John G. Fleagle
Department of Anatomical
Sciences, Health Sciences
Center, State University of
New York, Stony Brook, New
York 11794-8081, U.S.A.
E-mail:
jfleagle@mail.som.sunysb.edu
W. Scott McGraw
Department of Anthropology,
The Ohio State University,
1680 University Dr.,
Mansfield, Ohio 44907,
U.S.A. E-mail:
mcgraw.43@osu.edu
Received 3 July 2000
Revision received 1 October
2001 and accepted
10 October 2001
Skeletal and dental morphology of African
papionins: unmasking a cryptic clade
One of the more perplexing problems in primate systematics concerns
the phyletic relationships of the large African monkeys—Mandrillus
(including drills), Papio, Lophocebus and Cercocebus. For over twenty
years, there has been molecular evidence that mangabeys are an
unnatural group and that the terrestrial forms—Cercocebus—are the
sister taxon of Mandrillus, while the arboreal forms—Lophocebus—are
more closely allied with Papio. Nevertheless, most systematists
have been reluctant to accept this scheme due to the lack of morphological evidence. In this paper, we undertake a detailed analysis of
the scapula, humerus, radius, ulna, pelvis, femur and dentition of
papionin primates. We identify a host of features shared by Cercocebus
and Mandrillus to the exclusion of Lophocebus and Papio. The polarity
of characters is established by examining an outgroup comprised of
several species of Macaca. The features shared by Cercocebus and
Mandrillus are functionally related to specific feeding and locomotor
behaviors that include aggressive manual foraging, the processing of
hard-object foods and the climbing of vertical trunks. We hypothesize
that the ability to subsist on hard seeds and nuts gleaned from the
forest floor is a key adaptation for the Cercocebus–Mandrillus clade.
2002 Elsevier Science Ltd.
Keywords: Cercocebus,
Lophocebus, Papio,
Mandrillus, phylogeny,
anatomy.
Journal of Human Evolution (2002) 42, 267–292
doi:10.1006/jhev.2001.0526
Available online at http://www.idealibrary.com on
Introduction
Since their initial discovery by western
scientists 200 years ago, the larger African
cercopithecine monkeys have been placed in
two seemingly natural groups—mangabeys
and baboons (including mandrills and
geladas). Indeed many authors have only
recognized two genera—Cercocebus for the
mangabeys and Papio for the baboons,
geladas and mandrills (Thorington &
Groves, 1970; Szalay & Delson, 1979).
However, most recent authors tend to recognize more taxonomic diversity within the
two groups. While all mangabeys are longlimbed monkeys with hollow cheeks and
long tails, mangabeys are clearly divided into
two species groups that show consistent differences in cranial anatomy (Groves, 1978)
and ecology (e.g., Chalmers, 1968; Jones &
Sabater Pi, 1968; Quiris, 1975; Waser,
0047–2484/02/030267+26$35.00/0
1977, 1984; Homewood, 1978; Horn,
1987; Mitani, 1989; Olupot et al., 1994,
1997; Kingdon, 1997). One group, the
galeritus–torquatus–atys–agilis group retained
in the genus Cercocebus, contains predominantly terrestrial monkeys that are reported
to live in large groups with large home
ranges (Jones & Sabater Pi, 1968; Quiris,
1975; Homewood, 1978; Mitani, 1989).
The other species group, the albigena–
aterrimus group, contains more slender,
strictly arboreal monkeys with smaller home
ranges (Chalmers, 1968; Waser, 1977,
1984; Horn, 1987; Olupot et al., 1994,
1997). These mangabeys are frequently
placed in a separate genus, Lophocebus (e.g.,
Groves, 1978; Rowe, 1996; Fleagle, 1999).
Likewise, the baboons are generally
divided into three different genera: Papio for
the savannah baboons found throughout
2002 Elsevier Science Ltd.
268
Cercocebus
. . .
Mandrillus
Lophocebus
Theropithecus
Papio
Figure 1. Phylogeny of African papionin monkeys as evinced by molecular data (Disotell, 1994; Harris &
Disotell, 1998; Harris, 2000).
sub-Saharan Africa (Jolly, 1967; Groves,
1993); Theropithecus for the geladas of the
Ethiopian plateau (Jolly, 1972; Jablonski,
1993), and Mandrillus for the forestdwelling mandrill and drill of western and
western Central Africa (Grubb, 1973).
In 1976, Cronin & Sarich presented
immunological data showing that Cercocebus
(=Lophocebus) albigena was more closely
related to baboons and geladas than to C.
galeritus and Mandrillus. Dutrilleaux et al.
(1982) and Stanyon et al. (1988) demonstrated that Mandrillus and Cercocebus spp.
uniquely shared a rearrangement of chromosome 10 to the exclusion of Lophocebus.
More recently, Disotell and colleagues
have demonstrated from both mitochondrial genes (Disotell, 1994) and nuclear
genes (Harris & Disotell, 1998; Harris,
2000) that C. galeritus and Mandrillus
form one clade, while L. albigena forms
a separate clade with baboons (Papio
and Theropithecus), although the detailed
relationships within this later clade are
unresolved (Figure 1).
Despite the consistent molecular evidence
for over 20 years supporting a diphyletic
origin of mandrills and baboons from separate groups of mangabeys, morphological evidence in support of such a phylogeny has,
until recently, been very limited. Thus,
while several workers (e.g., Groves, 1978;
Nakatsukasa, 1994a,b, 1996) have documented cranial and postcranial differences within mangabeys and others have
documented differences among mandrills,
baboons and geladas (e.g., Jolly, 1970;
Szalay & Delson, 1979; Jablonski, 1993;
Ciochon, 1994), there has been remarkably
little morphological evidence put forth linking Cercocebus uniquely with Mandrillus (but
see Hill, 1970; Disotell, 1994; Groves,
2000) and Lophocebus uniquely with Papio
and/or Theropithecus.
In the course of a comparative study of
locomotor behavior and skeletal anatomy
within mangabeys, we uncovered numerous
osteological similarities that link Cercocebus
and Mandrillus, and contrast with the
conditions found in Lophocebus and Papio,
thus supporting the molecular phylogeny
(Fleagle & McGraw, 1998). Moreover, a
review of the ecological literature demonstrates that Cercocebus (the drill mangabeys
of Kingdon, 1997) and mandrills share
many characteristics of their feeding ecology
and locomotor behavior that accord with the
dental and skeletal features uniting these
genera. In an earlier paper (Fleagle &
McGraw, 1999), we identified features of
the postcranial anatomy and dentition that
unite Cercocebus and Mandrillus and distinguish them from Lophocebus and Papio. In
the present paper, we expand our comparative sample in several ways: we include drills
and geladas as well as mangabeys, mandrills
and baboons; we compare females of the
two clades as well as males; and we include
data on macaques, the outgroup to the
African papionins, in order to distinguish
primitive and derived states. In addition, we
offer functional hypotheses for these morphological differences based on what is
known about the behavior and ecology of
mangabeys and mandrills. Finally, we
present an adaptive scenario to explain
the evolution of this clade of large African
cercopithecines in terms of a unique set
of ecological adaptations for exploiting
resources on the forest floor of African
tropical forests.
Materials and methods
This study is based on examination of
approximately 150 individual skeletons of
the genera Cercocebus, Lophocebus, Papio,
Mandrillus, Theropithecus and Macaca from
collections in the American Museum of
Natural History, The Natural History
269
Museum of London, Powell–Cotton
Museum (Birchington, U.K.), Museum of
Comparative Zoology (Harvard), Royal
Museum of Central Africa (Tervuren,
Belgium), The Randall L. Susman
Collection at SUNY—Stony Brook, and
specimens collected from the Ivory Coast’s
Tai Forest.
We have compared the bones and teeth
of these monkeys using descriptive comparisons, quantitative, continuous measurements, and nonmetric assessments of crest
development. The quantitative measurements on individual bones and teeth are
illustrated in Figure 2. Measurements were
taken with digital calipers.
These monkeys are very sexually dimorphic, and comparing mixed samples of
sexually dimorphic species can often
obscure species-specific differences (see
Nakatsukasa, 1994a,b, 1996). Therefore,
we report several different quantitative
comparisons of the dental and postcranial
skeleton. Initial comparisons are between
the larger mixed species samples of
males from the two clades (i.e., Cercocebus
and Mandrillus vs. Lophocebus, Papio, and
Theropithecus). Subsequently, we compare
all individuals of the two clades by adding
the measurements from the smaller (and
more variable) samples of females of each
clade. Finally, in order to estimate which of
the character states is likely to be primitive
for papionins as a group, we compare the
same measurements from a small sample
of mixed species and sexes of Macaca to
the mixed species and mixed sex samples
of Cercocebus–Mandrillus clade and the
Lophocebus, Papio, Theropithecus clade.
Univariate dimensions were converted
into ratios to permit comparisons of
different sized individuals and species. In all
cases, means of the individual specimens from each of the two clades were
compared using Student’s t-test for differences between the means of the pooled
samples.
270
. . .
(a)
(b)
(c)
(d)
(e)
(f)
Figure 2. Measurements of (a) scapula, (b) humerus, (c) ulna, (d) pelvis, (e) tibial shaft and (f) premolars.
Comparative skeletal and dental
anatomy
Osteological differences between species
of mangabeys and between baboons,
mangabeys, and geladas have been
documented by other studies (e.g., Jolly,
1970, 1972; Ciochon, 1994; Nakatsukasa,
1994a,b, 1996; see also Gebo & Sargis,
1994). In most cases, these studies have
identified features that distinguish terrestrial
and arboreal taxa. In many of these features, the arboreal Lophocebus differs from
the more terrestrial Cercocebus, as well as
from Mandrillus, Papio and Theropithecus
(Nakatsukasa, 1994a,b, 1996; Fleagle &
McGraw, 1999). In this study we do not
discuss those arboreal–terrestrial features;
rather our goal is to document features that
seem diagnostic of the Cercocebus–Mandrillus
clade and distinguish them from the
271
Figure 3. Scapulae of (top, left to right) Mandrillus and Papio and (bottom, left to right) Cercocebus and
Lophocebus. Note the taller suprascapular fossa and deeper inferior scapular angle in Mandrillus and
Cercocebus.
morphological features of Lophocebus, Papio
and Theropithecus. In addition, we attempt to
identify synapomorphies and symplesiomorphies by comparisons with an outgroup
composed of specimens of several species of
Macaca.
Upper extremity
The scapulae of Cercocebus and Mandrillus
differ strikingly from those of Lophocebus
and Papio in being relatively deep in a
superior–inferior dimension relative to their
dorso-ventral length (Figures 3 and 4;
Tables 1 and 2). This overall shape difference is also associated with a more pronounced development of the origin of teres
major from the inferior scapular angle and a
relatively taller suprascapular fossa. Using
our relatively gross measurement of the ratio
of maximum scapula depth divided by
maximum scapula length, Mandrillus and
Cercocebus are statistically indistinguishable.
However, Lophocebus is very different from
Papio and more similar to Cercocebus and
Mandrillus. In the overall metric comparisons between clades, the Cercocebus–
Mandrillus clade has a significantly higher
index in comparisons of males and comparisons of mixed sex samples. The Macaca
outgroup is significantly different from
both the Lophocebus, Papio, Theropithecus
clade and the Cercocebus–Mandrillus clade.
However, there are considerable differences
among the values for individual specimens
and for the different genera, precluding any
clear determination of the primitive condition for this index. This is a gross index that
combines many aspects of scapula morphology. We suspect that a more detailed
272
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
. . .
(13)
(7)
(4)
(3)
(7)
(3)
(6)
(1)
(7)
(8)
(1)
(5)
(18)
(13)
(14)
(3)
(2)
(2)
60
70
80
90
100
110
120
Figure 4. Plot of scapula height index. In this and all subsequent plots, sample sizes for each sex are given
in parentheses. For each taxon and sex the index mean1 S.D. is depicted. For samples consisting of
three individuals or less, we have plotted each individual value.
analysis of scapular shape in these taxa
would reveal more functional differences
among these taxa.
Similar differences in scapular shape
have been reported in several other primate groups, notably between Presbytis melalophos and Presbytis (Trachypithecus) obscura
(Fleagle, 1977), and also between Pithecia
pithecia and Chiropotes satanas (Fleagle &
Meldrum, 1988). In these primates, the
short and deep scapula is associated with
clinging and climbing behavior and the
longer scapula with more quadrupedal
behavior. The deep scapular shape increases
the lever arm of teres major which is a major
extensor of the humerus. Both Mandrillus
and Cercocebus have been reported to regularly engage in vertical climbing up tree
trunks as they ascend from the forest floor in
search of arboreal foods (as noted by a
reviewer, this is evident in the Nature television program ‘‘Mask of the Mandrill’’).
Similarly, the much lower value of this index
for Papio and Theropithecus almost certainly
reflects the less frequent use of vertical
climbing habits in these open-country
terrestrial genera.
The most striking feature of the proximal
part of the humeri of Cercocebus and
Mandrillus is the exceptional development of
the deltoid plane which is relatively broad
with prominent, projecting crests medially
(deltopectoral crest) and laterally (deltotriceps crest) (Figures 5 and 6; Table 1).
Lateral to the deltotriceps crest, both
Cercocebus and Mandrillus show a relatively
broad excavation for the upper fibers of
brachialis that is not seen in Lophocebus and
Papio.
An index of the width of the deltoid
plane relative to the width of the head
of the humerus shows significantly higher
values for the Cercocebus–Mandrillus clade
in comparisons of either males or the
mixed sex samples (Table 2, Figure 7).
The Macaca sample is not significantly
different from the Cercocebus–Mandrillus
clade in this index, suggesting that the
broad deltoid plane is primitive for
papionins.
Table 1 Skeletodental indices of papionin taxa
Deltoid plane
wdth100/
humeral
head wdth
Sup crest
hgt100/
humerus
length
Ant artic
wdth100/
biepicondylar
wdth
Coronoid
wdth100/
Art notch
wdth
Ilium min
wdth100/
Max acetabular
diam
Tibia mid
AP diam100/
Mid ML diam
100
(P4 m-db-l/
M1 m-db-l)
100
(p4 m-db-l/
m1 m-db-l)
92·5, 6·3 (13)
98·1, 2·9 (7)
97·5, 5·6 (4)
87·5, 7·2(13)
78·4, 7·7 (7)
92·1, 6·7 (4)
37·5, 2·6 (12)
36·2, 1·6 (7)
37·3, 1·9 (4)
68·3, 2·6 (13)
76·1, 2·6 (7)
76·4, 3·2 (4)
55·3, 6·5 (12)
54·2, 4·3 (7)
51·3, 1·3 (4)
112·3, 7·1 (13)
106,
5·2 (7)
119·9, 4
(4)
129·6, 7·2(13)
131·3, 4·7 (7)
137·7, 10 (4)
86·6
81·7
34·8
(3)
67·8, 1·02 (4)
59·6, 0·77 (4)
115·7, 5·6
(5)
125·8, 14·4 (4)
35·01, 2·7 (7)
31·5
(3)
35·7, 4·1 (6)
20·9
(1)
36·3, 2·2 (7)
32·9, 2·6 (8)
35·4
(1)
32·9, 2·9 (5)
31·7, 2·4 (19)
30·6, 2 (13)
30·7, 2 (11)
32·9
(3)
23·2
(2)
26·8
(2)
71·2, 9·5 (7)
67·9
(3)
70·8, 7·1 (6)
72·8
(1)
74·9, 5·9 (7)
74,
3·2 (8)
80·2
(1)
74·5, 2·8 (5)
80·7, 10·7 (19)
79·2, 6·2 (13)
75·3, 4·3 (6)
59·8
(3)
77·3
(2)
74·2
(2)
51·6, 8·6 (8)
71·6
(3)
53·8, 5·7 (6)
60·3
(1)
46·7, 6·1 (7)
44·9, 4·6 (8)
54·7
(1)
54·8, 4·7 (5)
67·7, 6·7 (14)
63·4, 5·5 (12)
68·1, 4·2 (10)
72·4
(3)
65·3
(2)
67·8
(2)
124·1, 12·9 (7)
106·9
(3)
113·5, 5·5 (6)
100·9
(1)
114·3, 7·7 (7)
110·8, 7·1 (8)
103·7
(1)
104·9, 7·2 (5)
98·8, 6·7 (18)
95·3, 10·2 (12)
103·2, 5·3 (12)
118·4
(3)
100·4
(2)
104·1
(2)
124,
8·7 (8)
134·8
(3)
131·8, 9·1 (6)
154·7
(1)
129·7, 6·9 (7)
142·4, 19·7 (8)
128·6
(1)
126·9, 4·7 (5)
145·9, 8·1(15)
139·2, 9·9(11)
142·7, 5·9(11)
139·8
(3)
141·3
(2)
142·2
(2)
77·9, 128 (5)
77·4
(2)
86·7, 5·9 (8)
82·7, 7·8 (5)
78·7, 12
(8)
75,
4·6 (7)
84·2, 16·6 (10)
84·8, 3·3 (5)
81·5
(3)
83
(1)
81·2, 11·7 (4)
82·3
(2)
88·9, 12·8 (8)
86·04, 7·4 (5)
84·3, 7·9 (10)
79·1, 10
(4)
94·1, 15·1 (11)
88·4, 7·2 (5)
80·6
(3)
80·7
(1)
78·6
80·7
58·6,
57·2,
56,
56·7,
59,
54
82·4
(3)
86·6
(3)
58·9, 5·5 (15)
55·6, 7·9 (7)
60·1, 5·5 (14)
61·9, 11·5 (7)
57,
7·1 (5)
58·3
(2)
(3)
95·2, 4·7 (7)
86·1
(3)
87·9, 7·7 (6)
111·2
(1)
88·4, 4·9 (7)
89·4, 4·9 (8)
81·9
(1)
88·5, 2·4 (5)
88·4, 5·1 (18)
88·6, 5 (13)
74·2, 4·4 (14)
69
(3)
65·8
(2)
74·7
(2)
(3)
84·8, 10·1 (7)
70·3
(3)
82·8, 7·9 (6)
88·5
(1)
86·3, 3·6 (7)
77·9, 8·5 (8)
75·6
(1)
80·7, 4·3 (5)
74·8, 6·5(19)
73·9, 6·6(14)
69·4, 7·4(11)
75·8
(3)
63·4
(2)
70·5
(2)
5·4
4·8
5·5
5·8
7·6
(3)
(3)
(15)
(7)
(14)
(7)
(5)
(2)
Cercocebus torquatus (males)
Cercocebus torquatus (females)
Cercocebus agilis (males)
Cercocebus agilis (females)
Cercocebus atys (males)
Cercocebus atys (females)
Mandrillus sphinx (males)
Mandrillus sphinx (females)
Mandrillus leucophaeus (males)
Mandrillus leucophaeus (females)
Macaca fascicularis (males)
Macaca fascicularis (females)
Macaca nemestrina (males)
Macaca nemestrina (females)
Lophocebus albigena (males)
Lophocebus albigena (females)
Papio hamadryas (males)
Papio hamadryas (females)
Theropithecus gelada (males)
Theropithecus gelada (females)
Scap max
hgt100/
Max length
N.B. For each taxon we supply means, standard deviations and sample sizes. For samples less than three we provide only means and sample sizes. Abbreviations: ant=anterior,
artic=articular, diam=diameter, hgt=height, max=maximum, mid=midpoint, min=minimum, scap=scapula, sup=supinator, wdth=width.
273
274
Table 2 Significance values for comparisons of papionin taxa
Cercocebus–Mandrillus vs.
Lophocebus–Papio–Theropithecus
(males only)
<<0·001
<<0·001
0·009
<<0·001
<<0·001
0·384
<<0·001
<<0·001
<<0·001
0·078
<<0·001
<0·001
<0·001
0·005
<<0·001
<<0·001
<0·001
<<0·001
<<0·001
0·107
<<0·001
<<0·001
<<0·001
<<0·001
<<0·001
<<0·001
0·295
0·41
0·61
Macaca vs.
Cercocebus–Mandrillus
Macaca vs. Lophocebus–
Papio–Theropithecus
0·01
0·23
<<0·001
0·0004
0·002
<<0·001
0·001
. . .
Scapula max. height100/max.
length
Deltoid plane width100/humeral
head width
Sup. crest height100/humerus
length
Ant. articular width100/biepiconylar
width
Coronoid width100/articular notch
width
Ilium min. width100/max. acetabular
diameter
Tibia mid AP diameter100/mid ML
diameter
100(p4 m-db-l/m1 m-db-l)
100(P4 m-db-l/M1 m-db-l)
Cercocebus–Mandrillus vs.
Lophocebus–Papio–Theropithecus
(both sexes)
275
Figure 5. Humerii of (left to right) Mandrillus, Cercocebus, Lophocebus and Papio. Note the expanded
deltoid plane and more proximally extending supinator crest in Mandrillus and Cercocebus.
The distal part of the humeral shaft of
Cercocebus and Mandrillus differs from that
of Lophocebus and Papio in having a broader
brachialis flange and a more proximally
extending supinator crest (Figures 5 and 6,
8; Tables 1 and 2). The differences between
the Cercocebus–Mandrillus clade and the
Lophocebus, Papio, Theropithecus clade are
significant in both the all male and the
mixed sex comparisons. The Macaca sample
is not significantly different from the
Cercocebus–Mandrillus clade in this index
suggesting that a proximally extending
supinator crest is primitive for papionins.
Compared with Lophocebus and Papio,
Cercocebus and Mandrillus have a relatively
narrow distal articulation of the humerus
(Figures 5 and 6). Indices of anterior
276
. . .
Figure 6. Posterior view of (left to right) Mandrillus, Cercocebus, Lophocebus and Papio humerii. Note the
lateral projection of the delto-triceps crest and the narrower olecranon in Mandrillus and Cercocebus.
articular width divided by either the
biepicondyler width or humerus length
show significant differences between the
Cercocebus–Mandrillus
clade
and
the
Lophocebus, Papio, Theropithecus clade for
both the male only sample and the mixed
sex sample (Figure 9; Tables 1 and 2).
However, there is considerable variation
among individuals and species within each
clade. The Macaca sample is not significantly different from the Lophocebus, Papio,
Theropithecus clade in these indices, suggesting that the Cercocebus–Mandrillus condition
is derived.
On the humerus of Cercocebus and
Mandrillus there is a relatively more
277
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(13)
(7)
(4)
(3)
(7)
(3)
(6)
(1)
(7)
(8)
(1)
(5)
(19)
(14)
(11)
(3)
(2)
(2)
55
65
75
85
95
105
35
40
45
Figure 7. Plot of deltoid plane index.
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(12)
(7)
(4)
(3)
(7)
(3)
(6)
(7)
(8)
(1)
(5)
(19)
(13)
(11)
(3)
(2)
(2)
20
25
30
Figure 8. Plot of relative height of supinator crest.
prominent medial lip to the trochlea, and a
relatively narrower olecranon fossa with a
sharp, and deep lateral margin (Figures 5
and 6; Table 1). In an index of depth of the
medial trochlear lip divided by the width of
the olecranon fossa, the Cercocebus–
Mandrillus clade is significantly different
from the Lophocebus, Papio, Theropithecus
clade in both the male only comparisons and
mixed sex comparisons (Table 2). The
Macaca sample is significantly different from
the Cercocebus–Mandrillus clade, but not
from the Lophocebus, Papio, Theropithecus
clade, suggesting that the Cercocebus–
Mandrillus condition for this index is derived
for papionins.
278
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. assamensis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
. . .
(13)
(7)
(4)
(4)
(7)
(3)
(6)
(1)
(7)
(8)
(2)
(1)
(5)
(19)
(13)
(6)
(3)
(2)
(2)
50
60
70
80
90
100
Figure 9. Plot of relative anterior articular width of humerus.
The distinctive features of the proximal
part of the ulna of Cercocebus and Mandrillus
reflect those of the distal humerus. There is
a narrower coronoid process with a prominent medially facing surface for articulation
with the prominent trochlear lip compared
to the relatively broader coronoid process in
Lophocebus and Papio (Figure 10; Table 1).
In an index of coronoid width divided by
width of the radial notch, the values for the
Cercocebus–Mandrillus clade are significantly
lower than those for the Lophocebus, Papio,
Theropithecus clade in both the males only
comparisons and the mixed sex comparisons
(Table 2; Figure 11). The Macaca sample is
significantly different from either African
clade, but overlaps extensively with the
Cercocebus–Mandrillus clade suggesting a
narrow coronoid is primitive for papionins.
In Cercocebus and Mandrillus the interosseus line is prominent and flares ventrally as
far as the ventral surface of the ulnar shaft,
whereas in Lophocebus and Papio the interosseus line is less distinct and confined to the
lateral margin of the shaft (Figure 12; Table
3). Although not as distinctive as the ulnar
features, in Cercocebus and Mandrillus, the
interosseus (medial) border and the posterior border of the radius are more prominently developed than in Lophocebus and
Papio, and these sharp borders give the
radius a triangular rather than a rounded
cross-sectional shape (Figure 12; Table 3).
The forelimb bones of Cercocebus and
Mandrillus show a scapula with features
found in other anthropoid species that cling
and climb on vertical supports, a humerus
with a large brachialis muscle suggesting
adaptations for powerful elbow flexion, and
a narrow, stable elbow region. The forearm
shows pronounced crest development that
suggests prominent wrist or digital flexor
musculature. This osteological evidence
accords well with suggestions by Jolly
(1970) that mandrills seem to have much
larger forelimb flexors than other papionins.
Jolly (1967:437) further noted that mandrills and drills ‘‘are usually said to be forestfloor animals, but both show anatomical
signs of considerable greater adaptation to
tree climbing than in typical baboons
(Papio).’’
279
Figure 10. Proximal ulnae of (left to right) Mandrillus, Cercocebus, Lophocebus and Papio. Note the narower
coronoid in Mandrillus and Cercocebus.
In the absence of more detailed information about the musculature of these monkeys, we can nevertheless make some
hypotheses about the adaptive significance
of the unusual features of the forelimb of
Cercocebus and Mandrillus, based on their
naturalistic behavior that would seem to
require powerful forelimb flexion. Both are
terrestrial foragers that spend much of their
daily activity rummaging through leaf litter
on the forest floor in search of fallen fruits,
nuts and seeds, as well as animal material.
They are described as aggressive manual
foragers that rip apart rotten logs, open large
fruits, and harvest terrestrial herbaceous
vegetation (Hoshino, 1985; Harrison, 1988;
Schaaf et al., 1990; Bergmueller, 1998).
While this manual foraging may account for
the digital, carpal, and elbow flexion they
seem insufficient to account for the scapula
features. Both of these monkeys also climb
vertical tree trunks in search of food. It
seems likely that this activity would require
powerful humeral retraction as well as
elbow, carpal, and digital flexion.
Lower extremity
The pelvis of Cercocebus and Mandrillus differs from that of Lophocebus and Papio in
having an ilium that is relatively much
broader at its base (Figures 13 and 14;
Tables 1 and 2). The mean values of basal
ilium breadth relative to acetabulum
breadth are significantly higher for the
Cercocebus–Mandrillus clade than for the
Lophocebus, Papio, Theropithecus clade for
both the male only comparisons and the
mixed sex comparisons. However, there
is considerable spread in the values for
280
. . .
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(12)
(7)
(4)
(4)
(8)
(3)
(6)
(1)
(7)
(8)
(1)
(5)
(14)
(12)
(10)
(3)
(2)
(2)
30
40
50
60
70
80
90
Figure 11. Plot of index of coronoid width divided by radial notch width.
individuals and genera. The values for
Macaca are not significantly different from
those of the Cercocebus–Mandrillus clade,
suggesting that a relatively broad basal ilium
is primitive for papionins.
The femur of Cercocebus and Mandrillus is
generally characterized by a more prominent
gluteal tuberosity than in Lophocebus or
Papio (Table 3). Distally, the femur of
Cercocebus and Mandrillus has a patellar
groove in which the medial lip is prominent
and more similar in form and size to the
lateral tip, whereas in Lophocebus and Papio
the lateral lip is more prominent (Figure 15;
Table 2). Among other primates, climbing
species such as Pan troglodytes are characterized by a patellar groove with equal medial
and lateral margins or prominent medial
margins (Ward et al., 1995).
The tibial shaft of Cercocebus and
Mandrillus is relatively rounder than that of
Lophocebus and Papio in which the tibial shaft
is more elongate antero-posteriorly (Tables 1
and 2; Figure 16). The mean values of relative tibial compression are significantly lower
for the Cercocebus–Mandrillus clade than for
the Lophocebus, Papio, Theropithecus clade in
comparisons for both males only and mixed
sex samples. The values for Macaca are not
significantly different from those of the
Cercocebus–Mandrillus clade, suggesting that
a more rounded tibia shaft is primitive for
papionins.
The features of the hindlimb that distinguish Cercocebus and Mandrillus are less
extensive than those of the forelimb and less
amenable to interpretation by comparison
with other primates. Of the behaviors that
seem characteristic of the monkeys, it seems
most likely that vertical climbing up tree
trunks could select for a robust ilium, a
relatively more prominent medial patellar
margin and a more rounded tibial shaft.
Premolar morphology
Compared with virtually all other Old World
monkeys, Cercocebus and Mandrillus are
unusual in having relatively large upper and
lower posterior premolars that approach the
first molar in size (Figures 17–19; Table 1).
The ratio of posterior premolar area to first
281
Figure 12. Top: ulnae of (top to bottom) Mandrillus, Cercocebus, Lophocebus and Papio. Bottom: radii of
(top to bottom) Mandrillus, Cercocebus, Lophocebus and Papio. Note the more prominent interosseous
crests in the Mandrillus and Cercocebus ulnae and radii.
282
. . .
Table 3 Distribution of nonmetric characters among papionins
Interosseus crest (ulna)
Radial shaft shape
Gluteal tuberosity
Patellar groove: medial lip
Patellar groove: lateral lip
Cercocebus
Mandrillus
Papio
Lophocebus
Strong
Triangular
Prominent
Weak/round
Strong/sharp
Strong
Triangular
Prominent
Weak/round
Moderate
Weak
Rounded
Weak
Moderate
Weak
Weak
Rounded
Weak
Moderate
Moderate
Figure 13. Bony pelvis of (left to right) Mandrillus, Cercocebus, Lophocebus and Papio. The ilium of
Mandrillus and Cercocebus is much broader at its base (arrows) than those of Lophocebus and Papio.
molar area (both uppers and lowers) is
significantly higher for the Cercocebus–
Mandrillus clade than for the Lophocebus,
Papio, Theropithecus clade in comparisons
of either male only or mixed sex
samples (Table 2). Our relatively small
sample of Macaca (mostly M. nemestrina) is
not significantly different from the
Cercocebus–Mandrillus clade in this index,
suggesting that large premolars are primitive
for papionins.
Both Cercocebus and Mandrillus rely extensively on hard nuts and seeds that they
collect on the forest floor. Hoshino (1985;
see also Rogers et al., 1996) found that
these were the main food items of mandrills
during the period of fruit scarcity, and
Cercocebus in West Africa rely extensively on
283
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(13)
(7)
(4)
(5)
(1)
(7)
(3)
(6)
(1)
(7)
(8)
(1)
(5)
(18)
(12)
(12)
(3)
(2)
(2)
80
90
100
110
120
130
140
Figure 14. Plot of relative basal ilium width.
hard nuts that other primates cannot open
(Bergmueller, 1998; Range, 1998; Rutte,
1998). It seems most likely that the large
premolars in these species are adaptations
for cracking open hard nuts and seeds. In
addition, both Cercocebus and Mandrillus
have been reported to eat the pith and
bark of various grasses and they may also
use their premolars for stripping these
foods (Harrison, 1988). Lophocebus are also
reported to eat hard nuts such as palm
nuts, but they clearly lack the enlarged
premolars of Cercocebus. Unfortunately,
there are too few behavioral observations of
the details of food processing in these
species.
The adaptive niche of Cercocebus and
Mandrillus
In the sections above we have documented
morphological features that distinguish
Cercocebus and Mandrillus from Lophocebus
and Papio and support the molecular studies
that link these genera as sister taxa. The
former two genera are among the most
poorly known of all African primates
because they move rapidly on the ground
through dense vegetation and have large
ranges. Nevertheless, on the basis of studies
over many years, it is clear that these
monkeys share many unique behavioral and
ecological features that unite them and distinguish them from Lophocebus, Papio, and
Theropithecus.
Cercocebus has a broad distribution over
much of central Africa, while Mandrillus is
restricted to areas of Western and western
central Africa (Gartlan, 1970; Grubb, 1973;
Kingdon, 1997). Both Cercocebus and
Mandrillus are large, predominantly terrestrial forest-dwelling cercopithecines that
spend much of their daily activity foraging
through leaf litter on the forest floor. In
describing the foraging behavior of a horde
of mandrills in Gabon, Rogers et al.
(1996:304–305) noted, ‘‘When following
mandrill trails through the forest, it was
obvious that the leaf litter had been extensively disturbed, termite nests broken open
and rotten wood explored.’’ Likewise, in
describing Cercocebus agilis in Gabon, Quiris
284
. . .
Figure 15. Femora of (left to right) Mandrillus, Cercocebus, Lophocebus and Papio. Note the more equal
medial and lateral lips of the patellar groove in Mandrillus and Cercocebus.
(1975:366) observed, ‘‘Pour terminer,
notons que si C. galeritus (agilis) a ete
observe frequemment au sol, fouillant la
vase ou la litiere, il n’a jamais ete possible de
couvrir ce qu’il cherchait . . .’’. Similarly,
Cercocebus atys in the Tai forest, Ivory Coast
spends most of its foraging time sorting
through leaf litter and fallen trees on
the forest floor (McGraw, 1996; personal
observation).
Although Cercocebus and Mandrillus are
primarily terrestrial foragers, they also
climb trees to forage on ripe fruits. When
they ascend trees Cercocebus often do so by
vertical climbing up large trunks (McGraw,
1996; see also Quiris, 1975). Similar
285
Cercocebus torquatus
C. torquatus
C. agilis
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
Macaca fascicularis
M. fascicularis
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(13)
(7)
(4)
(4)
(8)
(3)
(6)
(1)
(7)
(8)
(1)
(5)
(15)
(11)
(11)
(3)
(2)
(2)
100
110
120
130
140
150
160
170
Figure 16. Plot of relative compression of the tibial shaft.
Figure 17. Maxillary dentition of (top, left to right) Cercocebus and Mandrillus and (bottom, left to right)
Lophocebus and Papio. Note the enlarged posterior premolars of Cercocebus and Mandrillus.
climbing behavior has been described for
mandrills (e.g., Hoshino, 1985; Harrison,
1988; Rogers et al., 1996).
Both Cercocebus and Mandrillus are most
commonly described as frugivores that feed
on fallen fruit and various types of animal
material on the forest floor. However, few
studies have made an effort to distinguish
nuts and seeds from fruits. Those that make
a distinction emphasize the importance of
seeds and nuts retrieved from the forest floor
in the diet of both these genera, whether diet
was determined by direct observation or
by analysis of feces. Hoshino (1985:265)
explicitly noted, ‘‘mandrills tended to consume fallen seeds after their fruiting period.
Thus mandrills can be called ‘seed eaters’,
while arboreal Cercopithecus monkeys and
chimpanzees can be called ‘pulp eaters’.’’
Likewise, Cercocebus is well-known for its
286
Cercocebus torquatus
C. torquatus
C. agilis
C. agilis
C. atys
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
. . .
(5)
(2)
(8)
(5)
(8)
(7)
(12)
(5)
(3)
(1)
(3)
(3)
(8)
(7)
(14)
(7)
(5)
(2)
40
50
60
70
80
90
100
110
105
115
Figure 18. Plot of relative upper premolar size.
Cercocebus torquatus
C. torquatus
C. agilis
C. agilis
C. atys
C. atys
Mandrillus sphinx
M. sphinx
M. leucophaeus
M. leucophaeus
M. nemestrina
M. nemestrina
Lophocebus albigena
L. albigena
Papio hamadryas
P. hamadryas
Theropithecus gelada
T. gelada
(4)
(2)
(8)
(5)
(10)
(4)
(11)
(5)
(3)
(1)
(3)
(3)
(8)
(7)
(14)
(7)
(5)
(2)
45
55
65
75
85
95
Figure 19. Plot of relative lower premolar size.
diet of seeds and hard nuts (e.g., Kingdon,
1997). Indeed, field workers have noted that
Cercocebus is most readily located in the
forest by listening for the sound of them
cracking nuts with their teeth (McGraw,
1996). These hard nuts and seeds seem
particularly important for these monkeys in
the dry season, when fruits are less common.
As Hoshino (1985:257) observed, ‘‘the
period of eating specific seeds was longer
than their fruiting period, because seeds are
more resistant to decomposition than pulp
and, thus, edible for a longer period. In
other words, mandrills could feed on fruit in
the form of fallen seeds, even when the
availability of fresh fruit (pulp) is poor.’’
Thus, it is the ability to eat hard seeds and
fruits that seems to permit these large monkeys to survive in times of low abundance of
ripe fruits. In addition, both Cercocebus and
Mandrillus have been reported to eat terrestrial herbaceous vegetation and various
grasses, especially during periods of low
fruit abundance. Indeed, there is considerable dietary overlap between Cercocebus and
Mandrillus in Cameroon (Caldecott et al.,
1996).
Although fallen seeds and nuts are a critical and relatively long lasting resource that
Cercocebus and Mandrillus can exploit better
than other species because of their dental
and locomotor abilities, they nevertheless seem to be sparsely distributed. Thus,
Cercocebus and Mandrillus are unusual in that
both regularly form supertroops or hordes of
large numbers of individuals from several
troops that join together and range very
widely during periods of low fruit availability
(Gartlan, 1970; Hoshino, 1985). Further
studies of the ecology of these species will
undoubtedly reveal additional behavioral
similarities (Shah, 1996).
Overall, there is abundant evidence that
Cercocebus and Mandrillus are a limited and
specialized clade of large cercopithecines
characterized by shared, unique adaptations
for terrestrial foraging on the forest floor for
fallen fruits, animal material and especially
hard seeds and nuts. It is this latter specialty
combined with their flexible social organization that is critical to their survival during
periods of low fruit availability.
The phylogeny of the
Cercocebus–Mandrillus adaptation
For each of the morphological features
that distinguishes the Cercocebus–Mandrillus
287
clade from the Lophocebus, Papio, Theropithecus clade, we evaluated the condition
in a mixed species, mixed sex sample of
macaques (M. fascicularis and M. nemestrina) in order to determine which morphologies are likely to be primitive and
which are derived among papionins. It
turns out that for most of the skeletal features (broad deltoid plane, high supinator
crest, broad ilium base, and less mediolaterally compressed tibial shaft), the
Macaca mean is not significantly different
from that of the Cercocebus–Mandrillus clade.
For another feature (coronoid width divided
by radial notch width) the means are statistically different, but there is considerable
overlap and the macaques are clearly more
similar to the Cercocebus–Mandrillus sample.
Likewise, the sample of Macaca nemestrina is
most similar to the Cercocebus–Mandrillus
sample in having relatively large premolars. In only one feature (relative articular
width of the humerus) is the Lophocebus,
Papio, Theropithecus clade more similar to
the macaque morphology. In scapular
shape, each of the three groups is distinctive.
This suggests that, overall, the morphological features that distinguish the Cercocebus–
Mandrillus clade are mostly primitive for
papionins, and it is the Lophocebus, Papio,
Theropithecus clade that is derived.
This simple phylogenetic analysis accords
well with observations of Caldecott et al.
(1996), who have noted extensive similarities in the ranging behavior and foraging
adaptations of pigtail macaques (M. nemestrina) with those of mandrills and drills, and
with Fooden’s (1980) biogeographic arguments that the pigtail macaques are part of
the initial macaque radiation into Asia. This
would suggest that the wide-ranging, forest
floor gleaning adaptation of Cercocebus,
mandrills, and pigtail macaques is the basal
adaptation of papionins, and that the more
arboreal (Lophocebus) and open-country taxa
(Papio and Theropithecus) show derived
adaptations.
288
. . .
Discussion
In an earlier paper we described morphological features of the dentition and skeleton
of males that characterize Cercocebus and
Mandrillus as a clade and distinguish them
from the other clade of African papionins
composed of Lophocebus, Papio and Theropithecus (Fleagle & McGraw, 1999). In this
paper we have provided more extensive
illustrations and broader analyses of these
morphological features by including females
and additional species in both clades. In
addition, we have compared the African
papionins with a small sample of macaques
in order to determine whether the
Cercocebus–Mandrillus features were primitive or derived for papionins. There are
numerous aspects of our analysis and
results that merit further discussion. These
include the analysis of female skeletons, the
diverse morphology within the Lophocebus,
Papio, Theropithecus clade, and the value
of primitive characters in identifying
adaptations.
In our morphological analyses described
above, the results from comparisons of the
larger, mixed sex samples always accorded
with the results of the all male comparisons
(see also Fleagle & McGraw, 1999), and
with the phylogeny of African papionins
based on biomolecular and genetic studies
(Cronin & Sarich, 1976; Hewett-Emett &
Cook, 1978; Disotell, 1994, 1996; Harris &
Disotell, 1998). However, it is evident from
the plots of the samples by sex and species
that the significant differences were largely
driven by the male sample. The cladespecific, and even species-specific patterns
were far less obvious among the female
specimens than among the males. Similar
results have been reported by others, including Nakatsukasa (1994a,b, 1996). We cannot explain why females do not show as
strong a pattern of skeletal differences as
males; however, there are several possible
reasons.
One possibility is that the differences are
the result of methodological problems in
either sampling or measuring. For the taxa
we measured, complete adult female skeletons are much less common than adult
male skeletons in the museum collections we
visited. As a result, female samples are more
likely to be a random sample from a variety
of populations rather than a large sample
from one population. Thus, the apparent
diversity in some of the female samples and
the differences from the males may reflect
geographic diversity. In addition, a single
misidentified specimen would have a much
greater effect on the characteristics of a
small sample.
It is also quite possible that the measurements of female specimens are less accurate
than those of males. Compared with male
skeletons, female skeletons generally have
less distinctive muscle markings. As a result
measurements that depend on identifying a
ridge caused by a muscle attachment may
well be more difficult to identify in female
skeletons. However, there is no evidence
from our results that this is the case. The
measurements based on muscle attachments
(e.g., supinator crest height) are not obviously more variable in the female skeletons,
except in Theropithecus. More generally,
however, it is well established that the metabolic and functional demands on the skeleton are often different for males and
females of dimorphic species so that
many biomechanical features show different
scaling relationships.
Even though we have been able to demonstrate significant differences between the
two clades of African papionins in several
skeletal and dental features, there is also
considerable variation within each of the
clades. This is particularly evident for the
Lophocebus, Papio, Theropithecus clade,
which includes taxa with very different patterns of substrate use. Lophocebus is a totally
arboreal rainforest monkey; Theropithecus is
an almost totally terrestrial monkey from the
289
high plateau of Ethiopia; and the widespread Papio is probably best characterized
as predominantly terrestrial, but individuals
of many populations regularly forage in
trees. In view of the considerable differences
in the ecology and locomotion of extant
members of this clade, reconstruction of the
evolutionary history of adaptations in this
clade is a debated and, in our view, largely
unresolved issue with many alternative
scenarios and difficult questions that need to
be addressed in the future.
For example, if the predominantly terrestrial locomotor habits of M. nemestrina,
Cercocebus, and Mandrillus are primitive for
African papionins, under what conditions
did the characteristic postcranial features of
Lophocebus and Papio evolve? Is Lophocebus
secondarily arboreal? This is a scenario
favored by Nakatsukasa (1996) in his study
of skeletal differences among mangabeys.
How does Theropithecus fit into the picture?
Delson & Dean (1993) have argued from
cranial evidence for similarities between
Theropithecus and Mandrillus, implying that
these taxa share features that are primitive
for African papionins. As with our comparisons of African papionins and macaques, it
seems almost certain that any scenario will
involve some type of mosaic of primitive and
derived features in all lineages. Moreover, as
the study by Delson & Dean (1993) emphasizes, attempts to answer these questions will
require a broader consideration of the fossil
record of this group, including Parapapio,
fossil baboons, and fossil Theropithecus as
well as more detailed functional studies of
the biomechanical significance of the
morphological features.
In this study we have attempted to identify the functional significance of the features
shared by Cercocebus and Mandrillus without
distinguishing between those features that
are primitive retentions from a common
papionin ancestry and those that are derived
for that clade. Our comparisons with a small
sample of Macaca, the outgroup to African
papionins, suggest that many of the features
shared by Cercocebus and Mandrillus are
probably primitive for all papionins. These
results need to be corroborated by
additional outgroup comparisons with a
larger sample of macaques, with fossil
macaques such as Procynocephalus, and with
a broader sample of other extant and fossil
cercopithecines.
While an appreciation of the phylogeny of
individual morphological and behavioral
features is valuable for enabling us to reconstruct the evolutionary history of papionins,
the phylogenetic status of individual morphological features as either primitive or
derived is not critical for our ability to
interpret their functional significance or
their status as adaptations. There is ample
evidence that both primitive and derived
features have mechanical and physiological
functions that are amenable to interpretation through comparative methods (e.g.,
Anthony & Kay, 1993; Ross et al., 2001).
Thus, as discussed above, it seems most
likely that the features shared by Cercocebus,
Mandrillus and Macaca nemestrina are part of
a longstanding, primitive papionin foraging
adaptation related to forest floor locomotion
and gleaning.
Summary
Although biomolecular studies of primate
systematics over the past 20 years have consistently demonstrated that mandrills are the
sister group of Cercocebus mangabeys rather
than being closely related to baboons and
geladas, there has been little morphological
or behavior evidence to support this
phylogeny.
Detailed comparisons of the skeletal
anatomy of Cercocebus, Mandrillus, Lophocebus and Papio reveal osteological features
of the scapula, humerus, ulna, radius,
pelvis, femur and tibia that unite Cercocebus
and Mandrillus, and distinguish them
from Lophocebus and Papio. In addition,
290
. . .
Cercocebus and Mandrillus have greatly
enlarged posterior premolars. The enlarged
premolars are an adaptation for cracking
open hard nuts that both of these monkeys
retrieve from the forest floor litter. The
forelimb features are best interpreted as
adaptations for extensive and forceful
manual foraging in forest floor litter as well
as vertical climbing of tree trunks. The hindlimb features are hypothesized as adaptations for vertical climbing. Comparisons
with an outgroup comprised of Macaca
species reveals that many features shared by
Cercocebus and Mandrillus including a broad
deltoid plane, proximally extending supinator crest, narrow coronoid, broad basal
ilium, less mediolaterally compressed tibial
shaft and expanded premolars, are primitive
for papionins.
Identification of these shared morphological features and a review of the literature show that mandrills and Cercocebus
mangabeys are similar in many aspects of
their behavior and ecology. Both are primarily wide-ranging terrestrial gleaners on
the forest floor where they manually search
through leaf litter and rotten logs in search
of arthropods, fallen fruit, nuts, and seeds.
They also ascend the trunk of trees in search
of arboreal food items.
Acknowledgements
We thank the following institutions and
individuals for allowing access to specimens
in their care: American Museum of Natural
History (Ross MacPhee), The Natural
History Museum of London (Paula
Jenkins), Field Museum of Natural History
(Bill Stanley), UC-Berkeley Laboratory
of Human Evolution (Clark Howell),
Museum of Comparative Zoology, Harvard
University (Maria Rutzmoser), PowellCotton Museum (John Harrison), Royal
Museum of Central Africa, Tervuren
(Wim Van Neer), SUNY-Stony Brook
(Randall Susman). We thank Ronald Noe,
the assistants and students of the Tai Forest
Monkey project for enabling a better
understanding of mangabey biology. Bill
Jungers provided valuable statistical advice
and Luci Betti-Nash skillfully produced the
figures. We thank the various ministries for
permission to carry out fieldwork in the
Ivory Coast; the Centre Suisse de Recherche
Scientific and its director, Dr Oliver
Girardin, provided logistical assistance in
Africa. This work was generously supported by a grant from the LSB Leakey
Foundation.
References
Anthony, M. R. L. & Kay, R. F. (1993). Tooth form
and diet in ateline and alouattine primates: Reflections
on the comparative method. Am. J. Science 293A,
356–382.
Bergmueller, R. (1998). Nahrungsoekologie und
Konkurrenz um Nahrungressourcen bei der
Raunchgrauen Mangabe (Cercocebus atys). Diploma,
University of Regensburg, Germany.
Caldecott, J. O., Feistner, A. T. C. & Gadsby, E. L.
(1996). A comparison of ecological strategies of
pig-tailed macaques, mandrills and drills. In (J. E. Fa
& D. G. Lindburg, Eds) Evolution and Ecology
of Macaque Societies, pp. 73–94. Cambridge:
Cambridge University Press.
Chalmers, N. R. (1968). Group composition, ecology
and daily activity of free living mangabeys in Uganda.
Folia primatol. 8, 247–262.
Ciochon, R. L. (1994). Evolution of the Cercopithecoid
Forelimb: Phylogenetic and Functional Implications
from Morphometric Analyses. Berkeley: University of
California Press.
Cronin, J. E. & Sarich, V. M. (1976). Molecular
evidence for dual origin of mangabeys among Old
World monkeys. Nature 260, 700–702.
Delson, E. & Dean, D. (1993). Are Papio baringensis R.
Leakey, 1969, and P. quadratirostris Iwamoto, 1982,
species of Papio or Theropithecus? In (N. Jablonski,
Ed.) Theropithecus: The Rise and Fall of a Primate
Genus, pp. 125–156. Cambridge: Cambridge
University Press.
Disotell, T. R. (1994). Generic-level relationships of
the Papionini (Cercopithecoidea). Am. J. phys.
Anthrop. 94, 47–57.
Disotell, T. R. (1996). The phylogeny of Old World
monkeys. Evol. Anthrop. 5, 18–24.
Dutrillaux, B., Courturier, J., Muleris, M., Lombard,
M. & Chauvier, G. (1982). Chromosomal phylogeny
of forty-two species or subspecies of cercopithecoids
(Primates Catarrhini). Ann. Genet. 25, 96–109.
Fleagle, J. G. (1977). Locomotor behavior and skeletal
anatomy of sympatric Malaysian leaf monkeys
(Presbytis obscura and Presbytis melalophos). Yearb.
phys. Anthrop. 20, 440–453.
Fleagle, J. G. (1999). Primate Adaptation and Evolution.
2nd edn. New York: Academic Press.
Fleagle, J. G. & Meldrum, D. J. (1988). Locomotor
behavior and skeletal morphology of two sympatric
Pitheciine monkeys, Pithecia pithecia and Chiropotes
satanas. Am. J. Primatol. 16, 227–249.
Fleagle, J. G. & McGraw, W. S. (1998). Skeletal
anatomy of African Papionins: function, phylogeny
or both? Am. J. phys. Anthrop. 105, 82–83.
Fleagle, J. G. & McGraw, W. S. (1999). Skeletal and
dental morphology supports diphyletic origin of
baboons and mandrills. Proc. natl. Acad. Sci. 96,
1157–1161.
Fooden, J. (1980). Classification and distribution of
living macaques (Macaca Lacepede, 1799). In (D. G.
Lindberg, Ed.) The Macaques: Studies in Ecology,
Behavior and Evolution, pp. 1–9. New York: Van
Nostrand-Reinhold.
Gartlan, J. S. (1970). Preliminary notes on the ecology
and behavior of the drill Mandrillus leucophaeus
Ritgen, 1824. In (J. R. Napier & P. H. Napier, Eds)
Old World Monkeys: Evolution, Systematics and
Behavior, pp. 445–480. New York: Academic Press.
Gebo, D. L. & Sargis, E. J. (1994). Terrestrial adaptations in the postcranial skeletons of guenons. Am.
J. phys. Anthrop. 93, 341–371.
Groves, C. P. (1978). Phylogenetic and population systematics of the mangabeys (Primates:
Cercopithecoidea). Primates 19, 1–34.
Groves, C. P. (1993). Order Primates. In (D. E. Wilson
& D. M. Reader, Eds) Mammalian Species of the
World: A Taxonomic and Geographic Reference,
pp. 243–277. Washington, DC: Smithsonian
Institute Press.
Groves, C. P. (2000). The phylogeny of the
Cercopithecoidea. In (P. F. Whitehead & C. J. Jolly,
Eds) Old World Monkeys, pp. 77–98. Cambridge:
Cambridge University Press.
Grubb, P. (1973). Distribution, divergence and
speciation of the drill and mandrill. Folia primatol. 20,
161–177.
Harris, E. E. (2000). Molecular systematics of the Old
World monkey tribe Papionini: analysis of the total
available genetic sequences. J. hum. Evol. 38, 235–
256.
Harris, E. E. & Disotell, T. R. (1998). Nuclear gene
trees and the phylogenetic relationship of the
mangabeys (Primates: Papionini). Mol. Biol. Evol.
15, 892–900.
Harrison, M. J. (1988). The mandrill in Gabon’s rain
forest—ecology, distribution and status. Oryx 22,
218–228.
Hewett-Emmett, D. & Cook, C. N. (1978). Atypical
evolution of papionin -haemoglobins and indication
that Cercocebus may not be a monophyletic genus. In
(D. J. Chivers & K. A. Joysey, Eds) Recent Advances
in Primatology, pp. 291–294. London: Academic
Press.
Hill, W. C. O. (1970). Primates: Comparative Anatomy
and Taxonomy, Vol. VIII, Cynopithecinae: Papio,
291
Mandrillus, Theropithecus. Edinburgh: Edinburgh
University Press.
Homewood, K. M. (1978). Feeding strategy of
the Tana mangabey (Cercocebus galeritus galeritus)
(Mammalia: Primates). J. Zool. Lond. 186, 375–391.
Horn, A. D. (1987). The socioecology of the black
mangabey (Cercocebus aterrimus) near Lake Tumba,
Zaire. Am. J. Primatol. 12, 165–180.
Hoshino, J. (1985). Feeding ecology of mandrills
(Mandrillus sphinx) in Campo Animal Reserve,
Cameroon. Primates 26, 248–273.
Jablonski, N. G. (1993). The phylogeny of
Theropithecus. In (N. G. Jablonski, Ed.) Theropithecus:
The Rise and Fall of a Primate Genus, pp. 209–224.
Cambridge: Cambridge University Press.
Jolly, C. J. (1967). The evolution of the baboons.
In (H. Vagborg, Ed.) The Baboon in Medical
Research, Vol. II, pp. 23–50. Austin: University of
Texas Press.
Jolly, C. J. (1970). The large African monkeys as an
adaptive array. In (J. R. Napier & P. H. Napier, Eds)
Old World Monkeys; Evolution, Systematics, and
Behavior, pp. 139–174. New York: Academic Press.
Jolly, C. J. (1972). The classification and natural
history of Theropithecus (Simopithecus) (Andrews,
1916), baboons of the African Plio-Pleistocene. Bull.
Br. Mus. (nat. Hist.), Geol. 22, 1–123.
Jones, C. & Sabater-Pi, J. (1968). Comparative ecology
of Cercocebus albigena (Gray) and Cercocebus torquatus
(Kerr) in Rio Muni, West Africa. Folia Primatol. 9,
99–113.
Kingdon, J. (1997). The Kingdon Field Guide to African
Mammals. New York: Academic Press.
McGraw, W. S. (1996). The positional behavior and
habitat use of six sympatric monkeys in the Tai
Forest, Ivory Coast. Ph.D. Dissertation, SUNY at
Stony Brook.
Mitani, M. (1989). Cercocebus torquatus: adaptive
feeding and ranging behaviors related to seasonal
fluctuations of food resources in the tropical rain
forest of south-western Cameroon. Primates 30, 307–
323.
Nakatsukasa, M. (1994a). Intrageneric variation of
limb bones and implications for positional behavior
in Old World monkeys. Z. Morph. Anthrop. 80,
125–136.
Nakatsukasa, M. (1994b). Morphology of the humerus
and femur in African mangabeys and guenons:
functional adaptations and implications for the
evolution of positional behavior. Afr. Study Monogr.
21, 1–61.
Nakatsukasa, M. (1996). Locomotor differentiation
and different skeletal morphologies in mangabeys
(Lophocebus and Cercocebus). Folia primatol. 66,
15–24.
Olupot, W., Chapman, C. A., Brown, C. H. & Waser,
P. M. (1994). Mangabey (Cercocebus albigena) population density, group size, and ranging: a twenty year
comparison. Am. J. Primatol. 32, 197–205.
Olupot, W., Chapman, C. A., Waser, P. M. & IsabiryeBasuta, G. (1997). Mangabey (Cercocebus albigena)
ranging patterns in relation to fruit availability and
292
. . .
the risk of parasite infection in Kibale National park,
Uganda. Am. J. Primatol. 43, 65–78.
Quiris, R. (1975). Ecologie et organisation sociale de
Cercocebus galeritus agilis dans le Nord-Est du Gabon.
Terre et Vie 29, 337–398.
Range, F. (1998). Sozialsystem and Konkurrenztyp
weiblicher Raugrauer mangaben (Cercocebus torquatus
atys). Masters Thesis, University Bayreuth,
Germany.
Rogers, M. E., Abernathy, K. A., Fontaine, B.,
Wickings, E. J., White, L. J. & Tutin, C. E. G.
(1996). Ten days in the life of a mandrill horde in the
Lope Reserve, Gabon. Am. J. Primatol. 40, 297–313.
Ross, C. F., Lockwood, C. A., Fleagle, J. G. & Jungers,
W. L. (2001). Adaptation and behavior in the primate fossil record. In (J. M. Plavcan, C. P. van
Schaik, R. F. Kay & W. L. Jungers, Eds) Reconstructing Behavior in the Primate Fossil Record, pp. 1–41.
New York: Plenum.
Rowe, N. (1996). The Pictorial Guide to the Living
Primates. East Hampton: Pogonias Press.
Rutte, C. (1998). Die Nahrungssuchstrategie der
Rauchgrauen Mangabe (Cercocebus torquatus atys).
Masters Thesis, University Erlangen, Germany.
Schaaf, C. D., Butynski, T. M. & Hearn, G. W. (1990).
The drill (Mandrillus leucophaeus) and other primates
in the Gran Caldera Volcanica de Luba: Results of a
survey conducted March 7–22, 1990. Unpublished
report, Government of the Republic of Equatorial
Guinea.
Shah, N. F. (1996). Preliminary study of two sympatric
mangabey species (Cercocebus galeritus agilis and
Cercocebus albigena) in the Dzanga-Sangha Reserve,
Central African Republic. XVIth Congress of the
International Primatological Society.
Stanyon, R., Fantini, C., Camperio-Ciani, A.,
Chiarelli, B. & Ardito, G. (1988). Banded karyotypes of 20 Papionini species reveal no necessary
correlation with speciation. Am. J. Primatol. 16,
3–17.
Szalay, F. S. & Delson, E. (1979). Evolutionary
History of the Primates. New York: Academic
Press.
Thorington, R. W. & Groves, C. P. (1970). An annotated classification of the Cercopithecoidea. In (J. R.
Napier & P. H. Napier, Eds) Old World Monkeys:
Evolution, Systematics and Behavior, pp. 629–648.
New York: Academic Press.
Ward, C. V., Ruff, C. B., Walker, A., Teaford,
M. F., Rose, M. D. & Nengo, I. O. (1995).
Functional morphology of Proconsul patellas from
Rusinga Island, Kenya, with implications for other
Miocene–Pliocene catarrhines. J. hum. Evol. 29,
1–20.
Waser, P. (1977). Feeding, ranging and group size in
the mangabey C. albigena. In (T. H. Clutton-Brock,
Ed.) Primate Ecology: Studies of Feeding and Ranging
Behavior in Lemurs, Monkeys and Apes, pp. 183–222.
New York: Academic Press.
Waser, P. (1984). Ecological differences and behavioral
contrasts between two mangabey species. In (P. S.
Rodman & J. G. H. Cant, Eds) Adaptations for
Foraging in Non-human Primates, pp. 195–216. New
York: Columbia University Press.
John Fleagle