Tail Architecture and Function of Cebupithecia sarmientoi a Middle Miocene Platyrrhine from La Venta Colombia.код для вставкиСкачать
THE ANATOMICAL RECORD 294:2013–2023 (2011) Tail Architecture and Function of Cebupithecia sarmientoi, a Middle Miocene Platyrrhine From La Venta, Colombia JASON M. ORGAN1,2* AND PIERRE LEMELIN3 Department of Surgery, Saint Louis University School of Medicine, St. Louis, Missouri 2 Center for Anatomical Science and Education, Saint Louis University School of Medicine, St. Louis, Missouri 3 Division of Anatomy, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada T6G 2H7 1 ABSTRACT Cebupithecia sarmientoi, an early pitheciine from the Middle Miocene of La Venta, Colombia, preserves an almost complete caudal vertebral sequence (18 vertebrae). Behavioral reconstructions for this taxon based on appendicular elements suggest a locomotor proﬁle similar to that of Pithecia for which vertical clinging postures and leaping behavior are frequently adopted. General tail morphology suggests some similarity with prehensile-tailed Cebus in the proximal tail region, although overwhelming similarity with nonprehensile-tailed Pithecia is evident in the distal tail region. Indices of caudal muscle attachment sites show marked similarities to nonprehensile-tailed platyrrhines, especially Pithecia. However, the cortices of Cebupithecia caudal vertebral bodies are thicker than those of most other nonprehensile-tailed New World primates. Mechanically, this would provide high resistance to bending and torsional stresses, falling within the range exhibited by prehensile-tailed monkeys. These results suggest that Cebupithecia may have employed its tail differently than most nonprehensile-tailed platyrrhines living today, behaviors that possibly involved tail-bracing or twisting during hindlimb (pedal grasping) suspensory behaviors. Such behaviors may serve as a preadaptive model for the full-ﬂedged evolution of below-branch tail suspension and prehensility seen in other New World primates. Anat Rec, 294:2013–2023, C 2011 Wiley Periodicals, Inc. 2011. V Key words: prehensile tail; caudal vertebrae; cross-sectional geometry; section modulus; pQCT; Platyrrhini; New World monkeys; primate evolution INTRODUCTION The holotype specimen of Cebupithecia sarmientoi (UCMP 38762)—an early pitheciine from the Middle Miocene deposits of the Villavieja Formation (Honda Group), La Venta, Colombia (12.5 ma) (Stirton, 1951; Stirton and Savage, 1951)—is 70% complete (Fig. 1) and preserves a nearly complete caudal vertebral sequence (18 vertebrae), affording the rare opportunity to examine tail architecture and function in a fossil New World monkey (Meldrum and Lemelin, 1991). Locomotor reconstructions of Cebupithecia based on limb evidence C 2011 WILEY PERIODICALS, INC. V Grant sponsor: National Science Foundation; Grant number: BCS-0550676; Grant sponsors: Johns Hopkins University, Saint Louis University. *Correspondence to: Jason M. Organ, Center for Anatomical Science and Education, Department of Surgery, Saint Louis University School of Medicine, 1402 S. Grand Blvd., M306, St. Louis, MO 63104. Fax: 314-977-5127. E-mail: firstname.lastname@example.org Received 15 September 2011; Accepted 16 September 2011 DOI 10.1002/ar.21504 Published online 1 November 2011 in Wiley Online Library (wileyonlinelibrary.com). 2014 ORGAN AND LEMELIN Fig. 1. The composite postcranial skeleton of Cebupithecia sarmientoi (UMCP 38762), modiﬁed from photo courtesy of Jeff Meldrum and John Fleagle. 2015 CEBUPITHECIA TAIL MORPHOLOGY suggest a combination of frequent clinging postures and leaping, similar to that observed for Pithecia (e.g., Davis, 1987; Fleagle and Meldrum, 1988; Ford, 1990; Meldrum et al., 1990; Meldrum and Lemelin, 1991; Meldrum, 1993), but without tarsal specializations indicative of hindlimb suspension typical of Chiropotes (Fleagle and Meldrum, 1988; Meldrum, 1993, 1998; but see Ford, 1990). Tail anatomy shows some similarity with prehensile-tailed Cebus1 more proximally; more distally, the tail exhibits overwhelming resemblance with nonprehensiletailed Pithecia (Meldrum and Lemelin, 1991). Ultimately, the tail of Cebupithecia was diagnosed as nonprehensile (Meldrum and Lemelin, 1991). These inferences about tail function in Cebupithecia are based on external dimensions without consideration of the internal architecture of the caudal vertebrae. Recent analysis of caudal vertebral structure among extant platyrrhines indicates that caudal vertebrae with relatively higher resistance to torsional and bending moments are found in prehensile-tailed species (Organ, 2010). The cortices of the caudal vertebrae of Cebupithecia appear thicker than most other nonprehensile-tailed platyrrhines (pers. obs.), suggesting they may be capable of resisting higher torsional and bending stresses. In light of the similarities to the Cebus tail in the proximal region and the apparently thicker cortices of the caudal vertebral bodies, the overall mechanical structure of the Cebupithecia tail requires re-examination. More speciﬁcally, this study tests the hypothesis that the thicker cortices of the Cebupithecia caudal vertebrae translate into stronger vertebrae. If the caudal vertebrae of Cebupithecia are indeed stronger in bending and torsion than their external dimensions would predict, it might suggest this taxon employed its tail in some form of bracing and/or suspensory (bending) behavior. MATERIALS AND METHODS Sample The caudal vertebrae of Cebupithecia were compared with a sample of extant platyrrhines from nine genera (four prehensile-tailed and ﬁve nonprehensile-tailed) totaling 70 individuals (Table 1). Much of this comparative sample is described in detail in Organ (2010). In 1 The tail of Cebus has been called ‘‘semiprehensile’’ by some authors (e.g., Napier and Napier, 1967; Emmons and Gentry, 1983) in order to distinguish it from the more derived prehensile tail found in atelines. Like the term ‘‘semibrachiation’’ (Napier and Napier, 1967), ‘‘semiprehensile’’ is uninformative and misleading as it may imply that the tail of Cebus represents an intermediate condition between a nonprehensile tail found in most primates and the prehensile tail of atelines. Although there are clear behavioral and morphological differences between the prehensile tails of atelines and Cebus (Rosenberger, 1983; Rosenberger and Strier, 1989; Lemelin, 1995; Bergeson, 1996; Organ, 2007, 2010; Organ et al., 2009, 2011), these lines of evidence support the contention of parallel evolution of this remarkable organ (i.e., tail prehensility evolved independently at least twice in New World primates). Moreover, mechanical demands placed upon the musculoskeletal elements of the ateline and Cebus tails result in strong structural similarities in caudal vertebrae and musculature (Organ, 2007, 2010; Organ et al., 2009), and both types of tails are capable of suspending the entire body weight of the animal (Bergeson, 1996). Thus, in the present mechanical analysis, the tail of Cebus is categorized and treated as being prehensile. TABLE 1. Comparative sample of extant platyrrhines Species Ateles fusciceps (N ¼ 6) Lagothrix lagotricha (N ¼ 8) Alouatta palliata (N ¼ 8) Cebus apella (N ¼ 8) Saimiri sciureus (N ¼ 8) Pithecia pithecia (N ¼ 7) Pithecia monachus (N ¼ 5) Chiropotes satanas (N ¼ 5) Aotus trivirgatus (N ¼ 7) Saguinus oedipus (N ¼ 7) Body Massa (g) Tail Type M: 8,890 F: 9,160 M: 7,280 F: 7,020 M: 5,959 F: 4,619 M: 3,650 F: 2,520 M: 779 F: 662 M: 1,940 F: 1,580 M: 2,610 F: 2,110 M: 3,100 F: 2,960 M: 813 F: 736 M: 418 F: 404 Prehensile Prehensile Prehensile Prehensile Nonprehensile Nonprehensile Nonprehensile Nonprehensile Nonprehensile Nonprehensile a All body masses taken from Smith and Jungers (1997) except Alouatta, taken from Organ (2010). that study, structural differences in the proximal caudal vertebrae were found between species of Pithecia, which may be associated with differences in tail-use behavior. However, the morphologic differences were based on small sample sizes for each Pithecia species. To this end, the present analysis adds nine individuals to the previous dataset, expanding the number of Pithecia pithecia and Pithecia monachus individuals in the sample, while also adding individuals of Chiropotes. The inclusion of Chiropotes in particular is important because this taxon may spend a signiﬁcant proportion of time in hindlimb suspension with the middle section of the tail draped over the substrate (i.e., tail-bracing) (Fleagle and Meldrum, 1988; van Roosmalen et al., 1988; Walker, 1993, 1996; Meldrum, 1998). Mechanical Structure of the Tail General tail morphology. The overall structural morphology of the tail of Cebupithecia was assessed ﬁrst by recording the total number of caudal vertebrae in each of the three regions of the tail: proximal, transitional, and distal (Fig. 2; see Organ (2010) for a complete review of tail osteology). Caudal vertebrae from the proximal tail region differ dramatically in morphology from those of the transitional and distal tail regions. In particular, these caudal vertebrae bear ventral and neural arches and a single pair of transverse processes and articulate with one another by zygapophyseal and intervertebral disc joints (in a fashion similar to lumbar vertebrae). Caudal vertebrae from transitional and distal tail regions are distinct from proximal ones by possessing two pair of transverse processes and only intervertebral disc articulations. Differences in intervertebral articular morphology contribute to differences in tail movements and ﬂexibility. In short, movements of the proximal region of the tail are primarily limited to ﬂexion-extension in a sagittal plane with more limited movements in other planes (Organ, 2010; see also Wada 2016 ORGAN AND LEMELIN Fig. 2. Schematic representations of the primate tail, showing distinctions between the prehensile tail of an ateline (A); prehensile tail of Cebus (B); and, a nonprehensile-tailed platyrhrine (C). Blue, proximal region; green, transitional region; tan, distal region; TV, transition vertebra; LV, longest caudal vertebra; MDV, mid-distal vertebra. See text for description of TV, LV, and MDV. (Reproduced with permission from Organ et al., 2009). et al., 1993), while the planes of motion between the transitional and distal region vertebrae are more varied. Prehensile-tailed platyrrhines tend to have longer proximal tail regions with more numerous vertebral elements than nonprehensile-tailed platyrrhines (Ankel, 1962, 1972; German, 1982; Schmitt et al., 2005; Organ, 2010), which increases movement and ﬂexibility of the proximal tail region (i.e., more zygapophyseal joints in series; see Ward, 1991, 1993; Shapiro, 1993, 1995). Once the relative number of vertebrae in each tail section was compiled, a single vertebra from each region was chosen for external dimensional (process expansion indices) and internal architectural (cross-sectional geometry) analyses: transition vertebra (TV) from the proximal region, longest caudal vertebra (LV) from the transitional region, and mid-distal vertebra (MDV) from the distal region of the tail. These vertebrae represent functionally analogous points within each tail that can be compared across taxa (Organ et al., 2009; Organ, 2010). These vertebrae were chosen because they likely experience different mechanical loads in prehensile and nonprehensile activities. The TV was chosen because it represents the vertebral element where possible movements of the tail change from being oriented primarily in ﬂexion/extension in a sagittal plane to being more multidirectional (including ﬂexion, extension, and axial rotation) (Organ, 2010; see also Wada et al., 1993). The LV is the longest vertebra in the caudal sequence and, therefore, is likely to experience higher bending moments than other caudal vertebrae; the MDV is located in the area where tail-wrapping around a substrate most likely occurs in most prehensiletailed platyrrhines (Organ, 2010). vertebrae in the distal region of the tail. If the total number of vertebrae is unknown, the placement of the MDV is less certain. Therefore, to analyze the structure of the MDV in Cebupithecia, it was necessary to estimate the total number of caudal vertebrae that would have been present in the living individual. Three reconstructions of total caudal vertebral number were used in this analysis. First, the tail of Cebupithecia was reconstructed to include 25 total caudal vertebrae, the purported primitive condition for primates (Schultz and Straus, 1945; Stirton, 1951). Next, because of its numerous postcranial similarities to Pithecia, the tail of Cebupithecia was reconstructed to a total of 23 caudal vertebrae, representing the average number of vertebrae in Pithecia (Organ, 2010). Finally, a prehensile tail model of 27 total caudal vertebrae was used, as it represents the average number of caudal vertebrae among all ateline and Cebus individuals in the sample (Organ, 2010). It is important, again, to note that only the distalmost vertebrae are missing from the Cebupithecia material. Therefore, choice of reconstruction method did not affect the position of the morphologically deﬁned TV and LV. Only the location of the MDV was affected by the choice of reconstruction method, and even then, only by a small margin. Cebupithecia Tail Reconstruction Methods The initial descriptions of the Cebupithecia axial skeleton identiﬁed 17 caudal vertebrae (Stirton, 1951; Stirton and Savage, 1951). A full analysis of the axial skeleton conducted by Meldrum and Lemelin (1991) identiﬁed an additional caudal vertebra, bringing the number to 18. However, the total number of caudal vertebrae for this taxon is still unknown, as the distal-most elements were not recovered. Missing distal elements of the tail does not present challenges to analyzing the TV and LV, as they are morphologically deﬁned based on articular morphology and length. However, the placement of the MDV within the vertebral sequence is determined by the total number of Indices of Caudal Muscle Attachments Caudal musculature of platyrrhine prehensile tails is more robust than that of nonprehensile tails, particularly in the ventral (ﬂexor) and lateral muscle compartments (Lemelin, 1995; Organ et al., 2009). Therefore, previous morphometric analyses of the external geometry of caudal vertebrae focused on the relative expansion of muscle attachment sites (especially transverse processes) as a generally successful means of discriminating between prehensile and nonprehensile vertebrae (German, 1982; Meldrum and Lemelin, 1991; Youlatos, 2003; Schmitt et al., 2005; Organ, 2010). Transverse processes serve as the sites of origin and insertion of mm. intertransversarii caudae, the prime lateral ﬂexors and rotators of the tail (Dor, 1937; Lemelin, 1995; Organ et al., 2009; Organ, 2010), as well as the ﬂeshy origins (in the proximal tail) of mm. ﬂexor caudae longus et brevis, the primary ﬂexors of the tail. An index of relative transverse process expansion was calculated to reﬂect the moment arm of lateral ﬂexion of mm. intertransversarii caudae and the robustness of mm. ﬂexor caudae longus et brevis (Organ, 2010). CEBUPITHECIA TAIL MORPHOLOGY The distal (tendinous) attachment site for mm. ﬂexor caudae longus et brevis on the hemal arch (which articulates with the hemal process on the proximoventral aspect of the caudal vertebra) is not shared with other muscles. The relative expansion of the hemal process discriminates prehensile from nonprehensile vertebrae better than indices of transverse process expansion, in part because transverse processes are inherently tied to muscle belly size in addition to mechanical advantage (Organ, 2010). Therefore, an index of hemal process expansion was also calculated to reﬂect the dorsoventral bending moment arm of mm. ﬂexor caudae longus et brevis, without the potentially confounding inﬂuence of muscle size (Organ, 2010). Three linear measurements of the TV, LV, and MDV were collected with Mitutoyo 15-cm sliding digital calipers (Mitutoyo Corp. USA, Aurora, IL) to the nearest 0.1 mm (Organ, 2010): a. Craniocaudal length (CCL): maximum craniocaudal length of the vertebra, measured in lateral view. b. Proximal transverse process breadth (PTPB): maximum breadth of the proximal transverse process, measured in superior view. c. Hemal process height (HPH): maximum height of the hemal process, measured from proximal transverse process to hemal process, in proximal view. These measurements were used to calculate transverse process expansion (TPE) and hemal process expansion (HPE) indices, following Organ (2010): a. TPE ¼ PTPB/CCL 100 b. HPE ¼ HPH/CCL 100 Cross-Sectional Geometric Methods All cross-sectional properties of caudal vertebrae exhibit similar patterns of variation across taxa (e.g., cross-sectional areas, second moments of area, section moduli) (Organ, 2007; Organ and Lemelin, 2009). This study focuses on the polar section modulus (Zp)—a direct measure of torsional strength and a proximate measure of (twice) average bending strength (Ruff, 2002; Organ, 2010)—because previous studies suggest it is the most appropriate singular measure of overall mechanical strength of a bony element (Ruff, 2002; Organ, 2010). Cross-sectional images of the TV, LV, and MDV of 61 individual monkeys were obtained at 50% of vertebral craniocaudal length with a Norland-Stratec XCT Research SAþ peripheral quantitative computed tomography (pQCT) scanner (Stratec Medizintechnik GmbH, Pforzheim, Germany) using the following parameters: pixel size ¼ 0.1 mm2, slice thickness ¼ 1 mm, scan speed ¼ 3 mm/sec. Polar section moduli were calculated from pQCT scans using native scanner software (XCT software 5.40), with a density threshold of 500 mg/cm3 to isolate cortical from cancellous bone and an internal software algorithm designed to segment bone types (described in detail in Organ, 2010). The Ellipse Model Method (EMM) of reconstructing cross-sectional geometry with biplanar radiographs (Ohman, 1993; Runestad et al., 1993; Lazenby, 1997, 2017 Fig. 3. Diagrams of measurements used to reconstruct vertebral cross-sectional properties using the Ellipse Model Method (EMM). Total subperiosteal and medullary breadths were measured from anteroposterior and mediolateral projection radiographs and used to calculate section properties using formulae found in the text. Only the AP projection is illustrated here. 1998; O’Neill and Ruff, 2004) was used instead of pQCT to estimate the polar section modulus of the caudal vertebrae of Cebupithecia and the nine pitheciine individuals added to the comparative dataset. In brief, this method uses cortical thickness of the vertebrae measured from orthogonal radiographs (posteroanterior and lateral projections) to geometrically derive cross-sectional geometric properties for a hollow ellipse using standard equations (Fig. 3) (O’Neill and Ruff, 2004): a. Second moment of area about mediolateral axis: Ix ¼ p/64[(AP3ML) – (ap3ml)]. b. Second moment of area about anteroposterior axis: Iy ¼ p/64[(APML3) – (apml3)]. c. Polar second moment of area: J ¼ Ix þ Iy. d. Polar section modulus: Zp ¼ J/(average diameter/ 2). Biplanar radiographs of the nine specimens of Pithecia and Chiropotes were captured at the Field Museum of Natural History (Chicago, IL) with a Philips Electronic Instruments, Radiﬂuor 120 (TORR X-Ray Corp., East Sussex, UK) x-ray source onto Kodak MX125 Ready Pack Film using a voltage of 50 kV and a current of 3.0 mA. The caudal vertebrae of Cebupithecia were radiographed with a Thermo Scientiﬁc Kevex PXS5-724EA MicroFocus (Thermo Fisher Scientiﬁc, Inc., Waltham, MA) portable x-ray source mounted above a PaxScan 4030 (Varian Medical Systems, Palo Alto, CA) amorphous-silicon digital imaging panel at the Division of Fishes, National Museum of Natural History, Smithsonian Institution (Washington, DC) using a voltage of 68 kV and a current of 0.098 mA. One disadvantage of EMM in reconstructing cross-sectional geometry is it tends to overestimate section properties compared with other noninvasive methods (O’Neill and Ruff, 2004). Therefore, in order to compare data collected with pQCT and EMM, a small subset (N ¼ 17) of the pQCT-scanned vertebrae was also examined using EMM in order to validate the pooling of data collected with different methods. 2018 ORGAN AND LEMELIN Fig. 4. Contributions from each tail region as a percentage (%) of total tail length in extant platyrrhines and in Cebupithecia sarmientoi. The values for Cebupithecia are based on two reconstruction models of total vertebral number: (A) primitive reconstruction model of 25 total caudal vertebrae and (B) Pithecia reconstruction model of 23 total caudal vertebrae. Comparison of Cebupithecia to Extant Sample Caudal Muscle Development Indices Statistical evaluation of caudal vertebral metrics for 61 individuals of the extant sample are presented in detail elsewhere (Organ, 2010). The additional specimens of extant piithecines included in this study do not change the overall patterns of species distributions or conclusions of the earlier analysis. Thus, because the purpose of the present study is to evaluate the Cebupithecia tail against the backdrop of the extant sample, this study does not rehash the statistical particulars of the Organ (2010) study. Instead, data for Cebupithecia caudal vertebrae are presented in the context of the extant data previously analyzed. Indices for three caudal vertebrae (TV, LV, and MDV) that can be used as proxies for caudal muscle development are presented in Figure 5. As previously reported (German, 1982; Youlatos, 2003; Organ, 2010), prehensiletailed taxa have generally more expanded transverse and hemal processes than nonprehensile-tailed taxa, with the differences becoming more drastic further distally within the vertebral sequence. Hemal process expansion of the TV, LV, and the three reconstructions of the MDV of Cebupithecia all fall within the nonprehensile tail range (Fig. 5a–c). Relative transverse process expansion also places Cebupithecia within the nonprehensile tail range for the LV and MDVs (Fig. 5d–e). In particular, Cebupithecia consistently resembles Pithecia in terms of relative expansion of transverse and hemal processes, and does not have muscle development indices anywhere near the ranges of prehensile-tailed species (Fig. 5). RESULTS General Tail Morphology Relative lengths of the three regions of the Cebupithecia tail are presented in Figure 4 based on two reconstructions of total vertebral number: primitive model of 25 caudal vertebrae (Fig. 4a) and a Pithecia model of 23 caudal vertebrae (Fig. 4b). In general, prehensile-tailed taxa tend to have a longer proximal region of the tail. As the reconstructed total number of caudal vertebrae for Cebupithecia increases (i.e., from 23 to 25), the relative length of the proximal region of the specimen decreases. In both reconstructions, the proximal region of the tail is relatively short, as in nonprehensile-tailed taxa. However, the reconstruction of the Cebupithecia tail based on a Pithecia model shows a proximal tail region that is relatively longer than nonprehensile-tailed Saimiri, and encroaching on the relative length seen in prehensile-tailed Cebus. Cross-Sectional Geometry of Caudal Vertebrae The natural logarithms (ln) of polar section moduli derived from pQCT were compared with those derived from EMM using the least-squares regression method. The regression coefﬁcient was compared to an isometric slope of 1.0 using a student’s t-test (Zar, 2010). Data from both methods are highly correlated (r ¼ 0.99) and the relationship is positively allometric (y ¼ 1.11 þ 0.05, p[isometry]<0.05), but with an intercept not different from zero (P[intercept]>0.05) (Fig. 6). Because of the high correlation between measures, data were pooled for the complete analysis, even though the relationship CEBUPITHECIA TAIL MORPHOLOGY 2019 Fig. 5. Notched box plots of caudal muscle attachment indices for the transition vertebra (TV), longest caudal vertebra (LV), and mid-distal vertebra (MDV). The middle horizontal waist of the box represents the median, the outer horizontal limits of the boxes are the hinges encompassing the interquartile range, and the whiskers represent all other data points except outliers more than 1.5 times the interquartile range. Notches extending out from the median represent 95% conﬁdence intervals around the median (McGill et al., 1978). Hemal process expansion index (A–C) and transverse process expansion index (D, E). Red boxes, prehensile-tailed taxa; blue boxes, nonprehensiletailed taxa. Cebupithecia is indicated by a single solid line for TV and LV indices, and by three solid lines for MDV indices: Pithecia reconstruction model of 23 caudal vertebrae (top line); primitive reconstruction model of 25 caudal vertebrae (middle line); prehensile-tailed monkey reconstruction model of 27 caudal vertebrae (bottom line). Cp, Cebupithecia; At, Ateles; Lg, Lagothrix; Al, Alouatta; Cb, Cebus; Sm, Saimiri; Pp, Pithecia pithecia; Pm, Pithecia monachus; Ch, Chiropotes; Ao, Aotus; Sg, Saguinus. between variables is positively allometric. The intercept of the regression equation is not different from zero, meaning no correction factors are needed to correct overestimated EMM-derived section moduli to their pQCTderived equivalence. However, the positively allometric relationship of the variables does imply an overestimation of polar section modulus in EMM-derived data compared to pQCT-derived data at the higher end of the body size range for the sample, and an underestimation of polar section modulus at the lower end of the size range. This is not problematic for the present analysis because the EMM-derived data included are for animals in the middle and lower end of the body size range. Therefore, it is possible that the EMM-derived data obtained for Cebupithecia (2.2 kg, Fleagle, 1999) represent a ‘‘minimum’’ polar section modulus, and results should be interpreted in this context. Figure 7 illustrates size-adjusted polar section moduli for three caudal vertebrae (TV, LV, and MDV) of the extant platyrrhine sample and Cebupithecia. All three caudal vertebrae are relatively stronger in torsion and bending in prehensile-tailed platyrrhines compared to most other taxa (Fig. 7a–c). The tails of Cebupithecia and Pithecia monachus have stronger TV and LV compared with all other nonprehensile-tailed taxa, with cross-sectional property values of these two caudal vertebrae falling within the range of Cebus (Fig. 7a–b). Depending on which reconstruction of total vertebral number is used for Cebupithecia, the strength of the MDV varies, however. If 23 caudal vertebrae are used to reconstruct the tail of Cebupithecia, the MDV appears as strong as that of Cebus (Fig. 7c, top line). With 25 caudal vertebrae, MDV strength of Cebupithecia falls between prehensile- and nonprehensile-tailed taxa (Fig. 7c, middle line); at 27 caudal vertebrae, MDV strength falls within the range of all nonprehensile-tailed platyrrhines, including both species of Pithecia (Fig. 7c, bottom line). It should be pointed out that 2020 ORGAN AND LEMELIN unlike the TV and LV, cross-sectional properties of the MDV of P. monachus do not differ from those of Pithecia pithecia (Fig. 7c). DISCUSSION Fig. 6. Bivariate plot comparing natural logarithms of polar section modulus measurements of 17 individuals (open circles) from the sample obtained through peripheral quantitative computed tomography (pQCT) and the Ellipse Model Method (EMM). Data are highly correlated (r ¼ 0.99) and the relationship between variables is positively allometric (y ¼ 1.11 þ 0.05, P[isometry]<0.05), but with an intercept not different from zero (P[intercept]>0.05). See text for more detail. Reconstructions of the positional behavior of Cebupithecia sarmientoi (UMCP 38762) include a signiﬁcant amount of clinging postures and leaping behavior (Davis, 1987; Fleagle and Meldrum, 1988; Meldrum and Fleagle, 1988; Ford, 1990; Meldrum and Lemelin, 1991; Meldrum, 1993). The degree of hindlimb suspension employed by Cebupithecia is equivocal, however, as some researchers pointed out the absence of anatomical correlates of such behavior in the foot (e.g., Fleagle and Meldrum, 1988; Meldrum and Lemelin, 1991; Meldrum, 1993), while others have stressed morphological aspects of the hindlimb consistent with suspensory and climbing behaviors (e.g., Davis, 1987; Ford, 1990). Yet, nearly all studies have agreed that Cebupithecia relied heavily on vertical and nonhorizontal supports. Previous analyses of the external morphology of the caudal vertebrae of Cebupithecia indicated a distinct resemblance to Cebus in the proximal region of the tail, and marked similarities to Pithecia in more distal region of the tail (Meldrum and Lemelin, 1991). The previous work by Meldrum and Lemelin (1991) also noted the stronger resemblance of Cebupithecia in many aspects of the axial skeleton (including caudal vertebral morphology) to Pithecia monachus over Pithecia pithecia. The data reported here conﬁrm that the caudal vertebrae of Cebupithecia resemble those found in the nonprehensile tail of Pithecia to the greatest degree, with closer resemblance to P. pithecia rather than P. monachus in terms of caudal muscle development. Novel analysis of the internal morphology of the Cebupithecia caudal vertebrae presented here also suggests distinct similarities to Cebus in the proximal (and transitional) tail region. As predicted, the thicker cortices of the caudal vertebrae of Cebupithecia translate to increased resistance to bending and torsion at the locations Fig. 7. Notched box plots of size-adjusted torsional strength of the transition vertebra (TV), longest caudal vertebra (LV), and mid-distal vertebra (MDV). The middle horizontal waist of the box represents the median, the outer horizontal limits of the boxes are the hinges encompassing the interquartile range, and the whiskers represent all other data points except outliers more than 1.5 times the interquartile range. Notches extending out from the median represent 95% conﬁdence intervals around the median (McGill et al., 1978). Red boxes, prehen- sile-tailed taxa; blue boxes, nonprehensile-tailed taxa. Cebupithecia is indicated by a single solid line for TV and LV indices, and by three solid lines for MDV indices: Pithecia reconstruction model of 23 caudal vertebrae (top line), primitive reconstruction model of 25 caudal vertebrae (middle line), prehensile-tailed monkey reconstruction model of 27 caudal vertebrae (bottom line). Cp, Cebupithecia; At, Ateles; Lg, Lagothrix; Al, Alouatta; Cb, Cebus; Sm, Saimiri; Pp, Pithecia pithecia; Pm, Pithecia monachus; Ch, Chiropotes; Ao, Aotus; Sg, Saguinus. CEBUPITHECIA TAIL MORPHOLOGY of the transition vertebra (proximal tail region) and longest caudal vertebra (transitional tail region). The transition and longest caudal vertebrae of P. monachus are stronger in torsion and bending than those of P. pithecia (Organ, 2010, and conﬁrmed here). As a matter of fact, the section moduli of the caudal vertebrae of P. monachus and Cebupithecia fall within the range of those of Cebus (Fig. 7a–b). Depending on the tail reconstruction model used, the caudal vertebra of Cebupithecia at mid-point of the distal tail region is either: (1) as strong as that of the prehensile-tailed Cebus (23 caudal vertebrae model); (2) intermediate in strength between that of prehensile and nonprehensile-tailed platyrrhines (25 caudal vertebrae model); (3) similar in strength to that of nonprehensiletailed taxa (27 caudal vertebrae model) (Fig. 7c). Thus, the internal architecture of the TV, LV, and MDV caudal vertebrae of Cebupithecia suggests functional adaptation to elevated stress levels that may be related to positional behaviors. Analyses of the postcranial elements of Cebupithecia aligned this fossil platyrrhine most closely with Pithecia (Davis, 1987; Ford, 1990; Fleagle and Meldrum, 1988; Meldrum and Lemelin, 1991; Meldrum, 1993), emphasizing the marked adaptation for powerful leaping typical of pitheciines (Pithecia and Chiropotes). However, as Ford (1990: 170) pointed out: ‘‘the particular mix of anatomical features seen in Cebupithecia is unlike that found in any living platyrrhine.’’ The data presented in this study show that the increased cortical thickness of the caudal vertebral bodies of Cebupithecia translates into higher section moduli (i.e., higher torsional and bending strength) compared to extant pitheciines. The relatively thickened cortices of the caudal vertebrae (which do not characterize the limb bones) suggest that the tail of this fossil pitheciine may have been used differently than that of Pithecia. Instead, Cebupithecia may have used its tail as a brace on nonhorizontal substrates during hindlimb suspensory behavior, in a manner similar to that observed in Chiropotes. This hypothesis can only be tested further directly with recovery of additional caudal vertebral material. Platyrrhine Tail Evolution A growing body of evidence suggests that the prehensile tail evolved twice in parallel among platyrrhine primates: once among the atelines (Ateles, Lagothrix, Brachyteles, and Alouatta) and once within the genus Cebus (Rosenberger, 1983; Rosenberger and Strier, 1989; Lemelin, 1995; Organ, 2007, 2010; Organ et al., 2009, 2011). Differences in tail-use behavior and morphology are the primary lines of evidence of this parallel evolution. During feeding, atelines regularly use tail suspension (with or without assistance from the hindlimbs), while Cebus primarily uses its tail to form a tripod with its hindlimbs (Bergeson, 1996; Garber and Rehg, 1999; Garber, 2011). During suspensory locomotion, Lagothrix and Ateles show extensive use of the prehensile tail, which is best characterized as a ‘‘ﬁfth limb’’ (Jenkins et al., 1978; Turnquist et al., 1999; Schmitt et al., 2005). These behavioral differences likely account for a number of morphological differences between ateline and Cebus prehensile tails (Organ, 2010). Atelines have relatively longer tails than Cebus (Rosenberger, 1983; Organ, 2007; Garber, 2011), which account for a higher percent- 2021 age of total body mass (Grand, 1977). At its distal end, the tail of atelines has much greater ﬂexor muscle mass relative to the extensors compared with Cebus and especially nonprehensile-tailed platyrrhines (Grand, 1977; Lemelin, 1995). And ﬁnally, the ateline prehensile tail possesses a glabrous friction pad on the ventrodistal surface of the tail with abundant mechanoreceptors sensitive to touch (Hill, 1962; Organ et al., 2011). The Cebus prehensile tail is completely covered in hair, and does not experience the same degree of tactile sensitivity (Organ et al., 2011). Hypothesized proximate causes to the evolution of platyrrhine prehensile tails have included accommodation of larger body sizes in the forest canopy (e.g., Napier and Napier, 1967; Grand, 1972, 1977), assistance in feeding behaviors (e.g., Mittermeier and Fleagle, 1976; Grand, 1972, 1984; Cant, 1986; Rosenberger, 1992; Bergeson, 1996, 1998), or response to mechanically compliant substrates in the forest canopy (Emmons and Gentry, 1983; Organ, 2010). Many of these proximate causes are also compatible with the tail-bracing behavior seen among pitheciines. There is compelling evidence that Chiropotes engages in tail-bracing during hindlimb suspension (Fleagle and Meldrum, 1988; van Roosmalen et al., 1988; Walker, 1993, 1996; Meldrum, 1998). If such behavior is also present in Pithecia monachus (see Organ, 2010), then it may explain some of the stronger caudal vertebrae (transition and longest) of this species compared with Pithecia pithecia. By the same token, similarities in cross-sectional properties of the caudal vertebrae between P. monachus and Cebupithecia suggest that tail-bracing behavior may be primitive for the pitheciine clade, with an ensuing loss in P. pithecia. The distribution of tail-bracing behaviors within the pitheciine clade is of interest in light of the hypothesis that this behavior may have been transitional from nonprehensile to ateline prehensile tails (Meldrum, 1998). In particular, Meldrum (1998: 151) argues ‘‘that hindlimb suspension, assisted by tail-bracing and/or twining, served as a transitional positional behavior in the evolution of the highly derived platyrrhine [ateline] prehensile tail.’’ Hindlimb suspensory behaviors are employed by a large number of primate and nonprimate mammals, and some of the skeletal correlates of these behaviors are well-established. In particular, mammals that suspend from their hindlimbs often have tarsal morphology indicative of enhanced plantarﬂexion and supination of the hindfoot, which together enable hindfoot reversal and/or hindlimb suspension (e.g., Jenkins and McClearn, 1984; Fleagle and Meldrum, 1988; Meldrum et al., 1997). It is important to point out that such specializations of the hindfoot are not present in Cebupithecia (Fleagle and Meldrum, 1988; Meldrum, 1993, 1998; but see Ford, 1990). Therefore, the tail-bracing behavior of Cebupithecia was probably different from that observed in Chiropotes in the sense that it was rarely or never used in the context of hindlimb suspension. The present analysis does not bring evidence to support or disprove the hypothesis that hindlimb suspension with assistance of the tail was a transitional behavior for the evolution of ateline prehensile tails. However, the presence of possible tail-bracing behavior in an early pitheciine is an important step toward understanding the evolution of tail-use behaviors in extant pitheciines—including Chiropotes and possibly Pithecia monachus. 2022 ORGAN AND LEMELIN ACKNOWLEDGEMENTS The authors thank A. Rosenberger for inviting them to contribute to this special issue on evolutionary morphology of the New World monkeys, and to L. Halenar for comments that signiﬁcantly improved this manuscript. They are grateful to P. Holroyd at the University of California Museum of Paleontology for access to the Cebupithecia specimen. For access to extant comparative material, they thank: R. Thorington, J. Mead, L. Gordon, and J. Jacobs of the Smithsonian National Museum of Natural History; R. MacPhee and E. Westwig of the American Museum of Natural History; R. Martin, W. Stanley, and T. Damitz of the Field Museum of Natural History; and K. Glander of Duke University. Finally, they thank M. Teaford and C. Ruff for access to research equipment, and K. Rose, A. Taylor and R. German for early discussions of this work. Dedicated to M.E. Organ. LITERATURE CITED Ankel F. 1962. Vergleichende Untersuchungen über die Skelettmorphologie des Greifschwanzes südamerikanischer Affen. Z Morphol Oekol Tiere 52:131–170. Ankel F. 1965. Der Canalis sacralis als indikator für die Länge der Caudalregion der Primaten. Folia Primatol 3:263–276. Ankel F. 1972. Vertebral morphology of fossil and extant primates. In: Tuttle R, editor. The functional and evolutionary biology of primates. Chicago: Aldine. p 223–240. Bergeson DJ. 1996. The positional behavior and prehensile tail use of Alouatta palliata, Ateles geoffroyi, and Cebus capucinus. Ph.D. dissertation. Washington University, St. Louis. Bergeson DJ. 1998. Patterns of suspensory feeding in Alouatta palliata, Ateles geoffroyi, and Cebus capucinus. In: Strasser E, Fleagle J, Rosenberger A, McHenry H, editors. Primate locomotion: recent advances. New York: Plenum Press. p 45–60. Cant JGH. 1986. Locomotion and feeding postures of spider and howling monkeys: ﬁeld study and evolutionary interpretation. Folia Primatol 46:1–14. Davis LC. 1987. Morphological evidence of positional behavior in the hindlimb of Cebupithecia sarmientoi (Primates: Platyrrhini). M.A. thesis, Arizona State University. Emmons LH, Gentry AH. 1983. Tropical forest structure and the distribution of gliding and prehensile-tailed vertebrates. Am Nat 121:513–524. Fleagle JG. 1999. Primate adaptation and evolution. 2nd ed. San Diego: Academic Press. Fleagle JG, Meldrum DJ. 1988. Locomotor behavior and skeletal morphology of two sympatric pitheciine monkeys, Pithecia pithecia and Chiropotes satanas. Am J Primatol 16:227–249. Ford SM. 1990. Locomotor adaptations of fossil platyrrhines. J Hum Evol 19:141–173. Ford SM, Davis LC. Systematics and body size: implications for feeding adaptations in New World monkeys. Am J Phys Anthropol 88:415–468. Garber PA. 2011. Primate locomotor positional behavior and ecology. In: Campbell CJ, Fuentes A, MacKinnon KC, Bearder SK, Stumpf RM, editors. Primates in perspective. 2nd ed. Oxford: Oxford University Press, p 548–563. Garber PA, Rehg JA. 1999. The ecological role of the prehensile tail in White-faced Capuchins (Cebus capucinus). Am J Phys Anthropol 110:325–339. German RZ. 1982. The functional morphology of caudal vertebrae in New World monkeys. Am J Phys Anthropol 58:453–459. Grand TI. 1972. A mechanical interpretation of terminal branch feeding. J Mammal 53:198–201. Grand TI. 1977. Body weight: Its relation to tissue composition, segment distribution, and motor function. I. Interspeciﬁc comparisons. Am J Phys Anthropol 47:211–240. Grand TI. 1984. Motion economy within the canopy: Four strategies for mobility. In: Rodman PS, Cant JGH, editors. Adaptations for foraging in nonhuman primates. New York: Columbia University Press. p 54–72. Hill WCO. 1962. Primates: comparative anatomy and taxonomy V. Cebidae: Edinburgh University Press. Jenkins FA, Jr., McClearn D. 1984. Mechanisms of hindfoot reversal in climbing mammals. J Morphol 182:197–219. Jenkins FA, Jr., Dombrowski PJ, Gordon EP. 1978. Analysis of the shoulder in brachiating spider monkeys. Am J Phys Anthropol 48:65–76. Lazenby RA. 1997. Bias and agreement for radiogrammetric estimates of cortical bone geometry. Invest Radiol 32:12–18. Lazenby RA. 1998. Second metacarpal geometry: rehabilitating a circular argument. Am J Hum Biol 10:747–756. Lemelin P. 1995. Comparative and functional myology of the prehensile tail in New World monkeys. J Morphol 224:351–368. Meldrum DJ. 1993. Postcranial adaptations and positional behavior in fossil platyrrhines. In: Gebo D, editor. Postcranial adaptation in nonhuman primates. DeKalb: Northern Illinois University Press. p 235–251. Meldrum DJ. 1998. Tail-assisted hindlimb suspension as a transitional behavior in the evolution of the platyrrhine prehensile tail. In: Strasser E, Fleagle J, Rosenberger A, McHenry H, editors. Primate locomotion: recent advances. New York: Plenum Press. p 145–156. Meldrum DJ, Fleagle JG. 1988. Morphological afﬁnities of the postcranial skeleton of Cebupithecia sarmientoi. Am J Phys Anthropol 75:249–250. Meldrum DJ, Lemelin P. 1991. Axial skeleton of Cebupithecia sarmientoi (Pitheciinae, Platyrrhini) from the Middle Miocene of La Venta, Colombia. Am J Primatol 25:69–89. Meldrum DJ, Fleagle JG, Kay RF. 1990. Partial humeri of two Miocene Colombian primates. Am J Phys Anthropol 81:413– 422. Meldrum DJ, Dagosto MD, White J. 1997. Hindlimb suspension and hindfoot reversal in Varecia variegate and other arboreal mammals. Am J Phys Anthropol 103:85–102. McGill R, Tukey JW, Larsen WA. 1978. Variations of box plots. Am Stat 32:12–16. Mittermeier RA, Fleagle JG. 1976. The locomotor and postural repertoires of Ateles geoffroyi and Colobus guereza, and a reevaluation of the locomotor category semibrachiation. Am J Phys Anthropol 45:235–256. Napier J, Napier PH. 1967. A handbook of living primates. London: Academic Press. Napier PH. 1976. Catalogue of Primates in the British Museum (Natural History). Part 1: Families Callitrichidae and Cebidae. London: British Museum (Natural History). Ohman JC. 1993. Computer software for estimating cross-sectional geometric properties of long bones with concentric and eccentric elliptical models. J Hum Evol 25:217–225. O’Neill MC, Ruff CB. 2004. Estimating human long bone cross-sectional geometric properties: a comparison of noninvasive methods. J Hum Evol 47:221–235. Organ JM. 2007. The functional anatomy of prehensile and nonprehensile tails of the Platyrrhini (Primates) and Procyonidae (Carnivora). PhD dissertation. Baltimore, Maryland: Johns Hopkins University School of Medicine. Organ JM. 2010. Structure and function of platyrrhine caudal vertebrae. Anat Rec 293:730–745. Organ JM, Lemelin P. 2009. Caudal vertebral structure of Cebupithecia sarmientoi. Am J Phys Anthropol S48:312. Organ JM, Muchlinski MN, Deane AS. 2011. Mechanoreceptivity of prehensile tail skin varies between ateline and cebine primates. Anat Rec 294:2064–2072. Organ JM, Teaford MF, Taylor AB. 2009. Functional correlates of ﬁber architecture of the lateral caudal musculature in prehensile and nonprehensile tails of the Platyrrhini (Primates) and Procyonidae (Carnivora). Anat Rec 292:827–841. Rosenberger AL. 1983. Tale of tails: parallelism and prehensility. Am J Phys Anthropol 60:103–107. CEBUPITHECIA TAIL MORPHOLOGY Rosenberger AL. 1992. The evolution of feeding niches in New World monkeys. Am J Phys Anthropol 88:525–562. Rosenberger AL, Strier KB. 1989. Adaptive radiation of the ateline primates. J Hum Evol 18:717–750. Ruff CB. 2002. Long bone articular and diaphyseal structure in Old World monkeys and apes. I. Locomotor effects. Am J Phys Anthropol 119:305–342. Runestad JA, Ruff CB, Nieh JC, Thorington RW, Teaford MF. 1993. Radiographic estimation of long bone cross-sectional geometric properties. Am J Phys Anthropol 90:207–213. Schmitt D, Rose MD, Turnquist JE, Lemelin P. 2005. Role of the prehensile tail during ateline locomotion: experimental and osteological evidence. Am J Phys Anthropol 126:435–446. Schultz AH, Straus WL, Jr. 1945. The numbers of vertebrae in primates. Proc Am Phil Soc 89:601–626. Shapiro L. 1993. Functional morphology of the vertebral column in primates. In: Gebo DL, editor. Postcranial adaptation in nonhuman primates. DeKalb: Northern Illinois University Press. p 121–149. Shapiro L. 1995. Functional morphology of indrid lumbar vertebrae. Am J Phys Anthropol 98:323–342. Stirton RA. 1951. Ceboid monkeys from the Miocene of Colombia. Bull Univ Calif Pub Geol Sci 28:315–356. Stirton RA, Savage DE. 1951. A new monkey from the La Venta Miocene of Colombia. Compilacion de los studios geol oﬁciles en Colombia, Servicio Geologico Nacional (Bogota) 7:345–356. 2023 Turnquist JE, Schmitt D, Rose MD, Cant JGH. 1999. Pendular motion in the brachiation of captive Lagothrix and Ateles. Am J Primatol 48:263–281. van Roosmalen MGM, Mittermeier RA, Fleagle JG. 1988. Diet of the Northern Bearded Saki (Chiropotes satanas Chiropotes): a Neotropical seed predator. Am J Primatol 14:11–35. Wada N, Hori H, Tokuriki M. Electromyographic and kinematic studies of tail movements in dogs during treadmill locomotion. J Morphol 217:105–113. Walker SE. 1993. Positional adaptations and ecology of the Pitheciini. Ph.D. dissertation. City University of New York. Walker SE. 1996. The evolution of positional behavior in the saki-uakaris (Pithecia, Chiropotes, and Cacajao). In: Norconk MA, Rosenberger AL, Garber PA, editors. Adaptive radiations of neotropical primates. New York: Plenum Press. p 335–367. Ward C. 1991. Functional anatomy of the lower back and pelvis of the Miocene hominoid Proconsul nyanzae from Mfangano Island, Kenya. PhD dissertation. Baltimore, Maryland: Johns Hopkins University School of Medicine. Ward C. 1993. Torso morphology and locomotion in Proconsul nyanzae. Am J Phys Anthropol 92:291–328. Youlatos D. 2003. Osteological correlates of tail prehensility in carnivorans. J Zool (Lond) 259:423–430. Zar JH. 2010. Biostatistical analysis. 5th ed. Upper Saddle River: New Jersey: Prentice Hall.