close

Вход

Забыли?

вход по аккаунту

?

Biogeochemical and craniometric investigation of dietary ecology niche separation and taxonomy of Plio-Pleistocene cercopithecoids from the Makapansgat Limeworks.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 135:121–135 (2008)
Biogeochemical and Craniometric Investigation of
Dietary Ecology, Niche Separation, and Taxonomy of
Plio-Pleistocene Cercopithecoids From the
Makapansgat Limeworks
Nicolaas H. Fourie,1,2* Julia A. Lee-Thorp,1,3 and Rebecca Rogers Ackermann1
1
Department of Archaeology, University of Cape Town, Cape Town, South Africa
Department of Anthropology, Center for the Advanced Study of Human Paleobiology,
The George Washington University, Washington, DC
3
Department of Archaeological Sciences, University of Bradford, Bradford, UK
2
KEY WORDS
stable isotopes; trace elements; enamel; craniofacial morphology; Makapansgat
ABSTRACT
Three sympatric fossil cercopithecoid
genera (Cercopithecoides, Parapapio, and Theropithecus)
occur in Members 3 and 4 at the Makapansgat Limeworks hominin locality, South Africa, and their presence
in a single ecosystem suggest a certain degree of ecological and/or dietary differentiation between taxa. Here,
we explore the extent of dietary niche separation amongst
these taxa using stable isotope (13C/12C, 18O/16O) and
trace-element (Sr, Ba, Ca) analyses of fossil tooth
enamel. In particular we searched for evidence of subtle
niche separation between the more closely related, morphologically similar taxa of the genus Parapapio, as
uncertainties exist around their taxonomy and taxonomic identification. Given these uncertainties, craniometric analyses were also performed to ground the dietary interpretations in a morphological context. The
results found no clear taxonomic signal in the craniometric data for the Parapapio sample, and further indicate
that this sample was no more variable morphologically
than a single, geographically circumscribed, extant
chacma baboon sample. In contrast, two overlapping dietary ecologies were found within this same Makapansgat
Parapapio sample. Additionally, two widely differing dietary ecologies were found within the Cercopithecoides
williamsi sample, while results for Theropithecus darti
indicate a predominantly C4 diet. Hence, although biogeochemical dietary indicators point towards distinct dietary ecologies within and between fossil genera at Makapansgat, within the genus Parapapio disjunctions exist
between the dietary categories and the taxonomic
assignment of specimens. Am J Phys Anthropol 135:121–
135, 2008. V 2007 Wiley-Liss, Inc.
A rich array of primate taxa are represented in the
Plio-Pleistocene deposits of the South African karstic
cave sites. Thus far, all recovered fossil primate specimens from South African Plio-Pleistocene sites are
thought to belong to the family Cercopithicidae. Two
sub-families, Cercopithicinae and Colobinae, are represented by four genera and up to 12 species, and one genus and one species, respectively (Freedman, 1957;
Brain, 1981; Delson, 1992, 1993). All the South African
Plio-Pleistocene Cercopithicinae genera belong to the
tribe Papionini (Freedman, 1957; Brain, 1981; Delson,
1992, 1993). During this time, sympatry/parapatry was
common, with up to five papionin taxa and one colobine
taxon sympatric/parapatric at some sites (Freedman,
1970). Yet at the same time there is limited morphological evidence for dietary/niche specialization.
No modern analogue to this situation exists among
extant papionin taxa. Most extant members of this tribe
in Africa are medium body-sized terrestrial primates.
There are currently far fewer extant papionin taxa in
Africa than there were in the Plio-Pleistocene, and these
are widely distributed across the continent; few are truly
sympatric or even parapatric. Only Papio h. hamadryas
and Theropithecus gelada co-occupy the same region,
facilitated by distinct niche separation and specific dietary specialization on the part of T. gelada (Dunbar and
Dunbar, 1974). Some Papio subspecies are parapatric
(Zinner et al., 2001), and in areas of parapatry, species
borders are either maintained or hybrid zones develop
(Nagel, 1970, 1973; Kummer, 1971; Phillips-Conroy and
Jolly, 1981; Alberts and Altmann, 2001).
This unusual ecological situation—abundant sympatry/parapatry combined with limited morphological evidence for niche separation—is exemplified at the Pliocene hominin locality of Makapansgat Limeworks in the
Limpopo Province of South Africa. Here, five different
fossil cercopithecoid taxa are thought to have been sympatric/parapatric. One is a colobine, while four taxa were
papionins, of which three (genus: Parapapio) appear to
have been very closely related. How did so many primate
taxa manage to cohabit in the same environment, which
C 2007
V
WILEY-LISS, INC.
C
Grant sponsors: National Research Foundation of South Africa,
Paleo-Anthropology Scientific Trust, University of Cape Town.
*Correspondence to: Nicolaas Hofmeyer Fourie, Department of
Anthropology, Center for the Advanced Study of Human Paleobiology, 2110 G Street, NW, Washington, DC 20052.
E-mail: nfourie@gwu.edu
Received 10 January 2007; accepted 20 July 2007
DOI 10.1002/ajpa.20713
Published online 16 October 2007 in Wiley InterScience
(www.interscience.wiley.com).
122
N.H. FOURIE ET AL.
they shared with a large variety and number of other
herbivorous and omnivorous fauna?
The above scenario is not absent among extant primates. For example, various species of arboreal guenons,
which are skeletally relatively indistinct from one
another, cohabit in Africa, and avoid competition by
occupying different vertical eco-niches and/or by following differing dietary ecologies (Gautier-Hion, 1980; Gautier-Hion and Gautier, 1986; Mitani, 1991; Schoeninger
et al., 1997; Ramakrishnan and Coss, 2001). However,
vertical niche differentiation is unlikely for the South
African Plio-Pleistocene fossil cercopithecoid taxa as they
are thought to have been terrestrial, or at least partially
terrestrial. For example, C. williamsi and at least some
of the three Parapapio taxa (Elton, 2001), as well as
T. darti, were likely terrestrial. Furthermore, all three
Parapapio taxa included substantial amounts of C4 resources in their diets implying that they spent much time
foraging terrestrially (Luyt, 2001; Codron et al., 2005).
Analysis of the postcranial morphology of C. williamsi, as
well as limited stable isotope evidence, suggest that this
fossil colobine may have been substantially adapted to a
terrestrial ecology (Birchette, 1981; Leaky, 1982; Benefit
and McCrossin, 1990; Delson, 1992; Luyt, 2001; Codron,
2003; Codron et al., 2005). Therefore, in order for all of
these taxa to be supported within the same ecosystem, it is
reasonable to expect a certain degree of dietary differentiation between them. If this was the case, these dietary differences should be detectable through chemical dietary
indicators, distinctive dental wear and/or subtle morphological differences between these taxa.
The material from Makapansgat presents a good opportunity to test for dietary differentiation between multiple sympatric/parapatric primates in the Plio-Pleistocene fossil record. Here, particular emphasis is placed on
the genus Parapapio, as its abundance (i.e., number of
specimens) and diversity (i.e., number of species) at
Makapansgat allow for assessment of intertaxa and
intrataxa dietary variation, which might provide evidence for subtle dietary niche separation among closelyrelated taxa. In South Africa this genus is currently represented by three morphologically similar species (Pp.
jonesi, Pp. broomi, and Pp. whitei) that differ primarily
in size. Pp. jonesi, Pp. broomi, and Pp. whitei, were originally distinguished on the basis of molar size, Pp. whitei
having the largest and Pp. jonesi having the smallest
molars (Broom, 1940). Subsequently, further refinements
have been made in the identification of these three species, but differences in molar size remain a fundamental
and discriminating difference. In other aspects of morphology, these species appear to be relatively homogenous. These three species have all been described from
the same deposits sat Makapansgat Limeworks
(Freedman, 1957, 1976; Brain, 1981) while the same species have also been found to be synchronous and sympatric or parapatric at a number of other fossil hominin
localities (Freedman, 1957, 1970, 1976; Brain, 1981; Benefit and McCrossin, 1990; Elton, 2001). They show no
obvious morphological adaptations that would suggest
different dietary ecologies. Ecological and dietary differences between these closely related taxa, if they exist,
may be identifiable through fine-grained analysis of
chemical dietary indicators.
Here we investigate the dietary ecology of three sympatric, contemporaneous fossil cercopithecoid genera
(Cercopithecoides, Parapapio, and Theropithecus) from
the Makapansgat Limeworks hominin locality in South
Africa, using stable isotope and trace-element techniques
for paleodietary analysis in combination with a quantitative morphometric analysis. Such an approach allows us
to ground our understanding of niche separation in the
context of morphological/taxonomic distance. A multilayered approach to paleodietary analysis is taken; a suite
biogeochemical paleodietary indicators (13C/12C, 18O/16O,
Sr, and Ba) are used to investigate subtle dietary differences between these taxa because each indicator is sensitive to dietary or behavioral components to which others
are insensitive. In this way it is hoped that subtle as
well as broader dietary differences between taxa will be
detected.
Biogeochemical dietary indicators
An animal’s tissues are made up of the food it eats.
Any biogeochemical patterns in an animal’s diet should
be reflected in its tissues; these biogeochemical dietary
signals can be preserved in tooth enamel for millions of
years (e.g., Luz and Kolodny, 1985; Lee-Thorp and Sponheimer, 2003). Stable isotopes and trace-element analyses of tooth enamel have proven to be highly effective
ways of investigating diet in both modern and fossil animal communities. Through comparison between taxa
within a single community relatively fine grained dietary
information can be retrieved using these methods.
Stable light isotopes as dietary tools are based on the
differential patterned discrimination of certain isotopes
in natural systems. Because of differences in the physicochemical properties of the isotopes of carbon and oxygen the heavier isotopes of carbon and oxygen are
treated differently during chemical reactions in the
global carbon and oxygen cycles. C4 photosynthesizing
plants, such as tropical grasses and certain sedges discriminate less strongly against 13C than C3 photosynthesizing plants like trees, shrubs, and herbs. C3 and C4
plants thus yield two distinct nonoverlapping ranges of
d13C values (Smith and Epstein, 1971). It is therefore
possible to distinguish between different dietary ecologies in ecosystems where this C3/C4 dichotomy exists.
Although a dominant influence on body water 18O/16O is
meteoric and environmental water isotope composition
within one locality, 18O/16O in body water, and hence in
tissues, can be influenced by the most important sources
of water for individual species. For instance, leaf water
can be significantly more enriched in 18O than meteoric
surface water. On the basis of this difference, obligate
drinkers can be discriminated from nonobligate drinkers
(Kohn, 1996; Sponheimer and Lee-Thorp, 1999; Carter,
2001). These isotopic ratios are archived in an animal’s
tissues, and are particularly well preserved in tooth
enamel (Luz and Kolodny, 1985; Lee-Thorp et al., 1989;
Cerling et al., 1997; Lee-Thorp and Sponheimer, 2003;
Lee-Thorp et al., 2003; Sponheimer et al., 2005a).
Certain trace element ratios have proven to be useful
dietary discriminators in natural ecosystems that can
discriminate between dietary resources to which stable
isotopes may be insensitive (Runia, 1987; Sillen, 1992;
Sillen et al., 1995; Burton et al., 1999; Sponheimer et al.,
2005a; Sponheimer and Lee-Thorp, 2006). Strontium (Sr)
and Barium (Ba) resemble Calcium (Ca) in their physical
and chemical properties and can replace Ca in metabolic
reactions. However, because Sr and Ba are larger atoms
than Ca they are discriminated against in the mammalian gut and during metabolic reactions. Hence their
American Journal of Physical Anthropology—DOI 10.1002/ajpa
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
presence is reduced at each trophic level (Elias et al.,
1982).
It has become clear that concentrations of strontium
and barium are not uniform in plants (Sillen, 1992). The
concentration Sr decreases during transport in the xylem
from the roots to the leaves. Leaves thus have lower Sr/
Ca ratios than roots and stems (Bowen and Dymond,
1955, 1956; Burton et al., 1999) and fruits may have
lower Sr/Ca ratios than leaves. Furthermore, grasses in
African ecosystems have been found to have higher concentrations of Sr than leaves of dicotyledonous plants
(Sponheimer and Lee-Thorp, 2006). Thus Sr/Ca ratios
can be used to discriminate between grass, leaf/fruit,
and underground plant organs (Runia, 1987; Sillen
et al., 1995; Burton et al., 1999; Sponheimer et al.,
2005a). Sponheimer et al. (2005a) have shown, by proxy,
that grasses have substantially higher Ba/Ca ratios than
roots and tubers, both of which yield high Sr/Ca ratios.
A combination of Sr/Ca and Ba/Ca ratios can thus be
useful in discriminating between the relative contribution of certain plants and plant parts to the diet that are
not distinct in their Sr/Ca or carbon isotope ratios. Sr/Ba
ratios have also been used as an expression of the relative amounts of Sr and Ba in an animal tissue (Burton
and Price, 1990, 1991; Gilbert et al., 1994; Sponheimer
and Lee-Thorp, 2006). By expressing the relative abundances of Sr and Ba as a Sr/Ba ratio, it has been found
that clearer distinctions can be made between certain dietary categories than by using Sr/Ca and Ba/Ca ratios
(Burton and Price 1990, 1991; Gilbert et al., 1994; Sponheimer and Lee-Thorp, 2006). It should be noted that for
certain dietary components there is still little data is
available (e.g., fruits). Importantly, Sr/Ca and Ba/Ca
intraspecific variability is known to be high within any
given taxon even in a limited area; hence there are limitations on the interpretations that can be made at the
level of individuals. Not withstanding these difficulties,
trace-elements can be useful as dietary indicators in
African ecosystems if carefully applied (Sillen, 1992;
Sillen et al., 1995; Sponheimer et al., 2005a; Sponheimer
and Lee-Thorp, 2006).
Taxonomic issues in dietary analyses
A number of studies have employed isotopic techniques to investigate the ecology of extant and Plio-Pleistocene primate communities or individual taxa (Lee-Thorp
et al., 1989; Schoeninger et al., 1997, 1998, 1999; Carter,
2001; Luyt, 2001; Sponheimer and Lee-Thorp, 2001;
Codron, 2003; Thackeray and Myer, 2004; Codron et al.,
2005). Some of these studies have pointed to puzzling bimodal distributions of dietary habits for at least two of
the taxa included in this study—C. williamsi and Parapapio sp. (Luyt, 2001; Codron et al., 2005). Extreme dietary variability is not unheard of amongst primate species, but it is unusual, and where it occurs is related to
broad geographical/environmental distinctions. Yet this
is not the case for the South African hominin-bearing
fossil sites that have been studied to date, as they represent environments comprised of reasonably similar habitats overall (Reed, 1997; Pickering et al., 2004). The
bimodality and variability of the dietary data reported
for some taxa (Luyt, 2001; Codron, 2003; Codron et al.,
2005) from these sites are therefore somewhat unusual
given the paleoenvironmental setting. These apparently
anomalous dietary patterns underscore taxonomic concerns raised by morphologists (Maier, 1970; Jones, 1978;
123
Brain, 1981; Delson, 1992; Thackeray and Myer, 2004).
This is a problem for isotopic analyses; when dietary
groupings do not correlate well with the taxonomic
assignment of specimens, it is unclear how to interpret
such disjunctions. Are the specimens identified incorrectly or are the taxonomic categories themselves problematic?
Some of these difficulties regarding the secure identification of specimens and possible flaws in the taxonomy
encountered in studies of fossil primate dietary ecology
could relate to the nature of the samples used in these
studies. Most biogeochemical dietary studies have used
isolated teeth or dentognathic fragments for analysis
(Codron, 2003; Codron et al., 2005). The identification of
such specimens, however, is difficult because there are
often no distinctive morphological features that can be
used to accurately distinguish between closely related
taxa. Difficulties regarding the correct taxonomic assignment of specimens and their implications for the interpretation of paleodietary analyses are likely to be prevalent in this sample given the uncertainties surrounding
the taxonomy of the South African Plio-Pleistocene Cercopithecoidea as a whole (e.g., Maier, 1970; Freedman,
1976; Delson, 1992).
If specimens are incorrectly identified, dietary patterns in the data are lost and dietary differences
between taxa are obscured by undue within-taxon variability. We attempted to circumvent this problem by
sampling only enamel from the teeth of complete or partially complete cranial specimens from which morphological craniometric measurements could also be taken. In
this way higher confidence can be placed in the taxonomic assignment of specimens. Furthermore, the craniometric data can themselves be analysed using multivariate statistical techniques, to further investigate
remaining concerns around the identification and classification of specimens. We argue that these precautions
will provide a more secure taxonomic context within
which to investigate dietary/ecological differences between these taxa.
This investigation therefore differs from previous
investigations of Plio-Pleistocene cercopithecoid dietary
ecology in a number of important ways. Firstly, a suite
of dietary indicators are used to investigate broad as
well as subtle dietary differences. Secondly, only complete or partially complete cranial specimens from a narrow sequence in a single site are sampled. Thirdly, specimens are subjected to morphometric analysis to manage
taxonomic concerns, and potentially to provide a more
secure taxonomic context for the interpretation of dietary data. It should be noted that the craniometric analysis conducted in this study does not attempt to definitively settle taxonomic concerns regarding the genus
Parapapio. It is done primarily to facilitate dietary analysis by correlating dietary groupings to independently
generated craniometric groupings. The results may, however, provide insights into broader taxonomic issues.
MATERIALS AND METHODS
Fossil and extant samples
All fossil primate specimens used in this study were
adults recovered from Makapansgat Limeworks Members 3 and 4. These deposits yielded a substantial number of Parapapio specimens assigned to three species
(Pp. whitei, Pp. broomi, and Pp. jonesi), as well as Theropithecus darti and Cercopithecoides williamsi specimens.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
124
N.H. FOURIE ET AL.
Tooth enamel was sampled from adult individuals
with erupted third molars to reflect adult (or near adult)
physiological and dietary conditions. For craniometric
comparative purposes, data were also collected from an
extant baboon sample consisting of 30 chacma baboon
crania housed in the Iziko Natural History Museum,
Cape Town. They represent a geographically restricted
sample of baboons from the Western Cape Province.
Stable isotope and trace-element analysis
Enamel powder was collected by abrading the buccal/
lingual surfaces of the third molar using diamond tipped
micro-drill, and wherever possible, broken enamel surfaces were sampled. Pretreatment procedures to remove
possible contaminants followed published methods
(Sponheimer, 1999). Briefly, (5 mg) enamel powder was
reacted with 1.5% NaClO solution in 2 ml centrifuge
tubes for 15 min, followed by removal of extraneous carbonates in a 0.1 M solution of CH3COOH for 10 min.
Samples were also given an acetone wash to remove possible traces of preservatives. Following each step, samples were alternatively centrifuged and rinsed in distilled water. Finally samples were dried in a freeze-drier.
CO2 was produced by reaction with 100% phosphoric
acid (H3PO4) at 708C, and cryogenically distilled in a
Kiel II autocarbonate device, and introduced into a Finnigan Matt 252 mass spectrometer (Finnigan, Bremen)
for measurement of the isotopic ratios of carbon and oxygen. As per convention, 13C/12C and 18O/16O ratios are
expressed in the delta convention in parts per thousand
(%) relative to the VPDB standard as follows:
dX ¼
Rsample Rstandard 1 3 1000
where X is 13C or 18O and R is the corresponding 13C/12C
or 18O/16O ratio. Values obtained were calibrated using
both IAEA and internal standards. Precision as determined by replicates of standards was 0.1% for 13C/12C
and 0.2% for 18O/16O.
The remaining sample, usually between 1 and 2 mg of
pretreated material , were weighed into teflon beakers
and dissolved in 1 ml of HF:HNO3 (3:1). The beakers
were closed and left on a hotplate for 48 h. The contents
of the beakers were then evaporated until incipient dryness. The sample residues were then twice dissolved in 2
ml of HNO3 and evaporated to dryness. The dried sample was then dissolved in a 5% HNO3 solution containing
10 ppb Re, Rh, In, and Bi as internal standards. The
samples were then analysed on the on a Perkin-Elmer
ELAN 6000 inductively coupled plasma mass spectrometer (ICP-MS) for Mn, Zn, Sr, Ba, and Ca content. Traceelement amounts are expressed in parts per million
(ppm). Results were standardized against artificial multielement standards. Replicate analysis of the international rock standard BHVO-1 gave a procedural error of
better than 3%.
Craniometric analysis
Three-dimensional craniometric landmark coordinate
data were collected from both the fossil material and the
comparative extant sample using a Microscribe 3-DX
digitizer and Inscibe-32 software (Immersion, San Jose,
CA). Twenty-four cranial landmarks (after Singleton,
2002) were digitized. These together with the inter-landmark distances for the fossil sample are listed in Table 1.
Linear dental measurements were also taken using a
pair of digital callipers, both to supplement the craniometric analysis, but also because the previous taxonomy
of many fossil specimens was based on differences in
dental dimensions. They consisted of maximum mesiodistal (anterior and posterior) and buccal-lingual measurements of the third molars of all specimens, extant
and fossil, that retained these teeth. Fossil crania were
often incomplete and were missing a number of landmarks on at least one side; many individuals shared few
variables, while few individuals shared many. Therefore,
to maximize both the number of specimens and the
number of variables in any given analysis, five subsets
consisting of a varying number and combination of landmarks and specimens were selected and all further morphometric analyses were done on these subsets. Interlandmark distances and Procrustes rotated landmark
data were subjected to a Principal Components Analysis
(PCA). Landmark configurations were Procrustes-transformed using PAST ver.1.34 morphometrics software
package (Hammer et al., 2005). Variable loadings from
the PCA’s on Procrustes rotated landmark data were
also judged against the geometric mean to determine
what aspects of shape variation are allometric.
PCA analyses were performed on fossil subsamples
alone, to investigate differences in the size and shape
between fossil specimens. Analyses were also performed
combining the extant chacma baboon sample and the fossil subsample in the same analysis. This allowed for the
fossil subsample and the extant sample to be compared
on aspects of size, shape, and variability in those components. Dental measurements of both the extant and fossil samples were also subjected to PCA and the results
were compared.
RESULTS
Stable isotope and trace-element analysis
The box and whiskers plot of d13C and d18O patterns
for the fossil cercopithecoid sample (see Fig. 1) incorporates the broader community isotope ecology of Makapansgat, using previously published data for browsing
bovids, and grazing bovids (Sponheimer, 1999; Sponheimer and Lee-Thorp, 1999; Sponheimer et al., 1999,
2001) as well as extant chacma baboons from the Waterberg. d13C values for modern samples have been
adjusted for the fossil fuel effect. All Parapapio and one
Cercopithecoides williamsi specimen’s d13C values fall in
between those for browsing and grazing fauna at Makapansgat Limeworks, indicating intermediate diets which
include both C3 and C4 sources. In general these values
are more enriched in 13C than the enamel carbonate of
the extant chacma baboons (P. h. ursinus) from a woodland savannah in the Waterberg, South Africa, about
100 km to the west of Makapansgat (Codron, 2003). In
that case baboons typically yielded narrow range of d13C
enamel carbonate values between 213.5 and 210% (n 5
10) suggesting relatively minor contributions of C4
plants (or possibly succulents) to their diets (Codron,
2003).
At Makapansgat Limeworks fossil browsers are more
enriched in 18O than fossil grazers (Sponheimer, 1999)
as seen in Figure 1, a pattern observed in several modern faunal assemblages in South and East Africa
(Cerling et al., 1997; Sponheimer and Lee-Thorp, 2001;
Sponheimer et al., 2005a). All the fossil cercopithecoid
specimens analysed in his study yielded enamel carbon-
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Prosthion
Nasospinale
Premax/max (R)
Rhinion
Superior Premaxilary Suture (R)
Zygomaxillary inferior (R)
Nasion
Zygomaxillary superior (R)
Fronto malare orbitale (R)
Orbital notch (R)
Fronto malare Temporale (R)
Glabella
Temporal/frontal/Parietal suture (R)
Bregma
Premax/max (L)
Premaxillary suture superior (L)
Zygomaxillary Inferior (L)
Zygomaxillary Superior (L)
Fronto malare orbitale (L)
Orbital notch (L)
Fronto malare temporale (L)
Temporal/frontal/ Parietal suture (L)
Porion (R)
Porion (L)
Landmark
PR
NS
PM
RH
FMN
ZI
NA
ZS
FMO
ON
FMT
GL
PT
BR
PM
FMN
ZI
ZS
FMO
ON
FMT
PT
P
P
Abbreviation
BR-PN(L)1
FMO(R)-GL2
FMO(R)-ON(L)2,3
FMO(R)-ON(R)2
GL-BR1
GL-ON(L)1,2,3
GL-TP(R)1
NA-ZS(R)2
NA-BR1
NA-FMO(R)2,3
NA-GL1,2,3
NA-ON(L)1,3
NA-ON(R)2,3
NA-PM(L)4,5
NA-TP(R)1
NA-ZS(L)2,4
NA-ZS(R)3
NS-NA4
NS-PM(L)4
NS-PM(R)4
NS-RH4
NS-S(L)4
ON(R)-GL2,3
ON(R)-ON(L)2,3
PM(L)-ZS(L)4
PM(R)-NA4,5
PM(R)-PM(L)4,5
PM(R)-RH4
PM(R)-ZS(L)4
PR-NA4,5
PR-NS4
PR-PM(L)4
PR-PM(R)4,5
PR-RH4
PR-ZS(L)4
RH-NA4
RH-PM(L)4
RH-ZS(L)4
TP(R)-BR1
TP(R)-ON(L)1
ZI(R)-FMO(R)
ZI(R)-GL2,3
ZI(R)-NA2
ZI(R)-ON(L)2,3
ZI(R)-ON(R)2,3
ZI(R)-ZS(R)2
ZI(R)-FMO(R)2,3
ZI(R)-GL2
ZI(R)-ON(L)2
ZI(R)-ON(R)2
Inter-Landmark
Distance
16.29
17.22
65.05
72.51
88.87
36.92
42.92
82.10
94.14
11.34
21.36
21.15
44.44
83.35
50.35
43.46
45.95
45.85
63.16
28.20
45.37
57.02
66.81
12.19
16.54
14.09
48.13
55.65
18.86
45.00
24.57
62.65
27.78
83.04
20.52
18.81
33.83
72.66
55.98
17.42
13.44
37.73
45.34
89.30
30.01
67.77
29.03
62.97
47.89
59.12
26.91
33.47
55.51
61.34
11.26
12.90
16.18
33.40
54.38
28.10
31.93
27.62
51.04
15.64
14.12
23.66
44.67
23.19
58.65
15.94
33.93
40.04
24.13
17.47
35.57
45.50
24.33
34.87
10.13
13.97
13.22
67.83
35.32
8.12
17.92
14.62
72.30
28.68
28.88
61.76
15.99
15.72
33.61
52.80
13.23
24.67
52.56
68.71
29.59
41.52
63.49
73.69
12.47
17.34
18.06
45.53
64.17
28.20
41.19
28.68
13.26
17.23
26.39
30.54
56.92
23.02
18.99
28.89
48.47
14.64
30.36
57.21
67.37
36.51
40.73
63.68
69.93
14.54
19.36
20.00
41.08
61.48
29.06
45.75
29.73
36.62
46.76
24.23
36.90
51.05
22.47
14.57
18.60
61.67
59.66
30.92
61.94
45.43
54.27
46.72
10.70
16.14
62.96
54.11
15.48
43.50
58.55
44.38
55.31
29.84
29.67
52.40
59.49
9.35
16.79
17.55
30.90
52.16
28.69
32.25
28.22
51.52
16.42
14.47
23.49
44.03
25.45
58.46
43.50
53.36
37.09
61.92
57.48
66.79
50.73
29.53
18.94
33.53
38.85
22.96
15.51
26.92
28.41
59.09
33.81
44.56
19.31
52.73
13.71
42.12
28.41
60.05
32.75
8.46
12.62
14.96
Parapapio
Parapapio
Parapapio Parapapio Parapapio Pp. broomi#
C. williamsi Parapapio Parapapio Parapapio
# MP34,5 sp. MP1675 sp. MP474,5 sp. M30784,5 sp. $ M30843,4,5 sp. $ MP2393,4,5 sp. M30795 sp. M31331 sp. MP2084 M30651,2,3
TABLE 1. Table showing anatomical landmarks, the abbreviations for each landmark and the inter-landmark distances for the fossil sample
PR
NS
PM
RH
FMN
ZI
NA
ZS
FMO
ON
FMT
GL
PT
BR
PM
FMN
ZI
ZS
FMO
ON
FMT
PT
P
P
Abreviation
67.34
36.50
51.77
22.48
61.83
17.50
49.31
28.37
70.87
35.37
11.96
18.12
16.60
53.23
28.37
15.34
31.47
47.72
62.61
38.33
66.00
59.28
75.26
55.09
31.91
21.74
36.32
45.62
26.19
51.68
39.46
52.46
23.42
47.81
15.19
52.22
28.58
54.89
38.04
8.10
17.11
16.30
54.12
28.58
17.02
31.35
44.69
61.89
39.95
62.80
57.42
69.21
52.21
29.38
23.13
34.00
42.40
23.82
BR-PN(L)1
FMO()R-GL2
FMO(R)-ON(L)2,3
FMO(R)-ON(R)2
GL-BR1
GL-ON(L)1,2,3
GL-TP(R)1
NA-ZS(R)2
NA-BR1
NA-FMO(R)2,3
NA-GL1,2,3
NA-ON(L)1
NA-ON(R)2,3
NA-PM(L)4,5
NA-TP(R)1
NA-ZS(L)4,5
NA-ZS(R)3
NS-NA4
NS-PM(L)4
NS-PM(R)4
NS-RH4
NS-S(L)4
ON(R)-GL2,3
ON(R)-ON(L)2,3
PM(L)-ZS(L)4
PM(R)-NA4,5
PM(R)-PM(L)4,5
PM(R)-RH4
PM(R)-ZS(L)4
PR-NA4,5
PR-NS4
PR-PM(L)4
PR-PM(R)4,5
PR-RH4
PR-ZS(L)4
RH-NA4
RH-PM(L)4
RH-ZS(L)4
TP(R)-BR1
TP(R)-ON(L)1
ZI(R)-FMO(R)
ZI(R)-GL2,3
ZI(R)-NA2
ZI(R)-ON(L)2,3
ZI(R)-ON(R)2,3
ZI(R)-ZS(R)2
ZI(R)-FMO(R)2,3
ZI(R)-GL2
ZI(R)-ON(L)2
ZI(R)-ON(R)2
Pp. broomi$
MP1701,2,3
45.69
55.49
44.86
6.20
16.58
60.99
55.64
18.37
41.75
61.13
Pp. jonsei#
MP751
37.09
61.92
57.48
66.79
50.73
29.53
18.94
33.53
38.85
22.96
27.99
27.35
55.32
17.00
18.88
22.40
48.42
15.51
26.92
44.97
62.03
32.30
32.87
59.01
64.27
9.01
17.78
19.66
31.26
56.23
33.27
30.45
32.68
32.75
8.46
12.62
14.96
60.06
17.84
17.62
68.96
66.68
31.13
38.74
64.36
62.32
70.52
55.24
29.47
23.22
35.82
42.44
28.15
32.23
33.31
83.03
19.07
17.60
29.98
65.46
15.00
28.37
67.11
89.73
35.30
39.17
76.25
95.84
13.83
19.93
20.55
41.84
78.32
54.31
38.84
42.51
37.56
4.96
13.31
15.80
89.44
33.31
28.41
37.31
37.31
40.08
23.62
Pp. whitei#
MP2212,3,4,5
13.96
65.27
Pp. whitei$
MP1195
13.71
59.09
33.62
44.56
19.31
Pp. whitei$
M30702,3,4,5
(L) and (R) refer to the left or right measurement respectively.
The super scripted numbers next to the inter-landmark distances and specimens denote the subset that they were analysed in.
The symbols # and $ next to the taxon name denote the sex of the specimen.
Prosthion
Nasospinale
Premax/max (R)
Rhinion
Superior Premaxilary Suture (R)
Zygomaxillary inferior (R)
Nasion
Zygomaxillary superior (R)
Fronto malare orbitale (R)
Orbital notch (R)
Fronto malare Temporale (R)
Glabella
Temporal/frontal/Parietal suture (R)
Bregma
Premax/max (L)
Premaxillary suture superior (L)
Zygomaxillary Inferior (L)
Zygomaxillary Superior (L)
Fronto malare orbitale (L)
Orbital notch (L)
Fronto malare temporale (L)
Temporal/frontal/ Parietal suture (L)
Porion (R)
Porion (L)
Landmark
TABLE 1. (Continued).
Pp. broomi#
MP21,2,3
Inter-Landmark
Distance
38.39
65.49
59.06
77.01
55.29
29.57
18.53
36.08
48.98
26.47
30.87
30.06
70.71
20.80
20.71
31.48
67.88
15.30
32.99
72.78
82.71
36.17
45.20
84.25
85.78
16.69
21.17
20.97
45.54
83.25
40.42
45.54
45.23
35.74
10.47
21.42
16.97
82.70
30.06
17.96
36.83
36.83
52.81
22.88
Pp. whitei#
MP2232,3,4,5
51.03
55.93
46.50
7.13
16.31
64.87
59.24
15.14
45.60
61.87
Pp. whitei
MP1641
15.09
13.37
79.02
78.46
26.03
76.02
T. darti$
MP2225
20.47
32.67
42.45
25.12
27.18
26.11
72.26
16.43
16.64
25.27
60.87
15.66
31.13
62.67
79.88
30.85
35.64
72.07
85.03
13.50
16.56
19.57
36.99
73.13
48.59
33.78
41.66
32.25
10.75
17.48
17.77
78.83
16.06
32.45
47.32
18.85
T. darti
M30733,4,5
127
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
18
ate d O values (1.1% to 25.8%, 21.8% 6 1.5%, n 5 19)
similar to the lower range of d18O values found for sympatric and contemporaneous fossil grazing taxa (3% to
25.7%, 21% 6 2%, n 5 23) from Makapansgat Limeworks (Sponheimer and Lee-Thorp, 1999). Browsing taxa
have significantly (P \ 0.05) higher d18O (5.3% to
25.3%, 1.5% 6 1.5%, n 5 28) compared with both grazing and cercopithecoid taxa. This may suggest that the
cercopithecoid taxa in this study derive significant proportion of their metabolic water from 18O depleted meteoric surface water. It is probable that these taxa were
obligate drinkers similar to extant Papio taxa. Notably
the fossil cercopithecoid specimens were most similar in
d18O to sympatric and contemporaneous fossil grazing
suids (23.1% 6 2%, n 5 5) and equids (21.9% 6 0.8%,
n 5 5) from Makapansgat Limeworks (Sponheimer and
Fig. 1. d13C (open boxes) and d18O (solid boxes) patterns for
the fossil cercopithecoid sample, browsers and grazers from
Makapansgat Limeworks Members 3 and 4 and extant baboons.
All d values reported in %. The boxes represent the 25th–75th
percentiles (with medians as small clear boxes), whiskers represent nonoutlier range, circles represent outliers and stars represent extremes. The two small solid boxes represent the individual values for the two C. williamsi specimens. [Color figure
can be viewed in the online issue, which is available at www.
interscience.wiley.com.]
Lee-Thorp, 1999). Isotope and trace-element patterns
suggest that suids include a substantial proportion of
rootstocks in their diet (Sillen, 1988; Sponheimer et al.,
2005a; Sponheimer and Lee-Thorp, 2006). The underground parts of plants resemble ground water in their
d18O values and are 18O-depleted relative to the above
ground parts (Epstein et al., 1977; Sternberg, 1989). A
large input of rootstocks from both C3 and C4 plants
could partly explain the combination of intermediate carbon isotope values and depleted oxygen isotope values
for these fossil cercopithecoids. Although faunivores
(hyenas in this case) are not shown in Figure 1, the
range of d18O values (23.9% to 22.2%, 23% 6 0.9%, n
5 3) (Sponheimer and Lee-Thorp, unpublished data)
overlaps with those of cercopithecoid taxa. The relatively
low d18O values of faunivores also suggest that animal
foods are depleted in 18O (Tredget et al., 1993; Sponheimer and Lee-Thorp, 1999). Animal protein or fats
could also have contributed to the low d18O values in
these fossil cercopithecoids.
Pp. whitei and Pp. jonesi tooth enamel carbonate
yielded similar mean d13C values of 26.6% 6 3.4% and
a range of 25.6% to 29.5% (n 5 5), and a mean 27% 6
0.7% and a range of 26.5% to 27.5% (n 5 2), respectively (Table 2 and Fig. 1). Both Pp. whitei and Pp. jonesi
enamel carbonate was generally but not significantly
more enriched in 13C than Pp. broomi which yielded an
average d13C value of 29% 6 1.3% and a range of
27.6% to 210% (n 5 3). One Pp. whitei specimen,
MP76 (see Fig. 1), yielded a relatively low d13C value of
29.5%, which was uncharacteristic of specimens which
had been identified as Pp. whitei. The d13C value of this
individual accounts for the large standard deviation
reported for the Pp. whitei group and the range overlap
with the Pp. broomi group. The group consisting of unassigned Parapapio specimens yielded a wide range of d13C
values (22.2% to 29.5%, 27.2% 6 2.4%, n 5 6), which
overlaps with Pp. whitei, Pp. jonesi, and Pp. broomi values. This would be expected if the unassigned specimens
included representatives from all species (or dietary
groupings). The high d13C value (22.2%) obtained for
M3147 alone accounts for large range of d13C values
reported for the unassigned Parapapio sample.
Pp. whitei (20.7% to 22.5%, 21.3% 6 07%, n 5 5)
and Pp. jonesi (20.7% to 22.2%, 21.4% 6 1.1%, n 5 2)
TABLE 2. Table showing stable isotope and trace-element ratio composition for each specimen
Specimen number
MP36
MP3A
M3070
MP119
MP164
MP223
MP76
MP151
MP170
MP2
MP173
MP75
MP208
M3084
MP239
M3079
M3133
M3147
MP222
Taxon
d13C (%)
d18O (%)
Sr/Ca
Ba/Ca
Sr/Ba
C. williansi
C. williansi
Pp. whitei
Pp. whitei
Pp. whitei
Pp. whitei
Pp. whitei
Pp. broomi
Pp. broomi
Pp. broomi
Pp. jonesi
Pp. jonesi
Parapapio sp.
Parapapio sp.
Parapapio sp.
Parapapio sp.
Parapapio sp.
Parapapio sp.
T. darti
24.27
28.17
27.58
27.28
26.37
25.65
29.52
29.46
27.59
29.98
27.45
26.47
27.93
26.33
29.45
29.13
27.83
22.18
22.4
21.8
22.78
21.17
21.39
20.68
22.54
20.85
21.35
21.89
1.14
22.23
20.71
21.31
21.6
22.38
25.84
21.35
20.99
24.35
0.2
0.28
0.4
0.28
0.24
0.23
0.41
0.54
0.18
0.39
0.17
0.18
0.42
0.27
0.54
0.98
0.43
0.07
0.16
0.14
0.15
2.7
1.73
2.83
1.85
0.08
0.12
0.08
0.07
0.29
0.07
0.12
0.07
0.27
0.19
0.08
0.22
2.74
3.49
7.26
2.79
1.32
2.63
1.55
5.77
1
2.79
12.51
1.97
American Journal of Physical Anthropology—DOI 10.1002/ajpa
128
N.H. FOURIE ET AL.
18
both yielded similar d O values to Pp. broomi (1.1% to
21.9%, 20.7% 6 0.6%, n 5 3) (see Fig. 1). Only MP2
yielded a higher d18O value (1.1%), but this value does
not fall outside the range of variation observed for other
fauna, fossil, or extant (Sponheimer and Lee-Thorp,
1999; Codron, 2003). The unassigned Parapapio specimens yielded similar d18O values (21% to 25.8%,
22.2% 6 1.8%, n 5 6) to all three Parapapio taxa, as
would be expected.
The two specimens of C. williamsi yielded very different d13C values indicating widely disparate inputs of C4
foods. MP3A has a d13C value of 28.2% suggesting a
diet largely composed of C3 resources. MP36 has a d13C
value of 24.3% suggesting a mixed diet dominated by
C4 plants. Figure 1 shows that both the C. williamsi
specimens yielded similar d18O values (21.8% and
22.8%, respectively, 22.3% 6 0.7%, n 5 2) to Parapapio specimens (1.1% to 25.8%, 21.6% 6 1.4%, n 5 16),
with no significant difference.
The single T. darti specimen analysed here yielded a
d13C value (22.4%) consistent with a largely C4 diet
similar to that of Theropithecus oswaldi (Lee-Thorp et
al., 1989b; Codron et al., 2005) and other fossil grazers
in Makapansgat valley. The d18O value for this specimen (MP222) was relatively low, overlapping with values of fossil grazing suids (Sponheimer and Lee-Thorp,
1999) and consistent with similarly lower d18O values
for T. oswaldi specimens from Swartkrans (Codron, 2003).
The Pp. broomi sample yielded generally but not significantly higher Sr/Ca ratios (0.41–0.23, 0.39 6 0.16, n
5 3) than Pp. whitei (0.40–0.20, 0.28 6 0.07, n 5 5) and
Pp. jonesi (0.39–0.18, 0.28 6 0.15 n 5 2) (see Fig. 3).
The unassigned Parapapio specimens yielded a range of
values (0.54–0.17, 0.32 6 0.16, n 5 5) representative of
those observed for all Parapapio taxa. There is substantial
overlap in the Sr/Ca ranges between the three Parapapio
taxa. All three taxa are broadly similar in their Sr/Ca ratios
although Pp. broomi tends to a higher mean Sr/Ca ratio.
Pp. whitei sample yielded generally but not significantly higher Ba/Ca ratios (0.16–0.07, 0.13 6 0.04, n 5
4) than Pp. broomi specimens (0.12–0.08, 0.09 6 0.02, n
5 4). The higher Ba/Ca ratio of the Pp. whitei sample is
in keeping with the stable carbon isotope data which
suggest a substantial C4/grass component in their diet.
The two Pp. jonesi specimens had widely different Ba/Ca
ratios from one another. MP75 yielded the highest Ba/Ca
ratio compared with all other fossil cercopithecoid specimens in this study, a relatively high Sr/Ca ratio and a
more depleted d13C value than the other Pp. jonesi specimen, MP173. The high Sr/Ca and Ba/Ca ratios for this
specimen are consistent with grass and rootstocks in the
diet; however, the magnitude of the Ba/Ca ratio as compared with other specimens remains unusual. M3147
also yielded an anomalously high Sr abundance. It
yielded an unusually high Sr and Sr/Ca ratio inconsistent with the range of Sr and Sr/Ca values obtained for
other fossil cercopithecoid specimens. M3147 yielded a
Sr abundance (325 ppm) more than three times that of
the average for the genus Parapapio (93 ppm). These
higher values are more consistent with granitic-dominated geological zones (Sponheimer and Lee-Thorp,
2006), and hence possibly this individual came from an
area with a different geology. Unassigned Parapapio
specimens (0.27–0.07, 0.14 6 0.09, n 5 5) yielded a
range of Ba/Ca ratios overlapping with all three Parapapio taxa as would be expected if it consisted of representatives of all three taxa.
Fig. 2. Sr/Ba patterns for the fossil cercopithecoid sample,
browsers, grazers, grazing suids, and carnivores from Makapansgat Limeworks Members 3 and 4. The boxes represent the
25th–75th percentiles (with medians as small clear boxes),
whiskers represent nonoutlier range, circles represent outliers
and stars represent extremes. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
Samples for trace-element analysis were often very
small as they represented the material left over after,
sometimes, multiple isotopic analyses. MP76 represents
one such very small sample. For this reason there was
not enough material to measure its Ba abundance, but
only yielded Sr and Ca abundances. Ba/Ca and Sr/Ba
ratios could therefore not be calculated for this specimen.
Figure 2 summarizes the Sr and Ba results as a Sr/Ba
ratio. Sr/Ba ratios clearly distinguish Pp. broomi from
Pp. whitei and Pp. jonesi samples; however differences
are not statistically significant. Pp. broomi yielded
higher mean Sr/Ba ratios (4.35 6 2.1, n 5 3) than Pp.
whitei and Pp. jonesi specimens together (2.21 6 0.64, n
5 6.). Pp. broomi also yielded Sr/Ba ratios, which were
relatively high compared with all other comparative
taxa, including the grazing suids, and only overlapped
with the higher end of the range of Sr/Ba ratios for
browsing bovids.
Craniometric analysis
The results of the PCA on the extant chacma baboon
sample showed that all five sets of analyses discriminate
among male and female specimens on the bases of the
subsets of measurements used, and that differences
among the specimens are primarily allometric. The only
exception was the PCA analysis of the Procrustes-rotated
coordinate landmark data for Subset III which did not
discriminate among males and female specimens. Distribution along the PC1 axis was found to be allometric but
not as a function of sex. PC1 loadings for the extant, fossil, and combined subsets were usually significantly and
moderately correlated to the geometric means of specimens. This implies that this axis for most subsets summarizes aspects of both size-related and size-independent
shape. Similarly PC2 usually summarized size-independent shape.
Subsets I–III consisted mainly of data regarding the orbital region, mid-face and aspects of the superior-anterior
areas of the cranium (NA, GL, TP, BR, ON, ZI, ZS, FMO,
GL). Subsets I–III were made up of relatively small num-
American Journal of Physical Anthropology—DOI 10.1002/ajpa
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
129
Fig. 3. Plot of PC1 vs. PC2 based on PCA of Subset V interlandmark distances for the fossil cercopithecoid subsample. (Pp.
whitei 5 4, Parapapio sp. 5 7, T. darti 5 2, C. williamsi 5 1).
Fig. 4. Plot of PC1 vs. PC2 based on PCA of Subset V Procrustes-rotated coordinate landmark data for the fossil cercopithecoid subsample. (Pp. whitei 5 4, Parapapio sp. 5 7,
T. darti 5 2, C. williamsi 5 1).
bers of specimens (6–9 specimens), most of which already
had a species assignment. PCA of inter-landmark distance data and Procrustes-rotated coordinate landmark
data revealed no affinities between specimens that could
be correlated to their current taxonomic assignment or
sex identification. No consistent affinities between specimens were observed across any of these subsets.
Subsets IV and V consisted mainly of data describing
aspects of the mid-face and the length of the muzzle
(PR, NS, PM, RH, NA, PM, ZS). These two subsets consisted of the largest number of specimens (10 and 14
specimens, respectively) of which most Parapapio specimens in either subset had not yet been given a species
designation. Subset IV and V consisted similar sets of
inter-landmark distance and Procrustes-rotated coordinate landmark data and PCA analyses of the two subsets yielded similar relationships between specimens.
PCA of the inter-landmark distance and Procrustesrotated coordinate landmark data for Subset V are presented below in Figures 3 and 4. PCA of both subsets of
data discriminated between specimens based on muzzle
length to some extent. In Figure 3 specimens with relatively shorter muzzles fall more positively on the PC1
axis, while specimens with relatively narrow muzzles
fall more positively along the PC2 axis. Two of the three
specimens that are more negatively oriented along the
PC1 axis (i.e., long muzzles) are male Pp. whitei specimens (MP221 and MP223), where as among specimens,
which are oriented more positively along this axis (i.e.,
shorter muzzles) there are four female specimens, to
assigned as Pp. whitei (M3070 and MP119) and to unassigned Parapapio specimens (M3079 and M239). The sex
of other specimens has not been determined. The two T.
darti specimens fall more negatively on the PC2 axis
and more positively along the PC1 axis distinguishing
them from the Parapapio sample a shaving somewhat
longer and broader muzzles. The C. williamsi specimen
does not appear to be distinct from the Parapapio sample although it is oriented to the edge of the Parapapio
distribution. The PCA of subset IV inter-landmark distances discriminated more clearly between the papionin
sample and C. williamsi.
PCA of Procrustes-rotated coordinate landmark data
for both subsets IV and V (see Fig. 4) revealed no dis-
tinct polarization of the specimens although there were
general similarities regarding the associations between
specimens to the PCAs of the inter-landmark distance
data. The PC1 axes summarized both size-dependant
and -independent shape whereas the PC2 axes summarized size-independent shape. The larger Parapapio specimens (MP221, MP223, and MP47) cluster toward the
left on the PC1 axis but at the same time they also associate with other Parapapio specimens which in the PCA
analysis appeared dissimilar in size to these specimens.
This would suggest that the unassigned Parapapio specimens, which cluster with the larger male Pp. whitei
specimens, although somewhat smaller, differ from the
other Parapapio specimens in size-independent shape of
the muzzle. Interestingly Subset V data are unable to
distinguish the T. darti specimen (M3073) and the C.
williamsi specimen (MP3) from the Parapapio subsample. Only MP222 is clearly dissociated from the Parapapio subsample in shape along the PC2 axis.
When the Parapapio subsamples were compared with
the extant chacma baboon sample in PCA of both interlandmark distances and Procrustes-rotated coordinate
landmark data they were not found to vary more than
the extant chacma sample. The Parapapio subsamples
were often distinct in both shape and size from the
extant chacma sample although some overlap with the
female components of the extant chacma sample was
observed. This may be due to similarities in aspects of
size and the length of the muzzle. Figure 5 represents a
typical result when the extant chacma and Parapapio
sample were analyzed together.
Turning to the dental measurements, the plot of the
first two PC axes (see Fig. 6), based on a PCA analysis
of the measurements of the three dimensions recorded
for the right third molars of Parapapio specimens which
preserved these dental dimensions, separate specimens
along the both the PC1 and PC2 axis. Specimens were
separated along taxonomic lines as would be expected.
Molar size should be a good predictor of taxon since
molar size was a key aspect in the historical identification of the taxa and the subsequent assignment of Parapapio specimens to each of the three taxa (Broom, 1940;
Freedman, 1957). Pp. whitei specimens (MP119, MP223,
and MP76) group together as expected. Examination of
American Journal of Physical Anthropology—DOI 10.1002/ajpa
130
N.H. FOURIE ET AL.
Fig. 5. Plot of PC1 vs. PC2 based on PCA of Subset V interlandmark distances for the extant baboon sample and Subset V
fossil cercopithecoid subsample (n 5 45, P. h. ursinus 5 31, Pp.
whitei 5 4, Parapapio sp. 5 7, T. darti 5 2, C. williamsi 5 1).
the raw data confirms that they have larger measurements on all three dimensions than all other specimens.
An unassigned Parapapio specimen (M3147) is also associated with this group and yielded measurements similar
to Pp. whitei specimens in this group.
Inspection of the individual measurements for each of
the three dimensions of the third molars confirms that
Pp. whitei specimens are larger than Pp. broomi specimens for buccul-lingual posterior and buccul-lingual anterior lengths. There is some overlap in mesio-distal distances between Pp. whitei specimens and Pp. broomi.
MP173 was the only Pp. jonesi specimen to preserve any
dental dimensions. Surprisingly it had a mesio-distal
length greater than any of the Pp. whitei and Pp. broomi
specimens, contrary to what would be expected from the
smallest Parapapio taxon. This dimension may have
been exaggerated by a crack running through the tooth
between the two cusps which may have contributed as
much as 1 mm to the measurement. Nonetheless, a mesiodistal measurement of 11.6 mm still places it in the range
of measurements found for Pp. whitei and Pp. broomi.
When the dimensions of the third molars of fossil Parapapio sample were compared with the extant chacma
baboon sample, the fossil Parapapio sample were somewhat smaller but showed considerable overlap with the
extant chacma sample along both axes. There also did
not appear to by any undue variability in the Parapapio
sample compared with the extant chacma sample.
DISCUSSION
Stable isotope and trace-element analysis of
fossil cercopithecoid specimens
The ecological position of fossil cercopithecoidea as a
group within the community ecology of Makapansgat
during Member 2 and 3 times is striking, as most taxa
appear to fill the gap between browsing and grazing
fauna. The Parapapio diet appears to be truly mixed,
including both C3 and C4 foods. Parapapio yielded relatively low d18O ratios, similar to water-dependant taxa
such as suids and equids. However, Parapapio’s d13C values were dissimilar to these taxa, and indicated a stronger C3 component in the diet. Given this combination,
Fig. 6. Plot of PC1 vs. PC2 based on PCA of three dimensions of the right third molar of selected Parapapio specimens
and the extant chacma baboon sample. (n 5 39, Pp. whitei 5 4,
Pp. broomi 5 3, Parapapio sp. 5 3, P. h. ursinus 5 29).
we can deduce that their C3 foods were unlikely to be
leaves but rather 18O-depleted rootstocks and fruits. The
combination of Sr/Ca, Ba/Ca and Sr/Ba ratios in the Parapapio sample resemble the results for extant warthogs
(Sponheimer et al., 2005a) and mole rats (Sponheimer
and Lee-Thorp, 2006). These animals include significant
but varying proportions of rootstocks in their diet. In
summary, the relatively low d18O and (high) d13C values
observed for Parapapio suggest that they were frequent
drinkers and/or partook of C3 or C4 rootstocks, while the
trace-element composition is consistent with a reliance
on the underground parts of plants.
Within the genus Parapapio, stable isotope and trace
element analyses identify two overlapping dietary groupings. Stable carbon isotope analysis identifies a 13Cenriched and a 13C-depleted group. The former consists
mostly of specimens attributed to Pp. whitei and Pp.
jonesi, while the latter is comprised of specimens attributed to Pp. broomi. There were no noticeable differences
between Parapapio taxa in d18O, suggesting that all
were obligate drinkers. Although these dietary groupings
are grossly correlated to taxonomic groupings, individual
specimens within Pp. whitei (MP76) and Pp. broomi
(MP170) yielded d13C values unlike those of the majority
of specimens within their respective taxa; MP76 (Pp.
whitei) yielded d13C more typical of Pp. broomi specimens and MP170 (Pp. whitei) yielded d13C more typical
of Pp. whitei specimens. This might indicate that the
two specimens have been incorrectly identified, since
their ecological affinities do not match their taxonomic
affinities. Alternatively, this could hint at dietary variability within each taxon, with the ranges of this variability currently poorly represented due to the small
sample sizes. Larger sample sizes might reveal greater
variability and in effect close the ecological gap between
different taxa. However, as the d13C data stand, they
suggest that Pp. whitei and Pp. jonesi included similar
amounts of C4 foods in their diets, and that Pp. broomi
included more C3 foods in its diet at Makapansgat.
A similar bimodal distribution of d13C values is
reported for Parapapio specimens from Sterkfontein,
where specimens attributed to Pp. broomi and Pp. jonesi
were more 13C-enriched than specimens attributed to
American Journal of Physical Anthropology—DOI 10.1002/ajpa
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
Pp. whitei (Codron et al., 2005). Furthermore, both
groups of 13C-enriched Parapapio specimens from Sterkfontein (n 5 7, mean 5 27.5% 6 1.1% d13C) and Makapansgat (n 5 6, mean 5 26.8% 6 0.8% d13C) indicates
a similar reliance on C4 resources.
It is worth noting that the taxonomic affinities of the
two dietary groups at Sterkfontein and Makapansgat differ. At Sterkfontein it is Pp. whitei that is characterized
by a more C3-oriented diet, whereas at Makapansgat it
is Pp. broomi that fits that dietary profile. Samples from
Sterkfontein were mainly derived from isolated teeth
and it was also noted that dietary groupings corresponded relatively poorly to the taxonomic assignment of
specimens. Therefore, it is possible that the lack of similarity between the two sites results from the problematic
taxonomic identification of Parapapio specimens from
isolated teeth at Sterkfontein. Alternatively, as indicated
by the craniometric analysis of the Makapansgat material, the three-species breakdown of Parapapio may not
reflect real biological units, and therefore much of the
noise and lack of correlation in the data may result from
this problem. More rigorous taxonomic investigations
may resolve some of these disparities.
Sr/Ca, Ba/Ca and Sr/Ba also identified two overlapping
dietary groupings within the fossil Parapapio sample,
corresponding to the dietary groupings observed for
d13C. Sr/Ca and Ba/Ca have been found to be relatively
variable (Sillen, 1988), especially in primary consumers
(Burton et al., 1999), and may produce a large range of
values for a specific taxon (Sillen, 1988). Here, the identification of patterning within such relatively variable
indicators may be considered to be an indication of real
dietary differences. Pp. whitei specimens tended to have
generally lower Sr/Ca ratios, higher Ba/Ca ratios and
lower Sr/Ba ratios than Pp. broomi specimens. This difference was particularly apparent in the distribution of
Ba/Ca and Sr/Ba ratios. Sponheimer and Lee-Thorp
(2006) report that mole rats, a taxon that largely feeds
off rootstocks (Kingdon, 1997), yielded much higher Sr/
Ba ratios than all other taxa. Hence, a plausible explanation for the high Sr/Ba observed for Pp. broomi would
again reflect a greater dependence on rootstocks in this
taxon than in Pp. whitei and Pp. jonesi.
Although Pp. whitei Sr/Ca ratios appeared to be generally low, Pp. broomi Sr/Ca ratios appeared to be quite
variable and did not cluster tightly. The combination of
d13C, Sr/Ca, Ba/Ca, and Sr/Ba ratios observed for the
various Parapapio taxa in the sample is consistent with
the interpretation that Pp. broomi specimens included
substantially more C3 rootstocks in their diet than Pp.
whitei specimens. The two Pp. jonesi specimens yielded
widely different Sr/Ca and Ba/Ca ratios to one another.
MP75 yielded a high d13C signal consistent with a significant C4 component in the diet and high Sr/Ca and Ba/
Ca ratios, while MP173 yielded a lower d13C value and
low Sr/Ca and Ba/Ca ratios. This latter combination has
not been identified elsewhere. Given the high natural
intra-specific variability of these trace-elements, and the
small sample sizes in this study, results from individual
specimens, however, should not be over-interpreted. Notwithstanding the need for larger sample sizes and the
nature of Sr/Ca and Ba/Ca variability, some of the variability in the combinations of dietary signals may be due
to specific dietary inputs not yet modelled in modern
African savannah ecosystems.
Taken together, the current stable isotope and traceelement data from Parapapio samples from Swartkrans,
131
Sterkfontein, and Makapansgat indicate the presence of
two identifiable but overlapping dietary regimes that are
not particularly well correlated to the current specimen
identifications (Codron, 2003; El-Zaatari et al., 2005).
Importantly, the apparent dietary groupings revealed by
the analyses presented here do not necessarily argue for
the presence of two distinct taxa at Makapansgat, but
only points out two different dietary ecologies. These differences may reflect dietary shifts that result from environmental changes over time. Several thousand if not
tens of thousands of years may be represented within
the Member 3 and 4 deposits at Makapansgat, during
which time environmental oscillations is likely to have
occurred.
The isotopic variability, and by extension the dietary
variability, of the Parapapio sample from Makapansgat
does not exceed the variability observed within several
extant populations of baboons from the Awash Valley,
Ethiopia, measured over several decades (Fourie et al.,
unpublished data). However, this comparison should not
be taken too far since the Awash Valley in Ethiopia represents a much more extreme and arid environment to
that which would have been present during Makapansgat Members 3 and 4 times. Further research on well
provenanced and securely identified samples is clearly
needed to confirm dietary characterizations and the nature dietary variability within these fossil taxa, to be
certain about the dietary ecology and niche separation
within Parapapio.
The results for the two C. williamsi specimens compared well with results for this same taxon reported by
Codron (2003) and Codron et al., (2005). One specimen
(MP3A) yielded a d13C value consistent with a mixed/C3oriented diet, while the other specimen (MP36) yielded a
d13C value consistent with a mostly C4 diet. Such a disparity in diets within a single taxon is unusual, and
implies that these individuals were exploiting two completely distinct ecological niches. It has been noted that
this taxon appears to be post-cranially adapted to a terrestrial lifestyle while retaining a dental morphology
typical of folivorous colobines (Benefit and McCrossin,
1990). Dental wear studies have also suggested a terrestrial diet including a substantial amount of grasses (ElZaatari et al., 2005). Some C4 dietary contribution is to
be expected, but a C4-dominated diet is not, given the
taxonomic affinities and dental morphology of this taxon.
Codron (2003) has argued that such a disparate signal is
unlikely to be due to temporal shifts in ecology, as such
a radical dietary shift over a relatively short period of
time (in evolutionary terms) seems unlikely. It has been
tentatively suggested that the existence of a second species of Cercopithecoides at Sterkfontein and Swartkrans
may be the most parsimonious explanation for the existence of two such distinctive dietary ecologies within the
C. williamsi sample (Codron, 2003; Codron et al., 2005).
The results of this study are consistent with this hypothesis, and extend the range of this observation to Makapansgat Members 3 and 4. The presence of two distinct
dietary ecologies within this (relatively) morphologically
homogenous taxon also has implications for paleoecological reconstructions. C. williamsi is an extinct colobine
monkey, and all extant colobine monkeys are strict arboreal folivores living in wooded environments. The presence of Cercopithecoides sp. in fossil assemblages has
therefore often been interpreted as indicating the presence of more closed wooded habitats in paleonenvironmental reconstructions (e.g., WoldeGabriel et al., 1994).
American Journal of Physical Anthropology—DOI 10.1002/ajpa
132
N.H. FOURIE ET AL.
Yet the data presented here for C. williamsi, in conjunction with the dietary data from Swartkrans and Sterkfontien (Codron et al., 2005), stress again that assumptions about ecological uniformitarianism based on taxonomic affinities may not always be valid and require testing.
Finally, the results of this study also offer some
insights into the dietary ecology of Theropithecus. The T.
darti (MP222) specimen yielded a d13C value consistent
with a predominantly C4 diet with a small C3 component. This specimen also yielded relatively high Sr/Ca
and Ba/Ca ratios consistent with a high proportion of
above ground plant parts. The d13C value for this specimen is nearly identical to the mean d13C values reported
for T. oswaldi from Swartkrans (Lee-Thorp et al., 1989b;
Codron et al., 2005). Taken together, these data question
interpretations—based on small molar morphology and
dental wear differences (Benefit and McCrossin, 1990)—
that T. darti had a more frugivorous and folivorous diet
than T. oswaldi. Rather, d13C and trace-element data
suggest that T. darti and T. oswaldi both relied heavily
on C4 grasses.
Craniometric assessment of fossil cercopithecoid
taxa from Makapansgat limeworks site
The main reason for choosing relatively complete
specimens for isotopic analysis, rather than fragmentary
remains or isolated teeth, was that it provided morphological context for interpreting the results. One of the
expectations was that any dietary distinctions found
would most likely occur along taxonomic, or at least morphological, lines. This was not the case. Certainly the
results of the craniometric analyses showed size and
shape differences among fossil cercopithecoids. Some of
these differences may reflect sexual dimorphism in the
sample. Analysis of Subsets III, IV, and V inter-landmark distance data appeared to discriminate between
larger and smaller fossil cercopithecoid specimens with
respect to the orbital region and the length of the muzzle
and mid-face. Larger specimens for these subsamples
were usually male, and smaller specimens were usually
female, although a large proportion of these subsamples
often consisted of unassigned and unsexed Parapapio
specimens. It is therefore not entirely clear whether the
polarization of specimens is specifically related to a sexually dimorphic pattern in the data or distinct taxonomic
morphotypes, although separation along the PC1 axis for
Subsets IV and V suggest the former. Similarly, analysis
of coordinate data were not able show any consistent
association between specimens of the same species or
sex. For example, in the analysis of Subset I coordinate
data the Pp. jonesi specimen MP75 and the Pp. whitei
specimen MP164 clearly paired together more closely
than to any other specimens with respect to the shape of
the anterior neurocranium.
Certain individual specimens stand out in their
extreme size. MP221 and MP223 (Pp. whitei, males)
were consistently identified as relatively large specimens
in analyses of distance data. This grouping of Pp. whitei
specimens away from other Parapapio specimens based
on size fits well with the definition of Pp. whitei as the
largest of the three Parapapio taxa at Makapansgat.
MP47 is a relatively large unassigned Parapapio specimen associated closely with MP221 and MP223 on PC1
and PC2 in the analyses of Subset IV and V inter-landmark distances. Surprisingly, MP170, a female Pp.
broomi specimen, was also larger than all conspecifics.
This specimen was most closely associated with the two
large male Pp. whitei specimens (MP221 and MP223) in
aspects of the orbital region and mid-face. However none
of these specimens differed consistently in shape from
other smaller Parapapio specimens. In analyses of coordinate data these ‘‘large’’ Parapapio specimens were not
separated from smaller Parapapio specimens. Analyses
of subsets II, IV, and V coordinate data did not indicate
that MP221 and MP223 shared any great affinities in
size-independent shape of the orbital region (Subset II)
and size-dependant shape of the muzzle (Subset IV and
V). MP221 and MP223 often shared greater shape affinities (both size-correlated and size-independent) with
other specimens than with each other. Only MP170
appeared to differ in both size and size-independent
shape of the anterior neurocranium (Subset I) from other
specimens, but appeared quite similar to the specimens
in shape and size of the orbital region and mid-face
(Subset II).
Despite the presence of consistent size and/or shape
differences among certain specimens throughout these
analyses, there was no consistent taxonomic patterning
to these differences. The Pp. broomi male (MP2) and the
Pp. broomi female (MP170) differed substantially in size
and shape of the anterior neurocranium from other Pp.
whitei, Pp. broomi, and Pp. jonesi specimens based on
the analysis of coordinate data, and also differed from
one another in the analysis of Subset I inter-landmark
distances. In the analyses of Subset II and III inter-landmark distances, they appear to associate with MP221
and MP223, indicating greater similarity in size and
shape with these two male Pp. whitei specimens than
with other Pp. broomi specimens. The Pp. jonesi specimen (MP75) did not appear to be smaller or different to
other Parapapio specimens regardless of their taxonomic
assignment. No consistent patterned differences between
Pp. whitei and Pp. broomi specimens were apparent in
the analysis of scaled (coordinate) and unscaled (distance) data across subsets.
When the extant and fossil samples are analyzed together two patterns emerge. First, there appears to be
some overlap between the two groups, with larger Parapapio specimens falling within the female extant chacma
baboon range. Larger specimens such as MP170 (Subset
II and III), MP47 (Subset III, IV, and V), MP221 (Subset
IV and V), and MP223 (Subset IV and V) regularly fall
within the range of the extant female chacma sample for
size and shape of the orbital regions, upper mid-face and
muzzle. However, Parapapio did appear distinctive from
the extant chacma baboon sample in the size and shape
of the anterior neurocranium; analyses of the coordinate
data showed more distinct shape differentiation between
the Parapapio sample and the extant chacma baboon
sample. This implies that Parapapio, although similar to
the chacma sample in certain anatomical regions of the
cranium, differed substantially and consistently from the
extant chacma baboon sample in both size and shape.
The second interesting result to emerge from the pooled
analyses is that the entire Parapapio sample did not usually appear to be more variable than the extant chacma baboon sample, in aspects of size or shape, even though
greater intra-specific variability in the fossil sample was
expected to some extent due to the inclusion of multiple
taxa and the possibility of temporal depth (Williams et al.,
2007). Additionally, the extant baboon sample is derived
from a geographically restricted population of a single subspecies of baboon, which is unlikely to be an adequate
American Journal of Physical Anthropology—DOI 10.1002/ajpa
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
model for size and shape variation across the fossil sample.
Despite this, the fossil Parapapio sample did not vary more
than the extant baboon sample for any of the anatomical
regions of the skull that were analyzed. Arguments for
multiple taxa based on excessive variation in the fossil Parapapio sample, at least within this sample from Makapansgat and for the variables analyzed here, are inconsistent
with the craniometric data presented here.
Finally, it is necessary to consider the dental dimensions of the fossil samples, especially as Parapapio species have been mainly defined and distinguished from
one another based on the relative size of the post-canine
dentition (e.g. Broom, 1940; Freedman, 1957; Maier,
1970; Freedman, 1976). In general, dental measurements correlated with taxonomic assignments, although
Pp. whitei and Pp. broomi show overlap in dental dimensions. However, this correlation has little meaning, as
comparing these dimensions can only yield results consistent with the previous taxonomic order, because the
taxa themselves have been defined by the dimensions of
the post-canine teeth.
Parapapio specimens also show substantial overlap
with the extant chacma baboon sample in the measured
dimensions of the M3. It is therefore likely that the
dimensions of the post-canine teeth of Parapapio may in
some cases not be distinguishable from other fossil
papionin taxa such as Papio robinsoni or P. h. ursinus.
Dental dimensions alone should therefore not be used to
distinguish between Parapapio and other papionin taxa.
Furthermore, it is implicit in the Parapapio literature
that small post canine teeth/tooth equals a small specimen. The three species are commonly described as representing three different size categories, when in fact those
size categories appear only to apply to the post canine
tooth size and not to the whole specimen. MP173,
although not included in the craniometric analysis, is a
relatively large specimen but it has small pre-molars
and therefore was assigned to the ‘‘smaller’’ of the three
Parapapio species. Similarly, MP2, identified as a Pp.
broomi male (Freedman, 1960), yielded some of the
smallest M3 measurements, while craniometric analysis
indicated it was of intermediate to large size relative to
other specimens in Subsets I, II, and III. Smaller postcanine teeth do not necessarily imply a small specimen.
The assumption that Pp. jonesi specimens are absolutely
smaller than Pp. broomi specimens, and that Pp. broomi
specimens are absolutely smaller than Pp. whitei specimens, does not appear to hold.
Taken together, the craniometric and dental analyses
indicate that the taxonomic assignment of specimens
within this sample, and by extension within and among
samples from other sites, may be substantially flawed.
The data did not yield any taxonomic signal which would
indicate the presence of multiple taxa within this sample, although the possibility that multiple taxa are present is not entirely discounted. Historical classification
criteria for Parapapio taxa and the taxonomic assignment of Parapapio specimens are clearly problematic.
Future research should review the taxonomy and the
individual taxonomic assignments of specimens using
large samples from various sites. This will help us to
understand the nature of the variation reported for the
fossil samples, and might reveal temporally correlated
trends between sites. Such a comprehensive review
using multiple sites and appropriately constructed
extant models will yield a more reliable and thoroughly
understood taxonomy. This is crucial if further research
133
into the biology and ecology of Plio-Pleistocene cercopithecoid taxa is to yield meaningful results.
CONCLUSION
This study has provided the first fine-grained dietary
data on well preserved and more securely identified partially complete cranial fossil cercopithecoid specimens, in
an attempt to circumvent taxonomic uncertainties inherent in using isolated teeth and dentognathic specimens.
Dietary indicators suggest that two overlapping dietary
ecologies are represented in the Parapapio sample.
These dietary groups are loosely correlated to specimens
attributed to Pp. broomi, which appear to have had a
mixed C3 diet with substantial input from rootstocks,
and to specimens attributed to Pp. whitei/Pp. jonesi,
which appear to have had a mixed C4 diet with a lesser
rootstock component. Yet craniometric analyses found no
clear taxonomic signal within this genus, and no consistent craniometric affinities could be demonstrated
between specimens assumed to belong to the same species. These analyses also highlighted inconsistencies in
the current taxonomic identification of specimens.
Although not discounting the possibility that more than
one taxon may be present in the Parapapio sample, the
data do not support such a conclusion. In the context of
the craniometric analyses, it would appear that two
overlapping dietary ecologies are present within a single
taxon. This is not necessarily surprising, as many papionin taxa are very adaptable generalists, capable of
exploiting a wide range of habitats and resources (Hall,
1961; DeVore and Hall, 1965; Hamilton et al., 1978). The
bimodality in the dietary signal of the Parapapio sample
from Makapansgat Members 3 and 4 may reflect a predictable dietary response to climatic and environmental
oscillations over time. Alternatively, it is also possible
that the two observed dietary ecologies may reflect the
presence of two ecologically distinct taxa in the Parapapio sample, in which case the variables (and variable
combinations) used for the craniometric analyses were
not appropriate for distinguishing between these closely
related taxa. These possibilities must be further tested
via morphometric analyses of larger samples of Parapapio from various sites, as well as more extensive dietary
analyses of well provenanced and securely identified
specimens. Additionally, the dietary data also suggest
that T. darti—or at least the individual sampled here—
did not have a diet that differed from T. oswaldi, and
that C. williamsi at Makapansgat exhibits two very different dietary ecologies, as has been found at other sites
(Luyt, 2001; Codron, 2003; Codron et al., 2005).
To sum, although not conclusive, the results of this
study suggest that Parapapio at Makapansgat could represent a single taxon with an eclectic and adaptable dietary ecology. These results have also demonstrated that
it possible to detect subtle dietary differences between
related taxa using stable isotope and trace-element tools.
However, it is important to remember that the information yielded by these techniques only reveals part of the
dietary, ecological, and behavioral complexity of the animal. Only through a synthesis of functional morphology,
dental wear, and biogeochemical techniques can a more
holistic picture of the ecology of the animals investigated
be arrived at, as each provides unique biological, evolutionary, and behavioral information. Future paleo-dietary studies should attempt to apply the full suite of
techniques to attain greater resolution and tease apart
American Journal of Physical Anthropology—DOI 10.1002/ajpa
134
N.H. FOURIE ET AL.
ecological relationships and dietary components in fossil
foodwebs.
ACKNOWLEDGMENTS
I owe a special thanks to my coauthors, Prof. Julia
Lee-Thorp and Dr. Becky Ackermann for their help,
patience, and advice. I also thank Dr. Matt Sponheimer,
Dr. Daryl Codron, Mr. John Lanham (for help with the
mass spec), Prof. Judith Sealy, Dr. Andreas Spath, and
his staff (ICP-MS) for their various contributions and
support. I thank Dr. Mike Raath of the Bernard Price
Institute for Palaeontological Research (BPI) and the
Department of Anatomical Sciences at the University of
the Witwatersrand for access to the fossil primate material from Makapansgat, and Ms. Denise Hamerton from
the Iziko Natural History Museum, Cape Town, for making the baboon collection at the museum available to me
for craniometric sampling. I am also very grateful to my
late uncle, Flip du Plessis and my aunt, Hermien du
Plessis, for their investment in my university career and
future. I also thank my parents, sister and extended
family for their support and love.
LITERATURE CITED
Alberts SC, Altmann J. 2001. Immigration and hybridization
patterns of yellow and anubis baboons in and around Amboseli, Kenya. Am J Primatol 53:139–154.
Benefit BR, McCrossin ML. 1990. Diet, species diversity and
distribution of African fossil baboons. Kroeber Anthropol Soc
Papers 70–71:77–93.
Birchette MG. 1981. Postcranial remains of cercopithecoides.
Am J Phys Anthropol 54:201.
Bowen HJM, Dymond J. 1955. Strontium and barium in plants
and soils. Proc R Soc London Ser B 144:355–368.
Bowen HJM, Dymond J. 1956. The uptake of calcium and strontium by plants from soils and nutrient solutions. J Exp Bot
7:264–272.
Brain CK. 1981. The fossil animals. In: Brain CK, editor. The
hunters or the hunted? An introduction to African cave
taphonomy. Chicago: University of Chicago Press. p 147–190.
Broom R. 1940. The South African Pleistocene cercopithecoid
apes. Ann Transv Mus 20:89–100.
Burton JH, Price TD. 1991. Paleodietary applications of barium
values in bone. In: Pernicka E, Wagner GA, editors. Proceedings of the 27th International Symposium on Archaeometry,
Heidelberg, 1990. Basel: Berkhauser Verlag. p 787–795.
Burton JH, Price D, Middleton WD. 1999. Correlation of bone
Ba/Ca and Sr/Ca due to biological purification of calcium.
J Arch Sci 26:609–616.
Carter ML. 2001. Sensitivity of stable isotopes (13C, 15N and
18
O) in bone to dietary specialization and niche separation
among sympatric primates in Kibale National Park, Uganda.
Unpublished Ph.D. thesis, University of Chicago.
Cerling TE, Harris JM, Ambrose SH, Leaky MG, Solounias N.
1997. Dietary and environmental reconstruction with stable
isotope analysis of herbivore tooth enamel from the Miocene
locality of Fort Ternan, Kenya. J Hum Evol 33:635–650.
Codron DM. 2003. Dietary ecology of chacma baboons (Papio
ursinus (Kerr, 1792)) and Pleistocene cercopithecoidea in savanna environments of South Africa. Masters thesis, University of Cape Town.
Codron D, Luyt J, Lee-Thorp JA, Sponheimer M, DeRuiter D,
Codron J. 2005. Utilization of savanna-based resources by
baboons during the Plio-Pleistocene. S Afr J Sci 101:254–
248.
Delson E. 1992. Evolution of Old World monkeys. In: Jones JS,
Martin RD, Pilbeam D, Bunney S, editors. Cambridge encyclopedia of human evolution. Cambridge: Cambridge University Press. p 217–222.
Delson E. 1993. Theropithecus fossils from Africa and India and
the taxonomy of the genus. In: Jablonski NG, editor. The rise
and fall of a primate genus. Cambridge: Cambridge University Press. p 157–189.
DeVore I, Hall KRL. 1965. Baboon ecology. In: DeVore I, editor.
Primate behavior. New York: Holt, Rinehart and Winston.
p 20–53.
Dunbar RIM, Dunbar EP. 1974. Ecological relations a niche separation between sympatric and terrestrial primates in Ethiopia. Folia Primatol 21:36–60.
Eisenhart WL. 1974. The fossil cercopithecoids of Makapansgat
and Sterkfontien. BA thesis, Department of Anthropology,
Harvard College.
Elias RW, Hirao Y, Patterson CC. 1982. The circumvention of
the natural biopurification of calcium along nutrient pathways by atmospheric inputs of industrial lead. Geochim Cosmochim Acta 46:2561–2580.
Elton S. 2001. Locomotor and habitat classifications of cercopithedoid post cranial material from Sterkfontein member 4.
Bolt’s Farm and Swartkrans members 1 and 2, South Africa.
Paleont Afr 37:115–126.
El-Zaatari S, Grine FE, Teaford MF, Smith HF. 2005. Molar
microwear and dietary reconstructions of fossil cercopithecoidea from the Plio-Pleistocene deposits of South Africa. J
Hum Evol 49:1–26.
Epstein S, Thompson P, Yapp CJ. 1977. Oxygen and hydrogen
isotopic ratios in plant cellulose. Science 198:1209–1215.
Freedman L. 1957. The fossil cercopithecoidea of South Africa.
Ann Transv Mus 23:122–213.
Freedman L. 1960. Some new fossil cercopithecoid specimens
from Makapansgat, South Africa. Palaeont Afr 7:7–45.
Freedman L. 1970. A new checklist of fossil cercopithecoidea of
South Africa. Paleont Afr 13:109–110.
Freedman L. 1976. South African fossil cercopithecoidea: a reassessment including a description of new material from
Makapansgat, Sterkfontien and Taung. J Hum Evol 5:297–315.
Gautier-Hion A. 1980. Seasonal variation of diet related to species an sex in a community of Cercopithecus monkeys. J Anim
Ecol 49:237–269.
Gautier-Hion A, Gautier J. 1986. Sexual dimorphism, social
units and ecology among sympatric forest guenons. Symp Soc
Study Hum Biol 24:61–77.
Gilbert C, Sealy J, Sillen A. 1994. An investigation of barium,
calcium and strontium as paleodietary indicators in the
Southwestern Cape, South Africa. J Arch Sci 21:173–184.
Gonfiantini R. 1965. Isotope effects in the evaporation of saltwater. Atti Soc Tuscana Nat Sci Pisa 72:5–22.
Hall KRL. 1961. Feeding habits of the chacma baboon. Adv Sci
17:559–567.
Ham R. 1995. Polyspecific association between grey-cheeked
mangabeys (Cercocebus albigena) and four sympatric primate
species in the Lopé Reserve, Gabon. Folia Primatol 64:100.
[Abstract from 1995 Spring Meeting Primate Society of Great
Britain, University of Edinburgh; April, 95].
Hamilton WJ III, Buskirk RE, Buskirk WH. 1978. Omnivory
and utilization of food resources by chacma baboons, Papio
ursinus. Am Nat 112:911–924.
Hammer O, Harper DAT, Ryab PD. 2001. PAST: Palaeontological
Statistics software package for education and data analysis.
Palaeontologica Electronica 4:(1).
Jones TR. 1978. The skull and mandible of the South African
baboon: a morphological study. Johannesburg: Witwatersrand
University Press.
Kingdon J. 1997. The Kingdon field guide to African mammals.
New York: Academic Press.
Kohn MJ. 1996. Predicting animal d18O: accounting for diet and
physiological adaptation. Geochim Cosmochim Acta 60:4811–
4829.
Koch PL, Turos N, Fogel ML. 1997. The effects of sample treatment on the isotopic integrity of carbonate in biogenic hydroxylapatite. J Arch Sci 24:417–429.
Kohn MJ. 1996. Predicting animal d18O: accounting for diet and
physiological adaptation. Geochim Cosmochim Acta 60:4811–
4829.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
DIETARY ECOLOGY OF FOSSIL CERCOPITHECOIDS
Kummer H. 1971. Primate societies: group techniques of ecological adaptation. Chicago: Aldine.
Leaky MG. 1982. Extinct large colobines from the Plio-Pleistocene of Africa. Am J Phys Anthropol 58:153–172.
Leaky MG. 1993. Evolution of Theropithecus in the Turkana basin. In: Jablonski N, editor. Theropithecus: the rise and fall of
a primate genus. Cambridge: Cambridge University Press.
p 85–125.
Lee-Thorp JA, Sponheimer M. 2003. Three case studies used to
reassess the reliability of fossil bone and tooth enamel isotope
signals for paleodietary studies. J Anthropol Archaeol 22:208–
216.
Lee-Thorp JA, Sponheimer M, van der Merwe NJ. 2003. What
do stable isotopes tell us about hominid dietary and ecological
niches in the Pliocene? Int J Osteoarcaeol 13:104–113.
Lee-Thorp JA, van der Merwe NJ, Brain CK. 1989. Isotopic evidence for dietary differences between two extinct baboon species from Swartkrans. J Hum Evol 18:183–190.
Luyt CJ. 2001. Revising paleoenvironment from the hominidbearing Plio-Pleistocene sites: new isotopic evidence from
Sterkfontein. Unpublished Masters thesis, University of Cape
Town.
Luz B, Kolodny Y. 1985. Oxygen isotope variations in phosphate
of biogenic apatites. IV. Mammal teeth and bones. Earth
Planet Sci Lett 75:29–36.
Luz B, Kolodny Y, Horowitz M. 1984. Fractionation of oxygen
isotopes between mammalian bone-phosphate and environmental drinking water. Geochim Cosmochim Acta 48:1689–1693.
Maguire JM. 1998. Makapansgat: a guide to the paleontological
and archaeological sites of the Makapansgat Valley. Pretoria:
Desktop Creations.
Maier W. 1970. New fossil Cercopithecoidea from the lower
Pleistocene cave deposits of the Makapansgat Limeworks,
South Africa. Paleont Afr 13:69–109.
Mitani M. 1991. Niche overlap and polyspecific association
among sympatric cercopithecoids in the Campo Animal
Reserve, Southwestern Cameroon. Primates 32:137–151.
Nagel U. 1970. Social organisation in a baboon hybrid zone. In:
Procedures of the 3rd International Congress on Primatology,
Basel: Karger. p 223–233.
Nagel U. 1973. A comparison of anubis baboons, hamadryas
baboons and their hybrids at a species order in Ethiopia. Folia
Primatol 19:104–165.
Newman TK, Jolly CJ, Rogers J. 2004. Mitochondrial phylogeny
and systematics of baboons (Papio). Am J Phys Anthropol
124:17–27
Partridge TC. 2000. Hominid bearing cave and tufa deposits. In:
Partridge TC, Maud RR, editors. The Cenozoic of Southern
Africa. New York: Oxford University Press. p 100–131.
Phillips-Conroy JE, Jolly CJ. 1981. Sexual dimorphism in two
subspecies of Ethiopian baboons (Papio hamadryas) and their
hybrids. Am J Phys Anthropol 56:115–129.
Pickering TR, Clarke RJ, Heaton JL. 2004. The context of
Stw573 an early hominid skull and skeleton from Sterkfontein Member 2 taphonomy and paleoenvironment. J Hum
Evol 46:279–297.
Ramakrishnan U, Coss RG. 2001. A comparison of sleeping behaviour of three sympatric primates. Folia Primatol 72:51–63.
Reed KE. 1996. The paleoecology of Makapansgat and other Afican Plio-Pleistocene hominid localities. Ph.D. dissertation,
State University of New York at Stony Brook.
Reed KE. 1997. Early hominid evolution and ecological change
through the African Plio-Pleistocene. J Hum Evol 32:289–
322.
Reed KE. 1998. Using large mammal communities to examine
ecological and taxonomic organisation and predict vegetation
in extant and extinct assemblages. Paleobiology 32:384–408.
Reid DJ, Schwartz GT, Dean C, Chandrasekera. 1998. A histological reconstruction of dental development in the common
chimpanzee, Pan troglodytes. J Hum Evol 35:427–448.
Runia LT. 1987. Strontium and calcium distribution in plants:
effect on paleodietary studies. J Arch Sci 14:599–608.
135
Schoeninger MJ, Iwaniec UT, Glander KE. 1997. Stable isotope
ratios indicate diet and habitat use in new world monkeys.
Am J Phys Anthropol 103:69–83.
Schoeninger MJ, Iwaniec UT, Nash LT. 1998. Ecological attributes recorded in stable isotope ratios of arboreal prosimian
hair. Oecologia 113:222–230.
Schoeninger MJ, Moore J, Sept JM. 1999. Subsistence strategies
of two ‘‘savanna’’ chimpanzee populations: the stable isotope
evidence. Am J Phys Anthropol 49:297–314.
Sillen A. 1988. Elemental and isotopic analysis of mammalian
fauna from southern Africa and their implications for paleodietary research. Am J Phys Anthropol 76:49–60.
Sillen A. 1992. Strontium-calcium ratios (Sr/Ca) of Australopithecus robustus and associated fauna from Swartkrans.
J Hum Evol 23:495–516.
Sillen A, Hall G, Richardson S, Armstrong R. 1995. Strontium
calcium ratios (Sr/Ca) and isotopic ratios (87Sr/86Sr) of
Australopithecus robustus and Homo sp. from Swartkrans.
J Hum Evol 28:277–285.
Singleton M. 2002. Patterns of cranial shape variation in the
Papionini (Primates: Cercopithecinae). J Hum Evol 42:547–578.
Smith BN, Epstein S. 1971. Two categories of 13C/12C ratios for
plants. Plant Physiol 47:380–384.
Sponheimer M. 1999. Isotopic ecology of the Makapansgat Limeworks fauna. Ph.D. dissertation, Rutgers University.
Sponheimer M, de Ruiter D, Lee-Thorp J, Spath A. 2005a. Sr/
Ca and early hominin diets revisited: new data from modern
and fossil tooth enamel. J Hum Evol 48:147–156.
Sponheimer M, Lee-Thorp JA. 1999. Oxygen isotope in enamel carbonate and their ecological significance. J Arch Sci 26:723–728.
Sponheimer M, Lee-Thorp JA. 2001. The oxygen isotope composition of mammalian enamel carbonate from Morea Estate,
South Africa. Oecologia 126:153–157.
Sponheimer M, Lee-Thorp JA. 2006. Enamel diagenesis at
South African Australopith sites: implications for paleoecolgical reconstruction with trace-elements. Geochim Cosmochim
Acta 70:1644–1654.
Sponheimer M, Lee-Thorp J, de Ruiter D, Codron D, Codron J,
Baugh T, Thackeray F. 2005b. Hominins, sedges and termites:
new carbon isotope data from the Sterkfontin Valley and
Kruger National Park. J Hum Evol 48:301–312.
Sponheimer M, Reed K, Lee-Thorp JA. 1999. Combining isotopic
and ecomorphological data to refine bovid paleodietary reconstruction: a case study from Makapansgat Limeworks hominid locality. J Hum Evol 36:705–718.
Sponheimer M, Reed K, Lee-Thorp JA. 2001. Isotopic paleoecology of Makapansgat Limeworks Perissodactyla. S Afr J Sci
97:327–329.
Sternberg LSL. 1989. Oxygen and hydrogen isotope ratios in
plant cellulose: mechanisms and applications. In: Rundel PW,
Ehleringer JR, Nagy KA, editors. Stable isotopes in ecological
research. New York: Springer, p 24–141.
Thackeray JF, Myer S. 2004. Parapapio broomi and P. jonesi
from Strekfontein: males and females of one species? Ann
Transv Mus 41:79–82.
Tredget EE, Forsyth N, Uji-Friedland A, Chambers M, Ghahary
A, Scott PG, Hogg AM, Burke JF. 1993. Gas chromatography
mass spectrometry determination of oxygen-18 in oxygen-18
labelled 4-hydroxyproline for measurement of collagen synthesis and intracellular degradation. J Chromotogr B Biomed Sci
Appl 612:7–19.
Williams FL, Ackermann RR, Leigh SR. 2007. Inferring PlioPleistocene southern African biochronology from facial affinities in Parapapio and other fossil papionins. Am J Phys
Anthropol 132:163–174.
WoldeGabriel G, White TD, Suwa G, Renne P, de Heinzeli J,
Hart WK. 1994. Ecological and temporal placement of early
Pliocene hominids at Aramis, Ethiopia. Nature 371:330–333.
Zinner D, Peleaz F, Torkler F. 2001. Distribution and habitat
associations of baboons (Papio hamadryas) in Central Eritrea.
Int J Primatol 22:397–413.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Документ
Категория
Без категории
Просмотров
0
Размер файла
281 Кб
Теги
taxonomy, makapansgat, limeworks, pleistocene, plio, investigation, craniometric, separating, dietary, ecology, cercopithecoids, niche, biogeochemical
1/--страниц
Пожаловаться на содержимое документа