Were Neandertal Humeri Adapted for Spear Thrusting Or Throwing? a Finite Element Study

University of Massachusetts Amherst ScholarWorks@UMass Amherst

Masters Theses Dissertations and Theses

November 2014

Were Neandertal Humeri Adapted for Spear Thrusting or Throwing? A Finite Element Study

Michael Anthony Berthaume University of Massachusetts Amherst

Follow this and additional works at: https://scholarworks.umass.edu/masters_theses_2

Part of the Biological and Physical Anthropology Commons, Biomechanical Engineering Commons, Biomechanics and Biotransport Commons, Evolution Commons, and the Other Ecology and Evolutionary Biology Commons

Recommended Citation Berthaume, Michael Anthony, "Were Neandertal Humeri Adapted for Spear Thrusting or Throwing? A Finite Element Study" (2014). Masters Theses. 72. https://doi.org/10.7275/5952474 https://scholarworks.umass.edu/masters_theses_2/72

This Open Access Thesis is brought to you for free and open access by the Dissertations and Theses at ScholarWorks@UMass Amherst. It has been accepted for inclusion in Masters Theses by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact [email protected].

WERE NEANDERTAL HUMERI ADAPTED FOR SPEAR THRUSTING OR THROWING? A FINITE ELEMENT STUDY

A Thesis presented

MICHAEL A. BERTHAUME

Submitted to the Graduate School of the University of Massachusetts Amherst in partial fulfillment of the requirements for the degree of

MASTER OF ARTS

September 2014

Department of Anthropology

WERE NEANDERTAL HUMERI ADAPTED FOR SPEAR THRUSTING OR THROWING? A FINITE ELEMENT STUDY

A Thesis Presented

MICHAEL A. BERTHAUME

Approved as to style and content by:

______Brigitte M. Holt, Chair

______Laurie R. Godfrey, Member

______Michael O. Sugerman, Member

______Thomas Leatherman, Department Head Department of Anthropology

DEDICATION

To my amazing family,

Mom, Dad, Andy, Maria, Angie, there is no way I could be where I am today without your amazing love and support, whether it be taking me in, helping me move, listening to me drone on about teeth and FEA of biological systems, or making me crack up at one of our three hour long family dinners.

Together, we make the craziest, happiest, and most powerful and loving group of people I know.

Thank you.

Insanity runs in my family. It practically gallops. -Cary Grant

ACKNOWLEDGEMENTS

When I started in the anthropology program at the University of Massachusetts,

Amherst, I had three goals. First, to learn about the paleoanthropological record.

Second, to learn how to teach. And third, to learn how to ask a relevant question. While

I will never be able to fully achieve these goals in my lifetime, I am considerably closer to achieving them than I was three years ago.

This thesis would not have been possible without the incredible support and guidance granted to me by my advisor, Dr. Brigitte Holt, and my committee members,

Drs. Laurie Godfrey and Michael Sugerman, or finite element training I received from

Drs. Ian Grosse, Elizabeth Dumont, and the Biomesh Workshop. For that I am forever in their debt.

I would like to extend my thanks to NESPOS for access to Regourdou 1’s scans,

Drs. Colin Shaw and Tim Ryan for access to the Norris Farms scans, and Dr. Brigitte

Holt for access to Grotte des Enfants 4’s scans. In addition, I would like to thank Dr.

Elizabeth R. Dumont and Dan Pulaski in the Mammology Department at UMass,

Amherst for access to the computer programs necessary to construct the finite element models, and Carl Jewell and Dr. Joe Hamill for help in collecting the EMG data in the

Biomechanics Lab at UMass, Amherst. Finally, I would like to thank the University of

Massachusetts, Amherst for granting the Faculty Research Grant/Healey Endowment grant to Dr. Brigitte Holt in order to obtain the scans of Grotte de Enfants 4, Prof.

Patrick Simon (Musée d'Anthropologie Préhistorique de Monaco) for access to GDE 4,

Centre d'Imagerie Médicale de Monaco for taking the CT scans.

ABSTRACT WERE NEANDERTAL HUMERI ADAPTED FOR SPEAR THRUSTING OR THROWING? A FINITE ELEMENT STUDY

SEPTEMBER 2014

MICHAEL A. BERTHAUME

B.S., UNIVERSITY OF MASSACHUSETTS AMHERST

M.A., UNIVERSITY OF MASSACHUSETTS AMHERST

Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST

Directed by: Professor Brigitte M. Holt

An ongoing debate concerning Neandertal ecology is whether or not they utilized long range weaponry. The anteroposteriorly expanded cross-section of

Neandertal humeri have led some to argue that their humeri were adapted to thrust hunting weapons, while the rounder cross-section of Late Upper Paleolithic humeri suggests modern humans threw their weapons. We test the hypothesis that Neandertal humeri were built to resist strains engendered by thrusting rather than throwing, using finite element (FE) models of one Neandertal, one Early Upper Paleolithic (EUP) human and three recent human humeri, representing a range of cross-sectional shapes and sizes. Electromyography and kinematic data and articulated skeletons were used to determine muscle force magnitudes and directions during three positions of spear throwing and three positions of spear thrusting. Maximum von Mises strains, an indicator of bone remodeling, were determined at the 35% and 50% cross-sections of all models. During throwing and thrusting, von Mises strains produced by the

Neandertal humerus fell roughly within or below those produced by the modern human humeri, although the Neandertal experienced significantly higher strains during one stage of thrusting. The EUP humerus performed similarly to the

Neandertal, but slightly poorer during spear thrusting. This indicates the Neandertal and EUP human humeri were just as well adapted at resisting strains during throwing as recent humans and just as well or worse adapted at resisting strains during thrusting as recent humans. We also did not find any correlation between strains and biomechanical metrics used to measure humeral adaptation in throwing and thrusting

(retroversion angle, Imax/Imin, J). These results failed to support the hypothesis that the shape of Neandertal humeri reflects thrusting loads, and suggest they were capable of using long distance weaponry.

vii

TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ...... v

ABSTRACT ...... vi

LIST OF TABLES ...... x

LIST OF FIGURES ...... xi

CHAPTER

1. INTRODUCTION ...... 1

1.1 Overview ...... 1 1.2 Background: Evidence of long range weapons by Neandertals ...... 1

1.2.1 Lithic Evidence ...... 1 1.2.2 Faunal Evidence ...... 4 1.2.3 Skeletal Evidence ...... 5

1.3 Hypotheses ...... 7

2. MATERIALS AND METHODS ...... 9

2.1 Materials and Sample ...... 9 2.2 Finite Element Model Construction ...... 12 2.3 EMG: Spear Thrusting ...... 19 2.4 Interpreting the Finite Element Models ...... 22

3. RESULTS ...... 25

3.1 Muscle Activation Patterns During Spear Thrusting...... 25 3.2 FEA Simulations ...... 25

4. DISCUSSION ...... 31

5. CONCLUSION ...... 35

APPENDIX: NUMERICAL STRAIN RESULTS DURING SPEAR THRUSTING AND THROWING...... 36

BIBLIOGRAPHY...... 39

viii

LIST OF TABLES

Table Page

2.1 Morphological characteristics of the five humeri used to create FE models ...... 10

3.1 R2 values for the Imax/Imin ratio, J, J normalized, and the retroversion angle vs. maximum von Mises strain at the 3 positions of spear throwing and spear thrusting at 35% and 50% of the length of the humerus. *=p<0.0126 ...... 29

3.2 Number of times one of the cross-sections was primarily in torsion, bending, or neither ...... 30

LIST OF FIGURES

Figure Page

1.1 Increasing robustness and ellipticity in humeral cross-sections...... 6

2.1 Neandertal, EUP and recent human humeri (left to right, Norris Farms humeri 820696, 820740, and 821230). Humeri are depicted to scale. ..10

2.2 Stages of spear throwing and spear thrusting modeled in the simulations ...... 16

2.3 Experimental set up for collecting EMG data. The “spear” was a pressure treated pine pole 84 inches long and 1.25 inches in diameter. The foam pad target was placed 40 inches off the ground and adhered to an oak table (72 inches high) using masking tape. The table was stood up on its side and braced against a concrete wall. The subjects back foot was placed 84 inches away from the target and remained stationary during the spear thrusting cycle, while the forward foot stepped forward...... 21

2.4 Two cross-sections of perfectly circular humeri. The one on the left is in pure torsion, and the one on the right is in pure bending...... 24

3.1 MVC% values for muscles in the left and right arms during a single spear thrusting cycle, averaged over one hundred trials (ten participants, ten trials per participant) ...... 26

3.2 Maximum von Mises strains 35% and 50% up the shaft of the humeri during three stages of spear throwing ...... 27

3.3 Maximum von Mises strains 35% and 50% up the shaft of the humeri during three stages of spear thrusting ...... 28

3.4 Cross-sections of humeri during spear thrusting and throwing, taken at the 35% and 50% marks. Warm colors indicate high strains and cool colors indicate low strains. During spear thrusting and throwing, the humerus is neither consistently in torsion nor consistently in bending ...... 30

CHAPTER 1

INTRODUCTION

1.1 Overview

The transition from the Middle to the Upper Paleolithic in Eurasia was marked by copious cultural and behavioral changes, earning it the status of a revolution

(Mellars, 1994; Bar-Yosef, 2002; Shea and Sisk, 2010). One such change was the introduction of long range weaponry, including spear throwers and bows and arrows.

Although wooden spears were likely a staple in the Paleolithic hunting toolkit prior to this transition, as is evidenced by the wooden spears found in Schöningen and

Saxony (Movius, 1950; Thieme, 1997; Rieder, 2003), it is unknown whether these spears were thrown or thrusted (Schmitt et al., 2003; Churchill and Rhodes, 2009). If the spears were thrown, this pushes the introduction of long range weaponry back to at least 400 thousand years ago (ka). This makes it likely that Neandertals utilized long range weaponry, enhancing their fitness by a) increasing the distance between hunter and prey, and b) bringing higher success rates in hunting megafauna and small, fleeting game such as birds and rabbits (Stiner et al., 1999, 2010; Bar-Yosef, 2002).

1.2 Background: Evidence of long range weapons by Neandertals

1.2.1 Lithic Evidence

Most components of long range weaponry, such as bone and wood, are poorly preserved in the fossil record. Conversely, lithics preserve well and can be used to infer the presence of projectile weapons (Thomas, 1978; Shea, 2006; Shea and Sisk,

2010). Neandertal lithics are dominated primarily by Mousterian and Levallois

1 technology and are manufactured from stone flakes (Jochim, 2011). These lithics were efficient multipurpose tools and were used for tasks such as chopping vegetation, butchering, and hide scraping. In addition, some were hafted onto wooden spears and used as weapons, but these spears were likely not thrown, as the lithics would have made the spears too top heavy.

Function can be inferred from lithics through physical experimentation, morphometric analyses (e.g. tip cross sectional area (TCSA) and perimeter (TCSP)), use-trace, and micro- and macro-fracture (Lombard and Phillipson, 2010).

Experimental studies on replica Levallois lithics showed that some would have made efficient points when hafted onto the end of a spear and thrust into an object. Results implied that there was an optimal shape for hafted points, where if they were too short and narrow or long and wide, they would fracture upon striking the target, making them suboptimal. Points that were short and wide, long and narrow, or halfway between short and long and halfway between wide and narrow points did not fracture, making them optimally shaped (Shea et al., 2001). From these results, it was hypothesized that, because some Neandertal lithics fell within this optimal range,

Neandertals may have thrusted spears with hafted lithics.

TCSA has been hypothesized to be a factor influencing whether or not a lithic was used as a long range weapon. Shea (2006) showed that his replicated, optimally shaped, Levallois spear points, which are hypothesized to be short range weapons, had higher TCSA values than Native American dart tips and arrow heads, which are known to function as long range weapons (Shea, 2006). Morphometric analyses revealed that Levallois and Mousterian points in the Levant had much higher TCSA values than the Native American dart tips and arrow heads, and TCSA values similar to the optimally shaped Levallois spear points (Shea, 2006; Villa et al., 2009).

Further analyses on prehistoric lithics showed that, according to TCSA values, although long range weapons such as arrows and darts were likely used in Africa beginning 50-100 ka, they were not present in the Levant until the Early Upper

Paleolithic (EUP, 45 ka), and then spread across Europe from East to West (Lombard and Phillipson, 2010; Shea and Sisk, 2010). TCSP studies have revealed results similar to TCSA studies, but claim to be more accurate, and that the high correlation between TCSA and TCSP is the only reason TCSA has been successful at identifying long range weapons in the past (Sisk and Shea, 2009, 2011).

While useful, these studies compared experimentally constructed Levallois points, optimized according to modern standards, to Native American finely constructed arrowheads and darts. This could be problematic, particularly if suboptimal points were used as spearheads in the UP, and if European and Levant UP projectile weapons utilized points more robust than Native American arrowheads and darts. In addition, these studies can only infer lithic function based on its design, and cannot say for certain what the lithic was used for. Use-trace and micro-fracture studies are particularly effective in determining how a lithic was employed, but are ineffective at determining whether or not a lithic was part of a long range weapon.

From use-trace studies, we know that Neandertals did haft at least some lithics, but hafted weapons can be components of both short (i.e. spear thrusting) and long range

(i.e. spear throwing) weapons (Boëda et al., 1999; Villa et al., 2009; Rots and Plisson,

2014).

Macro-fracture analysis is useful in inferring long range weapon use, as macro-fractures occur when a lithic hits a hard object when traveling at a high velocity. Macro-fracture analysis alone, however, cannot determine if a lithic was used as a long range weapon, as these types of fractures can occur when a hafted lithic

3 is thrown and hits a hard object, such as bone or a stone, or is thrust into a hard object, such as bone. While there is little macro-fracture evidence on Neandertals lithics, a recent study from the site of the Abri du Maras, France, reported that six Levallois points had macro-fractures indicative of striking a hard object (Hardy et al., 2013).

While some of these points had TCSA values that suggested they were hafted onto spears that were thrust into a hard object, others had lower TCSA values, indicating they may have been part of more complex long range weaponry, i.e. arrows or darts

(Hardy et al., 2013). Therefore, while it is rare, there is some evidence in the archaeological record that suggests Neandertals utilized long range weapons.

1.2.2 Faunal Evidence

The faunal record can also be used to infer the presence of long range weapons in a population. Through an exhaustive search of ethnographic studies, Churchill determined that long range weapons were more effective at catching small, fleeting prey, such birds and rabbits, than short range weapons (Churchill, 1993a, 1993b).

The presence of small, fast prey in the faunal record could therefore indicate the presence of long range weaponry (Stiner et al., 1999, 2010), although it is possible these animals could have been obtained using other methods, such as trapping.

Conversely, ethnographic studies suggest the presence of small and slow (e.g. tortoises) or large prey (e.g. reindeer) can reflect the use of long or short range weapons. The faunal record indicates that, over time, inland Neandertals continued to hunt primarily terrestrial megafauna, such as reindeer, and hunted small, fast game at low levels. Conversely, anatomically modern humans (AMHs) began to shift away from megafauna and developed a more diverse diet, including many new plants, rabbits, birds, and fresh water marine life (Kaufman, 2002; Finlayson and Carrión,

2007; Niven et al., 2012; Fa et al., 2013; Yravedra et al., 2014). Isotopic evidence

4 further supports these conclusions and indicates that inland Neandertals were hypercarnivorous, and therefore may be sticking to a diet that consists strictly of megafauna (Richards et al., 2001; Bocherens et al., 2005; Trinkaus et al., 2009).

However, unlike inland Neandertals, a more diverse diet can be found at

Neandertals sites in the Levant, where evidence of small, fleeting prey can be found in low levels. Although these levels are lower than the evidence of small, fleeting prey found at contemporary AMH sites in the Levant, indicating AMHs had an increased dietary breadth compared to Neandertals, the presence of some small, fleeting prey suggests Neandertals may have been using long range weaponry in the Levant.

1.2.3 Skeletal Evidence

Neandertal long range weapon use can also be inferred through the biological record. Long bone cross-sectional geometry is an indicator of physical activity, where bone remodeling causes individuals with higher levels of physical activity to have more robust bones compared to individuals with lower levels of activity (Ruff and

Hayes, 1983a, 1983b; Bertram and Swartz, 1991; Ruff et al., 2006). This is because high levels of physical activity, either through frequent, undemanding loads or infrequent, demanding loads, induce strains on the bone and cause it to remodel in a way that resists strains, making the bone more robust. This leaves a signature on the bone that, if read correctly, can be used to infer behavior and activity patterns. Cross- sectional geometrical parameters, such as second moment areas (Iy, Ix, Imin, Imax), ratios of second moment areas (Iy/Ix, Imax/Imin), total cortical area (CA), total area

(TA), and polar moment area (J) are frequently used to infer activity levels from long bones (Ruff et al., 1993; Trinkaus et al., 1994; Holt and Formicola, 2008; Ogilvie and

Hilton, 2011; Shaw, 2011a; Macintosh et al., 2013; Shaw and Stock, 2013).

Throwing puts a torsional force on the humerus (Evans et al., 1995; Callaghan et al., 2004), and causes athletes that regularly participate in throwing activities to have humeri with more robust, circular cross-sectional areas (i.e. humeri that have higher second polar moment areas, J, and a lower Imax/Imin ratio, at the 35% and 50% marks, see Fig. 1) (Bass et al., 2002; Shaw and Stock, 2009). It also causes athletes

Figure 1.1: Increasing robustness and ellipticity in humeral cross-sections.

that regularly throw, particularly from a younger age, to have more supinated humeri and as a result a higher retroversion angle (Pieper, 1998; Rhodes and Churchill,

2009). Work on humeral cross-sectional geometry by Churchill and colleagues has concluded that Neandertals not only have more robust humeri than AMHs, but they were also loading their humeri in a different manner, causing high levels of bilateral asymmetry (Trinkaus et al., 1994; Trinkaus and B. Ruff, 1999). Neandertals also, generally, have less circular and more elliptical cross-sectional geometries. The two

6 main hypotheses as to why these difference exist are 1) Neandertals regularly participated in spear thrusting, which applied a bending moment on the trailing humerus, causing it to become more robust and elliptical than the forward humerus

(Churchill et al., 1996; Churchill, 2002; Schmitt et al., 2003) and 2) Neandertals regularly participated in hide scraping, causing their dominant humerus to become more robust (Shaw et al., 2012).

Data from the skeletal record fail to converge on an exact date at which long range weapons would have become a part of the hunting toolkit in Eurasia. Data on cross-sectional geometry puts use of long range weaponry at the transition from the

Middle Upper Paleolithic (MUP) to the Late Upper Paleolithic (LUP), while data on retroversion angle puts use of long range weaponry earlier, in the MUP (Churchill and

Rhodes, 2009; Rhodes and Churchill, 2009). Both lines of evidence agree that

Neandertals likely did not use long range weaponry, but rather participated in spear thrusting, and do not address hide scraping.

1.3 Hypotheses

Here, we test whether or not Neandertals habitually participated in spear thrusting or spear throwing using finite element analysis (FEA). If Neandertals regularly participated in spear thrusting and not spear throwing, we would expect their humeri to exhibit lower strains during spear thrusting and higher strains during spear throwing compared to human humeri. If instead, Neandertals regularly engaged in spear throwing, we would expect their humeri to have strains within and/or below the range of modern humans during spear throwing, and strains within and/or above the range of modern humans during spear thrusting.

Cross-sectional geometry is used to describe both robusticity and shape. Here, robusticity is measured using the polar moment area, J, and shape is measured using

7 the Imax/Imin ratio, where humeri with more robust humeri have higher J values, and circular humeri have a Imax/Imin ratio closer to one. If the relationships between cross- sectional geometry and spear throwing/thrusting exist as hypothesized, there should be a negative correlation between robusticity and strain during throwing, a negative correlation between the shape and strain during thrusting, and a positive correlation between shape and strain during throwing.

Finally, it has been hypothesized that the humerus is primarily in bending during spear thrusting: we test this hypothesis through analysis of the strain distributions in the humerus. To date, the only work done to determine whether the rear humerus is primarily in bending has been conducted using 2D kinematics

(Schmitt et al., 2003). Our 3D models should be able to answer this question more accurately.

Shea and Sisk (2010) have recently argued that long range weaponry should be broken into two categories, simple and complex. This is because they believe complex, and not simple long range weaponry led to the Upper Paleolithic Revolution

(Shea and Sisk, 2010). While we recognize the benefits of this distinction, it is not being made here. Instead, we are testing whether or not it was likely that Neandertals threw weapons. All simple long range weapons (i.e. wooden spears) were thrown and only some complex weapons (i.e. spears with atlatls vs. bows and arrows and darts) were thrown (Cattelain, 1997). Therefore, our results cannot be used to make the distinction between simple and complex projectile weaponry.

CHAPTER 2

MATERIALS AND METHODS

To test our hypotheses, we constructed five finite element (FE) models of humeri from stacks of CT scans, oriented so the z-axis ran along the length of the bone (see Table 2.1). The FE method allows researchers to calculate stress and strain magnitudes and distributions in complex structures under complex loadings, making it ideal for testing these hypotheses. The models were of a Neandertal from the Middle

Paleolithic (MP) and an EUP human, and three recent anatomically modern humans from the Norris Farms collection, representing the extremes in modern human humeral cross-sectional geometry (see Fig. 2).

2.1 Materials and Sample

The right humerus of Regourdou 1 was used to construct the Neandertal model

(Senut, 1985; Vandermeersch and Trinkaus, 1995a). Scans were obtained courtesy of

Professor Roberto Macchiarelli through the NESPOS database (indexed Regourdou 4, www.nespos.org). Regourdou 1’s right humerus represents one of the best preserved

Neandertal humeri in the fossil record, allowing for minimal geometric assumptions to be made during FE model construction (Vandermeersch and Trinkaus, 1995a). In addition, analyses of bilateral asymmetry, tooth wear, and handedness suggest

Regourdou 1 was most likely right handed (Vandermeersch and Trinkaus, 1995b;

Volpato et al., 2012). Therefore, if this individual regularly participated in spear thrusting or throwing, the biomechanical signature would most likely be recorded in the right humerus (Vandermeersch and Trinkaus, 1995a; Shaw, 2011b).

Table 2.1: Morphological characteristics of the five humeri used to create FE models.

4 50% Measurements J (mm ) Normalized J Imax/Imin Retroversion (mm-1.33) Angle Regourdou 1 17149 902 1.78 44 Grotte de Enfants 4 19453 422 1.74 26 NF 820696 7389 597 2.08 30 NF 820740 15655 600 1.38 38 NF 821230 12741 513 2.28 42

4 35% Measurements J (mm ) Normalized J Imax/Imin Retroversion (mm-1.33) Angle Regourdou 1 14593 768 1.20 44 Grotte de Enfants 4 15652 340 1.34 26 NF 820696 6833 552 1.53 30 NF 820740 15757 604 1.13 38 NF 821230 10456 421 1.64 42

Figure 2.1: Neandertal, EUP and recent human humeri (left to right, Norris Farms humeri 820696, 820740, and 821230). Humeri are depicted to scale.

Unfortunately, the sex of Regourdou 1 remains unknown. However, body mass estimates have caused Regourdou 1 to be classified as a male in other studies (Ruff et al., 1997; Rhodes and Churchill, 2009), and the benefits of using the complete right humerus of Regourdou 1 outweigh this flaw.

Grotte des Enfants 4 (GDE 4)’s right humerus was used to construct the EUP model. GDE 4 is a Gravettian male dated to the EUP, and was discovered in the

Grimaldi cave complex located on the border of France and Italy (Holt and Formicola,

2008). Although it is possible that GDE 4 would have lived after Neandertals became extinct and thus does not represent an AMH that lived contemporary to Neandertals, it lived near the time of Neandertal extinction and was hypothesized to utilize long range weapons (Rhodes and Churchill, 2009). The scans used to construct GDE 4 were taken by B. Holt.

The Norris Farms collection was chosen as a comparison group because it represents a prehistoric population with multiple, complete humeri preserved. The skeletal remains come from Norris Farms #36 cemetery in the Morton Site Complex, have been attributed to the Oneota people in Illinois, USA, and date to A.D. 1300. A total of 294 human burials were recovered from the cemetery with a high level of preservation, containing individuals with a large age range (Milner and Smith, 1990;

Milner et al., 1991). A number of studies have been done on this population concerning violence (Milner et al., 1991), genetic drift (Stone and Stoneking, 1993), and ontogenetic changes trabecular bone (Ryan and Krovitz, 2006).

The three Norris Farms humeri chosen for this study were all adults and represented the extremes of the cross-sectional metrics being analyzed at the 50% cross-section. Humerus 820740 had the roundest and most robust humeri (highest J value and lowest Imax/Imin ratio), suggesting it was well adapted to resist strains imposed during spear throwing and not spear thrusting. Humeri 821230 and 820696 were chosen because they had the most elliptical humeri (highest Imax/Imin ratios), suggesting they were well adapted to resist strains imposed during spear thrusting and not spear throwing. However, humerus 821230 was gracile (low J) and humerus

820696 was robust (high J, see Table 2.1). Between the three humeri, all extremes in morphology were analyzed with the exception of circular and gracile (low J and

Imax/Imin). This is acceptable, as an individual with a gracile, circular humerus represents someone who had little bone remodeling during their lifetime, making it unlikely that their cross-sectional geometry would reveal any biomechanical weapons signature. CT for the Norris Farms scans were taken and provided by C.Shaw and

T.Ryan

2.2 Finite Element Model Construction

Finite element models require three inputs: geometry, material properties, and constraints. Geometries were built up from a stacks of CT scans with varying resolutions (Regourdou 1=0.3527 pixels/mm, GDE 4=0.1123 pixels/mm, and

NF820696, NF820740, and NF821230=0.0851 pixels/mm). Every third or fourth slice was used for the Norris Farms humeri, as their resolutions were too high for FE model construction. About 900 slices were used to construct each model.

To construct the models, the CT scans were imported into Mimics 13.0, where they were thresholded and edited slice by slice in order to create four volumes: proximal and distal sections of trabecular bone, the cortical bone, and the medullary cavity. A 3D surface model was created, exported as an *.stl (standard tessellation language) surface file, and imported into 3Matic where the surfaces were smoothed and the number of elements was reduced. An *.stl file was exported from 3Matic and imported into Geomagic Studio 2013 where the geometry was edited to remove overlapping triangles, self-intersections, and artificial holes in the model. Caution was taken to ensure the separation of the cortical shell, medullary cavity, and sections of trabecular bone, and to ensure no overlap existed between the four sections.

An *.stl file was exported from Geomagic and imported into 3Matic, where the number of elements were reduced and the surface model was used to construct a

12 meshed, solid model with four distinct volumes (cortical bone, medullary cavity, and proximal and distal trabecular bone sections). The volumetric model was exported as a Nastran file and imported into Strand7. Presently, there were two thin sections of cortical bone separating the volumes of trabecular bone from the medullary cavity.

This was necessary in order to create a volume for the medullary cavity that was distinct from the sections of trabecular bone. If left in place, these sections of cortical bone would artificially stiffen the model.

Nastran files of the two sections of trabecular bone and *.stl’s of the cortical shell, medullary cavity, and trabecular bone were exported from Strand7. The *.stl’s of the medullary cavity and trabecular bone sections were and imported into

Geomagic. Exporting *.stl’s and not just using the model already in Geomagic is necessary, as the model has been remeshed since it was last in Geomagic, changing node locations and element sizes and locations. The elements on the proximal and distal ends of the medullary cavity were deleted, and elements on the medullary cavity sides of the trabecular bone sections were copied, normals were reflected, and the elements were attached to the medullary cavity using the Fill Holes, Flat tools in

Geomagic. The Flat option in Fill Holes Bridge allowed holes to be filled without changing the existing mesh, while the tangent and curvature options would have changed the positions of existing nodes.

The *.stl of the edited medullary cavity was exported from Geomagic and imported into 3Matic, where it was used to construct a solid model. A Nastran file of the medullary cavity was exported from 3Matic and imported into Strand7. The

Nastran files of trabecular bone previously exported from Strand7 were also re- imported into Strand7. Nodes were zipped, and an *.stl of the surface of the three combined sections were exported from Strand7 as an *.stl and imported into 3Matic.

The *.stl of the cortical shell previously exported from Strand7 was also imported into

3Matic, and two solid meshes were constructed, one of the cortical bone one of the combined trabecular bone sections and medullary cavity. The solid mesh of the combined trabecular bone sections and medullary cavity was deleted, and the solid mesh of the cortical bone was exported as Nastran files from 3Matic and imported into Strand7. The two previously save Nastran files of the trabecular bone sections were also imported into Strand7, and nodes were zipped. This gave the final model, consisting of cortical bone, two sections of trabecular bone and a hollow medullary cavity.

Bone material properties are anisotropic and heterogeneous, meaning they are directionally and spatially dependent (Rho et al., 1998; Strait et al., 2005; Currey,

2006; Wang et al., 2006; Berthaume et al., 2012). Spatial variability in material properties can be captured by correlating material properties to bone density (Rho et al., 1995), however this is not possible with fossils, as the fossilization process changes the density of the bone. Conversely, directional variation in material properties is conserved during fossilization (Olesiak et al., 2010), but can only be measured through destructive or potentially destructive techniques (e.g. nanoindentation, microtensile, and ultrasonic testing (Cuy et al., 2002; Wang et al.,

2006; Beaupied et al., 2007; Constantino et al., 2010; Olesiak et al., 2010; Chung and

Dechow, 2011)). It is unfeasible to gather such information on human fossils from the Paleolithic.

In lieu of comprehensive material property datasets, homogeneous, isotropic material properties are assigned to FE models of biological systems. For our models, we assigned a Young’s modulus of 17.2 GPa and a Poisson’s ratio of 0.3 to the cortical bone (Boeree et al., 1993; Currey, 2006; Zadpoor, 2006). Material properties

14 of trabecular bone vary with species, anatomical region, and function, and are always significantly more compliant than cortical bone (Goldstein, 1987). Therefore, we assigned a Young’s modulus of 0.5 GPa and a Poisson’s ratio of 0.3 to the trabecular bone.

Stress and strain magnitudes vary considerably with variation in material property magnitudes and models (Berthaume et al., 2012), however we are operating under the assumption that, when alive, all five humeri had similar material property datasets, making any error introduced to the models by assigning homogeneous, isotropic material properties negligible. This makes the strain magnitudes reported in this paper inaccurate, but the patterns seen in the strain magnitudes (i.e. Regourdou 1 having higher or lower strains than the GDE 4) accurate.

Two sets of constraints were applied to mimic muscle and joint reaction forces. Eight muscle forces were applied to the models, representing the anterior, lateral, and posterior deltoids, medial and lateral triceps brachii, pectoralis major, infraspinatus and supraspinatus. Three phases of the throwing and three phases of the thrusting cycle were modeled, requiring a total of six sets of muscle forces. The phases of the throwing cycle modeled were the beginning of acceleration phase, the maximum level of muscle activity during the acceleration phase, and the deceleration phase. The phases of the thrusting cycle modeled were the beginning of the thrusting cycle and two points of maximum muscle activity and the end of the spear thrusting cycle (see

Fig. 3).

Muscle forces are vectors requiring two inputs: magnitude and direction.

Magnitudes were determined by multiplying maximum muscle force, determined using physiological cross-sectional data (PCSA), by percent muscle activation,

15 determined using percent maximum voluntary contraction (MVC%). PCSA values were taken from values published on the Visible Human Male (VHM)

Figure2.2: Stages of spear throwing and spear thrusting modeled in the simulations.

(Garner and Pandy, 2001). These PCSA values differ from others published in the literature, particularly in the deltoids, because the VHM was a 38 year old adult male and other published values tend to be on elderly individuals whose muscles have begun to degenerate.

It should be noted that these PCSA values do not take into account pennation angle (the angle between the muscle fibers and the distal tendon of the muscle), which causes a decrease in overall PCSA values

( ) ( ) ( 3) ( ) 2 푀 푔 ∗ cos 휃 푉 푐푚 ∗ cos 휃 푃퐶푆퐴(푐푚 ) = 푔 = 휌 ( ) ∗ 퐿 (푐푚) 퐿푓(푐푚) 푐푚3 푓

Where θ is the pennation angle, Lf is the normalized fiber length, M is the mass of the muscle, ρ is the muscle density, between 1.055 and 1.112g/cm3 depending on the percentage of formaldehyde used (Ward and Lieber, 2005), and V is the volume of the muscle. While pennation angles on middle aged persons in the upper limb are rare in the literature, data on the anterior, lateral, and posterior deltoids, infraspinatus, and supraspinatus show that pennation angles are on average 5-11° (Altobelli et al., n.d.).

This would only decrease the PCSA values of the VHM by 0.4-1.8%. The pectoralis major, however, has higher pennation angles of 26° for the clavicular portion and 32° for the sternocostal portion, which would decrease our PCSA values by 10-15%. As we could not find pennation angles on the triceps for middle aged individuals, and the published data was a mixture males and females, all of which were older than the

VHM, we chose to not take pennation angle into account.

MVC% values for throwing were taken from the literature, where electromyography (EMG) data was gathered on javelin throwers while throwing baseballs, and used to calculate MVC% (Illyés and Kiss, 2005). For this study, we assumed differences in MVC% for baseball and spear throwing were negligible

(Roach et al., 2013), making the data from Illyés and Kiss (2005) appropriate to use.

As no published MVC% data exists for spear thrusting, data was gathered in the

Biomechanics lab in the Kinesiology Department at the University of Massachusetts,

Amherst (see EMG: Spearing Thrusting section).

For throwing, muscle force directions were approximated using an articulated skeleton, where its right arm was placed in the three anatomical positions being modeled and pictures were taken from multiple angles. Simplified geometries used to represent the scapula, clavicle, and the olecranon process were constructed in

Geomagic, and, using the pictures as references, were placed in the correct anatomical

17 position relative to the head of one of the humeri (NF820696). Locations of the centroids of the muscle insertions were then recorded for all three stages of throwing.

Muscle force magnitudes and directions were used to apply muscle forces to the humeri using the tangential traction option in Boneload (www.biomesh.org) (Grosse et al., 2007). The same procedure was followed during thrusting, but the articulated skeleton was placed in the anatomical position that would occur midway through a spear thrusting cycle. The position of the humerus only changes a few degrees relative to the scapula and clavicle during spear thrusting cycle, (unpublished), so one humeral position could be used to represent all three positions of the spear thrusting cycle.

To minimize variation in muscle insertion locations from model to model, 3D vectors were calculated between the medial and lateral epicondyles and the centroids of the muscle insertions for the NF820696 humerus. Vectors were then calculated in

2D space, in the xy-plane, between the medial and lateral condyles, for all humeri and used to calculate the angle of rotation around the z-axis between each model and

NF820696. This angle of rotation was multiplied against the x and y components of the 3D vectors to properly orient the muscle insertions relative to the epicondyles. To normalize the z component of the 3D vectors, it was multiplied by the ratio of the lengths of each humeri to the length of NF820696. Vectors were calculated from the distal and not proximal ends of the humeri because it is assumed the ulna would be in the same position relative to the humerus and spinal column for each model. Given the multiple degrees of freedom present in the glenohumeral joints and few degrees of freedom present in the humeroulnar joint, this could only be achieved by ensuring the normals of the humeroulnar joint surface were pointing in the same direction.

Identical muscle forces were used for all models.

Zero-displacement constraints were applied to the distal ends of the humeri to mimic the reaction forces from the radius and ulna, and to the center of the head of the humeri to mimic the reaction force from the glenoid fossa. A local Cartesian coordinate system was placed with an xy-plane parallel to the surface of the bone at the distal end of the humeri, roughly at the center of the humeroulnar joint. Zero- displacement constraints were placed in the z-direction over the area of the humeroulnar and humeroradial joints. A local cylindrical coordinate system was placed at the glenohumeral joint with an xy-plane roughly parallel to the surface of the bone. Zero-displacement constraints were placed in the x- and z-directions of the coordinate system over the area of the glenohumeral joint, allowing the humerus to rotate about the joint but preventing the humerus from translating into or expanding within the joint. While this second set of constraints over constrained the proximal end of the humerus, it was necessary in order for the model to be in static equilibrium.

In addition, Saint-Venart’s principle states this over constraining of the proximal end of the humerus should not affect the results at the 35% and 50% marks (Ugural and

Fenster, 2003).

2.3 EMG: Spear Thrusting

Ten right handed male participants with no history of trauma to the upper limbs were recruited from the University of Massachusetts, Amherst. Participants were asked to remove their shirts, and Trigno Wireless EMG sensors, part of the

Delsys® SDK wireless system, were placed on the belly of the following muscles on both the right and left sides: anterior deltoid, lateral deltoid, posterior deltoid, triceps brachii, biceps brachii, infraspinatus, supraspinatus, and pectoralis major. Placement of the sensors was determined using www.seniam.org, Fig. 2 from Illyés and Kiss

(2005), and palpation. EMG data was collected with EMGworks at a sample rate of

1000 Hz. In addition, kinematic data was gathered via retro-reflective markers which were placed on the subject and along the length of the spear using double-sided and basic duct tape. Marker positions were recorded with infrared cameras, capturing 3D position over time, at a frequency of 20 points per second. MVC was measured by having each participant perform 16 actions that isolated each muscle for two seconds every six seconds over a 30 second period.

To mimic spear thrusting, participants were handed a pressure treated, cylindrical wooden pole made of pine (length= 84 in, diameter=1.25 in) and asked to stand on a 6 inch tall platform surrounded by infrared cameras. The pole had five kinematic markers on it to record its position in time and space, located at the proximal and distal most ends, and at three random locations along the length of the pole. A foam pad with a duct tape cross, the center of which was 40 inches off the ground, 84 in away from the participant, acted as a target and was adhered to a 6 foot oak table which was placed on its side and braced against a concrete wall (see Fig. 2).

Participants were asked to place their right hand as far back on the pole as possible and their left had on the pole at a location that felt comfortable. They were then asked to strike the target in three manners: single strike, strike and hold, and triple strike

(Schmitt et al., 2003).

For the single strike, participants were told to step towards the target with their left foot, keeping their right foot stationary, strike the target once with maximal force, and withdraw. For stab and hold, participants were told to step towards and strike the target, then attempt to push the pole through the target for three seconds (three metronome clicks), and withdraw. They were encouraged to push the pole along its long axis to minimize the bending forces being applied to the pole. For the triple

20 strike, participants were told to step towards the target and strike it as hard and as fast as they could three times in a row. Participants were asked to repeat each action 10 times, for a total of 30 trials. Each trial was recorded separately using EMGworks. In

Figure 2.3: Experimental set up for collecting EMG data. The “spear” was a pressure treated pine pole 84 inches long and 1.25 inches in diameter. The foam pad target was placed 40 inches off the ground and adhered to an oak table (72 inches high) using masking tape. The table was stood up on its side and braced against a concrete wall. The subjects back foot was placed 84 inches away from the target and remained stationary during the spear thrusting cycle, while the forward foot stepped forward.

order to abate the effects of fatigue, participants were given a minimum of 20 seconds rest after every trial, and a minimum of 60 seconds rest between each action. The study’s protocol was approved by the Department of Anthropology’s Internal Review

Board at the University of Massachusetts, Amherst. For the purposes of this study, only the MVC% data from the single spear thrusting cycle was applied to the FEA models.

Kinematic data was postprocessed using EMGworks. The marker at the end of the spear nearest the participant’s right hand was labeled, and its trajectories were exported as .tsv files. EMG data for all muscles were exported from EMGworks,

21 converted to .csv files, and imported into Matlab with the kinematic files. Each trial was individually postprocessed using an in-house written Matlab program.

The Matlab program first determined the beginning and end of the spear thrusting cycle. The beginning was classified as when the spear had positive velocity moving towards the target, and the end was when the spear had a velocity less than or equal to zero. The root mean squares (RMS) of the EMG signals were taken for the

MVC calculations, and the EMG data from the trials were processed with a moving window average of 50 ms. MVC% was calculated by dividing the EMG data from the trials by the maximum MVC value and multiplying the EMG data by 100. After all trials were processed, if the MVC% for a muscle was higher than 100%, indicating that the muscle was more than 100% active, the EMG data for that muscle across all trials was divided by the highest MVC% and multiplying by 100. The length of time to complete the spear thrusting cycle was normalized by the total time for each cycle, so zero percent corresponded to the beginning and 100% corresponded to the end of the spear thrusting cycle. The mean and standard deviation of the MVC% for each percent through the thrusting cycle was taken across all participants.

2.4 Interpreting the Finite Element Models

Two sets of results were of interest: strain magnitudes and distributions.

Strain magnitudes are hypothesized to cause bone to remodel, and if there is a biomechanical signature left in the humeri it would be due to bone remodeling. While strain magnitudes reported here do not reflect the actual magnitudes of strains experienced by the individuals being modeled, due to assumptions in material property and muscle force magnitudes and distributions, they are comparable between models. Experimentally, in vivo strain gage data is used to test hypotheses concerning

22 bone remodeling. Strain gages, however, only have the ability to measure principal and shear strains in the plane that they are placed. Conversely, FEA calculates shear and principal strains across all planes. This allows for the calculation of von Mises strain, which takes into account shear and principal strains

(1) 1 휀 = 퐸√ ((휀 − 휀 )2 + (휀 − 휀 )2 + (휀 − 휀 )2) 푣푀 2 1 2 2 3 3 1 where E is Young’s modulus, ε1 is the first principal strain, ε2 is the second principal strain, ε3 is the third principal strain, and εvm is von Mises strain. Shear strains are related to principal strains through the constitutive equations of solid mechanics

(Cook et al., 2001; Ugural and Fenster, 2003). As there is evidence that bone remodels in response to the highest levels of strain experienced by the bone, and von

Mises strain takes into account all aspects of strain, this makes von Mises strain suitable for testing our hypotheses. Moreover, there is experimental evidence that suggests bone remodels in response to high von Mises strains (e.g. (Mahnama et al.,

2013)). Therefore, we are interested in the maximum von Mises strains experienced

35% and 50% up the length of the humeri.

Strain distributions reveal whether the humerus is in torsion or bending. A cross-section of a humerus that is in pure torsion will have high von Mises strains distributed along the periosteal surface and low von Mises strains distributed evenly along the endosteal surface. A cross-section of a humerus that is in pure bending has two areas of high strains occurring on opposing ends of the periosteal surface and a band of low strains occurring between the bands of high strains, running through the endosteal surface (see Fig. 5).

It is well known that, during throwing, the humerus is in torsion (Evans et al.,

1995; Callaghan et al., 2004; Sabick et al., 2005) and it has been hypothesized that,

23 during spear thrusting, the humerus is bending (Schmitt et al., 2003). Therefore, our models should show that the humerus is primarily in torsion during throwing and, if

Figure 2.4: Two cross-sections of perfectly circular humeri. The one on the left is in pure torsion, and the one on the right is in pure bending.

the 2D kinematics results are correct, that the humerus is primarily in bending during spear thrusting.

CHAPTER 3

RESULTS

3.1 Muscle Activation Patterns During Spear Thrusting

Maximum voluntary contraction percentage (MVC%) values for the single strike spear thrust, averaged over one hundred trials (ten individuals, ten trials per individual), are shown graphically over the entire spear thrusting cycle in Fig. 6.

Only MVC% values for the trailing arm were used to calculate the muscle forces that were applied to the FE models, as the trailing arm has been hypothesized to take a much higher load than the forward arm during the spear thrusting cycle (Schmitt et al., 2003). Three positions in the spear thrusting cycle were chosen to calculate the muscle forces that would be applied to the FE models: 1%, the beginning of the cycle,

77%, the highest muscle activation for the biceps and pectoralis major, and 91%, the highest muscle activation for the posterior deltoid and supraspinatus, with high muscle activations for the triceps, lateral deltoid and infraspinatus. MVC% values for the strike and hold trials and multiple strike trials remain unanalyzed.

3.2 FEA Simulations

Under the hypothesis that Neandertals habitually participated in spear thrusting and not throwing, we expect the Neandertal humerus to experience higher strains during spear throwing and lower strains during spear thrusting. Throughout the entire throwing cycle, the Neandertal experienced maximum von Mises strains either within or below the range of the modern humans at both the 35% and the 50% sections. The EUP human also experienced strains within or below the

Figure 3.1: MVC% values for muscles in the left and right arms during a single spear thrusting cycle, averaged over one hundred trials (ten participants, ten trials per participant).

range of modern humans in positions one and two, and strains higher than the modern humans in position three at both the 35% and the 50% sections (Fig. 7, see appendix

A for numerical results). Similarly, the Neandertal experienced maximum von Mises strains either within or below the range of the modern humans during all positions of

Figure 3.2: Maximum von Mises strains 35% and 50% up the shaft of the humeri during three stages of spear throwing.

spear thrusting at the 50% section and during positions one and three at the 35% section. During position 2, the Neandertal experienced higher strains at the 35% section. The EUP human experienced strains within or below the range of modern humans at positions one and two, and higher than the modern humans in position

Figure 3.3: Maximum von Mises strains 35% and 50% up the shaft of the humeri during three stages of spear thrusting.

three at both the 35% and the 50% sections (Fig. 8, appendix A).

Cross-sectional geometric properties and retroversion angle are hypothesized to be efficient predictors of spear throwing and/or spear thrusting. In particular, the polar moment area, J, and the normalized polar moment area, J normalized by length to the 5.33 power (Trinkaus et al., 1994), have been observed to be higher in the dominant arm of people who habitually throw and is considered to be an indicator of throwing. J normalized and non-normalized was never correlated to maximum von

2 Table 3.1: R values for the Imax/Imin ratio, J, J normalized, and the retroversion angle vs. maximum von Mises strain at the 3 positions of spear throwing and spear thrusting at 35% and 50% of the length of the humerus. *=p<0.0126

Imax/Imin J J, Normalized Retroversion Throwing, 35% Angle Position 1 0.5492 0.1485 0.1217 0.1999 Position 2 0.2694 0.2685 0.05082 0.5486 Position 3 0.00428 0.04395 0.2269 0.6948

Throwing, 50% Position 1 0.6465 0.2628 0.2219 0.1153 Position 2 0.2085 0.1093 0.227 0.1663 Position 3 0.01323 0.7021 0.01916 0.5657

Thrusting, 35% Position 1 0.5856 0.075 0.862 0.02191 Position 2 0.01534 0.2273 0.1382 0.0868 Position 3 0.5538 0.3061 0.6487 0.1654

Thrusting, 50% Position 1 0.2605 0.03261 0.4244 0.0881 Position 2 0.06101 0.09078 0.2551 0.00976 Position 3 0.0000111 0.4251 0.003998 0.1213

Mises strains (Table 3.1). J normalized and non-normalized was also never correlated to maximum von Mises strains during spear thrusting. Retroversion angle is also hypothesized to be an indicator of habitual throwing, and was never correlated to maximum von Mises strains during any stage of throwing or thrusting. Furthermore, there was no correlation between maximum von Mises strains and the Imax/Imin ratio in either spear throwing or spear thrusting.

Strain distributions revealed that during spear throwing and spear thrusting, the humeri were occasionally in primarily in torsion, occasionally primarily in bending (Fig. 9, Table 3.2), and occasionally in a combination of both bending and thrusting.

Table 3.2: Number of times one of the cross-sections was primarily in torsion, bending, or neither. Torsion Bending Indeterminate Position 1 6 3 1 Throwing Position 2 5 3 2 Position 3 8 1 1 Position 1 5 4 1 Thrusting Position 2 4 5 1 Position 3 4 6 0

Figure 3.4: Cross-sections of humeri during spear thrusting and throwing, taken at the 35% and 50% marks. Warm colors indicate high strains and cool colors indicate low strains. During spear thrusting and throwing, the humerus is neither consistently in torsion nor consistently in bending.

CHAPTER 4

DISCUSSION

Our results fail to support the hypothesis that, in the case of Regourdou 1,

Neandertals habitually participated in spear thrusting and not spear throwing. If

Regourdou 1 regularly participated in spear thrusting and not spear throwing, we would have expected its humerus to have higher strains at the 35% and the 50% cross- sections than the recent humans during throwing and lower strains during thrusting.

This was not the case, as the Regourdou 1 regularly experienced strains either within or below the range of recent humans during the entire spear throwing cycle.

Regourdou 1 also experienced strains either within or below of the range of recent humans during all positions of spear thrusting at the 50% cross-section and during positions one and three at the 35% cross-section, and experienced strains higher than the range of modern humans during position 2 at the 35% cross-section. Together, these results indicate that, if Regourdou 1’s humerus has remodeled to resist strains imposed during hunting, Regourdou 1 likely participated in both spear throwing and thrusting, and was better adapted at throwing than thrusting.

The biological record does not agree on an exact date that long range weaponry began to be used in Europe (Churchill et al., 1996; Churchill and Rhodes,

2009; Rhodes and Churchill, 2009), but it does put the use of long range weaponry after the EUP and the time of GDE 4. In our simulations, GDE 4’s humerus experienced strain levels similar to the Neandertal relative to the recent human sample. Assuming GDE 4’s humerus has remodeled to resist loads applied during hunting, this implies that GDE 4 was utilizing a hunting toolkit similar to the

Neandertals and engaged in spear throwing at a level similar to Regourdou 1.

However, since GDE 4 experienced higher strains than Regourdou 1 and recent humans at stage 2 of spear thrusting, GDE 4’s humerus was not as efficient at resisting strains as Regourdou 1 and therefore may have participated in lower levels of spear thrusting.

Cross-sectional geometry of the humerus has been hypothesized to be an indicator of weapons technology use (Churchill, 1993b, 2002; Schmitt et al., 2003;

Churchill and Rhodes, 2009): our results did not support this hypothesis, as humeral robusticity (J) and shape (Imax/Imin) showed no correlation with strains. This may appear to contradict results that athletes that regularly participate in throwing have larger, more circular humeral cross-sections in their dominant arms compared to their non-dominant arms (e.g. (Shaw and Stock, 2009)). However while being a thrower implies an individual has a more robust, circular humerus, having a more robust, circular humerus does not imply that the person is a thrower. Therefore, the polar moment area and the Imax/Imin ratio are poor indicators of long range weapon use.

Humeral retroversion angle, which tends to be higher in habitual throwers, was uncorrelated to strains during spear thrusting and spear throwing. This is likely because high levels of humeral retroversion do not occur because the bone has remodeled to reduce strains, but reflect a lack of ontogenetic change, as retroversion angle tends to be higher in children and decreases with age. A high retroversion angle, when coupled with high bilateral asymmetry, indicates that during development, an individual was participating in a task that placed a high torsional load on their dominant, but not their non-dominant, arm. In the Paleolithic, this torsional load could have occurred from a variety of activities, such as spear thrusting,

32 spear throwing, or hide scraping (Pieper, 1998; Yamamoto et al., 2006), and work by

Rhodes and Churchill (2009) implies this may have begun during the MP.

Finally, it has been hypothesized that the humerus is primarily in bending during spear thrusting (Churchill et al., 1996; Schmitt et al., 2003): these results do not support this conclusion. Instead, because certain strong and highly active muscles

(e.g. the anterior deltoid and pectoralis major) do not insert into the humerus strictly in the coronal plane, and instead wrap around the humerus and insert anteriorly and laterally, the humerus is twisted as it is pulled medially. This applies a torsional load to be applied to the humerus. Therefore, location of muscle attachment sites appears to control whether or not the humerus is primarily in torsion, bending, or a combination of the two during spear thrusting. This provides further evidence as to why cross-sectional humeral shape is a poor indicator of weapons use in the fossil record.

It is possible that, within a group that habitually participates in throwing, cross-sectional size and shape will be efficient at separating out individuals that throw with more frequency or force from individuals that throw less or with less frequency.

However, when analyzing the fossil record, this distinction cannot be made a priori, making the use of cross-sectional geometry for determining weapons use inefficient.

Our results support the idea that Neandertals and EUP humans may have participated in spear throwing, although our small sample size makes it impossible to make any group level conclusions. Furthermore, it is possible that the Neandertal and

EUP humeri have not remodeled to resist strains during hunting, but have remodeled to resist strains due to another activity, such as hide scraping, that converged upon a humeral morphology that is also efficient at resisting strains during spear throwing. It is likely that, in order to survive through the Paleolithic in Europe, some type of

33 clothing would have been necessary (White, 2006). Shaw et al. (2012) argued that

Neandertals may have needed clothes, and been manufacturing said clothing out of hides. Hide scraping is a laborious task, and has been known to take over eight hours to process one hide (Oakes, 1988). Since multiple hides would need to be processed per individual each year, this task could have caused the humerus to remodel and left a biomechanical signature. Hide scraping could explain the robustness and shape of

Neandertal humeri better than spear throwing/thrusting, and could also explain the bilateral asymmetry observed in Neandertal humeri (Shaw et al., 2012).

Basic assumptions were made in our model construction and interpretations which, if violated, could change the results of this study. Firstly, as in the case of cross-sectional geometry and functional anatomy, bone material properties were assumed to be constant across all humeri, and any changes in material properties would have minimal changes on the strains compared to changes in geometry.

However, we know that bone material properties vary with age and anatomical region

(Currey and Butler, 1975; Currey, 2006; Wang et al., 2006; Dechow et al., 2010;

Chung and Dechow, 2011) and these properties do not preserve in fossils (Olesiak et al., 2010) making it impossible to assign completely accurate material properties to the models. Furthermore, when bone remodels, it can change bone material properties: therefore, bone remodeling could change the bone’s material properties, morphology, or both. We also assigned muscle force of equal magnitudes to all models, ignoring possible variations in muscle size (Grosse et al., 2007). We feel that this is appropriate because, when comparing individuals with various cross-sectional geometries, differences in muscle loadings are not taken into account.

CHAPTER 5

CONCLUSION

Our results fail to support the hypothesis that Neandertals habitually participated in spear thrusting and not spear throwing. Conversely, they support the idea that Neandertal humeri were efficient at resisting strains experienced during spear thrusting and spear throwing. Our EUP human humerus experience similar humeral strain levels as our Neandertal, indicating that, if the Neandertal and EUP human humeri had remodeled to resist strains during hunting activities, EUP humans and Neandertals may have had similar hunting toolkits. Furthermore, cross-sectional geometry, particularly size and shape, and humeral retroversion angle proved to be ineffective at predicting humeral strains during weapons use, and the humerus is not primarily in bending during spear thrusting.

These results contradict previous notions concerning the biomechanics of spear thrusting and the use of morphological properties to infer Neandertal and EUP human behavior. They also depict how more advanced, comprehensive methods such as FEA can be used to more accurately test hypotheses than more simplified ones such as cross-sectional geometry, but only when correctly applied. Finally, our results suggests that Neandertals may have been more advanced, particularly in their hunting toolkit, than has previously been hypothesized, and that they may have even been using similar technologies to sympatric anatomically modern humans.

APPENDIX

NUMERICAL STRAIN RESULTS DURING SPEAR THRUSTING AND THROWING

Table A.1: Strains at 35% and 50% the length of the humerus during Position 1 of spear throwing and thrusting.

Throwing, 50% Position 1 microtrains von Mises von Mises Comp Tensile min max 820696 -555 412 21 606 Norris 820740 -320 369 13 370 Farms 821230 -1032 1038 131 1064 Regourdou 1 -125 148 28 182 Grotte des Enfants 4 -170 193 9 215

Throwing, 35% Position 1 microtrains von Mises von Mises Compressive Tensile min max 820696 -148 167 35 204 Norris 820740 -132 158 25 167 Farms 821230 -521 543 101 637 Regourdou 1 -141 159 17 187 Grotte des Enfants 4 -121 160 3 158

Thrusting, 50% Position 1 microtrains von Mises von Mises Compressive Tensile min max 820696 -291 263 10 290 Norris 820740 -160 171 3 171 Farms 821230 -145 149 42 170 Regourdou 1 -127 156 4 155 Grotte des Enfants 4 -143 135 27 151

Thrusting, 35% Position 1 microtrains von Mises von Mises Compressive Tensile min max 820696 -103 103 6 104 Norris 820740 -77 77 6 81 Farms 821230 -130 125 47 159 Regourdou 1 -53 58 3 58 Grotte des Enfants 4 -105 118 6 139

Table A.2: Strains at 35% and 50% the length of the humerus during Position 2 of spear throwing and thrusting.

Throwing, 50% Position 2 microtrains von Mises von Mises Compressive Tensile min max 820696 -1352 1288 134 1368 Norris 820740 -1577 1786 92 1780 Farms 821230 -2741 2789 142 2804 Regourdou 1 -761 818 63 868 Grotte des Enfants 4 -416 361 14 418

Throwing, 35% Position 2 microtrains von Mises von Mises Compressive Tensile min max 820696 -575 621 43 730 Norris 820740 -578 619 10 621 Farms 821230 -1042 1170 77 1223 Regourdou 1 -555 659 32 760 Grotte des Enfants 4 -116 138 85 145

Thrusting, 50% Position 2 microtrains von Mises von Mises Compressive Tensile min max 820696 -232 212 132 232 Norris 820740 -91 100 3 98 Farms 821230 -228 259 11 255 Regourdou 1 -168 181 31 203 Grotte des Enfants 4 -224 244 28 286

Thrusting, 35% Position 2 microtrains von Mises von Mises Compressive Tensile min max 820696 -52 78 5 80 Norris 820740 -35 57 4 57 Farms 821230 -97 115 7 112 Regourdou 1 -139 151 8 194 Grotte des Enfants 4 -290 292 11 322

Table A.3: Strains at 35% and 50% the length of the humerus during Position 3 of spear throwing and thrusting.

Throwing, 50% Position 3 microtrains von Mises von Mises Compressive Tensile min max 820696 -211 214 34 219 Norris 820740 -117 139 15 140 Farms 821230 -78 95 15 96 Regourdou 1 -124 136 25 149 Grotte des Enfants 4 -274 275 46 328

Throwing, 35% Position 3 microtrains von Mises von Mises Compressive Tensile min max 820696 -148 157 22 181 Norris 820740 -78 88 7 90 Farms 821230 -50 64 11 70 Regourdou 1 -111 116 9 139 Grotte des Enfants 4 -237 267 14 337

Thrusting, 50% Position 3 microtrains von Mises von Mises Compressive Tensile min max 820696 -385 352 70 387 Norris 820740 -179 230 13 230 Farms 821230 -273 331 13 327 Regourdou 1 -190 259 11 257 Grotte des Enfants 4 -206 172 17 223

Thrusting, 35% Position 3 microtrains von Mises von Mises Compressive Tensile min max 820696 -138 169 12 181 Norris 820740 69 100 4 102 Farms 821230 -111 159 6 158 Regourdou 1 -71 92 58 93 Grotte des Enfants 4 -141 168 8 180

BIBLIOGRAPHY

Altobelli, G.C., Eng, C.M., Abrams, G.D., Mathewson, M.A., Gokhim, D.S., Fakhouri, N.S., Lieber, R.L., Taylor, A.B., Ward, S.R., n.d. Glendohumeral and scapulothoracic muscle architecture in middle-aged and elderly individuals. Journal of Biomechanics.

Bar-Yosef, O., 2002. The Upper Paleolithic Revolution. Annual Review of Anthropology. 31, 363–393.

Bass, S.L., Saxon, L., Daly, R.M., Turner, C.H., Robling, A.G., Seeman, E., Stuckey, S., 2002. The effect of mechanical loading on the size and shape of bone in pre-, peri-, and postpubertal girls: a study in tennis players. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research. 17, 2274–80.

Beaupied, H., Lespessailles, E., Benhamou, C.-L., 2007. Evaluation of macrostructural bone biomechanics. Joint, bone, spine : revue du rhumatisme. 74, 233–9.

Berthaume, M.A., Dechow, P.C., Iriarte-Diaz, J., Ross, C.F., Strait, D.S., Wang, Q., Grosse, I.R., 2012. Probabilistic finite element analysis of a craniofacial finite element model. Journal of Theoretical Biology. 300, 242–53.

Bertram, J.E.A., Swartz, S.M., 1991. The “law of bone transformation”: a case of crying Wolff? Biological Reviews. 66, 245–273.

Bocherens, H., Drucker, D.G., Billiou, D., Patou-Mathis, M., Vandermeersch, B., 2005. Isotopic evidence for diet and subsistence pattern of the Saint-Césaire I Neanderthal: review and use of a multi-source mixing model. Journal of human evolution. 49, 71–87.

Boëda, E., Geneste, J., Griggo, C., Mercier, N., Muhesen, S., Reyss, J., Taha, A., Valladas, H., 1999. A Levallois point embedded in the vertebra of a wild ass (Equus africanus): hafting, projectiles and Mousterian hunting weapons. Antiquity. 73, 394–402.

Boeree, N.R., Dove, J., Cooper, J.J., Knowles, J., Hastings, G.W., 1993. Development of a degradable composite for orthopaedic use: mechanical evaluation of an hydroxyapatite-polyhydroxybutyrate composite material. Biomaterials. 14, 793– 796.

Callaghan, E.B., Bennett, D.L., El-Khoury, G.Y., Ohashi, K., 2004. Ball-thrower’s fracture of the humerus. Skeletal radiology. 33, 355–8.

Cattelain, P., 1997. Hunting during the Upper Paleolithic: Bow, Spearthrower, or both? In: Knecht, H. (Ed.), Projectile Technology, Interdisciplinary Contributions to Archaeology. Springer US, Boston, MA, pp. 213–240.

Chung, D.H., Dechow, P.C., 2011. Elastic anisotropy and off-axis ultrasonic velocity distribution in human cortical bone. Journal of anatomy. 218, 26–39.

Churchill, S.E., 1993a. Weapon Technology, Prey Size Selection, and Hunting Methods in Modern Hunter-Gatherers: Implications for Hunting in the Palaeolithic and Mesolithic. Archeological Papers of the American Anthropological Association. 4, 11–24.

Churchill, S.E., 1993b. Weapon development and hunting, Middle Pleistocene to Early Holocene. Archaeological Papers of the American Anthropologist Association. 12–24.

Churchill, S.E., 2002. Of assegais and bayonets: Reconstructing prehistoric spear use. Evolutionary Anthropology: Issues, News, and Reviews. 11, 185–186.

Churchill, S.E., Rhodes, J.A., 2009. The Evolution of the Human Capacity for “Killing at a Distance”: The Human Fossil Evidence for the Evolution of Projectile Weaponry. In: Hublin, J.-J., Richards, M.P. (Eds.), The Evolution of Hominin Diets, Vertebrate Paleobiology and Paleoanthropology. Springer Netherlands, Dordrecht, pp. 201–210.

Churchill, S.E., Weaver, A.H., Niewoehner, W.A., 1996. Late Pleistocene human technological and subsistence behavior: functional interpretations of upper limb morphology. Quaternaria Nova. vi, 413–447.

Constantino, P., Lawn, B., Lee, J., Lucas, P., Morris, D., Smith, T., 2010. Properties of tooth enamel in great apes, Biological Sciences Faculty Research.

Cook, R.D., Malkus, D.S., Plesha, M.E., Witt, R.J., 2001. Concepts and applications of finite element analysis, 4th ed. John Wiley and Sons, New York.

Currey, J.D., 2006. Bones: Structure and Mechanics. Princeton University Press.

Currey, J.D., Butler, G., 1975. The mechanical properties of bone tissue in children. The Journal of bone and joint surgery. American volume. 57, 810–4.

Cuy, J.L., Mann, A.B., Livi, K.J., Teaford, M.F., Weihs, T.P., 2002. Nanoindentation mapping of the mechanical properties of human molar tooth enamel. Archives of Oral Biology. 47, 281–291.

Dechow, P.C., Wang, Q., Peterson, J., 2010. Edentulation alters material properties of cortical bone in the human craniofacial skeleton: functional implications for craniofacial structure in primate evolution. Anatomical record (Hoboken, N.J. : 2007). 293, 618–29.

Evans, P.A., Farnell, R.D., Moalypour, S., McKeever, J.A., 1995. Thrower’s fracture: a comparison of two presentations of a rare fracture. Emergency Medicine Journal. 12, 222–224.

Fa, J.E., Stewart, J.R., Lloveras, L., Vargas, J.M., 2013. Rabbits and hominin survival in Iberia. Journal of human evolution. 64, 233–41.

Finlayson, C., Carrión, J.S., 2007. Rapid ecological turnover and its impact on Neanderthal and other human populations. Trends in Ecology & Evolution. 22, 213–222.

Garner, B.A., Pandy, M.G., 2001. Musculoskeletal model of the upper limb based on the visible human male dataset. Computer methods in biomechanics and biomedical engineering. 4, 93–126.

Goldstein, S.A., 1987. The mechanical properties of trabecular bone: Dependence on anatomic location and function. Journal of Biomechanics. 20, 1055–1061.

Grosse, I.R., Dumont, E.R., Coletta, C., Tolleson, A., 2007. Techniques for modeling muscle-induced forces in finite element models of skeletal structures. Anatomical record (Hoboken, N.J. : 2007). 290, 1069–88.

Hardy, B.L., Moncel, M.-H., Daujeard, C., Fernandes, P., Béarez, P., Desclaux, E., Chacon Navarro, M.G., Puaud, S., Gallotti, R., 2013. Impossible Neanderthals? Making string, throwing projectiles and catching small game during Marine Isotope Stage 4 (Abri du Maras, France). Quaternary Science Reviews. 82, 23– 40.

Holt, B.M., Formicola, V., 2008. Hunters of the Ice Age: The biology of Upper Paleolithic people. American journal of physical anthropology. Suppl 47, 70–99.

Illyés, Á., Kiss, R.M., 2005. Shoulder muscle activity during pushing, pulling, elevation and overhead throw. Journal of Electromyography and Kinesiology. 15, 282–289.

Jochim, M., 2011. The Upper Paleolithic. In: Milisauskas, S. (Ed.), European Prehistory: A Survey. Springer, New York, pp. 67–118.

Kaufman, D., 2002. Re-evaluating Subsistence Skills of Levantine Middle and Upper Palaeolithic Hunters: A Comparison of the Faunal Assemblages. Oxford Journal of Archaeology. 21, 217–229.

Lombard, M., Phillipson, L., 2010. Indications of bow and stone-tipped arrow use 64 000 years ago in KwaZulu-Natal, South Africa. Antiquity. 84, 635–648.

Macintosh, A.A., Davies, T.G., Ryan, T.M., Shaw, C.N., Stock, J.T., 2013. Periosteal versus true cross-sectional geometry: a comparison along humeral, femoral, and tibial diaphyses. American journal of physical anthropology. 150, 442–52.

Mahnama, A., Tafazzoli-Shadpour, M., Geramipanah, F., Mehdi Dehghan, M., 2013. Verification of the mechanostat theory in mandible remodeling after tooth extraction: animal study and numerical modeling. Journal of the mechanical behavior of biomedical materials. 20, 354–62.

Mellars, P., 1994. The Upper Paleolithic Revolution. In: Cunliffe, B. (Ed.), The Oxford Illustrated Prehistory of Europe. New York, pp. 42–78.

Milner, G., Anderson, E., Smith, V., 1991. Warfare in Late Prehistoric West-Central Illinois. American antiquity. 56, 581–603.

Milner, G., Smith, V., 1990. Oneota human skeletal remains. In: Santure, S., Harn, A., Esarey, D. (Eds.), Archaeological Investigations at the Morton Vilage and Norris Farms 36 Cemetery, Reports of Investigations 45. Illinois State Museum, Springfield, pp. 111–148.

Movius, H.L., 1950. A wooden spear of third interglacial age from lower Saxony. Southwestern Journal of Anthropology. 6, 139–142.

Niven, L., Steele, T.E., Rendu, W., Mallye, J.-B., McPherron, S.P., Soressi, M., Jaubert, J., Hublin, J.-J., 2012. Neandertal mobility and large-game hunting: the exploitation of reindeer during the Quina Mousterian at Chez-Pinaud Jonzac (Charente-Maritime, France). Journal of human evolution. 63, 624–35.

Oakes, J.E., 1988. Copper and Caribou Inuit skin clothing production.

Ogilvie, M.D., Hilton, C.E., 2011. Cross-sectional geometry in the humeri of foragers and farmers from the prehispanic American Southwest: exploring patterns in the sexual division of labor. American journal of physical anthropology. 144, 11–21.

Olesiak, S.E., Sponheimer, M., Eberle, J.J., Oyen, M.L., Ferguson, V.L., 2010. Nanomechanical properties of modern and fossil bone. Palaeogeography, Palaeoclimatology, Palaeoecology. 289, 25–32.

Pieper, H.-G., 1998. Humeral Torsion in the Throwing Arm of Handball Players. Am. J. Sports Med. 26, 247–253.

Rho, J.Y., Hobatho, M.C., Ashman, R.B., 1995. Relations of mechanical properties to density and CT numbers in human bone. Medical Engineering & Physics. 17, 347–355.

Rho, J.-Y., Kuhn-Spearing, L., Zioupos, P., 1998. Mechanical properties and the hierarchical structure of bone. Medical Engineering & Physics. 20, 92–102.

Rhodes, J.A., Churchill, S.E., 2009. Throwing in the Middle and Upper Paleolithic: inferences from an analysis of humeral retroversion. Journal of Human Evolution. 56, 1–10.

Richards, M.P., Pettitt, P.B., Stiner, M.C., Trinkaus, E., 2001. Stable isotope evidence for increasing dietary breadth in the European mid-Upper Paleolithic. Proceedings of the National Academy of Sciences of the United States of America. 98, 6528–32.

Rieder, H., 2003. Der große Wurf der frühen Jäger: Nachbau altsteinzeitlicher Speere. Biologie in unserer Zeit. 33, 156–160.

Roach, N.T., Venkadesan, M., Rainbow, M.J., Lieberman, D.E., 2013. Elastic energy storage in the shoulder and the evolution of high-speed throwing in Homo. Nature. 498, 483–6.

Rots, V., Plisson, H., 2014. Projectiles and the abuse of the use-wear method in a search for impact. Journal of Archaeological Science.

Ruff, C., Holt, B., Trinkaus, E., 2006. Who’s afraid of the big bad Wolff?: “Wolff’s law” and bone functional adaptation. American journal of physical anthropology. 129, 484–98.

Ruff, C.B., Hayes, W.C., 1983a. Cross-sectional geometry of Pecos Pueblo femora and tibiae--a biomechanical investigation: I. Method and general patterns of variation. American journal of physical anthropology. 60, 359–81.

Ruff, C.B., Hayes, W.C., 1983b. Cross-sectional geometry of Pecos Pueblo femora and tibiae--a biomechanical investigation: II. Sex, age, side differences. American journal of physical anthropology. 60, 383–400.

Ruff, C.B., Trinkaus, E., Holliday, T.W., 1997. Body mass and encephalization in Pleistocene Homo. Nature. 387, 173–176.

Ruff, C.B., Trinkaus, E., Walker, A., Larsen, C.S., 1993. Postcranial robusticity in Homo. I: Temporal trends and mechanical interpretation. American journal of physical anthropology. 91, 21–53.

Ryan, T.M., Krovitz, G.E., 2006. Trabecular bone ontogeny in the human proximal femur. Journal of human evolution. 51, 591–602.

Sabick, M.B., Kim, Y.-K., Torry, M.R., Keirns, M.A., Hawkins, R.J., 2005. Biomechanics of the shoulder in youth baseball pitchers: implications for the development of proximal humeral epiphysiolysis and humeral retrotorsion. The American journal of sports medicine. 33, 1716–22.

Schmitt, D., Churchill, S.E., Hylander, W.L., 2003. Experimental Evidence Concerning Spear Use in Neandertals and Early Modern Humans. Journal of Archaeological Science. 30, 103–114.

Senut, B., 1985. Computerized tomography of a neanderthal humerus from le Regourdou (Dordogne, France): Comparisons with modern man. Journal of Human Evolution. 14, 717–723.

Shaw, C.N., 2011a. Is “hand preference” coded in the hominin skeleton? An in-vivo study of bilateral morphological variation. Journal of Human Evolution. 61, 480– 487.

Shaw, C.N., 2011b. Is “hand preference” coded in the hominin skeleton? An in-vivo study of bilateral morphological variation. Journal of Human Evolution. 61, 480– 487.

Shaw, C.N., Hofmann, C.L., Petraglia, M.D., Stock, J.T., Gottschall, J.S., 2012. Neandertal humeri may reflect adaptation to scraping tasks, but not spear thrusting. PloS one. 7, e40349.

Shaw, C.N., Stock, J.T., 2009. Habitual throwing and swimming correspond with upper limb diaphyseal strength and shape in modern human athletes. American journal of physical anthropology. 140, 160–72.

Shaw, C.N., Stock, J.T., 2013. Extreme mobility in the Late Pleistocene? Comparing limb biomechanics among fossil Homo, varsity athletes and Holocene foragers. Journal of human evolution. 64, 242–9.

Shea, J., Davis, Z., Brown, K., 2001. Experimental Tests of Middle Palaeolithic Spear Points Using a Calibrated Crossbow. Journal of Archaeological Science. 28, 807–816.

Shea, J.J., 2006. The origins of lithic projectile point technology: evidence from Africa, the Levant, and Europe. Journal of Archaeological Science. 33, 823–846.

Shea, J.J., Sisk, M.L., 2010. Complex projectile technology and Homo sapiens dispersal into Western Eurasia. PaleoAnthropology. 100–122.

Sisk, M.L., Shea, J.J., 2009. Experimental use and quantitative performance analysis of triangular flakes (Levallois points) used as arrowheads. Journal of Archaeological Science. 36, 2039–2047.

Sisk, M.L., Shea, J.J., 2011. The African Origin of Complex Projectile Technology: An Analysis Using Tip Cross-Sectional Area and Perimeter. International Journal of Evolutionary Biology. 968012, 8.

Stiner, by M.C., Munro, N.D., Surovell, T.A., 2010. The Tortoise and the Hare.

Stiner, M.C., Munro, N.D., Surovell, T.A., Tchernov, E., Bar-Yosef, O., 1999. Paleolithic population growth pulses evidenced by small animal exploitation. Science. 283, 190.

Stone, A.C., Stoneking, M., 1993. Ancient DNA from a pre-Columbian Amerindian population. American journal of physical anthropology. 92, 463–71.

Strait, D.S., Wang, Q., Dechow, P.C., Ross, C.F., Richmond, B.G., Spencer, M.A., Patel, B.A., 2005. Modeling elastic properties in finite-element analysis: how much precision is needed to produce an accurate model? The anatomical record. Part A, Discoveries in molecular, cellular, and evolutionary biology. 283, 275– 87.

Thieme, H., 1997. Lower Palaeolithic hunting spears from Germany. Nature. 385, 807–10.

Thomas, D.H., 1978. Arrowheads and Atlatl Darts: How the Stones Got the Shaft. American antiquity. 43, 461–72.

Trinkaus, E., B. Ruff, C., 1999. Diaphyseal Cross-sectional Geometry of Near Eastern Middle Palaeolithic Humans: The Tibia. Journal of Archaeological Science. 26, 1289–1300.

Trinkaus, E., Churchill, S.E., Ruff, C.B., 1994. Postcranial robusticity in Homo. II: Humeral bilateral asymmetry and bone plasticity. American journal of physical anthropology. 93, 1–34.

Trinkaus, E., Soficaru, A., Dobos, A., Constantin, S., Zilhao, J., Richards, M.P., 2009. Stable Isotope Evidence for Early Modern Human Diet in Southeastern Europe: Peştera cu Oase, Peştera Muierii and Peştera Cioclovina Uscată*. MATERIALE ŞI CERCETĂRI ARHEOLOGICE. V, 5–14.

Ugural, A.C., Fenster, S.K., 2003. Advanced strength and applied elasticity, Fourth. ed. Prentice Hall PTR, Upper Saddle River, New Jersey.

Vandermeersch, B., Trinkaus, E., 1995a. The postcranial remains of the Régourdou 1 Neandertal: the shoulder and arm remains. Journal of Human Evolution. 28, 439–476.

Vandermeersch, B., Trinkaus, E., 1995b. The postcranial remains of the Régourdou 1 Neandertal: the shoulder and arm remains. Journal of Human Evolution. 28, 439–476.

Villa, P., Boscato, P., Ranaldo, F., Ronchitelli, A., 2009. Stone tools for the hunt: points with impact scars from a Middle Paleolithic site in southern Italy. Journal of Archaeological Science. 36, 850–859.

Volpato, V., Macchiarelli, R., Guatelli-Steinberg, D., Fiore, I., Bondioli, L., Frayer, D.W., 2012. Hand to mouth in a neandertal: right-handedness in Regourdou 1. PloS one. 7, e43949.

Wang, Q., Strait, D.S., Dechow, P.C., 2006. A comparison of cortical elastic properties in the craniofacial skeletons of three primate species and its relevance to the study of human evolution. Journal of human evolution. 51, 375–82.

Ward, S.R., Lieber, R.L., 2005. Density and hydration of fresh and fixed human skeletal muscle. Journal of biomechanics. 38, 2317–20.

White, M.J., 2006. Things to do in Doggerland when you’re dead: surviving OIS3 at the northwestern-most fringe of Middle Palaeolithic Europe. World Archaeology. 38, 547–575.

Yamamoto, N., Itoi, E., Minagawa, H., Urayama, M., Saito, H., Seki, N., Iwase, T., Kashiwaguchi, S., Matsuura, T., 2006. Why is the humeral retroversion of throwing athletes greater in dominant shoulders than in nondominant shoulders? Journal of shoulder and elbow surgery / American Shoulder and Elbow Surgeons ... [et al.]. 15, 571–5.

Yravedra, J., Panera, J., Rubio-Jara, S., Manzano, I., Expósito, A., Pérez-González, A., Soto, E., López-Recio, M., 2014. Neanderthal and Mammuthus interactions at EDAR Culebro 1 (Madrid, Spain). Journal of Archaeological Science. 42, 500–508.

Zadpoor, A.A., 2006. Finite element method analysis of human hand arm vibrations. International Journal of Scientific Research. 16, 391–395.