- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
On a basis of a method for muscle cross-sectional area estimation from cortical bone area that was previously developed (Slizewski et al. Anat Rec 2013; 296:1695–1707), we reconstructed muscle cross-sectional area at 65% of radius length for a sample of Neolithic human remains from the Linear Pottery Culture (ca. 5,700–4,900 years BC). Muscle cross-sectional area estimations for the Neolithic sample were compared to in vivo measurements from a recent human sample. Results demonstrate that the Neolithic individuals had larger muscle cross-sectional area relative to radius length than the contemporary humans and that their forearms were more muscular and robust. We also found significant differences in relative muscle cross-sectional area between Neolithic and recent children that indicate different levels of physical stress and isometric activities. Our results fit into the framework of studies previously published about the sample and the Linear Pottery Culture. Therefore, the new approach was successfully applied to an archaeological sample for the first time here. Results of our pilot study indicate that muscle cross-sectional area estimation could in the future supplement other anthropological methods currently in use for the analysis of postcranial remains. Anat Rec, 297:1103–1114, 2014. © 2014 Wiley Periodicals, Inc.
As bone is the major and often only source of information for Prehistoric Anthropologists and Paleoanthropologists, additional knowledge about physical abilities of individuals represented by their skeletal remains could be derived from estimation of soft tissue properties. Muscle cross-sectional area estimation can provide information on activity patterns of an individual or a group, and is an alternative approach to the estimation of force production capability and activity patterns based on visual scoring of enthesopathies (for discussion and approaches see, e.g., Baumann, 1926; Plummer, 1984; Stirland, 1993; Churchill and Morris, 1998; Robb, 1998; Stirland, 1998; Wilczak, 1998; Weiss, 2003; Henderson and Alves Cardoso, 2013). Estimation of muscle cross-sectional area could also improve other methods such as calculations of body mass or reconstructions of the body composition and visual appearance from skeletal remains.
Muscle cross-sectional area estimation is based on the theory of a linear relationship between muscle and bone within an individual, called the “muscle–bone unit” (Schoenau and Frost, 2002; Schoenau, 2005). It is widely accepted today that bone cross sections provide valuable insights into the mechanical properties of long bones, and give information about the peak forces (bending and torsion) that have been placed on the bone during lifetime (Lovejoy et al., 1976; Ben-Itzhak et al., 1988; Grine et al., 1995; Churchill et al., 1996; Pfeiffer and Zehr, 1996; Pearson and Grine, 1997; Katzenberg and Saunders, 2000; Stock and Shaw, 2007). It has been demonstrated by medical studies that muscle cross-sectional area and cortical bone cross-sectional area are strongly correlated in all so far examined living mammals (Runge et al., 2002). Therefore, muscle cross-sectional area estimation can be a first step toward the assessment of force production capability from the skeletal remains of individuals and/or populations. Maximum forces placed on the bone by muscle stimulate the buildup of both the cortical bone and muscle cross-sectional area (Rauch et al., 2002; Schoenau and Frost, 2002). The individual grows stronger in terms of both the muscle and bone properties (Neu et al., 2001) and as a result also becomes stronger in terms of his/her capability to produce forces like punch, grip, or tensile. It is important to note that muscle and bone mass buildup are primarily stimulated by isometric forces and not by endurance activities as since the latter do not cause significant increases of muscle cross-sectional area and bone strength (Schoenau and Frost, 2002). Therefore, estimations of muscle cross-sectional area will rather provide us with information on activities that required muscular peak forces than on the overall physical activity of an individual.
Muscle cross-sectional area alone may not be mistaken as a direct equivalent for the “strength” of an individual. In the forearm, “strength” in terms of grip or tensile force is the result of both muscle cross-sectional area and forearm length (Fricke et al., 2009). Due to the leverage effect an individual with longer limbs is provided with a biomechanical advantage and needs a smaller muscle cross-sectional area for generating equal forces than an individual with shorter limbs. Therefore, estimations of muscle cross-sectional area have to be standardized by forearm length to provide information on the physical capabilities of the forearm and on the stresses placed on the individual during lifetime (Ruff et al., 1993; Churchill, 1994; Trinkaus and Churchill, 1999; Schoenau et al., 2000; Schoenau, 2005; Fricke et al., 2009). Muscle cross-sectional area of the forearm standardized by radius length provides us with a surrogate for forearm force production capability and tells us about the stresses placed on the forearm of the individual during lifetime. High values for muscle cross-sectional area standardized by forearm length indicate both a relatively large muscle cross-sectional area proportionately to the limb length and high stresses placed frequently on the forearm of the individual.
An approach towards muscle cross-sectional area estimation of the forearm from the cortical bone of the radius has been presented in a previous article, and was based on an in vivo study in a living human sample (Slizewski et al., 2013). The present work is a follow-up study that presents the first application of this newly published method to an archaeological skeletal sample. The aim of the current study is to demonstrate how muscle cross-sectional area estimation works in practice, which information can be derived from it for archaeological and anthropological investigations, and to evaluate the limitations of the method.
- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
Average relative radius length within the Neolithic sample is 23.91 cm for adult men and 21.24 cm for adult women, with the smallest radius being 13.8-cm long (child) and the longest measuring 27.4 cm (adult male). Muscle cross-sectional area could be estimated for 115 radii (58 individuals). One measurement of a female left radius was invalid due to matrix filling within the shaft at the 65% location. Accordingly, %DA could only be calculated for 57 individuals. For muscle cross-sectional area estimation, the valid measurement of the right radius of the female individual was used under the restriction that it could not be determined whether or not this was the dominant side.
Calculations of %DA values for the rest of the sample are displayed in Table 2. Differences between the two radii of one individual are remarkably small within the Mühlhausen sample. The %DA values of 85.96% of the entire sample range below 2.5%. Only in two males, the difference between the radii is clearly pronounced with %DA values higher than 5%. Both males have been classified as right handed based on the %DA values. For further analysis, all individuals have been classified as left or right handed based on the tendency given by %DA and only the radius, which was classified as non “dominant” was included in the analysis. This has been done to assure comparability to the DONALD sample, which only includes the nondominant arms of the participants (self-stated). However, it has to be pointed out that the classifications of handedness provided here for the Mühlhausen sample cannot be considered as reliable determinations of the dominant arms as the differences between most radii of the sample are too low to be regarded as significant. The threshold for reliable determinations of handedness based on %DA has been identified as being 5% for the ulna (Shaw, 2011) and only 12.28% of the individuals of the Mühlhausen sample show a difference in cortical area higher than 5%.
Table 2. Results of %DA calculations for the Neolithic sample
|0,1 – 2,5%||17||8||15||9||x||x||49|
|2,5 – 5%||3||x||x||2||x||x||5|
|5 – 100%||x||x||2||x||x||x||2|
When burial areas separate the sample, it shows that the majority of females buried in Area I are classified as right handed, whereas the majority of males from Area I are classified as left handed. For the individuals from Area II this effect is reversed.
Muscle cross-sectional area was estimated for all radii classified as “nondominant” based on the %DA values presented above. Mean values for muscle cross-sectional area (mm2) estimated by the equation quoted in the methods section are displayed in Table 3. Mean muscle cross-sectional area values of the Mühlhausen sample are lower than those of the recent German sample for all age classes except for Infans I. The range for muscle cross-sectional area values is also larger in the recent than in the Neolithic sample, probably an effect of larger sample size. The difference in mean muscle cross-sectional area is most pronounced between recent and Neolithic mature individuals; with the mean muscle cross-sectional area of recent mature individuals ranging 727.46 mm2 above that of the Neolithic group.
Table 3. Absolute mean vales in mm2 for measured muscle cross-sectional area (DONALD) and estimated muscle cross-sectional area (Mühlhausen)
| ||DONALD sample||Mühlhausen Sample|
|Age Class||N||Min. (mm2)||Max. (mm2)||Mean (mm2)||N||Min. (mm2)||Max. (mm2)||Mean (mm2)|
This picture changes dramatically when muscle cross-sectional area values are standardized by radius length. Although the range of variation is still significantly larger within the modern sample, the Neolithic sample provides clearly higher mean standardized values than the recent one within all age classes (see Table 4). Also, all minimum values for estimated muscle cross-sectional area standardized by radius length are significantly higher within the Neolithic sample than minimum values for muscle cross-sectional area standardized by radius length from the modern sample. For Infans I and II children, this also applies to maximum values, even though the difference is less pronounced here, while maximum values for adult and mature individuals are higher within the recent sample. The difference in mean values between the modern and the Neolithic sample is largest between the Infans II groups and becomes smaller with each age class upward.
Table 4. Muscle cross-sectional area (mm2) standardized by radius length (mm)
| ||DONALD sample||Mühlhausen Sample|
When separated by sexes, the Neolithic groups range on average above the recent Germans in all age classes (see Fig. 4). While for Neolithic males the muscle cross-sectional area standardized by radius length index is highest within the Infans I and II age classes, the index increases from Infans II on with each age class in the modern human males. Standardized muscle cross-sectional area values of Infans I children range above Infans II and juveniles in modern males and females and in Neolithic males. In Neolithic females, the Infans II age class has higher standardized muscle cross-sectional area values than the Infans I group. In every subsample, the mature age class provides higher muscle cross-sectional area index values than the adult.
Although sexual dimorphism in standardized muscle cross-sectional area values is generally small within the living human sample and increases from Infans II on with age, it is pronounced between Neolithic males and females and decreases with age. Only in the mature age group, the difference between males and females is slightly higher (+0.15) in the modern than in the Neolithic group. Although sexual dimorphism is very small between Infans II males and females of the modern sample (1.36), it is marked between boys and girls of the same age class in the Neolithic sample (8.1).
Figure 5 displays the standardized muscle cross-sectional area index separate by burial areas. Even though sample composition in matters of age, sex, and number of individuals differs between Areas I and II and is not representative in the current study, it is recognizable that the Infans I and II groups have generally higher muscle cross-sectional area index values than adults and juveniles in both areas. The small sizes of the subsamples for Areas II and I, and the lack of female subadults from Area II, make it difficult to analyze differences in sexual dimorphism between the areas.
- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
The major aim of the present study was to demonstrate how muscle cross-sectional area estimations work on a skeletal sample from an archaeological context. The results of the analysis described above have shown that muscle cross-sectional area estimations can provide us with valuable information on the body composition which complement and expand data derived from the skeletal remains by other methods.
However, as this was a pilot study for muscle cross-sectional area estimation from skeletal remains and the focus was, therefore, on methodology, the Neolithic sample was primarily selected to fit the general framework for muscle cross-sectional area estimation best. This provides us with several limitations for the results presented above. Sample sizes were small with only 29 individuals from each burial area, and sample compositions are biased and not representative for age, sex, and burial positions within the graveyards. With this kept in mind, the results of muscle cross-sectional area estimations still allow for a number of interpretations.
The determinations of “handedness” based on %DA indicate primarily a consistent strain on both forearms for the majority of Neolithic individuals from both areas, potentially related to a range of tasks, which required the equal involvement of both arms. Small bilateral asymmetry (compared to modern or historic reference populations) of the upper limbs has been found previously in other Neolithic samples (Marchi et al., 2006; Molnar, 2006; Lorkiewicz et al., 2009; Stock et al., 2011).
The buildup of cortical bone and muscle size is primarily induced by repeated activities that place peak forces on the bones—in the present analysis the forearm. As already described above, “strength”—in the forearm expressed by, for example, grip or tensile force—is the result of both limb length and muscle properties (leverage effect). In the present study, this means that the Neolithic population had a physical disadvantage for a number of biomechanical performances compared to the recent human sample because of their on average significantly shorter forearms. Therefore, a Neolithic individual with a shorter forearm would have needed a larger muscle cross-sectional area to generate the same forces as a recent individual with a longer forearm. Higher mean values for muscle cross-sectional area standardized by forearm length are accordingly not necessarily related to the capability of producing higher forces but to a higher level of isometric activities performed by that person, which corresponds to a physically more challenging lifestyle. Muscle cross-sectional area estimations standardized by limb length also provide us with data on the body composition of an individual. High values for muscle cross-sectional area standardized by radius length indicate muscular forearms, whereas low values indicate a more gracile body shape.
The results of both absolute and standardized muscle cross-sectional area estimations show that the range of variation within age classes and sexes is significantly higher in the recent human sample than within the Neolithic. This higher range of variation within the modern sample is probably an effect of both an overall larger sample size (695 recent Germans vs. 58 Neolithic individuals) and/or a more diverse range of activities within the living human sample. Although recent German middle class individuals can be involved either in physically demanding (e.g., craftsmen) or physically nondemanding labor (e.g., white-collar workers), and may be or may be not exercising (e.g., weight training at a fitness center), it is supposed that during the Linear Pottery Culture the dominant subsistence strategy combined husbandry and agriculture supported by a regional varying dependence in hunting and gathering (see, e.g., Jeunesse, 1987, 2000; Uerpmann and Uerpmann, 1997; Kind, 1998; Bogucki, 2000; Stephan, 2005; Schade-Lindig and Schade 2006). Even though there is evidence indicating that division of labor and social differentiation became manifest during the Linear Pottery Culture (Veit, 1996), both gathering and agriculture will involve a generally higher level of mechanical loadings placed on the forearm (e.g., by carrying fruits or plowing) than office work. Also, the production of stone tools requires high strike forces and is a biomechanically challenging task for the upper limbs (Williams et al., 2010). Therefore, even though the different sample sizes will certainly have influenced the results, it seems reasonable to hypothesize that in terms of mechanical loadings the activities performed by individuals of the Mühlhausen group were probably more similar to each other than the mechanical loadings produced by the diverging lifestyles of the individuals from the living human sample (office work vs. physical training or labor).
As has been demonstrated by the standardized muscle cross-sectional area values, the mechanical loadings placed on the forearms of the Neolithic sample were on average higher within all age classes, indicating a generally higher level of frequently performed mechanically challenging tasks for the Mühlhausen people. This also applies to the minimum values, which suggests that even those individuals from the Mühlhausen site who had a relatively low involvement in isometric tasks were still much more frequently performing physically demanding activities than individuals from the lowest range of the modern sample. Results for muscle cross-sectional area standardized by radius length clearly demonstrate that muscular activity level of the forearm was significantly higher in the Neolithic than in the recent human sample, as one would expect. Muscle cross-sectional area standardized by radius length shows that the disadvantage of shorter forearms was compensated in the Neolithic population by the buildup of larger muscle cross-sectional areas, which must have enabled them to produce the same or higher forces than the recent individuals. The Mühlhausen population was, therefore, more frequently and extensively involved in exertion activities than the living human sample.
The most drastic differences in muscle cross-sectional area index in the present study were found between recent children and the Neolithic subadult sample. While the Neolithic Infans II individuals have the highest mean muscle cross-sectional area values standardized by radius length of the Mühlhausen sample, modern Infans II children have the smallest muscle cross-sectional area index values of the modern sample. Accordingly, Neolithic Infans II children were significantly more isometrically active than recent children. The results also imply that the amount of regular, high exertions was proportionally higher within childhood activities than within adult activities of the Mühlhausen people. This might be an effect of children being involved in adult activities as the infant body would have responded with the buildup of proportionally more bone and a larger muscle cross-sectional area than the adult body performing the same tasks, because the stresses placed on the infant forearms would have been higher due to the shorter limb length. Within the Infans II age class Neolithic boys have significantly larger proportional muscle cross-sectional areas than contemporary girls of the same age, while the values for modern boys and girls of the same age are equally high. This implies that the Mühlhausen Infans II might have been involved in some age and sex specific activities, which placed peak forces on their forearms. The high values for muscle cross-sectional area within the young age groups of the Mühlhausen sample correspond to the theory of Witwer-Backofen and Tomo (2008) that childhood during the Linear Pottery Culture was a severe stress phase. An auxological study about the Linear Pottery Culture, that also included 38 subadult individuals from Stuttgart-Mühlhausen, showed that the children from this region in Germany reached the mid-growth spurt earlier than other children from the same time period (Welte and Wahl, 2010). In recent children, the mid-growth spurt takes place at about 5.5–6.5 years. It is related to the maturation of the adrenal cortex and characterized by a phase of accelerated growth (Stratz, 1926; Knussmann, 1996). The shown earlier mid-growth spurt for the Mühlhausen children could potentially be related to developmental disorder (documented by enamel hypoplasia in the Mühlhausen sample) caused by malnutrition, diseases, or physical overload. However, the reasons for the peak in muscle cross-sectional area in the Infans II age class and its relation to the earlier mid-growth spurt have to remain speculative at this point. A larger sample size and a more specific study set up would be required to derive at any reliable conclusions here.
The overall trend for increasing muscle cross-sectional area relative to radius length within the mature age class could be primarily related to hormonal changes, as it is present in both Neolithic and recent mature individuals and, therefore, unlikely to be linked to a specific task. A significant influence of the sexual hormones estrogen and testosterone on the muscle–bone unit has been described previously (Rauch and Schoenau, 2001; Schoenau et al., 2001; Schoenau, 2005), and the estrogen level in women decreases about 40% to 60% during menopause at an age of 45–55 years (Minkin and Wright, 1997), while the testosterone level in men decreases slightly from the age of 30 onward about 1% each year (Brawer, 2004). But, increasing muscle cross-sectional area values beyond the age of 40 are an unusual finding as various studies have described a general trend toward decreasing muscle mass and bone strength with age (e.g., Lang, 2011; Ławniczak and Kmieć, 2012; Sornay-Rendu et al., 2012). Conversely, it has been demonstrated that regular exercise continuous from younger to older ages results in significantly larger muscle cross-sectional area and the ability to produce higher forces in mid-ages and old-ages (Wroblewski et al., 2011; Leskinen et al., 2012;). Furthermore, it has been suggested that there is no peak level of bone mass related to age and that muscle cross-sectional area should always be related to bone length and not to age (Rauch et al., 2002; Schoenau, 2005). According to a recent study, 56% of German males aged over 18 and 38% of German females over 18 are regularly engaged in physical activities that make them sweat or lose their breath for at least 2.5 hr/week (Lampert et al., 2012), which might be one factor contributing to the high values for muscle cross-sectional area in the mid-ages of the sample. For Neolithic mature individuals it could be hypothesized that they were still regularly involved in isometric activities stressing their forearms as part of their everyday lives.
Muscle cross-sectional area estimation has provided us with information on the degree of isometric activities involving the forearm within the Mühlhausen sample. A special situation for the younger age classes was identified. The forearms of the Mühlhausen people were very muscular and “robust” compared to recent German average.
In the present study, only muscle cross-sectional area of the forearm was estimated. In the future, further in vivo studies in living human samples will have to provide a basis for estimating muscle cross-sectional area of other limb bones. As this was the first assessment of muscle cross-sectional area in a prehistoric sample no dataset from another culturally, temporally, or geographically related skeletal sample was available for comparison and all data of the Mühlhausen sample were compared to a recent human sample. More information could be derived from additional muscle cross-sectional area estimations on skeletal samples from other early Neolithic sites.
We demonstrated a method for muscle cross-sectional area estimation from the radius, which can be applied as an additional or alternative method for estimating physical strength and robusticity of individuals and populations in Prehistoric Anthropology. The present study has shown that muscle cross-sectional area estimations can provide anthropologists with new perspectives on skeletal samples and provide information on soft tissue properties otherwise inaccessible for human skeletal material.