вход по аккаунту


Articular to diaphyseal proportions of human and great ape metatarsals.

код для вставкиСкачать
Articular to Diaphyseal Proportions of Human
and Great Ape Metatarsals
Damiano Marchi*
Department of Evolutionary Anthropology, Duke University, Durham, NC 27708-0383
forefoot; metatarsophalangeal joint; australopithecine locomotion; cross-sectional
geometry; metatarsus
This study proposes a new way to use
metatarsals to identify locomotor behavior of fossil hominins. Metatarsal head articular dimensions and diaphyseal strength in a sample of chimpanzees, gorillas, orangutans, and humans (n 5 76) are used to explore the relationships of these parameters with different locomotor
modes. Results show that ratios between metatarsal
head articular proportions and diaphyseal strength of
the hallucal and fifth metatarsal discriminate among
extant great apes and humans based on their different
locomotor modes. In particular, the hallucal and fifth
metatarsal characteristics of humans are functionally
related to the different ranges of motion and load patterns during stance phase in the forefoot of humans in
Studies of australopithecine foot morphology have generally relied upon comparisons between the pedal skeleton of Au. afarensis and those of extant hominoids, usually chimpanzees, gorillas, orangutans, and humans.
One region of the foot that has been extensively investigated to establish the amount of bipedal adaptation in
early australopithecines is the forefoot and in particular
the metatarsophalangeal joints (MTPJs), (Stern and Susman, 1983; Susman et al., 1984; Latimer and Lovejoy,
1990a,b; Duncan et al., 1994; Griffin, 2009; Griffin and
Richmond, 2010). For example, it has been demonstrated
that modern human MTPJ morphology provides great
dorsiflexion of the foot during bipedal locomotion, in contrast to nonhuman hominoid MTPJs that provide great
plantarflexion of the foot due to the use of the foot in
climbing and grasping (Stern and Susman, 1983; Susman et al., 1984; Latimer and Lovejoy, 1990a,b).
While the relationship between different forefoot morphology and differing locomotor behaviors among extant
hominoids is well understood, the study of the forefoot of
Au. afarensis in a comparative context with extant great
apes by different authors has produced differing conclusions. Some authors interpret the degree of adaptation of
the Au. afarensis foot to bipedality as derived toward the
human condition (Latimer and Lovejoy, 1990a,b) while
others have proposed that the foot was mostly ape-like
(Stern and Susman, 1983; Susman, 1983; Susman et al.,
1984). In an attempt to provide a quantitative assessment of the functional morphology of australopithecine
MTPJs, Duncan et al. (1994) applied computer imaging
technology to the study of metatarsal head morphology
in extant hominoids and Au. afarensis. They found that
the measurement of metatarsal head morphology in the
dorsoplantar plane does not differentiate among locomotor behaviors in hominoids. Therefore, they concluded
that this characteristic is not a conclusive means of evaluating the function of Au. afarensis metatarsals (MTs).
C 2010
bipedal locomotion. This method may be applicable to
isolated fossil hominin metatarsals to provide new information relevant to debates regarding the evolution of
human bipedal locomotion. The second to fourth metatarsals are not useful in distinguishing among hominoids. Further studies should concentrate on measuring
other important qualitative and quantitative differences
in the shape of the metatarsal head of hominoids that
are not reflected in simple geometric reconstructions of
the articulation, and gathering more forefoot kinematic
data on great apes to better understand differences in
range of motion and loading patterns of the metatarsals.
Am J Phys Anthropol 143:198–207, 2010. V 2010 WileyC
Liss, Inc.
A more recent study (Griffin and Richmond, 2010)
found that modern humans show more dorsal canting of
the proximal pedal phalanges than do extant apes. However, human values overlap with great ape values, and
the Au. afarensis from Hadar (A.L. 333–115 g) proximal
phalangeal joint orientation overlaps with that of both
modern humans and African apes. The authors concluded that dorsal canting may not be the only skeletal
signal of joint extension (Griffin and Richmond, 2010).
One of the reasons for the lack of significance in the
comparisons carried out using MT head properties may
be that these studies only considered ‘‘dorsiflexion indicators’’ on the MTPJs. However, as previously suggested
(Susman, 1983; Susman et al., 1984; Duncan et al.,
1994), enhanced dorsal excursion of Au. afarensis MTPJs
compared to the MTPJs of the great apes would not
mean that ranges of MTPJ plantarflexion would have
been significantly limited in Au. afarensis. Therefore, we
need an MT property that includes both dorsiflexion and
plantarflexion information to effectively assess Au. afarensis forefoot function.
Another reason for the lack of significance of previous
studies may be that locomotor behavior normally affects
both articular properties and strength of a structure
Additional Supporting Information may be found in the online
version of this article.
*Correspondence to: Damiano Marchi, Department of Evolutionary Anthropology, Duke University, Durham, NC 27708-0383.
Received 24 April 2009; accepted 19 February 2010
DOI 10.1002/ajpa.21306
Published online 21 April 2010 in Wiley Online Library
Fig. 1. Pictorial representation of the measurements used in
this analysis, as shown on a human metatarsal: a,b, metatarsal
length; c,d, dorsoplantar periosteal diameter at midshaft; e,f,
mediolateral periosteal diameter at midshaft; g,h, maximum
dorsoplantar head breadth; i,j, mediolateral head breadth.
(Ruff, 2002). While joint surfaces transmit only compressive loads and more importantly determine range of
motion and joint stability during motion (Godfrey et al.,
1991; Rafferty and Ruff, 1994; Hamrick et al., 2000;
Drapeau, 2008), the primary role of the diaphysis is to
resist bending and torsional loads (Rubin and Lanyon,
1982; Demes et al., 2001). That applies to MT diaphyses
as well. Wunderlich’s (1999) study on plantar pressure
distribution and Marchi’s (2005) study (Marchi and Borgognini-Tarli, 2004) on cross-sectional geometric (CSG)
properties of hominoid MTs provided evidence of the
relationship existing between MT diaphyseal strength
and different locomotor modes in extant hominoids (but
see Griffin and Richmond, 2005), even though some overlap between species was present. Further, Ruff (2002)
demonstrated that limb long bone articular and crosssectional proportions combined can distinguish between
locomotor modes among primates, at least to the level of
broad locomotor differences.
In this study I propose the use of maximum dorsoplantar (DP) and mediolateral (ML) diameter (see Fig. 1) and
total estimated surface area of MT head to provide a
means to quantify the total amount of excursion present
at the MTPJs, as constituted by both dorsiflexion and
plantarflexion. Even if the use of maximum DP diameter
implies a loss of information on ‘‘pure dorsiflexion’’ at
the level of the MTPJs, the lack of conclusive results
from the studies reviewed above that attempted to quantify MTPJ dorsiflexion to separate extant hominoids suggest the need for different skeletal estimators of forefoot
function. Further, I propose to combine MT head proportions with cross-sectional diaphyseal properties in an
attempt to control for the load bearing function of MTs.
This will allow us to explore how much of the differences
between hominoids in MT articular size is due to MTPJ
excursion and how much to MT load bearing patterns.
This study tests the hypothesis that distal articular size
relative to mechanical load on the MTs varies between
species (chimpanzee, gorilla, orangutan, and human)
and digital rays according to different locomotor behaviors. Changes in the articular/diaphyseal ratio may be
driven by differences in mechanical loading of the bone
(without a change in joint excursion, for example) as
well as differences in joint function.
The first prediction of this study is that humans will
have relatively greater MT I head proportions (area and
diameters) than great apes. This prediction is based on
the observation that the first MTPJ hyperextension at
toe-off in humans is greater than in great apes (Latimer
and Lovejoy, 1990a,b).
The second prediction is that orangutans will have relatively smaller MT I head proportions than all of the
other hominoids based on the observation of the small
degree of movement present at the level of their hallucal
MTPJ (Rose, 1988). Further, I predict that orangutans
will have greater MT II-V head proportions to midshaft
strength based on the observations of their hook-like foot
that requires mobile articulations (Tuttle, 1970; Rose,
1988) and relatively low diaphyseal strength (Marchi,
The third prediction is that African great apes will
have MT head to midshaft strength proportions in
between those of humans and orangutans. This prediction is based on the observation that African great ape
locomotion is a compromise between arboreal and terrestrial locomotion (Tuttle, 1970; Tuttle et al., 1998).
Because of that they have MTs more adapted to the arboreal environment than humans and more adapted to
the terrestrial environment than orangutans. This compromise will produce a forefoot with characteristics in
between those of the other two species.
The sample
The sample analyzed here is taken from a larger sample already described in detail elsewhere (Marchi, 2005).
Sample sizes for each species are listed in Table 1.
Because of the differences in the proportion of cortical
to total area between subadults and adults (Ruff et al.,
1994), the sample consists of only adults for each species.
Only individuals with complete long bone epiphyseal
fusion and no signs of senescence (e.g., arthritic changes,
osteoporosis) were selected.
Sex was known for all the great apes. For the humans,
sex had to be assigned. The good state of preservation of
the sample permitted a reliable sex attribution by means
of pelvic and cranial traits (Brothwell, 1981; Bruzek,
2002). Right and left MT I-V were measured for each
individual and then averaged.
Articular dimensions
Linear dimensions of the diaphysis and articulation
are listed in Table 2 and shown in Figure 1. Articular
dimensions were chosen to represent the major dimensions of each articular surface and as input in the following geometric formula for calculating total surface areas:
metatarsal head surface area 5 DP head breadth 3 ML
head breadth.
The estimation of articular surfaces using formulae
based on modeling of surfaces as ovals, rectangles, partial spheres, and cylinders has been utilized in previous
studies (Runestad, 1997; Egi, 2001; Ruff, 2002). Direct
comparisons of geometric estimations and measurements
from latex molds taken from the same articulations perAmerican Journal of Physical Anthropology
TABLE 1. Study sample
8 (6 captive, 2 wild-shot)
10 (all wild-shot)
5 (1 captive, 4 wild-shot)
11 (5 captive, 6 wild-shot)
5 (1 captive, 4 wild-shot)
7 (3 captive, 4 wild-shot)
From the Anthropologische Staatssammlung at the Universität München, Germany. Medieval humans (7th C. AD) from a
German necropolis (Neuburg, Donau).
From the Schultz collection and the primate collection at the
Universität Zürich Irchel, Switzerland; and from the Zoologische Staatssammlung at the Universität München, Germany.
Pan troglodytes troglodytes, P. t. schweinfurthii, P. t. sp.
Gorilla gorilla gorilla.
Pongo pygmaeus pygmaeus, Pongo p. abelii, Pongo p. sp.
formed on humeral and femoral articulations of primates
and nonprimate mammals have been shown to be reasonably accurate (Rafferty, 1996; Egi, 1999; Wunderlich,
1999). Therefore, I have used a surface area estimation
based on the roughly rectangular contour of the MT
TABLE 2. Abbreviations and description of structural properties
of metatarsals
Cross-sectional diaphyseal dimensions
One cross-section at a level equal to 50% of the MT
length was taken. The orientation of reference axes and
the section location follow Marchi (2005). Bones were oriented in a standardized position and polysiloxane molds
were placed around the diaphysis to record the subperiosteal contours. Estimates of endosteal contours were
obtained from measurements of biplanar radiographs of
the diaphysis. In the absence of a CT scanner, this
method yields reasonably accurate results (O’Neill and
Ruff, 2004). Scion image macros modeled after the
SLICE program (Nagurka and Hayes, 1980) were used
to calculate standard cross-sectional properties.
The cross-sectional properties measured in this study
were the cortical area (CA) and the polar section modulus (Zp). The CA is proportional to axial compressive and
tensile strength and Zp to torsional strength and average
bending strength. Zp can be approximated from J, the
polar second moment of area, as J0.73 (Ruff, 1995, 2002).
In long bones of the hindlimb, the primary direction of
bending during active quadrupedal locomotion is
expected to be in the anteroposterior (AP) plane, based
on the orientation of the hindlimb and in vivo strain
gauge measurements in quadrupeds (e.g., Carter et al.,
1981; Demes et al., 2001). Therefore, to distinguish species that engage more frequently in running and leaping,
Zx (AP bending strength) has been used (Ruff, 2002) as a
measure of diaphyseal strength in the femur and tibia.
There are no strain gauge studies on the MTs. However,
the strains developed during locomotion in the forefoot
are likely to be more variable (Wunderlich, 1999). Therefore, Zp (twice average bending strength) was used as a
measure of diaphyseal strength in the MTs. Analyses
were also carried out with simple external breadth
dimensions, specifically, the average of DP and ML
breadths of the MTs midshaft. All properties and their
abbreviations are listed in Table 2.
Statistical methods
Following Ruff (2002), natural log ratios of within
bone properties were compared. This method is valid if
American Journal of Physical Anthropology
The distance between the centre of the
proximal articulation and the most
prominent part of the distal
articulation (head) (Fig. 1a,b).
Metatarsal 1 to 5 dorsoplantar
periosteal diameter at 50% of the MT
length with the bone held in
standardized position (Fig. 1c,d).
Metatarsal 1 to 5 mediolateral
periosteal diameter at 50% of the MT
length with the bone held in
standardized position (Fig. 1e,f).
Metatarsal 1 to 5 average dorsoplantar
and mediolateral breadths at 50% of
the MT length.
Metatarsal 1 to 5 head dorsoplantar
breadth. The maximum dorsoplantar
breadth of the MT head not including
the epicondyle behind the dorsal
aspect of the head (Fig. 1g,h).
Metatarsal 1 to 5 head mediolateral
breadth. The maximum mediolateral
breadth of the MT distal articulation
with the bone held in palmar view
(Fig. 1i,j).
Metatarsal 1 to 5 surface area.
Metatarsal 1 to 5 midshaft torsional
See text for the formula used to calculate articular surface areas.
See text for explanation of Zp.
animals with similar locomotor behaviors are isometric
in structural proportions. Ruff (2002) also suggested that
the best way to test this hypothesis would be between
primates of different body size involved in similar locomotion. In this study, the two species that are involved
in the most similar locomotion are the chimpanzee and
the gorilla. However, it is hard to argue for ‘‘functional
equivalence’’ between these African apes. In fact, gorillas
and chimpanzees vary in their locomotor behavior
(Remis, 1993, 1995; Doran 1996, 1997). Therefore, I
tested for isometry in each species.
Appendices 1–4 show the results of log-log bivariate
reduced major axis regressions (see, Ruff, 2000 for reasons why this method is appropriate) between a variety
of structural properties of MTs for humans, chimpanzees, gorillas, and orangutans. Comparison of observed
slopes to predicted isometric values were carried out following Hofman (1988). Because of the difference in units
of measurement, the expected isometric slopes for each
variable are as follows: surface areas to Zp have a theoretical isometric slope of 2/3 (0.667), linear diameters to
Zp of 1/3 (0.333), surface areas to average midshaft diameter 2/1 (2.000), and linear diameters to average midshaft diameter 1/1 (1.000). The Tukey test of multiple
comparison was used to test for differences in species
mean ratios (Appendices 1–4). Group differences were
considered significant at the P \ 0.05 level. Results
revealed for the most part no significant differences
between the observed slopes and the expected isometry
for each of the variables within each group. This supports the use of log-ratios of structural properties of MTs
to assess proportional differences between species.
Differences in locomotion between captive and wildshot animals might be expected (Marchi, 2005, 2007).
TABLE 3. Within-bone distal articulation to midshaft cross-sectional geometric proportions of metatarsals
Mean SD (n 5 19)
Head surface area/section modulus
2.328bb (0.15)
2.118b (0.18)
2.285b (0.18)
2.375b (0.14)
2.126b (0.15)
Head dorsoplantar breadth/section modulus
1.167b (0.07)
1.264b (0.10)
1.380b,c (0.11)
1.436a,b (0.10)
1.300b (0.10)
Head mediolateral breadth/section modulus
1.165a (0.12)
0.859c (0.10)
0.909b (0.11)
0.942b (0.08)
0.829b,c (0.10)
Mean SD (n 5 15)
Mean SD (n 5 11)
Mean SD (n 5 24)
See Table 2 for abbreviations. All ratios were natural log-transformed, e.g., ln (MTISA/MTIZp0.667).
Letters indicate results of Tukey multiple comparisons tests: possession of the same letter 5 nonsignificant difference (P [ 0.05)
between groups; a 5 highest mean, b 5 next highest mean, etc.
n 5 18.
n 5 15.
n 5 19.
n 5 20.
n 5 17.
n 5 12.
n 5 21.
Given that the chimpanzee sample contains 11 captive
individuals out of 19 and the orangutan sample four captive individuals out of 12, sample size would be appreciably reduced if only wild-shot animals were considered.
To determine whether there were significant differences
between wild-shot and captive subsamples, statistical
tests were performed among subsamples. Because of the
nonnormal distributions (Kolmogorov-Smirnov test),
Mann-Whitney U-tests between wild-shot and captive
(and Kruskal-Wallis analysis of rank for wild-shot only,
see below) articular to cross-sectional proportions of MTs
(log-ratios) were carried out for chimpanzees and orangutans. Only one captive individual was included in the
gorilla sample and its values were within the range of
values of those of the other individuals (Supporting Information Tables S1 and S2). Comparisons of MT proportions among the complete samples of the different species analyzed in this study were carried out using Tukey
multiple comparison tests. Box plots are used to represent graphically group data distributions. All statistical
analyses were done using STATISTICA 7.
(Supporting Information Tables S1 and S2) show that
there are few significant differences across species
among the pooled sample and the wild-shot only sample.
Moreover, the general distribution for all the tested variables (e.g., the species with the highest or lowest value
for a variable) did not change between the pooled sample
and the wild-shot sample. The comparisons of box plots
and medians for the pooled sample (Figs. 2–4) and the
wild-shot sample (Supporting Information Figs. S1–S3)
also show the same general pattern for all the variables.
Although it is important to be aware of variation
between wild-shot and captive animals, the same patterns observed for the pooled and wild-shot chimpanzee
samples analyzed here suggest that reliable information
can be gleaned from a pooled sample. Therefore, I have
pooled wild-shot and captive chimpanzees for all further
analyses to maintain a sufficiently large sample size.
Within-bone articular to cross-sectional
diaphyseal proportions
Wild-shot vs. captive
Results of Tukey multiple comparison tests between
species for within-bone proportions of MTs are given in
Tables 3 and 4. Figures 2–4 show boxplots of some of the
variables for MTs.
Mann-Whitney U-tests on the articular to cross-sectional proportions and diameters of the MTs (logged
ratios), (Supporting Information Tables S3 and S4) show
that wild-shot orangutans do not significantly differ from
captive orangutans. Wild-shot chimpanzees were different from captive chimpanzees in 11 out of 30 ratios
(wild-shot animals have relatively larger head proportions than captive animals). Tukey multiple comparison
tests on the pooled sample (Tables 3 and 4) and KruskalWallis analysis of ranks for the wild-shot only sample
Metatarsal I. Metatarsal I head surface area to midshaft strength (Zp) is presented in Figure 2a. In both the
comparison of medians and group means using Tukey
tests (Table 3), orangutans (quadrumanous) show the
smallest head relative surface area for MT I, while
humans (bipeds) show the largest head relative surface
area for MT I. African great apes (knuckle-walkers) have
values in between those of orangutans and humans. All
differences between groups are significant (P \ 0.05).
Grouping is similar using the DP articular breadth of
American Journal of Physical Anthropology
TABLE 4. Within-bone distal articulation to midshaft average diameter proportions of metatarsals
Mean SD (n 5 19)
Head surface area/average midshaft breadth
20.331ab (0.06)
20.426b (0.08)
20.332b (0.09)
20.294b (0.07)
20.408b (0.08)
Head dorsoplantar breadth/average midshaft breadth
20.330a,b (0.05)
20.223b (0.09)
20.097b,c (0.10)
20.047a (0.10)
20.172b (0.11)
Head mediolateral breadth/average midshaft breadth
20.333a (0.10)
20.628b (0.09)
20.568b (0.11)
20.540b (0.08)
20.643b,c (0.11)
Mean SD (n 5 15)
Mean SD (n 5 11)
Mean SD (n 5 25)
See Table 2 for abbreviations. All ratios were natural log-transformed, e.g., ln (MTISA0.5/MTIAV).
Letters indicate results of Tukey multiple comparisons tests: possession of the same letter 5 nonsignificant difference (P [ 0.05)
between groups; a 5 highest mean, b 5 next highest mean, etc.
n 5 19.
n 5 16.
n 5 20.
n 5 21.
n 5 18.
n 5 12.
n 5 22.
the MT head (Fig. 3a). When ML articular breadth is
used, more overlapping between species is present.
Orangutans show the relatively smallest ML head
breadth (P \ 0.05) and African apes and humans cannot
be separated on the basis of this head property (Fig. 4a;
Table 3).
Metatarsal I articular to average midshaft diameter
results are shown in Table 4. As for the articular to Zp
comparisons, orangutans have the relatively smallest
MT I head for both the area and the ML and DP
breadths (P \ 0.05). However, humans and African great
apes cannot be separated for any of the MT I head properties when they are compared with midshaft diameters
rather than Zp.
Metatarsal II–V. Head surface area to midshaft
strength results show that for MT II orangutans and
humans are not significantly different from each other
and have significantly larger relative surface areas than
African great apes. Orangutans have the largest relative
surface areas for MT III-V in both the comparison of
medians (Fig. 2c,d) and group means using Tukey tests
(P \ 0.05, Table 3). Humans have the smallest relative
surface area for MT V (P \ 0.05). African apes group together (Fig. 2b–d; Table 3) and have the smallest relative surface area for MT II, and are not significantly different from humans for MT III and IV. For MT V African
great apes show values in between those of orangutans
and humans (P \ 0.05).
In the DP articular breadth to midshaft strength comparisons, orangutans have among the deepest MT heads
for MT II-V, even though they are significantly deeper
than the other species only for MT V. African great apes
overlap to different degrees with humans and orangutans. Only for MT II do they form a group with relative
DP breadth significantly smaller than orangutans and
American Journal of Physical Anthropology
humans. A similar result is present for ML articular
breadth to midshaft strength comparisons (Figs. 3b–d,
4b–d; Table 3).
Metatarsal II-V articular to midshaft average diameter
results (Table 4) show more overlap among species to the
extent that it is difficult to separate them. However,
orangutans still show the largest relative surface area
for MT III-V (P \ 0.05) and humans show the smallest
relative surface area for MT V (P \ 0.05). More overlapping is present in the other comparisons.
The use of external diaphyseal breadths rather than
section moduli often produces more overlap between
groups, in agreement with a previous study on limb long
bones (Ruff, 2002). Therefore, I will discuss only the
results obtained from the MT head properties to midshaft diaphyseal strength proportions.
Metatarsal head proportions to midshaft strength
Metatarsal I. The present study shows that, as predicted, humans have the largest MT I head surface area
and DP breadth relative to midshaft strength among
hominoids. It may be argued that the relatively larger
MT I head proportions may be related to higher loading
on the MT I in humans. However, expressing MT I articular size relative to MT I diaphyseal strength should
standardize for the greater mechanical load on the
human MT I. Further, only a relative increase in DP
breadth, not ML breadth of the human MT I head has
been found. If the larger MT I relative head surface was
due to higher joint reaction force an increase in both DP
and ML breadths would be expected. This result pro-
Fig. 2. Metatarsal head surface area relative to midshaft diaphyseal strength. I, metatarsal I; II, metatarsal II; III, metatarsal
III; IV, metatarsal IV; V, metatarsal V.
vides evidence that the greater relative surface area of
the human MT I can be attributed especially to an
enlarged joint along the dorsoplantar axis rather than
the mediolateral axis, and therefore to a relatively larger
total dorsoplantar range of movement at the MTPJ in
humans than in great apes.
As predicted, orangutans have the smallest MT I surface area, DP and ML breadth to midshaft diaphyseal
strength proportions among the hominoids studied here.
The bony features of the orangutan foot enable it to
function as a suspensory hook-like organ, which is very
different from the plantigrade foot of African great apes
(Tuttle, 1970; Tuttle et al., 1998). The orangutan hallux
is short, does not participate in the hook and is not important for power gripping (Tuttle, 1970; Rose, 1988;
Tuttle et al., 1998). Accordingly, a previous study found
that orangutans have lower hallucal midshaft diaphyseal
strength than African great apes and humans (Marchi,
2005). Because of the low midshaft hallucal strength the
low MT I head proportions to midshaft strength proportions reflect the small degree of MTPJ movement present at this joint (Tuttle, 1970; Rose, 1988).
African ape MT I surface area and DP breadth relative
to midshaft strength is in between those of humans and
orangutans as predicted. Gorillas and especially chimpanzees have a foot that is adapted to grasping in an arboreal environment (Schultz, 1963; Tuttle, 1970; Duncan
et al., 1994), but it is also more specialized for locomotion on the ground than the orangutan foot (Schultz,
1963; Tuttle, 1970; Rose, 1988; Wang and Crompton,
2004). However, because the African ape hallux is also
specialized for opposability necessary in an arboreal
American Journal of Physical Anthropology
Fig. 3. Metatarsal head dorsoplantar breadth relative to midshaft diaphyseal strength. I, metatarsal I; II, metatarsal II; III,
metatarsal III; IV, metatarsal IV; V, metatarsal V.
environment, it is not suitable for the toe-off mechanism
observed in humans. These results agree with the evidence that the African ape MTPJ is biomechanically
more adapted to an arboreal environment than that of
humans, and more adapted to a terrestrial environment
than that of orangutans (Tuttle, 1970; Susman, 1983;
Susman et al., 1984; Latimer and Lovejoy, 1990a; Tuttle
et al., 1998).
Metatarsal II–V. Contrary to expectations, humans and
orangutans show the highest MT II head to midshaft
strength proportions among the hominoids studied here,
and are not significantly different from each other. The
two species are at the two extremes of the range of hominoid locomotor behaviors: humans are permanently terrestrial while orangutans are the most arboreal species.
American Journal of Physical Anthropology
Because of the functional similarity of MT I and MT II
for propulsion during gait (Griffin, 2009) the high relative MT head properties of MT II in humans could be
explained as for MT I. However, it could be argued that
the reason for the high human relative MT II head properties is the weak MT II diaphysis in humans (Griffin
and Richmond, 2005). It has been observed that human
MT II diaphyseal strength is not different from that of
orangutans and gorillas (Marchi, 2005) and gorillas
show smaller MT II head properties to diaphyseal
strength than humans. Therefore, the weak diaphysis
alone cannot explain these results. Probably, a combination of diaphyseal robusticity and MT head proportions
unique to humans is responsible for these results.
The head of MT II in orangutans is adapted for plantarflexion, a range of motion that is functionally advan-
Fig. 4. Metatarsal head mediolateral breadth relative to midshaft diaphyseal strength. I, metatarsal I; II, metatarsal II; III,
metatarsal III; IV, metatarsal IV; V, metatarsal V.
tageous for arboreal climbing and suspensory locomotion
(Rose, 1988). Orangutans’ great relative MT II head surface area and DP breadth may be explained as the consequence of their increased forefoot mobility in an arboreal environment. In summary, even if the similarities
between humans and orangutans can be explained on
the basis of our knowledge of their locomotion, these
results show that the relative MT II head surface area
cannot distinguish between them and therefore is not
suitable for fossil applications.
Orangutans show the largest MT III-V head surface
area and breadths (MT IV DP breadth excepted) to midshaft proportions among hominoids, as predicted. During
suspensory and cautious quadrumanous locomotion,
orangutans generally hold onto branches with at least
three cheiridia (Davenport, 1967; Tuttle, 1970; Cant,
1987) that require very mobile metacarpophalangeal
articulations (Tuttle, 1970). Rose (1988) proposed that
the circular outline of the proximal articular surface of
the proximal phalanges in the hand may indicate that
orangutans are involved in extensive adduction/abduction at the level of the metacarpophalangeal articulation.
These features may be true of the pedal phalanges as
well (Rose, 1988). The generally large DP and ML head
breadth to midshaft strength of orangutans’ MTs found
in the present study agree with the large range of
adduction/abduction at MTPJ articulation proposed by
Rose (1988). The increase in relative MT head surface
area would potentially increase the range of maximum
stability during flexion/extension and abduction/adduction at the MTPJ providing greater surface of contact
between the head of MTs and the proximal articulation
of the proximal phalanges at any angle of plantarflexion
and dorsiflexion of the forefoot. However, we must be
American Journal of Physical Anthropology
aware of the results from a recent study on proximal
pedal phalanges dorsal canting and MTPJ kinematics
(Griffin, 2009) that suggest that it may not always be
sufficient to base locomotor prediction on joint morphology. More in vivo kinematic data are needed to understand if the relative MT head properties analyzed in this
study can be entirely explained in terms of increased
mobility at the MTPJs.
Humans have the smallest relative MT V head surface
area among all the hominoids, while orangutans have
the largest and African great apes have values in
between the other two species. Ape results agree with
expectations and the general findings that orangutans
have more mobile articulations in the foot (Tuttle, 1970;
Rose, 1988) and relatively low diaphyseal strength
(Marchi, 2005), and that African great apes have a foot
that is adapted to grasping in an arboreal environment
but also more specialized for locomotion on the ground
than the orangutan foot (Schultz, 1963; Tuttle, 1970;
Rose, 1988; Duncan et al., 1994; Wang and Crompton,
2004). Human results can be understood by looking into
the differences in MT V diaphyseal strength among hominoids. It has been found that African great apes and
orangutans do not show any difference in MT V diaphyseal robusticity (Marchi, 2005). Therefore, the functional
interpretation of the ratio of MT V head proportion to
midshaft strength provided here, focusing on the different forefoot range of movement due to different locomotor habits among extant great apes, can be supported.
However, humans show the most robust MT V diaphysis
among hominoids (Marchi, 2005). Therefore, the relatively smallest MT V ratio among hominoids that is
found in humans is probably more a consequence of the
different patterns of load shift during stance phase in
humans and apes (Wunderlich, 1999)—which produce
more robust MT V diaphyses in humans—than of the
range of movement at the level of the fifth MTPJ. More
studies on the total range of excursion at the level of the
MTPJs (and inclusion of Asian great apes) are needed to
fully understand the relationship between mobility and
morphology at the MTPJ of lateral MTs within hominoids.
Inferring fossil hominin locomotion
The results of this study show that MT I and MT V
articular to diaphyseal strength proportions can be used
as morphological indicators of forefoot function in extant
hominoids (hylobatids excluded). The large surface area
and DP diameter relative to midshaft strength of MT I
of humans compared to great apes is functionally related
to the relatively higher range of motion of this joint in
bipedal locomotion (Stern and Susman, 1983; Latimer
and Lovejoy, 1990a,b; Duncan et al., 1994). The small
surface area and DP diameter relative to midshaft
strength of MT V of humans compared to great apes is
functionally related to the difference in load patterns
during stance phase in the foot of humans and apes that
results in more robust MT V diaphyses in humans (Wunderlich, 1999; Marchi, 2005).
This morphological indicator of forefoot propulsion can
be applied to fossil hominin forefoot bones such as the
recently described complete australopithecine MT V
(Zipfel et al., 2009) and others described in the past
(Stern and Susman, 1983; Susman and Brain, 1988; Susman and de Ruiter, 2004) to provide new evidence about
early hominin locomotor adaptations. The method may
American Journal of Physical Anthropology
also be applied to the recently described MT I of Ardipithecus ramidus (Lovejoy et al., 2009). If the proposed
upright walking adaptations of Ar. ramidus provided by
recent studies on the fossil material (White et al., 2009)
prove to be correct, then new hypotheses about the evolution of hominin bipedal locomotion should be considered. The application of the method developed in this
study to the Ar. ramidus material may provide further
evidence on the locomotory adaptation of this important
This study investigated the relationships between
metatarsal (MT) head and dyaphyseal strength proportions among great apes and humans. The goal was to
provide a method that may be applied to early hominin
fossils to better understand their forefoot function.
The results of the present study show that MT I and
MT V articular to cross-sectional diaphyseal proportions
provide statistically significant results that can be used
to distinguish between extant hominoids. In particular,
they can be used to distinguish between the human-like
forefoot and great ape-like forefoot. Therefore, this
method can be applied to fossil MT I and MT V to provide new evidence of forefoot function in early hominins.
However, MT II-IV are not suitable to distinguish among
hominoids based on their locomotor behavior. There may
be two reasons for these results. First, forefoot kinematic
data on great apes are rare. More data are necessary for
more accurately determining joint range of motion at the
metatarsophalangeal joint. Second, these results may
reflect important qualitative and quantitative differences
in the shape of the MT head of hominoids that cannot be
assessed in studies that use a simple geometric reconstruction of the articulation. It is therefore necessary for
future studies to collect kinematic data of the forefoot of
nonhuman hominoids (Griffin, 2009) and, as already suggested in the past (Duncan et al., 1994), to better quantify the topography of the MT head, for example using
multidimensional morphometrics (Proctor et al., 2008),
to include all the characteristics of this articulation in
studies of locomotor reconstructions.
The authors expresses his gratitude to Karin Isler,
University of Zurich, Irchel, Switzerland and Gisela
Grupe, University of Munich, Germany for access to
specimens. He also thanks Rebecca Cuddahee, Deborah
Cunningham, Tracy Kivell, and Denny Wenscott for
helpful comments during the preparation of the manuscript; Chris Ruff and two anonymous reviewers for the
useful comments provided during the reviewing process.
He finally expresses his gratitude to Kirk Johnson and
Amy Schreier for editing the final version of this article.
Brothwell DR. 1981. Digging up bones. Oxford: Oxford University Press.
Bruzek J. 2002. A method for visual determination of sex using
the human hip bone. Am J Phys Anthropol 117:157–168.
Cant JGH. 1987. Effects of sexual dimorphism in body size of
feeding postural behavior of Sumatran orangutans (Pongo
pygmaeus). Am J Phys Anthropol 74:143–148.
Carter DR, Vasu R, Spengler DM, Dueland RT. 1981. Stress
fields in the unplated and plated canine femur calculated
from in vivo strain measurements. J Biomech 14:63–70.
Davenport RK. 1967. The orang-utan in Sabah. Folia Primatol
Demes B, Qin YX, Stern JT, Larson SG, Rubin CT. 2001. Patterns of strain in the macaque tibia during functional activity.
Am J Phys Anthropol 116:257–265.
Doran DM. 1996. Comparative positional behavior of the African apes. In: McGrew MC, Marchant LF, Nishida T, editors.
Great ape societies. Cambridge: Cambridge University Press.
p 213–224.
Doran DM. 1997. Ontogeny of locomotion in mountain gorillas
and chimpanzees. J Hum Evol 32:323–344.
Drapeau MSM. 2008. Articular morphology of the proximal ulna
in extant and fossil hominoids and hominins. J Hum Evol
Duncan AS, Kappelman J, Shapiro LJ. 1994. Metatarsophalangeal joint function and positional behavior in Australopithecus
afarensis. Am J Phys Anthropol 93:67–81.
Egi N. 1999. Functional morphology of the limb bones in recent
carnivores and its application to North American Hyaenodontid Creodonta, Ph.D. thesis. Baltimore: Johns Hopkins University.
Egi N. 2001. Body mass estimates in extinct mammals from
limb bone dimensions: the case of North American hyaenodontids. Paleontology 44:497–528.
Godfrey LR, Sutherland M, Boy D, Gomberg N. 1991. Scaling of
limb joint surface areas in anthropoid primates and other
mammals. J Zool Lond 223:603–625.
Griffin NL. 2009. Comparative forefoot kinematics and bone
architecture in extant hominids. Ph.D. dissertation, George
Washington University.
Griffin NL, Richmond BG. 2005. Cross-sectional geometry of the
human forefoot. Bone 37:253–260.
Griffin NL, Richmond BG. 2010. Joint orientation and function
in great ape and human proximal pedal phalanges. Am J
Phys Anthropol 144:116–123.
Hamrick MW, McPherron AC, Lovejoy CO, Hudson J. 2000.
Femoral morphology and cross-sectional geometry of adult
myostatin-deficient mice. Bone 3:343–349.
Hofman MA. 1988. Allometric scaling in palaeontology: a critical
survey. Hum Evol 3:177–188.
Latimer B, Lovejoy CO. 1990a. Hallucal tarsometatarsal joint in
Australopithecus afarensis. Am J Phys Anthropol 82:125–133.
Latimer B, Lovejoy CO. 1990b. Metatarsophalangeal joints of
Australopithecus afarensis. Am J Phys Anthropol 83:13–23.
Lovejoy CO, Latimer B, Suwa G, Asfaw B, White TD. 2009.
Combining prehension and propoulasion: the foot of Ardipitehcus ramidus. Science 326:72e1–72e8.
Marchi D. 2005. The cross-sectional geometry of the hand and
foot bones of the Hominoidea and its relationship to locomotor
behavior. J Hum Evol 49:743–761.
Marchi D. 2007. Relative strength of the tibia and fibula and
locomotor behavior in hominoids. J Hum Evol 53:647–655.
Marchi D, Borgognini-Tarli SM. 2004. Cross-sectional geometry
of the limb bones of the Hominoidea by biplanar radiography
and moulding techniques. J Anthropol Sci 82:89–102.
Nagurka ML, Hayes WC. 1980. An interactive graphics package
for calculating cross-sectional properties of complex shapes. J
Biomech 13:59–64.
O’Neill MC, Ruff CB. 2004. Estimating human long bone crosssectional geometric properties: a comparison of noninvasive
methods. J Hum Evol 47:221–235.
Proctor DJ, Broadfield D, Proctor K. 2008. Quantitative threedimentional shape analysis of the proximal hallucial metatar-
sal articular surface in Homo, Pan, Gorilla, and Hylobates.
Am J Phys Anthropol 135:216–224.
Rafferty KL. 1996. Joint design in primates: external and subarticular properties in relation to body size and locomotor
behavior, Ph.D. thesis, Johns Hopkins University.
Rafferty KL, Ruff CB. 1994. Articular structure and function in
Hylobates, Colobus, and Papio. Am J Phys Anthropol 94:395–
Remis M. 1993. The positional behavior of lowland gorillas
(Gorilla gorilla gorilla) in the Dzanga-Sangha Reserve,
Central African Republic. Am J Primatol 30:344 (abstract).
Remis MJ. 1995. Effects of body size and social context on the
arboreal activities of lowland gorillas in the Central African
Republic. Am J Phys Anthropol 97:413–433.
Rose MD. 1988. Functional anatomy of the cheiridia. In:
Schwartz JH, editor. Orang-utan biology. New York: Oxford
University Press. p 299–310.
Rubin CT, Lanyon LE. 1982. Limb mechanics as a function of
speed and gait: a study of functional strains in the radius and
tibia of horse and dog. J Exp Biol 101:187–211.
Ruff CB. 1995. Biomechanics of the hip at birth in early Homo.
Am J Phys Anthropol 98:527–574.
Ruff CB. 2000. Body size, body shape, and long bone strength in
modern humans. J Hum Evol 38:269–290.
Ruff CB. 2002. Long bone articular and diaphyseal structure in
Old World monkeys and apes. I: locomotor effects. Am J Phys
Anthropol 119:305–342.
Ruff CB, Walker A, Trinkaus E. 1994. Postcranial robusticity in
Homo. III: Ontogeny. Am J Phys Anthropol 93:35–54.
Runestad JA. 1997. Postcranial adaptations for climbing in Loridae (Primates). J Zool (Lond) 242:261–290.
Schultz AH. 1963. Relations between the length of the main
parts of the foot skeleton in primates. Folia Primatol 1:150–
Stern JT, Susman RL. 1983. The locomotor anatomy of Australopithecus afarensis. Am J Phys Anthropol 60:279–317.
Susman RL. 1983. Evolution of the human foot: evidence from
Plio-Pleistocene hominids. Foot Ankle 3:365–376.
Susman RL, Brain TM. 1988. New first metatarsal (SKX 5017)
from Swartkrans and the gait of Paranthropus robustus. Am
J Phys Anthropol 77:7–15.
Susman RL, de Ruiter DJ. 2004. New hominin first metatarsal
(SK 1813) from Swartkrans. J Hum Evol 47:171–181.
Susman RL, Stern JT, Jungers WL. 1984. Arboreality and
bipedality in the Hadar hominids. Folia Primatol 43:113–156.
Tuttle RH. 1970. Postural, propulsive, and prehensile capabilities in the cheiridia of chimpanzee and other great apes. In:
Bourne GH, editor. The chimpanzee, Vol.2. Basel/New York:
Karger. p 167–253.
Tuttle RH, Hallgrimsson B, Stein T. 1998. Heel, squat, stand,
stride. Function and evolution of hominoid feet. In: Strasser
E, Fleagle J, Rosenberger A, McHenry H, editors. Primate
locomotion. Recent advances. New York: Plenum. p 435–448.
Wang WJ, Crompton RH. 2004. Analysis of the human and ape
foot during bipedal standing with implications for the evolution of the foot. J Biomech 37:1831–1836.
White TD, Asfaw B, Beyene Y, Haile-Selassie Y, Lovejoy CO,
Suwa G, WoldeGabriel G. 2009. Ardipithecus ramidus and
the paleobiology of early hominids. Science 326:75–86.
Wunderlich RE. 1999. Pedal form and plantar pressure distribution in anthropoid primates. Ph.D. dissertation. State University of New York at Stony Brook, New York.
Zipfel B, DeSilva JM, Kidd RS. 2009. Earliest complete hominin
fifth metatarsals—implications for the evolution of the lateral
column of the foot. Am J Phys Anthropol 140:532–545.
American Journal of Physical Anthropology
Без категории
Размер файла
378 Кб
ape, proportional, diaphyseal, human, articular, great, metatarsal
Пожаловаться на содержимое документа