вход по аккаунту


Effect of mobility on femur midshaft external shape and robusticity.

код для вставкиСкачать
Effect of Mobility on Femur Midshaft External
Shape and Robusticity
Daniel J. Wescott*
Department of Anthropology, University of Missouri-Columbia, Columbia, Missouri 65211
cross-sectional shape; biomechanics; long bones; subsistence strategy
This study investigates differences in
femur midshaft shape, robusticity, and sexual dimorphism
derived from external measurements between a broad
range of prehistoric and historic North American populations with different subsistence strategies and inferred
levels of mobility. The sample was divided into six groups
to test whether observed femur midshaft variables follow
the patterns predicted based on archaeologically and his-
torically determined subsistence and mobility data. The
results suggest significant variation in femur midshaft
shape and robusticity in all populations, and that inferred
mobility levels do not correspond consistently with femur
midshaft structure in either males or females. Results do,
however, support the prediction that sexual dimorphism is
generally greater in more mobile populations. Am J Phys
Anthropol 130:201–213, 2006. V 2005 Wiley-Liss, Inc.
Mobility behavior in prehistoric populations provides
information on numerous aspects of culture and society,
including subsistence strategy, sociopolitical organization, trade, demography, enculturation, and territoriality
(Kelly, 1992). One avenue that has been used frequently
to detect the level of terrestrial logistic mobility (TLM;
daily distance covered by an individual or group from
the residence and back) within and between populations
is the biomechanical analysis of femur midshaft structure. Many researchers (e.g., Holt, 2003; Larsen, 1997;
Larsen et al., 1995; Ruff, 1987, 1994, 1999; Stock and
Pfeiffer, 2001, 2004) argued that the TLM of a population can be estimated using the ratio of the anteroposterior (Ix) to mediolateral (Iy) second moments of area of
the femur midshaft. Larsen et al. (1996) and Ruff (1987,
2000) contended that the Ix/Iy ratio, or ‘‘mobility index,’’
is greater in hunter-gatherers than in horticulturalists,
since the former subsistence practice requires more longdistance travel, especially among males, than does the
latter. Furthermore, hunting-gathering requires a higher
degree of sexual division of labor than does horticulture,
which results in a greater degree of femur midshaft sexual dimorphism among hunter-gatherers (Bridges, 1995;
Larsen, 1997; Ruff, 1987, 2000).
Studies showed that during activities such as running
or climbing, contraction of the hamstrings and quadriceps muscles causes high anteroposterior (a-p) bending
loads from the femur midshaft to the tibia midshaft
(Morrison, 1968, 1969, 1970). As a result of these activityrelated mechanical loads, modeling modifies the femoral
midshaft from its basic circular cross-section to one that
is a-p elongated. Therefore, groups with low TLM should
have less directional loading and thus greater femoral
midshaft circularity than highly mobile populations.
Likewise, females, who are generally less mobile than
males, should have more circular femora than males,
especially in hunter-gatherer populations. This is the
basis for using femur midshaft shape to estimate mobility patterns.
Several researchers, using archaeological evidence to
determine TLM patterns, contended there is a strong correlation between lower limb structure and TLM within
specific populations (Holt, 2003; Ruff, 1987; Ruff and
Larsen, 1990; Ruff et al., 1993; Stock and Pfeiffer, 2001,
2004). Specifically, Stock and Pfeiffer (2001) analyzed differences in upper and lower limb bone strength between
a highly mobile Later Stone Age African sample and a
marine mobile (watercraft) sample from the Andaman
Islands. They found that the African population exhibited significantly stronger lower limb bones, while the
Andaman Islanders displayed more robust upper limb
bones, leading them to claim support for the tenet that
mobility patterns can be reconstructed using limb bone
morphology. More recently, Stock and Pfeiffer (2004)
examined differences in long bone structure between forest and coastal Later Stone Age foragers in South Africa,
and found that males in both biomes were extremely
mobile, but females in the forest biome were more mobile
than coastal females. They concluded that sex differences in mobility levels vary among hunter-gatherers,
depending on environmental resources. Holt (2003) used
femoral midshaft cross-sectional geometry to test the
archaeological hypothesis that mobility decreased from
the Early Upper Paleolithic to Mesolithic periods in
Europe. Her results indicated that femur midshaft bending strength decreased through time, suggesting greater
sedentism in the later European populations. All these
studies support the association between long bone structure and TLM.
The use of femoral midshaft robustness/strength and
shape to infer levels of logistic mobility has not gone
without criticism (e.g., Bertram and Swartz, 1991; Frost,
1997; Jurmain, 1999; Lovejoy et al., 2003; Rockhold,
1998). The main objection to interpreting behavior from
C 2005
*Correspondence to: Daniel J. Wescott, Department of Anthropology, University of Missouri-Columbia, 107 Swallow Hall, Columbia,
MO 65211. E-mail:
Received 11 August 2004; accepted 19 April 2005.
DOI 10.1002/ajpa.20316
Published online 19 December 2005 in Wiley InterScience
TABLE 1. Sample characteristics by subsistence adaptation
Broad-spectrum hunter-gatherer
Woodland hunter-gatherer
Marine hunter-gatherer
Equestrian hunter-gatherer
Incipient horticulturalist
Village horticulturalist-hunter
Maize-dependent horticulturalist
Early modern industrialist
Late modern industrialist
Cultural-temporal provenience
Female N
Male N
Paleoamerican, Archaic, Great Basin foragers
Early and Middle Woodland
Coahuiltecan, Karankawa
Historic Great Plains equestrian
Late Woodland, Fremont
Great Plains prehistoric and historic horticulturalist
Southwest Pueblo, Texas Caddo
Pre-1900 American
Post-1900 American
biomechanical evidence is that there are numerous
cultural, environmental, and biological factors affecting
long bone geometry that have not been controlled for in
biomechanical studies, and that many of these factors
may have a greater influence on bone morphology than
mechanical loads. Therefore, the effects of mobility on
long bone cross-sectional strength and shape are probably not universal.
Most studies examining the effect of TLM on femur
morphology focused on changes occurring among related
groups within a single region. This study differs in that
it uses femur midshaft robustness and shape and sexual
dimorphism derived from external measurements to test
predictions of the biomechanical model using populations
from a number of regions throughout North America.
Only a few other studies examined a broad range of populations using a biomechanical approach. For example,
Ruff (1987) examined femur midshaft external a-p and
mediolateral (m.l) breadth ratios (a-p/m-l) in a variety of
archaeological and modern groups, and found a reduction in sexual dimorphism with the transition from hunting-gathering to horticulture, and a further reduction as
populations became industrialized. He asserted that this
reduction indicates a change in sex-specific mobility patterns. Later, Ruff (1999) compared sex differences in Ix/
Iy ratio between hunter-gatherers and horticulturalists
in the Southeast, Southwest, and Great Plains, finding
that nearly all hunter-gatherer groups exhibited greater
sexual dimorphism in femur shape than horticultural
populations. Collier (1989) compared Australian Aborigines to other modern groups, and found that Australian
Aborigines have narrow diaphyses relative to their
length (i.e., more gracile femora) compared to RomanoBritish, American white, Arikara Native American, and
Eskimo populations. He concluded that the gracility of
Australian Aborigines was related to genetic and economic rather than behavioral factors. However, Collier
(1989) found that Aboriginal males have an a-p/m-l ratio
greater than urban populations and similar to other
hunter-gatherers. Pearson (2000) also examined diverse
populations, but his research focused on the effects of climate and lifestyle (mobile and sedentary) on long bone
robusticity. His results suggest that robusticity varies as
greatly with climate as with lifestyle. While all these
studies investigated a broad range of populations, the
study by Ruff (1987) was limited to examining differences in sexual dimorphism among three broad subsistence strategies (hunter-gatherers, horticulturalists,
and industrialists), Collier (1989) concentrated on the
robusticity of Australian Aborigines, and Pearson (2000)
focused only on differences in robusticity among two
broadly defined lifestyles (mobile hunter-gatherers and
sedentary agriculturalists and industrialists). In the
present study, sexual dimorphism in femur midshaft
structure and sex-specific differences in femoral midshaft
shape and robusticity are compared among a wide range
of North American populations with six different TLM
Based on the predictions of the biomechanical model,
males in groups with high TLM should exhibit more
robust femora, with a greater a-p/m-l ratio (Ruff, 1987).
Sexual dimorphism should also be greater in highly
mobile groups, since males are more likely to travel long
distances during subsistence activities than females
(Ruff, 1987; Stock and Pfeiffer, 2004). In order to test
these predictions, two specific questions are addressed in
this study: 1) Do males in populations with high TLM
consistently exhibit greater femoral midshaft robusticity
and greater a-p expansion than males in less mobile populations? and 2) Is there more sexual dimorphism in
groups with high TLM than in those that are less
A large sample of prehistoric, early historic, and modern groups from North America was used (Table 1). Femoral external dimensional data were obtained from the
University of Tennessee/Smithsonian Institution (UT/SI)
postcranial database (Wescott, 2001), the Terry Collection (Trotter, 1981), and the Forensic Data Bank (FDB)
(Ousley and Jantz, 1998). The UT/SI database is composed of individuals dating from the Paleoamerican to
historic periods (primarily Native Americans) from the
Great Basin, Great Plains, Southwest, Southeast, Midwest Prairie, and Texas Gulf Coast of the United States.
However, samples from the Great Plains predominate
the sample. Age and sex were estimated for individuals
in this database using standard osteological methods
(Bass, 1995; Ubelaker, 1989). The Terry Collection and
FDB contain data on 19th and 20th century Americans
(primarily American blacks and whites) with known biographical information. Only adult specimens (i.e., fused
epiphyses) were used in this study.
Size and shape variables
Femur maximum length (FML), maximum head diameter (FHD), midshaft anteroposterior diameter (APM),
and midshaft mediolateral diameter (MLM) were collected and recorded to the nearest millimeter. External
dimensions were used to calculate: 1) femur midshaft
diaphyseal shape (FMS) by dividing the a-p by the m-l
diameter (APM/MLM), and 2) femur midshaft robusticity
American Journal of Physical Anthropology—DOI 10.1002/ajpa
(FMR) by adding midshaft diameters, multiplying the
sum by 100 and dividing by femur head size (100 *
(APM þ MLM)/FHD). These indices approximately reflect the cross-sectional properties of the bone, and provide a measure of the bone’s strength and average resistance to bending (Cole, 1994). An FMS ratio of 1.0 indicates that the diaphysis is the same size in both planes
(circular). A ratio greater than 1.0 indicates that the
cross-section is elongated along the a-p plane, and suggests that the bone was subjected to greater bending
loads in that direction. Greater bending loads in the m-l
direction are indicated if the ratio is less than 1.0 (Ruff,
1987). The FMR index offers a measure of bone strength
and approximates the ratio of torsional strength (J) divided by body mass used in many biomechanical studies
(Cole, 1994; Pearson, 2000).
Traditionally, diaphyseal thickness is standardized by
bone length when calculating the robusticity index, but
this does not control for differences in body mass when
comparing populations of different body proportions.
Femur head size has a much greater correlation with
body weight than does femur length (Lieberman et al.,
2001; McHenry, 1988; Ruff et al., 1991), and several
authors (Lieberman et al., 2001; Ruff, 2000) suggested
that one way to control for differences in body size is to
standardize cross-sectional properties by articular size
(e.g., FHD). For this reason, femur robusticity is standardized here by femur head diameter and not length.
Cross-sectional geometric data obtained from physical
sections of the bone or computed tomography (CT) scans
are preferred over external dimensions for biomechanical
analyses (Larsen, 1997; Ruff, 1987, 2000, 2002; Trinkaus
and Ruff, 2000), but it is not feasible to obtain the large
sample size of cross-sectional data needed to examine a
broad range of populations. While possibly not as precise
as true cross-sectional data, the examination of femur
midshaft shape and robusticity based on external measurements does provide a reliable substitute for cross-sectional data, and was successfully used to draw behavioral conclusions that were also supported by cross-sectional data (Bridges et al., 2000; Cole, 1994; Larsen,
1981; Ruff, 1987; Wescott, 2001).
Mobility scores
To examine the effects of mobility on long bone crosssectional morphology, the sample was subdivided into
nine subsistence strategies with varying degrees of
TLM inferred from archaeological and historical evidence
(Table 1): broad-spectrum hunter-gatherers (BHG), woodland hunter-gatherers (WHG), marine hunter-gatherers
(MHG), equestrian hunter-gatherers (EHG), incipient
horticulturalists (IHH), village horticulturalists-hunters
(VHH), maize-dependent horticulturalists (MDH), early
modern industrialists (EMI), and late modern industrialists (LMI). Each of the nine adaptations was scored for
TLM level (Table 2), with higher numbers representing a
greater level of TLM than lower numbers.
The BHG adaptation is represented by the Paleoamericans, Archaic groups, and Great Basin hunter-gatherers. A BHG adaptation requires a highly mobile lifestyle
compared to other groups (Hemphill and Larsen, 1999;
Hofman et al., 1996), and therefore received a mobility
score of 5. The Archaic sample is derived primarily from
the Great Plains. Archaeological evidence suggests that
Archaic bands exploited numerous ecological zones and
traveled great distances to procure food and raw materi-
TABLE 2. Mobility (TLM) scores for each
subsistence adaptation
Woodland hunter-gatherer
Marine hunter-gatherer
Equestrian hunter-gatherer
Incipient horticulturalist
Village horticulturalist-hunter
Early modern industrialist
Late modern industrialist
Very high
Very low
Extremely low
als for tool production (Frison, 1998). Archaic population
densities also suggest that these hunter-gatherers were
extremely nomadic (Carlson, 1998). Likewise, Ruff (1999,
2000) found that BHG populations from the Great Basin
exhibit a level of sexual dimorphism in femur shape
more similar to Early European Upper Paleolithic samples than to other North American hunter-gatherers
(Woodland). Ruff (1999, 2000) argued that this similarity
is a consequence of mobility patterns.
The WHG are represented by Early and Middle variants of the Woodland tradition, primarily from the
Great Plains (Wescott, 2001), while the Karankawa and
Coahuiltecan of the Texas Gulf Coast (Meadows Jantz
et al., 2001) are used to represent the MHG subsistence
strategy. Both the WHG and MHG were slightly less
mobile than the BHG, and received a mobility score of 4.
Big-game (bison) hunting was important for the WHG in
the Great Plains, but a focus on smaller, more local,
game animals is also apparent (Johnson and Johnson,
1998), especially in the Central Plains and Southern
Plains, where bison were less numerous (Blakeslee,
1994). Horticulture was becoming increasingly important
during the Woodland period, but appears to have been
relatively unimportant in the Great Plains until the
beginning of the Late Woodland (Johnson, 2001). The
use of local lithic materials by Woodland groups also
suggests that these groups were less mobile than earlier
hunter-gathers. Likewise, the MHG fished, hunted marine animals, and gathered local wild flora, and were
shown archaeologically and ethnographically to be relatively mobile but focused on local resources (Meadows
Jantz et al., 2001). The WHG and MHG lifestyles would
likely require less logistical mobility than that seen in
BHG, but greater TLM than horticulturalists or industrialists.
The EHG adaptation was practiced by the protohistoric and historic equestrian groups of the Plains (i.e.,
Blackfoot, Comanche, Crow, and Sioux). Great Plains
equestrians are characterized as highly mobile bands
that relied primarily on horses for logistic and residential mobility (Wedel and Frison, 2001). While some Great
Plains tribes had much greater access to horses than
others, the Sioux, Comanche, Blackfoot, and Crow that
make up most of the EHG sample had abundant horses.
For these tribes, the horse was so important that they
adapted their cultures to it (Carlson, 1998). Because
these groups relied primarily on the horse for logistical
mobility, they were given a mobility score of 2.
The Late Plains Woodland tradition and Fremont culture characterize the IHH adaptation type (Blakeslee,
1994; Cultrain and Stafford, 1999). Late Woodland
American Journal of Physical Anthropology—DOI 10.1002/ajpa
groups in the Great Plains supplemented wild seed
plants, nuts, and game animals with cultigens (Adair,
1996; Blakeslee, 1994; Johnson and Johnson, 1998). The
archaeological record suggests that they were more sedentary than earlier Woodland groups, but more mobile
than later horticultural groups. The Fremont are generally considered maize horticulturalists, but they were
probably much less dependent on domestic cultigens
than the Anasazi and other maize-dependent Southwestern groups (Cultrain and Stafford, 1999). Stable isotope
analysis of Fremont skeletons from the Salt Lake region
even revealed that some groups subsisted on diets primarily comprised of wild plants (Cultrain and Stafford,
1999). This suggests that the Fremont had a mobility
pattern similar to Late Woodland populations on the
Great Plains. The mixed subsistence strategy of the IHH
required less mobility than seen in hunter-gatherers, but
greater mobility compared to equestrian nomads or sedentary maize-dependent horticulturalists (Wood, 1998).
Therefore, this subgroup was assigned a TLM score of 3.
The VHH adaptation was practiced by the prehistoric
Plains Village and Middle Missouri traditions and the
historic Plains horticultural tribes. These groups were
primarily horticulturalists, but are archaeologically and
historically known as only semisedentary populations.
Besides growing food crops, the VHH groups also hunted
bison at least twice a year (Blakeslee, 1994; Parks,
2001). The bison hunts generally involved the entire village and frequent travel by foot of over 160 km (Blakeslee,
1994). As a result, the VHH are given a mobility score of
3. Archaeological and historical accounts of VHH groups
indicate that their level of mobility is consistent with
that of IHH populations, but probably greater than that
of equestrian nomads.
Maize-dependent horticulturalists (MDH) are represented by Pueblo groups from the Southwest and the
Texas Caddo. These groups were sedentary and probably
not as logistically mobile as the BHG, WHG, MHG, IHH,
or VHH, and therefore are given a mobility score of 2.
Southwest horticulturalists are known archaeologically
and historically as among the most sedentary Native
American populations, and were often used by Ruff
(1987, 1999, 2000) as sedentary populations. The
archaeological and historical record also suggests that
the Caddo were sedentary horticulturalists who grew a
variety of crops and hunted small local fauna. The
Caddo and Southwest horticulturalists are grouped
because they practiced a very similar subsistence strategy and even participated in a common trade network.
While the MDH groups supplemented their diet by hunting local fauna, they most likely did not travel any farther by foot than the EHG. Furthermore, unlike the
VHH, obtaining bison does not appear to have been
important in these groups until the introduction of
horses and firearms (Drass, 1998). Therefore, the MDH
are given a TLM score of 2.
Nineteenth century Americans from the Terry Collection and historic archaeological sites represent the EMI
adaptation. The occupational practices of the EMI are
highly variable and include slaves, laborers, missionaries, and soldiers. The EMI are also a very genetically
heterogeneous group. However, they are used in this
study as a single group because they represent a population with low TLM. Furthermore, because Ruff (1987),
Collier (1989), and Pearson (2000) also used the EMI as
examples of very sedentary groups, it is important that
they be included in this study for comparison. The EMI
are given a mobility score of 1 to indicate the low level of
mobility in this group compared to hunter-gatherer and
horticultural populations (Ruff, 1987).
Finally, the LMI pattern is typified by 20th century
American groups that exhibit extremely low mobility
due to modern forms of transportation. The sample is
drawn primarily from the Forensic Data Bank and designated as Americans with birthdates after AD 1900. Like
the EMI, there is great genetic and occupational variation in the LMI sample. As a group they are far less
mobile than the EMI because of their reliance on modern
forms of transportation. Rockhold (1998) showed a significant secular change in femur shape and robusticity
among modern Americans, suggesting that the LMI
should not be grouped with the EMI. The LMI are given
the lowest TLM score possible, i.e., 0.
Data analysis
The femoral data were used to compare sexual dimorphism and sex-specific differences among groups with
different mobility scores. Variation in degree of sexual
dimorphism among subsistence groups was compared to
help sort out sex differences in TLM, while sex-specific
comparisons were used to examine differences in mobility without the potential effect of sexual dimorphism.
Group comparisons were performed using least
squares regression and analysis of variance (ANOVA).
Regression reveals the relationship between long bone
variables and the ordinal codes for mobility. ANOVA was
used to examine differences between groups without
regard to the ordinal order of the mobility level. Differences between populations were subsequently contrasted
using Tukey’s multiple comparison test, that assesses
which groups differ statistically (Sokal and Rohlf, 2000).
Effect of mobility on sexual dimorphism
Statistically significant (P 0.05) sexual dimorphism
is present in all raw osteometric dimensions (FML,
FHD, APM, and MLM) and midshaft structural variables (FMS and FMR) (Table 3). The absolute greater size
of males is reflected in larger mean values for FML,
FHD, APM, and MLM. There is a significant mobility/
sex interaction for FMS, with sexual dimorphism generally increasing with mobility level. However, sexual
dimorphism in FMS only reaches statistical significance
in mobility levels 0, 2, 3, and 5 (Table 3; Fig. 1). In the
least mobile group (0), females exhibit a significantly
greater a-p/m-l ratio than males, and in the most mobile
group (5), males have a significantly greater FMS than
females. Femur midshaft robusticity also shows a significant mobility/sex interaction. Females are more robust
than males in the femur midshaft among the least
mobile groups (0, 1, and 2), while males are more robust
than females in the more mobile groups (Table 3; Fig. 1).
However, sexual dimorphism is only significant for two
of the five groups (1 and 2).
Effect of mobility on sex-specific samples
Femur shape (FMS) and robusticity (FMR) do not
show any strong correlation with TLM scoress in either
sex (Table 4; Figs. 2, 3). For males, there is a significant overall mobility effect for FMS, but paired-group
comparisons show that the pattern does not follow mobility levels. Furthermore, only the EMI group (mobility
American Journal of Physical Anthropology—DOI 10.1002/ajpa
TABLE 3. Descriptive statistics by mobility level
Significant overall sex effect.
Significant overall mobility effect in combined sex sample.
Significant mobility/sex interaction.
slightly more robust than MDH groups, but not significantly. Among hunter-gatherer males, WHG groups are
more robust than BHG populations. The LMI sample is
also more robust than the EMI group. Among females,
there is no apparent within-subsistence strategy pattern.
Fig. 1. Sexual dimorphism in femur midshaft shape and
robusticity by mobility level. Percent sex difference: [(male
mean female mean)/female mean] 3 100. *Statistically significant (P 0.05) sexual dimorphism.
score 1) is significantly different from other groups.
There is also a significant overall mobility effect for FMS
in females, but only because the LMI differ from other
groups. Femur midshaft robusticity is nearly significant
among males, but again the pattern did not conform to
the predictions of the biomechanical model. There is a
significant mobility effect for FMR in females, but like
males, group means do not correspond to mobility scores.
An examination of changes through time within subsistence strategies also does not show a significant
change in FMS or FMR due to mobility level. Among
males, BHG have a greater FMS than later WHG, and
IHH and VHH have a greater index than later MDH,
but none of these differences reach the level of statistical
significance. Likewise, there is no significant pattern in
FMR. Within horticulturalists, the IHH and VHH are
Biomechanical analyses of long bone morphology have
been used extensively by anthropologists to reconstruct
subsistence-based mobility patterns in prehistoric and
fossil populations (e.g., Brock and Ruff, 1988; Holt, 2003;
Larsen et al., 1995, 1996; Ruff and Hayes, 1983; Ruff,
1987, 1994, 1999; Stock and Pfeiffer, 2004). According to
the biomechanical model, sedentary groups should
exhibit more circular and gracile femoral diaphyses and
less sexual dimorphism in femur midshaft morphology
than highly mobile populations. However, these morphological differences were not found consistently in all populations (e.g., Bridges, 1989; Bridges et al., 2000), suggesting that the effect of mobility on femur midshaft
structure may not be universal. Genetic, climatic, terrain
type, other occupational factors, and age may influence
femur midshaft shape and robusticity as much or more
than mobility patterns. Therefore, this study was designed to investigate the effect of mobility on femur midshaft shape, robustness, and sexual dimorphism by
examining variation among groups with different TLM
Use of external measurements
Femur midshaft diaphyseal shape and robusticity
were derived using external dimensions in this study,
which several researchers claimed are insufficient for
biomechanical studies because they do not take into
account the internal architecture of the bone (Larsen,
1997; Ruff, 1987, 2000; Trinkaus and Ruff, 2000). However, cross-sectional properties are influenced by exter-
American Journal of Physical Anthropology—DOI 10.1002/ajpa
1B > 4BC > 5BC > 2BC > 3C
0A > 1AB > 5AB > 4AB > 2B
3B > 1B > 2B > 4B > 5B
1AB > 5AB > 3AB > 2AB > 0B
3B > 2B > 1B > 4B > 5B
1AB > 0A > 4AB > 5AB > 3B
Fig. 2. Midshaft femur shape ratio by mobility level for
females. Mean value is indicated; whiskers extend 1 standard
deviation. See Table 4 for paired comparisons.
Groups with same superscript letter do not differ significantly.
5B > 3B > 2B
> 3BC > 5CD > 2D
> 5AB > 1B > 2B
3B > 5B > 2B
0A > 4AB > 1B
3A > 2A > 1A
4B > 1B >
1AB > 4BC
3AB > 4AB
4A > 1A >
3A > 2A >
5A > 0A >
Paired group differences1
TABLE 4. Effect of mobility on femoral size, shape, and robusticity
Paired group differences1
Fig. 3. Midshaft femur shape ratio by mobility level for
males. Mean value is indicated; whiskers extend 1 standard
deviation. See Table 4 for paired comparisons.
nal dimensions, leading numerous authors (Bridges
et al., 2000; Jungers and Minns, 1979; Pearson, 2000;
Rockhold, 1998; Ruff, 1987; Wescott, 2001) to argue that
femur mishaft a-p and m-l dimensions can be used as
reliable substitutes for some cross-sectional properties in
biomechanical studies. Ruff (1987) indicated that even
though external dimensions do not provide information
on cortical bone thickness, the same general patterns of
femur cross-sectional shape should be expected for indices based on external breadth dimensions as for those
derived from geometric properties. Bridges et al. (2000)
also found that the femur midshaft shape index calculated from external dimensions and cross-sectional values showed nearly the same patterns of change among
archaeological populations from Illinois. Additionally,
several studies using external dimensions to reconstruct
behavior were supported by cross-sectional data. Larsen
(1981) found a significant difference in a-p and m-l external dimensions between prehistoric hunter-gatherers
and horticulturalists in Georgia that was later supported
American Journal of Physical Anthropology—DOI 10.1002/ajpa
by cross-sectional data on these same groups (Ruff et al.,
1984). Cole (1994) used external measurements to examine changes in long bone shape, robusticity, and sexual
dimorphism with the transition from hunting-gathering
to horticulture in the Northern Plains. His results were
nearly identical to those of Ruff (1994) using cross-sectional data on the same groups. This indicates that
external dimensions can be used with validity for addressing many questions regarding long bone structure.
Additionally, several studies explicitly tested the
strength of the relationship between external dimensions
and cross-sectional properties. Jungers and Minns (1979)
compared the shape index derived from external measurements to Ix/Iy obtained from CT scans on three different primate species. They found a highly significant correlation (r2 ¼ 0.986) between femur midshaft index and
Ix/Iy, and concluded that external dimensions offer a
means of testing hypotheses regarding mechanical morphology (Jungers and Minns, 1979). However, it should
be pointed out that Jungers and Minns (1979) used three
different species and only examined a total of seven
specimens. It is possible that the high correlation they
found may not persist within a single species using a
larger sample size. Other studies based on larger human
samples showed a strong correlation between external
dimensions and cross-sectional properties in humans.
Pearson (2000) examined the relationship between J
(estimation of torsional strength) predicted from external
measurements and actual J derived from cross-sectional
CT scans. His results suggest a strong relationship
between the two properties, with a squared correlation
of 0.898. Likewise, Rockhold (1998) compared external
indices (a-p/m-l) and cross-sectional properties (Ix/Iy)
from 131 directly sectioned femora (data from Ruff,
1994), and found a squared correlation of 0.762. Wescott
(2001) also found high correlations between FMS and Ix/Iy
(r2 ¼ 0.690) and between J and estimated J (r2 ¼ 0.829),
using over 400 CT cross-sectioned femora. The relatively
high correlation between FMS and Ix/Iy in these studies
indicates that between 70–98% of the variation within
and between populations in Ix/Iy can be captured by
external dimensions. External measurements do not
always explain all of the variation captured by crosssectional data, and therefore may be more likely to produce overlap between groups (Ruff, 2002). However, as
these studies empirically demonstrated, external dimensions are reliable for testing many biomechanical
This assertion does not mean that estimations of
cross-sectional geometric properties based on external
dimensions should be compared to those derived from
other more direct methods (e.g., direct sectioning, CT
scans, or latex cast method), because they do not provide
a high degree of agreement. For example, Stock (2002a)
pointed out that correlation coefficients derived through
regression analysis may conceal significant differences
between measurements obtained using different methods. To estimate how accurately external dimension indices (FMS) can reproduce ‘‘true’’ shape derived from
cross-sectional data, the mean directional ([(estimated
shape true shape)/true shape] * 100) and absolute
(|[(estimated shape true shape)/true shape]| * 100)
percent prediction errors between the two methods were
calculated for 150 CT-derived femoral cross-sections, following O’Neill and Ruff (2004). The mean directional
and absolute percentage errors are 1.10 (SD ¼ 12.55)
and 9.72 (SD ¼ 7.98), respectively. This shows that esti-
mations of femur shape (FMS) based on external dimensions consistently underestimate the true cross-sectional
property (Ix/Iy). While this error is high, it is similar to
the error found by O’Neill and Ruff (2004) for the ellipse
model method using a-p and m-l radiographs to determine cortical thickness, a procedure that has been used
extensively to address questions in biomechanical studies. Therefore, while external dimensions cannot be used
to accurately estimate cross-sectional properties, they
appear to provide a reasonable and cost-effective alternative for testing biomechanical hypotheses, but the results cannot be directly compared to those based on
cross-sectional data.
Effect of mobility on sexual dimorphism
The purpose of this study was to test whether
observed femoral midshaft morphology determined from
external dimensions follows the patterns predicted for
populations based on archaeological and ethnographical
mobility data. The biomechanical approach predicts that
high TLM should affect midshaft shape (FMS), resulting
in greater a-p expansion of the shaft, especially among
males, and that sexual dimorphism should be greater in
more mobile populations. Results show that sexual
dimorphism in femoral midshaft shape and robusticity is
generally greater in more mobile populations, as predicted, but sex-specific variation in FMS and FMR does
not correspond too well with mobility, even within subsistence groups.
Ruff (1987) found that sexual dimorphism in femur
midshaft shape was greatest among hunting-gathering
populations, was reduced in horticulturalists, and was
nearly nonexistent in industrial groups. His industrial
group was similar to the EMI sample used in this study.
If LMI is removed, the general pattern of FMS and FMR
among hunter-gatherers, horticulturalists, and industrialists found in this study is similar to that observed by
Ruff (1987), and could be interpreted as differences in
mobility. Over the last 150 years, however, Americans
have undergone significant changes in femur length,
midshaft shape, and robusticity (Rockhold, 1998;
Wescott, 2001). Femur midshaft shape (FMS) has
increased significantly through time, especially in
females, due to a decrease in the midshaft mediolateral
diameter (MLM) rather than an increase in anteroposterior diameter (APM) (Rockhold, 1998; Wescott, 2001). If
FMS truly reflects differences in mobility, then these
results are difficult to explain. Because of modern transportation, one would expect the femoral midshafts of
modern Americans to become more circular through time
as a result of decreased a-p bending strength. Since this
is not the case, researchers should use caution when
using femur shape, at least when it is derived from
external measurements, to interpret mobility patterns.
Anthropologists must analyze why and how the a-p/m-l
ratio differs between populations, because factors other
than mobility may play an important role in the overall
shape of the femur at midshaft.
The surprising differences between the LMI and other
groups may be the result of task specialization among
recent populations, but the conclusion that femur external midshaft shape should be used with caution to infer
TLM level is also supported by differences between
northern and southern Great Plains horticulturalists
(Ruff, 1994; Wescott, 2001). Archaeological evidence suggests that southern and northern Plains horticulturalists
American Journal of Physical Anthropology—DOI 10.1002/ajpa
practiced nearly identical subsistence strategies and had
a similar level of TLM (Wood, 1998), yet populations
from the southern Great Plains exhibit a significantly
greater FMS (males, 1.18; females, 1.12) than linguistically related groups farther north (males, 1.11; females,
1.05) (Wescott, 2001). This conclusion was also supported
by Ruff (1994) using cross-sectional properties (Ix/Iy).
Like the LMI, the main difference between the southern
and northern Great Plains horticulturalists is not due to
an increase in the a-p plane, but rather a decrease in
the m-l plane (Wescott, 2001).
One question that arises from these results is, why
are highly mobile hunter-gatherers more sexually dimorphic in FMS and FMR than are less mobile horticulturalists or industrialists? Ruff (1987) argued that declining
sexual dimorphism through time is due to a reduction in
a-p bending strength among males, combined with virtually no change among females. Hunting-gathering males
are more likely to engage in activities (e.g., hunting,
fishing, and trapping) that require long-distance travel,
but as populations changed to a horticultural subsistence
strategy, male activities (e.g., house-building, cooking,
soil preparation, and tending crops) require less TLM
(Ruff, 1987). Activities among females in all types of subsistence strategies require less mobility than among
males (Ruff, 1987).
In general, the results of this study appear to support
the assertion of Ruff (1987), but the difference in FMS
between males and females is not significant for mobility
level 4 (MHG and WHG), suggesting that the trend seen
in sexual dimorphism is open to question. There are several possible explanations for this result. One concerns
the MHG sample. The Karankawa and Coahuiltecan
skeletons used in this study are from a mission cemetery. Isotope analysis and historical records indicate that
these populations primarily practiced a marine huntinggathering lifestyle (Meadows Jantz et al., 2001), but they
may have also participated in mission work, resulting in
a lower TLM than inferred. However, even if the MHG
are removed, the percent sex difference does not change,
remaining at approximately 4%. A second reason could
be that Early and Middle Woodland groups (WHG) were
practicing more horticulture than hunting-gathering,
which would result in a lower level of mobility among
males. However, most archaeological evidence suggests
that horticulture was extremely limited among Early
and Middle Woodland groups, especially on the Great
Plains (Blakeslee, 1994). Another explanation is that the
mobility levels used in this study are too finely seriated.
It is difficult to determine with precision the TLM level
of groups such as equestrian hunter-gatherers. To examine this possibility, the study sample was subdivided into
three less arbitrary mobility groups. The highly mobile
BHG and WHG were combined and compared to the sedentary MDH and the very sedentary EMI. The results
show that the hunter-gatherers and horticulturalists
exhibit statistically significant sexual dimorphism, while
the modern industrialists do not. Hunter-gatherers exhibit nearly 9% (P < 0.0001) sexual dimorphism in FMS,
as opposed to about 5% (P ¼ 0.0004) in horticulturalists
and almost none (P ¼ 0.6344) in EMI. These results are
very consistent with those found by Ruff (1987), but are
also consistent with the results based on six mobility
Of course, even with the apparent relationship
between mobility and degree of sexual dimorphism, it
could be that the trend in sexual dimorphism is not
related to sexual division of labor. Ruff et al. (1993) documented a steady decline in femur midshaft cortical area
and robusticity in Homo from the early Pleistocene to
the late Holocene. Additionally, Holt (2003) examined
changes in mobility from the Early Upper Paleolithic to
the Mesolithic, and found that femur midshaft circularity increased in both sexes as mobility decreased in
Europe. She suggested that a reduction in a-p bending
strength, combined with a low level of sexual dimorphism, reflects a decrease in mobility but no change in
sexual division of labor. However, Ruff (1987) argued
that femur shape is not significantly different between
females of different subsistence strategies, because
female chores almost always require less mobility than
male chores. These conclusions indicate that the trend in
sexual dimorphism may not be related to any specific
activity, but rather to some other environmental, biological, or genetic factors.
Effect of mobility on sex-specific samples
While the degree of sexual dimorphism may be associated with level of mobility, nonsignificant sex-specific differences among mobility groups suggest that at least in
some populations, long-distance travel is not reflected in
femur midshaft shape or robusticity. Again, there are
several possible explanations for these results. Like
before, the finely categorized mobility scores used in this
study may obscure the relationship between mobility
and femur shape, especially when using external dimensional data that may produce more overlap between
groups than cross-sectional data. However, even when
the number of mobility groups is reduced to three, there
is still no significant relationship between mobility and
femur shape (FMS) or robusticity (FMR) in either males
or females. Sedentary populations (EMI) exhibit greater
femur circularity among males, but there are no significant differences between mobile hunter-gatherers (BHG
and WHG) and less mobile horticulturalists (MDH).
Another problem might be associated with comparing
groups practicing different subsistence strategies. Most
biomechanical studies focus on variation between samples practicing different subsistence strategies in the
same region, and Ruff (2000, p. 86) argued that ‘‘populations undergoing direct transitions from one subsistence
strategy to another provide the best controlled tests’’ for
hypotheses about sexual division of labor and mobility
patterns. Stock (2002b), however, claimed that the comparison of groups with different subsistence strategies
may be misleading because of pervasive changes in diet,
disease, and other factors associated with any subsistence transition. Leach (2003), for example, argued that
humans have undergone some of the same changes in
body size and skeletal robusticity with the transition to
horticulture as seen in domesticated animals. For example, there is often a reduction in stature with the transition from hunting-gathering to horticulture in many
regions. This suggests that investigators examining the
effects of mobility should focus on changes within subsistence strategies, as done by Holt (2003) and Stock and
Pfeiffer (2001, 2004).
If changes in mobility within subsistence strategies
are examined in the present study, there appears to be a
slight reduction in femur shape through time within
hunter-gatherers and within horticulturalists among
males (Fig. 3). However, this trend is not statistically
significant in either group, and does not exist in females
American Journal of Physical Anthropology—DOI 10.1002/ajpa
(Fig. 2). Additionally, among industrial societies, FMS
has significantly increased through time in both sexes
(Figs. 2, 3). Studies on secular change in recent Americans suggest that the alterations in morphology are similar in all Americans, regardless of ancestry (Rockhold,
1998; Wescott and Jantz, 2005). It is possible that shape
and robusticity measurements based on external dimensions are not sensitive enough to statistically detect
these changes, but this is unlikely due to the great variation in all groups. Additionally, Larsen (1995) argued
that skeletal changes (e.g., stature) associated with the
adoption of horticulture depend so much on local circumstances that no clear trend is apparent. Bone morphology is influenced by both proximate (changes during an
individual’s life due to activity patterns, health, and diet)
and ultimate (genetic changes due to selection and drift)
causes. It is possible that differences between huntergatherers and horticulturalists reflect major health and
diet changes or significant genetic changes due to population expansion, but this further suggests that the factors affecting morphology are probably not universal and
depend on local circumstances.
A final factor that may influence the lack of statistically significant correlations between FMS and FMR and
mobility patterns is that the pooled samples are not
genetically, nutritionally, or culturally homogeneous. However, it is not uncommon for researchers to group genetically and culturally diverse groups by subsistence
(e.g., Ruff, 1987) in biomechanical studies. Furthermore,
Wescott (2005) showed that all Native American groups
consistently exhibit a similar shape in the subtrochanteric region of the femur. He also found no significant
difference between American blacks and whites in subtrochanteric shape. This suggests that pooling the samples is probably not the problem. The lack of correlation
between femur structure and mobility most likely suggests that there is no universal effect of mobility on
femur midshaft shape and robusticity.
Since we know that long bone diaphyseal architecture
is influenced by numerous factors, should we expect
mobility to be reflected in femur cross-sectional structure? Lovejoy et al. (2002, 2003) and Ohman and Lovejoy
(2001, 2003) argued that the cross-sectional shape of the
femur shaft is primarily a reflection of the growth plate
shape, which is controlled by regulatory genes, and
therefore variation among populations in long bone
shape is a product of developmental differences between
populations. This is supported in part by population differences in proximal femur shape (Gilbert and Gill,
1990; Wescott, 2005) and the appearance of adult shape
early in growth and development (Lovejoy et al., 2002;
Gill, 2001). However, Holden and Ward (2003) found no
support for the relationship between midshaft and
growth plate shape using external dimensions. Clearly,
more research in this area is needed.
Ohman and Lovejoy (2003) and Lovejoy et al. (2003)
also contended that many of the changes in bone structure cited in biomechanical studies do not conform to the
predictions of Wolff ’s law. They challenged the notion
that expansion of both the periosteal and endosteal surfaces of the playing arm of tennis players (Jones et al.,
1977), which is frequently cited as an example of Wolff ’s
law, does not increase strength while conserving bone, as
would be predicted by a biomechanical model. On the
other hand, Holt et al. (2004) argued that if the effect of
age is considered, the results of Jones et al. (1977) correspond closely with the predictions of the biomechanical
model. However, since the current study only uses external dimensions, predictions about Wolff ’s law cannot be
Even if we assume that differences between populations in femur structure are due to plastic responses to
mechanical loads, for which there is a significant body of
supporting research (discussions in Frost, 2003; Lieberman, 1997; Martin et al., 1998), we are still left with the
question of whether femur shape, at least derived from
external dimensions, can be used to estimate TLM.
There are numerous cultural, nutritional, and environmental differences between populations that could
greatly affect mechanical loading on the lower limbs and
therefore obscure the patterns of sexual dimorphism and
cross-sectional morphology.
Pearson (2000), for example, investigated long bone
robusticity in groups from different climates and with
different subsistence practices, and found that populations in cold environments are more robust than those
from warm climates. Furthermore, cold-climate huntergatherers are generally more robust than are sedentary
groups (horticulturalists and industrialists), but warmenvironment hunter-gatherers are more gracile than
warm-climate sedentary groups. Pearson (2000) concluded that warm-climate hunter-gatherers are less
robust because they have a greater reliance on gathering
plant foods, compared to hunter-gatherers in cold climates who obtain a significant amount of their calories
from meat. Among hunter-gatherers, Binford (1980)
made a distinction between ‘‘foragers’’ and ‘‘collectors’’
based on logistical and residential mobility patterns. It
could be that warm-climate hunter-gatherers should be
considered foragers with high residential mobility (frequent campsite moves) but low logistical mobility (infrequent resource-extraction trips). Cold-climate huntergatherers, on the other hand, may exhibit low residential
mobility but high logistic mobility, i.e., the pattern seen
in collectors. This suggests that while mobility may be
reflected in femur morphology, climatic influence on
mobility type and intensity must also be taken into
account. Even so, mobility differences between populations should be reflected more in femur midshaft shape
(FMS) than robusticity (FMR), since long-distance travel
and running cause a-p directional loads on the femur.
Climatic differences in femur shape have not yet been
The physical terrain in which a group lives may also
thwart our interpretation of behavior based on femur
structure (Stock and Pfeiffer, 2004; Ruff, 1999; Wescott,
2001). Wescott (2001) found no support for the association between terrain and femur cross-sectional morphology, but Ruff (1999) asserted that terrain may be one of
the most significant factors affecting lower limb bone
robusticity. In fact, in a later paper, Ruff (2000) argued
that once terrain is factored out, the effect of subsistence
strategy seems to greatly decline. In a comparison of
populations from mountainous (Great Basin and Southwest) and relatively flatter (South Dakota and Georgia)
regions, Ruff (1999) observed that populations in mountainous regions have significantly more robust femora
than populations with similar subsistence practices in
other regions. He found no differences in Ix/Iy ratio associated with terrain type, which implies that terrain type
may affect femur strength but not shape. If so, significant dissimilarities in femur midshaft shape between populations may indicate differences in mobility
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Frost (1983, 1987, 1988, 1997, 1999, 2003) suggested
that the level of TLM may not be reflected in femur
cross-sectional morphology because running does not
cause sufficient mechanical loading to produce architectural changes in long bones. The largest mechanical
loads placed on bones are from the contraction of
muscles (Frost, 1997, 1999). Frost (1997) argued that
high-intensity activities (those that require strong
muscles) produce strains large enough to activate bone
modeling, but running does not, regardless of its frequency. He contended that frequent loads on the lower
limbs caused by running would intensify microdamage
and therefore increase remodeling, but loads would not
be great enough to activate modeling. Based on the
‘‘mechanostat’’ model of Frost (1987, 2003), the differences within and between hunter-gatherers and horticulturalists (observed by Ruff, 1987, 1994, 1999; Ruff and
Hayes, 1983; Ruff and Larsen, 1990; Ruff et al., 1984)
might be better explained by differences in sudden ‘‘maximal accelerations of the body from standing to top running speed’’ (Frost, 1999, p. 448). Frost (1999) argued
that unlike long-distance running, activities that require
sudden accelerations (e.g., soccer) place greater strains
on the lower limbs. This suggests that variation within
and between hunter-gatherers and horticulturalists may
reflect differences in hunting practices rather than TLM.
Several anthropologists also found that variation in
the intensity of activities performed by individuals during normal daily chores affects femur diaphyseal robusticity (Bridges, 1989; Bridges et al., 2000; Larsen, 1981,
1987; Ruff et al., 1984). Bridges (1989) found that femur
strength increased in both males and females with the
introduction of agriculture in the American Southeast,
which she reasoned was due to a general increase in
intensity of a variety of subsistence-related activities.
Likewise, long bone strength increased among females
with intensification of native seed crops in west-central
Illinois, but then decreased again slightly with the introduction of maize, which Bridges et al. (2000) argued may
have been due to the relative ease of processing maize
relative to native seeds. Larsen (1981, 1987) and Ruff
et al. (1984) found a different pattern in long bone robusticity and shape with the transition from hunting-gathering to horticulture on the Georgia coast, which they
also interpreted as related to intensity of activity. A
decrease in long bone robustness and sexual dimorphism
was observed with the shift in subsistence among Georgia populations, suggesting that the intensity of activities and sexual division of labor decreased with the
introduction of horticulture in this region (Larsen, 1981,
1987; Ruff et al., 1984). After Spanish contact in Florida,
however, humeral strength increased in Native populations, while the femoral shaft became more circular (Ruff
and Larsen, 1990). Ruff and Larsen (1990) interpreted
these changes as indicative of an increase in overall
workload but a decrease in mobility among these populations as they moved to missions.
Finally, variation in the age of commencement of particular adult activities, something that can probably
never be precisely determined in prehistoric populations,
may also complicate interpretations of behavior (Pearson
and Lieberman, 2004). One consistent dilemma for biomechanical interpretations is that there appears to be
considerable regional variation in the effects of horticulture on long bone cross-sectional morphology. In the
American Southwest and Great Basin, femoral strength
appears to decline after the transition from hunting-
gathering to horticulture (Brock and Ruff, 1988; Larsen
et al., 1995; Ruff, 1999; Ruff and Hayes, 1983; Ruff and
Jones, 1981), but in the Great Plains there are no significant changes in long bone strength with this subsistence
shift (Cole, 1994; Ruff, 1994; Wescott, 2001). There also
seem to be inconsistencies in the Southeastern United
States. Ruff and Larsen (1990) and Ruff et al. (1984)
observed a decline in femoral strength among males and
almost no change in females associated with the adoption of horticulture along the Georgia coast. In Tennessee, Boyd and Boyd (1989) found no significant differences between Archaic and Mississippian populations.
Bridges (1991), however, discovered that femur strength
was greater in Mississippian than Archaic groups in
Alabama. While such differences may suggest considerable regional variation in how the adoption of horticulture affected the human skeleton, Knüsel (1993) argued
that cultural variation in the age at which adult activities begin may also explain these disparities. Interestingly, modeling is primarily a phenomenon of growth,
and as the skeleton reaches maturity, modeling
decreases to a trivial level (Frost, 1985). This suggests
that at least part of the variation between populations in
long bone architecture and sexual dimorphism is a
reflection of childhood activities (Lovejoy et al., 2003). It
also implies that populations that vary in the age at
which children become involved in adult activities may
differ significantly in femur cross-sectional architecture,
regardless of TLM patterns. Likewise, sexual dimorphism in femur midshaft structure may be affected by
the timing of commencement of adult activities, since
females mature earlier than males. For example, consider two hypothetical populations (A and B) where
males and females begin adult activities at the same
age. If children begin adult activities at an earlier age in
A than B, population A may exhibit greater sexual
dimorphism than population B, even though there are
no differences in TLM or workload.
The results of this study and its implications strongly
suggest a need to examine the relationship between
loads caused by long-distance travel and femur shape in
living humans. The current study cannot directly evaluate the effectiveness of the Ix/Iy ratio in estimating mobility, but the results suggest that mobility does not have
a consistent effect on FMS or FMR. Other factors such
as genetics, climate, terrain, and age at which adult
activities commence probably have as much or more of
an effect on femur structure. These factors may obscure
the effects of mobility or may mimic it, and therefore
caution should be employed when evaluating mobility
patterns based on femur structure, at least when external dimensions are used.
The biomechanical model predicts greater sexual
dimorphism in high TLM groups than less mobile ones,
and that males in highly mobile groups will consistently
have greater femur midshaft a-p elongation due to directional loading associated with long-distance travel. This
study is based on external dimensions that may be less
sensitive than cross-sectional properties at detecting the
relationship between bone strength and mobility patterns, especially when examining changes within subsistence groups. Nevertheless, the results show that femur
midshaft shape and robusticity do not consistently correspond to the expectations of the biomechanical model.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
The results show that sexual dimorphism is greater
among more mobile populations, and that changes in
sexual dimorphism may be a strong indicator of behavior/activity change. However, how much this has to do
with sex-specific chores is still open to question. Within
each sex, however, there appears to be no significant correlation between femur midshaft shape or robusticity
and mobility level. This pattern is expected for females
but not males. Although this study does not directly test
the power of the relationship between the Ix/Iy ratio and
mobility patterns, since it uses external dimensional
data, it suggests the need for examining a broad range
of groups when using cross-sectional geometric data.
Furthermore, femur shape and robusticity based on
external measurements should be used with caution
when making interpretations about the level of TLM
within and between populations. Differences in femur
shape, especially differences in sexual dimorphism of
shape, between populations and within populations over
time probably indicate a change in activity patterns, but
these changes may be caused by numerous factors.
I thank Doug Owsley, Richard Jantz, and P. Wiley for
providing some of the data used in this study. I also
thank Lyle Konigsberg, David Gerard, and Andrew
Kramer for help on this project, and Deborah Cunningham, Lee Lyman, Clark Larsen, and the anonymous
reviewers for suggestions and comments.
Adair MJ. 1996. Woodland complexes in the central Great
Plains. In: Hofman JL, editor. Archaeology and paleoecology
of the central Great Plains. Arkansas Archeological Survey
research series no. 48. Fayetteville: Arkansas Archeological
Survey. p 101–122.
Bass WM. 1995. Human osteology: a laboratory and field manual. Columbia, MO: Missouri Archaeological Society.
Bertram JEA, Swartz SM. 1991. The ‘‘law of bone transformation’’: a case of crying Wolff? Biol Rev 66:245–273.
Binford LR. 1980. Willow smoke and dogs’ tails: hunter-gatherer
settlement systems and archaeological site formation. Am
Antiq 45:4–20.
Blakeslee DJ. 1994. The archaeological context of human skeletons in the northern and central Plains. In: Owsley DW, Jantz
RL, editors. Skeletal biology in the Great Plains: migration,
warfare, health, and subsistence. Washington, DC: Smithsonian. p 9–32.
Boyd DC, Boyd CC. 1989. A comparison of Tennessee Archaic
and Mississippian maximum femoral lengths and midshaft
diameters: subsistence change and postcranial robusticity.
Southeast Archaeol 8:107–116.
Bridges PS. 1989. Changes in activities with the shift to agriculture in the southeastern United States. Curr Anthropol 30:
Bridges PS. 1991. Skeletal evidence of changes in subsistence
activities between the Archaic and Mississippian time periods
in northwestern Alabama. In: Powell ML, Bridges PS, Mires
AMW, editors. What mean these bones? Studies in Southeastern bioarchaeology. Tuscaloosa: Alabama University Press.
p 89–101.
Bridges PS. 1995. Skeletal biology and behavior in ancient
humans. Evol Anthropol 4:112–120.
Bridges PS, Blitz JH, Solano MC. 2000. Changes in long bone
diaphyseal strength with horticultural intensification in westcentral Illinois. Am J Phys Anthropol 112:217–238.
Brock SL, Ruff CB. 1988. Diachronic patterns of change in
structural properties of the femur in the prehistoric American
Southwest. Am J Phys Anthropol 112:217–238.
Carlson DR. 1998. The Plains Indians. College Station, TX:
Texas A&M University Press.
Cole TM III. 1994. Size and shape of the femur and tibia in
northern Plains Indians. In: Owsley DW, Jantz RL, editors.
Skeletal biology in the Great Plains: migration, warfare,
health, and subsistence. Washington, DC: Smithsonian Institution Press. p 219–234.
Collier S. 1989. The influence of economic behaviour and environment upon robusticity of the post-cranial skeleton: a comparison of Australian Aborigines and other populations.
Archaeol Oceania 24:17–30.
Cultrain JB, Stafford TW. 1999. Stable-carbon isotopes and
Great Salt Lake wetlands diet: toward an understanding of
the Great Basin formative. In: Hemphill BE, Larsen CS, editors. Prehistoric lifeways in the Great Basin wetlands: bioarchaeological reconstruction and interpretation. Salt Lake
City: University of Utah Press. p 53–83.
Drass RR. 1998. The southern Plains villagers. In: Wood WR,
editor. Archaeology on the Great Plains. Lawrence, KS: University of Kansas Press. p 415–455.
Frison GC. 1998. The northwestern and northern Plains
Archaic. In: Wood WR, editor. Archaeology on the Great
Plains. Lawrence, KS: University of Kansas Press. p 140–172.
Frost HM. 1983. A determinant of bone architecture: the minimum effective strain. Clin Orthop 175:286–292.
Frost HM. 1985. The ‘‘new bone’’: some anthropological potentials. Yrbk Phys Anthropol 28:211–226.
Frost HM. 1987. Bone ‘‘mass’’ and the ‘‘mechanostat’’: a proposal. Anat Rec 219:1–9.
Frost HM. 1988. Vital biomechanics: proposed general concepts
for skeletal adaptations to mechanical usage. Calcif Tissue Int
Frost HM. 1997. Why do marathon runners have less bone than
weight lifters? A vitial-biomechanical view and explanation.
Bone 20:183–189.
Frost HM. 1999. An approach to estimating bone and joint loads
and muscle strength in living subjects and skeletal remains.
Am J Hum Biol 11:437–455.
Frost HM. 2003. Bone’s mechanostat: a 2003 update. Anat Rec
Gilbert R, Gill GW. 1990. A metric technique for identifying
American Indian femora. In: Gill GW, Rhine S, editors.
Skeletal attribution of race: methods for forensic anthropology. Maxwell Museum of Anthropology anthropological papers
no. 4. Albuquerque: University of New Mexico. p 97–99.
Gill GW. 2001. Racial variation in the proximal and distal
femur: heritability and forensic utility. J Forensic Sci 46:791–
Hemphill BE, Larsen CS, editors. 1999. Prehistoric lifeways in
the Great Basin wetlands. Salt Lake City: University of Utah
Hofman JL, Logan B, Adair M. 1996. Prehistoric adaptation
types and research problems. In: Hofman JL, editor. Archaeology and paleoecology of the central Great Plains. Arkansas
Archeological Survey research series no. 48. Fayetteville:
Arkansas Archeological Survey. p 203–220.
Holden AD, Ward CV. 2003. The effects of epiphyseal shape
on metacarpal diaphyseal proportions in hominoids: implications for long bone growth. Am J Phys Anthropol [Suppl] 36:116.
Holt BM. 2003. Mobility in Upper Paleolithic and Mesolithic
Europe: evidence from the lower limb. Am J Phys Anthropol
Holt BM, Ruff CB, Trinkaus E. 2004. The Wolff ’s law debate:
throwing out the water, but keeping the baby. Am J Phys
Anthropol [Suppl] 37:115–116.
Johnson AE. 2001. Plains Woodland tradition. In DeMallie RJ,
editor. Handbook of North American Indians: Plains. Washington, DC: Smithsonian Institution. p 159–172.
Johnson AM, Johnson AE. 1998. The Plains Woodland. In: Wood
WR, editor. Archaeology on the Great Plains. Lawrence, KS:
University of Kansas Press. p 201–234.
Jones HH, Priest JD, Hayes WC, Tichenor CC, Nagel DA. 1977.
Humeral hypertrophy in response to exercise. J Bone Joint
Surg [Am] 59:204–208.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Jungers WL, Minns RJ. 1979. Computed tomography and biomechanical analysis of fossil long bones. Am J Phys Anthropol
Jurmain R. 1999. Stories from the skeleton: behavioral reconstruction in human osteology. Amsterdam: Gordon and Breach.
Kelly RL. 1992. Mobility/sedentism: concepts, archaeological
measures, and effects. Annu Rev Anthropol 21:43–66.
Knüsel CJ. 1993. On the biomechanical and osteoarthritic differences between hunter-gatherers and agriculturalists. Am J
Phys Anthropol 91:523–527.
Larsen CS. 1981. Functional implications of postcranial size
reduction on the prehistoric Georgia coast, USA. J Hum Evol
Larsen CS. 1987. Bioarchaeological interpretations of subsistence economy and behavior from human skeletal remains.
In: Schiffer MB, editor. Advances in archaeological method
and theory. San Diego: Academic Press. p 339–445.
Larsen CS. 1995. Biological change in human populations with
agriculture. Annu Rev Anthropol 24:185–213.
Larsen CS. 1997. Bioarchaeology: interpretating behavior from
the human skeleton. New York: Cambridge University Press.
Larsen CS, Ruff CB, Kelly RL. 1995. Structural analysis of the
Stillwater postcranial human remains: behavioral implications of articular joint pathology and long bone diaphyseal
morphology. Anthropol Pap Am Mus Nat Hist 177:107–133.
Larsen CS, Ruff CB, Griffin MC. 1996. Implications of changing
biomechanical and nutritional environments for activity and
lifeway in the eastern Spanish borderlands. In: Baker BJ,
Kealhofer L, editors. Bioarchaeology of Native American
adaptation in the Spanish borderlands. Gainesville: University Press of Florida. p 95–125.
Leach HM. 2003. Human domestication reconsidered. Curr
Anthropol 44:349–368.
Lieberman DE. 1997. Making behavioral and phylogenetic inferences from hominid fossils: considering the developmental influences of mechanical forces. Annu Rev Anthropol 26:185–210.
Lieberman DE, Devlin MJ, Pearson OM. 2001. Articular area
responses to mechanical loading: effects of exercise, age, and
skeletal location. Am J Phys Anthropol 116:266–277.
Lovejoy CO, Meindl RS, Ohman JC, Heiple KG, White TD.
2002. The Maka femur and its bearing on the antiquity of
human walking: applying contemporary concepts of morphogenesis to the human fossil record. Am J Phys Anthropol
Lovejoy CO, McCollum MA, Reno PL, Rosenman BA. 2003.
Developmental biology and human evolution. Annu Rev
Anthropol 32:85–109.
Martin RB, Burr DB, Sharkey NA. 1998. Skeletal tissue
mechanics. New York: Springer.
McHenry HM. 1988. New estimates of body weight in early
hominids and their significance to encephalization and megadontia in ‘‘robust’’ australopithecines. In: Grine FE, editor.
Evolutionary history of the ‘‘robust’’ australopithecines. New
York: Aldine de Gruyter. p 133–148.
Meadows Jantz L, Jantz RL, Herrmann NP, Sparks CS, Weisensee KE, Kopp D, Owsley DW, Kardash R, Clark R. 2001. The
Refugio Mission: an analysis of the human skeletal remains.
Center for Archaeological Research report, University of
Texas-San Antonio: University of Texas Press.
Morrison JB. 1968. Bioengineering analysis of force actions
transmitted by the knee joint. Biomed Eng 3:164–170.
Morrison JB. 1969. Function of the knee joint in various activities. Biomed Eng 4:573–580.
Morrison JB. 1970. The mechanics of the knee joint in relation
to normal walking. J Biomech 3:51–61.
Ohman JC, Lovejoy CO. 2001. The shape of a long bone’s shaft:
bending stress or growth plate form? Am J Phys Anthropol
[Suppl] 32:115.
Ohman JC, Lovejoy CO. 2003. Asymmetry in the humeri of tennis
players: ‘‘Wolff’s law’’ or not? Am J Phys Anthropol [Suppl] 36:
O’Neill MC, Ruff CB. 2004. Estimating human long bone crosssectional geometric properties: a comparison of noninvasive
methods. J Hum Evol 47:221–235.
Ousley SD, Jantz RL. 1998. The forensic data bank: documenting skeletal trends in the United States. In: Reichs KJ, editor.
Forensic osteology: advances in the identification of human
remains, 2nd ed. Springfield, IL: Charles C. Thomas. p 441–
Parks DR. 2001. Arikara. In: DeMallie RJ, editor. Handbook of
North American Indians: Plains. Washington, DC: Smithsonian Institution. p 365–390.
Pearson OM. 2000. Activity, climate, and postcranial robusticity:
implications for modern human origins and scenarios of adaptive change. Curr Anthropol 41:569–607.
Pearson OM, Lieberman DE. 2004. The aging of Wolff ’s law:
ontogeny and responses to mechanical loading in cortical
bone. Yrbk Phys Anthropol 47:63–99.
Rockhold LA. 1998. Secular change in external femoral measures from 1840 to 1970: a biomechanical interpretation. M.A.
thesis, University of Tennessee-Knoxville.
Ruff CB. 1987. Sexual dimorphism in human lower limb bone
structure: relationship to subsistence strategy and sexual
division of labor. J Hum Evol 16:391–416.
Ruff CB. 1994. Biomechanical analysis of northern and southern
Plains femora: behavioral implications. In: Owsley DW, Jantz
RL, editors. Skeletal biology in the Great Plains: migration,
warfare, health, and subsistence. Washington, DC: Smithsonian Institution Press. p 235–246.
Ruff CB. 1999. Skeletal structure and behavioral patterns of
prehistoric Great Basin populations. In: Hemphill BE, Larsen
CS, editors. Prehistoric lifeways in the Great Basin wetlands:
bioarchaeological reconstruction and interpretation. Salt Lake
City: University of Utah Press. p 290–320.
Ruff CB. 2000. Biomechanical analyses of archaeological human
skeletons. In: Katzenberg MA, Saunders SR, editors. Biological anthropology of the human skeleton. New York: Wiley.
p 71–102.
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, Hayes WC. 1983. Cross-sectional geometry of Pecos
Pueblo femora and tibiae—a biomechanical investigation: II:
sex, age, and side differences. Am J Phys Anthropol 60:383–
Ruff CB, Jones HH. 1981. Bilateral asymmetry in cortical bone
of the humerus and tibia—sex and age factors. Hum Biol
Ruff CB, Larsen CS. 1990. Postcranial biomechanical adaptations to subsistence strategy changes on the Georgia coast.
Anthropol Pap Am Mus Nat Hist 68:94–120.
Ruff CB, Larsen CS, Hayes WC. 1984. Structural changes in
the femur with the transition to agriculture on the Georgia
coast. Am J Phys Anthropol 64:125–136.
Ruff CB, Scott WW, Liu AYC. 1991. Articular and diaphyseal
remodeling of the proximal femur with changes in body mass
in adults. Am J Phys Anthorpol 86:397–413.
Ruff CB, Trinkaus E, Walker A, Larsen CS. 1993. Postcranial
robusticity in Homo. 1. temporal trends and mechanical interpretation. Am J Phys Anthropol 91:21–53.
Sokal RR, Rohlf FJ. 2000. Biometry: the principles and practice of statistics in biological research. San Francisco: W.H.
Stock J. 2002a. A test of two methods of radiographically
deriving long bone cross-sectional properties compared to
direct sectioning of the diaphysis. Int J Osteoarchaeol 12:
Stock J. 2002b. Climatic and behavioural influences on postcranial robusticity among Holocene foragers. Ph.D. dissertation,
University of Toronto.
Stock J, Pfeiffer S. 2001. Linking structural variability in long
bone diaphyses to habitual behaviors: foragers from the
southern African Later Stone Age and the Andaman Islands.
Am J Phys Anthropol 115:337–348.
Stock J, Pfeiffer S. 2004. Long bone robusticity and subsistence behaviour among Later Stone Age foragers of the forest
and fynbos biomes of South Africa. J Archaeol Sci 31:999–
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Trinkaus E, Ruff CB. 2000. Comment to Pearson OM. Activity, climate, and postcranial robusticity. Curr Anthropol 41:
Trotter M. 1981. Robert J. Terry. 1871–1966. Am J Phys Anthropol 56:503–508.
Ubelaker DH. 1989. Human skeletal remains: excavation, analysis, interpretation, 2nd ed. Washington, DC: Taraxacum.
Wedel WR, Frison GC. 2001. Environment and subsistence. In:
DeMallie RJ, editor. Handbook of North American Indians:
Plains. Washington, DC: Smithsonian Institution. p 44–60.
Wescott DJ. 2001. Structural variation in the humerus and
femur in the American Great Plains and adjacent regions: dif-
ferences in subsistence strategy and physical terrain. Ph.D.
dissertation, University of Tennessee-Knoxville.
Wescott DJ. 2005. Population variation in femur subtrochanteric shape. J Forensic Sci 50:286–293.
Wescott DJ, Jantz RL. 2005. Assessing craniofacial secular
change in American whites and blacks using geometric
morphometry. In: Slice D, editor. Modern morphometrics in
physical anthropology, volume V: developments in primatology: progress and prospects. New York: Kluwer. p 231–245.
Wood WR. 1998. Introduction. In: Wood WR, editor. Archaeology
on the Great Plains. Lawrence, KS: University of Kansas
Press. p 1–15.
American Journal of Physical Anthropology—DOI 10.1002/ajpa
Без категории
Размер файла
204 Кб
femur, effect, robusticity, shape, external, midshaft, mobility
Пожаловаться на содержимое документа