Changes in skeletal robusticity in an iron age agropastoral group The samnites from the Alfedena necropolis (Abruzzo Central Italy).код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 144:119–130 (2011) Changes in Skeletal Robusticity in an Iron Age Agropastoral Group: The Samnites From the Alfedena Necropolis (Abruzzo, Central Italy) V.S. Sparacello,1* O.M. Pearson,1 A. Coppa,2 and D. Marchi3 1 Department of Anthropology, University of New Mexico, Albuquerque, NM Dipartimento di Biologia Animale e dell’Uomo, Università ‘‘Sapienza’’ di Roma, Italy 3 Department of Evolutionary Anthropology, Duke University, Durham, NC 2 KEY WORDS cross-sectional geometry; humeral asymmetry; mobility; pastoralism; Neolithic ABSTRACT Cross-sectional geometrical (CSG) properties of an Iron Age Samnite group from the Alfedena necropolis (Abruzzo, Italy, 2600–2400 B.P.) are compared with a Ligurian Neolithic sample (6000–5500 B.P.). In the period under examination, Samnites were organized in a tribal confederation led by patrilinear aristocracies, indicating incipient social stratiﬁcation. In comparison, Neolithic society lacked clear signs of social hierarchy. The subsistence of both groups was mainly based on pastoralism and agriculture, but changes in habitual behavior are expected due to the socio-economic transformations that characterized the Iron Age. The Samnites’ warlike ideology suggests that unimanual weapon-use and training would have become frequent for males. The intensiﬁcation of agriculture and the adoption of transhumant pastoralism, performed by a smaller subset of the population, likely led to a lower average level of logistic mobility. The strongly genderized ideology of the period suggests a strict sexual division of labor, with women primarily performing sedentary tasks. CSG properties based on periosteal contours were calculated for humeri, femora, and tibiae (N 5 61). Results corroborated the expectations: Alfedena males show substantial humeral bilateral asymmetry, indicating prevalent use of one arm, likely due to weapon training. In both sexes lower limb results indicate reduced mobility with respect to the Neolithic group. Sexual dimorphism is signiﬁcant in both humeral asymmetry and lower limb indicators of mobility. Although both groups could be broadly deﬁned as agropastoral based on archeological and historical evidence, CSG analysis conﬁrmed important differences in habitual behavior. Am J Phys Anthropol 144:119–130, 2011. V 2010 Wiley-Liss, Inc. The purpose of this research is to use biomechanical analysis to help interpret changes in activity patterns that occurred with the deep social and economical transformations of the Metal Ages, particularly the Italian Iron Age. This period saw a dramatic demographic expansion, agricultural intensiﬁcation, a novel pattern of land use, and a level of social stratiﬁcation unknown in Italy previously. In comparison, Neolithic pastoralists in Italy had a society lacking indicators of social hierarchy (Guidi, 2000; Robb, 2007). By the Iron Age, a more structured, proto-urban society had developed, with indications of incipient state formation (Peroni, 1989). Although both Neolithic and Iron Age subsistence were based on agriculture and pastoralism, changes in social structure and economy are expected to have led to substantial changes in activity patterns. Testing if those hypothetical changes in habitual behavior are mirrored in the skeletal biology of Iron Age groups may greatly contribute to our understanding of a crucial period of human history. Changes in habitual stresses and activities are inferred here through the comparison of long bone mechanical parameters obtained from cross-sectional diaphyseal geometry (CSG) in two Italian samples: an Iron Age sample from the Alfedena necropolis (2600–2400 B.P.; Parise Badoni and Ruggeri Giove, 1980) and a Middle Neolithic sample from the Finale Ligure area (6000– 5500 B.P; Maggi, 1997). Relevant bioarchaeological parameters such as the degree of mobility of a population, the dissimilarity between male and female activities, and the preferential use of one arm can be qualitatively predicted from material and cultural evidence drawn from archeology, ethnology, and history. CSG provides a means to evaluate biological evidence for habitual activity, and the quantitative nature of its results provides a more deﬁned framework for comparisons between human groups. The integration of numerical data derived from biomechanical analysis of skeletal remains with ethnographic and archaeological information has proven to be a useful tool in reconstructing the physical and behavioral environment of past and recent populations (e.g., Bridges, 1989; Ruff et al., 1993; Churchill et al., 1996, 2000; Ruff, 1999, 2000a; Holt, 2003). Cortical bone is primarily responsive to strain prior to sexual maturity (Pearson and Lieberman, 2004), suggesting that inferences on activity patterns drawn from adult skeletal properties should be limited to these groups on which individuals participate in adult economic behaviors sufﬁciently early. However, as discussed by Ruff et al. (2006b), CSG also reﬂect in part adult activities, and C 2010 V WILEY-LISS, INC. C *Correspondence to: Vitale Stefano Sparacello, Department of Anthropology MSC01-1040, 1 University of New Mexico NM871310001, Albuquerque, NM. E-mail: Vito@unm.edu Received 24 December 2009; accepted 11 June 2010 DOI 10.1002/ajpa.21377 Published online 17 August 2010 in Wiley Online Library (wileyonlinelibrary.com). 120 V.S. SPARACELLO ET AL. adolescent activities likely parallel young adult activities in many preindustrial societies, such as the ones analyzed in this article. Although long bones from industrial groups are consistently less robust (relatively less rigid or strong) than agriculturalist and foraging groups (Ruff et al., 1993), no general pattern of changes in robusticity has been documented in those studies concentrating on the adoption or intensiﬁcation of agriculture (Ruff et al., 1984; Bridges, 1989; Ruff, 1999, 2008; Bridges et al., 2000 Wescott, 2006). Some studies found a decrease in relative long bone strength, particularly in males (Ruff et al., 1984; Larsen et al., 1995), while others documented an increase in female humeral robusticity, especially on the left side (Bridges, 1989). Many societies also show a trend toward a reduction in the degree of lateralization in the upper limb after a shift to agriculture, presumably from decreases in hunting activities performed via throwing and an increase in bimanually intensive activities, such as the use of grinding stones (Bridges, 1989; Trinkaus et al., 1994; Churchill et al., 1996, 2000; Marchi et al., 2006). However, speciﬁc activities such as woodworking or warfare-related weapon training may have a role in developing a certain degree of humeral asymmetry in nonforaging economies (Rhodes and Knüsel, 2005; Sparacello and Marchi, 2008). Hunter-gatherers also tend to have less circular femoral diaphyses than agriculturalists (Bridges, 1989; Ruff, 1999; Bridges et al., 2000), which is thought to reﬂect decreased mobility among the agriculturalists. However, Robb (1994a) hypothesized that, to the degree that they remain highly mobile, herders may retain shape indices similar to those of hunter-gatherers. His hypothesis has been conﬁrmed by recent CSG analyses (Marchi et al., 2006; Marchi, 2008; Sparacello and Marchi, 2008). The variability of these patterns suggests that CSG characteristics are not related in a simple, uniform way to broad changes in subsistence, such as the passage to a production economy or the intensiﬁcation of agriculture. Rather, CSG results mirror the complexity of human adaptations, and reﬂect the particular conﬁgurations that local subsistence economies take on the basis of social, technological, and environmental factors. It therefore becomes important to base predictions about CSG parameters on as much information as possible, integrating theoretical expectations with hypotheses based on the available archaeological, ethnological, and historical evidence. Iron age and Alfedena socio-economic context The Iron Age population considered here represents the Samnite people. This well-deﬁned ethnic proto-historic group shared a common language (Oscan) and law, a tribal federal political organization, and a strong sense of self-identiﬁcation in a territory and people (Salmon, 1967). The sample studied here derives from the skeletal series of the Alfedena necropolis, which is located in the high valley of the Sangro River (Abruzzo, Central Italy). With the Iron Age (ﬁrst millennium B.C.) archeologists recognize a sharp demographic increase in the Italian peninsula. The number and the size of sites belonging to the Iron Age outweigh all the preceding phases of prehistory, and include several large necropoli containing thousands of burials distributed over a few centuries (D’Agostino, 1974, Peroni, 1979). Many settlements American Journal of Physical Anthropology lasted for centuries, and were often fortiﬁed with walls, or guarded from a system of hilltop fortiﬁed outposts. One example of such large and durable settlements is the site of Alfedena, which consists of a fortiﬁed hilltop settlement and a vast necropolis of over 1,400 burials in an area of about 35,000 m2 (Mariani, 1901). In the Iron Age, a more intensive and stable form of agriculture was adopted, probably in response to demographic pressure (Boserup, 1965, 1975). In the Italian peninsula, agricultural intensiﬁcation is inferred from the shift of settlements away from grazing lands into areas suitable for seed crops, by the appearance and increased production of high quality metal objects such as metal sickles and axes, and by the development of agricultural techniques such as the use of the draught horse, the plough, and the four-wheeled cart (Peroni, 1979). Pastoralism was still economically important, as suggested by the presence of numerous tools for processing secondary animal products in the archaeological assemblages of Italian Iron Age sites (Salmon, 1967). Most of the Samnites’ territory is mountainous, barren, and unsuitable for agriculture. Although farming activities are reported by archaeologists at Alfedena, the most important economic activity appears to have been stock-raising (Salmon, 1967). Herding took the form of annual, short- distance ‘‘vertical’’ transhumance. Herds were moved to the highlands during the summer and down to valleys during the winter (D’Ercole, personal communication). The Italian Iron Age saw an increase in social complexity, with the rise of aristocracies organized in tribal confederacies based on multi-patrilinear alliances (Pallottino, 1991; Robb, 1994b). The societies become more stratiﬁed, with the formation of a strong warrior aristocracy whose sphere of inﬂuence transcended the single village (Peroni, 1979). At Alfedena, burial patterns reﬂect the presence of an aristocracy held together by extended kin coalitions. Tombs were arranged in family circles, and a number of male and female tombs with rich grave goods can be distinguished. The number of contemporary circles with rich burials demonstrates that in each community many family lines were part of the dominant group (Vida Navarro, 1993). Multi-patrilinear alliances are corroborated by studies of Alfedena cranial epigenetic traits, suggesting a close-kin relationship between males buried in the same funerary circle (Bondioli et al., 1986). Increased social complexity was accompanied by changes in the ideology of proto-historic societies. The ideology of the aristocratic society of Iron Age proto-historic populations, including the Samnites, was based on extolling warfare and male audacity (Robb, 1994b). Warrior paraphernalia constitute the main male grave goods. At Alfedena, weapons of offense such as maces, swords, axes, spears, and javelins were commonly found as male grave goods (Parise Badoni and Ruggeri Giove, 1980), and often show signs of use in combat or training (Tagliamone, 1999). Several lines of evidence suggest that warfare and weapon use were not only an ideological and cultural matter, but were also part of the everyday activity repertoire among Iron Age communities. Accounts from early Roman historians report that Samnites ‘‘defended their settlements with the sturdy right arms of their men rather than with walls,’’ and ‘‘educated their boys in the Spartan manner’’ (Salmon, 1967; p 30). In central Italy, wars between neighboring tribal states were common through the ﬁrst millennium B.C. (D’Agostino, 1974; Pallottino, 1991). Before being subju- CSG CHANGES WITH IRON AGE SUBSISTENCE SHIFT gated by Romans, Samnites tended to raid their neighbors for loot rather than trading with them. Historical accounts suggest that plundering expeditions and retaliation by the victims were common (Salmon, 1967). Skeletal evidence of warfare activities can be found in the remarkable incidence of sword injuries and cranial trauma in many Iron Age necropoli, including Alfedena, especially in males (Macchiarelli et al., 1981; Robb, 1997; Paine et al., 2007). The nature and the severity of most of the Alfedena traumas (especially large blade and small projectile wounds) suggest that they were the result of interpersonal violence. Given the evidence of kin networks between males of the same community, it is likely that injuries were inﬂicted from males of rival groups, further supporting a scenario of a high level of inter-community warfare. Differences in activity between sexes in the Iron Age were a consequence of the social stratiﬁcation of a society in which moral coalitions were based on class and gender (Robb, 1994b). Primary production activities were performed by males, while females were relegated to time-consuming sedentary roles such as secondary products and textile processing, child rearing, and craftwork (Ehrenberg, 1989; Robb, 1994b). Expectations of the comparison with the Neolithic sample Although both Iron Age and Neolithic people performed pastoralism and agriculture (Marchi et al., 2006; Sparacello and Marchi, 2008), Iron Age socio-economic and technological changes are expected to have had an inﬂuence on habitual behavior. We expect to be able to detect those changes in CSG skeletal robusticity patterns. Theoretical predictions and archaeological evidence lead to the following testable predictions: 1. Alfedena males are expected to show a high level of humeral asymmetry in cross-sectional robusticity. Several historical accounts are suggestive of frequent use of unimanual weapons in battle and training. The comparison with the Ligurian Neolithic males should be informative, because this group shows a remarkable level of asymmetry possibly due to frequent use of stone hatchets in various activities (Sparacello and Marchi, 2008). 2. Alfedena females’ humeral asymmetry is expected to be higher in comparison to their Neolithic counterparts. Neolithic females show humeral asymmetry much lower than the values shown by sedentary modern samples. This has been attributed to daily stressful activities such as the use of bimanual stone querns for cereal grinding (Marchi et al., 2006; Sparacello and Marchi, 2008). In the Iron Age, the use of animal-driven querns for cereal grinding was introduced (Lynch and Rowland, 2005), which suggests that the use of manual stone querns was a less common activity by that time. Finding higher lateralization in Alfedena females would corroborate the hypothesis that the introduction of animal-driven querns freed Iron Age females from the strenuous task of manual cereal processing. This should result also in a decrease in humeral robusticity in the Alfedena female sample. 3. Both males and females of the Alfedena sample should show lower CSG indicators of mobility with respect to the Neolithic sample. Archaeological and historical evi- 121 dence strongly suggests that Alfedena economy was based on pastoralism and that agriculture was a secondary activity (Salmon, 1967; D’Ercole, personal communication). However, theoretical expectations for the Iron Age relative to the Neolithic predict agricultural intensiﬁcation due to demographic increase. With the transition to a state-level society, it is expected that herders and farming communities would become two parts of a more complex economy (Chang and Koster, 1986) in which herding forms the base of aristocracy’s wealth (Robb, 1994a). When this happens, herding is performed by specialized herders, a small subset of the population (Bonte, 1981). If this was the situation among the Samnites, most Alfedena males would have been involved in low-mobility agricultural tasks, and, at a population level, CSG indicators of mobility should be low. Females should show low mobility levels, being involved in sedentary, time-consuming processing of secondary products. In contrast, Neolithic people are considered to be more mobile, and it is likely that herding tasks were shared among most of the members of the community. Accordingly, lower limb CSG properties of the Neolithic Ligurian sample suggest high levels of mobility, especially in males (Marchi et al., 2006; Marchi, 2008). 4. Differences in activity patterns between sexes are expected in the CSG properties of the Alfedena sample. If upper limb robusticity was shaped by a gendered prestige activity such as weapon use and training, we expect to detect its effects only in males. This should lead to high sexual dimorphism in humeral bilateral asymmetry. Females’ indicators of mobility should be lower than males’, especially considering the ideal of restricting females to sedentary tasks that was present among Iron Age people (Robb, 1994b). MATERIALS AND METHODS The skeletal series studied here derives from excavations started at Alfedena in 1974. Archaeologists excavated 2,824 m2 of the necropolis, discovering 132 burials interred over a period of about 150 years between 2600 and 2400 BP (Parise Badoni and Ruggeri Giove, 1980). The burials were arranged in three well-deﬁned funerary circles, which reﬂect family/clan relationship among some individuals, as demonstrated also by the study of cranial epigenetic traits (Bondioli et al., 1986; Rubini, 1996). Numerous bioanthropological papers have been published regarding this skeletal series (see Rubini, 1996; Paine et al., 2007), which is considered a good cross-section of the Samnite population over a small time window. Grave good richness shows some variability across funerary circles and individuals, with the constant of the presence of offense weaponry in male tombs. Some of the requirements of CSG research restrict the sample of available bones. Only full adults (as judged by fully closed epiphyses) not showing sign of advanced senescence or manifest pathological conditions were sampled. Sex was inferred in previous research on the basis of the morphometric characteristics of the pubic bone (Washburn, 1948; Acsadi and Nemeskeri, 1970; Day 1975), and of the skull (Giles and Elliot, 1963; Boulinier, 1968; Acsadi and Nemeskeri, 1970; Demoulin, 1972). Close agreement between the two methods was found (Coppa and Macchiarelli, 1982; Coppa et al., 1990). Age was determined in previous research using osteometric and dental characteristics (Coppa and Macchiarelli, 1982) American Journal of Physical Anthropology 122 V.S. SPARACELLO ET AL. TABLE 1. Composition of the Alfedena Iron Age sample Specimen no. Sex 1 3 4 5 6 9 18 19 21 35 36 39 40 41 53 66 67 68 73 78 82 83 84 86 88 89 90 91 97 98 102 105 109 112 114 115 116 117 119 121 126 130 132 8 10 37 49 65 69 70 76 79 93 110 113 118 122 124 127 128 Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Female Female Female Female Female Female Female Female Female Female Female Female Female Female Female Female Female a TABLE 2. Composition of the Ligurian Neolithic sample Right humerus Left humerus Femur Tibia – – Xa X X X – X X X X – Xa X – X X X Xa X X – X X X X X X X X – X X Xa X X – X X X X X X X Xa – – X Xa X X X X – – X X X – X – X X – – X – Xa – X Xa X X – X X Xa X – X – Xa Xa X X – X X Xa – – X X X X – X X X X X X X X – – X X Xa X Xa X – – – Xa – – X X – X X X – X X – – – – – – X X X X X – X X X – – – – X X X X X – – X X X – X X X X X X X X X X X X X – X X X X X X X – X – X X X X – X X – – – X X – X X X X – X X X – X X X X X X X X X – X X X – X X X X X X X X – X – X X X X X X X – X X X Humerus with head damage (see text). following Miles (1963), Acsadi and Nemeskeri (1970), and Masset (1971), and by the analysis of vertical sections of humeral and femoral heads (Nemeskeri et al., 1960). To be positioned on the reference planes following Ruff and Hayes (1983) and Ruff (2002), bones must be fairly complete and present epiphyses in good condition. As a result, only 60 individuals (43 males and 17 females) American Journal of Physical Anthropology Specimen no. Sex Arene Candide 2 Tiné Arene Candide 7 Arene Candide 8 Arene Candide EVI Arene Candide IX Arene Candide XIIIb Arma dell’Aquila 2b Bergeggi 2b Bergeggi 3b Bergeggi 4b Cavernetta Boragni 2b Pollera 10 Pollera 13 Pollera 22 Pollera 30 Pollera 32b Pollera 6246 Arene Candide EVI Arene Candide XII Arma dell’Aquila 1 Arma dell’Aquila V Bergeggi 5b Cavernetta Boragni 1b Pollera 12 Pollera 14 Pollera 1 Tiné Pollera 33 Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Male Female Female Female Female Female Female Female Female Female Female a b Right Left humerus humerus Femur Tibia X X X X X X X Xa Xa X X X X X X Xa X X Xa Xa Xa X X X X X Xa Xa X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Humerus with head damage (see text). Individual added in this research. could be studied in this research, and not all bones were available for all individuals (Table 1). Thus, the samples evaluated for different locations and in different analyses (e.g., humeral bilateral asymmetry vs. humeral robusticity) were often not composed by the same individuals. This may potentially introduce noise into the results. The Ligurian Neolithic sample includes 10 additional individuals that augment the series used in previous research (Marchi, 2008; Sparacello and Marchi, 2008), for a total of 27 (17 males and 10 females, Table 2). Polysiloxane molds of the periosteal contour were taken at mid-distal humerus (35% bone length from the distal end), and midshaft femur and tibia (50% bone length), using bone lengths deﬁned by Ruff (2002). When present, both humeri were sampled to evaluate the degree of upper limb asymmetry. Since the lower limb does not present signiﬁcant lateralization (Auerbach and Ruff, 2004), only the right side was sampled when available, the left otherwise. Cross-sectional properties were calculated using a version of the program SLICE (Nagurka and Hayes, 1980), provided by Dr. Michael Black and adapted as a macro routine inserted in Scion Image release Beta 4.03. The variable expressing overall diaphyseal robusticity discussed in this study is calculated from the polar second moment of area J, corresponding to the torsional and (twice) average bending rigidity of the beam, raised to the power of 0.73 (Ruff, 1995, 2000b). The value of J0.73 is proportional to the section modulus (Zp), which can be more precisely calculated dividing J by the average radius of the section. Although J0.73 is proportional to rather than strictly equivalent to Zp, we will refer to it as Zp, as done in previous research (e.g., Maggiano et al., 2008). 123 CSG CHANGES WITH IRON AGE SUBSISTENCE SHIFT TABLE 3. Skeletal samples used to construct the regression equations Humeri Femora Tibiae Method Reference Ligurian Neolithic 30 16 13 Latex Cast Methoda Noli Medieval Alfedena Iron Age Neuburg Medieval 42 21 15 Direct sectioning Marchi et al., 2006; Marchi, 2008; Sparacello and Marchi, 2008 Sparacello and Marchi, 2008, and unpublished data Bondioli et al., 1986; Macchiarelli, 1988 Marchi, 2004 a Fibulae 30 Direct sectioning 37 Latex Cast Method O’Neill and Ruff, 2004. TABLE 4. Regression equations used to predict actual cross-sectional geometric properties from the periosteal section Model parameters and coefﬁcients c J Imax/Imind Ix/Iye Error n r2 Slope Intercept P %SEEa Mean absolute PPEb 204 174 173 0.999 0.996 0.994 0.994 1.012 1.001 20.002 20.007 20.0035 0.0000 0.0000 0.0000 4.07% 1.74% 1.85% 2.85% 1.26% 1.37% %SEE 5 exp (SEE 1 4.6052) 2 100; based on natural log-transformed data (Ruff, 2003). Absolute PPE 5(|estimate 2 true|/true) 3 100. Based on back-transformed natural log estimated and natural log true values. c ln J vs. ln periosteal J. d ln Imax/Imin vs. ln periosteal Imax/Imin. e ln Ix/Iy vs. ln periosteal Ix/Iy. a b The mechanical loading on long bones is a function of physical activity, bone length, and body mass (Ruff, 2000b). To isolate the effects of activity, the estimate of overall bone strength Zp was scaled for size dividing by bone mechanical length and body mass (Ruff, 2000b). Body mass was estimated in both samples from the femoral head superoinferior diameter (FH) using the formula from Grine et al. (1995). Unstandardized J was used to calculate humeral asymmetry. Given their correspondence with mobility levels in the lower limb, we will also discuss shape indices. For the femur, we consider both the ratio of Imax (maximum second moment of area) to Imin, (minimum second moment of area), and the ratio between Ix (second moment of area in the anterior–posterior plane) and Iy (second moment of area in the medio-lateral plane). In the tibia, the shape index Ix/Iy may be difﬁcult to measure precisely due to the more problematic positioning of this bone. Therefore, only the shape index Imax/Imin will be discussed. In the Alfedena sample, and for the Ligurian individuals added in this research (Table 2), CSG properties were estimated from periosteal contours using regression equations. The method is derived from the recent research of Stock and Shaw (2007), who analyzed the relationship between actual polar second moment of area (J) and shape indices (Imax/Imin and Ix/Iy) and a section’s external dimensions (diameters, circumference, subperiosteal area, and periosteal Imax/Imin and Ix/Iy). Their research demonstrated that subperiosteal area and periosteal shape indices can predict actual J and actual shape indices reasonably well. To obtain CSG properties from the mould of the periosteal contour, we developed similar equations based on actual CSG properties (hollow bone) and periosteal contour (solid bone), using data from skeletal samples previously studied by the authors (Table 3). We decided to develop our own equations, instead of using the ones published by Stock and Shaw, to minimize the error involved in estimating true cross-sectional properties from external dimensions. Our equations are based on skeletal samples that are geographically, temporally, and genetically the closest to the Alfedena sample among the CSG data available. Moreover, we could include in the regression equation 30 femora of the Alfedena sample that were sectioned at midshaft for previous investigations (Bondioli et al., 1986). Of these 30 femora, 23 are the contralateral side of the individuals studied here. Data from the remaining seven femora were not included in this study because the protocol used for positioning did not strictly correspond to the guidelines provided in Ruff and Hayes (1983) and Ruff (2002). We tested the regression method using the equations based on the Ligurian Neolithic and Noli Medieval samples on the Alfedena sectioned femora used as a control sample. We concluded that using bone, sex, and period speciﬁc equations does not produce appreciable differences in CSG estimates (Sparacello and Pearson, in press). We therefore used here a pooled-bone, pooled-sex equation already used in a research article currently in press, which also analyzes patterns of change in ﬁbular robusticity (Marchi et al., in press, and Table 4). Mean percent prediction errors obtained in the regressions, fall within the range of error generally accepted in CSG research when other methods to obtain cross-sections are used (Stock, 2002; O’Neill and Ruff, 2004; Stock and Shaw, 2007). In the Alfedena sample, 15 individuals had a damaged humeral head (Table 1). In these cases, the bone’s length (necessary to locate the 35% level) was estimated from the length of the humerus of the contralateral side using a regression equation calibrated on the rest of the sample (R2 5 0.923). The same procedure was applied in previous research to the eight individuals of the Neolithic sample presenting humeral head damage (R2 5 0.974), (Sparacello and Marchi, 2008). Statistical evaluation of differences in humeral, femoral, and tibial robusticity between the Alfedena and Neolithic samples was generally carried out using the Student’s t test. When the subsample under examination showed signiﬁcant deviation from normality (Lilliefors test), the nonparametric Mann-Whitney U tests were used. The same tests were used to evaluate differences American Journal of Physical Anthropology 124 V.S. SPARACELLO ET AL. TABLE 5. Differences among periods and sexes in humeral robusticity (Zp) Ligurian Neolithic Males Right Left Females Right Left Differences between sexesc Right Left Alfedena Iron Age Mean (n) SD Mean (n) SD Difference between periods 52.23a (14) 50.60 (14) 8.05 6.76 56.74 (35) 48.02 (30)c 8.92 6.52 NSb NSd 43.94 (9) 43.16 (8) 5.34 4.75 48.28 (12) 47.16 (10) 7.97 7.44 NSb NSb P \ 0.01b P \ 0.01b P \ 0.01b NSd Zp (J0.73) is standardized dividing by the product of body mass and bone mechanical length. Student’s t test. c Lilliefors normality test signiﬁcant (a 5 0.05). d Mann-Whitney U test. a b TABLE 6. Differences among periods and sexes in humeral bilateral asymmetry Ligurian Neolithic Alfedena Iron Age Median (n) Median (n) 18.52a (13) 3.46 (8) P \ 0.05d 24.3 (28)b 10.57 (10)b P \ 0.01c Males Females Difference between sexesc Difference between periods P \ 0.05c P \ 0.05c BA 5 Bilateral asymmetry calculated as: [((J max 2 J min)/J min) 3 100]. Nonstandardized J used for calculation. Lilliefors normality test signiﬁcant (a 5 0.05). c Mann-Whitney U test. d Student’s t test. a b TABLE 7. Humeral bilateral asymmetry of nonstandardized polar moment of inertia (J) among different human groups Median (n)a Males Middle Upper Paleolithicb Late Upper Paleolithicb Ligurian Neolithic Alfedena Iron Age Ligurian Medieval Fishergate Medievalc Amerindian Georgia Coastd Modern Euroamericanb Tennis playersd Females Middle Upper Paleolithic Late Upper Paleolithic Ligurian Neolithic Alfedena Iron Age Ligurian Medieval Amerindian Georgia Coast Modern Euroamerican Tennis players Upper and lower quartiles 58.7 51.2 18.52 24.3 12.53 20.28 9.90 7.5 74.56 (7) (9) (13) (28) (10) (19) (19) (19) (34) 39.9–96.4 45.7–97.4 6.04–24.65 15.30–42.83 4.02–17.13 8.03–26.29 – 3.1–21.8 – 27.3 22.3 3.46 10.57 6.78 4.94 11.7 39.14 (2) (5) (8) (10) (8) (18) (19) (11) 2.4–27.3 14.5–23.6 1.74–10.80 8.83–17.03 1.31–10.15 – 3.0–19.4 – Bilateral asymmetry calculated as: [((max 2 min)/min) 3 100]. Data from Churchill et al., 2000. Data provided by Dr. Jill Rhodes. d Data from Trinkaus et al., 1994. a RESULTS Humerus Tables 5 and 6 report temporal differences in humeral cross-sectional robusticity, bilateral asymmetry, and sexual dimorphism. Humeral robusticity (Zp) standardized for body size shows no changes among temporal periods in either sex. Within temporal period, sexual dimorphism (males more robust) is signiﬁcant on both sides in the Neolithic sample and on the right side in the Alfedena sample (Table 5). When compared to their Neolithic counterparts, both sexes of the Alfedena sample show signiﬁcantly higher bilateral asymmetry in nonstandardized robusticity (Table 6). Males are signiﬁcantly more lateralized than females for humeral J in both the Neolithic and Alfedena samples (Table 6). Table 7 compares bilateral asymmetry in CSG robusticity among different human groups. The male Alfedena sample is the most lateralized after Paleolithic huntergatherers and tennis players; Alfedena females show a level of lateralization comparable to modern industrialized people. b c in percentage data such as sexual dimorphism and bilateral asymmetry. Statistical analyses were performed with STATISTICA 7 (Statsoft, 2004). In addition to the samples studied here, broader comparisons of humeral bilateral asymmetry use data kindly provided by Dr. Jill Rhodes or drawn from the literature (Trinkaus et al., 1994; Churchill et al., 2000). American Journal of Physical Anthropology Lower limb Table 8 reports temporal differences in femoral crosssectional properties and sexual dimorphism. No signiﬁcant difference in cross-sectional robusticity is present among temporal group in either sex. However, the Alfedena sample (both sexes) has lower shape indices, and the difference is signiﬁcant for male Imax/Imin and Ix/Iy and female Ix/Iy. Femoral sexual dimorphism in the Alfedena sample is not signiﬁcant for either robusticity or shape values. 125 CSG CHANGES WITH IRON AGE SUBSISTENCE SHIFT TABLE 8. Differences among periods and sexes in diaphyseal geometric properties of femora Ligurian Neolithic Males Zp Imax/Imin Ix/Iy Females Zp Imax/Imin Ix/Iy Differences between sexesc Zp Imax/Imin Ix/Iy Alfedena Iron Age Mean (n) SD Mean (n) SD Difference between periods 96.56a (14) 1.49 (14) 1.34 (13) 13.75 0.23 0.2 100.11 (27)b 1.21 (27) 1.03 (27) 11.89 0.13 0.14 NSc P \ 0.0001d P \ 0.00001d 94.31 (10) 1.26 (10) 1.21 (10) 12.97 0.18 0.21 90.65 (15) 1.18 (15)b 0.95 (15) 16.64 0.14 0.15 NSd NSc P \ 0.01d NSd P \ 0.05d NSd NSc NSc NSd Zp (J0.73) is standardized dividing by the product of body mass and bone mechanical length. Lilliefors normality test signiﬁcant (a 5 0.05). c Mann-Whitney U test. d Student’s t test. a b TABLE 9. Differences among periods and sexes in diaphyseal geometric properties of tibiae Ligurian Neolithic Males Zp Imax/Imin Females Zp Imax/Imin Differences between sexesc Zpc,d Imax/Imin Alfedena Iron Age Mean (n) SD Mean (n) SD Difference between periodsa 105.07a (15) 2.53 (15) 13.97 0.44 94.52 (33) 2.27 (33) 9.95 0.37 P \ 0.01b P \ 0.05b 93.59 (10) 2.20 (10) 12.1 0.26 84.13 (14) 1.96 (14)c 16.35 0.43 NSb P \ 0.05d P \ 0.1b P \ 0.05b P \ 0.05b P \ 0.01d a Zp (J0.73) is standardized dividing by the product of body mass and bone mechanical length. Student’s t test. c Lilliefors normality test signiﬁcant (a 5 0.05). d Mann-Whitney U test. b Neolithic males show higher shape indices than females. Probably due to small sample size, the difference is signiﬁcant for Imax/Imin and not for Ix/Iy. Table 9 reports temporal differences in tibial cross-sectional properties and sexual dimorphism. Both sexes of the Alfedena sample show lower robusticity and shape indices when compared to their Neolithic counterparts. The difference is statistically signiﬁcant for male Zp adjusted for body size, and for Imax/Imin in both sexes. Differences between sexes within temporal period generally include higher Zp adjusted for body size and Imax/Imin in males. The difference is statistically signiﬁcant for the Ligurian Neolithic’s Imax/Imin (P \ 0.1 in Zp, probably due to the small sample size), and for both Zp and Imax/Imin in the Alfedena sample. DISCUSSION Humerus The lateralization index quantiﬁes the higher mechanical strength of the dominant arm, because of its preferential use in stressful activities. Modern industrialized people are considered here to show asymmetry values because of the physiological handedness in a context of absence of stressful activities, either unimanual or bimanual. This average value of asymmetry is not equal to zero, but between 8 and 12% (Trinkaus et al., 1994; Shaw and Stock, 2009). In the context of generally higher robusticity in preindustrial with respect to industrialized samples, lower asymmetry values in preindustrial samples can be considered as indicative of stressful bimanual tasks and higher values indicative of stressful asymmetric activities. High lateralization is common in groups whose behavioral repertoire included frequent and stressful unimanual tasks like asymmetric sports and throwing activities (Trinkaus et al., 1994; Churchill et al., 2000; Shaw and Stock, 2009). As noted by Churchill et al. (1996) and Rhodes and Knüsel (2005), the high loading rates and intermittent character of training correspond to the pattern of activities that best stimulate osteogenic response (Burr et al., 2002; Robling et al., 2002). Previous research demonstrated the higher robusticity of the Neolithic Ligurian people relative to other agricultural and industrialized groups (Sparacello and Marchi, 2008). The Alfedena sample shows within-sexes humeral robusticity comparable to Neolithic people. This suggests that both groups conducted a lifestyle placing high levels of mechanical stresses on the upper limb. It is therefore likely that their level of humeral asymmetry is a consequence of stressful activity patterns. Alfedena males are greatly lateralized, signiﬁcantly more so than Neolithic males. The remarkable level of asymmetry in Neolithic males was attributed to land American Journal of Physical Anthropology 126 V.S. SPARACELLO ET AL. clearing for agriculture, which in horticultural societies is a male activity (Ehrenberg, 1989). Since Liguria lacks natural pastures, it has been proposed that tree pollarding was practiced to procure branches, twigs, and leaves which constituted additional fodder to herds, as demonstrated by micromorphological soil analyses (Macphail et al., 1997; Maggi and Nisbet, 2000). Small eclogite stone hatchets characterize Neolithic layers from Liguria, and wear patterns indicate that they were used for woodworking (Starnini and Voytek, 1997). Among Alfedena people, manual deforestation was not a common activity because their multicrop agriculture had advanced well beyond the slash and burn stage. Herds were moved between seasonal pastures, and providing fodder through tree pollarding was not necessary. Despite this, Alfedena males show higher lateralization, which is therefore likely not due to hatchet use. When compared with other human groups, Alfedena males are about three times more asymmetric than modern Euroamericans and are the most lateralized after professional tennis players and Paleolithic hunters. However, Alfedena values are very distant from the values expressed by tennis players, and are about half of the values shown by Paleolithic hunters, whose activity patterns include the most extreme degree of preferential use of the dominant arm, most likely because their hunting strategy was based on hand-thrown projectile weapons (Trinkaus et al., 1994; Churchill et al., 1996, 2000). The level of asymmetry at Alfedena seems comparable to— yet somewhat higher than—the pattern shown by the medieval Fishergate group (Rhodes and Knüsel, 2005). Other similarities between the two samples include the presence of swords as grave goods, and a high incidence of perimortal sword wounds in the skeletal series. In the Medieval period relevant to the Fishergate sample, English legal proclamations prescribed that every man was required to have and be proﬁcient in some weapon type, according to their social status (Rhodes and Knüsel, 2005). Given the shared characteristics between the two samples, it is therefore likely that their similar level of asymmetry is due to frequent weapon training. Thus, Alfedena males’ high humeral asymmetry corroborates the hypothesis that weapon training was a common activity among Iron Age males, and ﬁts with the reconstruction of a warlike society in which such prestige activities were part of the habitual behavioral repertoire. The lateralization of the female Alfedena sample is similar to the modern industrialized sample. This, coupled with the high robusticity values shown by this group, suggests a balanced presence of unimanual and bimanual stressful activities. When compared with Neolithic females, Alfedena females show a signiﬁcantly higher lateralization. We suggest that this pattern is the skeletal corroboration of the changes in cereal processing technology inferred from archaeology. Cross-culturally, processing vegetable foods is a female activity (Ehrenberg, 1989). In the Neolithic, cereal grinding was performed by means of bimanual grinding stones, and most likely took hours every day (Robb, 2007). This strenuous and time-consuming task placed equally high mechanical loads on females’ upper limbs, and contributed to generate an average level of asymmetry close to zero (Marchi et al., 2006; Sparacello and Marchi, 2008). A correlation between manual cereal grinding and female low asymmetry has been found also in other researches (Bridges et al., 2000; Weiss, 2009). In the Iron Age, cereal processing was performed mainly through the use of animalAmerican Journal of Physical Anthropology driven querns (Lynch and Rowland, 2005). This made the use of bimanual querns for cereal processing less common, resulting in a female level of asymmetry close to the one shown by modern sedentary samples. However, the high level of robusticity in Alfedena females— comparable to the one shown by Neolithic females—suggests a balanced presence of stressful unimanual and bimanual activities, rather than absence of stressful activities. This result contradicts our expectation that Alfedena females should have shown lower robusticity when compared to Neolithic females, and hints that in the Iron Age certain subsistence-related activities—such as textile production, food processing, and possibly other hitherto unrecognized tasks—placed a high degree of mechanical stress on female upper limbs. Lower limb Abruzzo and Liguria are both very mountainous areas; comparisons between the two groups factor out the possible inﬂuence of terrain on lower limb robusticity (Ruff, 1999). According to Ruff (2000a), once the inﬂuence of terrain is factored out, the effect of different subsistence activities on femoral robusticity greatly declines. Since the Ligurian and Alfedena people shared the part of their behavioral repertoires related to pastoral and agricultural activities, it is not surprising that their femoral robusticity values are similar within sexes between temporal groups. In contrast to femoral torsional robusticity, the shape of lower limb midshaft sections may correlate better with mobility levels (i.e., daily travel). Biomechanical studies show that activities such as climbing and running cause high AP bending loads from the femoral to the tibial midshaft (e.g., Burr et al., 1996). Therefore, elliptical-shaped (and AP-oriented) lower limb midshafts are often considered characteristic of highly mobile populations. Bioarchaeological research is consistent with this framework, showing that lower limb shape indices correlate with mobility patterns (Ruff, 1999; Holt, 2003; Stock, 2006). In addition to lower limb shape indices, tibial robusticity also shows a correlation with mobility. This is likely due to the fact that distal elements show greater mass optimization between tissue economy and safety factors (Stock, 2006). Thus, it is expected that more sedentary groups will have on average lower shape indices in the femur and tibia, and lower tibial robusticity when compared to mobile groups. Archaeological and historical evidence strongly suggests that in both groups under examination ovine stockbreeding was the most important economic activity, while agriculture was a marginal activity (Salmon, 1967; Marchi et al., 2006; Sparacello and Marchi, 2008; D’Errico, personal communication). However, we predicted a decrease in average mobility in the Alfedena sample because of the theoretical expectation of alterations in activity patterns due to Iron Age socio-economic changes. Results indicate that ratio of (average male)/ (average female) femoral Ix/Iy is 1.11 in the Ligurians, and 1.08 in the Alfedena sample. These values fall in the range near the ‘‘transition’’ between hunter-gatherers and agriculturalists, as measured in a number of samples (Ruff, 2008). This reﬂects the mixed subsistence economy of both our samples, and as expected the Alfedena value is slightly lower. The difference in mobility patterns between the two periods is more apparent when CSG CHANGES WITH IRON AGE SUBSISTENCE SHIFT analyzing within-sexes comparisons: in the Alfedena group all CSG indicators of mobility are lower when compared to the sex-matched Neolithic samples, and most differences are statistically signiﬁcant. Both femoral and tibial properties depict a coherent scenario of decreased mobility in Alfedena males and females with respect to their Neolithic counterparts. Results therefore conﬁrm the theoretical expectations of socio-economic changes in the Iron Age: the economic importance of herding among the Alfedena people did not result in a mobile lifestyle at a population level, at least when compared to the Neolithic. This is likely due to two contributing factors. Although there is little emphasis on agriculture in the Samnite archaeological and historical record, due to demographic pressure, agricultural and pastoral intensiﬁcation led to a more sedentary lifestyle overall in comparison to the Neolithic. Cross-culturally, with the shift from ‘‘hoe and the digging stick’’ horticulture typical of the Neolithic to ‘‘plough and oxen’’ agriculture typical of later periods, males took over most of the sedentary agricultural tasks. At the same time, the intensive exploitation of secondary animal products among Samnites led to an increase of time-consuming and sedentary activities such as textile production, which were generally performed by females (Ehrenberg, 1989). Moreover, the degree to which people were involved in the high-mobility tasks typical of herding may have been different in the two groups under examination. Although both groups compared here can be broadly deﬁned as pastoral (Chang and Koster, 1986), their subsistence probably took different forms, resulting in less average mobility in the Iron Age. In the Neolithic, short-range transhumance of small ﬂocks was integrated into an agricultural-based subsistence economy. Herding was part of a complex of shared habitual subsistence activities which involved each member of the village more or less equally (Robb, 2007). This contributed to greater average mobility in the population, especially in males who usually took care of herding (Dyson-Hudson and DysonHudson, 1980). In the Iron Age, transhumance functioned as part of a capitalistic market economy forming the base of aristocracy’s wealth (Robb, 1994a). Pastoralism was performed by a small subset of specialized herders, and become a distinct and separated activity with respect to farming activities (Bonte, 1981; Chang and Koster, 1986). This process was still incipient in the short-distance, vertical transhumance of Alfedena Samnites, and become fully developed only later, after Roman conquest (Salmon, 1967; Palasciano, 1999; D’Ercole, personal communication). However, our results showing lower indicators of mobility in males with respect to the Neolithic are compatible with a scenario where only a small subset of the male population was involved in the moving of herds to the pastures. This leads to lower average indicators of mobility at a population level, because mobile individuals are less likely to be sampled. The inclusion in the sample of individuals that performed high mobility tasks (herders in this case) should result in high variability in the male lower limb shape indices. Past research detected increased variability in femoral midshaft Ix/Iy in the Native American population of La Florida due to postcontact forced labor, which resulted in increased variation in activity patterns (Ruff and Larsen, 2001). In our comparisons, male SD in shape indices is lower when compared with their Neolithic counterpart. A possible explanation for this result is that few if any herders were sampled in our research. 127 Sexual dimorphism Differences in gender roles and behavior typical of the Western world had its roots in the Neolithic (Ehrenberg, 1989). Most ethnoanthropological studies of pastoral societies evidence males’ involvement in herding activities, while women’s activities appear to be mainly domestic (childcare, household activities), agricultural, and involving the processing of secondary products (see Dyson-Hudson and Dyson-Hudson, 1980). On this basis, sexual dimorphism in skeletal properties is expected to be detectable in both samples compared here. The Italian Neolithic ideology appears to have been weakly genderized. The similarity in grave goods between sexes suggests that the Neolithic culture conceptualized gender as an opposition between balanced and complementary factors (Robb, 1994b, 2007). Some evidence of space genderization consists in male-associated imagery located in the periphery of occupational areas, while the female focus appears to have been closer to villages. Male activities probably included hunting and herding further aﬁeld, while females were more involved in activities performed close to the village, such as food processing, gathering, and agriculture (Robb, 1994b, 2007). This suggests greater male mobility in an overall mobile context. In horticultural societies, in which hoes or digging sticks are used for seeding, women are usually almost wholly responsible for agriculture, while men perform land clearing (Ehrenberg, 1989). This should generate males with more robust and lateralized humeri, and that is especially true in Liguria, where manual deforestation practices were important for both land clearing and fodder procurement. Sexual dimorphism in CSG properties of the Ligurian Neolithic group has been discussed elsewhere (Marchi et al., 2006; Sparacello and Marchi, 2008; Marchi, 2008). The present analysis including additional individuals substantially conﬁrms previous ﬁndings, and corroborates the above expectations. Males are signiﬁcantly more robust and lateralized than females; moreover, males show some evidence of greater mobility (signiﬁcantly higher femoral Imax/Imin, higher tibial Zp, and signiﬁcantly higher tibial Imax/Imin) in a generally mobile group. The ideology of Alfedena people was strongly genderized. The aristocratic stratiﬁcation of Iron Age protostate societies was held together by moral coalitions based on class and gender (Robb, 1994b). According to the masculine ideal celebrated in material art through paintings, and in literature, males were involved in all the activities where animal tenure was involved, which included plough agriculture. Moreover, weapon possession and use were male prestige-related activities. The feminine role in such a gendered society consisted of performing ﬁne handiwork in seclusion (Robb, 1994b). Accordingly, Iron Age female burials were marked by spindle whorls, suggesting that wool processing constituted a great part of their activity repertoire (Bartoloni, 1988; Bietti Sestieri, 1993). However, Samnites followed a rough peasant life, and only a small subset of wealthy women could have afforded to perform light jobs such as ﬁne handcrafting (Robb, 1994b). As suggested also by the high level of humeral robusticity in absolute terms shown by Alfedena females, most of the women were likely involved in the stressful and time-consuming—yet sedentary—tasks typical of secondary products processing, including vegetable and animal food processing, and weaving. American Journal of Physical Anthropology 128 V.S. SPARACELLO ET AL. Results obtained in this research are consistent with the non-CSG based reconstruction: males show statistically signiﬁcantly higher lateralization, and signiﬁcantly more robust right humeri. This is consistent with the hypothesized frequent weapon training among males. Lower mobility in the female sample is suggested by signiﬁcantly lower average tibial robusticity and shape index. The femoral indicators of mobility are not signiﬁcantly different between sexes. This is likely due to the fact that the Alfedena group was sedentary, and therefore femoral shape indices are close to circularity. Tibial indicators of mobility allow for a greater range of differences between groups, being based on robusticity measures and on shapes normally far from circularity. A greater sensitivity of tibial CSG properties to variations in mobility levels may also be due to the fact that the distal limb segment is less inﬂuenced by variations in body shape (Ruff et al., 2006a), and shows greater optimization for tissue economy (Stock, 2006). Alfedena females therefore show evidence of being very sedentary in an overall nonmobile population. CONCLUSIONS The biomechanical analysis of the Alfedena skeletal series corroborated our expectations about subsistence changes for the shift from Neolithic to Iron Age societies. The comparison of Alfedena people with a group from earlier prehistory with similar subsistence base provided an example of how increased socio-economic complexity contributes in determining changes in the habitual behavior of agropastoral groups. Although archaeological and historical evidence stress the economic importance of pastoralism among Samnites, we expected a decrease in mobility when compared to the Neolithic due to agricultural intensiﬁcation and incipient pastoral specialization. Accordingly, lower limb robusticity and shape indices in the later population mirror a decrease in the average level of mobility in both males and females, which is consistent with an economic system in which high mobility pastoral tasks were performed by a small subset of the population, and most people were less mobile. The analysis of upper limb robusticity and asymmetry conﬁrmed that males performed stressful unimanual activities, which produces skeletal evidence supporting frequent weapon training, a prestige-related male activity typical of Iron Age warlike societies. The increase in female lateralization with respect to the Neolithic likely mirrors the shift from manual to animal-driven cereal processing. The remarkable sexual dimorphism in both upper and lower limb CSG properties mirrors a genderized society, where males performed most of the production and status-related activities and females were relegated to time-consuming sedentary tasks. Results of the comparison of cross-sectional properties provided skeletal quantitative data on the above parameters, which can be used to construct a comparative framework for other prehistoric groups. This allows for further explorations of subtle differences in the behavioral adaptations of agricultural and pastoral groups due to changing environments, and socio-economic levels of complexity. ACKNOWLEDGMENTS The authors thank Robert R. Paine for suggesting the idea for this study. Thanks to Jill Rhodes for providing American Journal of Physical Anthropology her raw data on the Fishergate sample and for useful comments during the writing of this article. Heather Edgar, Roberto Macchiarelli, Martin Muller, and Sherry Nelson provided valuable comments during the writing of the article. Domenico Mancinelli was very helpful during data collection at L’Aquila. They thank Vincenzo D’Ercole, Inspector Archaeologist for the Abruzzo Region, for providing useful information about Alfedena archeological context and Samnite subsistence. Thanks to the two anonymous reviewers and to the editor of this journal for the useful comments that signiﬁcantly improved the earliest version of this article. LITERATURE CITED Acsadi G, Nemeskeri J. 1970. History of human life span and mortality. Budapest: Akademiai Kiado. Auerbach BM, Ruff CB. 2004. Human body mass estimation: a comparison of ‘‘morphometric’’ and ‘‘mechanical’’ methods. Am J Phys Anthropol 125:331–342. Bartoloni G. 1988. A few comments on the social position of women in the protohistoric coastal areas of western Italy made on the basis of a study of funerary goods. Riv Antropol 66:317–336. Bietti-Sestrieri AM. 1993. The Iron Age community of Osteria dell’Osa. Cambridge: Cambridge University Press. Bondioli L, Corruccini RS, Macchiarelli R. 1986. Familial segregation in the Iron Age community of Alfedena, Abruzzo, Italy, based on osteodental trait analysis. Am J Phys Anthropol 71:393–400. Bonte P. 1981. Ecological and economic factors in the determination of pastoral specialization. J Asian Afr Stud 16:33–49. Boserup E. 1965. The conditions of agricultural growth: the economics of agrarian change under population pressure. London: Allen and Unwin. Boserup E. 1975. The impact of population growth on agricultural output. Quart J Agr Econ 89:257–270. Boulinier G. 1968. La determination du sexe des cranes humains a l’aide dex functions discriminates. Bull Mem Soc Anthropol Paris 3:301–316. Bridges PS. 1989. Changes in activities with the shift to agriculture in the southeastern United States. Curr Anthropol 30:385–394. 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. Burr DB, Milgrom C, Fyhrie D, Forwood M, Nyska M, Finestone A, Hoshaw S, Saiag E, Simkin A. 1996. In vivo measurement of human tibial strains during vigorous activity. Bone 18:405–410. Burr DB, Robling AG, Turner CH. 2002. Effects of biomechanical stress on bones in animals. Bone 30:781–786. Chang C, Koster HA. 1986. Beyond bones: toward an archaeology of pastoralism. Adv Archaeol Method Theory 9:97–146. Churchill SE, Formicola V, Holliday TW, Holt B, Schumann BA. 2000. The Upper Paleolithic population of Europe in an evolutionary perspective. In: Roebroeks W, Mussi M, Svoboda J, Fennema K, editors. Hunters of the golden age-the midupper Palaeolithic of Eurasia 30,000–20,000 bp. Leiden: University of Leiden. p 31–57. Churchill SE, Weaver AH, Niewoehner WA. 1996. Late Pleistocene human technological and subsistence behavior: functional interpretations of upper limb morphology. Quaternaria Nova 6:413–447. Coppa A, Colarossi P, Danubio ME, Mancinelli D, Petrone PP. 1990. Aspetti paleodemograﬁci in campioni di popolazione adulta dell’Italia Centrale durante l’Età del Ferro. Antropol Contemp 13:179–191. Coppa A, Macchiarelli R. 1982. The maxillary dentition of the iron-age population of Alfedena (middle-Adriatic area, Italy). J Hum Evol 11:219–235. CSG CHANGES WITH IRON AGE SUBSISTENCE SHIFT D’Agostino, B. 1974. Il mondo periferico della Magna Grecia. In: D’Agostino B, Arias PE, Colonna G, editors. Popoli e civiltà dell’Italia antica. Roma: Biblioteca di Storia Patria. p 177–272. Day MH. 1975. Sexual differentiation in the innominate bone studied by multivariate analysis. Ann Hum Biol 2:143–151. Demoulin F. 1972. Importance de certaines measures craniennes (en particulier de la longueur sagittale de la mastoide) dans la determination sexuelle des cranes. Bull Mem Soc Anthropol Paris 9:259–264. Dyson-Hudson R, Dyson-Hudson N. 1980. Nomadic pastoralism. Annu Rev Anthropol 9:15–61. Ehrenberg M. 1989. Women in prehistory. Norman: University of Oklahoma Press. Giles E, Elliot O. 1963. Sex determination by discriminant function analysis of crania. Am J Phys Anthropol 21:53–58. Grine FE, Jungers WL, Tobias PV, Pearson OM. 1995. Fossil Homo femur from Berg Aukas, northern Namibia. Am J Phys Anthropol 26:67–78. Guidi A. 2000. Preistoria della complessità sociale. Bari: Laterza. Holt BM. 2003. Mobility in Upper Paleolithic and Mesolithic Europe: evidence from the lower limb. Am J Phys Anthropol 122:200–215. 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 77:107–133. Lynch AJ, Rowland CA. 2005. The history of grinding. Littleton: Society for Mining Metallurgy and Exploration. Macchiarelli R. 1988. Age-related rates and patterns of cortical bone involution in past human populations: a protohistorical Italian example. Riv Anthropol 66:55–76. Macchiarelli R, Salvadei L, Dazzi M. 1981. Paleotraumatologia cranio-celebrale nella comunità protostorica di Alfedena (VI-V sec. a.C., area medio-adriatica). Antropol Contemp 4:239–243. Macphail RI, Courty MA, Hather J, Wattez J. 1997. The soil micromorphological evidence of domestic occupation and stabling activities. In: Maggi R, editor. Arene Candide: a functional and environmental assessment of the Holocene sequence (excavations Bernabó Brea-Cardini 1940–1950). Mem Ist Ital Paleont Um 5:53–88. Maggi R. 1997. The radiocarbon chronology. In: Maggi R, editor. Arene Candide: a functional and environmental assessment of the Holocene sequence (excavations Bernabò Brea-Cardini 1940–50). Mem Ist Ital Paleont Um 5:31–52. Maggi R, Nisbet R. 2000. Alberi da foraggio e scalvatura neolitica: nuovi dati dalle Arene Candide. In: Pessina A, Muscio G, editors. La neolitizzazione tra oriente e occidente, atti convegno di studi. Udine: Edizioni del Museo Friulano di Storia Naturale. p289–308. Maggiano IS, Schultz M, Kierdorf H, Sierra Sosa T, Maggiano CM, Tiesler Blos V. 2008. Cross-sectional analysis of long bones, occupational activities and long-distance trade of the Classic Maya from Xcambó—archaeological and osteological evidence. Am J Phys Anthropol 136:470–477. Marchi D. 2004. Cross-sectional geometry of the limb bones of the Hominoidea: its relationships with locomotion and posture. Ph.D. dissertation, University of Pisa, Pisa, Italy. Marchi D. 2008. Relationships between lower limb cross-sectional geometry and mobility: the case of a Neolithic sample from Italy. Am J Phys Anthropol 137:188–200. Marchi D, Sparacello VS, Holt BM, Formicola V. 2006. Biomechanical approach to the reconstruction of activity patterns in Neolithic western Liguria, Italy. Am J Phys Anthropol 131:447–455. Marchi D, Sparacello VS, Shaw CN. Mobility and lower limb robusticity of a pastoralist Neolithic population from northwestern Italy. In: Pinhasi R, Stock JT, editors. Human bioarchaeology of the transition to agriculture. New York: WileyLiss (in press). Mariani L. 1901. ‘‘Auﬁdena,’’ ricerche archeologiche e storiche del Sannio settentrionale. Roma: Acc Naz Dei Lincei. Masset C. 1971. Erreurs systematiques dans la determination de l’age par les sutures craniennes. Bull Mem Soc Anthropol Paris 7:85–105. 129 Miles AEW. 1963. The dentition in the assessment of individual age in skeletal material. In: Brothwell DR, editor. Dental anthropology. New York: Pergamon. p 191–209. Nagurka ML, Hayes WC. 1980. An interactive graphics package for calculating cross-sectional properties of complex shapes. J Biotech 13:59–64. Nemeskeri J, Harsanyi L, Acsadi G. 1960. Methoden zur diagnose des Lebensalters von Skelettfunden. Anthropol Anz 24:70–95. O’Neill MC, Ruff CB. 2004. Estimating human long bone crosssectional geometric properties: a comparison of noninvasive methods. J Hum Evol 47:221–235. Paine R, Mancinelli D, Ruggieri M, Coppa A. 2007. Cranial trauma in Iron Age Samnite agriculturists, Alfedena, Italy: implications for biocultural and economic stress. Am J Phys Anthropol 132:48–58. Palasciano I. 1999. Le lunghe vie erbose (Tratturi e pastori del Sud). Lecce: Capone Editore. Pallottino M. 1991. History of Earliest Italy. Ann Arbor: University of Michigan Press. Parise Badoni F, Ruggeri Giove M. 1980. Alfedena. La necropoli di Campo Consolino. Scavi 1974–1979. Chieti: Soprintendenza Archeologica dell’Abruzzo. 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. Peroni R. 1979. From Bronze Age to Iron Age: economic, historical, and social considerations. In: Ridgway D, Francesca R, editors. Italy before the Romans: the Iron Age, Orientalizing, and Etruscan periods. London: Academic Press. p7–30. Peroni R. 1989. Protostoria dell’Italia continentale. La penisola italiana nelle età del Bronzo e del Ferro. Popoli e Civiltà dell’Italia Antica Vol. 9. Roma: Biblioteca di Storia Patria. Rhodes JA, Knüsel CJ. 2005. Activity related skeletal change in medieval humeri: cross-sectional and architecture alterations. Am J Phys Anthropol 128:536–546. Robb J. 1994a. Skeletal signs of activity in the Italian Metal Ages: methodological and interpretative notes. Hum Evol 3:215–229. Robb J. 1994b. Gender contradictions: moral coalitions and inequality in prehistoric Italy. J Eur Archaeol 2:20–49. Robb J. 1997. Violence and gender in early Italy. In: Martin D, Frayer D, editors. Troubled times: violence and warfare in the past. Toronto: Gordon and Breach. p 111–144. Robb J. 2007. The Early Mediterranean village—agency, material culture, and social change in Neolithic Italy. Cambridge: Cambridge University Press. Robling AG, Hinant FM, Burr DB, Turner CH. 2002. Improved bone structure and strength after long-term mechanical loading is greatest if loading is separated into short bouts. J Bone Miner Res 17:1545–1554. Rubini M. 1996. Biological homogeneity and familial segregation in the Iron Age population of Alfedena (Abruzzo, Italy), based on cranial discrete traits analysis. Int J Osteoarchaeol 6:454–462. Ruff CB. 1995. Biomechanics of the hip and birth in early Homo. Am J Phys Anthropol 98:527–574. 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. 2000a. 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. 2000b. Body size, body shape, and long bone strength in modern humans. J Hum Evol 38:269–290. Ruff CB. 2002. Long bone articular and diaphyseal structure in Old World monkeys and apes. I: locomotor effects. Am J Phys Anthropol 119:305–342. Ruff CB. 2003. Long bone articular and diaphyseal structure in Old World monkeys and apes. II: estimation of body mass. Am J Phys Anthropol 120:16–37. Ruff CB. 2008. Biomechanical analyses of archeological human skeletal samples. In: Katzenberg MA, Saunders SR, editors. Biological anthropology of the human skeleton, 2nd ed. New York: Wiley. p 183–206. American Journal of Physical Anthropology 130 V.S. SPARACELLO ET AL. Ruff CB, Hayes WC. 1983. Cross-sectional geometry of the Pecos Pueblo femora and tibiae—a biomechanical investigation. I: methods and general pattern of variation. Am J Phys Anthropol 60:359–381. Ruff CB, Holt BM, Sládek V, Berner M, Murphy WA, Nedden D zur, Seidler H, Reicheis W. 2006a. Body size, body shape, and long bone strength of the Tyrolean ‘‘Iceman.’’ J Hum Evol 51:91–101. Ruff CB, Holt BM, Trinkaus E. 2006b. Who’s afraid of the big bad Wolff? ‘‘Wolff ’s law’’ and bone functional adaptation. Am J Phys Anthropol 129:484–498. Ruff CB, Larsen CS. 2001. Reconstructing behavior in Spanish Florida: the biomechanical evidence. In: Larsen CS, editor. Bioarchaeology of Spanish Florida: The Impact of Colonialism. Gainesville: University Press of Florida. p 113–145. 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, Trinkaus E, Walker A, Larsen CS. 1993. Postcranial robusticity in Homo. I: temporal trends and mechanical interpretation. Am J Phys Anthropol 91:21–54. Salmon ET. 1967. Samnium and the Samnites. Cambridge: Cambridge University Press. Shaw CN, Stock JT. 2009. Habitual throwing and swimming correspond with upper limb diaphyseal strength and shape in modern human athletes. Am J Phys Anthropol 140:160–172. Sparacello VS, Marchi D. 2008. Mobility and subsistence economy: a diachronic comparison between two groups settled in the same geographical area (Liguria, Italy). Am J Phys Anthropol 136:485–495. Sparacello VS, Pearson OM. The importance of accounting for the area of the medullary cavity in cross-sectional geometry: a test based on the femoral midshaft. Am J Phys Anthropol. DOI: 10.1002/ajpa.21361 American Journal of Physical Anthropology Starnini A, Voytek B. 1997. New lights on old stones. In: Maggi R, editor. Arene Candide: a functional and environmental assessment of the Holocene sequence (excavations Bernabó Brea-Cardini 1940–50). Mem Ist Ital Paleont Um 5:427–511. Statsoft. 2004. Statistica. Tulsa: Statsoft. Stock JT. 2002. A test of two methods of radiographically deriving long bone cross-sectional properties compared to direct sectioning of the diaphysis. Int J Osteoarchaeol 12:335–342. Stock JT. 2006. Hunter-gatherer postcranial robusticity relative to patterns of mobility, climatic adaptation, and selection for tissue economy. Am J Phys Anthropol 131:194–204. Stock JT, Shaw CN. 2007. Which measures of skeletal robusticity are robust? A comparison of external methods of quantifying diaphyseal strength to cross-sectional geometric properties. Am J Phys Anthropol 134:412–423. Tagliamone G. 1999. Le Armi: Lo sviluppa di una societa aristocratica; il ruolo delle armi. In: Piceni Popolo d’Europa. Roma: Editrice De Luca. p 112–114. Trinkaus E, Churchill SE, Ruff CB. 1994. Postcranial robusticity in Homo. II: humeral bilateral asymmetry and bone plasticity. Am J Phys Anthropol 93:1–34. Vida Navarro M. 1993. Warriors and weavers: sex and gender in Early Iron Age graves from Pontecagnano. J Accordia Res Center 3:67–100. Washburn SL. 1948. Sex differences in the pubic bone. Am J Phys Anthropol 6:199–207. Weiss E. 2009. Sex differences in humeral bilateral asymmetry in two hunter-gatherer populations: California Amerinds and British Columbian Amerinds. Am J Phys Anthropol 140:19– 24. Wescott DJ. 2006. Effect of mobility on femur midshaft external shape and robusticity. Am J Phys Anthropol 130:201–213.