Barranca primate (but not in other early Miocene platyrrhines or Branisella) and in Saimiri and Saguinus among extant platyrrhines. The cingulum is notched in the hypoexid, where a strong cuspule is present. Wear facets on the distolingual corner of M1 and continuing onto the trigonid of M2 indicate that M1 had a hypocone. The description of M2 is based on MPEF 5146 (Fig. 17G,F). The morphology of the tooth is similar to that of the M1. The M2 crown, short mesiodistally and broad buccolingually (Table 3), has low cusp relief and smooth enamel. The trigonid is short, constituting about 33% of the mesiodistal length. The occlusal surface of the trigonid is quite compressed, with the crown bulging buccally and lingually such that the ratio of protoconidmetaconid breadth to trigonid breadth is 52%. The trigonid is two-cusped; unlike on M1, there is no paraconid. The metaconid is positioned more lingually and not as much distally to the protoconid compared with the conguration on M1. Other aspects of the trigonid of M2 resemble those of M1. Wear surface X on the distal trigonid is absent. The trigonid and talonid are essentially identical in height but the trigonid is wider than the talonid (versus similar breadths on M1). The entoconid is large and positioned more nearly lingually to the hypoconid than on M1.

The conguration of the distal margin of the talonid is very similar to that described on the M1. Distolingually, the hypoconulid is separated from the entoconid by a deep sulcus. The cristid obliqua is oriented more mesiodistally than in M1 and reaches the protoconid, leaving the hypoexid more shallow than on M1. As with M1, the lingual margin of the talonid basin is closed but notched. The buccal cingulum is much weaker than on M1 and spans only the hypoexid, disappearing lingual to the protoconid and hypoconid. The cingulum is notched at several points, but no discrete cuspules are present. As with M1, wear facets on the distolingual corner of M2 give a clear indication that M2 had a hypocone. The M1 is greater in overall size (i.e., area) than M2, but in length, M1 and M2 are similar. From the broken remnants of MPEF 5146, it appears that M3 was slightly shorter mesiodistally than M2, and substantially narrower. Upper incisors (MACN CH 359, a left I1 [Fig. 18A]). No upper incisors from late Oligoceneeearly Miocene platyrrhines have been reported previously, so our comparisons are restricted to extant platyrrhines. The I1 has a rounded, oval outline in occlusal view. A complete lingual cingulum is present. Although somewhat obscured by wear, it appears that a basal lingual cusp was absent. If so, this absence is a resemblance to callitrichines, Callimico, Aotus, and Saimiri. Upper canine (MACN CH 361, a right canine [Fig. 18B]; MACN CH 1302, a left canine [Fig. 18C]; MACN CH 1303, a right canine). The occlusal outline of MACN CH 361 is a tear-drop-shaped oval. The crown is tall and slender. A deep mesial groove is present on the mesial aspect extending onto the root, a condition seen on Branisella and all early Miocene platyrrhines. There is a complete lingual cingulum, but it is a narrow shelf, not bulging as in some living platyrrhines such as Ateles. The shelf is similarly proportioned in Dolichocebus, Branisella, Carlocebus, and Homunculus. The Soriacebus canine has a weaker lingual cingulum. Specimens MACN CH 1302 and 1303 have a similar morphology to the tooth just described, except that the crowns are less projecting and smaller overall. These two sizes and shapes suggest that Dolichocebus canines were sexually dimorphic. Upper premolars (MACN CH 1012, a left P2; MACN CH 864, a right P3; MACN CH 878 [Fig. 18D], a right P4; MACN CH 1011, a right dp4). The P2 is single-rooted as inBranisella, Tremacebus, and most other platyrrhines (except Soriacebus and, among extant platyrrhines, Aotus, Alouatta, and Callicebus, which have two-rooted P2s). In buccal view the tooth is lancelet-shaped and bulges at the cervix. The lingual cingulum is well developed and was probably complete, but lacks a protocone. As in all extant platyrrhines, crests run from the paracone to the mesial and distal borders of the crown. These are asymmetric, with the mesial crest shorter than the distal. The P3 (MACN CH 864) is two-rooted. In addition to the single lingual root, there appears to be a single buccal root that is broken close to the cervix. The occlusal outline is nearly oval, but broader buccally than lingually. A lingual swelling is present, here interpreted as a small protocone. Strikingly, like Branisella but unlike other platyrrhines, the

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

343

Fig. 18. Upper incisors, canines, and premolars of Dolichocebus gaimanensis: (A) MACN CH 359, a left I1, in occlusolingual view; (B) MACN CH 1302, left canine, in occlusolingual view; (C) MACN CH 361, right canine, in occlusolingual view; (D) MACN CH 878, a right P4, in occlusal view (mesial is to the left). Scale bar 3 mm.

protocone is situated on the mesial margin of the crown. A small hypocone is present. The distal prole of the tooth is smoothly convex, not concave (waisted) as in Callithrix and Cebuella. The P2 occlusal area is much less than P3 area. The P4 (MACN CH 878) is also two-rooted, nearly oval in occlusal outline, with a convex distal prole. It has a strong protocone. A large hypocone is present on a very strong lingual cingulum. The large hypocone and pronounced cingulum are features shared with Soriacebus, Dolichocebus, Homunculus, and Carlocebus; the cusp is smaller or absent in extant platyrrhines. The cingulum begins at the distal crown margin and wraps completely around the lingual base of the protocone. This conguration resembles that of Callicebus, which has a very strong lingual cingulum and a smaller hypocone, although in some Callicebus the hypocone is often lingual

to the cingulum rather than connected with it. The tooth has a small parastyle and metastyle but lacks conules. The postprotocrista is weak and reaches toward the base of the hypocone. Such a short, weak postprotocrista is characteristic of all Oligoceneeearly Miocene platyrrhines but is not seen in extant platyrrhines, with the exception of Callithrix and Cebuella). In extant platyrrhines, the crest is strong and reaches distolingually to the distal margin of the crown. A groove is present between protocone and hypocone. This morphology resembles that of Carlocebus and Soriacebus (and Callicebus, although the groove between the protocone and hypocone is shallower than in the latter taxa). The P4 is small compared to M1; the ratio of crown areas is 0.59, resembling Oligocene/Miocene Branisella, Carlocebus, and Homunculus, as well as Callimico and marmosets among extant platyrrhines. This condition is distinctly different from that in Saimiri, in which P4 is enlarged. The dp4 (MACN CH 1011) is a three-rooted tooth. The lingual root is splayed lingually, which suggests that this is a deciduous tooth. The buccal tooth roots are broken off. There is a strong preprotocrista. The hypocone is small and distolingual to the protocone. A strong prehypocrista encloses the talon basin lingually. A very small swelling on the preprotocrista may be a paraconule; there is no trace of a metaconule. The parastyle and mesostyle are very small and associated with a very weak buccal cingulum. The strong lingual cingulum, leading from the small hypocone, wraps part way around the protocone before disappearing mesiolingual to that cusp. A postprotocrista is strong and joins the prehypocrista. A lateral transverse crista is continuous with the postprotocrista. Upper molars (MACN CH 876, a left M1; MACN CH 356, a left M2 [Fig. 19C,D,E]; MACN CH 357, a right M3 [Fig. 19A]). In extant platyrrhines, the metacone is smaller than the paracone on M2, whereas they are similar in size on M1. On this basis, MACN CH 876 is here identied as an M1 and MACN CH 357 as an M2. The M1 and M2 are three-rooted. The M1 is heavily worn and many of the details of its crown morphology are obliterated. Most of the details of occlusal anatomy will be based on the M2 structure, but where the condition is determinate on M1, this will be noted. The M2 has a strong preprotocrista connected with a small paraconule. Paraconules are an unusual feature in extant platyrrhines, found only in Callimico. Early Miocene Homunculus and Carlocebus share this characteristic with Dolichocebus. African Eocene/Oligocene parapithecids have paraconules, as do oligopithecids and some propliopithecids. A metaconule is absent. This cusp is present in parapithecids and variable in extant platyrrhines. A pericone is absent on both teeth. The hypocone is large and is distal and slightly lingual to the protocone on both teeth. A prehypocrista is present but weak. On M2, there is a small mesostyle, but this cusp is not attached to the premetacrista or postparacrista. A mesostyle is absent on M1. Both teeth have a weak and discontinuous buccal cingulum, but both have a strong lingual cingulum. On M1, the lingual cingulum disappears lingual to the protocone. On M2, it wraps all the way around the lingual aspect of the protocone and reaches to the preprotocrista. On both teeth, a strong postprotocrista runs toward the base of the metacone

344

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

lingual end of a crest running from the paracone. A small metaconule is present. There is no pericone. The hypocone is present but much smaller than on the other molars. There is no prehypocrista. The tooth lacks a mesostyle and there is no buccal cingulum. A lingual cingulum is present but disappears mesially at the base of the protocone. A strong postprotocrista runs toward the base of the metacone where it joins the lateral posterior transverse crista. Body size Table 5 provides estimates of body mass from lower rst and second molar dimensions based on formulae published by Conroy (1987), Kay and Simons (1980), and Kay and Meldrum (1997). Mass determinations range from 1149 to 1754 g. Perhaps the most appropriate estimate is derived from M1 areas and body weights of 15 platyrrhine species, yielding the following least-squares regression (r2 0.935): ln female body weight ln M1 area (1.565) 3.272. Based on this equation, the estimated weight of MPEF 5146 would be 1554 g; if MACN CH 866 is an M1, it would come from an animal estimated to weigh 1527 g, about the same size as the living masked titi monkey (Callicebus personatus). Likewise, the length of the skull (prosthioneinion) of Dolichocebus (~68 mm) falls within the range of skull lengths of Callicebus personatus (mean of 10 males 71.7 mm; mean of 10 females 68.9 mm) (Hershkovitz, 1990). Mean C. personatus weights are 1270 g (males) and 1378 (females) (Hershkovitz, 1990). Thus, the estimates from lower molar dimensions and cranial lengths are concordant. Phylogenetic analysis Two runs were undertaken using PAUP and the charactertaxon matrix in Appendix 2. In the rst run, extant taxa were constrained by a molecular tree topology for extant platyrrhines and catarrhines and with the position of all extinct taxa unconstrained. Thus, an extinct taxon may link wherever it is most parsimonious. In the second set of runs, we analyzed the data set with the topology of extant platyrrhines and catarrhines unconstrained. The two runs yielded essentially the same result. The backbone-constrained tree yields one most-parsimonious tree (Fig. 20A). The tree is 66440 steps. The consistency index is 0.343, the retention index is 0.526, and the rescaled consistency index is 0.180. The tree unites the crown platyrrhines as a clade to the exclusion of all of the extinct taxa studied except for Proteropithecia, which is linked to the sakis and uakaries, as previously reported by Kay et al. (1998a). Branisella is positioned as the most basal stem platyrrhine. All of the early Miocene taxa examined form a clade, within which Dolichocebus joins Carlocebus as its sister taxon, with Tremacebus and Soriacebus more distantly related.

Fig. 19. Upper molars of Dolichocebus gaimanensis: (A, B) MACN CH 357, a left M3, in occlusal and occlusodistal views; (CeE) MACN CH 356, a left M2, in occlusolingual stereopair, occlusal stereopair, and mesial views. Scale bar 3 mm.

where it joins the lateral posterior transverse crista. The M2 is slightly smaller than M1. The M3 has a single lingual root and a twinned buccal root. There is a strong preprotocrista. A paraconule appears at the

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Table 5 Estimated body mass of Dolichocebus gaimanensis from M1 and M2 measurements Specimen M1 area (mm2) M2 length (mm) Body-mass estimate (g) Conroys all-primate regression for M1 1322 (383e4568) 1296 (377e4461) Conroys anthropoid regression for M1 1754 (778e3953) 1724 (766e3879) Kay and Simons M2 regression 1262 1149

345

Kay and Meldrums platyrrhine M1 area 1554 1527

MPEF 5146 MACN CH 866

13.53 12.87

3.90 4.03

Notes: Formulae for M1 area are from Conroy (1987) and Kay and Meldrum (1997); M2 length formula is from Kay and Simons (1980).

A
Tarsius Aegyptopithecus zeuxis Hylobates Miopithecus talapoin Presbytis melalophos Apidium phiomense Simonsius grangeri Callicebus Pithecia Chiropotes Cacajao Proteropithecia Aotus Saimiri Cebus Saguinus Callithrix Cebuella Callimico Leontopithecus Alouatta Ateles Lagothrix Brachyteles Dolichocebus gaimanensis Carlocebus carmeni Tremacebus harringtoni Soriacebus ameghinorum Branisella boliviana Catopithecus browni Proteopithecus sylviae 1000 changes
Fig. 20. (A) Maximum-parsimony analysis of the cranial and dental matrix, as constrained by the molecular backbone. The results are depicted as a phylogram in which the branch lengths are proportional to the number of character changes. The scale for 1000 changes is given at the bottom. The character weight for each character is 100, so the value 1000 is equivalent to 10 changes in a two-state character. (B) 50% majority consensus of the 11 shortest trees, when parsimony is relaxed by 130 steps.

346

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Tarsius Aegyptopithecus zeuxis Hylobates Miopithecus talapoin 100 Presbytis melalophos Apidium phiomense Simonsius grangeri Callicebus Pithecia 100 100 Chiropotes Cacajao Proteropithecia Aotus Saimiri Cebus Saguinus 91 64 100 Callithrix Cebuella Callimico Leontopithecus Alouatta 55 Ateles Lagothrix Brachyteles Dolichocebus gaimanensis 73 Carlocebus carmeni Soriacebus ameghinorum Tremacebus harringtoni Branisella boliviana Catopithecus browni Proteopithecus sylviae
Fig. 20 (continued).

By iteratively relaxing parsimony (i.e., by examining the concurrence of nodes in trees of successively greater lengths, or Bremer support [Bremer, 1994]), we examined the stability of this phylogenetic pattern. Six further trees were identied up to 66,530 steps that continue to preserve the integrity of the crown platyrrhines, and recognize a cluster of late Oligoceneeearly Miocene taxa as stem platyrrhines (i.e., Branisella, Tremacebus, Dolichocebus, Soriacebus, and Carlocebus). However, only a Dolichocebus-Carlocebus clade among the stem taxa receives 100 percent support (the 50% majorityrule consensus of these 7 trees is shown in Fig. 20B). The unity of crown Platyrrhini, excluding any of the fossil taxa in our sample (again with the exception of Proteropithecia), breaks down at the fourteenth tree (>66,580 steps).

With respect to the second set of runs (extant taxa unconstrained by molecular data), trees recovered from the cranial and dental data set are very similar to those found in the analysis in which the extant taxa were constrained by molecular data. We obtained a single most-parsimonious tree with the unconstrained cranial and dental data. The tree agrees with the molecular phylogenies in the following features: (1) Tarsius is sister to anthropoids; (2) a platyrrhine-catarrhine split is supported; (3) late Oligocene and early Miocene taxa are placed as stem platyrrhine taxa; (4) a crown platyrrhine clade (with Proteropithecia) is present; (5) a callitrichine clade includes Callimico; (6) a pitheciine clade is supported; and (7) an atelid clade is supported. The morphological tree differs from that of the molecular tree in a number of ways as

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

347

Tarsius Aegyptopithecus zeuxis Hylobates Miopithecus talapoin Presbytis melalophos Apidium phiomense Simonsius grangeri Callicebus Aotus Cebus Pithecia Chiropotes Cacajao Proteropithecia Alouatta Brachyteles Lagothrix Ateles Saimiri Saguinus Callimico Callithrix Cebuella Leontopithecus Dolichocebus gaimanensis Carlocebus carmeni Tremacebus harringtoni Soriacebus ameghinorum Branisella boliviana Catopithecus browni Proteopithecus sylviae 1000 changes
Fig. 21. Maximum-parsimony analysis of the cranial and dental matrix unconstrained by a molecular backbone.

indicated in Fig. 21. Thus, while the results are by no means identical, we feel secure in assuming that the cranial and dental data are sufciently reliable to support phylogenetic inferences for the fossils. Discussion Character evolution Three views offered in the literature as to the afnities of Dolichocebus relative to living platyrrhines are: (1) Dolichocebus is a relative of callitrichines (Kraglievich, 1951; Horovitz, 1999); (2) Dolichocebus is a sister taxon of Saimiri (Rosenberger, 1979a); and (3) Dolichocebus is a stem platyrrhine (Hershkovitz, 1982; Fleagle and Bown, 1983). Except for the study by Horovitz (1999), proponents of these views single out one character or a small suite of characters in support of their hypothesis. Our approach, following and amplifying

that of Horovitz and colleagues, has been to examine more comprehensively the cranial and dental characters. Before discussing our results, however, it is appropriate to consider the basis for previous claims about the afnities of Dolichocebus. The callitrichine hypothesis. All claims that Dolichocebus is a sister to extant Callitrichinae are based on dental evidence. The original basis for the claim that Dolichocebus is related to callitrichines is Kraglievichs (1951) suggestion that Dolichocebus shared with Callitrichinae the loss of the third molar. Hershkovitz (1970) noted that the pertinent region of the type skull was broken away but concluded that there probably were three molars. The new fossil material and reexamination of the type skull using CT-imaging show that Dolichocebus had three molars, not two as in stem callitrichines. The palate of the type skull is broken away in the pertinent area, but the maxillary sinus is continuous and extensive above the roots of M2 and must have continued behind this tooth, leaving ample room for an M3. Fleagle and Bown (1983) described an isolated M3 of Dolichocebus (MACN CH 357), and one of the mandibular specimens described above has three molars and the other has an M2 with an interproximal facet for M3. Horovitz (1999) suggested that Dolichocebus, along with late Oligocene Branisella and several early to middle Miocene platyrrhines, were sister taxa to extant Callitrichinae (see also Takai et al.s [2000] views about Branisella). She cited three characters in support of callitrichine afnities of Dolichocebus. The rst of these is a large P3 protoconid relative to the P4 protoconid (vs. subequal in size). This character cannot be evaluated based on the known Dolichocebus material, as recognized by Horovitz (1999). The second character is the absence of a P4 hypoconid (vs. present). Our examination of this character (number 129 in our matrix) does not support Horovitzs observation that extant callitrichines lack a P4 hypoconid. The P4 hypoconids are generally present in extant platyrrhines, but absent or cristiform in extant atelids. Like other nonatelid platyrrhines, Dolichocebus possesses a P4 hypoconid (MACN CH 898). The nal character cited by Horovitz is the absence of a P4 entoconid (vs. present). Our examination of this character (number 123 in our matrix) conrms Horovitzs observation that extant callitrichines have a lingual crest bordering the P4 talonid but lack an entoconid. However, the only known P4 of Dolichocebus (MACN CH 898) has a discrete P4 entoconid. The Saimiri/cebine hypothesis. It has been suggested that Dolichocebus is related to cebines (Cebus and Saimiri) and specically to Saimiri. Rosenberger proposed several synapomorphies that link cebines with Dolichocebus: (1) narrow nasal bones, an elongate interorbital process of the frontal, and vaulted frontals without a depressed glabella; and (2) an elongate (dolichocephalic) braincase. CT scans reveal that the precise location and size of the frontal and nasal bones cannot be determined due to the poor preservation of the interorbital region. Additionally, the glabellar region is missing from the type skull. Values of braincase shape (100 braincase length/braincase breadth) of Dolichocebus, Tremacebus, and some extant platyrrhines are given in Table 6. Using this index, the longer

348

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Table 6 Braincase shape in selected living and extinct platyrrhines Taxa Pitheciinae Callicebus torquatus lugens1 Male Female Callicebus moloch1 Male Female Callicebus personatus1 Male Female Aotinae Aotus trivirgatus2 (mixed sex) Cebinae Saimiri boliviensis3 Male Female Cebus albifrons cesarae4 Male Female Callitrichinae Cebuella pygmaea5 (mixed sex) Callithrix jacchus5 (mixed sex) Callithrix argentata5 (mixed sex) Saguinus fuscicollis5 (mixed sex) Saguinus mystax5 (mixed sex) Leontopithecus rosalia5 (mixed sex) Callimico goeldii5 (mixed sex) Atelidae Alouatta seniculus4 Male Female Early Miocene taxa Tremacebus harringtoni2 Dolichocebus gaimanensis2 Braincase length (n) Braincase breadth (n) Shape*

53.5 (41) 52.7 (31) 52.4 (24) 52.0 (32) 57.2 (11) 55.7 (9) 51.5 (10)

36.8 (40) 35.5 (31) 35.0 (23) 34.3 (31) 36.4 (11) 35.9 (10) 33.9 (10)

145 149 150 152 157 155 152

54.5 (11) 52.6 (7) 78.6 (18) 74.9 (10) 30.9 (62) 40.5 (23) 39.6 (45) 42.1 (161) 45.6 (28) 47.3 (10) 45.8 (11)

36.0 (11) 36.1 (7) 50.5 (18) 50.8 (10) 21.1 (60) 25.6 (24) 25.7 (45) 26.1 (164) 28.7 (28) 28.4 (13) 29.8 (10)

151 146 156 147 146 158 154 161 159 167 154

74.4 (7) 67.5 (7) 51.0 (1) 52.0 (1)

56.6 (7) 49.8 (7) 34.0 (1) 32.0 (1)

131 136 150 163

Data sources: 1 Hershkovitz (1990: 88, 92, 95); 2 Hershkovitz (1974: 30); 3 Hershkovitz (1984); 4 Hershkovitz (1949: 391); 5 Hershkovitz (1977: Appendix Table 1). * Shape 100 (brain case length/braincase breadth).

and narrower the braincase, the larger the index. The values in the table show that living cebines are not dolichocephalic and that Dolichocebus, if anything, resembles tamarins in braincase shape. Representative species of Saimiri and Cebus range from 147 to 151. This range is encompassed by that of marmosets and falls within the range of three species of Callicebus. Living cebines are not especially long-headed. The ratio for Dolichocebus is 163. However, we suspect that this value is inated given that the skull was mediolaterally compressed postmortem. Even taken as the true value, Dolichocebus resembles Saguinus and Leontopithecus more closely than either Saimiri or Cebus.

Proposed specic similarities shared between Dolichocebus and Saimiri are based on features of a lightly built masticatory system and the structure of orbit (Rosenberger, 1979a, b; Fleagle and Rosenberger, 1983; Rosenberger et al., 1990). Turning rst to the masticatory system, it is difcult to determine whether or not Dolichocebus did in fact have a lightly built masticatory system (Hershkovitz, 1982) and, if it did, whether this is a shared derived similarity with Saimiri. Rosenberger (1979a) based his claim on the following points that Saimiri and Dolichocebus are said to share: (1) gracile anterior and posterior roots of the zygomatic arch, suggesting a weak masseter muscle; (2) a slender pyramidal process of the palatine, suggesting weakly developed pterygoid muscles; (3) temporal line on parietal neither deeply etched nor rugose, suggesting a weak temporalis muscle; (4) a transversely shallow temporal fossa (i.e., the cross-sectional area of the temporal fossa is small); (5) entoglenoid fossa shallow and postglenoid process absent medially, suggesting reduced mediolateral component of force in the temporomandibular joint. With respect to rst point, the zygomatic arches, which serve as the origin for the masseter, are broken away, as are both roots of the zygomata bilaterally. Regarding the second point, the pterygoid plates, which provide the origin for the medial pterygoid musculature, together with the pyramidal processes that support them, are not preserved. (The palate is broken posteriorly so that even the M3s are absent). On the third and fourth points, the temporal lines (marking the origin of the temporalis muscle from the sides of the skull) are weak, as Rosenberger (1979a) described, but the temporal lines are higher on the skull (closer to the midline) suggesting either that the temporalis was more extensive or that the braincase was smaller. Moreover, in the absence of zygomatic arches, it is impossible to determine the cross-sectional area of the temporal fossa, because the arches dene its lateral border. The weak temporalis muscle markings (point four) do resemble Saimiri, but they also resemble all other small-bodied platyrrhines. On the last point, Rosenberger (1979a) correctly stated that Dolichocebus and Saimiri both have a shallow glenoid fossa, but this is true for most platyrrhines, especially all of the small-bodied taxa. In agreement with Rosenberger, the postglenoid process of Dolichocebus is quite small, as in Saimiri, but this similarity also extends to Aotus and callitrichines (see footnote 3). A strong postglenoid process is associated with posterior expansion of the articular surface of the temporomandibular joint in primates and ungulates that have a large transverse excursion of the mandible during the power stroke of mastication (Wall, 1998). Conversely, having a weak process may tell us less about the amount of muscular force engendered by the power stroke and more about the extent of transverse excursion during that stroke. The portions of the mandible that serve as insertions for the muscles of mastication are not preserved in any specimen. In summary, the extent of similarity between Dolichocebus and Saimiri has been overstated by reference to anatomical details that are not preserved. Similarities that can be documented in the masticatory apparatus of Dolichocebus and

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

349

Saimiri can also be observed in other small-bodied platyrrhines. Some of these similarities may have more to do with the extent of mandibular excursion and less to do with masticatory force per se. Several aspects of the orbital region also led Rosenberger (1979a) to suggests that Saimiri and Dolichocebus might be sister taxa. The most impressive features that have been suggested to link Saimiri and Dolichocebus are the narrow interorbital pillar and the interorbital fenestra.5 After careful preparation using a UV light to distinguish bone from matrix, Rosenberger (1979a) concluded that there was no bone between the orbits and inferred that an interorbital fenestra was present in life. CT scans reveal that, supercially, the interorbital region is poorly preserved. However, cross sections indicate that much of the interorbital region of Dolichocebus is composed of matrix rather than bone. They also suggest that Dolichocebus possessed a cupular sinus (as in Cebus, but not Saimiri) and that the interorbital region was probably narrow. The narrow interorbital region is indeed a possible derived similarity shared with Saimiri, as Rosenberger suggested. However, the CT scans neither conrm nor preclude the presence of an interorbital fenestra in Dolichocebus. Saimiri possesses a number of other anatomical peculiarities (probably derived features) of the cranium, such as a relatively horizontal nuchal plane and the absence of a maxillary sinus. Dolichocebus, in contrast, has a more obliquely oriented nuchal plane and a large maxillary sinus. While both of the latter features are probably symplesiomorphic for platyrrhines as a whole and do not exclude Dolichocebus from the ancestry of Saimiri, neither do they reinforce the impression that Dolichocebus was especially Saimiri-like. The orbits of Dolichocebus are less convergent than those of Saimiri and other noncallitrichine platyrrhines. The more divergent orbits of Dolichocebus are also seen in early Miocene Tremacebus and may be another trait of stem platyrrhines that persists in callitrichines.

conicting hypotheses is overtly cladistic and involves several assumptions: (1) We used outgroup comparisons to establish polarities of the characters under consideration. For this purpose we selected a variety of African late Eocene and early Oligocene taxa as outgroups. (2) Where molecular phylogenies are sufciently well corroborated, we used them to constrain our conclusions. (3) We eschewed character-weighting schemes. There seems to be no satisfactory reason for character weighting. For example, there is no reason to believe that loss of the hypocone on the premolars is any more (or less) likely to occur than loss of the same feature on the molars or loss of the third molars. Yet historically, phylogenetic assessment has placed greater reliance (weight) on loss of molar hypocones and third molars and given scant attention to presence or absence of premolar hypocones. However, there is simply no a priori reason for believing that some features are better indicators of phylogeny than others. (4) We did not assume that certain suites of cranial or dental features found together in a clade of living platyrrhines evolved together and simultaneously as a single adaptive pattern or developmental pattern. For example, it has been proposed that loss of the third molar and loss of molar hypocones, along with a number of features of postcranial anatomy and several life-history traits, are the result of phyletic dwarng and the constraints of small body size on diet, foraging patterns, and reproductive strategy in callitrichines (Ford, 1980). This hypothesis might lead us to assume that we had simply scored the same character several times. Three things that we know about the evolution of the Anthropoidea run counter to the expectations of this particular hypothesis. First, the ancestors of callitrichines may not have been larger than they are today, and therefore callitrichines may not be phyletic dwarfs. While the ancestry of callitrichines is too poorly known to be certain, many Eocene/Oligocene African stem anthropoids and some Miocene noncallitrichine Colombian platyrrhines were within the size range of living callitrichines. Second, callitrichine-sized Anthropoidea are not constrained to have this particular pattern of morphology and behavior. The platyrrhine Patasola from the Miocene of Colombia (Kay and Meldrum, 1997) falls within the size range of living callitrichines but has a hypocone and third molar. Third, several of the dental traits of this character complex are not always linked with small body size in platyrrhines. Lagonimico, another Miocene Colombian primate (Kay, 1994), has lost the hypocone and greatly reduced the third molar but is much larger than any living callitrichine. Conversely, Callimico goeldii is sister to Callithrix among living platyrrhines and falls within the size range of living callitrichines, but it retains (or has regained) the third molar and hypocone. Character evolution and the place of Dolichocebus within the Platyrrhini. A stem anthropoid is an extinct taxon that is

Broader phylogenetic inferences To reassess of the phylogenetic position of Dolichocebus, we undertook a broader phylogenetic analysis. As we have seen in the descriptive anatomy presented above, Dolichocebus presents a confusing mix of similarities to many extant and extinct platyrrhine taxa. Put another way, phylogenetic conclusions based upon a single character or on a few selected characters are very likely to be inaccurate. Overall measures of homoplasy, such as the consistency index and the retention index (e.g., Maddison and Maddison, 2001), indicate high levels of homoplasy within our data set of cranial and dental characters. Accordingly, a few characters can be found to support almost any phylogenetic hypothesis. Our choice among these
It should be added that even though possession of an interorbital fenestra is unusual in primates, it demonstrably does occur in at least one other extant primate that we have examineddsome specimens of the catarrhine Erythrocebus (e.g., Duke Lemur Center Osteology specimen 064) have a large fenestra. Thus, even this feature exhibits homoplasy.
5

350

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

more closely related to living Anthropoidea than to the haplorhine Tarsius, but one that is outside the crown-anthropoid clade, which includes all living anthropoids, their last common ancestor (LCA), and all extinct taxa that are descendants of that LCA. Our maximum-parsimony phylogeny (Figs. 20, 21) places late Eocene and early Oligocene African Proteopithecus as a stem anthropoid. Early Oligocene African Aegyptopithecus is a stem catarrhine. The analysis places the parapithecids Apidium and Simonsius as stem catarrhines. However, a more inclusive analysis with related, more primitive parapithecid taxa such as Qatrania suggests that parapithecids are a stem anthropoid clade, and it reveals that a number of the molar characters that link Apidium and Simonsius with catarrhines are homoplasies instead of synapomorphies (Kay et al., 1997; Ross et al., 1998; Kay et al., 2004c; Seiffert et al., 2004). The analysis is equivocal as to whether late Eocene African Catopithecus (Oligopithecidae) is a stem catarrhine or a stem anthropoid; for reasons given by Kay and Delson (2000), we consider Catopithecus to be a stem anthropoid. Our maximum-parsimony trees (Figs. 20, 21) suggest that Bolivian late Oligocene Branisella, Patagonian early Miocene Dolichocebus and Tremacebus, and Patagonian late early Miocene Carlocebus and Soriacebus are stem platyrrhines. Additionally, there is weak evidence for a stem platyrrhine clade of the Patagonian taxa comprising Dolichocebus, Tremacebus, Carlocebus, and Soriacebus. Latest early Miocene Homunculus, under study elsewhere, may also belong to such a clade (Kay et al., 2005). Likewise, early Miocene Chilecebus may be a stem platyrrhine (Flynn et al., 1995) but is unavailable to us for study. Table 7 summarizes the status of 41 cranial and dental features we consider most informative of the above phylogenetic relationships. Figure 22 is a simplied version of Figs. 20 and 21 for the purposes of our discussion. Before discussing these characters, it should be emphasized that all of the traits being discussed demonstrate examples of independent parallel acquisition under virtually every phylogenetic scenario. A measure of such parallelism is the consistency index (CI; (Maddison and Maddison, 2001). A CI of 1.0 indicates that no homoplasy has occurred. The characters in Table 7 have CIs ranging from 0.17 to 1.00. Homoplasy is present in nearly all informative characters in our most-parsimonious phylogeny. This result should come as no surprise, considering the pervasiveness of this phenomenon in dental, cranial, and postcranial characters (Sanchez-Villagra and Williams, 1998; Williams, 2007). The character states listed in Table 7 as the platyrrhine and catarrhine character states represent the reconstructed states of the crown platyrrhine and catarrhine nodes (nodes 1e4 in Fig. 22), that is, the states of the LCAs of the crown groups, as reconstructed from the parsimony analysis. We use the terms shared derived and derived in place of synapomorphy and apomorphy, respectively. The terms primitive and shared primitive are substituted for plesiomorphic or symplesiomorphic, respectively. We rst consider the overall pattern of anthropoid evolution based on a contrast between the character states of stem anthropoids and our hypothetical

reconstructions of the ancestral condition of crown catarrhine and platyrrhine LCAs (nodes 3 and 4 in Fig. 22). Secondly, we evaluate the evidence for whether Aegyptopithecus is a crown or a stem catarrhine (i.e., a branch of the lineage segment between nodes 1 and 3). Thirdly, we consider the evidence for whether Dolichocebus and the other late Oligocene and early Miocene New World taxa Branisella, Tremacebus, Carloccebus, and Soriacebus are crown or stem platyrrhines (a branch or branches along the lineage segment between nodes 1 and 2 in Fig. 22). Fourthly, we examine the evidence supporting the existence of distinct clades of stem platyrrhines (support for node 4 in Fig. 22). Stem anthropoids, crown catarrhines, and crown platyrrhines contrasted. A number of the craniodental characters of the stem anthropoids of late Eocene and early Oligocene of AfricadApidium, Simonsius, Proteopithecus, and Catopithecusddiffer from the crown anthropoid LCA (node 1) or from the morphotypes of the crown catarrhine or crown platyrrhine LCAs (nodes 2 and 3, respectively):  The arrangement of bones at pterion (Character 9) varies among anthropoids (Ashley-Montague, 1933). The frontal contacts the alisphenoid at pterion in Tarsius and all crown catarrhines, whereas in the crown platyrrhine LCA, the zygomatic contacts the frontal. Based on the Tarsius outgroup, we infer that alisphenoid-frontal contact is the primitive stem-anthropoid state, although the condition has not been established in any stem anthropoid (see also the discussion of this feature in Fleagle and Kay, 1987).  The infraorbital foramen of the maxillary bone is variably positioned in anthropoids relative to the maxillary teeth (Character 14). In Catopithecus, it is posteriorly positioned, whereas it is more anteriorly positioned in Apidium, Simonsius, and Proteopithecus. The foramen is positioned quite far forward relative to the maxillary cheek teeth, a derived condition in the crown platyrrhine LCA (Horovitz, 1999), whereas in the crown catarrhine LCA, the foramen is more posteriorly positioned.  The ascending wing of the premaxilla (Character 21) is also variably developed, being broad in Catopithecus (Simons and Rasmussen, 1996) but narrow in Simonsius and Proteopithecus (Simons, 2001). Crown platyrrhine and catarrhine LCAs have an anteroposteriorly reduced wing of the premaxilla. This apparently shared derived state of the two crown groups evolved independently, as evidenced by the broad ascending wing of the premaxilla in the stem catarrhine Aegyptopithecus (Fleagle and Kay, 1983).  In the stem anthropoids Proteopithecus, Catopithecus, and to a lesser degree Simonsius, the maxillary postcanine tooth rows diverge posteriorly (Character 25). This condition is also inferred in the crown platyrrhine LCA, whereas the crown catarrhine LCA is derived in having more parallel-sided tooth rows.  The stem anthropoid Proteopithecus has an abbreviated medial pterygoid plate, whereas Simonsius and the crown catarrhine LCA have broad medial pterygoid plates

Table 7 Selected cranial and dental character distributions Character Consistency index Character states in late Eocene African anthropoids ? Posteriorly positioned (Catopithecus); anteriorly positioned (Proteopithecus) Broad (Catopithecus); narrow (Proteopithecus) Diverge slightly Short medial plate (Proteopithecus); unknown (Catopithecus) Less than 55 (Apidium, Simonsius, Catopithecus, and Proteopithecus) Medially positioned Annular Extensively ossied Aegyptopithecus Character states of crown Catarrhini Frontal-alisphenoid contact Posteriorly positioned Dolichocebus Character states of crown Platyrrhini Zygomatic-parietal contact More anteriorly positioned

Character 9: Pterion Charcter 14: Position of infraorbital foramen Character 21: Breadth of ascending wing of the premaxilla Character 25: Divergence of palatal tooth rows Character 27: Size of medial pterygoid plate Character 43: Orbital convergence Character 57: Position of external carotid foramen Character 66: Ectotympanic Character 74: Extent of ossication of tentorium cerebelli Character 75: Cartmills canal Character 77: Relative brain size Character 80: Mandibular depth Character 81: Mandibular symphyseal fusion Character 82: Position of coronoid process Character 111: P2

0.75 0.29

Frontal-alisphenoid contact Posteriorly positioned

? More anteriorly positioned

0.50

Broad

Narrow

Narrow

Narrow

0.45 0.40

Diverge slightly Long medial plate

Diverge slightly Long medial plate

Diverge slightly ?

Diverge more widely Short medial plate R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

0.40

Greater than 65

Greater than 65

Less than 65

Greater than 65

0.67 0.50 0.20

Ventrally positioned Annular Minimally ossied

Ventrally positioned Tubular Minimally ossied

Medially positioned Annular Extensively ossied

Medially positioned Annular Extensively ossied

0.33 0.22 0.25 0.50 0.20 0.50

Open Unknown Shallow Unfused Well above condyle Second premolar present (Proteopithecus); absent (Catopithecus) P3e4 two-rooted P3 metaconid absent or trace Distolingual Transverse P4 talonid and trigonid lengths z equal P3 and P4 of similar height P3 not enlarged More elongate

Partially occluded Small Deep Fused Well above condyle Second premolar absent

Absent Larger Deep Fused At or slightly above condyle Second premolar absent

Open Larger ? ? ? Second premolar present

Open Larger Deep Fused At or slightly above condyle Second premolar present

Character 113: P3e4 roots Character 122: P3 metaconid Character 128: P3 protocristid Character 130: P4 protocristid Character 139: Premolar talonid Character 144: Relative heights of P3 vs. P4 Character 145: Size of P3 Character 146: P4 shape (MD/BL)

0.67 0.56 0.33 0.33 0.17 0.50 0.38 0.50

P3e4 two-rooted P3 metaconid absent or trace Distolingual Transverse P4 talonid elongate P3 projects above P4 P3 enlarged Short and broad

P3e4 two-rooted P3 metaconid absent or trace Distolingual Transverse P4 talonid elongate P3 projects above P4 P3 enlarged More elongate

P3e4 single-rooted Small P3 metaconid Distolingual Distolingual P4 talonid very short P3 and P4 of similar height P3 not enlarged Short and broad

P3e4 single-rooted Large P3 metaconid Transverse Transverse P4 talonid very short P3 and P4 of similar height P3 not enlarged Short and broad 351 (continued on next page)

Table 7 (continued) Character Consistency index Character states in late Eocene African anthropoids Two-rooted Narrow Trigonid higher than talonid Open lingually Protocristid runs towards metaconid Absent Present Small Moderate To base of trigonid Aegyptopithecus Character states of crown Catarrhini Two-rooted Narrow Trigonid and talonid heights z equal Closed lingually Runs into hypoexid Present Present, large Large Large To base of trigonid Dolichocebus Character states of crown Platyrrhini Single-rooted Absent Trigonid and talonid heights z equal Closed lingually Protocristid runs towards metaconid Absent

352

Character 152: M3 root number Character 162: M3 heel Character 164: M1e2 trigonid-to-talonid height Character 166: M1 trigonid lingual wall Character 169: Protocristid orientation Characters 172, 173: Wear facet X Character 175: Postentoconid sulcus Character 176: M1 hypoconulid Character 177: M2 hypoconulid Characters 183, 184: M1e2 cristid obliqua terminus Character 190: M1e2 distal fovea Character 196: M1 absolute size (estimate of body size) Character 213: Upper canine occlusion

0.33 0.57 0.40 0.50 0.50 1.00, 0.50 0.44 0.43 0.46 0.33, 0.25

Two-rooted Narrow Trigonid and talonid heights z equal Closed lingually Runs into hypoexid Present Present, large Large Large To base of trigonid

Two-rooted ? Trigonid and talonid heights z equal Closed lingually Protocristid runs towards metaconid Absent Present Small Small Extends onto posterior trigonid wall or reaches protocristid Present Small Upper canine occludes with P2

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Absent Absent Absent Extends onto posterior trigonid wall or reaches protocristid Absent Small Upper canine occludes with P2

0.40 0.25 0.50

Absent Very small Upper canine occludes with P2 (Proteopithecus); with P3 (Catopithecus) Two (Catopithecus); three (Proteopithecus) Three (Catopithecus); three (Proteopithecus) Hypocone absent Waist between protocone and paracone Strong

Present Large Upper canine occludes with P3

Present Large Upper canine occludes with P3

Character 217: Number of P3 roots Character 218: Number of P4 roots Character 228: P4 hypocone development Character 233: Prole of P3 distal crown margin Character 236: Development of P3e4 buccal cingulum

0.36 0.44 0.57 0.25 0.50

Three Three Small hypocone present Straight or convex distal crown margin Absent

Three Three Hypocone absent Straight or convex distal crown margin Absent

Two Two Hypocone enlarged Straight or convex distal crown margin Absent

Two Two Hypocone small or absent Straight or convex distal crown margin Absent

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

353

Stem & Crown Catarrhini

Crown Catarrhines

Stem & Crown Platyrrhini

Crown Platyrrhines

crown catarrhine node

Dolichocebus Tremacebus Carlocebus Soriacebus

crown platyrrhine node

4 ? Stem catarrhine taxa (including Aegyptopithecus) Branisella Stem platyrrhines

crown anthropoid node

Stem anthropoid taxa (includes Apidium, Simonsius,Catopithecus, Proteopithecus)

Fig. 22. Simplied cladogram illustrating the crown and stem platyrrhines. Labeled nodes are discussed in the text.

(Character 27). In the crown platyrrhine LCA, the medial pterygoid plate is extremely reduced to a low ridge or even absent. In the stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, the orbital convergence is less than 65 (Character 43), whereas in the crown platyrrhine and catarrhine LCAs, orbital convergence is less than 65 (Ross, 1995). In the stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, and in the crown platyrrhine LCA, the internal carotid artery enters the middle ear through a foramen positioned ventromedially on the bulla (Character 57), whereas in the crown catarrhine LCA, the canal is more medially positioned relative to bullar landmarks (Ross, 1994). The ectotympanic is annular in the stem anthropoids Apidium, Simonsius, Catopithecus, and Proteopithecus (Character 66), a condition retained in the crown platyrrhine LCA, whereas the crown catarrhine LCA is derived in having a laterally expanded, collar-shaped ectotympanic. In the stem anthropoids Simonsius, Catopithecus, and Proteopithecus the tentorium cerebelli is partially ossied (Character 74) (Kay et al., 2008). The crown platyrrhine LCA shares the stem anthropoid condition, whereas the derived condition in the crown catarrhine LCA is for the tentorium to be unossied.

 In the stem anthropoids Apidium and Simonsius, Cartmills canal (Character 75) is absent, whereas it is present in Catopithecus, Proteopithecus, and all crown platyrrhines (Kay et al., 2008). A derived condition in crown catarrhines is for the canal to be absent.  Overall brain size (endocranial volume) cannot be established in Catopithecus or Proteopithecus, but it is smalldin the strepsirrhine rangedin Simonsius (Character 77). The crown catarrhine and platyrrhine LCAs both have relatively large endocranial volumes compared with Simonsius, Tarsius, and extant strepsirrhines. This apparently shared derived state of the two crown groups evolved independently, as evidenced by the relatively small endocranial volume in the stem catarrhine Aegyptopithecus (Radinsky, 1977; Simons, 1993; Simons et al., 2007).  In the stem anthropoids Apidium, Simonsius, Catopithecus, and Proteopithecus, the mandibular corpora are shallow (Character 80). The catarrhine and platyrrhine LCAs both have deep mandibular corpora.  The mandibular symphysis is fused in Apidium and Simonsius but unfused in Catopithecus and Proteopithecus (Character 81). All catarrhines and platyrrhines have fused mandibular symphyses as adults (Ravosa and Hogue, 2003).  In the stem anthropoids Apidium, Catopithecus, and Proteopithecus, the tip of the coronoid process reaches high above the condyle (Character 82). In the crown catarrhine

354

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

and platyrrhine LCAs, the coronoid process is at or slightly below the level of the mandibular condyles. The stem anthropoids Apidium, Simonsius, and Proteopithecus retain three permanent premolars (Character 111). All crown platyrrhines share the primitive anthropoid three-premolar condition. Crown catarrhines and, convergently, Catopithecus show a reduction in the number of premolars from three to two. In the stem anthropoids Apidium, Simonsius, Catopithecus, and Proteopithecus the lower third and fourth premolars are two-rooted (Character 113). All crown catarrhines retain the primitive arrangement, but all crown platyrrhines have the derived single-rooted P3 and P4 (Kay, 1980). Stem anthropoids Apidium, Simonsius, Catopithecus, and Proteopithecus have a small or absent P3 metaconid (Character 122). Crown catarrhines also have a reduced P3 metaconid. This condition appears not to be simply the result of the loss of the second premolar and attendant crown simplication in the latter because, as noted, Proteopithecus, which possesses three premolars, also has a simple P3 without the metaconid. The crown platyrrhine LCA is derived in having a large P3 metaconid (Kay, 1980). Stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus and the crown catarrhine LCA have distolingually oriented P3 protocristids (Character 128). In contrast, the crown platyrrhine LCA is derived in having a transversely oriented P3 protocristid (Kay, 1980). The stem anthropoids Proteopithecus and Catopithecus have transversely oriented P4 protocristids, whereas Apidium and Simonsius have more distolingually oriented P4 protocristids (Character 130). The crown catarrhine and platyrrhine LCAs have transversely oriented P4 protocristids like Proteopithecus and Catopithecus. In the stem anthropoids Proteopithecus and Catopithecus, the ratios of P4 talonid to trigonid mesiodistal lengths are approximately equal (Character 139). The crown catarrhine LCA has an elongate talonid and short trigonid. The crown platyrrhine LCA has a mesiodistally short P4 talonid. Thus, crown catarrhines, on one hand, and crown platyrrhines, on the other hand, have attained the two different derived states compared with the primitive condition seen in Catopithecus and Proteopithecus. In the stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, the overall size and projective heights of P3 and P4 are similar (Characters 144, 145). The crown platyrrhine LCA shares these stem anthropoid conditions, whereas the crown catarrhine LCA is derived in having an enlarged P3, and its prototoconid projects well above that of P4. The stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus have mesiodistally elongate P4s (Character 146), as does the crown catarrhine LCA. In contrast, the crown platyrrhine LCA is derived in having a mesiodistally short and buccolingually broad P4. In the stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, as well as the crown catarrhine

LCA, M3 has two roots (Character 152) and M3 crowns have a third lobe featuring a hypoconulid (Character 162). The crown platyrrhine LCA is derived in having the M3 roots reduced to one, often with a gure-eight cross section, and the M3 hypoconulid and third lobe are absent (Kay, 1980). The stem anthropoids Proteopithecus and Catopithecus exhibit the presumed anthropoid primitive condition of having the molar trigonids raised above the talonids (Character 164). The crown catarrhine and platyrrhine LCAs share the derived condition in which the M1 and M2 trigonids are not or are only slightly raised above the talonids. The parapithecids Apidium and Simonsius are derived in having the trigonid and talonid of similar heights; however, more primitive parapithecids such as Qatrania exhibit the stemanthropoid condition, demonstrating that this character evolved homoplastically in parapithecids and crown catarrhines. In the stem anthropoids Catopithecus and Proteopithecus, the protocristids of M1 and M2 run towards the metaconids, and M1e2 lack Phase II wear facet X (Characters 169, 172, 173). The crown catarrhine LCA exhibits a derived condition of having M1e2 protocristids deected distally into the hypoexid and in having a distinct M1e2 Phase II wear facet X. The crown platyrrhine LCA retains the primitive condition of these characters. The parapithecids Apidium and Simonsius have M1e2 protocristids deected distally into the hypoexid and a distinct M2 Phase II wear facet X (M2 only). More primitive parapithecids such as Qatrania exhibit the stem-anthropoid conditions, demonstrating that these characters evolved homoplastically in parapithecids and crown catarrhines. Stem anthropoids Simonsius, Proteopithecus, and Catopithecus have a weak but distinct M1e2 postentoconid sulcus. The crown catarrhine LCA is derived in having broad, open sulci between the entoconids and hypoconulids of M1e2 (Character 175). Additionally, M1 and M2 have a pronounced fovea distal to the entoconid for receipt of the enlarged hypocone (Character 190). The postentoconid sulcus and distal fovea are absentdderived conditions in the crown platyrrhine LCA (Kay, 1980). The parapithecid Apidium, in parallel with basal catarrhines, has a broad, open postentoconid sulcus and distal fovea, but these traits are not present in Simonsius or primitive parapithecids, and thus we infer the resemblance between Apidium and crown catarrhines to be homoplastic. Proteopithecus and Catopithecus have M1e2 hypoconulids of moderate size (Characters 176, 177). The crown catarrhine LCA has large M1e2 hypoconulids. A derived feature of the crown platyrrhine node is the absence of M1e2 hypoconulids (Kay, 1980). The parapithecid Apidium, in parallel with basal catarrhines, has large hypoconulids. In stem anthropoids and the crown catarrhine LCA, the M1e2 cristid obliquas run to the base of the posterior trigonid wall (Characters 183, 184). The crown platyrrhine LCA exhibits a derived condition in having the M1e2

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

355

cristid obliquas extending onto the posterior wall of the trigonid reaching towards the protoconid (Kay, 1980). Using M1 size as a surrogate for body size (Character 196), the stem anthropoids Proteopithecus and Catopithecus were very small animals compared with all crown catarrhines and smaller than the crown platyrrhine LCA. The stem anthropoids Apidium, Simonsius, and Proteopithecus share with crown platyrrhines an occlusal arrangement whereby the upper canine occludes with P2 (Character 213). Crown catarrhines are derived in having the upper canine in occlusion with P3. Catopithecus shares the C-P3 occlusal arrangement with catarrhines, one of a very few possible shared derived cranial or dental features that could support a link between oligopithecids and catarrhines (Simons and Rasmussen, 1996). The stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus have three-rooted P3e4sdtwo buccal roots and one lingual root (Characters 217, 218). This primitive condition is present in the crown catarrhine LCA. The crown platyrrhine LCA is derived in having two-rooted P3s and P4sdone buccal and one lingual root (Kay, 1980). In stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, as well as the crown platyrrhine and catarrhine LCAs, a P4 hypocone is absent or very weakly developed (Character 228).

condition, demonstrating that this character evolved homoplastically in parapithecids and crown anthropoids.  The P3s of the stem anthropoids Proteopithecus and Catopithecus have concave, waisted distal crown margins (Character 233). Crown anthropoids have P3 crowns with straight or slightly convex distal margins. The parapithecids Apidium and Simonsius homoplastically have P3 crowns with straight or slightly convex distal margins.  In the stem anthropoids Proteopithecus and Catopithecus, the buccal cingulum of P3e4 is well developed (Character 236). In the crown catarrhine and crown platyrrhine LCAs, the buccal cingulum is reduced or absent. The parapithecids Apidium and Simonsius homoplastically have weakly developed buccal cingula. Aegyptopithecus is a stem catarrhine. Aegyptopithecus exhibits an interesting mixture of crown-catarrhine traits and primitive retentions suggesting that it is a stem catarrhine. Aegyptopithecus and the crown catarrhine LCA (node 3 in Fig. 22) share thirteen derived character states relative to stem anthropoids and the crown anthropoid LCA:  Aegyptopithecus and the crown catarrhine LCA have parallel-sided tooth rows (Character 25). In stem anthropoids, stem platyrrhines, and the crown platyrrhine LCA, the postcanine tooth rows diverge posteriorly.  Aegyptopithecus and the crown catarrhine LCA have pronounced medial pterygoid plates (Character 27). Stem anthropoids, stem platyrrhines, and the crown platyrrhine LCA have an abbreviated medial pterygoid plate. The condition is unknown in Dolichocebus. Tremacebus has a reduced medial pterygoid plate. (Examination of micro-CTs of the skull of Tremacebus shows that an apparently enlarged medial pterygoid plate is reconstructed from plaster.)  In Aegyptopithecus and the crown catarrhine LCA, the carotid foramen is positioned ventrally on the bulla (Character 57), whereas the canal enters ventromedially in stem anthropoids, the stem platyrrhines Dolichocebus and Tremacebus, and nodal crown platyrrhines.  In Aegyptopithecus and the crown-catarrhine node, the tentorium cerebelli has an unossied petrosal root (Character 74), whereas in stem anthropoids, the tentorium is partially ossied (Kay et al., 2008). The stem platyrrhines Dolichocebus and Tremacebus and the crown platyrrhine LCA share the stem-anthropoid condition.  Aegyptopithecus, crown catarrhines, and, convergently, Catopithecus have a reduced number of premolars, from three to two (Character 111). Stem anthropoids, stem platyrrhines, and crown platyrrhines retain P2.  Aegyptopithecus and crown catarrhines have an enlarged P3 relative to P4, with its prototoconid projecting well above that of P4 (Characters 144, 145). Stem anthropoids, including oligopithecids with two premolars, are more primitive in having similar projective heights of the P3 and P4 protoconids, with P3 not enlarged relative to P4. Branisella, Dolichocebus, Carlocebus, and the

Crown-anthropoid features. A number of derived character states of the crown Anthropoidea LCA have been identied in our analysis (node 1 in Fig. 22). These features are found in taxa identied in our analysis as stem catarrhines (Aegyptopithecus) and platyrrhines (Branisella, Dolichocebus, Soriacebus, and Carlocebus) and in the crown catarrhine and platyrrhine LCAs:  All crown anthropoids have a fused mandibular symphysis and most have deep mandibular corpora. The stem anthropoids Proteopithecus and Catopithecus have unfused and shallow mandibles (Characters 80, 81). The late Oligocene Branisella has a shallow mandible. This suggests that deep mandibles may have evolved independently in the crown catarrhines and platyrrhine LCAs.  In the crown anthropoid LCA, the coronoid process of the mandible is close to the level of the mandibular condyle, whereas in stem anthropoids, the tip of the coronoid process projects high above the mandibular condyle (Character 82).  The crown anthropoid LCA has M1e2 trigonids and talonids of similar height (Character 164), whereas stem anthropoids have the molar trigonids raised above talonids.  In the stem anthropoids Proteopithecus and Catopithecus, the M1 trigonids are open lingually (Character 166). The crown platyrrhine and catarrhine LCAs have the M1 trigonid enclosed lingually by a premetacristid. The parapithecids Apidium and Simonsius are derived in having the M1 trigonid enclosed lingually; however, more primitive parapithecids such as Qatrania exhibit the stem anthropoid

356

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

crown platyrrhine LCA share these stem-anthropoid conditions. Aegyptopithecus and all crown catarrhines exhibit a derived condition of having M1e2 protocristids deected distally into the hypoexid and in having a distinct M1e2 Phase II wear facet X (Characters 169, 172, 173). In stem anthropoids, stem platyrrhines (Dolichocebus, Branisella, Carlocebus, and Soriacebus), and the crown platyrrhine LCA, the M1e2 protocristids run towards the metaconids and M1e2 Phase II wear facet X is absent. Aegyptopithecus and the crown catarrhine LCA share the derived condition in having broad, open sulci between the M1e2 entoconids and hypoconulids (Character 175). Stem anthropoids and the stem platyrrhines Branisella, Dolichocebus, Carlocebus, and Soriacebus have weaker and less distinct postentoconid sulci on M1e2. The latter is a more primitive condition than that inferred for the crown platyrrhine LCA. At the latter node, the postentoconid sulcus is absent. Aegyptopithecus and its smaller relatives Propliopithecus and Moeropithecus are similar to the crown catarrhine LCA in being larger in body size (Character 196). In Aegyptopithecus and crown catarrhines, the upper canine occludes with P3 (Character 213). In most stem anthropoids, the stem platyrrhines Branisella, Carlocebus, Dolichocebus, and Soriacebus, and all crown platyrrhines, the upper canine occludes with P2.

of the last common ancestor of living platyrrhines (the lineage segment between nodes 1 and 2 in Fig. 22). This hypothesis gains support from the presence in these taxa of retained stem-anthropoid character states found in combination with some derived crown-platyrrhine character states. Early Miocene platyrrhines share six derived features with the crown platyrrhine LCA.  The stem platyrrhines Branisella, Dolichocebus, Carlocebus, and Soriacebus share the derived platyrrhine condition of having single-rooted P3s and P4s (Character 113).  The mesiodistally short, buccolingually broad P4 of the crown platyrrhine LCA (Character 146) is also found in Dolichocebus, Soriacebus, and Branisella.  The crown platyrrhine LCA had the derived condition where the M1e2 cristid obliqua extends onto the posterior wall of the trigonid, reaching towards the protoconid (Characters 183, 184). Dolichocebus, Branisella, Carlocebus, and Soriacebus share this derived condition.  The crown platyrrhine LCA has two-rooted P3s and P4s (Characters 217, 218). Branisella, Carlocebus, Tremacebus, and Dolichocebus also have the typical platyrrhine two-rooted third and fourth premolars. Soriacebus has three-rooted P3s and P4s. At the same time, late Oligocene and early Miocene platyrrhines depart from the crown platyrrhine LCA, and are more like stem anthropoids, in retaining nine primitive character states:  In the crown platyrrhine LCA, the zygomatic and temporal bones contact at pterion, covering the contact between the the frontal and the alisphenoid (Character 9). The arrangement of these bones cannot be determined for Dolichocebus, but in Tremacebus, based upon our examination of the type skull, and reportedly in another latest middle Miocene platyrrhine, Homunculus (Tauber, 1991), conrmed by personal observations, the primitive frontal-alisphenoid contact is retained.  A derived condition in the crown platyrrhine LCA is for the infraorbital foramen to be positioned quite far forward relative to the cheek teeth (Character 14). Dolichocebus shares the derived crown-platyrrhine condition. However Tremacebus and Branisella have a more posteriorly positioned infraorbital foramen.  In stem anthropoids Apidium, Simonsius, Proteopithecus, and Catopithecus, the orbital convergence is less than 65 (Character 43), whereas in the crown platyrrhine and catarrhine LCAs, the orbital convergence is less than 65 (Ross, 1995).  Overall brain size (endocranial volume) cannot be established for Catopithecus or Proteopithecus, but it is smalldin the strepsirrhine rangedin Simonsius (Character 77). The crown catarrhine and platyrrhine LCAs both have relatively large endocranial volumes, but as noted, this apparently shared derived state of the two crown groups evolved independently, as evidenced by the relatively small

We identify three craniodental features where Aegyptopithecus still retains the stem anthropoid condition, suggesting that it is a stem rather than a crown catarrhine:  All crown catarrhines have expanded, collar-shaped ectotympanics (Character 66), whereas the ectotympanic is annular in stem anthropoids, stem platyrrhines, and all crown platyrrhines. Aegyptopithecus retains the primitive annular condition.  In crown catarrhines, Cartmills venous canal is absent (Character 75). The stem catarrhine Aegyptopithecus has an intermediate condition in which the canal is blind-ended or incomplete (Kay et al., 2008). In stem anthropoids, stem platyrrhines, and all crown platyrrhines (Kay et al., 2008) the canal is open. This is another case where Aegyptopithecus is less derived than crown catarrhines, consistent with it being a stem catarrhine.  The crown catarrhine and platyrrhine LCAs both have relatively large endocranial volumes (Character 77), but as noted, this is clearly an independent evolutionary event because the stem catarrhine Aegyptopithecus has a comparatively small endocranial volume (Simons et al., 2007). Branisella, Dolichocebus, Tremacebus, Carlocebus, and Soriacebus are stem platyrrhines. A number of cranial and dental characters suggest that late Oligocene Branisella and early Miocene Dolichocebus, Tremacebus, Carlocebus, and Soriacebus branched from the platyrrhine stem before the appearance

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

357

endocranial volume in the stem catarrhine Aegyptopithecus. There also is evidence that the stem platyrrhine Homunculus also had a relatively small endocranial volume (Kay et al., 2006). The derived enlarged P3 metaconid of the crown platyrrhine LCA (Character 122) is not observed in Dolichocebus, Branisella, Carlocebus, or Soriacebus, which have small but distinct P3 metaconidsdintermediate between the stem anthropoid condition and the derived condition of the crown platyrrhine LCA. The derived transversely oriented protocristid on the P3 of crown platyrrhines (Character 128) is not seen in Dolichocebus, Carlocebus, or Soriacebus, which have distolingually oriented P3 protocristids. However, Branisella has independently gained (or a Dolichocebus, Carlocebus, and Soriacebus clade has lost) the derived platyrrhine condition. The derived condition of the crown platyrrhine LCA of possessing a single-rooted M3 without a third lobe (Characters 152, 162) is not seen in stem platyrrhines. Dolichocebus, Carlocebus, and Soriacebus have a single M3 root with a gure-eight cross section. A small M3 third lobe with a small hypoconulid is present in Branisella, Carlocebus, and Soriacebus. The surface anatomy of the M3 cannot be evaluated in the available Dolichocebus specimens. The postentoconid sulcus is derivedly absent in the crown platyrrhine LCA (Character 175). The stem platyrrhines Branisella, Dolichocebus, Carlocebus, and Soriacebus still retain a distinct M1e2 postentoconid sulcus. The derived absence of an M1e2 hypoconulid in the crown platyrrhine LCA (Characters 176, 177) differs from a more primitive condition in stem platyrrhines. Branisella has a hypoconulid comparable in size to those of stem anthropoids. Dolichocebus, Carlocebus, and Soriacebus possess greatly reduced hypoconulidsdmore derived than in Branisella but less advanced than in crown platyrrhines.

Two additional features suggest that the Patagonian taxa form their own clade:  The crown anthropoid LCA has marginally placed premolar cusps (Character 137), as do Branisella and the crown platyrrhine and catarrhine LCAs. In contrast, Dolichocebus, Carlocebus, and Soriacebus have basally inated P4s.  Carlocebus, Dolichocebus, and Soriacebus are derived in having strongly developed P4 hypocones. By contrast, crown platyrrhines, stem anthropoids, and stem and crown catarrhines lack or have very weakly developed P4 hypocones (Character 228). Branisella likewise lacks a P4 hypocone. Contrasting views of platyrrhine evolution The implications of the phylogenetic analyses are obvious with respect to the choice between the deep time versus layered hypotheses of platyrrhine evolution. Our examination of the anatomy of Dolichocebus and our broader phylogenetic analysis supports the hypothesis that Branisella, Dolichocebus, Tremacebus, Carlocebus, and Soriacebus are stem platyrrhines. Branisella exhibits some dental similarities with callitrichines (Takai et al., 2000) and other crown platyrrhines, but it also possesses a number of similarities with stem anthropoids (e.g., the large, lingually placed hypoconulid). Autapomorphies of the dentition (high-crowned teeth that were worn rapidly) suggest that Branisella occupied a possibly terrestrial or scansorial adaptive niche, which are not found among living platyrrhines (Kay et al., 2001). Dolichocebus is not specially related to Saimiri, nor is it a cebine. Indeed, the proposed similarities between Dolichocebus and cebines either do not exist (e.g., dolichocephaly; see above) or are not established by the fossils (e.g., interorbital fenestration and the structure of glabella; the structure of the nasals and interorbital process of the frontal cannot be determined for Dolichocebus). Tremacebus, other than having somewhat enlarged orbits, shows few derived similarities to Aotusdthe orbits are divergent, as as in basal cebids, the olfactory lobes are not as enlarged as in Aotus, the palate is shallow (vs. deep), and the tooth rows diverge posteriorly (Hershkovitz, 1974; Kay et al., 2004a). Soriacebus exhibits dental and gnathic similarities to the most specialized pitheciines (mesiodistally compressed and procumbent incisors, robust canines, lowcrowned molars with poorly developed shearing crests, and a deep jaw with an apparently expanded angle), but many details of the dental structure are more primitive, and there are a number of crossing-specializations (Simpson, 1961: 102) that demonstrate an independent entry into the seed-predation niche currently occupied by extant Cebus and the Pitheciinae (Kay, 1990; Meldrum and Kay, 1997; Kay et al., 1998a). Thus, all four of these late Oligocene and early Miocene taxa possess a combination of plesiomorphic character states that more closely resemble the nearest platyrrhine outgroupsdAfrican late Eocene Catopithecus and Proteopithecus and Oligocene-to-Recent catarrhinesdwhile showing distinctive platyrrhine synapomorphies. Such a combination of primitive traits and platyrrhine

Evidence for cladogenesis among stem platyrrhines. Among the taxa examined here, there is some evidence, although not as yet compelling, that Branisella branched rst from the stem-platyrrhine lineage segment, followed by a southern or Patagonian clade consisting of Dolichocebus, Tremacebus, Carlocebus, and Soriacebus. Two features suggest that Branisella was rst to branch from the stem-platyrrhine lineage segment:  Dolichocebus, Carlocebus, and Soriacebus are derived in having distolingually oriented P4 protocristids (Character 130). By contast, the stem anthropoids Proteopithecus and Catopithecus, the stem platyrrhine Branisella, and the crown platyrrhine LCA have transversely oriented protocristids on P4.  Dolichocebus, Carlocebus, and Soriacebus possess the derived mesiodistally reduced P4 talonid of the crown platyrrhine LCA (Character 139). However, Branisella is more primitive in retaining the primitive elongate P4 talonid.

358

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

synapomorphies is precisely what would be expected in a cluster of stem taxa. The deep time hypothesis posits that most or all of the platyrrhine family clades extend back at least 20 million years or earlier. However, this hypothesis conates adaptation with phylogeny. For example, it is argued that nocturnality is such an unusual adaptive pattern for an anthropoid that it could only have evolved once in platyrrhine evolution. Tremacebus, like Aotus, has enlarged orbits and must have been nocturnal. Therefore, according to the deep time hypothesis, Tremacebus must be an aotine. Likewise, it is argued that seed predation and seed-pod husking is such an unusual adaptive pattern that it could only have evolved once in platyrrhine evolution. Thus, because Soriacebus, like pitheciines, is adapted to seed-husking and seed predation, it must be a pitheciine. Rosenberger (2002) framed the debate in terms of parsimony. He rejected the stem-group or separate-clade phylogenies, stating that [these hypotheses] require a high degree of parallel or convergent evolution (Rosenberger, 2002: 158). We agree. The layered hypothesis supported here does require substantial homoplasy, but not as much as the deep time hypothesis, which is why we prefer our interpretation over hisdboth entail a substantial amount of homoplasy, but the layered hypothesis calls for less. That so much parallel evolution could have occurred in the platyrrhine clade is quite plausible based upon what we know about primate evolution generally. Homoplasy is a very common phenomenon in primate evolution (Lockwood and Fleagle, 1999). We nd numerous examples of multiple independent occupations of adaptive niches among primates (Heesy and Ross, 2004). Eocene North American Omomyidae show repeated independent acquisition of enlarged lower central incisors as food-acquisition tools (Williams and Covert, 1994). Nocturnality and enlarged orbits occurred independently among several Malagasy strepsirrhines, as well as in Eocene omomyids and adapids. Indeed, among the extant platyrrhines, prehensile tails evolved twicedin atelids and Cebus (Rosenberger, 1983)dand so did seed predation (Cebus and pitheciines). Moreover, folivory evolved at least twice, once in Brachyteles and once in Alouatta. Why then should it be difcult to accept the proposition that seed predation occurred independently in pitheciines and Soriacebus or that nocturnality, or at least crepuscularity, evolved independently in Aotus and Tremacebus? Our phylogenetic analysis suggests that some extinct clades of platrrhines independently occupied similar adaptive niches to those occupied by crown platyrrhines. As Hershkovitz (1977) observed, there was no single adaptive radiation of platyrrhines. What we see today is a layered pattern of specializations, the consequence of independent evolution at different times over a long platyrrhine career on the South American continent. Molecular-branch times for the living platyrrhine families support the layered model. Pitheciine seed predators came on the scene at least three million years later than the nocturnal ancestors of Aotus and at least six million years earlier than the origin of gum feeding in Callithrix and Cebuella (Schneider et al., 2001). The fossil evidence also suggests that extinct platyrrhines may have occupied niche

space not present in the living representatives of the clade. For example, there is suggestive evidence that Branisella was scansorial (see above). Acrecebus, a poorly known cebine from the late Miocene of the Amazon Basin, may have attained the size of a small chimpanzee (Kay and Cozzuol, 2006). Summary and conclusions Dolichocebus gaimanensis comes from deposits of the Sarmiento Formation in Chubut Province, Argentina, near the town of Gaiman. The geology of the Sarmiento Formation in the type locality was reviewed and a scheme of subdivisions of the formation presented for that place. All specimens of Dolichocebus come from the Trelew Member. This primate and its associated mammalian fauna date to the early Miocenedabout 20 Madand are attributed to the Colhuehuapian South American Land Mammal Age. All Dolichocebus fossil material was described using conventional surface anatomy and micro-CT data from the cranium. The new material and newly imaged internal anatomy of the skull demonstrate that most anatomical characters hitherto supposed to support a phyletic link between Dolichocebus and either callitrichines or squirrel monkeys are either indeterminate or absent. A set of phylogenetic analyses using parsimony was undertaken on Dolichocebus, several fossil platyrrhines, all living platyrrhine genera, and a suite of better-known living and fossil anthropoid outgroups. All parsimony results point to the conclusion that Dolichocebus is a stem platyrrhine, as are late Oligocene Branisella and early Miocene Tremacebus, Soriacebus, and Carlocebus. The outcome of these analyses also highlights the pervasiveness of homoplasy in dental and cranial characters. This is not surprising because large amounts of homoplasy are present in previous analyses at all levels of biological organization, from DNA to aspects of adult bony morphology and behavior. Platyrrhine evolution often is conceived of as a single, ancient adaptive radiation. Our phylogenetic investigation suggests a very different pattern, wherein several extinct platyrrhine clades independently occupied adaptive niches similar to those occupied by crown platyrrhines. Simply stated, there was no single adaptive radiation of platyrrhines. What we see in todays platyrrhine assemblage is the latest in a series of radiations over a long platyrrhine career in the Neotropics. Review of our own and other available phyolgenetic data suggests a more layered evolutionary pattern. Several independent extinct clades lled modern platyrrhine niche space in the Oligocene and early Miocene. Modern platyrrhine families and subfamilies appear subsequently over a nine-million-year interval in the Miocene. Acknowledgements We beneted greatly from discussions with Chris Wall, James Rossie, Walter Hartwig, and input from the editor and several unidentied reviewers. We thank the curators at the

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

359

Museo Nacional de Ciencias Naturales, Buenos Aires, and the Museo de La Plata for access to material, and we acknowledge the contribution of the personnel from these museums to various eld projects over the past 25 years. RFK especially thanks Richard Madden, Guiomar Vucetich, and Alfredo Carlini for help and inspiration in the laboratory and on the outcrop. This research was supported by grants to JGF and RFK by the National Science Foundation and the National Geographic Society. The most recent grant is NSF BCS Award 0090255 to RFK.

inferior orbital ssure does not reach the zygomatic arch; 1 the ventrolateral limit of the inferior orbital ssure reaches the zygomatic arch. 9. (Unordered; weight 100). Zygomaticoparietal contact at pterion (Kay et al. [2004c] cranial character 46): 0 no postorbital closure; 1 zygomatic-parietal contact; 2 alisphenoid-frontal contact. Lacrimal region 10. (Weight 100). Position of lacrimal foramen (Kay et al. [2004c] cranial character 30): 0 outside orbital margin; 1 within the orbit or on the rim. 11. * (Weight 50). Extraorbital exposure of the lacrimal: 0 lacrimal fossa is completed anteriorly by maxilla; 1 lacrimal has some facial exposure; 2 lacrimal contacts nasal (excludes maxilla-frontal contact). 12. (Weight 100). Zygomatic-lacrimal contact (Kay et al. [2004c] cranial character 26): 0 present on ventral orbital rim; 1 absent on ventral orbital rim. 13. (Unordered; weight 100). Contact between lacrimal and palatine bones (Kay et al. [2004c] cranial character 28): 0 contact present; 1 lacrimal and palatine separated, contact between frontal and maxilla (or in some taxa, by a small os planum of the ethmoid); 2 separated by a large os planum. Facial region 14. * (Weight 50). Position of the infraorbital foramen relative to the Frankfurt horizontal plane (Horovitz [1999] character 26): 0 posterior to P4; 1 positioned above P3e4; 2 positioned above P2. 15. * (Weight 50). Angle of cranial kyphosis (Lieberman et al., 2000): 0 140 ; 1 >140 ; <155 degrees; 2 !155 . 16. (Weight 100). Nasal fossa width (Horovitz [1999] character 25): 0 narrower than the palate width; 1 broader than the palate width. 17. (Weight 100). Nasal capsule (Maier, 1980): 0 processus alaris superior present; 1 processus alaris superior absent. 18. (Weight 100). Snout length (Kay et al. [2004c] cranial character 37): 0 long snout; 1 short snout. 19. (Weight 100). Maxilla depth (Kay et al. [2004c] cranial character 38): 0 deep; 1 shallow. 20. (Weight 100). Interincisor diastema width (Kay et al. [2004c] cranial character 42): 0 broad and wider than that of extant haplorhines; 1 narrow, haplorhine-like. 21. (Weight 100). Ascending wing of premaxilla (Kay et al. [2004c] cranial character 49): 0 narrow; 1 broad. Temporomandibular region 22. * (Weight 50). Postglenoid foramen (Horovitz [1999] character 12): 0 absent; 1 small; 2 large. 23. (Weight 100). Temporomandibular joint morphology (Kay et al. [2004c] cranial character 40): 0 biconcave

Appendix 1: Cranial and dental characters and character states used in the phylogentic analyses Characters followed by an asterisk (*) are ordered. Ordered characters are scaled by the number of character states, such that the sum of the steps in the morphocline equals 100. This weight is indicated as weight number. A score of 9 is used if information is unavailable due to a lack of material or if the character does not apply to a particular taxon. Cranial characters adopted from Horovitz (1999) and Kay et al. (2004c) are so designated. Many of these cranial characters ultimately are drawn from the following sources: ForsythMajor (1901), Pocock (1925), Hill (1957), Le Gros Clark (1959), Du Brul (1965), Hershkovitz (1974, 1977), Cartmill (1978), Rosenberger (1979b), Conroy (1981), MacPhee and Cartmill (1986), Ross (1993, 1994, 1998), Kay (1994), MacPhee et al. (1995), Horovitz (1997, 1999), Kay et al. (1997, 2004c), Horovitz and MacPhee (1999), Kay and Kirk (2000), and Lieberman et al. (2000). Dental characters designated with an a sufx are new characters. Other dental characters are as in previous analyses (Kay et al., 2004c). Cranial characters Paranasal sinuses 1. (Weight 100). Ethmofrontal sinus (Type II): 0 present; 1 absent. 2. (Weight 100). Splenofrontal sinus (Type I): 0 present; 1 absent. 3. (Weight 100). Maxillary sinus: 0 present; 1 absent. 4. (Weight 100). Anterior ethmiodal sinus: 0 present; 1 absent. 5. (Weight 100). Sphenoidal sinus: 0 present; 1 absent. Zygomatic region 6. * (Weight 50). Zygomaticofacial foramen (Horovitz [1999] character 27): 0 small relative to M1 breadth; 1 large relative to M1 breadth; 2 very large relative to M1 breadth. 7. (Weight 100). Zygomatic arch position (Horovitz [1999] character 23): 0 above the alveolar border of the maxilla; 1 below the alveolar border. 8. (Weight 100). Extent of inferior orbital ssure (Horovitz [1999] character 84): 0 ventrolateral limit of the

360

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

and transversely wide; 1 anteroposteriorly oriented trough. 24. * (Weight 50). Postglenoid process size (100 postglenoid process length/prosthioneinion length) (Kay et al. [2004c] cranial character 41): 0 weak or absent (<0.39); 1 strong (!0.39, <0.69); 2 very strong (!0.69). Pterygoid and palatal region 25. * (Weight 50). Palate shape: 0 V-shaped (the distance between lingual surfaces of the upper canines divided by the distance between the lingual surfaces of the upper second molars is <0.39); 1 intermediate (ratio values of !0.39, 0.64); 2 approaches parallel (ratio values > 0.64). 26. (Weight 100). Interpterygoid fossa (Du Brul, 1965): 0 deep; 1 shallow. 27. * (Weight 50). Length of medial pterygoid plate (Kay et al. [2004c] cranial character 36): 0 long medial pterygoid plate extending one-third to one-half of the distance to the anterior surface of the bulla enclosing a large fossa between medial and lateral pterygoids; 1 short but distinct from lateral pterygoid plate for its entire dorsoventral extentdventrally there is a hamular process, and more dorsally the plate merges with the lateral plate or if distinct, the fossa is slitlike; 2 medial pterygoid plate entirely absent or reduced to a low rugositydonly the hamulus is present. 28. (Unordered; weight 100). Encroachment of the auditory bulla on the pterygoid fossa (Kay et al. [2004c] cranial character 17): 0 no encroachment; 1 encroachment by the anterior accessory cavity; 2 present and formed by the tympanic cavity. 29. (Unordered; weight 100). Nature of contact between the lateral pterygoid plate and the bullar wall (Kay et al. [2004c] cranial character 18): 0 absent; 1 laminar; 2 abutting. 30. (Weight 100). Extent of contact between the lateral pterygoid plate and the bullar wall (Kay et al. [2004c] cranial character 19): 0 slight; 1 very extensive. 31. (Weight 100). Pyramidal process of palate and postalveolar notch: 0 no postalveolar notch between the pyramidal process and the maxillary tuberosity; 1 offset from maxillary tuberosity by a distinct postalveolar notch. 32. * (Weight 50). Mediolateral position of pyramidal processes (100 times the ratio of interpyramidal breadth to outer M1 palate breadth) (Kay et al. [2004c] cranial character 35): 0 medially placed ( 43); 1 intermediate (>43, 64) 2 laterally placed (>64). 33. (Weight 100). Posterior palatine torus (Kay et al. [2004c] cranial character 34): 0 present; 1 absent. 34. * (Weight 50). Posterior nasal spine (Kay et al. (2004c) cranial character 33): 0 reduced or absent; 1 small but distinct; 2 robust and long. 35. (Weight 100). Posterior extent of the turbinates: 0 extend posterior to the palatine; 1 completely anterior to the palatine.

36. (Weight 100). Angle of the incisive canal in palate: 0 obliquely oriented with respect to the plane of the palate; 1 more closely resembles a right angle with the palate. Temporal fossa 37. (Weight 100). Temporal emissary foramen (Horovitz [1999] character 20): 0 present and large; 1 small or absent. Nucal region 38. * (Weight 50). Paroccipital processes: 0 forms a distinct shelf or process; 1 forms a raised ridge; 2 weak or absent. 39. (Weight 100). Pneumatization of mastoid (Kay et al. [2004c] cranial character 3): 0 absent; 1 present. Orbital region 40. * (Weight 50). Lateral cranial prole at glabella: 0 depressed; 1 at; 2 convex. 41. (Weight 100). Interorbital fenestra: 0 absent; 1 present. 42. * (Weight 50). Size of orbits (Kay et al. [2004c] cranial character 23): 0 small; 1 large; 2 extremely large. 43. (Weight 50). Orbital convergence (Ross, 1995): 0 less than 55 ; 1 55 e65 , 2 greater than 65 . 44. * (Weight 50). Interorbital breadth: 0 narrow; 1 broad; 2 extremely broad. 45. * (Weight 100). Exposure of vomer in orbit: 0 unexposed; 1 exposed. 46. (Weight 100). Postorbital closure (Kay et al. [2004c] cranial character 24): 0 none; 1 postorbital bar present; 2 postorbital septum present. 47. (Weight 100). Composition of the postorbital septum (Kay et al. [2004c] cranial character 25): 0 zygomatic forms most of the septum; 1 frontal forms most of the septum. 48. (Weight 100). Position of interorbital constriction relative to olfactory tract (Kay et al. [2004c] cranial character 27): 0 absent; 1 present below olfactory tract. 49. (Weight 100). Foramen rotundum (Kay et al. [2004c] cranial character 29): 0 superior orbital ssure transmits maxillary nerve; 1 separate foramen (f. rotundum) for maxillary nerve. 50. (Weight 100). Metopic suture in adult (Kay et al. [2004c] cranial character 31): 0 unfused; 1 fused. Ear region 51. (Weight 100). Cochlear housing as exposed in middle ear (Horovitz [1999] character 15): 0 singular; 1 dual. 52. (Weight 100). Transbullar septa (Horovitz [1999] character 14): 0 middle ears ventrolateral region without septa; 1 anteroventral region with septa. 53. (Unordered; weight 100). Transverse septum arising from the cochlear housing (Kay et al. [2004c] cranial character 1): 0 absent; 1 present and forming the lateral wall of an anterior accessory cavity pneumatized

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

361

from the tympanic cavity; 2 present and forming the lateral wall of an anterior accessory cavity pneumatized from the epitympanic recess. 54. (Weight 100). Extent of pneumatization of anterior accessory cavity (Kay et al. [2004c] cranial character 2): 0 anterior accessory cavity lies anterior to the tympanic cavity and is not trabeculated; 1 anterior accessory cavity extends medial to the tympanic cavity, and is trabeculated. 55. (Weight 100). Presence or absence of perbullar pathway for the internal carotid artery (Kay et al. [2004c] cranial character 4): 0 absent; 1 present and formed exclusively by the petrosal bone. 56. (Weight 100). Anteroposterior location of posterior carotid foramen in bulla (Kay et al. [2004c] cranial character 5): 0 posterior to line joining midpoints of tympanic bones; 1 anterior to this line. 57. * (Weight 50). Mediolateral position of posterior carotid foramen in bulla (Kay et al. [2004c] cranial character 6): 0 medial; 1 midline of the bulla; 2 lateral. 58. (Weight 100). Ventrodorsal position of the carotid foramen in the bulla (Kay et al. [2004c] cranial character 7): 0 dorsal, adjacent to basioccipital or mastoid bone; 1 ventral. 59. (Weight 100). Position of posterior carotid foramen relative to fenestra cochleae (Kay et al. [2004c] cranial character 8): 0 posterior; 1 ventral; 2 anterior. 60. (Weight 100). Position of the internal carotid canal relative to the fenestra cochleae (Kay et al. [2004c] cranial character 9): 0 runs across ventral lip of the fenestra cochleae, shielding it from ventral view when a canal is present; 1 internal carotid canal does not shield the fenestra cochleae from ventral view. 61. (Weight 100). Position of the portion of the internal carotid/promontory artery (or its accompanying nerves) lying on the promontorium anterior to the fenestra cochleae (Kay et al. [2004c] cranial character 10): 0 on ventrolateral surface of promontorium; 1 contacting only the cupula of the cochlea. 62. (Unordered; weight 100). Size of stapedial and promontory canals (Kay et al. [2004c] cranial character 11): 0 both stapedial and promontory canals are large; 1 stapedial slightly smaller than promontory; 2 stapedial highly reduced or absent altogether; 3 stapedial larger than promontory; 4 both promontory and stapedial canals absent. 63. (Weight 100). Morphology of promontory canal, when present (Kay et al. [2004c] cranial character 12): 0 open trough; 1 complete canal. 64. (Weight 100). Canal for internal carotid artery or nerves (Kay et al. [2004c] cranial character 13): 0 absent; 1 present. 65. (Weight 100). Position of ventral edge of the tympanic bone (Kay et al. [2004c] cranial character 14): 0 intrabullar, or aphaneric; 1 extrabullar or phaneric. 66. (Weight 100). The shape of the tympanic bone (Kay et al. [2004c] cranial character 15): 0 ribbonlike or

only slightly expanded; 1 laterally expanded into a collar or tube. 67. (Unordered; weight 100). Morphology of annular bridge (Kay et al. [2004c] cranial character 16): 0 linea semicircularis or partial annular bridge formed on a entotympanic bulla; 1 Linea semicircularis formed on a petrosal bulla; 2 a complete annular bridge. 68. (Weight 100). Flange of basioccipital overlapping medial bulla wall (Kay et al. [2004c] cranial character 20): 0 absent or minimal; 1 extensive. 69. (Weight 100). Basioccipital stem (Beard and McPhee, 1994): 0 narrow; 1 broad. 70. (Weight 100). Suprameatal foramen (Kay et al. [2004c] cranial character 21): 0 absent; 1 present, small, and in the posterior root of the zygomatic arch; 2 present, large, and above the external auditory meatus. 71. (Weight 100). Patent parotic ssure (Kay et al. [2004c] cranial character 22): 0 present; 1 absent. 72. (Weight 100). Enclosure of intratympanic portion of the facial nerve in a bony canal (Kay et al. [2004c] cranial character 47): 0 no canal, facial runs in a sulcus; 1 bony canal present. 73. (Weight 100). Epitympanic crest (Kay et al. [2004c] cranial character 48): 0 absent; 1 present. Brain and internal cranial characters 74. (Weight 100). Tentorium cerebelli ossication (Horovitz [1999] character 13): 0 absent; 1 present. 75. (Weight 100). Vascular canal connecting sigmoid sinus with subarcuate fossa (Horovitz [1999] character 17): 0 absent; 1 present. 76. * (Weight 50). Size of olfactory bulbs: 0 large olfactory lobe; 1 moderate olfactory lobe; 2 small olfactory lobe. 77. * (Weight 50). Relative brain size: 0 small; 1 large; 2 very large. Mandible 78. (Weight 100). Symphyseal orientation: 0 more horizontal orientation of planum alveolare; 1 more vertically oriented relative to planum alveolare. 79. * (Weight 50). Lateral prole of mandible (ratio of mandible depth [measured buccally] at P2 and M2): 0 superior and inferior border of the mandibular corpus are essentially parallel from the premolar to the mandibular angle ( 1.26); 1 the inferior border deepens posteriorly (>1.26, 1.72); 2 hyperdeep (>1.72). 80. (Weight 100). Mandibular corpus depth (Kay et al. [2004c] cranial character 45): 0 shallow; 1 deep. 81. (Weight 100). Symphyseal fusion in young adult (Kay et al. [2004c] cranial character 39): 0 absent; 1 present. 82. * (Weight 50). Coronoid height relative to condyle: 0 very far above condyle; 1 above the level of condyle; 2 slightly above or equal to coronoid. 83. * (Weight 50). Condyle height relative to tooth row (Kay et al. [2004c] cranial character 44): 0 at level of tooth row; 1 slightly above; 2 well above tooth row.

362

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

84. * (Weight 50). Angle of the mandible: 0 hookshaped angle; 1 moderately expanded angle; 2 extremely expanded angle. 85. (Weight 100). Depth of the coronoid-condylar notch: 0 deep; 1 shallow. Lower Teeth Incisors 86. i1.* (Weight 33). Lower incisor number: 0 three; 1 two; 2 one: I1 present, I2 absent; 3 lower incisors absent. 87. i2. (Weight 100). Lower incisor occlusal arrangement: 0 edges wear at, producing an arcuate battery from lateral perspective; 1 cusp tips staggered. 88. i3. (Weight 100). Lower incisor crown spacing: 0 no spaces; 1 spaces present between crowns. 89. i4. (Weight 100). I2-C diastema: 0 present; 1 absent. Comment: The presence of a diastema is variable in platyrrhines and never exceeds in breadth the mesiodistal dimension of the I2. 90. i5.*(Weight 50). I1e2 size (ratio of I1e2 area to M1 area): 0 very small ( 0.69); 1 moderately sized (!0.70, 1.07); 2 large (>1.07). 91. i6.* (Weight 33). I1:I2 proportions (ratio of I1 area to I2 area): 0 I1 much smaller than I2 (<0.71); 1 I1 smaller than I2 (!0.71, <0.78); 2 I1 almost as large as I2 (!0.78, <1.00); 3 I1 larger than I2 (!1.01). 92. i7.* (Weight 50). I1 crown width (spatulate incisors only): 0 considerably wider (MD) than root (spatulate): 1 narrow at apex, but still wider than root; 2 styliform (crown apex approximately the same width as the cervical margin). 93. i8. (Weight 100). I2 crown cross-sectional shape (ratio of MD to BL): 0 rounded oval (!0.64); 1 mesiodistally compressed (<0.64). 94. i9.* (Weight 50). Lower incisor crown height (crown heights judged from cemento-enamel junction to crown tip on the buccal surface): 0 low-crowned; 1 moderately high-crowned; 2 high-crowned. 95. i10. (Weight 100). I1e2 crown buccal outline: 0 gently curved in lateral perspective; 1 acutely curved. 96. i11.* (Weight 50). Lower incisor roots: 0 erect or vertical; 1 slightly procumbent; 2 very procumbent. 97. i12.* (Weight 50). Lower incisor crowns: 0 erect or vertical; 1 procumbent; 2 very procumbent. 98. i13.* (Weight 50). Tooth comb: 0 absent; 1 with three teeth; 2 with two teeth. 99. i14. (Weight 100). I1 crown shape: 0 spatulate; 1 lanceolate, pointed. 100. i15. (Weight 100). I2 heel development (a lingual swelling at the base of crown): 0 heel absent; 1 heel present. 101. i16. (Weight 100). Incisor lingual enamel: 0 well developed; 1 poorly developed or absent. 102. i17.* (Weight 50). Lower incisor lingual cingulum: 0 absent to weak; 1 strong but incomplete; 2 strong and complete.

103. i19.* (Weight 50). I1 area to M1 area: 0 I1 very small (ratio 0.32); 1 moderately enlarged (>0.32, 0.40); 2 very enlarged (> 0.40). Canines 104. c1.* (Weight 33). Female canine size (area relative to molars): 0 very small (C/M1 < 0.40); 1 moderate (!0.4, <0.80); 2 large (!0.80, 1.20); 3 very large (!1.20) 105. c2.* (Weight 50). Canine dimorphism (square root of male canine area divided by square root of female canine area): 0 low (<1.07); 1 moderate (!1.07, <1.17); 2 high (!1.17). 106. c3. (Weight 100). Canine cross-sectional shape: 0 rounded oval (MD:BL >1.00, <1.90); 1 MD compressed (ratio ! 1.90). 107. c4. (Weight 100). Canine lingual-crest development: 0 rounded; 1 sharp. 108. c5. (not ordered; Weight 100). Canine paracristid (not scored if species has canine incorporated into a tooth comb): 0 oblique to occlusal plane; 1 nearly horizontal to occlusal plane; 2 forms part of cropping mechanism with I1e2. 109. c6.* (Weight 50). Canine height (females): 0 low, squat; 1 narrow, short; 2 tall, at or above tooth row. Premolars 110. p1. (Weight 100). P1/P1: 0 present; 1 absent 111. p2. (Weight 100). P2/P2: 0 present; 1 absent. 112. p3. (Weight 100). P2 roots: 0 single; 1 double. 113. p4.* (Weight 50). P3e4 roots: 0 P3 single, P4 single; 1 P3 single, P4 double; 2 P3 double, P4 double. 114. p5.* (Weight 50). Premolar crowding (overlapping of crowns): 0 no crowding; 1 slightly crowded; 2 very crowded. 115. p6.* (Weight 50). P3 paraconid: 0 large; 1 small; 2 absent or extremely small. 116. p7.* (Weight 50). P4 paraconid: 0 large; 1 small; 2 absent or extremely small. 117. p9.* (Weight 33). P4 paraconid position: 0 mesial to protoconid; 1 mesiolingual, between protoconid and metaconid; 2 mesial to metaconid; widely spaced from metaconid; 3 twinned with metaconid. 118. p11.* (Weight 50). P3e4 cristid obliqua: 0 absent; 1 weak; 2 strong. 119. p13. (Unordered; weight 100). P2 protoconid height and shape: 0 slender, projects above protoconids of P3e4; 1 massive, projects above protoconids of P3e4; 2 not projecting, in line with P3; 3 extremely short, shorter than P3. 120. p14. (Weight 100). P4 metaconid position: 0 close to protoconid; 1 widely spaced from protoconid. 121. p15.* (Weight 50). P2 metaconid size: 0 absent or trace; 1 small; 2 large. 122. p16.* (Weight 50). P3 metaconid size: 0 absent or trace; 1 small; 2 large. 123. p17.* (Weight 50). P4 metaconid size: 0 absent or trace; 1 small; 2 large, almost as tall as protoconid.

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

363

124. p18. (Weight 100). P4 trigonid lingual wall: 0 basin closed by a premetacristid; 1 open with premetacristid absent or short. 125. p19.* (Weight 50). P3 entoconid and lingual talonid crest: 0 absent; 1 lingual talonid crest present but an entoconid does not stand out above it; 2 entoconid is a small, discrete cusp. 126. p20.* (Weight 50). P4 entoconid and lingual talonid crest: 0 absent; 1 lingual talonid crest present but an entoconid does not stand out above it; 2 entoconid is a small, discrete cusp. 127. p21. (Weight 100). P4 lateral and medial protocristid: 0 continuous between metaconid and protoconid; 1 discontinuous between metaconid and protoconid. 128. p22. (Weight 100). P3 lateral protocristid orientation: 0 transversely oriented; 1 distolingually oriented. 129. p23a. (Weight 100). P4 lateral protocristid: 0 present; 1 absent. 130. p23. (Weight 100). P4 lateral protocristid orientation: 0 transversely oriented; 1 distolingually oriented. 131. p24. (Weight 100). P3e4 posterior trigonid wall: 0 complete (taxa without metaconids are assigned this character state); 1 deeply notched 132. p25. (Weight 100). P3e4 hypoconid size: 0 large; 1 cristiform, small, or absent. 133. p26. (Weight 100). P3e4 hypoconid (or distal terminus of oblique cristid) position: 0 distal to protoconid; 1 distal to metaconid, or between protoconid and metaconid. 134. p26a. (Weight 100). P4 talonid breadth: 0 narrow; 1 broad. 135. p27. (Weight 100). P4 hypocristid shearing development: 0 weak or absent; 1 strong. 136. p28.* (Weight 50). P2 buccal cingulum development: 0 absent; 1 incomplete, broken at protoconid and hypoconid; 2 complete. 137. p29. (Weight 100). Lower premolar ination: 0 cusps marginal, not basally inated; 1 crown surfaces constricted, cusp margins sloping. 138. p30.* (Weight 50). P4 exodaenodonty: 0 not exodaenodont; 1 slightly exodaenodont; 2 very exodaenodont. 139. p31.* (Weight 33). P4 talonid length (ratio of midline MD of trigonid to MD of talonid): 0 extremely short or nonexistent (tri:tal ! 1.61); 1 short (much shorter than trigonid) (tri:tal ! 1.27, <1.61); 2 equal or slightly shorter in length to trigonid (tri:tal ! 0.92, < 1.27); 3 talonid longer than trigonid (tri:tal < 0.91). 140. p34. (Weight 100). P4 anterobuccal cingulum development: 0 absent or trace; 1 strong. 141. p36.* (Weight 50). P4 postprotoconid ridge: 0 weak or absent; 1 present; 2 very strong. 142. p37.* (Weight 50). P4 postmetaconid ridge: 0 weak or absent; 1 moderate; 2 very strong. 143. p40.* (Weight 50). P4 paraconid height: 0 low; 1 moderate; 2 high (nearly as high as protoconid). 144. p41.* (Weight 50). P3e4 protoconid height: 0 P3 much lower than P4; 1 P3 equal or slightly lower than P4 ; 2 P3 higher than P4 .

145. p42.* (Weight 25). P3 to P4 area: 0 0.45e0.59; 1 0.60e0.69; 2 0.70e0.79; 3 >0.80, <1.10; 4, ! 1.10. 146. p43.* (Weight 20). P4 MD/BL: 0 <0.95; 1 !0.96, <1.14; 2 !1.15, <1.20; 3 !1.21, <1.35;4 !1.36, <1.46; 5 >1.47. 147. p44.* (Weight 20). P4 to M1 area: 0 <0.62; 1 !0.63, <0.72; 2 !0.73, <0.82; 3 !0.83, <0.92; 4 !0.93, <1.02; 5 >1.03. 148. p45. (unordered; weight 100). P3e4 root orientation: 0 P3e4 roots aligned mesiodistally;1 P3 root shifted laterally, P4 mesial root aligned mesiodistally; 2 P3 roots aligned mesiodistally, P4 mesial root shifted laterally. Score as missing if roots are single. Molars 149. m1. (Weight 100). M3/M3: 0 present; 1 absent. 150. m2. (Weight 100). M1 root number: 0 one; 1 two. 151. m3. (Weight 100). M2 root number: 0 one; 1 two. 152. m4. (Weight 100). M3 root number: 0 one; 1 two. 153. m6.* (Weight 50). M2 trigonid width (ratio of BL of trigonid and talonid): 0 much wider than talonid (!1.11); 1 widths similar (<1.11, >0.90); 2 much narrower than talonid ( 0.90). 154. m6a. (Weight 100). M1 trigonid length: 0 M1 trigonid short on the lingual side; 1 M1 elongate on the lingual side. 155. m7.* (Weight 50). M3 trigonid width (based on relative BL breadths): 0 much wider than talonid (!1.20); 1 trigonid and talonid widths similar ( 1.20, >1.05); 2 trigonid narrower than talonid (<1.05). 156. m8.* (Weight 33). M1 paraconid position: 0 mesial to protoconid; 1 mesiolingual, between protoconid and metaconid; 2 mesial to metaconid but widely spaced from it; 3 twinned with metaconid. 157. m9.* (Weight 33). M2 paraconid position: 0 mesial to protoconid; 1 mesiolingual, between protoconid and metaconid; 2 mesial to metaconid but widely spaced from it; 3 twinned with metaconid. 158. m10.* (Weight 33). M3 paraconid position: 0 mesial to protoconid; 1 mesiolingual, between protoconid and metaconid; 2 mesial to metaconid but widely spaced from it; 3 twinned with metaconid. 159. m11. (Weight 100). M1 parastylid: 0 absent; 1 present. 160. m12.* (Weight 50). Molar metastylids: 0 absent; 1 small; 2 large. 161. m13. (Weight 100). M3 hypoconulid: 0 single; 1 double 162. m14.* (Weight 50). M3 heel: 0 absent; 1 narrower than talonid; 2 approximately equal in width to talonid. 163. m15.* (Weight 50). Molar enamel surface: 0 smooth; 1 slightly crenulated; 2 highly crenulated. 164. m16.* (Weight 50). M1 trigonid height (ratio of trigonid height to talonid height): 0 higher than talonid (!1.20); 1 slightly higher than talonid (!1.10, <1.20); 2 trigonid and talonid of similar height (<1.10).

364

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

165. m17.* (Weight 50). M1e2 cusp relief (ratio of protoconid height to hypoconid height, measured buccally): 0 low (<1.20); 1 moderate (!1.20, <1.35); 2 high (>1.35). 166. m18. (Weight 100). M1 trigonid lingual conguration: 0 open; 1 closed. 167. m19. (Weight 100). M1 metaconid position: 0 lingual to protoconid; 1 slightly distolingual to protoconid. 168. m20.* (Weight 50). M1e2 paraconid development: 0 absent; 1 small; 2 large. 169. m21. (Weight 100). M1e2 lateral protocristid orientation: 0 runs toward metaconid; 1 runs toward hypoexid. 170. m22. (unordered; weight 100). M1 distal trigonid wall: 0 complete; 1 deeply notched by protoconid/metaconid sulcus; 2 medial and lateral protocristid do not meet but no sulcus is discerned. 171. m23. (Unordered; weight 100). M2 distal trigonid wall: 0 complete; 1 deeply notched by a sulcus between protoconid and metaconid; 2 medial and lateral protocristid do not meet but no sulcus is present. 172. m24a. (Weight 100). M1 wear facet X: 0 present; 1 absent. 173. m24. (Weight 100). M2 wear facet X: 0 present; 1 absent. 174. m25.* (Weight 50). M1e2 entoconid: 0 absent or very low, 1 lower than metaconid; 2 large. 175. m26.* (Weight 50). M1e2 postentoconid sulcus: 0 prominent; 1 shallow sulcus; 2 absent. 176. m27.* (Weight 33). M1 hypoconulid size: 0 large; 1 moderate; 2 small; 3 absent. 177. m28* (Weight 33). M2 hypoconulid size: 0 large; 1 moderate; 2 small; 3 absent. 178. m29.* (Weight 33). M3 hypoconulid size: 0 large; 1 moderate; 2 small; 3 absent. 179. m30.* (Weight 50). M1e2 hypoconulid position: 0 twinned to entoconid; 1 near midline; 2 slightly buccal to midline. 180. m31.* (Weight 50). M1e2 cristid obliqua development: 0 weak (rounded); 1 strong (trenchant); 2 very strong (trenchant). 181. m32.* (Weight 50). M1 cristid obliqua orientation: 0 reaches trigonid wall at a point distal to protoconid; 1 reaches trigonid wall at a point distolingual to protoconid; 2 reaches trigonid wall at a point distal to metaconid. 182. m33.* (Weight 50). M2 cristid obliqua orientation: 0 reaches trigonid wall at a point distal to protoconid; 1 reaches trigonid wall at a point distolingual to protoconid; 2 reaches trigonid wall at a point distal to metaconid. 183. m34. (unordered; weight 100). M1 cristid obliqua terminus: 0 runs to base of trigonid; 1 runs part way up the distal trigonid wall; 2 connects with protoconid tip or protocristid; 3 connects with metaconid. 184. m35. (unordered; weight 100). M2 cristid obliqua terminus: 0 runs to base of trigonid; 1 runs part way

up the distal trigonid wall; 2 connects with protoconid tip or protocristid; 3 connects with metaconid. 185. m36. (unordered; weight 100). M3 cristid obliqua terminus: 0 runs to base of trigonid; 1 runs part way up the distal trigonid wall; 2 connects with protoconid tip or protocristid; 3 connects with metaconid. 186. m37. (Weight 100). M1e2 centroconid development: 0 present; 1 absent. 187. m38.* (Weight 50). M1e2 hypocristid development: 0 absent or seen only as a trace; 1 weak; 2 strong. 188. m39.* (Weight 50). M3 hypocristid development: 0 absent or seen only as a trace; 1 weak; 2 strong. 189. m40.* (Weight 50). M1e2 talonid lingual conguration: 0 open; 1 closed, notched lingually; 2 closed, no notch. 190. m41. (Weight 100). M1e2 distal fovea: 0 absent; 1 present. 191. m44. (Weight 100). Molar cusp ination: 0 cusps not inated, marginally positioned; 1 very inated. 192. m45.* (Weight 50). M1e2 buccal cingulum development: 0 absent to trace; 1 partial, broken at protoconid and hypoconid; 2 complete. 193. m46.* (Weight 50). M1 hypoexid depth: 0 very shallow; 1 moderate; 2 deep. 194. m47.* (Weight 50). M2 hypoexid depth: 0 very shallow; 1 moderate; 2 deep. 195. m53.* (Weight 25). M2 length/M3 length: 0 M3 much longer than M2 (0.71e0.80); 1 M3 longer than M2 (0.81e0.90); 2 M3 equal to M2 (0.91e1.00); 3 M3 smaller than M2 (1.01e1.12); 4 M3 much smaller than M2 (!1.13). Score as 5 if M3 is absent. 196. m54.* (Weight 33). M1 length: 0 <2.5 mm; 1 !2.5, <3.8 mm; 2 !3.8, 6.0 mm; 3 !6.0. 197. m55.* (Weight 33). M1 MD/BL: 0 1.00e1.15; 1 1.16e1.22; 2 1.23e1.32; 3 >1.33. 198. m57. (Weight 100). M1e2 entoconid position relative to hypoconid: 0 transverse to hypoconid; 1 distal to hypoconid. Upper teeth Incisors 199. I1.* (Weight 50). I1e2 interstitial contact: 0 absent; teeth widely spaced; 1 present as narrow contact; 2 I2 tightly packed against I1, I1 preparacrista abbreviated. 200. I2. (Weight 100). I1-I1 interstitial contact: 0 present; 1 absent: a wide space occurs in the midline between these teeth. 201. I3. (Weight 100). I2-C diastema: 0 present; 1 absent. 202. I4.* (Weight 50). I1 area: I2 area: 0 areas approximately equal ( 1.00); 1 I1 slightly larger than I2 (>1.00, <1.40); 2 I1 much larger than I2 (>1.40). 203. I5.* (Weight 50). I1 size (I1 area:M1 area): 0 small ( 0.50); 1 moderate (>0.50, <0.56); 2 large (!0.56). 204. I6.* (Weight 50). I1 occlusal shape (MD/BL): 0 rounded oval (<1.05); 1 buccolingually compressed (>1.05, <1.30); 2 extremely compressed (>1.30).

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

365

205. I7.* (Weight 50). I2 occlusal shape (MD/BL): 0 rounded oval ( 1.05); 1 slightly compressed (>1.05, <1.30); 2 extremely compressed !1.30). 206. I8.* (Weight 50). I1 crown shape: 0 spatulate; no apparent occlusal cusp, mesial and distal edges continuous and rounded; 1 semispatulate, central cusp present but blunt with discernable mesial and distal occlusal crests; 2 central occlusal cusp pointed, occlusal edges steep. 207. I9. (Weight 100). I1 lingual fovea: 0 simple; 1 dual with midcrown pillar. 208. I10. (Weight 100). I1 occlusal edge orientation (spatulate incisors only): 0 occlusal edge orthogonal to long axis of root; 1 occlusal edge wears at a steep angle to long axis of root; 2 crown with pronounced mesial asymmetry ( mesial process) in unworn state. 209. I11.* (Weight 50). I1e2 lingual cingulum: 0 weak, discontinuous; 1 narrow, continuous; 2 strong. 210. I12. (Weight 100). I1 basal lingual cusp: 0 absent; 1 present. 211. I13. (Weight 100). I1e2 buccal cingulum: 0 absent; 1 present. Canines 212. C1. (Weight 100). Canine cross-sectional shape (gap coded; ratio of maximum length in the occlusal plane to maximum breadth in the occlusal plane at right angles to maximum length): 0 oval (!1.16); 1 rounded (<1.16). 213. C2. (unordered, weight 100). Upper canine occlusion: 0 canine wears against P1 or P2; 1 canine wears against P2; 2 canine wears against P2 or P3; 3 canine wears against P3. 214. C3. (Weight 100). Canine mesial groove (females): 0 shallow or absent; 1 deep. 215. C4.* (Weight 50). Canine lingual cingulum: 0 weak or absent; 1 strong; 2 very strong. Premolars 216. P1.* (Weight 50). P2 root number: 0 one; 1 two; 2 three. If tooth is absent, taxon scored as 9. 217. P2.* (Weight 50). P3 root number: 0 one; 1 two; 2 three. 218. P3.* (Weight 50). P4 root number: 0 one; 1 two; 2 three. 219. P4.* (Weight 50). P2 to P3 area: 0 P2 very small ( 0.85); 1 P2 small (>0.85, <0.95); 2 P2 equal (!0.95). If tooth is absent, taxon scored as 9. 220. P5.* (Weight 33). P4 to M1 area: 0 P4 < < M1 ( 0.66); 1 P4 < M1 (>0.66, 0.76); 2 P4 M1 4 1 (0.77e1.05); 3 P > M (>1.06). 221. P6.* (Weight 50). P2 occlusal shape (MD/BL): 0 buccolingually broad (<0.80); 1 round (>0.80, <1.05); 2 mesiodistally elongate (>1.05). 222. P8. (Weight 100). P3e4 trigon/talon proportions: 0 trigon and talon proportions similar; 1 trigon much shorter than talon, with the protocone situated on the mesial aspect of the crown. Comment: Character modied from that of Kay et al. (2004c). On P4, some

species have the protocone in a mesial position with either a long postprotocrista or a short postprotocrista and separate hypocone. An example of this would be Brachyteles. Others (e.g., Callithrix) have the protocone more centrally placed. Concomitantly, this means that the talon is either large, as in Brachyteles, or much smaller, as in Callithrix. 223. P9. (Weight 100). P3 protocone: 0 present; 1 absent. 224. P10. (Weight 100). P4 metacone: 0 absent; 1 present. 225. P11. (Weight 100). P4 protocone: 0 low relative to paracone; 1 high relative to paracone. 226. P12. (Weight 100). P2 protocone: 0 present as discrete cusp; 1 absent or indistinguishable from lingual cingular ridge. 227. P13.* (Weight 33). Premolar hypocones: 0 absent; 1 present on P4 only; 2 present on P3e4; 3 present on P2e4. 228. P13a.* (Weight 50). P4 hypocone: 0 absent or trace; 1 bump on postprotocone crista or postcingulum; 2 distinct cusp on distal margin. 229. P14.* (Weight 50). P4 paraconule: 0 large; 1 small; 2 absent. 230. P15. (Weight 100). P3e4 parastyles: 0 present; 1 weak or absent. 231. P16. (Weight 100). P3e4 metastyles: 0 weak or absent; 1 present. 232. P17. (Weight 100). P3e4 postprotocrista: 0 strong, reaches the distal margin, and joins the postcingulum; 1 weak, short. 233. P18. (Weight 100). P2e3 prole of distal crown margin: 0 convex, smoothly rounded; 1 concave, waisted between buccal and lingual cusps. 234. P19. (Weight 100). P3e4 lingual cingulum: 0 absent or weak; 1 strong. 235. P20. (Weight 100). P4 metacone: 0 absent; 1 present. 236. P21. (Weight 100). P3e4 buccal cingulum: 0 absent or weak; 1 strong. Molars 237. M1.* (Weight 50). M1e2 root count: 0 three, three; 1 three, two; 2 two, two. 238. M2.* (Weight 50). M3 root count: 0 three; 1 two; 2 one. 239. M3.* (Weight 50). M2 shape (ratio of BL/MD): 0 very transverse (>1.65); 1 transverse (<1.65, >1.30); 2 squared ( 1.30). 240. M4.* (Weight 50). Ratio of M1 area to M2 area: 0 M1 > > M2 (!1.40); 1 M1 > M2 (<1.40, >1.0); 2 M1 M2 ( 1.0). 241. M5.* (Weight 50). M1e2 Nannopithex-fold: 0 absent; 1 weak; 2 strong. 242. M6.* (Weight 50). M1e2 pseudohypocone: 0 absent; 1 small; 2 large. 243. M7. (Weight 100). M1e2 metaconule: 0 single (or absent); 1 double. 244. M8.* (Weight 50). M1e2 paraconule: 0 absent; 1 small; 2 large.

366

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

245. M9.* (Weight 50). M1e2 preprotoconule: 0 absent; 1 weak; 2 strong. 246. M10.* (Weight 50). M1 hypocone size: 0 large; 1 small; 2 absent. 247. M11.* (Weight 50). M2 hypocone size: 0 large; 1 small; 2 absent. 248. M12. (Weight 100). M1e2 hypocone position: 0 distal, slightly lingual to protocone; 1 distal, far lingual to protocone. 249. M13.* (Weight 50). M1e2 prehypocrista: 0 absent; 1 weak; 2 strong, reaches to the postprotocrista, encloses the talon lingually. 250. M13a. (Weight 100). M1e2 prehypocrista orientation: 0 buccolingually towards postprotocrista; 1 buccally towards metaconule. This character is an autapomorphy of Cebus among extant platyrrhines but is also seen in Acrecebus (Kay and Cozzuol, 2006). 251. M14.* (Weight 50). M3 prehypocrista development: 0 absent; 1 weak; 2 strong, reaches to postprotocrista to enclose the talon lingually. 252. M15. (Weight 100). M1 or M2 paraconule position: 0 attached to preprotocrista; 1 not attached to preprotocrista. 253. M16.* (Weight 50). M1e2 metaconule size: 0 absent; 1 small to moderate; 2 large. 254. M17. (Unordered; weight 100). M1e2 mesostyle size: 0 absent; 1 present, attached to ectocrista; 2 present on buccal cingulum. 255. M18. (Unordered; weight 100). M1e2 postprotocrista: 0 strong, runs to base of metaconule or metacone; 1 strong but short, does not reach base of metacone; 2 absent.

256. M19. (Weight 100). M1e2 lateral posterior transverse crista: 0 sharp; 1 indistinct. 257. M20.* (Weight 50). P4eM1 pericone: 0 absent; 1 small; 2 large. 258. M22.* (Weight 50). M1e3 lingual cingulum development: 0 absent; 1 weak, broken; 2 strong, complete. 259. M24.* (Weight 50). M1e2 buccal cingulum development: 0 absent; 1 weak; 2 strong. 260. M27. (Weight 100). M1e2 premetaconule cristae: 0 absent or weak; 1 strong. 261. M28. (Weight 100). M1e2 postmetaconule cristae: 0 absent or weak; 1 strong. 262. M30.* (Weight 50). M3 paraconule: 0 absent; 1 small to moderate; 2 large. 263. M31.* (Weight 50). Molar protocone lingual ination: 0 not inated; 1 slightly inated; 2 very inated. 264. M33.* (Weight 100). M2 buccal expansion of paracone: 0 no expansion; 1 expanded. 265. M34.* (Weight 50). M3 metacone: 0 absent or very small; 1 moderate (but smaller than paracone); 2 large (equal to paracone). 266. M36.* (Weight 50). M3 hypocone: 0 absent or very small; 1 small; 2 large. 267. M44.* (Weight 50). M1e3 anterior cingulum: 0 strong, complete, long; 1 strong, short; 2 weak or absent. 268. M46.* (Weight 50). M1 size relative to M3 (based on the ratio of areas of each tooth): 0 M1 ! 2.5 times the size of M3 (scored as 0 when M3 is absent); 1 M1 < 2.5, ! 1.5 times M3; 2 M1 < 1.5 times M3.

Appendix 2. Character-taxon matrix for characters described in Appendix 1

Taxon 1 CR1 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix 1 0 ? ? ? ? 1 1 1 0 1 1 1 ? {01} {01} 1 1 0 1 1 1 1 CR2 1 ? ? ? ? ? 0 1 0 0 1 ? 1 ? 1 1 1 0 0 0 1 1 1 CR3 0 0 ? ? ? ? 0 1 0 0 0 1 1 ? 0 0 0 0 0 0 0 0 0 CR4 1 0 ? ? ? ? 1 1 0 1 1 ? 1 ? 0 1 1 0 1 0 {01} 1 {01} CR5 1 ? ? ? ? ? 0 1 {01} 0 1 ? 1 ? 1 0 0 0 1 0 1 1 0 CR6 0 0 0 0 0 2 {01} 0 0 1 0 0 0 ? 0 0 {12} 1 0 1 1 2 2 CR7 1 0 0 0 0 0 0 0 0 1 0 0 0 ? 0 0 0 0 0 0 0 0 0 CR8 1 0 ? 0 0 0 1 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 CR9 1 2 ? ? ? ? 1 1 1 1 1 1 1 ? 1 1 1 1 1 {12} 1 {12} 1 10 CR10 0 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 CR11 0 0 ? ? ? ? 0 0 0 1 0 0 0 ? 0 0 0 0 0 2 1 2 1 CR12 1 1 ? ? 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 CR13 2 ? ? ? ? ? 2 2 2 2 2 2 2 ? 2 2 2 2 2 2 2 2 2 CR14 1 0 ? 1 2 1 2 2 2 2 2 2 2 ? 2 2 2 2 2 0 1 1 1 CR15 0 2 ? ? ? 2 0 0 0 2 2 2 2 ? 0 1 1 1 1 2 2 2 1 CR16 0 0 ? ? 0 0 0 0 0 0 0 0 1 ? 0 0 0 0 0 0 0 0 0 17 CR17 ? ? ? ? ? ? ? 0 ? ? 1 ? ? ? 0 ? ? ? ? 1 0 ? ?

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Appendix 2 (continued ) Taxon 1 CR1 Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon ? ? ? ? ? 1 1 1 18 CR2 ? ? ? ? ? 0 1 1 CR3 ? 0 0 ? ? 0 1 1 20 CR4 ? ? ? ? ? 0 1 1 CR5 ? 0 ? ? ? 0 1 1 CR6 ? 0 0 ? ? 0 0 0 CR7 ? ? ? ? ? 0 0 0 CR8 ? 0 ? ? ? 0 0 0 CR9 ? ? 2 ? ? 2 2 2 10 CR10 ? 1 1 ? ? 1 1 1 CR11 ? ? 1 ? ? 0 0 0 CR12 ? 1 1 ? ? 1 1 1 CR13 ? 2 ? ? ? 2 2 2 30 CR14 ? 2 1 ? ? 1 1 0 CR15 ? 0 0 ? ? 2 1 1 CR16 ? 0 0 ? ? 0 0 0

367

17 CR17 ? ? ? ? ? ? ? ? 34

CR18 CR19 CR20 CR21 CR22 CR23 CR24 CR25 CR26 CR27 CR28 CR29 CR30 CR31 CR32 CR33 CR34 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 1 0 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 35 0 0 0 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 1 1 1 1 ? 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? ? ? ? ? 1 1 1 0 1 ? 1 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? 0 0 0 2 1 ? 1 1 2 1 {12} {12} 0 {01} 0 1 ? 1 1 1 0 1 0 0 0 0 ? 2 1 ? ? ? 0 0 1 0 ? 0 0 0 0 0 0 1 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 40 0 1 ? 1 1 1 1 0 1 1 1 {12} 1 ? 0 {01} 0 0 {01} 2 1 2 2 ? 0 0 ? ? 1 1 1 0 2 ? 2 2 1 1 1 {12} 1 1 2 2 ? 1 0 {01} {01} 0 2 1 1 1 ? 2 2 ? ? 2 1 2 1 1 ? 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 0 1 1 1 1 ? ? ? ? ? 1 1 1 1 0 ? ? 1 0 1 0 0 2 1 1 1 ? {12} 2 2 1 2 2 1 2 {12} ? ? 2 ? ? 0 0 0 1 0 ? ? 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 1 0 ? ? 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 2 0 1 ? ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 0 0 1 0 1 1 1 1 0 1 0 0 0 0 0 ? 1 1 1 1 1 0 0 0 0 ? ? 0 ? ? 0 0 0 0 2 {12} {12} 2 1 {12} 1 2 2 2 {12} 2 ? 1 1 1 1 1 1 {12} 1 2 ? 2 2 ? ? 2 2 2 0 0 ? 0 0 0 1 1 1 1 {01} 1 1 ? 1 1 1 1 1 1 1 1 1 ? 0 0 ? ? 1 1 1 50 1 2 ? ? 2 1 1 1 0 {01} 1 1 0 ? {01} {01} 0 0 1 {01} 1 1 {01} ? ? 0 ? ? 1 1 1 51

CR35 CR36 CR37 CR38 CR39 CR40 CR41 CR42 CR43 CR44 CR45 CR46 CR47 CR48 CR49 CR50 CR51 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix ? 1 ? ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? 1 ? ? ? 1 1 1 0 1 1 1 1 ? 1 1 0 ? ? ? ? 1 1 1 1 1 1 1 1 ? 1 1 0 0 ? ? ? 1 2 2 2 2 1 0 0 ? 2 2 0 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 0 0 ? 1 1 0 1 2 1 0 0 1 1 ? 0 0 0 0 ? ? ? 0 0 1 0 0 0 0 0 ? 0 0 2 {012} 0 0 0 0 2 0 0 0 0 0 0 ? 0 0 0 2 1 1 1 0 2 2 2 2 2 2 2 ? 1 1 0 1 1 1 1 1 1 0 0 1 0 0 0 ? 1 1 0 ? ? ? ? ? 0 1 1 0 0 0 0 ? 0 0 2 2 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 ? 1 0 ? 1 0 ? 1 ? ? 1 ? ? 1 0 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 1 1 1 0 ? ? ? 1 1 1 1 1 1 (continued on next page)

368 Appendix 2 (continued) Taxon 35

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

40

50

51

CR35 CR36 CR37 CR38 CR39 CR40 CR41 CR42 CR43 CR44 CR45 CR46 CR47 CR48 CR49 CR50 CR51 Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 1 1 1 0 0 0 0 ? ? ? ? ? 1 1 1 52 1 1 1 1 1 1 1 ? ? ? ? ? 1 1 1 1 1 1 0 0 0 0 ? 1 1 ? ? 1 1 1 2 2 0 0 1 1 1 ? 2 2 ? ? 2 2 2 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 0 0 0 0 1 1 1 ? ? 0 ? ? 0 1 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 0 0 0 0 0 0 0 0 0 1 ? ? 0 0 0 1 1 1 2 2 2 2 ? 1 1 ? ? 2 2 2 60 1 1 1 1 0 0 1 ? 0 1 ? ? 1 0 1 0 0 0 0 0 0 0 ? ? ? ? ? 0 0 0 2 2 2 2 2 2 2 ? 2 2 ? ? 2 2 2 0 0 0 0 0 0 0 ? ? 0 ? ? 0 0 0 1 1 1 1 1 1 1 ? 1 ? ? ? 1 1 1 1 1 1 1 1 1 1 ? ? ? ? ? 1 1 1 1 1 1 1 1 1 1 ? ? 1 ? ? 1 1 1 1 1 1 0 0 0 0 ? 1 1 ? ? 0 0 0 68

CR52 CR53 CR54 CR55 CR56 CR57 CR58 CR59 CR60 CR61 CR62 CR63 CR64 CR65 CR66 CR67 CR68 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 0 0 1 ? 0 0 0 0 0 0 0 0 0 ? 0 1 1 0 1 0 0 0 0 ? ? ? ? ? 0 0 0 69 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 70 0 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 1 0 ? 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 2 1 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 1 2 2 2 ? 2 2 ? ? 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? 0 0 0 2 2 2 ? ? ? 2 2 2 2 2 2 2 ? 2 2 2 2 2 2 2 2 2 ? 2 ? ? ? 2 2 2 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 80 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? ? ? ? ? 1 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 0 ? 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? 0 0 0 85

CR69 CR70 CR71 CR72 CR73 CR74 CR75 CR76 CR77 CR78 CR79 CR80 CR81 CR82 CR83 CR84 CR85 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus 0 1 ? 1 1 1 1 1 1 1 2 0 ? 0 0 0 0 0 0 0 0 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 ? ? 0 0 0 0 0 0 0 ? 1 1 1 1 0 0 1 0 1 0 2 2 0 2 2 2 2 1 2 ? ? ? 0 1 2 2 2 0 0 ? ? ? 0 1 2 2 1 1 1 1 1 1 1 1 1 ? 1 0 0 0 0 0 0 0 0 {01} 1 0 1 0 0 0 0 1 1 1 1 0 1 1 0 0 1 1 1 1 1 2 0 0 0 0 ? 1 1 1 0 1 2 1 2 1 1 2 2 2 2 1 1 1 ? 1 1 {12} 1 1 2 1 1 1 0 0 ? 0 1 0 0

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Appendix 2 (continued ) Taxon 69 70 80

369

85

CR69 CR70 CR71 CR72 CR73 CR74 CR75 CR76 CR77 CR78 CR79 CR80 CR81 CR82 CR83 CR84 CR85 Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 1 ? ? 1 1 1 86 i1 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 2 1 1 1 1 3 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 103 i19 Tarsius Aegyptopithecus zeuxis 0 0 c1 1 1 c2 0 2 c3 0 0 c4 0 0 c5 0 0 c6 0 2 i2 1 0 0 ? ? ? 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 ? ? 0 0 0 0 0 i3 ? 0 0 0 ? ? 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 ? 0 ? 0 0 0 i4 1 1 1 ? ? ? 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 ? ? 1 1 1 1 1 0 ? ? ? 0 0 0 ? ? 0 {01} 1 1 ? 0 1 ? ? 0 0 0 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 ? ? ? 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 ? 1 ? ? ? 1 1 1 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? 0 0 0 90 i5 ? 0 0 ? ? ? 2 {12} 2 0 1 1 1 ? 2 2 1 0 1 0 2 0 2 0 0 ? 2 ? 1 2 1 i6 0 ? ? ? ? 0 2 {01} 1 1 1 0 0 ? 1 2 2 2 2 0 0 0 1 {12} ? ? 0 ? 1 2 1 i7 ? 0 0 0 ? ? 0 0 0 1 2 2 2 2 0 2 2 0 0 0 0 0 0 ? 0 ? 2 ? 0 1 1 i8 1 0 0 0 ? ? 0 0 0 0 0 1 1 1 0 1 1 0 0 0 0 0 0 0 0 ? 1 ? 0 0 0 110 p1 1 1 p2 0 1 p3 0 ? p4 1 2 p5 2 0 p6 2 2 p7 2 2 p9 p11 i9 2 1 0 1 ? ? 0 0 0 1 2 2 2 2 1 2 2 0 0 0 0 0 0 0 0 ? 2 ? 0 0 0 i10 0 0 0 0 ? ? 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 i11 1 0 0 0 ? ? 0 0 0 0 1 1 1 ? 0 1 1 0 0 0 0 0 0 1 ? ? 1 ? 0 0 0 i12 1 0 0 0 ? ? 0 0 0 0 1 1 1 ? 0 1 1 0 0 0 0 0 0 0 ? ? 1 ? 0 0 0 i13 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 i14 1 0 0 0 ? ? 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 1 1 1 ? 0 0 0 1 0 1 1 1 1 ? 1 1 ? ? 0 0 0 2 2 2 ? 2 2 2 2 2 2 2 ? 2 ? 2 ? ? ? 0 0 0 2 2 2 ? 2 2 2 2 2 2 2 2 2 ? 2 2 ? ? 2 2 2 2 2 2 ? 0 0 0 1 1 0 2 ? 2 ? ? ? ? ? 2 2 2 1 1 1 ? 1 0 0 1 0 1 1 1 1 ? ? ? ? ? 1 1 1 1 1 1 ? 0 0 0 0 0 1 1 1 1 0 0 ? ? ? 0 0 0 1 1 1 ? 1 0 0 1 1 1 1 1 1 0 ? ? 1 1 0 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 ? ? ? ? 1 1 1 1 2 1 ? 1 1 1 0 0 2 2 2 2 ? ? ? ? ? 2 2 2 2 2 2 ? 2 1 1 2 2 2 2 2 2 ? ? ? ? ? 2 2 2 100 i15 ? 1 1 1 ? ? 1 1 1 1 0 0 0 0 1 0 0 1 1 1 1 1 1 1 1 ? 0 ? 1 1 1 i16 0 0 0 0 ? ? 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 2 2 2 ? 1 0 0 1 1 2 2 2 2 ? ? ? ? ? 1 1 1 1 1 1 ? 0 1 1 0 1 1 1 1 1 ? ? ? ? ? 1 1 1 102 i17 0 1 2 ? ? ? 2 2 2 0 0 0 0 0 2 2 1 2 2 1 1 2 2 2 2 ? 0 ? ? ? 0 119 p13

? 1 3 ? 1 ? (continued on next page)

370 Appendix 2 (continued) Taxon 103 i19 Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 0 0 0 0 2 2 2 0 1 1 1 ? 2 2 1 0 1 0 2 0 2 0 0 0 2 ? 1 2 2 120 p14 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos 1 1 0 1 1 1 1 1 1 1 1 1 1 ? 0 0 0 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 p15 0 ? 1 ? 0 0 {01} 0 {12} {01} {01} 1 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? ? c1 1 1 ? 1 1 2 3 1 2 3 3 3 2 {12} 1 {12} 2 1 2 1 2 1 1 ? 3 1 2 1 1

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

110 c2 2 2 2 ? 0 2 1 0 1 0 2 ? 0 0 0 0 0 2 1 2 2 1 2 ? ? ? 0 2 2 c3 0 0 ? 0 0 0 0 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 c4 0 0 ? 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 c5 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 c6 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 ? 2 ? 2 2 2 p1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 p2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 p3 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 ? ? ? p4 2 2 2 2 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 2 2 2 130 p16 0 0 1 0 0 1 2 2 2 {12} 2 2 2 ? {12} 0 1 2 {01} 2 2 2 2 1 1 ? 1 1 0 0 0 p17 1 2 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 p18 1 0 1 {01} 1 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 1 0 0 1 ? 0 0 0 0 0 p19 0 0 0 0 0 1 1 2 1 2 2 2 2 ? 1 0 0 1 1 {12} 1 1 1 1 1 ? 1 1 1 1 1 p20 0 1 0 2 1 2 2 2 {12} 2 2 1 2 ? 1 1 1 1 1 2 1 2 2 1 2 ? 2 {12} 1 1 1 p21 0 0 1 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 1 0 0 0 ? 0 0 0 0 0 p22 1 1 1 1 1 1 0 0 0 0 0 0 0 ? 0 1 1 0 0 0 0 0 0 1 1 ? 1 1 1 1 1 p23a 0 0 1 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 1 0 0 0 ? 0 0 0 0 0 p23 1 0 1 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 1 ? 1 1 0 0 0 p24 0 0 1 0 0 1 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 p25 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 1 1 1 1 0 0 ? 0 ? ? ? ? p26 1 0 1 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 ? ? ? ? 0 0 ? 0 0 ? ? ? P26a 0 1 0 1 0 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 0 1 1 1 ? 1 ? 1 1 1 p27 0 0 0 0 1 0 0 0 0 0 1 1 1 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 p5 0 0 2 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 2 2 2 p6 2 1 2 1 2 2 2 2 2 2 2 ? 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 p7 2 1 1 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 2 2 2 2 2 2 ? 2 1 2 2 2 p9 ? 2 2 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? p11 0 1 1 0 1 0 1 1 2 2 2 ? 1 1 1 1 1 1 1 1 1 1 1 ? 1 ? ? ? ?

119 p13 0 ? 0 2 2 2 1 2 2 2 1 2 0 2 2 0 2 2 2 2 2 2 2 ? 1 2 ? ? ? 136 p28 2 ? 1 ? 2 0 0 0 0 0 0 0 0 0 {01} {01} 0 0 0 0 0 0 0 0 0 ? 0 2 ? ? ?

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Appendix 2 (continued ) Taxon 137 p29 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 0 1 1 0 1 1 0 0 1 0 0 0 0 1 0 1 1 0 1 0 1 0 0 0 1 ? 1 1 0 0 0 154 m6a Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 ? ? 1 m7 1 1 1 1 0 0 1 0 0 1 1 1 1 ? ? ? ? 0 ? {12} 1 1 1 1 ? ? 1 m8 1 ? ? 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? m9 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? m10 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? m11 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 P30 1 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 p31 0 3 1 2 2 3 1 0 2 2 3 3 3 ? 1 2 2 3 2 0 2 0 1 3 0 ? 1 2 3 3 3 140 p34 1 0 0 0 1 0 0 0 0 0 0 0 0 ? 0 1 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 p36 1 1 0 1 0 0 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 0 ? ? 1 1 1 1 1 p37 1 1 0 1 1 0 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 p40 ? ? ? 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ? ? ? ? ? 160 m12 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 m13 0 0 0 0 0 0 ? ? ? ? ? ? ? ? ? ? ? ? ? 0 ? 0 ? 0 ? ? ? m14 1 1 2 1 {01} 1 0 0 0 0 0 0 0 ? ? ? ? 0 ? 0 0 0 0 {01} ? ? 2 m15 0 0 0 0 0 0 0 0 0 0 1 2 2 1 0 0 0 0 0 0 0 0 0 0 0 ? 0 m16 0 2 2 0 1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 ? 2 m17 2 1 1 1 2 1 1 2 1 1 0 0 0 ? 2 2 0 2 2 2 0 1 0 1 1 ? 1 m18 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 {01} 1 0 1 1 1 ? 1 m19 1 0 1 0 0 1 0 1 0 1 0 0 0 1 0 1 1 0 0 1 0 1 0 0 1 ? 1 m20 m21 p41 0 2 1 1 1 1 1 1 1 1 1 1 1 ? 1 2 2 1 1 1 1 1 1 1 1 ? 2 1 2 2 2 p42 1 4 3 3 1 3 2 3 3 3 3 3 3 ? 3 3 2 3 3 3 3 3 3 1 3 ? 3 3 4 4 3 p43 1 0 2 3 1 1 0 0 0 0 0 0 0 ? {01} {01} 1 0 0 {01} 0 0 0 1 1 ? 1 ? 2 2 1 p44 0 1 {12} 1 0 2 1 3 3 0 1 3 3 ? 2 3 2 2 4 {12} {12} 1 1 3 2 ? 2 ? 1 0 1 p45 1 1 2 2 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 ? ? 0 ? 1 1 1 m1 0 0 0 0 0 0 0 0 0 0 0 0 0 ? {01} 1 1 0 1 0 0 0 0 0 0 ? 0 0 0 0 0 150 m2 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 0 0 1 1 1 ? 1 1 1 1 1 m3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 0 1 0 1 1 1 1 ? 1 1 1 1 1 m4 1 1 1 1 1 1 0 0 0 0 0 ? ? ? ? ? ? 0 ? 1 0 ? 0 1 1 ? 0 ? 1 1 1

371

153 m6 1 1 1 1 1 0 1 {01} {01} 1 1 1 1 1 {01} 0 1 1 1 {12} 1 1 1 1 1 ? 1 1 1 1 1 170 m22

2 0 0 0 1 0 0 1 1 1 0 0 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 2 0 0 0 {01} 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 ? ? ? 0 0 0 (continued on next page)

372 Appendix 2 (continued) Taxon 154 m6a Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon ? 0 0 0 171 m23 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 0 0 1 0 0 1 0 0 0 0 2 2 2 0 0 0 0 0 2 0 0 0 0 0 0 ? 0 0 0 0 0 188 m39 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles 2 1 1 2 2 1 0 0 0 0 0 0 0 ? ? ? ? 0 ? 1 1 m40 2 1 1 2 2 1 2 2 2 1 2 2 2 2 2 2 2 2 2 {12} 2 m24a 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 0 0 0 m7 ? 1 1 1

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

160 m8 ? ? ? ? m9 ? ? ? ? m10 ? ? ? ? m11 0 0 0 0 m12 0 0 0 0 m13 ? 0 ? 0 m14 ? 1 0 1 m15 0 0 0 0 180 m24 1 0 0 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 0 0 0 190 m41 0 1 1 0 0 1 0 0 0 {01} 0 0 0 0 0 0 0 0 0 0 0 m44 0 1 1 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 m45 2 {12} 1 0 1 0 0 1 0 {01} 0 0 0 1 1 0 0 0 0 0 0 m46 1 1 2 1 1 1 1 1 1 1 0 0 0 1 1 1 2 1 1 2 1 m47 1 1 1 1 1 1 1 1 0 1 0 0 0 1 0 1 1 1 0 2 1 m53 0 1 2 3 3 2 3 4 3 3 3 3 3 ? {45} 4 5 5 5 1 3 m54 0 3 1 0 0 1 1 1 2 1 2 1 2 2 1 0 0 1 1 3 2 m55 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 {01} 1 1 {01} {01} 0 m57 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 {01} 0 I1 2 ? ? 1 ? ? 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 m25 1 1 1 1 1 2 1 1 1 1 1 0 0 1 0 0 1 0 0 1 1 2 1 1 1 ? 0 1 1 2 2 m26 2 0 0 1 1 1 2 2 2 {12} 2 2 2 2 2 2 2 2 2 {12} 2 2 2 1 0 ? 1 0 0 2 2 m27 3 0 0 1 1 0 3 2 3 {23} 3 3 3 3 3 3 3 3 {23} {23} 2 3 3 1 2 ? 2 2 0 3 3 m28 3 0 0 1 1 0 3 2 3 {23} 3 3 3 3 3 3 3 3 {23} {23} 3 3 3 1 2 ? 2 ? 0 3 3 m29 0 0 0 1 1 1 3 1 2 1 2 2 2 ? ? ? ? 3 ? 1 2 2 2 1 ? ? 1 ? 0 2 0 m30 ? 1 1 0 0 1 ? 1 ? 1 ? ? ? ? ? ? ? ? 0 0 ? ? ? 0 1 ? 1 1 1 ? ? m31 2 0 0 0 0 1 1 1 0 1 0 0 0 1 1 1 1 2 1 2 0 2 1 1 1 ? 0 1 0 2 2 m32 0 0 2 0 0 1 0 1 0 1 0 0 0 0 {01} 1 1 0 0 1 0 1 0 1 1 ? 0 1 0 0 0 m33 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 {01} 0 1 0 0 0 ? 0 1 0 0 0 m34 0 0 0 0 0 0 2 0 2 2 2 2 2 2 1 0 0 0 1 1 1 0 0 1 1 ? 1 1 0 0 0 m35 0 0 0 0 0 0 2 0 2 2 2 2 2 2 1 0 0 0 1 1 2 0 2 1 2 ? 1 1 0 0 0 200 I2 0 ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 I3 0 ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 I4 2 ? ? 0 1 ? 2 2 1 1 2 2 2 ? 2 2 1 1 1 0 1 I5 0 ? ? 0 ? ? 2 2 2 1 1 2 2 ? 1 2 0 0 0 0 2 m36 0 0 0 0 0 0 2 1 2 2 2 2 2 ? ? ? ? 0 ? 2 2 2 2 0 ? ? 1 ? 0 0 0 m37 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 m16 2 2 2 2 m17 ? 0 0 0 m18 1 1 1 1 m19 1 0 0 0 m20 0 0 0 0 m21 0 1 1 1

170 m22 0 1 0 0 187 m38 2 1 1 2 2 1 1 2 1 1 0 0 0 1 1 1 1 2 1 2 1 2 1 1 1 ? 1 2 1 2 2 204 I6 0 ? ? 1 ? ? 2 1 0 1 1 1 1 ? 1 1 2 2 1 0 1

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Appendix 2 (continued ) Taxon 188 m39 Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 2 1 2 ? ? 1 ? 1 2 2 205 I7 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 1 ? ? 1 ? ? 0 ? 0 0 1 0 1 ? 1 0 1 1 1 1 0 1 1 ? ? ? ? ? ? ? ? 222 P8 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia 1 1 1 ? ? ? 1 1 1 1 1 1 1 ? P9 1 0 0 0 0 0 0 0 0 0 0 0 0 ? P10 0 0 0 0 0 0 0 0 0 0 0 1 1 ? P11 0 0 0 0 0 0 0 0 0 0 0 0 0 ? P12 1 ? 0 ? 0 0 0 0 0 0 0 0 ? ? P13 0 2 3 0 0 0 0 0 1 1 {01} 0 1 ? P13a 0 1 1 0 0 0 0 0 {01} 1 1 1 1 ? P14 2 2 0 2 ? 0 2 2 2 2 2 2 2 ? I8 2 ? 0 0 0 ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? 0 0 0 I9 0 ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 1 1 1 ? 0 ? ? ? ? ? ? I10 ? ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? ? ? ? I11 2 ? ? 2 ? ? 2 2 2 2 1 1 1 ? 1 0 0 0 1 1 1 ? 1 ? 1 ? ? ? ? ? ? m40 0 2 1 1 ? 2 2 1 0 0 190 m41 0 0 0 1 ? 0 1 1 0 0 m44 0 0 0 1 ? 1 0 1 0 0 m45 0 0 0 1 ? 0 0 0 0 0 210 I12 0 ? ? 0 ? ? 0 0 1 1 1 1 1 ? 0 0 0 0 0 0 1 1 1 ? 0 ? ? ? ? ? ? I13 1 ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 ? 0 ? ? ? ? ? ? C1 1 0 ? 0 0 ? {01} 1 1 1 1 0 0 1 0 {01} 0 0 0 0 1 1 1 0 0 ? 0 ? 0 0 0 C2 1 3 1 3 1 1 1 1 1 1 1 1 1 ? 1 1 1 1 1 1 1 1 1 1 1 ? 1 1 3 3 3 230 P15 1 1 0 0 0 1 1 1 1 1 1 0 0 ? P16 0 0 0 0 1 0 0 0 1 0 0 1 1 ? P17 0 0 0 0 1 0 0 0 0 0 0 0 0 ? P18 1 0 0 1 1 0 0 0 0 0 0 0 0 ? P19 0 1 0 0 0 0 0 1 0 1 0 0 0 ? P20 0 0 0 0 ? 0 0 0 0 0 0 0 0 ? P21 M1 C3 1 1 ? 1 1 ? 0 1 0 0 0 0 0 1 1 ? 1 ? 1 ? 1 ? ? 1 1 ? 1 1 1 1 1 C4 1 1 ? 0 1 ? 1 2 2 2 ? 1 1 1 0 0 0 ? 0 0 2 ? ? 1 1 ? 0 1 2 0 0 P1 0 ? 2 0 ? 1 1 0 ? 1 0 ? ? ? 0 0 0 0 0 1 0 ? ? 0 0 ? 1 ? ? ? ? P2 2 2 2 1 2 2 1 0 1 1 0 ? ? ? 0 0 0 1 1 {12} 0 1 0 1 1 1 2 {01} 2 2 2 P3 2 2 2 2 2 2 1 1 1 1 2 ? ? ? 0 1 1 0 1 {012} 0 1 0 1 1 ? 2 1 2 2 2 P4 0 ? 1 ? ? 0 2 1 2 1 1 1 1 ? 0 2 0 2 1 0 1 0 2 0 0 ? 1 ? ? ? ? m46 1 1 1 1 ? 0 1 1 1 1 m47 1 1 1 0 ? 0 1 1 1 1 m53 3 3 3 3 ? 3 4 3 3 2 m54 3 2 1 1 ? 1 2 3 3 3 m55 1 0 0 1 ? 2 0 ? ? ? m57 0 0 0 0 ? 0 0 0 0 0 I1 1 ? ? 1 ? ? ? 1 1 1 200 I2 0 0 ? 0 ? ? ? 0 0 0 I3 0 0 ? ? ? ? ? 0 0 0 I4 0 1 ? ? ? ? ? ? ? ? I5 0 2 ? ? ? ? ? ? ? ? 220 P5 0 1 1 1 2 1 {01} {12} 2 0 1 2 2 ? 1 0 {01} 0 1 0 1 {01} 1 0 0 ? 1 0 1 0 0

373

204 I6 0 1 ? 1 ? ? ? ? ? ? 221 P6 1 ? 0 ? 0 0 0 0 0 0 1 0 0 ? 0 1 1 0 1 1 0 1 0 1 0 ? 0 ? ? ? ? 238 M2

0 0 0 1 1 0 0 0 0 0 0 0 0 ? (continued on

0 0 0 0 0 0 0 0 0 1 0 0 1 2 1 2 1 2 0 1 1 2 ? ? ? ? ? ? next page)

374 Appendix 2 (continued) Taxon 222 P8 Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 0 0 0 0 1 0 0 0 0 1 1 ? 0 0 0 0 ? 239 M3 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos Taxon 1 1 1 1 0 2 2 {01} 1 {12} 2 2 2 ? 0 1 1 1 {12} 2 2 2 2 1 0 1 1 1 2 2 2 256 M19 Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus 0 0 1 1 1 1 0 M20 0 0 2 0 0 0 0 P9 0 0 0 0 0 0 0 0 0 0 0 ? 0 ? 0 0 0 240 M4 1 2 2 2 1 1 1 1 1 1 1 1 1 ? 0 0 0 0 0 2 {12} 1 {12} 1 2 1 1 ? 2 2 2

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

230 P10 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 P11 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 P12 {01} 1 1 0 0 0 0 0 0 0 1 ? 0 ? ? ? ? P13 0 0 0 0 0 {012} 0 0 0 0 2 ? 2 {123} 0 0 0 P13a 0 0 0 0 0 {01} 0 0 0 0 2 ? 2 2 0 0 0 P14 2 2 2 2 2 2 2 2 2 2 2 ? 2 2 2 2 2 P15 1 0 0 1 1 1 1 0 1 0 0 ? 1 1 1 1 1 P16 0 0 0 0 0 0 0 1 0 1 1 ? 0 0 ? 0 0 P17 0 1 1 0 0 0 0 0 0 1 1 ? 1 1 0 0 0 P18 1 1 1 1 0 0 0 0 0 0 0 ? 0 ? 0 0 0 250 M5 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M6 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 M7 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 ? ? ? ? 0 0 0 M8 2 1 2 1 0 2 0 0 0 {01} 0 0 0 ? 0 0 0 1 0 0 0 0 0 0 1 ? 0 1 0 0 0 M9 0 1 2 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 1 ? 0 ? 0 0 0 260 M22 2 2 2 2 2 2 {01} M24 2 1 1 2 2 {12} 0 M27 0 0 0 0 0 0 0 M28 0 0 0 0 0 0 0 M30 0 0 2 0 0 1 0 M31 0 1 1 0 0 1 0 M33 0 1 1 0 0 1 0 M34 2 0 0 1 0 0 0 M36 0 0 1 0 0 0 1 M44 0 1 1 1 1 1 2 M10 2 0 0 1 1 0 0 0 0 0 0 0 0 ? 2 2 2 1 2 0 0 0 0 1 0 1 1 0 0 0 0 M11 2 0 0 1 1 0 0 0 0 0 0 1 0 ? 2 2 2 2 2 0 0 0 0 1 0 1 1 0 0 0 0 M12 ? 0 1 0 0 0 0 0 0 0 0 0 0 ? ? ? ? 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 M13 0 0 1 0 0 0 2 1 2 2 2 1 1 ? ? ? ? 0 ? 1 1 0 0 0 1 ? 1 {012} 2 2 2 M13a ? ? 0 ? ? ? 0 0 1 0 0 0 0 ? ? ? ? 0 ? 0 0 ? ? ? 0 0 0 0 0 0 0 M14 0 0 1 0 0 0 0 0 0 0 0 0 0 ? ? ? ? 0 ? 0 1 1 0 0 ? ? 0 0 ? ? ? M15 0 0 1 0 ? 1 ? ? ? ? ? ? ? ? ? ? ? 0 ? ? ? ? ? ? 0 ? ? ? ? ? ? M16 1 1 2 0 {01} 1 0 0 1 {01} 0 0 0 ? 0 0 0 1 0 {01} 0 {01} 0 {01} 0 ? 0 ? 0 0 0 M17 0 0 0 0 0 0 0 0 0 0 0 0 0 ? 2 2 2 2 2 1 0 2 0 {02} 2 ? 0 0 0 0 0 P19 0 0 0 0 1 0 0 0 0 1 1 ? 1 1 ? 0 0 P20 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 1 1 1 P21 0 0 0 0 0 {01} 0 1 0 0 0 ? 0 0 0 0 0 M1 0 0 0 0 0 0 2 ? ? 0 0 1 0 {01} 0 0 0

238 M2 ? ? ? 1 ? 2 2 ? 2 1 1 2 ? ? ? ? ? 255 M18 0 0 2 0 0 2 0 0 1 1 1 1 1 ? 0 0 0 0 0 1 1 1 1 1 0 ? 1 0 0 1 1 268 M46 2 2 1 2 0 0 1

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Appendix 2 (continued ) Taxon 256 M19 Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella pygmaea Callimico goeldii Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates lar Miopithecus talapoin Presbytis melalophos ? 1 1 1 1 1 ? 0 0 1 0 0 1 1 1 1 1 0 ? 1 0 0 0 0 M20 1 0 0 0 0 0 ? 0 0 0 0 1 0 0 0 0 0 0 1 {01} 0 0 0 0 M22 2 {12} 2 1 0 0 ? 1 2 2 2 2 {01} 0 0 0 2 2 2 2 0 0 0 0 M24 {01} 0 0 0 0 0 ? 1 1 1 1 1 1 0 0 0 {01} 0 ? 0 {01} 0 0 0 260 M27 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 M28 0 0 0 0 0 0 ? 0 0 0 0 ? 0 0 0 0 0 0 ? 0 0 ? ? ? M30 0 0 0 0 0 0 ? ? ? ? 0 ? 0 0 0 0 0 1 ? 0 0 0 0 0 M31 0 1 0 1 1 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 1 0 M33 1 1 0 0 0 0 ? {01} {01} 0 0 0 1 0 0 0 0 1 ? 1 ? ? 0 0 M34 0 1 0 0 0 0 ? ? ? ? 0 ? 1 1 1 1 0 2 ? 0 0 ? 2 2 M36 {01} 1 {01} 0 1 0 ? ? ? ? 0 ? {01} 1 1 1 0 1 1 0 1 ? 2 2 M44 1 2 1 2 2 2 ? 2 1 1 1 1 2 2 2 2 2 1 2 2 2 2 2 2

375

268 M46 0 1 1 1 {12} {12} ? 0 0 0 0 0 2 2 2 2 1 1 0 1 1 2 2 2

376

Appendix 3. Specimens examined

Taxon Branisella boliviana Tremacebus harringtoni Carlocebus carmenensis Alouatta caraya Alouatta palliata Type specimen, Tucuman, Argentina Atelidae Atelidae A. caraya group A. palliata group FMNH 44330, 44331, USNM 240406, 240407, 240408, Kay lab See Fleagle, 1990 20 50 Higher-level taxon Subtaxon Cranial material examined Dental measurements (sample size) From Takai et al., 2000 Dental material examined MNHN Bol V 3447, 3468, 5310, 5311 Type specimen, Tucuman, Argentina MACN SC 11, 113, 266, 400 USNM 238675, 390645, 390648, 518237, 518238, 518233

Alouatta pigra Alouatta belzebul Alouatta guariba

Atelidae Atelidae Atelidae

A. palliata group A. seniculus group A. seniculus group

22 20 18

BMNH 11.7.27.1 USNM 544508, MSP 8067, 8071 USNM 518256, 38392,113436, 518249, 518248, 518246/5, 518253, 518255, 518257, 518247, MSP 5868, 2346

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

Alouatta macconnelli Alouatta seniculus Ateles fusciceps Ateles geoffroyi

Atelidae Atelidae Atelidae Atelidae

A. seniculus group A. seniculus group Atelinae Atelinae USNM 14162, 14128/12128, USNM 14161, Kay lab specimen

22 51 7 44

Ateles marginatus Ateles paniscus Brachyteles cf. arachnoides

Atelidae Atelidae Atelidae

Atelinae Atelinae Atelinae

FMNH 2138, 21385 USNM 259474

29 10

Lagothrix cana

Atelidae

Atelinae

Lagothrix lagotricha Lagothrix poeppigii Callimico goeldii

Atelidae Atelidae Callitrichinae

Atelinae Atelinae

USNM 545885, 194342, Kay lab, FMNH 98053, 98052 USNM 303323, 464991, 573934, 528737, 583199, FMNH 98034, FMNH 60354, USNM 336310, 336311 23 2

USNM 284932, 291240, 291242, 291243 USNM 108529, 244869, 13048, 14161, 14154, 14116, 13297, 171075, 284948, 291056, 291057, 291060, 291064, 315840, 337694, 337698, 337707 MSP 1933019329 MSP 5277 ANS 2744, 2685, BMNH 48.10.25.1, USNM 259474, MSP 1159, 1863, 3533, 1158, 2236, 11100, 11098, 19360, 11104, 1160 USNM 398460, 397775, 194341, 194342, 399302, 269839, 311217, MSP 917, 5533, 11233 ANS 2872, USNM 194342, 398460, 397963, 398460, MSP 917 USNM 395455, 303323

Cebuella pygmaea

Callitrichinae

37

USNM 319513, 337948, 336302-336314 (12 specimens) 336311, 336313, 336314, 336323, 337325 ANS 13497, USNM 395454, 534283, 546320 USNM 518558-518561 (4 specimens), 518564, 518566e518568

Leontopithecus chrysomelas Leontopithecus rosalia

Callitrichinae Callitrichinae

Kay lab skull USNM 546318, 546321, 546317, 546322, 546323, FMNH 57838, 46165 Callithrix group

4 18

Callithrix penicillata

Callitrichinae

16

Callithrix geoffroyi Callithrix jacchus Callithrix argentata Saguinus midas Saguinus mystax Saguinus fuscicollis Saguinus geoffroyi Saguinus leucopus Cebus apella

Callitrichinae Callitrichinae Callitrichinae Callitrichinae Callitrichinae Callitrichinae Callitrichinae Callitrichinae Cebidae

Callithrix group Callithrix group Mico group S. midas group S. mystax group S. nigricollis group S. oedipus group S. oedipus group C. apella group

FMNH 134473, 134474 UT 0434, USNM 503895, 503886

10

USNM 518554, 518555 USNM 23459, 239458, 239457, 239561 USNM 393807 USNM 543489, 397242, 397873, 397877, 397930 ANS 16565, USNM 518576, 397302, skull in Kay collection USNM 301653, 336296 USNM 281630 USNM 397959, 397979

57 USNM 544383, 544382 USNM 398731, 336299, UT 0635, FMNH 24194, 24195 34 42 USNM 296639, 361020, 338958, 268935, 33956, 296638, 296640, 338957, 38959, 361019, FMNH 20000, 461384, 46182 40

Cebus libidinosus

Cebidae

C. apella group

Cebus albifrons Cebus capucinus Cebus olivaceus Saimiri boliviensis

Cebidae Cebidae Cebidae Cebidae

C. capucinus group C. capucinus group C. capucinus group S. boliviensis group

Kay lab skull USNM 291234, 283945 USNM 396265, 397316, 397252, FMNH 55498, 55499, 55500 USNM 291092, 3693149, 3963148, 291045 Kay lab skull

USNM 518281, 518283, 518336, 518341, 518347, 518349, 518350, 518466, 518471 USNM 398445 USNM 261913

53 46

Saimiri oerstedti Saimiri sciureus

Cebidae Cebidae

S. sciureus group S. sciureus group

35 USNM 194345, 241409, 393795, 398676, 398691, 398704, 398708, 398709, 546267, 393795 44 31 ANS 19813, USNM 171078, 396157, 396429, 396715, 396716, 396720, 396722, 396724, BMNH 45.6.17.6, 73.4.23.2 BMNH 8.5.9.8, 26.5.5.12, 26.5.5.10, 26.5.5.17 BMNH 3.9.5.2, USNM 518221 ANS 14306, USNM 256213

Aotus sp. Aotus lemurinus Aotus trivirgatus

Cebidae Cebidae Cebidae

Grey-neck Aotus group Grey-neck Aotus group

Kay lab skull FMNH 68858, 68860 USNM 514971, 464370, 503709

Callicebus cupreus Callicebus moloch Callicebus ornatus Callicebus personatus Callicebus torquatus

Pitheciidae Pitheciidae Pitheciidae Pitheciidae Pitheciidae

C. moloch group C. moloch group C. moloch group C. moloch group C. torquatus group

FMNH 25335, 25334 25 USNM 283924, 283953 USNM 406410, 406411, 406412, 406415, 406416, 406417, 406414, 388149, 388149, 406567, 406568, 406569, 406570 FMNH 94926, 94927 USNM 549519, 338964 20

Chiropotes albinasus Chiropotes satanas Pithecia monachus

Pitheciidae Pitheciidae Pitheciidae

Pitheciinae Pitheciinae Pitheciinae

34

USNM 546262 USNM338962, 406593, 406594 USNM 461918, 461919, 545891, 518224, 518223 (continued on next page)

378

Appendix 3 (continued) Taxon Pithecia pithecia Higher-level taxon Pitheciidae Subtaxon Pitheciinae Cranial material examined USNM 339658, 339660, 374743, 374744, 374745, 374749, 374748, 374763, 374767, FMNH 50883, 46176, Kay lab skull FMNH 88814, FMNH 88819 USNM 406423, 406422 USNM 488081, 488084, 546331 FMNH 56763, Duke osteological collection USNM 111990 USNM 395340, 220338 USNM 113174 DPC 5401, 6642, CGM 85785 YPM 23978, 25972, 23968 DPC 18651 DPC 11388 DPC 14095 Dental measurements (sample size) 25 Dental material examined ANS 1263, USNM 339659

Cacajao calvus Cacajao melanocephalus Tarsius bancanus borneanus Tarsius syrichta Hylobates lar Miopithecus talapoin Presbytis melalophos Aegyptopithecus zeuxis

Pitheciidae Pitheciidae Tarsiidae Tarsiidae Catarrhini Catarrhini Catarrhini Catarrhini

Pitheciinae Pitheciinae

19 USNM 256216

USNM 282761, Duke Osteological collection 40 39 31 Metrics from Kay et al., 1981 Metrics from Kay (unpublished) Metrics from Kay (unpublished) Metrics from Simons et al., 1999 Metrics from Simons et al.,1999 R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Skull in Kay collection Kay collection DPC 1112, CGM 40237, DPC 1112, CGM 40237, other specimens cited in Kay et al., 1981. DPC 3080, 3871; YPM 21018, 21016 DPC 1009, 1091, 2807, type specimen CGM 26912, 1118, 3110 CGM 42200, DPC FN 95-1039, FN 95771, FN 95-894 CGM 42200, 42206, DPC 95-1039

Hylobatidae Cercopithecidae Cercopithecidae Propliopithecidae

Apidium phiomense Simonsius ( Parapithecus) grangeri Catopithecus browni Proteopithecus sylviae

Stem Anthropoidea Stem Anthropoidea Stem Anthropoidea Stem Anthropoidea

Parapithecidae Parapithecidae Oligopithecidae Proteopithecidae

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382

379

Appendix 4: Percentage completeness of the data set

Taxon Number of characters missing* Total Tarsius Aegyptopithecus zeuxis Apidium phiomense Catopithecus browni Proteopithecus sylviae Simonsius grangeri Aotus Saimiri Cebus Callicebus Pithecia Chiropotes Cacajao Proteropithecia Saguinus Callithrix Cebuella Callimico Leontopithecus Alouatta Ateles Brachyteles Lagothrix Branisella boliviana Dolichocebus gaimanensis Tremacebus harringtoni Soriacebus ameghinorum Carlocebus carmeni Hylobates Miopithecus talapoin Presbytis melalophos 14 37 63 51 65 53 12 12 14 11 12 22 20 207 25 27 26 10 27 11 12 21 19 102 55 197 108 145 38 34 33 Cranial 4 9 42 35 25 15 3 2 4 3 2 7 4 85 2 3 3 0 4 2 2 5 3 80 31 34 84 84 3 2 2 Dental 10 28 21 16 40 38 9 10 10 8 10 15 16 122 23 24 23 10 23 9 10 16 16 22 24 163 24 61 35 32 31 Percentage complete (%) Total 94.8 86.2 76.5 81.0 75.7 80.2 95.5 95.5 94.8 95.9 95.5 91.8 92.5 22.8 90.7 89.9 90.3 96.3 89.9 95.9 95.5 92.2 92.9 61.9 79.5 26.5 59.7 45.9 85.8 87.3 87.7 Cranial 95.3 89.4 50.6 58.8 70.6 82.4 96.5 97.6 95.3 96.5 97.6 91.8 95.3 0.0 97.6 96.5 96.5 100.0 95.3 97.6 97.6 94.1 96.5 5.9 63.5 60.0 1.2 1.2 96.5 97.6 97.6 Dental 94.5 84.7 88.5 91.3 78.1 79.2 95.1 94.5 94.5 95.6 94.5 91.8 91.3 33.3 87.4 86.9 87.4 94.5 87.4 95.1 94.5 91.3 91.3 88.0 86.9 10.9 86.9 66.7 80.9 82.5 83.1

* Missing data includes characters that cannot be scored or are not observed either because pertinent material was not examined or anatomical material is unknown.

References
` ` Ameghino, F., 1894. Enumeration synoptique des especes de mammiferes ` fossiles des formations Eogenes de Patagonie. Bol. Acad. Nac. Cien. Cordoba 13, 259e451. Ashley-Montague, M.F., 1933. The anthropological signicance of the pterion in the primates. Am. J. Phys. Anthropol. 13, 159e336. Barroso, C.M.L., Schneider, H., Schneider, M.P.C., Sampaio, I., Harada, M.L., Czelusniak, J., Goodman, M., 1997. Update on the phylogenetic systematics of New World monkeys: Further DNA evidence for placing pygmy marmoset (Cebuella) within the genus Callithrix. Int. J. Primatol. 18, 651e674. Bashir, A., Ye, C., Price, A.L., Bafna, V., 2005. Orthologous repeats and mammalian phylogenetic inference. Genome Res. 15, 998e1006. Beard, K.C., MacPhee, R.D.E., 1994. Cranial anatomy of Shoshonius and the antiquity of Anthropoidea. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 55e98. Berggren, W.A., Kent, D.V., Swisher III, C.C., Aubry, M.-P., 1995. A revised Cenozoic geochronology and chronostratigraphy. In: Berggren, W.A., Kent, D.V., Aubry, M.-P., Hardenbol, J. (Eds.), Geochronology, Time Scales, and Global Stratigraphic Correlation. Society for Sedimentary Geology, Tulsa, pp. 129e212. Bordas, A.F., 1939. Diagnosis sobre algunos mamiferos de als capas con Colpodon del valle del Rio Chubut. Physis 14, 413e433. Bordas, A.F., 1942. Anotaciones sobre un Cebidae fosil de Patagonia. Physis 19, 265e269. Bown, T.M., Larriestra, C.N., Powers, D.W., Naeser, C.W., Tabbutt, K., 1988. New information on age, correlation, and paleoenvironments of fossil platyrrhine sites in Argentina. J. Vert. Paleontol. 8, 9A.

Bremer, K., 1994. Branch support and tree stability. Cladistics 10, 295e304. Cartmill, M., 1978. The orbital mosaic in prosimians and the use of variable traits in systematics. Folia Primatol. 30, 89e114. Cartmill, M., Kay, R.F., 1978. Craniodental morphology, tarsier afnities, and primate suborders. In: Chivers, D.J., Joysey, K.A. (Eds.), Recent Advances in Primatology: Evolution. Academic Press, London, pp. 205e214. Cartmill, M., MacPhee, R.D.E., Simons, E.L., 1981. Anatomy of the temporal bone in early anthropoids with remarks on the problem of anthropoid origins. Am. J. Phys. Anthropol. 56, 3e22. Conroy, G., 1980. Evolutionary signicance of cerebral venous patterns in paleoprimatology. Z. Morphol. Anthropol. 71, 125e134. Conroy, G., 1981. Cranial asymmetry in ceboid primates: the emissary foramina. Am. J. Phys. Anthropol. 55, 187e194. Conroy, G.C., 1987. Problems of body-weight estimation in fossil primates. Int. J. Primatol. 8, 115e137. Cozzuol, M.A., Fordyce, R.E., Jones, C.M., 1991. La presencia de Eretiscus tonnii (Aves, Spheniscidae) en el Mioceno temprano de Nueva Zelandia y Patagonia. Ameghiniana 28, 406. Dagosto, M., Gebo, D.L., 1994. Postcranial anatomy and the origin of the Anthropoidea. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 567e594. Du Brul, E.L., 1965. The skull of the lion marmoset, Leontideus rosalia Linnaeus: A study in biomechanical adaptation. Am. J. Phys. Anthropol. 23, 261e276. Eizirik, E., Murphy, W.J., Springer, M., OBrien, S.J., 2004. Molecular phylogeny and dating of early primate divergences. In: Ross, C., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Kluwer Academic/Plenum Publishers, New York, pp. 45e64.

380

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Anthropoid Origins: New Visions. Kluwer Academic/Plenum Publishers, New York, pp. 649e682. Hershkovitz, P., 1949. Mammals of northern Colombia. Preliminary report no. 4: monkeys (Primates) with taxonomic revisions of some forms. Proc. U.S. Nat. Mus. 98, 323e427. Hershkovitz, P., 1970. Notes on Tertiary platyrrhine monkeys and description of a new genus from the late Miocene of Colombia. Folia Primatol. 12, 1e37. Hershkovitz, P., 1974. A new genus of late Oligocene monkey (Cebidae, Platyrrhini) with notes on postorbital closure and platyrrhine evolution. Folia Primatol. 21, 1e35. Hershkovitz, P., 1977. Living New World Monkeys (Platyrrhini) with an Introduction to Primates, vol. 1. University of Chicago Press, Chicago. Hershkovitz, P., 1981. Comparative anatomy of platyrrhine mandibular cheek teeth dpm4, pm4, m1, with particular reference to those of Homunculus (Cebidae), and comments on platyrrhine origins. Folia Primatol. 35, 179e217. Hershkovitz, P., 1982. Supposed squirrel monkey afnities of the late Oligocene Dolichocebus gaimanensis. Nature 298, 201e202. Hershkovitz, P., 1984. Taxonomy of squirrel monkeys genus Saimiri: a preliminary report with description of a hitherto unnamed form. Am. J. Primatol. 7, 155e210. Hershkovitz, P., 1990. Titis, New World monkeys of the genus Callicebus (Cebidae, Platyrrhini): A preliminary taxonomic review. Fieldiana Zool. 55, 1e109. Hill, W.C.O., 1957. Primates. Comparative Anatomy and Taxonomy III: Pithecoidea, Platyrrhini, Hapalidae. University Press, Edinburgh. Hillis, D.M., 1999. SINEs of the perfect character. Proc. Natl. Acad. Sci. U.S.A. 96, 9979e9981. Hoffstetter, R., 1980. Los monos platirrinos (Primates): origen, extension, lo genia, taxonomia. Actas II Congreso Argentino de Paleontologa y Bioes tratigrafa y I Congreso Latinoamericano de Paleontologa, 1978. Buenos Aires, pp. 291e303. Horovitz, I., 1997. Platyrrhine systematics and the origin of Greater Antilles monkeys. Ph.D. Dissertation, State University of New York. Horovitz, I., 1999. A phylogenetic study of living and fossil platyrrhines. Am. Mus. Novit. 3269, 1e40. Horovitz, I., MacPhee, R.D.E., 1999. The quaternary Cuban platyrrhine Paralouatta varonai and the origin of Antillean monkeys. J. Hum. Evol. 36, 33e68. Houle, A., 1999. The origin of platyrrhines: An evaluation of the Antarctic scenario and the oating island model. Am. J. Phys. Anthropol. 109, 541e559. Kay, R.F., 1977. The evolution of molar occlusion in Cercopithecidae and early catarrhines. Am. J. Phys. Anthropol. 46, 327e352. Kay, R.F., 1980. Platyrrhine origins: a reappraisal of the dental evidence. In: Ciochon, R., Chiarelli, B. (Eds.), Evolutionary Biology of the New World Monkeys and Continental Drift. Plenum Press, New York, pp. 159e188. Kay, R.F., 1990. The phyletic relationships of extant and fossil Pitheciinae (Platyrrhini, Anthropoidea). J. Hum. Evol. 19, 175e208. Kay, R.F., 1994. Giant tamarin from the Miocene of Colombia. Am. J. Phys. Anthropol. 95, 333e353. Kay, R.F., Campbell, V.M., Rossie, J.B., Colbert, M.W., Rowe, T., 2004a. The olfactory fossa of Tremacebus harringtoni (Platyrrhini, early Miocene, Sacanana, Argentina): Implications for activity pattern. Anat. Rec 281A, 1157e1172. Kay, R.F., Cozzuol, M.A., 2006. A new platyrrhine from the Solim~es Formao tion (late Miocene, Acre State, Brazil) with comments about other Miocene monkeys from that region. J. Hum. Evol. 50, 673e686. Kay, R.F., Delson, E., 2000. Oligopithecidae. In: Delson, E., Tattersall, I., Van Couvering, J.A., Brooks, A. (Eds.), Encyclopedia of Human Evolution and Prehistory, second ed. Garland Publishing Co., New York, pp. 490e 493. Kay, R.F., Fleagle, J.G., Simons, E.L., 1981. A revision of the African Oligocene apes of the Fayum Province. Egypt. Am. J. Phys. Anthropol. 55, 293e322. Kay, R.F., Johnson, D.J., Meldrum, D.J., 1998a. A new pitheciin primate from the middle Miocene of Argentina. Am. J. Primatol. 45, 317e336.

Fleagle, J.G., 1985. New primate fossils from Colhuehuapian deposits at Gaiman and Sacanana, Chubut Province, Argentina. Ameghiniana 21, 266e274. Fleagle, J.G., 1990. New fossil platyrrhines from the Pinturas Formation, southern Argentina. J. Hum. Evol. 19, 61e85. Fleagle, J.G., Bown, T.M., 1983. New primate fossils from late Oligocene (Colhuehuapian) localities of Chubut Province, Argentina. Folia Primatol. 41, 240e266. Fleagle, J.G., Bown, T.M., Swisher III, C.C., Buckley, G.A., 1995. Age of the Pinturas and Santa Cruz formations. VI Congreso Argentino de Paleontologia y Bioestratigraa, Trelew, pp. 129e135. Fleagle, J.G., Kay, R.F., 1983. New interpretations of the phyletic position of Oligocene hominoids. In: Ciochon, R.L., Corruccini, R.S. (Eds.), New Interpretations of Ape and Human Ancestry. Plenum Press, New York, pp. 181e210. Fleagle, J., Kay, R.F., 1987. The phyletic position of the Parapithecidae. J. Hum. Evol. 16, 483e532. Fleagle, J.G., Kay, R.F., 1989. The dental morphology of Dolichocebus gaimanensis, a fossil monkey from Argentina. Am. J. Phys. Anthropol. 78, 221. Fleagle, J.G., Kay, R.F., Anthony, M.R.L., 1997. Fossil New World monkeys. In: Kay, R.F., Madden, R.H., Cifelli, R.L., Flynn, J.J. (Eds.), Mammalian Evolution in the Neotropics. Smithsonian Institution Press, Washington, DC, pp. 473e495. Fleagle, J.G., Powers, D.W., Conroy, G.C., Watters, J.P., 1987. New fossil platyrrhines from Santa Cruz Province, Argentina. Folia Primatol. 48, 65e77. Fleagle, J.G., Rosenberger, A.L., 1983. Cranial morphology of the earliest anthropoids. In: Sakka, M. (Ed.), Morphologie Evolution Morphogenese du Crane et Origine de lHomme. Centre National de Recherche Scientique, Paris, pp. 141e155. Fleagle, J.G., Tejedor, M.F., 2002. Early platyrrhines of southern South America. In: Hartwig, W.C. (Ed.), The Primate Fossil Record. Cambridge University Press, Cambridge, pp. 161e173. Flynn, J.J., Swisher III, C.C., 1995. Cenozoic South American land mammal ages: correlation to global geochronologies. In: Berggren, W.A., Kent, D.V., Aubry, M.-P., Hardenbol, J. (Eds.), Geochronology, Time Scales, and Global Stratigraphic Correlation. Society for Sedimentary Geology, Tulsa, pp. 317e333. Flynn, J.J., Wyss, A.R., Charrier, R., Swisher, C.C., 1995. An early Miocene anthropoid skull from the Chilean Andes. Nature 373, 603e607. Ford, S.M., 1980. Callitrichids as phyletic dwarfs, and the place of the Callitrichidae in Platyrrhini. Primates 21, 31e43. Ford, S.M., 1986. Systematics of the New World monkeys. In: Swindler, D.R., Erwin, J. (Eds.), Comparative Primate Biology, Vol. 1: Systematics, Evolution, and Anatomy. Alan R. Liss, New York, pp. 73e135. Ford, S.M., 1994. Primitive platyrrhines? Perspectives on anthropoid origins from platyrrhine, parapithecid, and preanthropoid postcrania. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 595e676. Forsyth-Major, C.I., 1901. On some characters of the skull in the lemurs and monkeys. Proc. Zool. Soc. Lond. 1901, 129e153. Gebo, D., Simons, E.L., 1987. Morphology and locomotor adaptations of the foot in early Oligocene anthropoids. Am. J. Phys. Anthropol. 74, 83e101. Gebo, D.L., Simons, E.L., Rasmussen, D.T., Dagosto, M., 1994. Eocene anthropoid postcrania from the Fayum, Egypt. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 203e234. Harada, M.L., Schneider, H., Schneider, M.P.C., Sampaio, I., Czelusniak, J., Goodman, M., 1995. DNA evidence on the phylogenetic systematics of New World monkeys: Support for the sister grouping of Cebus and Saimiri from two unlinked nuclear genes. Mol. Phylogenet. Evol. 4, 331e349. Hartwig, W.C., 1994. Patterns, puzzles, and perspectives on platyrrhine origins. In: Corruccini, R.S., Ciochon, R. (Eds.), Integrative Paths to the Past: Paleoanthropological Advances in Honor of F. Clark Howell. Prentice Hall, Englewood Cliffs, pp. 69e94. Hauser, D.L., Presch, W., 1991. The effect of ordered characters on phylogenetic reconstruction. Cladistics 7, 243e265. Heesy, C.P., Ross, C., 2004. Mosaic evolution of activity pattern, diet, and color vision in haplorhine primates. In: Ross, C., Kay, R.F. (Eds.),

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Kay, R.F., Kirk, E.C., 2000. Ostological evidence for the evolution of activity pattern and visual acuity in primates. Am. J. Phys. Anthropol. 113, 235e262. Kay, R.F., Kirk, E.C., Malinzak, M., Colbert, M.W., 2006. Brain size, activity pattern, and visual acuity in Homunculus patagonicus, an early Miocene stem platyrrhine: the mosaic evolution of brain size and visual acuity in Anthropoidea. J. Vert. Paleontol. 26, 83Ae84A. Kay, R.F., MacFadden, B.J., Madden, R.H., Sandeman, H., Anaya, F., 1998b. Revised age of the Salla beds, Bolivia, and its bearing on the age of the Deseadan South American Land Mammal Age. J. Vert. Paleontol. 18, 189e199. Kay, R.F., Madden, R.H., Mazzoni, M., Vucetich, M.G., Re, G., Heizler, M., Sandeman, H., 1999b. The oldest Argentine primates: rst age determinations for the Colhuehuapian South American Land Mammal Age. Am. J. Phys. Anthropol. 28 (Suppl.), 166. Kay, R.F., Madden, R.M., Vucetich, M.G., Carlini, A.A., Mazzoni, M.M., Re, G.H., Heizler, M., Sandeman, H., 1999a. Revised age of the Casamayoran South American Land Mammal Agedclimatic and biotic implications. Proc. Natl. Acad. Sci. U.S.A. 96, 13235e13240. Kay, R.F., Meldrum, D.J., 1997. A new small platyrrhine from the Miocene of Colombia and the phyletic position of Callitrichinae. In: Kay, R.F., Madden, R.H., Cifelli, R.L., Flynn, J.J. (Eds.), Vertebrate Paleontology in the Neotropics. Smithsonian Institution Press, Washington, DC, pp. 435e458. Kay, R.F., Ross, C.F., Williams, B.A., 1997. Anthropoid origins. Science 275, 797e804. Kay, R.F., Ross, J., Simons, E.L., 2008. The basicranial anatomy of African Eocene/Oligocene anthropoids. Are there any clues for platyrrhine origins? In: Fleagle, J.G., Gilbert, C.C. (Eds.), Elwyn L. Simons: A Search for Origins. Springer, New York, pp. 125e158. Kay, R.F., Rossie, J.B., Colbert, M.W., Rowe, T., 2004b. Observations on the olfactory system of Tremacebus harringtoni (Platyrrhini, early Miocene, Sacanana, Argentina) based on high resolution X-ray CT scans. Am. J. Phys. Anthropol. 35 (Suppl.), 123e124. Kay, R.F., Simons, E.L., 1980. Comments on the adaptive strategy of the rst African Anthropoidea. Z. Morphol. Anthropol. 71, 143e148. Kay, R.F., Simons, E.L., 1983. Dental formulae and dental eruption patterns in Parapithecidae (Primates, Anthropoidea). Am. J. Phys. Anthropol. 62, 363e375. Kay, R.F., Vizcaino, S.F., Tauber, A., Bargo, M.S., Williams, B.A., Luna, C., Colbert, M.W., 2005. Three newly discovered skulls of Homunculus patagonicus support its position as a stem platyrrhine and establish its diurnal arboreal folivorous habits. Am. J. Phys. Anthropol. 40 (Suppl.), 127. Kay, R.F., Williams, B.A., 1994. Dental evidence for anthropoid origins. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 361e446. Kay, R.F., Williams, B.A., Anaya, F., 2001. The adaptations of Branisella boliviana, the earliest South American monkey. In: Plavcan, J.M., van Schaik, C., Kay, R.F., Jungers, W.L. (Eds.), Reconstructing Behavior in the Primate Fossil Record. Kluwer Academic/Plenum Publishers, New York, pp. 339e370. Kay, R.F., Williams, B.A., Ross, C.F., Takai, M., Shigehara, N., 2004c. Anthropoid origins: a phylogenetic analysis. In: Ross, C.F., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Kluwer Academic/Plenum Publishers, New York, pp. 91e135. Ketcham, R.A., Carlson, W.D., 2001. Acquisition, optimization, and interpretation of x-ray computed tomographic imagery: applications to the geosciences. Comp. Geosci. 27, 381e400. Kraglievich, J.L., 1951. Contribuciones al conocimiento de los primates fosiles de la Patagonia. I. Diagnosis previa de un nuevo primat fosil del Oligoceno superior (Colheuhuapiano) de Gaiman, Chubut. Comm. Inst. Nac. Inv. Cien. Nat. II, Cien. Zool. 11, 57e82. Le Gros Clark, W.E., 1959. The Antecedents of Man. Edinburgh University Press, Edinburgh. Lieberman, D.E., Ross, C., Ravosa, M.J., 2000. The primate cranial base: ontogeny, function, and integration. Yearb. Phys. Anthropol. 43, 117e169. Lockwood, C.A., Fleagle, J.G., 1999. The recognition and evaluation of homoplasy in primate and human evolution. Yearb. Phys. Anthropol. 42, 189e232.

381

Mabee, P., 1989. Assumptions underlying the use of ontogenetic sequences for determining character state order. Trans. Am. Fish. Soc. 118, 151e158. Mabee, P., 2000. The usefulness of ontogeny in interpreting morphological characters. In: Wiens, J.J. (Ed.), Phylogenetic Analysis of Morphological Data. Smithsonian Institution Press, Washington, D. C, pp. 84e114. MacFadden, B.J., 1990. Chronology of Cenozoic primate localities in South America. J. Hum. Evol. 19, 7e22. MacFadden, B.J., Campbell, K.E., Cifelli, R.L., Siles, O., Johnson, N.M., Naeser, C.W., Zeitler, P.K., 1985. Magnetic polarity stratigraphy and mammalian fauna of the Deseadan (late Oligocene-early Miocene) Salla Beds of northern Bolivia. J. Geol. 93, 223e250. MacPhee, R.D.E., Cartmill, M., 1986. Basicranial structures and primate systematics. In: Swindler, D.R., Erwin, J. (Eds.), Comparative Primate Biology, Vol. 1: Systematics, Evolution, and Anatomy. Alan R. Liss, New York, pp. 219e275. MacPhee, R.D.E., Horovitz, I., Arredondo, O., Jimenez Vazquez, O., 1995. A new genus for the extinct Hispaniolian monkey Saimiri bernensis Rmoli, 1977, with notes on its systematic position. Am. Mus. Novit. 3134, 1e21. Maddison, D.R., Maddison, W.P., 2001. MacClade 4: Analysis of Phylogeny and Character Evolution. Version 4.03. Sinauer Associates, Sunderland. Maier, W., 1980. Nasal structures in old and new world primates. In: Ciochon, R., Chiarelli, B. (Eds.), Evolutionary Biology of the New World Monkeys and Continental Drift. Plenum Press, New York, pp. 219e241. Marshall, L.G., 1985. Geochronology and land-mammal biochronology of the transamerican faunal interchange. In: Stehli, F.G., Webb, S.D. (Eds.), The Great American Biotic Interchange. Plenum Press, New York, pp. 49e85. Martin, R.D., 1990. Primate Origins and Evolution: A Phylogenetic Reconstruction. Princeton University Press, Princeton, New Jersey. Meireles, C.M., Czelusniak, J., Ferreira, H.S., Schneider, M.P.C., Goodman, M., 1999a. Phylogenetic relationships among Brazilian howler monkeys, genus Alouatta (Platyrrhini, Anthropoidea), based on g1-globin pseudogene sequences. Genet. Mol. Biol. 22, 337e344. Meireles, C.M., Czelusniak, J., Schneider, M.P.C., Muniz, J.A.P.C., Brigido, M.C., Ferreira, H.S., Goodman, M., 1999b. Molecular phylogeny of ateline New World monkeys (Platyrrhini, Atelinae) based on gammaglobin gene sequences: Evidence that Brachyteles is the sister group of Lagothrix. Mol. Phylogenet. Evol. 12, 10e30. Meldrum, D.J., 1993. Postcranial adaptations and positional behavior in fossil platyrrhines. In: Gebo, D.L. (Ed.), Postcranial Adaptations in Nonhuman Primates. Northern Illinois University Press, DeKalb, pp. 235e251. Meldrum, D.J., Kay, R.F., 1997. Nuciruptor rubricae, a new pitheciin seed predator from the Miocene of Colombia. Am. J. Phys. Anthropol. 102, 407e427. Mickevich, M.F., 1982. Transformation series analysis. Syst. Zool. 31, 461e 478. Mickevich, M.F., Weller, S.J., 1990. Evolutionary character analysis: tracing character change on a cladogram. Cladistics 6, 137e170. Mouri, T., 1988. Intergeneric comparison of ceboid cranial measurements. Kyoto University Overseas Research Reports of New World Monkeys VI, pp. 23e31. Nixon, K.C., Davis, J.I., 1991. Polymorphic taxa, missing values and cladistic analyses. Cladistics 7, 233e241. Pascual, R., Odreman Rivas, O., 1971. Evolucion de las comunidades de los vertebrados del terciario Argentina, los aspectos paleozoogeogracos y paleoclimaticos relacionados. Ameghiniana 7, 372e412. Patterson, B., Pascual, R., 1972. The fossil mammal fauna of South America. In: Keast, A., Erk, F.C., Glass, B. (Eds.), Evolution, Mammals, and Southern Continents. State University of New York Press, Albany, pp. 247e309. Pocock, R.I., 1925. Additional notes on the external characters of some platyrrhine monkeys. Proc. Zool. Soc. Lond. 1925, 27e47. Poux, C., Chevret, P., Huchon, D., de Jong, W.W., Douzery, E., 2006. Arrival and diversication of caviomorph rodents and platyrrhine primates in South America. Syst. Zool. 55, 228e244. Radinsky, L., 1977. Early primate brains: fact and ction. J. Hum. Evol. 6, 79e86. Ravosa, M.J., Hogue, A., 2003. Function and fusion of the mandibular symphysis in mammals: a comparative and experimental perspective.

382

R.F. Kay et al. / Journal of Human Evolution 54 (2008) 323e382 Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Kluwer Academic/ Plenum Publishers, New York, pp. 157e182. Simmons, N., 1993. The importance of methods: Archontan phylogeny and cladistic analysis of morphological data. In: MacPhee, R.D.E. (Ed.), Primates and Their Relatives in Phylogenetic Perspective. Plenum Press, New York, pp. 1e61. Simons, E.L., 1993. New endocasts of Aegyptopithecus: oldest well-preserved record of the brain in Anthropoidea. Am. J. Sci. 293-A, 383e390. Simons, E.L., 2001. The cranium of Parapithecus grangeri, an Egyptian Oligocene anthropoidean primate. Proc. Natl. Acad. Sci. U.S.A. 98, 7892e7897. Simons, E.L., 2004. The cranium and adaptations of Parapithecus grangeri, a stem anthropoid from the Fayum Oligocene of Egypt. In: Ross, C., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Kluwer Academic/ Plenum Publishers, New York, pp. 183e217. Simons, E.L., Plavcan, J.M., Fleagle, J.G., 1999. Canine sexual dimorphism in Egyptian Eocene anthropoid primates: Catopithecus and Proteopithecus. Proc. Natl. Acad. Sci. U.S.A. 96, 2559e2562. Simons, E.L., Rasmussen, D.T., 1996. Skull of Catopithecus browni, an early Tertiary catarrhine. Am. J. Phys. Anthropol. 100, 261e292. Simons, E.L., Seiffert, E.R., Ryan, T.M., Attia, Y., 2007. A remarkable female cranium of the early Oligocene anthropoid Aegyptopithecus zeuxis (Catarrhini, Propliopithecidae). Proc. Natl. Acad. Sci. U.S.A. 104, 8731e8736. Simpson, G.G., 1935. Early and middle Tertiary geology of the Gaiman region, Chubut, Argentina. Am. Mus. Novit. 775, 1e29. Simpson, G.G., 1961. Principles of Animal Taxonomy. Columbia University Press, New York. Simpson, G.G., 1980. Splendid Isolation: The Curious History of South American Mammals. Yale University Press, New Haven. Singer, S.S., Schmitz, J., Schwiegk, C., Zischler, H., 2002. Molecular cladistic markers in New World monkey phylogeny (Platyrrhini, Primates). Mol. Phylogenet. Evol. 26, 490e501. Slowinski, J.B., 1993. Unordered versus ordered characters. Syst. Biol. 42, 155e165. Springer, M., Teeling, E., Madsen, O., Stanhope, M., deJong, W.W., 2001. Integrated fossil and molecular data reconstruct bat echolocation. Proc. Natl. Acad. Sci. U.S.A. 98, 6241e6246. Swofford, D.L., 2002. PAUP. Phylogenetic Analysis Using Parsimony, Version 4.0b10 (Altivec). Sinauer Associates, Sunderland. Szalay, F.S., Delson, E., 1979. Evolutionary History of the Primates. Academic Press, New York. Takai, M., Anaya, F., Shigehara, N., Setoguchi, T., 2000. New fossil materials of the earliest New World monkey, Branisella boliviana, and the problem of platyrrhine origins. Am. J. Phys. Anthropol. 111, 263e281. Tarling, D.H., 1980. The geologic evolution of South America with special reference to the last 200 million years. In: Ciochon, R.L., Chiarelli, B. (Eds.), Evolutionary Biology of the New World Monkeys and Continental Drift. Plenum Press, New York, pp. 1e42. Tauber, A., 1991. Homunculus patagonicus Ameghino, 1891 (Primates, Ceboidea), Mioceno Temprano, de la costa Atlantica Austral, Prov. de Santa Cruz, Republica Argentina. Acad. Nac. Cien. Cordoba, Argentina 82, 1e32. Tauber, A.A., Kay, R.F., Luna, C., 2004. Killik Aike Norte, una localidad clasica de la Formacion Santa Cruz (Mioceno temprano-medio), Patagonia, Argentina. Ameghiniana Resumenes 2004, 26. Tejedor, M.F., Tauber, A.A., Rosenberger, A.L., Swisher III, C.C., Palacios, M.E., 2006. New primate genus from the Miocene of Argentina. Proc. Natl. Acad. Sci. U.S.A. 103, 5437e5441. Wall, C.E., 1998. The expanded mandibular condyle of the Megaladapidae. Am. J. Phys. Anthropol. 103, 263e276. Williams, B.A., 2007. Comparing levels of homoplasy in the primate skeleton. J. Hum. Evol. 52, 480e489. Williams, B.A., Covert, H.H., 1994. New early Eocene anaptomorphine primate (Omomyidae) from the Washakie Basin, Wyoming, with comments on the phylogeny and paleobiology of anaptomorphines. Am. J. Phys. Anthropol. 93, 323e340.

In: Ross, C., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Kluwer Academic/Plenum Publishers, New York, pp. 413e462. Ray, D.A., Xing, J., Hedges, D.J., Hall, M.A., Laborde, M.E., Anders, B.A., White, B.R., Stoilova, N., Fowlkes, J.D., Landry, K.E., Chemnik, L.G., Ryder, O.A., Batzer, M.A., 2005. Alu insertion loci and platyrrhine primate phylogeny. Mol. Phylogenet. Evol. 35, 117e126. Rosenberger, A.L., 1978. Loss of incisor enamel in marmosets. J. Mammal. 59, 207e208. Rosenberger, A.L., 1979a. Cranial anatomy and implications of Dolichocebus, a late Oligocene ceboid primate. Science 279, 416e418. Rosenberger, A.L., 1979b. Phylogeny, evolution and classication of New World monkeys (Platyrrhini, Primates). Ph.D. Dissertation, City University of New York. Rosenberger, A.L., 1980. Gradistic views and adaptive radiation of platyrrhine primates. Z. Morphol. Anthropol. 71, 157e163. Rosenberger, A.L., 1981. Systematics: the higher taxa. In: Coimbra-Filho, A.F., Mittermeier, R.A. (Eds.), Ecology and Behavior of Neotropical Primates, vol. I. Academia Brasileira de Ciencias, Rio de Janeiro, pp. 9e27. Rosenberger, A.L., 1982. Supposed squirrel monkey afnities of the late Oligocene Dolichocebus gaimanensis, reply to P. Hershkovitz. Nature 298, 202. Rosenberger, A.L., 1983. Tale of tails: parallelism and prehensility. Am. J. Phys. Anthropol. 60, 103e107. Rosenberger, A.L., 1988. Platyrrhini. In: Tattersall, I., Delson, E., Van Couvering, J., Brooks, A. (Eds.), Encyclopedia of Human Evolution and Prehistory. Garland Publishing, New York, pp. 456e459. Rosenberger, A.L., 1992. Evolution of New World monkeys. In: Jones, S., Martin, R., Pilbeam, D.R. (Eds.), The Cambridge Encyclopedia of Human Evolution. Cambridge University Press, Cambridge, pp. 209e216. Rosenberger, A.L., 2000. Pitheciinae. In: Delson, E., Tattersal, I., Van Couvering, J., Brooks, A. (Eds.), Encyclopedia of Human Evolution and Prehistory, second ed. Garland Publishing Co, New York, pp. 562e563. Rosenberger, A.L., 2002. Platyrrhine paleontology and systematics: the paradigm shifts. In: Hartwig, W.C. (Ed.), The Primate Fossil Record. Cambridge University Press, Cambridge, pp. 151e159. Rosenberger, A.L., Setoguchi, T., Shigehara, N., 1990. The fossil record of callitrichine primates. J. Hum. Evol. 19, 209e236. Ross, C.F., 1993. The functions of the postorbital septum and anthropoid origins. Ph.D. Dissertation, Duke University. Ross, C.F., 1994. The craniofacial evidence for anthropoid and tarsier relationships. In: Fleagle, J.G., Kay, R.F. (Eds.), Anthropoid Origins: The Fossil Evidence. Plenum Press, New York, pp. 469e547. Ross, C.F., 1995. Allometric and functional inuences on primate orbit orientation and the origins of the Anthropoidea. J. Hum. Evol. 29, 210e228. Ross, C., Williams, B.A., Kay, R.F., 1998. Phylogenetic analysis of anthropoid relationships. J. Hum. Evol. 35, 221e306. Rossie, J.B., 2006. Ontogeny and homology of the paranasal sinuses in Platyrrhini (Mammalia: Primates). J. Morphol. 206, 1e40. Roth, S., 1899. Aviso preliminary sobre mamiferos mesozoicos encontrados en Patagonia. Rev. Mus. de La Plata 9, 381e388. ` Saban, R., 1963. Contribution a letude de los temporal des Primates. De scription chez lHomme et les Prosimiens. Anatomie comparee et phyloge nie. Mem. Mus. Natl. Hist. Nat. (France) N. Ser. A. Zool. 29, 1e378. Sanchez-Villagra, M., Williams, B.A., 1998. Levels of homoplasy in the evolution of the mammalian skeleton. J. Mammal. Evol. 5, 113e126. Schneider, H., Canavez, F.C., Sampaio, I., Moreira, M.A.M., Tagliaro, C.H., Seuanez, H.N., 2001. Can molecular data place each Neotropical monkey in its own branch? Chromosoma 109, 515e523. Schneider, H., Sampaio, I., Harada, M.L., Barroso, C.M.L., Schneider, M.P.C., Czelusniak, J., Goodman, M., 1996. Molecular phylogeny of the New World monkeys (Platyrrhini, Primates) based on two unlinked nuclear genes: IRBP interon 1 and e-globin sequences. Am. J. Phys. Anthropol. 100, 153e179. Seiffert, E.R., Simons, E.L., Simons, C.V.M., 2004. Phylogenetic, biogeographic, and adaptive implications of new fossil evidence bearing on crown anthropoid origins and early stem catarrhine evolution. In: Ross, C.,