Citation
Skeletal analysis of the formation of the bicondylar angle

Material Information

Title:
Skeletal analysis of the formation of the bicondylar angle
Creator:
Mitchell, Mary Shirley
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Master's ( Master of arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Anthropology, CU Denver
Degree Disciplines:
Anthropology
Committee Chair:
Warrener, Anna
Committee Members:
Orr, Caley
Musiba, Charles

Notes

Abstract:
The femoral bicondylar angle is often associated with bipedality in the fossil record because it forms during ontogeny as the result of biomechanical stress. However, the exact mechanisms responsible for its formation remain unknown. Two dominant theories based on static models are assumed in the literature, one focused on skeletal architecture at the proximal femur and one on architecture at the distal femur. The proposed relationships have never been statistically analyzed. This study aims to investigate several of the skeletal measurements at the proximal and distal femur often associated with the theories on the formation of the bicondylar angle, in order to determine if correlations do exist between them and the degree of bicondylar angle in adulthood.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Mary Shirley Mitchell. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
SKELETAL ANALYSIS OF THE FORMATION OF
THE BICONDYLAR ANGLE by
MARY SHIRLEY MITCHELL B.A., University of Hawai’i, 2003
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Arts Anthropology Program
2018


©2018
MARY SHIRLEY MITCHELL
ALL RIGHTS RESERVED


This thesis for the Master of Arts degree by Mary Shirley Mitchell has been approved for the Anthropology Program by
Anna Warrener, Chair Caley Orr Charles Musiba
Date: December 15, 2018


Shirley Mitchell, Mary (M.A., Anthropology)
Skeletal Analysis of the Formation of the Bicondylar Angle Thesis directed by Assistant Professor Anna Warrener
ABSTRACT
The femoral bicondylar angle is often associated with bipedality in the fossil record because it forms during ontogeny as the result of biomechanical stress. However, the exact mechanisms responsible for its formation remain unknown. Two dominant theories based on static models are assumed in the literature, one focused on skeletal architecture at the proximal femur and one on architecture at the distal femur. The proposed relationships have never been statistically analyzed. This study aims to investigate several of the skeletal measurements at the proximal and distal femur often associated with the theories on the formation of the bicondylar angle, in order to determine if correlations do exist between them and the degree of bicondylar angle in
adulthood.
The form and content of this abstract are approved. I recommend its publication.
Approved: Anna. G. Warrener


V
ACKNOWLEDGEMENTS
This process began under the always wry and compelling tutelage of Dr. Caley Orr, whos edits along the way kept me laughing even when I wanted to cry. Thank you for everything. I would never have finished it if not for the patient efforts of Dr. Anna Warrener, whose firm hand and empathetic ear guided me as I stumbled down an unknown path. I am indebted to you for always telling me what I needed to hear, especially when it wasn’t what I wanted to hear. To my Will, your support has meant the world to me and I could ask for nothing more in a teammate. My parents Richard and Jackie are, as always, my two anchors in the stormiest of seas.


VI
TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION 1
Background 2
Wolff’s Law of Bone Remodeling 2
Anatomical Description of the Pelvis and Femur 5
Biomechanics of the Lower Limb During Locomotion 6
Competing Models for the Formation of the Bicondylar Angle 13
Proximal Femur 13
Distal Femur 16
The Fossil Record 17
Hypotheses 24
II. MATERIALS AND METHODS 26
Anatomical Measurements 26
Analysis 30
Results 31
III. DISCUSSION 43
Conclusion 47
References
48


1
CHAPTER I INTRODUCTION
Bipedality is considered one of humanity's defining traits. Its evolution marked a turning point in our species by changing the way we moved across the landscape and freeing the hands of our ancestors, eventually allowing them to create and effectively wield tools, (Marzke et al. 1988) and increasing their capacity to carry supplies over long distances (Carvalho et al. 2012). Many osteological changes were necessary to habitually balance on two legs, and the development of the bicondylar angle of the femur has long been associated with remodeling of the skeleton. Defined as “the angle between an axis through the shaft of the femur and a line perpendicular to the infracondylar plane” (Shefelbine etal. 2002, pg. 765), the bicondylar angle brings the body center of mass over the foot allowing for a smoother and more stable gait (Tardieu and Trinkaus 1994). Because it forms as a response to mechanical stress during ontogeny, bicondylar angle is useful for identifying bipedality in hominins. The formation of this angle is thought to result from mechanical strain during locomotion at the proximal and distal ends of the femur. At the proximal end, femoral neck shaft length and angle interact with pelvic width, influencing magnitude of bicondylar angle. At the distal end of the femur, stress at the knee joint, particularly during early development, results in the asymmetrical growth of the lateral and medial condyles leading to an offset in femoral shaft (Tardieu and Trinkaus 1994).
The mechanisms by which bone responds to mechanical stress to form the bicondylar angle are complex. At birth, babies are genu varum, or bow-legged, with a bicondylar angle of 0°. However as they begin to walk the angle forms in response to the new strains being placed on the proximal and distal epiphysis at the growth plates and the bicondylar angle is thought to reach stasis at around age seven (Tardieu and Trinkaus 1994; Shefelbine etal. 2002). The fossil record suggests that this pattern of ontogenetic bone formation was the same in extinct populations of hominins (Tardieu and Trinkaus 1994). Comparative studies of human and extant ape locomotion allow a better understanding of the mechanical stresses that influence bone


shape during development. This information can be applied to the fossil record allowing greater understanding of the mobility strategies practiced throughout our evolution.
2
Femoral length, biacetabular width, neck shaft length and angle, and femoral condylar size all have been presumed to be determinants of bicondylar angle by various researchers (Ruff 1995, Shefelbine etal. 2002, Lovejoy etal. 2002, Tardieu etal. 2006), but no studies have systematically assessed patterns of correlation between these traits and many of these bone features which are not fully developed by the age of seven when bicondylar angle if thought to reach stasis. The goal of this study is to better understand the relationship between key features of the femur and pelvis and to test the assumptions of how mechanical stress at the proximal and distal ends of the femur are responsible for the formation of the bicondylar angle in humans. In the current model, an increase in biacetabular width or a decrease in neck shaft angle should lead to a greater bicondylar angle by pushing the proximal femur out laterally. A larger medial condyle should also correlate with larger bicondylar angles by creating more asymmetry at the growth plate during ontogeny, causing increased asymmetrical growth in the femoral shaft. Longer femoral length should correlate with a lesser degree of bicondylar angle as it will decrease the adducting moment at the knee and place less stress on the femoral condyles.
The results of this analysis may be useful for understanding the plasticity of bicondylar angle beyond the juvenile period if traits like femoral length and biacetabular width, which are still developing well into adolescence, are strongly correlated with its magnitude. Additionally, a better understanding of osteological features associated with bicondylar angle could help elucidate locomotor behavior in extinct hominins given the mosaic nature of the postrcranial elements currently known in the fossil record (Lovejoy, 2005).
Background
Wolff’s Law of Bone Remodeling
Wolff’s Law states that morphological changes within bone are the result of load bearing, and that altered mechanical forces result in bone modification during life (Wolff, 1892). Wolff


also proposed the Trajectorial Theory, describing mathematically the lines of compression and tension within internal structure of the bone exposed to physical stressors. The credibility of
3
Wolff’s Laws has been debated, particularly regarding the mechanisms dictating bone growth and the accuracy of the mathematical models of stress. However, the concept of skeletal remodeling in response to stress is generally accepted and has been supported by decades of observational and experimental data (Chamay and Tschantz, 1972; Woo etai, 1981; Mullender and Huiskes, 1995; Jang and Kim, 2008; Barak etai, 2011). In particular, Wolffs Law has frequently been used in analyses of the proximal femur to understand its plasticity during ontogeny (Hammer 2015; Hammer 2016).
To understand how remodeling of the femur due to mechanical stress occurs, the processes of long bone formation and growth will briefly be described. Long bone growth is the result of two different but homologous types of bone growth called endochondral ossification and appositional growth. Endochondral ossification is responsible for the longitudinal growth of the bone. In utero, mesenchymal cells begin to clump together to form an early cartilage structure after which cellular death leaves a lattice for bone building cells to begin the process of ossification and transformation into bone. The process concludes with the calcification of all the cartilage save that on the articular surfaces (Mackie et at. 2008). The primary site of ossification is the bone collar around the middle of the shaft, and the secondary site of ossification is at the epiphyses. This leaves the pliable growth plate with proliferating chondrocytes at both ends of the shaft where bone growth continues until vertical growth has ceased and ossification concludes (Kronenberg 2003).
Appositional growth determines the thickness of the bone. The process is the same as endochondral ossification, except the chondrocyte cells secrete extracellular matrix on the peripheral surface of the developing bone beneath the perichondrium. Appositional bone growth continues throughout development and persists until age 20 to maintain an optimal cortical bone size and structure relative to mechanical stress (Bronner et ai, 2010).


Skeletal remodeling in adulthood is a continuous process regulated by three types of bone-dwelling cells, namely the osteoclasts, osteoblasts, and osteocytes. (Sims and Vrahnas,
4
2014). Osteoclasts break down existing bone by demineralizing sections of bone through secreted enzymes, breaking down organic elements like collagen and inorganic elements like calcium and phosphorus (Teitelbaum, 2000). Osteoblasts build new bone by constructing an organic matrix which is mineralized by laying down hydroxyapatite which hardens the bone (Capulli et ai, 2014). Osteocytes are osteoblasts that have been entrapped within mineralized bone and become the living bone cell. Osteocytes use small channels within the bone called canaliculi to carry communicative chemicals to osteoclasts, osteoblasts, and other osteocytes relaying information about the state of the bone and regulating bone destruction and formation (Bonewald, 2011; Capulli et ai, 2014).
The process by which the cells translate mechanical loads into electrochemical activity is called mechanotransduction and is most often attributed to osteocyte function (Wang et al., 2007; Bonewald, 2014). Mechanical loading causes tissue deformation, but also prompts fluid flow within the physical passages of the bone utilized by the osteocytes for communication. This fluid movement is believed to stimulate the highly reactive osteocytes into releasing signaling molecules to osteoclasts and osteoblasts (Ehrlich and Lanyon, 2002; Bonewald, 2014). The result is localized bone growth dependent on the type and locus of the loading pressures (Meakin etal., 2014).
Load bearing functional-related pressures from birth to adolescence have a significant effect on the development of cortical bone in the femur, influencing the strength of the bone throughout life (Ruff 2002; Tanck et ai 2006). Long bones respond to various levels of physical activity through structural changes, regardless of age, sex, or ethnicity (Wetzsteon et ai 2011), and are most reactive to dynamic mechanical loading, as opposed to static mechanical loading (Nordstrom etal. 1998; Meakin etal. 2014). Particularly important are the dynamic loads


5
experienced in bipedality as the mechanical strains of walking and running are transmitted through the long bones of the legs and distributed through the pelvis.
Anatomical Description of the Pelvis and Femur
The bony pelvic-femoral complex is formed by the right and left os coxae and femora. The pelvis consists of two os coxae, jointed at the sacrum posteriorly and the pubic symphysis anteriorly. The pelvic brim separates the greater pelvis from the lesser pelvis, and defines the pelvic inlet. The greater pelvis is made of the iliac blades, and the lesser pelvis consist of ischium, os pubis, and sacrum. The hip joints are ball and socket joints where the acetabulum articulates with the round femur head. The bony pelvis functions as an anchor for 45 muscles that allow movement of the trunk and lower limbs (Gruss and Schmitt 2015), and as a mechanical structure that absorbs and transmits loads during standing, locomotion, and childbirth (Levin 1997).
The femur is the longest and heaviest bone in the body. Its epiphyses are composed of spongy trabecular bone encased in a superficial sheath of dense cortical bone. The proximal terminus includes the round head which articulates with the acetabulum, a thinner neck shaft, and the greater and lesser trochanters which are the main attachment sites for muscles originating on the pelvis and trunk. The distal end consists of the lateral and medial condyles which articulate with the tibia to form the knee joint. The intercondylar fossa serves as the site of attachment for most of the ligaments of the knee joint, and the patella sits between the condyles on the anterior surface within the patellofemoral groove. The femoral shaft is comprised of a cortical outer table of bone that contains a medullary cavity housing bone marrow. The human femur is triangular in cross-section with a distinctive posterior projection called the linea aspera. In healthy adults, the femoral shaft exhibits an anterior bowing. The femur serves as the attachment point or origination site of 23 different muscles which exert force on the hip and knee during locomotion. In humans, the femur helps transmit and absorb ground reaction and


6
gravitational forces during bipedality (Warrener2017) and aids in stability by bringing the center of mass underneath the body (Lovejoy 2005).
Several major muscle groups are involved in gait at the hip joint. The gluteus muscles (minimus, medius, and maximus) originate on the outer ilium and attach at the greater trochanter and gluteal tuberosity and serve to extend and abduct the hip as well as stabilizing the body in the mediolateral plane. The adductor muscles (magnus, brevis, longus) originate at the pubis and attach at the linea aspera along the femoral shaft, and are responsible for hip adduction and internal rotation. The iliopsoas originates at the lower spine and inner ilium and attaches at the lesser trochanter. The iliopsoas is the strongest hip flexor. The rectus femoris originates at the anterior inferior iliac spine and attaches at the base of the patella; it is part of the quadricep group and acts as a knee extensor. Finally, the hamstring muscles (semimembranosus, semitendinosus and biceps femoris) originate at the ischial tuberosity and attach posteriorly to the knee joint and act as hip extensors and knee flexors.
Biomechanics of the Lower Limb During Locomotion
Biomechanical stress and bone loading is described in terms of the mechanical forces acting on the bone at any one time (Fig. 1). The human gait cycle in walking begins at heel strike with one foot and ends when the same foot heel strikes again, and is therefore made of two steps. During walking it consists of 60% stance phase in which the foot is on the ground (20% of that is double support where both feet are on the ground) and 40% swing phase in which the foot is in the air during limb advancement (Yu et al. 2010). During running forces are greater and gait cycle consists of approximately 40% stance phase and 60% swing phase (including about 20% double float in which both feet are off the ground) dependent on speed (Dugan and Baht 2005). Stance phase is where most loading occurs in the femur, with mainly compressive force at the head, tensile forces at the greater trochanter, and both compressive and tensile forces at the lesser trochanter (Taylor et al. 1996; Duda et al. 1997).


7
Mechanical Loading of Bone
Compression Tension Shear Torsion Bending
8
Figure 1. Biomechanical stress as it relates to bone. Republished from One of the body’s hardest structures. Shelton, Michael (2015) [PowerPoint Slides] http://slideplayer.com/slide/4698157/.
In the study of human bipedality, the gluteal muscles have been of particular interest because of their role in pelvic equilibrium (Stern and Susman 1981, Lovejoy 2005, Al-lmam 2017, Warrener 2017). The gluteus minimus and medius serve to keep the hip upright, preventing a drop and lateral shift in the center of gravity, while the gluteus maximus is the primary stabilizer of the trunk over the lower limb (Lovejoy 2005). These muscles function in great apes as powerful hip extensors during quadrupedal locomotion, but their demand in bipedal locomotion as hip abductors would have changed the shape and orientation of the muscles and ilium (Stern and Susman 1981, Lovejoy 1988, Kozma etal. 2018, discussed below).
Joint Reaction Force (JRF) is the force generated within a joint as a response to the forces acting upon it, and in the hip it is the result of balancing the moments of body weight and abductor tension. The equilibrium formula for the hip joint states that the abductor force times the moment arm to the axis of rotation (roughly center of the femoral head) will equal the body weight times the moment arm from the centerline of the body to the axis of rotation. Anything that affects the length of the moment arm will impact the JRF. If we increase or decrease a


8
measurement on one side of the formula, a reactionary increase or decrease of a related measurement must occur in order to maintain biomechanical stability in the joint.
For instance, an increase in biacetabular width would serve to increase the body weight moment arm from the axis of rotation to the center of the body, which would increase the moment on the hip. To balance the equilibrium formula, there can be three reactionary changes. Internally, the body weight would need to decrease to maintain equilibrium. Externally, either a decrease in neck shaft angle would lengthen the external moment arm or the abductor muscles would have to generate increased force (Fig. 2). These longer moment arms would also increase the distance from the center of the body to the proximal femur, creating a more lateral orientation, greater valgus at the knee and thereby angulation of the femur, and greater bicondylar angles (Lovejoy et al. 2002). The need to balance hip JRF is what many researchers consider the driving force in pelvic/femoral morphology.
A.


9


10
Figure 2. Hip joint reaction force and equilibrium formula reactions. Panel A. Balanced equilibrium formula. Panel B\ Increased body weight moment arm and reactionary decreased body weight. Panel C: Increased body weight moment arm and reactionary increased abductor moment arm. Panel D: Increased body weight moment arm and reactionary increased abductor force.
Many researchers have suggested that mechanical stress is reflected in the architecture of trabecular bone inside the femoral head and neck to better absorb compressive, shear, and tensile strain (Bousson etal. 2006; Baum etal. 2010; Metzger et al. 2015). Lang et al. (1997) found a direct correlation between trabecular bone density and bone strength, and Weinbaum et al. (1994) found that shear stress in bovine humeri followed the lines of trabecular density. Stresses experienced by bipeds would most likely be very different, but trabecular bone has shown to be highly responsive to mechanical stress in humans (Duncan and Turner 2995; Metzger et al. 2015; Roberts etal. 2017). Studies of this kind for the knee joint in humans are few, but Roberts et al. (2017) reported responsive changes in the trabecular bone of the tibia in response to pathological gait patterns in those suffering from knee osteoarthritis.
However, Nawathe etal. (2015) argue that trabecular and cortical bone share loads, and that cortical bone supports up to 90% of the frontal-plane bending moment in stance loading at


11
the proximal femur. Holzer et al. (2009) found that the contribution of trabecular bone to overall bone strength was < 10%. The load bearing properties of trabecular vs. cortical bone under various strains in the femur is still a topic of debate (Nawathe et al. 2015), but one aspect to the understanding of bicondylar angle development as both proximal and distal epiphysis are composed of trabeculae.
The bicondylar angle (Fig. 3) is typically measured in the coronal plane, and it is the angle from which the line through the center of the femur shaft meets the line of the infracondylar plane, or the line on which the condyles sit flat (Shefelbine et al. 2002). This angle adducts the distal femur (Fig. 4), which lessens the adduction moment in the knee joint reducing transverse shear stress and allowing the trunk of the body to sit directly over the center of mass (Tardieu and Trinkaus 1994; Javois etal. 2009). Hypothetically this angle serves to reduce lifetime wear of the knee joint and stabilizes it during the stance phase of locomotion.


12
Figure 3. Bicondylar angle = a. Republished from "Ontogeny of the knee joint in humans, great apes and fossil hominids: pelvi-femoral relationships during postnatal growth in humans." Tardieu, Christine, and Holger Preuschoft (1996) Folia Primatologica. Originally illustrated in “An Introduction to Human Evolutionary Anatomy.” Aiello, Leslie, and Christopher Dean (1990) Academic Press.
Figure 4. The effect of a higher and lower bicondylar angle on the adducting moment arm of the knee. Republished from “Relationship between foot function and medial knee joint loading in people with medial compartment knee osteoarthritis.” Levinger et at. (2013) Journal of Foot and Ankle Research.
It’s widely accepted that the bicondylar angle forms during childhood in response to the mechanical stressors of bipedality (Tardieu 1996, Tardieu and Damsin 1997; Tardieu 2010), and a study by Tardieu (1996) suggests that it reaches stasis around the age of seven. Modern human babies display no bicondylar angle, and it increases as he/she begins walking bipedality (or doesn’t as in cases of paraplegia or otherwise immobile children who maintain an angle of 0 degrees) (Tardieu and Trinkaus 1994). This epigenetic phenomenon is most important during youth and adolescence while the femoral growth plate is still unossified. However, levels of


13
bipedal mobility impact the degree of angle throughout a person's life, and the angle will decrease in adults who lose mobility or increase later if mobility is gained (Tardieu 2010).
How plastic the bicondylar angle is during adulthood is of interest because many of the skeletal components credited with determining the degree of angle are not fully formed by age seven. Postcranial skeletal development continues into adulthood, with the femur fusing at the epiphysis and reaching stasis around age seventeen (Tanner 1990) and pelvic bones reaching statis around age twenty-one (Coleman 1969; Gonzalez et al. 2009). If skeletal architecture is a causal factor in the degree of bicondylar angle, its continued growth past the accepted age of stasis for bicondylar angle may suggest that bicondylar angle formation is more plastic than previously assumed, or that skeletal architecture does not influence bicondylar angle to the extent that has been accepted within the current model.
The degree of bicondylar angle in modern populations is often cited as being between 8-11 degrees (Tardieu and Trinkaus 1994), however there may be some question as to the variation around that mean in modern adult populations. Igbigbi and Sharrif (2005) found in a study of modern Malawians a range of between 1.5 and 12 degrees of bicondylar angle in both sexes and Pandya etal. (2008) reported bicondylar angles between 2 and 14 degrees in adults of the Gujarat region of India. This study produced a range of between 3 and 13 degrees. If this wide variation is indicative of the population at large, our cited means and comparisons against the fossil populations for whom we have a limited sample may not be as meaningful as previously assumed.
Competing Models for the Formation of Bicondylar Angle
Proximal Femur
Proximally, a key biomechanical factor is the femoral neck shaft angle (NSA). NSA and neck length are thought to be determinants of bicondylar angle by increasing or decreasing the moment arm of the hip abductors at the hip joint (Lovejoy et al. 2002). While neck length is genetically regulated, neck shaft angle changes during ontogeny due to mechanical stress


14
(Houston and Zaleski 1967; Trinkaus 1993). Neck angles at birth are around 150° and slowly decrease over time, eventually reaching stasis in adolescence with a great degree of variability but usually between 140° and 120° (Henriksson 1980; Trinkaus 1993). Many influencing factors have been suggested for the development of NSA, but one of the more debated in the literature has been latitude.
Climate-reliant body types have been well documented in all animal species as a result of Bergmann and Allen’s Rules, which state that bodies evolved in colder climates will display shorter, stockier builds accompanied by shorter limbs, while bodies evolved in warmer climates will display longer, leaner builds with protracted limbs. This is a phenomena observable in the animal kingdom and demonstrable in both fossil and modern human populations (Ruff 1994; Holliday and Falsetti 1995; Porter 1999; Ruff 2002). Weaver (2003) and Gillian etal. (2013) reported what they interpreted to be a clear correlation between modern human neck shaft angle and latitude. They argue that wider, stockier builds result in greater biacetabular distances, which would place the greater biomechanical strain on the proximal femur during ontogeny, decreasing NSA. Ruff also postulates that greater biacetabular width would specifically increase the mediolateral bending of the neck shaft, forcing a reduction in NSA (Ruff 1995).
Femoral neck structure is the result of the arcuate nature of the trabeculae and asymmetry of the cortical bone (Bonneau et al. 2012). Its development remains mysterious in part because of its wide variation (Anderson and Trinkaus 1998), and researchers have long attempted to explain it as a means of increasing femoral rotation at the hip (Kay et al. 2000) Specifically, “the more varus orientation of the femoral neck, or the decrease in its neck-shaft angle, acts to reduce the moment at the hip joint, tending to sublux the femoral head and thereby produces a more stable joint” (Anderson and Trinkaus 1998, pg. 281).
However, the suggestion that NSA might be greater in warmer adapted builds and vice versa was approached and dismissed by Trinkaus (1993; 1994; Anderson and Trinkaus 1998)


15
in various studies of modern human populations, whose fluctuation he ascribed to differences in mobility rather than body shape. He argued that populations with greater mobility would place greater mechanical strain on the head of the femur, therefore lowering NSA. Further complicating the variable of neck shaft angle is its antiversion, or the position of the neck anterior to the shaft in the coronal plane (Bulagouda et al. 2014). To solve for this, many (but not all) researchers measure NSA in the coronal plane rather than three dimensions, especially when working with scans over material remains (Bonneau etal. 2012). Some discrepancies in the research may stem from the lack of a standardized measurement protocol for NSA (Bonneau et al. 2012), but much of the debate stems from the biomechanical cause and effect of strain during ontogeny. The lack of consensus in the literature is indicative of the complexity involved in NSA development.
Presumably this theory of NSA being related to latitude would link smaller NSAs with shorter overall femurs and wider biacetabular widths and larger NSAs with longer femurs and smaller biacetabular widths, but more studies would be needed to demonstrate a correlation between NSA and climate adapted body types. Some manner of relationship might exist as a function of a working pelvic-femoral complex, but the extent to which they influence one another remains unknown. If NSA and bicondylar angle prove to be correlated, the nature of that association with climate-reliant body types would require much more investigation than what is currently available in the literature.
The current model of biacetabular width and its relationship to bicondylar angle is problematic in part because it is static, presuming that biomechanical stressors primarily act vertically through the body. A dynamic model would need to take into account the biomechanical forces at work in the joint from all directions and during all phases of the gait cycle. Under the static model an increase in biacetabular width would simply result in an increase in hip abductor force or decrease in NSA to balance the moments. As an example, higher bicondylar angles in women over men are widely accepted in both modern human and


16
ancestral populations (Lovejoy 2005). This sexual dimorphism is tied to the obstetrical dilemma, which theorizes that the expanded pelvic dimensions associated with birthing increasingly encephalized neonates during Homo evolution was constrained by mechanical aspects of a bipedal gait (Ruff 2017). In a purely static model, the wider hips needed to give birth would result in larger biacetabular widths and therefore higher degrees of bicondylar angle.
Greater biacetabular distances in humans do not develop in females over males until they reach puberty as the result of hormonal changes (Coleman 1969). Should bicondylar angle prove to reach a majority stasis by the age of seven as Tardieu suggests (1996), then purely skeletal arguments do not explain the discrepancy between higher bicondylar angles in adult women over men. A more dynamic model of biomechanical stress would be needed to satisfactorily describe the forces behind the sexual dimorphism seen in bicondylar angles. However, this static model based on skeletal architecture remains the predominantly used model in fossil analysis (although this may be changing, see Warrener etal. 2015).
Distal Femur
An alternative model for the formation of bicondylar angle is that the compressive stress produced at the knee joint during adduction in the stance phase when a child begins to walk bipedally is the driving force behind the development of the bicondylar angle, compelling a response from the bone within the medial condyle to grow larger to reduce the moment at the knee (Lovejoy and Heiple 1973; Pauwels 1980; Lovejoy etal. 2002). Given the responsive nature of trabecular bone as discussed earlier, and the fact that both the amount and thickness of trabeculae as well as the density of cortical bone increase as children begin to walk (Ryan and Krovitz 2006), this theory has gained traction. Furthermore, medial condyles are demonstrably larger proximodistally than lateral condyles in humans (Lovejoy 2005).
However, it is not just the condyles that are asymmetrical, but the long bone axis itself. Tardieu and Preuschoft (1996) suggest that the bicondylar angle seen in orangutans (discussed below) is due only to medial condylar height, but that in humans the uneven growth pattern


17
takes place in the growth plate at the medial and lateral side of the metaphysis (Tardieu and Preuschoft 1996; Tardieu etal. 2006). She argues that compressive force on the medial condyle in ontogeny causes it to grow larger than the lateral condyle, resulting in uneven growth at the medial metaphysis growth plate. This uneven growth then translates to anterorposterior widening of the lateral growth plate, and an anterorposterior elongation of the lateral shaft. The resulting oval shape of the lateral condyle serves to stabilize the knee by providing a greater radius of curvature for the tibia to articulate with at full extension of the knee (Tardieu et al. 2006). She further postulates that the prominence of the lateral lip of the femoral trochlea in humans (while genetically determined now) initially formed to stabilize the patella after the development of the bicondylar angle in ontogeny, and eventually became genetically coded. That the bicondylar angle was not similarly assimilated genetically has never been fully addressed, and no studies have yet been concluded on the lateral growth of the femoral shaft in children.
Support for the distal condyle model is found in rates of medial compartment osteoarthritis in the knee joint, which are extremely common (Kerrigan etal. 2002). It has been shown that compressive force in adulthood is most often experienced at the medial side of the knee joint at up to 2.5 times the compressive force experienced in the lateral side (Kerrigan et al. 2002). This medial compression is widely considered the result of the varus torque experienced during walking (Kerrigan etal. 2002, Purcell 2013), and the varus knee displayed during early childhood would create compressive stress on the medial condyle. The higher the valgus in the knee, the more this compressive force moves to the lateral side of the knee joint which can be seen in the high rates of lateral compartment knee osteoarthritis in adults who possess a valgus knee deformity (Felson et al. 2013).
The Fossil Record
While the earliest hominins show adaptations to bipedal locomotion, fossil discoveries over the past decade have revealed the mosaic nature of adaptations to this form of locomotion.


18
While humans are obligate bipeds, all extant hominins including gorillas, orangutans, chimps, and bonobos engage in forms of situational bipedalism (Kimura 1996; Videan and McGrew 2002; Thorpe and Crompton 2006). A comparative approach allows greater interpretation of the hominin fossil record with regards to the variety of morphological features that appear to be associated with bipedalism but also those that indicate different forms of bipedal progression in extinct hominins. The anatomy of extant apes and monkeys used for comparative analysis of hominin fossils is common practice (Cartmill and Milton 1977; Fleagle etal. 1981; Gebo 1996; Ruff 2002; Harcourt-Smith etal. 2004; Locke etal. 2011).
While imperfect in that extant species will never be an exact proxy for extinct species, comparative analysis allows the identification of osteological traits associated with specific biomechanical strains incurred during various mobility strategies. Those traits can then be identified on fossil elements to indicate similar forms of mobility. For instance, arguments on the mobility strategies of the last common ancestor (LCA) find support in the various locomotive strategies of modern primates.
Richmond etal. (2001) summarizes the LCA hypotheses that have gained the most traction amongst paleoanthropologists in the last decades into four categories. Arboreal quadrupeds would have used above-branch pronograde quadrupedalism like most extant anthropoids. Terrestrial quadrupeds would have walked on land quadrupedally using various hand positioning like digitigrady (weight on the toe), palmigrady (weight on the flat palm), fistwalking (weight on the back of the proximal phalanges), and knuckle-walking (weight on the back of the middle phalanges). Finally, antipronograde and hylobatian climbers would have practiced an orthograde arboreal locomotion with significant fore- and hindlimb flexibility.
Among modern great apes, only the orangutan is fully arboreal (Locke etal. 2011). Eastern gorillas, while occasionally arboreal, employ mainly quadrupedal terrestrial knucklewalking (Doran 1997) while lowland gorillas are much more combination arboreal and terrestrial (Remis 1995). Chimpanzees and bonobos engage in both suspensory arboreal and terrestrial


19
quadruped knuckle-walking behavior at different stages of ontogeny, often practicing more forelimb dominant arboreal orthogrady as infants and moving to hindlimb dominant quadrupedalism during adolescence and adulthood (Pontzer et al. 2014). This wide width of locomotive strategies practiced by our closest genetic relatives allows for in-depth study of those strategies on bone architecture, and specifically in pelvic and femoral morphology.
Orangutans are the only other great ape that display a bicondylar angle, around 6° (Tardieu and Preuschoft 1996). Tardieu and Preuschoft (1996) argue that the development of this angle differs than that of humans in that it is solely the result of a superoinferior lengthening of the medial condyle rather than uneven growth in the femoral shaft. However, if the antipronograde and/or hylobatian hypotheses are considered and bipedalism arose from an arboreal context rather than a terrestrial one as some researchers have suggested (Senut 2006; Thorpe et al. 2007; Thorpe et al. 2014) it stands to reason that the ontogenetic formation of the bicondylar angle in orangutans may be related to the biomechanical stress of their selective bipedality. No studies yet exist on the growth and formation of the bicondylar angle in orangutans.
Orrorin tugenensis is the first femoral fossil we have from the hominin clade, and it dates to 5.7-6 mya (Richmond and Jungers 2008). The samples consist of three proximal femoral fragments, one of which still has an intact head. While no pelvic fossils currently exist for Orrorin tugenensis, the femoral fragments have several markers that suggest bipedality. According to Pickford et al. (2002) the neck is elongated and on par with both Australopithecus and modern humans over apes. Further, the femoral head is large compared to shaft diameter which is closer to modern humans than either Australopithecus or great apes, and the presence of the Obturator externus groove, where the Obturator externus muscle inserts and acts as a lateral rotator of the hip during bipedal movement in humans and which is not present in extant apes. Most relevant to this study, O. tugenensis displays a pronounced gluteal tuberosity suggestive of heightened gluteal activity and extension of the acetabulofemoral joint, which brings the thigh in


20
line with the body during a bipedal stance. There is no mention of bicondylar angle in O. tugenensis, presumably due to the lack of a full shaft.
Some researchers argue that O. tugenensis femoral morphology is actually closer to that of modern humans over Australopithecus, and therefore may be a closer relative to Homo and exclude Australopithecus as a direct relative (Pickford etal. 2002). Other researchers suggest that O. tugenensis morphology shows intermediate architecture indicative of a basal hominin directly related to both Australopithecus and Homo (Richmond and Jungers 2008). Still others propose that the mosaic of arboreal phalenx and bipedal femoral morphology supports the hypothesis of an arboreal origin to hominin bipedality, whenever that may have occurred (Senut 2006). Without the pelvic fossils the full locomotive strategies of O. tugenensis may never be known, but by the outlined biomechanical model the longer neck and larger femoral head suggest a bicondylar angle may have been present.
Ardipithecus ramidus is temporally the next hominin fossil with both pelvic and femoral elements available in Homininae, dating to 4.4 mya (WoldeGabriel etal. 2009). Two proximal femora have been recovered, but with no head, neck, or greater trochanter and as with O. tugenensis there is no mention of a possible bicondylar angle (Lovejoy et al. 2009). Lovejoy et a\. (2009) describes the pelvic fossils as already undergoing the repositioning of the gluteal muscles indicated by the shape and flare of the ilium as well as the appearance of the anterior inferior iliac spine, unique to hominins (see also Kozma etal. 2018). These features of the false pelvis in Ar. ramdus are often discussed as derived for bipedality, while the true pelvis of Ar. ramdus is described as being more archaic with a long ischial ramus seen in African apes and associated with hamstring usage in active climbing. Lovejoy etal. (2009) argues that this is indicative of a combined mobility strategy employed by Ar. ramdus.
However, the mosaic of derived features in the upper pelvis and primitive features of the lower pelvis continues into Australopithecus and is not only related to locomotion. Australopithecines have several surviving femoral and pelvic fossils both whole and partial, and


21
most follow a pattern of morphology that has been described as: large and laterally flared illia, long ischium, extremely wide interacetabular widths, short femur lengths, large bicondylar angles, long femoral necks with lower neck shaft angles, small femoral heads, and an elliptical lateral condyle in the sagittal plane (Lovejoy etal. 1970; Berge 1994; Stern 2000; Lovejoy et al. 2002; Haile-Selassie etal. 2010; DeSilva etal. 2013).
Many of these features have been tied to some form of bipedality, yet they must also be considered in the context of parturition. Much of the obstetrical information about Australopithecus comes from two fossils: AL 288-1, an Au. afarensis also known as Lucy, which has a preserved innominate bone including a pubic ramus and the entire sacrum allowing a reconstruction of the key measurement of the pelvic inlet (Berge et al. 1984) and Sts 14, a “complete” but damaged Au. africanus pelvis from Sterkfontein whose reconstruction has been a topic of contention due to extreme taphonomic compression (Berge and Goularas 2010). The classic form of the Australopithecus pelvis, supported by these and other partial pelvic fossils, is platypelloid (although see Kibii et al. 2011 for Au. sediba pelvic dimensions, which may differ) (Tague and Lovejoy 1986; Hauslerand Schmid 1995; Berge and Goularas 2010). This pelvic shape presents with wide biacetabular widths and comparatively short femurs, creating extreme lateral placement of the proximal femur and large bicondylar angles in comparison with modern Homo (Fig. 5).
The platypelloid pelvis and great interacetabular distances have been argued to be the result of bipedality free of the restraints of birthing encephalized infants (Kibii et al. 2011). Ruff (1998; 2017) postulates that these proportions would have caused the longer neck shafts and lower NSA’s seen in australopithecines. His study suggests that this would have required they walk with more lateral deviation of their centers of gravity to reduce the joint reaction forces of the hips. However, Lovejoy (2005) has argued that longer neck lengths would have served to balance the joint reaction forces at the hip and allowed for a more human-like bipedal gait. Warrener et al. (2015) conducted a study suggesting that interacetabular distance was not


22
correlated with external torque which the hip abductors must oppose, and therefore hip width and abductor mechanics are not related in the way the current static model postulates. In response, Ruff (2017) argued that locomotor costs are not the only factor in determining pelvic and femoral architecture and more study was required.
Early Homo is characterized by an overall increase in body size, but generally retains the mosaic morphology characteristic of australopithecines including the mediolaterally wide pelvic shape (Churchill and Vansickle 2017). The Gona pelvis dated to 0.9 -1.4 mya and attributed to Homo erectus is the exception, as it is the only early Homo pelvic fossil available that displays the posterior/anterior widening of the pelvic midplane and inlet often associated with infant encephalization (Simpson etal. 2008). However, the Gona pelvis shows other morphological anomalies from Homo erectus, most importantly a small body size closer to that of Australopithecus, which has led some researchers to question its taxonomy and the selective pressures resultant in the oval pelvic shape (Ruff 2010).
The fossil record consistently begins to display anterior/posterior pelvic widening during the middle Pleistocene in fossils attributed to Homo heidelbergensis and Homo neanderthalensis, although many remain mediolateraly wide (Rak and Arensberg 1987; Arsuaga etal. 1999; Rosenberg etal. 2006; Bonmatf etal. 2010). Modern human ratios were not dependability reached until the emergence of Homo sapien (Franciscus 2009). Complete pelvic fossils to the degree that dimensions can be obtained are relatively rare (Bonmatf et at. 2010), so there remains a possibility that the anterior/posterior pelvic widening may have evolved earlier in Homo. This widening also correlates in the fossil record with a reduction in overall femoral bicondylar angle, which researchers argue lends support to the theory that the rounder pelvic inlets developed in the middle Pleistocene.
Posterior/anterior widening in the pelvis results in smaller biacetabular distances, which in turn would theoretically produce smaller femoral bicondylar angles (Tardieu 2010). This reduction in angle is aided by the increase of body size seen in Homo, which theoretically


23
lengthens the femur and reduces stress at the proximal and distal joints. These effects can be seen in the fossil femur of a hominin from the early Pleistocene in Dmanisi, Georgia designated early Homo. Although lacking pelvic elements, the fossils bicondylar angle falls within the accepted range of Australopithecus (Lordkipanidze et al. 2007). In contrast we see the bicondylar angles of Neanderthals begin to approach those seen in modern human populations (Fig. 5), even given their shorter stature (Duarte etal. 1999; Arsuaga etal. 2007).
Articular
bicondylar angle Sex Source
Australopithecus
A.L. 129-la (13") —
A L 333-4 <»") —
A.L 333w-66 (10*) —
St* 34 (15*) — Lovejoy, 1978
TM 1513 (14*) — Lovejoy, 1978
KNM ER 993 (16*1 — Walker. 1973
Homo cf. habilis
KNM ER 1472 (13“) — Day etal., 1975
KNM ER 1481A 10" — Day et al., 1975
Late archaic Homo
La Ferraaaie 2 12*. ir F Heim, 1982a
Fond-dc-ForOt 1 6” M? Twieaaelmnnn, 1961
Neundertnl 1 8.5* M Schwalbe, 1901
Spy 2 9* M9 Twieaaeltnunn, 1961
Tnbun 1 (125*) F
Early modern Homo
Cro-Magnon 4328 7* —
Cro-Magnon 4329 9" F?
Minatogawa 1 12" M Baba and Kndo, 1982
Minutogawu 3 9". 10” F Baba and Endo, 1982
Minatogawa 4 9" F Baba and Endo. 1982
Paviland 1 9* M
Pfedmostl 3 9". 9" M Matiegka, 1938
Pfedmoatl 4 7", 7" F Matiegku, 1938
Pfedmoatl 9 10". 10* M Matiegka, 1938
Pfedmoatl 10 13*. 14" F Matiegka, 1938
Pfedmoatl 14 12". 14" M Matiegka, 1938
Iai Rochet to 1 9" M Klaatach and Luatig. 1914
Skhul 4 9". 10* M McCown and Kaith, 1938
1 Value* In parent hear* arc approximate, cither due to dumnge to the femoral condyle* (e.g., KNM ER (Ifl.'l and Tnbun 11 or the preaervation of relatively little of the diatal femoral dtaphyai* (e g., A L •pecimeiw, Sl» 34. and TM 1613). Unlea* noted otherwme, meaauremonU are from the author*' meaaurementa on the original apecirnen* Right and left value* are provided, ok avuilubl*
Figue 5. Republished from “Early Ontogeny of the Human Femoral Bicondylar Angle.” Tardieu, Christine and Erik Trinkaus (1994) American Journal of Physical Anthropology.
If biomechanical stress acting upon skeletal architecture is a driving factor behind bicondylar
angle formation, the proposed anteroposterior expansion and correlating reduction in bicondylar


24
angle make sense and should be reflected in modern populations. However, if the bulk of the formation lies at the stress and architecture of the distal femur, this correlation may not have any causal relationship. If neither prove driving factors behind bicondylar angle, a deeper investigation of skeletal fossil elements and modern biomechanical strain and bone relationships may be needed in order to draw any conclusion about the mobility patterns of fossil hominins.
Hypotheses
Two distinct theories dominate the literature on bicondylar angle formation. One is based on the dimensions at the pelvis and proximal femur and is often used in fossil analysis, and one is based on the dimensions of the distal femur to explain growth during ontogeny. These models suggest that biomechanical strain acting upon and constrained by these skeletal dimensions is the ultimate driver behind the formation and degree of bicondylar angle seen in adults. The proposed correlations have been widely accepted, but never statistically analyzed.
This study aims to investigate the correlations between the skeletal relationships that have been proposed to dictate the degree of femoral bicondylar angle under the two distinct hypotheses of its formation.
Neck shaft angle (NSA) will be lower and the bicondylar angle will be higher in individuals that display a wider biacetabular width. A wider width would increase the moment at the hip, causing the neck shaft angle to lower during ontogeny to stabilize the joint. This would in turn push the proximal femur out laterally and create a greater bicondylar angle.
Height and area of the medial condyle will correlate with greater bicondylar angles by creating more asymmetry in the knee joint, increasing the adduction of the distal femur thereby increasing bicondylar angle.
The biomechanical models described above suggest the need to control for some variables, namely sex, femur length, and body mass. If the proximal hypothesis proves true, we should


25
see in the data a negative correlation between NSA and bicondylar angle and a positive correlation between biacetabular width and bicondylar angle. If the distal model proves true, we expect to see a positive correlation between medial condylar ridge height and area, and bicondylar angle.


26
CHAPTER II
MATERIALS AND METHODS
The data set for this study consists of 28 magnetic resonance images (MRIs) originally collected for Dr. Anna Warrener’s PhD dissertation (2011). Each subject was a physically fit, recreational runner between the ages of twenty-one and thirty-three. The lower body of each was scanned from the fourth lumbar vertebrae to the mid-metatarsals, and four overlapping sections were scanned isotropically at 1.7mm resolution from pelvis to feet then appended. The lower limbs were kept in anatomical orientation with a leg board and dividers. Each subjects feet were positioned in dorsiflexion and immobilized using a footboard.
MRIs were analyzed using the open source imaging software Image J with Fiji plugin version 2.0.0-rc-65/1.51s (Schindelin etal. 2012). Cropped and appended images were inspected in coronal, sagittal, and transverse planes in order to record landmarks in 3D X (medial/lateral), Y (vertical), and Z (posterior/anterior) coordinates. Each plane was assessed independently, however the X, Y, Z coordinate grid was consistent throughout all three. All distances were calculated in 3D, however angles were obtained in both 3D and planar measurements in order to make them comparable to the existing literature. Distance measurements were calculated as:
d=V (x2 - x1)2 + (y2 - y1)2 + (z2 - z1)2 using an online calculator (Furey 2018).
Anatomical Measurements
The measurements of the pelvis taken were biacetabular width, bicristal width, biishial diameter, and obstetrical conjugate. Biacetabular width was measured in two ways: from one center of the femur head to the other (method of collection below), and from the first showing of the bone table in the sagittal plane after moving through the femur head. Bicristal width was also measured in the sagittal plane from the first showing of one iliac crest to the other. Bi-ishial


27
diameter was taken in the transverse plane from the first showing of each ischium moving cranially. The obstetrical conjugate was measured from the sacral promontory to the deepest showing of the pubic symphysis.
The measurements of the femur taken for both right and left were femoral length, femoral head diameter, neck length, neck angle, bicondylar angle, height of the medial condyle, and area of the medial condyle. Femur length was measured from the first showing of the superior aspect of the femoral head in the transverse plane to the last showing of the inferior aspect of the medial condyle (McHenry and Corruccini 1978; Doyle and Windsor 2011).
Femoral head diameter was found by drawing a best fit circle using the circle tool in all three planes (Fig. 6), averaging the area, and calculating diameter using the formula D = 2Va/tt (Athapattu et al. 2013).
Results
----|AFii------|x--------|y------
1 1709.270 119.794 168.221
Figure 6. Best fit circle around the femoral head.
Both neck length and neck angle were calculated by using the circle tool to find the centroid of the femoral head in all three planes, then averaging them to get the X, Y, and Z of the center of the femoral head. The line through the femoral shaft was drawn by finding the transverse slice at 20% and 80% of the shaft length, then finding the X, Y, and Z coordinates of each centroid through the shaft using the polygon tool. Finally, a line was drawn from edge to edge through the thinnest part of the neck in the coronal plane to find a center point through which to draw the line from the center of the femur head to the intersect at the line through the


28
shaft. The length of this line is the true neck length and the angle at the neck shaft intersect is the neck angle. Mechanical neck length was the distance calculated from the first showing of the femoral head in the transverse plane to the first showing of the greater trochanter in the sagittal plane moving medially (Lovejoy 1973).
Bicondylar angle was calculated by finding the X, Y, and Z coordinates of the first inferior showing of both the lateral and medial condyles in the transverse plane, then drawing a line from one point to the other in the coronal plane to represent the infracondylar plane (Fig. 7). Where the infracondylar plane and shaft line meet is the bicondylar intersect (Bl), and Bl - 90° = bicondylar angle (adapted for MRI from Shefelbine et al. 2002).
Figure 7. Plotted points and resulting lines through the neck shaft angle and bicondylar angle. The height of the medial condyle, or condylar ridge, was calculated by taking the Y
coordinate of the crest at the medial condyle in the coronal plane and calculating the distance from the Y of the inferior showing of the condyle. The area of the medial condyle was taken in


29
the transverse plane using the polygon tool at the slice just before medial and lateral condyle join together (Fig. 8).
Figure 8. Area of the medial condyle in the transverse plane.
Table 1. Variable definitions and source material.
Measurement Method Source
Biacetabular Width (acetabulum) I Distance between first showing of each acetabulum in the sagittal plane. 1 Tague1989
Biacetabular Width (femoral heads) Distance between the center of the femoral heads. Ruff 1995
Bicristal width Distance between first showing of each iliac crests in the sagittal plane. Tague1989
Bi-ischial diameter Distance between first showing of each ischium in the transverse plane. Tague 1989
Obstetrical conjugate Distance between the sacral promontory and deepest point of pubic symphysis. Tague1989
Femoral length Measured from the first showing of the femur head to the last showing of the medial condyle. McHenry and Corruccini 1978; Doyle and Windsor 2011


30
Table 1 cont’d.
Femoral head diameter Averaged out from best fit circle in all three planes and calculated from diameter. Athapattu et al. 2013
True femoral neck length Distance from the center of the femoral head to intersect of shaft. Michelotti and Clark 1999.
Mechanical neck length Distance from the first showing of the greater trochanter in the sagittal plane to the first showing of the femoral head in transverse plane. Lovejoy 1973
Femoral neck angle Angle between center of the femoral head, neck shaft intersect, and infracondylar plane. Michelotti and Clark 1999.
Medial condylar height Distance from the medial condylar crest to the last showing in the transverse plane.
A/P lateral and medial condyles Anterior to posterior measurement of the thickest part of the condyles in the sagittal plane.
Bicondylar angle Angle between the last showing of the medial Adapted from Shefelbine et al.
condyle, the infracondylar plane, and the 2002
neck/shaft intersect - 90°.
Analysis
To test for normality of the data, histograms, skewness, and kurtosis were run, and a Shapiro-Wilk test was conducted for all variables. Residuals scatter plots were built for each relationship to test for homoscedasticity. T-tests were performed to compare left and right side measurements as well as 3D and 2D angle measurements.
Variables were tested via scatter plots and parametric bivariate Pearson’s R two-tailed analysis with accompanying Confidence Intervals (Cl) using a Fisher’s z transform at 95%.


31
Pearson’s R for the two control variables of femur length and body mass, and a Student’s t-test for the control variable of sex were conducted against the independent and dependent variables to assess which were significant. Multiple regression analyses were then conducted on all model relationships with control variables when appropriate. All analyses were performed in IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp.
A sensitivity power analysis was conducted using G*Power Version 3.1 in order to establish the effect size (See Figure 1). Type I error was set to 0.05 and power was set to 0.8 according to Cohen’s (1988) parameters established for acceptable risk of Type I and Type II error.
Results
Histograms showed some visual deviation from the classic bell curve, but all values for skewness were within the -2 and +2 cutoff for normal univariate distribution. All values for kurtosis also fell within this cut off except true neck length of the right femurs which returned a value of 2.031. The Shapiro-Wilk test, shown to be the most powerful test for normality (Razali and Wah 2011), showed all values as failing to reject normal distributions over the 0.05 cut off except for the true neck length of the right femurs, which produced a result of .037. Residuals scatter plots each displayed the random distribution suggestive of a linear relationship.
A Pearson’s R was initially conducted to determine if any of the variables correlated alone (table 2). None showed any significant relationship except obstetrical conjugate with left bicondylar angle (r=0.386, p=0.042) and right bicondylar angle (r=0.374, p=0.05). These values only barely reach the p=0.05 cut off for significance and show only weak correlation in /--values of 0.3. In order to ensure no variables were distorting the results, control variables needed to be identified and a multiple regression analysis conducted.


32
Table 2. Pearson’s R variable relationship analysis results.
Variables Tested r = Sig.p =
Biacetabular Width (Acetabulum)/Left Bicondylar Angle i 0.068 I 0.733
Biacetabular Width (Acetabulum)/Right Bicondylar Angle 0.174 0.376
Biacetabular Width (Femur Heads)/Left Bicondylar Angle 0.132 0.503
Biacetabular Width (Femur Heads)/Right Bicondylar Angle 0.079 0.689
Left Femoral Length/Left Bicondylar Angle 0.05 0.801
Right Femoral Length/Right Bicondylar Angle 0.074 0.707
Left Mechanical Neck Length/Left Bicondylar Angle 0.02 0.918
Right Mechanical Neck Length/Right Bicondylar Angle 0.262 0.178
Left Condyle Area/Left Bicondylar Angle 0.072 0.716
Right Condyle Area/Right Bicondylar Angle -0.069 0.727
Left Neck Angle/Left Bicondylar Angle 0.106 0.59
Right Neck Angle/Right Bicondylar Angle 0.25 0.199
Bicristal Distance/Left Bicondylar Angle 0.25 0.2
Bicristal Distance/Right Bicondylar Angle 0.109 0.581
Bituberous Diameter/Left Bicondylar Angle 0.063 0.751
Bitubrous Diameter/Right Bicondylar Angle 0.217 0.267
Obstetrical Conjugate/Left Bicondylar Angle 0.386 0.042
Obstetrical Conjugate/Right Bicondylar Angle 0.374 0.05


33
Paired sample t-tests were conducted to determine which measurements were statistically different from one another and if one side could serve as proxy for both. If the p-values fell above the p = 0.05 significance level, the null hypothesis could not be rejected and the measurements were statistically the same. There was a significant difference between right neck angle in 3D (M= 132.158, SD=4.476) and left neck angle in 3D (M= 136.141, SD=4.574); f=4.54, p=<0.001. There was also a significant difference between right neck angle in 2D (M= 134.132, SD=4.447) and left neck angle in 2D (M= 137.68, SD=4.279); f=4.271, p=<0.001. Further, there was a significant difference between left neck angle in 3D (M= 136.141,
SD=4.574) and left neck angle in 2D (M= 137.68, SD=4.279); f=-6.489, p=<0.001 as well as right angle in 3D (M= 132.158, SD=4.476) and right neck angle in 2D (M= 134.132, SD=4.447); t=-6.827, p=<0.001. These differences require that all variables tested against these measurements must be done in both 2D and 3D, on both the left and right side individually. Left medial condylar ridge {M=37.407, SD=3.381) and right medial condylar ridge {M=36.376,
SD=3.696) were also significantly different from one another; f=2.812, p=0.009 requiring they be tested independently. All other results showed no significant difference between both sides or 2D and 3D angles (table 3) and could therefore use one representative side or angle measurement.
Table 3. Paired T-test variable comparison results._______________________
Variable Pair M = SD = t = Sig.p =
Right Bicondylar Angle 2D I 7.051 I 2.085 I -1.837 I 0.077
Left Bicondylar Angle 2D 7.79 2.656
Left Bicondylar Angle 2D 7.79 2.656 -1.823 0.079
Left Bicondylar Angle 3D 7.606 2.445
Right Bicondylar Angle 2D 7.051 2.085 0.866 0.394
Right Bicondylar Angle 3D 7.375 3.076


34
Table 3 cont’d.
Left Mechanical Neck Length 80.927 7.555 -1.024 0.315
Right Mechanical Neck Length 82.243 8.069
Left True Neck Length 52.55 6.58 2.024 0.053
Right True Neck Length 50.74 5.294
Left Neck Angle 3D 136.141 4.574 4.54 <0.001
Right Neck Angle 3D 132.158 4.476
Left Neck Angle 2D 137.68 4.279 4.271 <0.001
Right Neck Angle 2D 134.132 4.447
Left Neck Angle 2D 137.68 4.279 -6.489 <0.001
Left Neck Angle 3D 136.141 4.574
Right Neck Angle 2D 134.132 4.447 -6.827 <0.001
Right Neck Angle 3D 132.158 4.476
Left Femoral Length 458.673 4.575 -0.015 0.988
Right Femoral Length 458.682 4.682
Left Medial Condylar Ridge 37.405 3.381 -2.812 0.009
Right Medial Condylar Ridge 36.376 3.696
Area Left Medial Condyle 1,1119.95 179.085 1.531 0.138
Area Right Medial Condyle 1,091.67 193.053
The Pearson’s R for the control variable body mass (table 4) showed that it was significantly correlated with mechanical neck length (r=0.55, p=0.002), right neck angle in 3D (r=-0.513, p=0.005) and 2D (r=-0.404, p=0.033), femur length (r=0.525, p=0.004), left medial condylar ridge (r=0.406, p=0.032) and right medial condylar ridge (r=0.404, p=0.032), and area


35
of the medial condyle {r= 0.711, p=<0.001). Pearson’s R for the control variable of femur length (table 5) showed it was significantly correlated with bicristal distance (r=0.57, p=0.002), true neck length (r=0.402, p=0.035) and mechanical neck length (r=0.637, p=<0.001), left medial condylar ridge (r=0.451, p=0.016) and right medial condylar ridge (r=0.385, p=0.043), and area of the medial condyle (r=0.654, p=<0.001). Student’s t-test for the control variable of sex (table 6) showed it was significantly correlated with biacetabular width from the acetabulum (f=-2.121, p=0.044), bituberous diameter {t=-3.662, p=0.001), mechanical neck length (f=4.005, p=<0.001), femur length {t=3.292, p=0.003), left medial condylar ridge (f=3.213, p=0.003) and right medial condylar ridge (t=2.862, p=0.003), and area of the medial condyle (t=3.287, p=0.003). All significantly correlated control variables were then introduced to the multiple regression analysis when their correlated variables were tested.
Table 4. Pearson’s R for control variable body mass.
Variable Body Mass/ r = Sig.p =
Biacetabular Width (acetabulum) i -0.258 I 0.185
Biacetabular Width (femur heads) 0.121 0.539
True Neck Length 0.113 0.567
Mechanical Neck Length 0.55 0.002
Left Neck Angle 3D -0.009 0.965
Right Neck Angle 3D -0.513 0.005
Femoral Length 0.525 0.004
Left Medial Condylar Ridge 0.406 0.032
Right Medial Condylar Ridge 0.404 0.033
Area Medial Condyle 0.711 <0.001


36
Table 4 cont’d
Left Neck Angle 2D -0.063 0.748
Right Neck Angle 2D -0.404 0.033
Bicondylar Angle -0.124 0.529
Bituberous Diameter -0.23 0.24
Obstetrical Conjugate -0.175 0.372
Bicristal Distance 0.373 0.051
Table 5. Pearson’s R for control variable femoral length.
Variable Femur Length/ r = Sig.p:
Biacetabular Width (from acetabulum) i -0.334 I 0.082
Biacetabular Width (from femur heads) 0.201 0.304
True Neck Length 0.401 0.035
Mechanical Neck Length 0.637 <0.001
Left Neck Angle 3D -0.112 0.572
Right Neck Angle 3D -0.228 0.242
Left Medial Condylar Ridge 0.451 0.016
Right Medial Condylar Ridge 0.385 0.043
Area Medial Condyle 0.654 <0.001
Left Neck Angle 2D -0.14 0.478
Right Neck Angle 2D -0.15 0.446


37
Table 5 cont’d.
Bicondylar Angle 0.05 0.801
Bituberous Diameter -0.294 0.129
Obstetrical Conjugate 0.063 0.751
Bicristal Distance 0.57 0.002
Table 6. Student’s t-test for control variable sex.____________________________
Variables M = SD = t = Sig.p =
Biacetabular Width Male 119.396 6.478 -2.121 0.044
(acetabulum) Female 124.606 6.489
Biacetabular Width (femur Male 175.155 4.614 -0.841 0.408
heads) Female 176.884 6.235
True Neck Length Male 52.185 6.887 -0.31 0.759
Female 52.971 6.461
Mechanical Neck Length Male 85.193 6.201 4.005 <0.001
Female 76.004 5.879
Left Neck Angle 3D Male 135.051 4.646 -1.377 0.18
Female 137.399 4.322
Right Neck Angle 3D Male 131.306 5.243 -1.086 0.288
Female 133.141 3.324
Left Neck Angle 2D Male 136.739 4.36 -1.264 0.218
Female 138.766 4.079
Right Neck Angle 2D Male 133.676 5.114 -0.576 0.57
Female 134.658 3.665


38
Table 6 cont’d.
Femoral Length Male 470.675 23.102 3.292 0.003
Female 444.825 17.55
Left Medial Condylar Ridge Male 39.053 3.588 3.213 0.003
Female 35.504 1.842
Right Medial Condylar Ridge Male 38.03 3.843 2.862 0.008
Female 34.469 2.472
Area Medial Condyle Male 1217.79 146.029 3.287 0.003
Female 1022.11 157.322
Bicondylar Angle Male 7.672 3.195 -0.247 0.807
Female 7.925 1.981
Bituberous Diameter Male 90.235 11.328 -3.662 0.001
Female 107.107 13.06
Bicristal Distance Male 266.195 16.318 0.488 0.629
Female 263.433 13.116
Obstetrical Conjugate Male 129.054 10.139 -1.489 0.149
Female 134.809 10.275
Once control variables were identified for each relationship, an individual multiple regression analysis was run on each proposed relationship (table 7, table 8). For the proximal model, two measurements of biacetabular width, true and mechanical neck length, and both 2D and 3D neck shaft angle were tested against bicondylar angle. In the distal model, medial condylar height and area were tested against bicondylar angle. None of the multiple regressions showed a significant relationship between any variables.


39
Table 7. Multiple Regression Analysis: Proximal Model. Length measurements in mm, mass in kg, angles in degrees.__________________________________________________
Left Bicondylar Angle Coefficients B Std. Error Beta t Sig.p =
(Constant) 1 -6.358 I 24.597 I I -0.258 I 0.779
Biacetabular Width (acetabulum) 0.029 0.091 0.074 0.313 0.757
Sex 0.328 1.564 0.063 0.21 0.836
Left Femoral Length 0.018 0.032 0.161 0.547 0.59
Body Mass -0.06 0.074 -0.218 -0.817 0.423
Left Mechanical Neck Length 0.037 0.114 0.104 0.322 0.751
Left Neck Angle 3D 0.023 0.132 0.04 0.175 0.863
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 6 I 21 I 0.174 I 0.047 I 0.981
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 5.577 I 25.419 I I 0.219 I 0.828
Biacetabular Width (acetabulum) 0.000 0.09 -0.001 -0.005 0.996
Sex -0.318 1.445 -0.061 -0.22 0.828
Left Femoral Length -0.002 0.035 -0.016 -0.052 0.959
Body Mass -0.05 0.07 -0.179 -0.712 0.484
Left True Neck Length 0.128 0.102 0.317 1.257 0.223
Left Neck Angle 3D 0.001 0.127 0.002 0.008 0.993
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 6 I 21 I 0.431 I 0.11 I 0.85
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 1.347 I 27.192 I I 0.05 I 0.961
Biacetabular Width (acetabulum) -6.573 0.09 0.00 -0.001 0.999
Sex -0.355 1.43 -0.68 -0.249 0.806
Left Femoral Length -0.001 0.035 -0.006 -0.019 0.985
Body Mass -0.051 0.069 -0.183 -0.733 0.472
Left True Neck Length 0.123 0.103 0.305 1.196 0.245


40
Table 7 cont’d.
Left Neck Angle 2D 0.031 0.136 0.049 0.226 0.824
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 6 I 21 I 0.441 I 0.112 I 0.843
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I -11.1 I 25.955 I I -0.428 I 0.673
Biacetabular Width (acetabulum) 0.027 0.091 0.071 0.301 0.766
Sex 0.234 1.545 0.045 0.152 0.881
Left Femoral Length 0.019 0.032 0.172 0.582 0.567
Body Mass -0.059 0.074 -0.215 -0.809 0.428
Left Mechanical Neck Length 0.029 0.115 0.082 0.252 0.803
Left Neck Angle 2D 0.06 0.14 0.096 0.426 0.675
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 6 I 21 I 0.2 I 0.054 I 0.973
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 4.901 I 24.829 I I 0.197 I 0.845
Biacetabular Width (femur heads) 0.014 0.108 0.028 0.127 0.9
Left Neck Angle 3D -0.014 0.121 -0.025 -0.118 0.907
Left True Neck Length 0.127 0.096 0.315 1.325 0.198
Left Femoral Length -0.009 0.024 -0.085 -0.379 0.708
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 4 I 23 I 0.559 I 0.089 I 0.695
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I -0.003 I 26.919 I I 0.000 I 1.00
Biacetabular Width (femur heads) 0.012 0.107 0.023 0.107 0.915
Left Neck Angle 2D 0.021 0.13 0.034 0.164 0.871
Left True Neck Length 0.121 0.097 0.299 1.244 0.226
Left Femoral Length -0.008 0.025 -0.07 -0.311 0.759
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =


41
Table 7 cont’d.
4 23 0.562 0.089 0.692
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I -15.27 I 26.552 I I -0.575 I 0.571
Biacetabular Width (femur heads) 0.067 0.116 0.136 0.581 0.567
Left Femoral Length 0.011 0.034 0.102 0.327 0.747
Left Neck Angle 2D 0.053 0.14 0.086 0.381 0.707
Left Mechanical Neck Length 0.038 0.115 0.107 0.327 0.747
Sex 0.099 1.559 0.019 0.063 0.95
Body Mass -0.066 0.074 -0.239 -0.895 0.381
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 6 I 21 I 0.244 I 0.065 I 0.957
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I -10.396 I 25.048 I I -0.415 I 0.682
Biacetabular Width (femur heads) 0.07 0.117 0.142 0.601 0.555
Left Femoral Length 0.01 0.034 0.088 0.283 0.78
Left Neck Angle 3D 0.014 0.133 0.023 0.103 0.919
Left Mechanical Neck Length 0.046 0.115 0.131 0.402 0.691
Sex 0.201 1.573 0.038 0.128 0.9
Body Mass -0.067 0.074 -0.242 -0.903 0.377
'able 8. Multiple Regression Analysis: ngles in degrees. Distal Model. Length measurements in mm, mas
Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p =
(Constant) I 4.461 I 14.281 I I 0.312 I 0.758
Left Medial Condylar Ridge -0.107 0.196 -0.136 -0.544 0.592
Body Mass -0.048 0.069 -0.174 -0.698 0.492
Sex 0.053 1.391 0.01 0.038 0.97
Left Femoral Length 0.023 0.029 0.208 0.786 0.44


42
Table 8 cont’d.
ANOVA and Model Summary df1 df2 F R2 Sig. p =
I 4 I 23 I 0.293 I 0.048 I 0.88
Left Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 2.678 I 14.214 I I 0.188 I 0.852
Area Left Medial Condyle 0.005 0.005 0.329 0.929 0.363
Body Mass -0.099 0.083 -0.356 -1.191 0.247
Sex 0.523 1.383 0.098 0.378 0.709
Left Femoral Length 0.012 0.033 0.101 0.356 0.725
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 4 I 21 I 0.455 I 0.08 I 0.768
Right Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 3.844 I 10.512 I I 0.366 I 0.718
Right Medial Condylar Ridge -0.058 0.13 -0.103 -0.446 0.659
Body Mass -0.081 0.052 -0.371 -1.55 0.135
Sex -0.138 1.033 -0.034 -0.134 0.895
Right Femoral Length 0.024 0.021 0.283 1.148 0.263
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 4 I 23 I 0.84 I 0.128 I 0.514
Right Bicondylar Angle Coefficients I B I Std. Error I Beta I t I Sig. p =
(Constant) I 0.279 I 10.352 I I 0.027 I 0.979
Area Right Medial Condyle 0.002 0.004 0.198 0.588 0.563
Body Mass -0.11 0.065 -0.504 -1.699 0.104
Sex 0.14 1.055 0.033 0.133 0.895
Right Femoral Length 0.025 0.023 0.28 1.092 0.287
ANOVA and Model Summary I df1 I df2 I F I R2 I Sig. p =
I 4 I 21 I 0.938 I 0.152 I 0.461


43
CHAPTER III DISCUSSION
The assumption that genetically determined elements of skeletal structure combined with mobility results in the degree of bicondylar angle have long been accepted as cause and effect. While Tardieu's work (1996; 2010) has proven that mobility is a driving factor in the formation of the bicondylar angle, this study indicates that the architectural elements long assumed to be a restraining parameter may in fact have less of an impact than previously presumed.
In the proximal hypothesis of this study, the most discussed relationship in the context of the fossil record has been that of biacetabular width and bicondylar angle, often as it pertains to australopithecines (Lovejoy etal. 2002). Two measurements for biacetabular width were analyzed, both from the center of the femur heads and from the first showing of the bone table in the acetabulum. Using both measurements, the scatter plots below (Fig. 9 and Fig. 10) suggest a lack of relationship confirmed by the Pearson’s R. The lack of a clear linear relationship between these two elements suggests that wider hips may not be able to account for the extreme bicondylar angles found in fossil hominin populations, and instead they may only be explicable through questions of mobility strategies, i.e. an altered form of bipedal gait, or a combination of bipedality and some other form of locomotion.


44
Biacetabular Width (acetabulum)
Biacetabular Width (acetabulum)
Figure 9: Biacetabular width as taken from the bone table of the acetabulum, and left bicondylar angle (left) and right bicondylar angle (right). Measurements in mm, angles in degrees.
Biacetabular Width (femur heads) Biacetabular Width (femur heads)
Figure 10: Biacetabular width as taken from the center of the femur heads, and left bicondylar angle (left) and right bicondylar angle (right). Measurements in mm, angles in degrees.
The condylar ridge and medial condyle area also showed a clear lack of relationship (Fig. 11 and Fig. 12) which suggest that Tardieu’s theory on biomechanical stress in the condyles is insufficient to explain the role of condylar involvement in bicondylar angle formation in its entirety. Presumably plasticity at the condyles is a driving force behind the degree of angle, but so too might the stress in the shaft itself at the growth plate play a larger than anticipated role if each of these major theories fail to produce significance in the measurements
of this study.


45
Left Medial Condyle Ridge
Right Medial Condyle Ridge
Figure 11: Condylar ridge and left bicondylar angle (left) and right bicondylar angle (right). Measurements in mm, angles in degrees.
4>
I
15
0
800.00 1,000.00 1,200.00 1,400.00
.• • •
Left Medial Condyle Area
Right Medial Condyle Area
Figure 12. Medial condyle area and left bicondylar angle (left) and right bicondylar area (right). Measurements in mm, angles in degrees.
Many researchers have suggested that Australopithecus engaged in a vastly different form of bipedality than is practiced by modern humans, including the possibility of an obligate upright bent-hip, bent-knee gait similar to that occasionally implemented by modern gorillas and chimpanzees (Stern and Susman 1983; Kullmer et a\. 2003; Sockol etal. 2007; Raichlan etal. 2010; Zipfel et at. 2011). If true, the development of a femur exposed to disparate biomechanical pressures during ontogeny than seen in modern populations cannot be predicted, and may result in extreme bicondylar angles due to some differing form of biomechanical stress at play in the architecture.
However, the significant results between bicondylar angle and obstetrical conjugate do suggest a relationship between pelvic dimension and bicondylar angle formation. Although


46
these results are only barely within the significance cut off (p = 0.042 and p = 0.05), they suggest that pelvic morphology correlate with degree of bicondylar angle through an unexpected pelvic dimension. The results show that expansion of the pelvis in the posterioranterior axis increase the degree of bicondylar angle largely determined in the mediolateral axis. This is the opposite of previous assumptions that this posterioanterior pelvic expansion in Homo led to smaller biacetabular widths and lower bicondylar angles in the fossil record.
One possibility is that posterioanterior expansion of the pelvis would also move the acetabulum anterior, creating a greater degree of anteversion at the femoral head. In order to retain the ~ 16° anteversion required to avoid hip dysplasia (Sugano etal. 1998) and shaft fractures (Braten et al. 1993), the trochanter of the femur would be forced outward laterally, which could result in a greater degree of bicondylar angle. It is also possible that this skeletal element is associated with some other pelvic measurement not taken for this study that would increase bicondylar angle.
A third explanation for the correlation between bicondylar angle and obstetrical conjugate may lie in the method of data collection. Angles for this study were calculated originally in 3D coordinates using X, Y, and Z points, and were then recalculated in the coronal plane in order to make them comparable to existing literature (Lovejoy and Heiple 1970; Tardieu and Damsin 1997; Shefelbine etal. 2002). This recalculation did shift the ranges by several degrees (i.e. the largest bicondylar angle went from 19.7 to 13.44). While this recalculation from 3D to 2D did not reach significance in difference using a Peason’s R, this practice of collection throughout the literature may change if measurements were taken in 3D coordinates. Furthermore, the effect size as calculated by the power analysis was 0.37, indicating that that power of this study was small. Further research with a larger dataset may yield different results.
However, the most likely inference is that the static model is inadequate to predict biomechanical loading of the hip and femur, and therefore to explain femoral bicondylar angle


47
development. The results of this study suggest that skeletal architecture is not a driving force behind the degree of bicondylar angle seen in adults. Therefore, a dynamic model of biomechanical forces and a deeper understanding of osteological reaction to those forces is needed if we are to understand how mobility is reflected in bone.
Conclusion
The presence of a bicondylar angle is a vital indicator of bipedality throughout the fossil record, and yet the process by which it develops remains undetermined. Without a thorough understanding of how and why this angle forms, the modes of bipedality practiced by our ancestors in comparison to our own can never be fully ascertained. The fossil record is incomplete, and the equivalence of osteological elements to various kinds of mobility requires an intimate understanding of the formation of those elements during ontogeny.
The current model of bicondylar angle formation rests on a static view of the interplay within the pelvic-femoral complex. The results of this study suggest that it has insufficient explanatory power to determine the mobility strategies practiced by our hominin ancestors or by current human populations. Understanding the biomechanical stresses that result in the formation of the bicondylar angle requires a more complex model that integrates a dynamic approach to mobility, as well as a deeper understanding of the forces at work in ontogeny and throughout adulthood. If skeletal architecture plus Newtonian forces alone cannot explain bicondylar angle formation, as is suggested by the results of this study, a more composite and integrative model of formation is required to understand the circumstances that lead to a bicondylar angle.


48
References
Aiello, Leslie, and Christopher Dean. 1990. An introduction to human evolutionary anatomy. Academic Press.
Al-lmam, A. 2017. "Anatomical Adaptations for Obligate Bipedal Locomotion in Humans." International Journal of Anatomy and Physiology 3 (1): 63-68.
Anderson, John Y., and Erik Trinkaus. 1998. "Patterns of sexual, bilateral and interpopulational variation in human femoral neck-shaft angles." The Journal of Anatomy 192 (2): 279-285.
Arsuaga, Juan-Luis, Carlos Lorenzo, Jose-Miguel Carretero, Ana Gracia, Ignacio Martinez, Nuria Garcia, Jose-Maria Bermudez de Castro, and Eudald Carbonell. 1999. "A complete human pelvis from the Middle Pleistocene of Spain." Nature 399 (6733): 255.
Arsuaga, J. L., V. Villaverde, R. Quam, I. Martinez, J. M. Carretero, C. Lorenzo, and A. Gracia. 2007. "New Neandertal remains from Cova Negra (Valencia, Spain)." Journal of Human Evolution 52 (1): 31-58.
Athapattu, Meuru, Amir Hossein Saveh, Jamaluddin Mahmud, Mohammad Alrashidi, Bin Wang, and Mahmoud Chizari. 2013. "Accuracy of Measuring Methods on the Femoral Head." Procedia Engineering 68: 83-87.
Barak, M. M., D. E. Lieberman, and J.J. Hublin. 2011. “A Wolff in sheep's clothing: trabecular bone adaptation in response to changes in joint loading orientation.” Bone 49 (6): 1141-1151.
Baum, T., J. Carballido-Gamio, MB Huber, D. Muller, R. Monetti, C. Rath, F. Eckstein, etal. 2010. “Automated 3D trabecular bone structure analysis of the proximal femur-prediction of biomechanical strength by CT and DXA.” Osteoporosis International 21 (9): 1553-64.
Berge, Christine, Rosine Orban-Segebarth, and Peter Schmid. 1984. "Obstetrical interpretation of the australopithecine pelvic cavity." Journal of Human Evolution 13 (7): 573-587.
Berge, Christine. 1994. "How did the australopithecines walk? A biomechanical study of the hip and thigh of Australopithecus afarensis." Journal of Human Evolution 26 (4): 259-273.


49
Berge, Christine, and Dionysis Goularas. 2010. "A new reconstruction of Sts 14 pelvis (Australopithecus africanus) from computed tomography and three-dimensional modeling techniques." Journal of Human Evolution 58 (3): 262-272.
Bonewald, L. F. 2011. “The amazing osteocyte.” Journal of Bone and Mineral Research 26 (2): 229-238.
Bonewald, L. F. 2014. “Osteocytes: An exquisitely sensitive mechanosensory cell.” Osteoporosis International 25 (S3).
Bonmatf, Alejandro, Asier Gomez-Olivencia, Juan-Luis Arsuaga, Jose Miguel Carretero, Ana Gracia, Ignacio Martinez, Carlos Lorenzo, Jose Maria Bermudez de Castro, and Eudald Carbonell. 2010. "Middle Pleistocene lower back and pelvis from an aged human individual from the Sima de los Huesos site, Spain." Proceedings of the National Academy of Sciences 107 (43): 18386-18391.
Bonneau, Noemie, Paul- Antoine Libourel, Caroline Simonis, Laurent Puymerail, Michel Baylac, Christine Tardieu, and Olivier Gagey. 2012. "A three- dimensional axis for the study of femoral neck orientation." Journal of Anatomy 221 (5): 465-476.
Bousson, Valerie, Anthony Bras, F. Roqueplan, Yan Kang, David Mitton, Sami Kolta, C. Bergot, Wafa Skalli, Eric Vicaut, W. Kalender, Klaus Engelke, and J.D. Laredo. 2006. “Volumetric quantitative computed tomography of the proximal femur: Relationships linking geometric and densitometric variables to bone strength. Role for compact bone.” Osteoporosis international: a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 17: 855-64.
Braten M. A., T. E. Terjesen, I. V. Rossvoll. 1993. “Torsional deformity after intramedullary nailing of femoral shaft fractures. Measurement of anteversion angles in 110 patients.” The Journal of Bone and Joint Surgery. British volume. 75 (5):799-803.
Bronner, F., M. C. Farach-Carson, H. I. Roach, & SpringerLink (Online service). 2010. Bone and Development. London: Springer-Verlag.
Bulagouda, R. S., S. D. Desai, D. G. Gannur, and U. C. Nuchhi. 2014. "A Study on the Neck Shaft Angle of Femur in-100 Macerated Bones." Indian Journal of Forensic Medicine & Toxicology 8(1): 116.


50
Capulli, M., R. Paone, and N. Rucci. 2014. “Osteoblast and osteocyte: games without frontiers.” Archives of Biochemistry and Biophysics 561 (3-12).
Cartmill, Matt, and Katharine Milton. 1977. "The lorisiform wrist joint and the evolution of ‘brachiating’ adaptations in the Hominoidea." American Journal of Physical Anthropology 47 (2): 249-272.
Carvalho, Susana, Dora Biro, Eugenia Cunha, Kimberley Hookings, William C. McGrew, Brian G. Richmond, and Tetsuro Matsuzawa. 2012. "Chimpanzee carrying behaviour and the origins of human bipedality." Current Biology 22 (6): R180-R181.
Chamay, A., P. Tschantz. 1972. “Mechanical influences in bone remodeling. Experimental research on Wolff's law.” Journal of Biomechanics 5 (2): 173-180.
Churchill, Steven Emilio, and Caroline Vansickle. 2017. "Pelvic morphology in Homo erectus and early Homo." The Anatomical Record 300 (5): 964-977.
Cohen, Jacob. 1988. "Statistical power analysis for the behavioral sciences. 2nd."
Coleman, William H. 1969. "Sex differences in the growth of the human bony pelvis." American Journal of Physical Anthropology 31 (2): 125-151.
DeSilva, Jeremy M., Kenneth G. Holt, Steven E. Churchill, Kristian J. Carlson, Christophers. Walker, Bernhard Zipfel, and Lee R. Berger. 2013. "The lower limb and mechanics of walking in Australopithecus sediba" Science 340 (6129): 1232999.
Doran, D. M. 1997. "Ontogeny of locomotion in mountain gorillas and chimpanzees." Journal of Human Evolution 32 (4): 323-344.
Doyle, Anthony J., and Sandra Winsor. 2011. "Magnetic resonance imaging (MRI) lower limb length measurement." Journal of Medical Imaging and Radiation Oncology 55 (2): 191-194.
Duarte, Cidalia, Joao Mauricio, Paul B. Pettitt, Pedro Souto, Erik Trinkaus, Hans Van Der Plicht, and Joao Zilhao. 1999. "The early Upper Paleolithic human skeleton from the Abrigo do Lagar Velho (Portugal) and modern human emergence in Iberia." Proceedings of the National Academy of Sciences 96 (13): 7604-7609.


51
Duda, Georg N., Erich Schneider, and Edmund YS Chao. 1997. "Internal forces and moments in the femur during walking." Journal of Biomechanics 30 (9): 933-941.
Dugan, Sheila A., and Krishna P. Bhat. 2005. "Biomechanics and analysis of running gait." Physical Medicine and Rehabilitation Clinics 16 (3): 603-621.
Duncan, R. L., and C. H. Turner. 1995. "Mechanotransduction and the functional response of bone to mechanical strain." Calcified Tissue International 57 (5): 344-358.
Ehrlich, P. J., and L. E. Lanyon. 2002. “Mechanical strain and bone cell function: a review.” Osteoporosis International 13 (9): 688-700.
Felson, David T., Jingbo Niu, K. Douglas Gross, Martin Englund, Leena Sharma, T. Derek V. Cooke, AN Guermazi et al. 2013. "Valgus malalignment is a risk factor for lateral knee osteoarthritis incidence and progression: findings from the Multicenter Osteoarthritis Study and the Osteoarthritis Initiative." Arthritis & Rheumatism 65 (2): 355-362.
Fleagle, John G., JackT. Stern, William L. Jungers, Randall L. Susman, Andrea K. Vangor, and James P. Wells. 1981. "Climbing: a biomechanical link with brachiation and with bipedalism." Symposia of the Zoological Society of London 48: 359-375.
Franciscus, Robert G. 2009."When did the modern human pattern of childbirth arise? New insights from an old Neandertal pelvis." Proceedings of the National Academy of Sciences 106 (23): 9125-9126.
Furey, Edward. 2017. "3D Distance Calculator"; from https://www.calculatorsoup.com - Online Calculator Resource.
Gebo, Daniel L. 1996. "Climbing, brachiation, and terrestrial quadrupedalism: historical precursors of hominid bipedalism." American Journal of Physical Anthropology 101 (1): 55-92.
Gilligan, Ian, Supichya Chandraphak, and Pasuk Mahakkanukrauh. 2013. “Femoral neck- shaft angle in humans: Variation relating to climate, clothing, lifestyle, sex, age and side.” Journal of Anatomy 223 (2): 133-51.


52
Gonzalez, Paula N., Valeria Bernal, and S. Ivan Perez. 2009. "Geometric morphometric approach to sex estimation of human pelvis." Forensic Science International 189 (1-3): 68-74.
Gruss, Laura Tobias, and Daniel Schmitt. 2015. "The evolution of the human pelvis: changing adaptations to bipedalism, obstetrics and thermoregulation." Philosophical Transactions of the Royal Society 370 (1663): 20140063.
Hammer, A. 2015. “The paradox of wolff's theories.” Irish Journal of Medical Science 184 (1): 13-22. doi: 10.1007/s11845-014-1070-y
Hammer, A. 2016. “Wolff: Straight not curved.” Irish Journal of Medical Science (1971 -) 186 (4): 939-946.
Haile-Selassie, Yohannes, Bruce M. Latimer, Mulugeta Alene, Alan L. Deino, Luis Gibert, Stephanie M. Melillo, Beverly Z. Saylor, Gary R. Scott, and C. Owen Lovejoy. 2010. "An early Australopithecus afarensis postcranium from Woranso-Mille, Ethiopia." Proceedings of the National Academy of Sciences 107 (27): 12121-12126.
Harcourt- Smith, W. E., and L. C. Aiello. 2004. “Fossils, feet and the evolution of human bipedal locomotion.” Journal of Anatomy 204 (5): 403-416.
Hausler, Martin, and Peter Schmid. 1995. "Comparison of the pelves of Sts 14 and AL288-1: implications for birth and sexual dimorphism in australopithecines." Journal of Human Evolution 29 (4): 363-383.
Henriksson, Lars. 1980. "Measurement of femoral neck anteversion and inclination: a radiographic study in children." Acta Orthopaedica Scandinavica 51 (186): 1-59.
Holliday, T. W., and A. B. Falsetti. 1995. “Lower limb length of European early modern humans in relation to mobility and climate.” Journal of Human Evolution 29: 141-153.
Holzer, Gerold, Gobertvon Skrbensky, Lukas A. Holzer, and Wolfgang Pichl. 2009. "Hip fractures and the contribution of cortical versus trabecular bone to femoral neck strength." Journal of Bone and Mineral Research 24 (3): 468-474.
Houston, C. Stuart, and Witold A. Zaleski. 1967. "The shape of vertebral bodies and femoral necks in relation to activity." Radiology 89 (1): 59-66.


53
Igbigbi, P. S., and M. Sharrif. 2005. "The bicondylar angle of adult Malawians." American Journal of Orthopedics 34 (6): 291-294.
Jang, I. G., and I. Y. Kim. 2008. “Computational study of Wolff's law with trabecular architecture in the human proximal femur using topology optimization.” Journal of Biomechanics 41 (11): 2353-2361.
Javois, C., C. Tardieu, B. Lebel, R. Seil, C. Hulet, Soc Francaise Arthroscopie, and Societe frangaise d'arthroscopie. 2009. “Comparative anatomy of the knee joint: Effects on the lateral meniscus.” Orthopaedics & Traumatology: Surgery & Research 95 (8): 49-59.
Kay, Robert M., Kai A. Jaki, and David L. Skaggs. 2000. "The effect of femoral rotation on the projected femoral neck-shaft angle." Journal of Pediatric Orthopaedics 20 (6): 736-739.
Kerrigan, D. Casey, Jennifer L. Lelas, Joyce Goggins, Greg J. Merriman, Robert J. Kaplan, and David T. Felson. 2002. "Effectiveness of a lateral-wedge insole on knee varus torque in patients with knee osteoarthritis." Archives of Physical Medicine and Rehabilitation 83 (7): 889-893.
Kibii, Job M., Steven E. Churchill, Peter Schmid, Kristian J. Carlson, Nichelle D. Reed, Darryl J. De Ruiter, and Lee R. Berger. 2011. "A partial pelvis of Australopithecus sediba." Science 333 (6048): 1407-1411.
Kimura, Tasuku. 1996. "Centre of gravity of the body during the ontogeny of chimpanzee bipedal walking." Folia Primatologica 66 (1-4): 126-136.
Kozma, Elaine E., Nicole M. Webb, William EH Harcourt-Smith, David A. Raichlen, Kristiaan D'Aout, Mary H. Brown, Emma M. Finestone, Stephen R. Ross, Peter Aerts, and Herman Pontzer. 2018. "Hip extensor mechanics and the evolution of walking and climbing capabilities in humans, apes, and fossil hominins." Proceedings of the National Academy of Sciences 201715120.
Kronenberg, H. M. 2003. “Developmental regulation of the growth plate.” Nature 423 (6937): 332.
Kullmer, O., F. Schrenk, B. Dorrhoffer. 2003. “High resolution 3D-image analysis of ape, hominid and human footprints.” Courier Forsch Senkenberg 243: 85-91.


54
Lang, T. F., J. H. Keyak, M. W. Heitz, P. Augat, Y. Lu, A. Mathur, and H. K. Genant. 1997. "Volumetric quantitative computed tomography of the proximal femur: precision and relation to bone strength." Bone 21 (1): 101-108.
Levin, S. A. M. 1997. "10. A different approach to the mechanics of the human pelvis: tensegrity." In Movement, Stability, and Lower Back Pain, edited by Vleeming atal, 157-167. Churchill Livingstone, London.
Levinger, P., H. B. Menz, A. D. Morrow, J. R. Bartlett, J. A. Feller, and N. R. Bergman. 2013. “Relationship between foot function and medial knee joint loading in people with medial compartment knee osteoarthritis.” Journal of Foot and Ankle Research 6 (1): 33.
Locke, D. P., L. W. Hillier, L. W. C. Warren et al. 2011. “Comparative and demographic analysis of orangutan genomes.” Nature 469: 529-533.
Lordkipanidze, David, Tea Jashashvili, Abesalom Vekua, Marcia S. Ponce de Leon, Christoph PE Zollikofer, G. Philip Rightmire, Herman Pontzeretal. 2007. "Postcranial evidence from early Homo from Dmanisi, Georgia." Nature 449 (7160): 305.
Lovejoy, C. Owen, and Kingsbury G. Heiple. 1970. "A reconstruction of the femur of Australopithecus africanus." American Journal of Physical Anthropology 32 (1): 33-40.
Lovejoy, C. Owen, Kingsbury G. Heiple, and Albert H. Burstein. 1973. "The gait of Australopithecus." American Journal of Physical Anthropology 38 (3): 757-779.
Lovejoy, C. Owen. 1988. "Evolution of human walking." Scientific American 259 (5): 118-125.
Lovejoy, C. Owen, Richard S. Meindl, James C. Ohman, Kingsbury G. Heiple, and Tim D. White. 2002. "The Maka femur and its bearing on the antiquity of human walking: applying contemporary concepts of morphogenesis to the human fossil record." American Journal of Physical Anthropology 119 (2): 97-133.
Lovejoy, C. Owen. 2005. "The natural history of human gait and posture: part 2. Hip and thigh." Gait & Posture 21 (1): 113-124.
Lovejoy, C. Owen. 2007. "The natural history of human gait and posture: Part 3. The knee." Gait & Posture 25 (3): 325-341.


Lovejoy, C. Owen, Gen Suwa, Scott W. Simpson, Jay H. Matternes, and Tim D. White. 2009. "The great divides: Ardipithecus ramidus reveals the postcrania of our last common ancestors with African apes." Science 326 (5949): 73-106.
55
Mackie, E. J., Y. A. Ahmed, L. Tatarczuch, K. Chen, and M. Mirams. 2008. “Endochondral ossification: How cartilage is converted into bone in the developing skeleton.” International Journal of Biochemistry and Cell Biology 40(1): 46-62.
Marzke, Mary W., Julie M. Longhill, and Stanley A. Rasmussen.1988. "Gluteus maximus muscle function and the origin of hominid bipedality." American Journal of Physical Anthropology 77 (4): 519-528.
McHenry, Henry M., and Robert S. Corruccini. 1978. "The femur in early human evolution." American Journal of Physical Anthropology 49 (4): 473-487.
Meakin, L. B., J. S. Price, and L. E. Lanyon. 2014. “The contribution of experimental in vivo models to understanding the mechanisms of adaptation to mechanical loading in bone.” Frontiers in Endocrinology 5.
Metzger, Thomas A., Stephen A. Schwaner, Anthony J. LaNeve, Tyler C. Kreipke, and Glen L. Niebur. 2015. “Pressure and shear stress in trabecular bone marrow during whole bone loading.” Journal of Biomechanics 48 (12): 3035-43.
Michelotti, J., and J. Clark. 1999. “Femoral neck length and hip fracture risk.” Journal of Bone and Mineral Research 14 (10): 1714-1720.
Mirza, Saqeb B., Douglas G. Dunlop, Sukhmeet S. Panesar, Syed G. Naqvi, Shafat Gangoo, and Saif Salih. 2010. "Basic science considerations in primary total hip replacement arthroplasty." The Open Orthopaedics Journal 4: 169.
Mullender, M. G., and R. Huiskes. 1995. “Proposal for the regulatory mechanism of Wolff's law.” Journal of Orthopaedic Research 13 (4): 503-512.
Nawathe, S., B. P. Nguyen, N. Barzanian, H. Akhlaghpour, M. L. Bouxsein, and T. M. Keaveny. 2015. “Cortical and trabecular load sharing in the human femoral neck.” Journal of Biomechanics 48 (5): 816-822.


Nordstrom, P., Pettersson, U., and R. Lorentzon. 1998. “Type of physical activity, muscle strength, and pubertal stage as determinants of bone mineral density and bone area in adolescent boys.” Journal of Bone and Mineral Research 13 (7): 1141-1148.
56
Pauwels, Friedrich. 1980. Biomechanics of the locomotor apparatus: contributions on the functional anatomy of the locomotor apparatus. Berlin: Springer-Verlag.
Pandya, A. M., T. C. Singel, M. M. Patel, and D. V. Gohil. 2008. "A study of the femoral bicondylar angle in the Gujarat region." Journal of the Anatomical Society of India 57 (2): 131-134.
Pickford, Martin, Brigitte Senut, Dominique Gommery, and Jacques Treil. 2002. "Bipedalism in Orrorin tugenensis revealed by its femora." Comptes Rendus Palevol 1 (4): 191-203.
Pontzer, Herman, David A. Raichlen, and PeterS. Rodman. 2014. "Bipedal and quadrupedal locomotion in chimpanzees." Journal of Human Evolution 66: 64-82.
Porter, A. M. W. 1999. "Modern human, early modern human and Neanderthal limb proportions." International Journal of Osteoarchaeology 9 (1): 54-67.
Purcell, Maureen W. 2013. "Sex Differences in the Femur and Acetabulum: Biomechanical Analysis with Forensic Significance."
Raichlen, David A., Adam D. Gordon, William EH Harcourt-Smith, Adam D. Foster, and Wm Randall Haas Jr. 2010. "Laetoli footprints preserve earliest direct evidence of human-like bipedal biomechanics." PLoS One 5 (3): e9769.
Rak, Yoel, and Baruch Arensburg. 1987. "Kebara 2 Neanderthal pelvis: first look at a complete inlet." American Journal of Physical Anthropology 73 (2): 227-231.
Razali, Nornadiah Mohd, and Yap Bee Wah. 2011. "Power comparisons of shapiro-wilk, kolmogorov-smirnov, lilliefors and anderson-darling tests." Journal of Statistical Modeling and Analytics 2 (1): 21-33.
Remis, Melissa. 1995. "Effects of body size and social context on the arboreal activities of lowland gorillas in the Central African Republic." American Journal of Physical Anthropology 97 (4): 413-433.


Richmond, Brian G., David R. Begun, and David S. Strait. 2001. "Origin of human bipedalism: The knuckle- walking hypothesis revisited." American Journal of Physical Anthropology 116 (S33): 70-105.
57
Richmond, Brian G., and William L. Jungers. 2008. "Orrorin tugenensis femoral morphology and the evolution of hominin bipedalism." Science 319 (5870): 1662-1665.
Roberts, B. C., L. B. Solomon, G. Mercer, K. J. Reynolds, D. Thewlis, and E. Perilli. 2017. "Joint loading and proximal tibia subchondral trabecular bone microarchitecture differ with walking gait patterns in end-stage knee osteoarthritis." Osteoarthritis and Cartilage 25 (10): 1623-1632.
Rosenberg, Karen R., Lu Zune, and Christopher B. Ruff. 2006. "Body size, body proportions, and encephalization in a Middle Pleistocene archaic human from northern China." Proceedings of the National Academy of Sciences of the United States of America 103 (10): 3552-3556.
Ruff, Christopher B. 1994. "Morphological adaptation to climate in modern and fossil hominids." American Journal of Physical Anthropology 37 (S19): 65-107.
Ruff, Christopher B. 1995. "Biomechanics of the hip and birth in early Homo." American Journal of Physical Anthropology 98 (4): 527-574.
Ruff, Christopher. 1998. "Evolution of the hominid hip." In Primate locomotion, pp. 449-469. Springer, Boston, MA.
Ruff, Christopher B. 2002. “Variation in human body size and shape.” Annual Review of Anthropology 31: 211-232.
Ruff, Christopher B. 2002. "Long bone articular and diaphyseal structure in Old World monkeys and apes. I: locomotor effects." American Journal of Physical Anthropology 119 (4): 305-342.
Ruff, Christopher. 2010. "Body size and body shape in early hominins-implications of the Gona pelvis." Journal of Human Evolution 58 (2): 166-178.
Ruff, Christopher. 2017. "Mechanical constraints on the hominin pelvis and the ‘obstetrical dilemma’." The Anatomical Record 300 (5): 946-955.


58
Ryan, Timothy M., and Gail E. Krovitz. 2006. "Trabecular bone ontogeny in the human proximal femur." Journal of Human Evolution 51 (6): 591-602.
Schindelin, Johannes, Ignacio Arganda-Carreras, Erwin Frise, Verena Kaynig, Mark Longair, Tobias Pietzsch, Stephan Preibisch et al. 2012. "Fiji: an open-source platform for biological-image analysis." Nature Methods 9 (7): 676.
Senut, Brigitte. 2006. "Arboreal origin of bipedalism." In Human Origins and Environmental Backgrounds, pp. 199-208. Springer, Boston, MA.
Shefelbine, S. J., C. Tardieu, and D. R. Carter. 2002. “Development of the femoral bicondylar angle in hominid bipedalism.” Bone 30 (5): 765-770.
Shelton, Michael. 2015. One of the body’s hardest structures. [PowerPoint Slides] Retrieved from http://slideplayer.com/slide/4698157/.
Simpson, Scott W., Jay Quade, Naomi E. Levin, Robert Butler, Guillaume Dupont-Nivet, Melanie Everett, and Sileshi Semaw. 2008. "A female Homo erectus pelvis from Gona, Ethiopia." Science 322 (5904): 1089-1092.
Sims, N., and C. Vrahnas. 2014. “Regulation of cortical and trabecular bone mass by communication between osteoblasts, osteocytes and osteoclasts.” Archives of Biochemistry and Biophysics 561: 22-28.
Sockol, Michael D., David A. Raichlen, and Herman Pontzer. 2007. "Chimpanzee locomotor energetics and the origin of human bipedalism." Proceedings of the National Academy of Sciences 104 (30): 12265-12269.
Stern, JackT., and Randall L. Susman. 1981. "Electromyography of the gluteal muscles in Hylobates, Pongo, and Pan: implications for the evolution of hominid bipedality." American Journal of Physical Anthropology 55 (2): 153-166.
Stern Jr., JackT., and Randall L. Susman. 1983. "The locomotor anatomy of Australopithecus afarensis." American Journal of Physical Anthropology 60 (3): 279-317.
Stern, Jack T. 2000. "Climbing to the top: a personal memoir of Australopithecus afarensis. Evolutionary Anthropology: Issues, News, and Reviews 9 (3): 113-133.


59
Sugano, Nobuhiko, Philip C. Noble, Emir Kamaric, Joseph K. Salama, Takahiro Ochi, and Hugh S. Tullos. 1998. "The morphology of the femur in developmental dysplasia of the hip." The Journal of Bone and Joint Surgery. British Volume 80 (4): 711-719.
Tague, Robert G., and C. Owen Lovejoy. 1986. "The obstetric pelvis of AL 288-1 (Lucy)." Journal of Human Evolution 15 (4): 237-255.
Tague, Robert G. 1989. "Variation in pelvic size between males and females." American Journal of Physical Anthropology 80 (1): 59-71.
Tanck, E. J. M., G. J. Hannink, R. Ruimerman, P. Buma, E. H. Burger, and R. Huiskes. 2006. “Cortical bone development under the growth plate is regulated by mechanical load transfer.” Journal of Anatomy 208 (1): 73-79.
Tanner, J. M. 1990. “The human growth curve.” In Human Biology edited by G. A. Harisson, J. M. Tanner, D. R. Pilbeam, and P. T. Baker. Oxford: Oxford University Press, pp. 320-360.
Tardieu, Christine, and Erik Trinkaus. 1994. "Early ontogeny of the human femoral bicondylar angle." American Journal of Physical Anthropology 95 (2): 183-195.
Tardieu, Christine, and Holger Preuschoft. 1996. "Ontogeny of the knee joint in humans, great apes and fossil hominids: pelvi-femoral relationships during postnatal growth in humans." Folia Primatologica 66 (1-4): 68-81.
Tardieu, C., and J. P. Damsin. 1997. “Evolution of the angle of obliquity of the femoral diaphysis during growth—correlations.” Surgical and Radiologic Anatomy 19 (2): 91-97.
Tardieu, Christine, Yann Glard, Emmanuelle Garron, Christophe Boulay, J- L. Jouve, Olivier Dutour, Gilles Boetsch, and Gerard Bollini. 2006. "Relationship between formation of the femoral bicondylar angle and trochlear shape: independence of diaphyseal and epiphyseal growth." American Journal of Physical Anthropology 130 (4): 491-500.
Taylor, M. E., K. E. Tanner, M. A. R. Freeman, and A. L. Yettram. 1996. "Stress and strain distribution within the intact femur: compression or bending?." Medical Engineering & Physics 18 (2): 122-131.


60
Teitelbaum, S. L. 2000. “Bone resorption by osteoclasts.” Science 289 (5484): 1504-1508.
Thorpe, Susannah KS, and Robin H. Crompton. 2006. "Orangutan positional behavior and the nature of arboreal locomotion in Hominoidea." American Journal of Physical Anthropology 131 (3): 384-401.
Thorpe, Susannah KS, Roger L. Holder, and Robin H. Crompton. 2007. "Origin of human bipedalism as an adaptation for locomotion on flexible branches." Science 316 (5829): 1328-1331.
Thorpe, Susannah KS, Juliet M. McClymont, and Robin H. Crompton. 2014. "The arboreal origins of human bipedalism." Antiquity 88 (341): 906-914.
Trinkaus, Erik. 1993. "Femoral neck-shaft angles of the Qafzeh-Skhul early modern humans, and activity levels among immature Near Eastern Middle Paleolithic hominids." Journal of Human Evolution 25 (5): 393-416.
Videan, Elaine N., and W. C. McGrew. 2002. "Bipedality in chimpanzee (Pan troglodytes) and bonobo (Pan paniscus)'. testing hypotheses on the evolution of bipedalism." American Journal of Physical Anthropology 118 (2): 184-190.
Wang, Y., L. M. McNamara, M. B. Schaffler, and S. Weinbaum. 2007. “A model for the role of integrins in flow induced mechanotransduction in osteocytes.” Proceedings of the National Academy of Sciences 104 (40): 15941-15946.
Warrener, Anna. 2011. "Pelvic shape, hip abductor mechanics and locomotor energetics in extinct hominins and modern humans."
Warrener, Anna G., Kristi L. Lewton, Herman Pontzer, and Daniel E. Lieberman. 2015. "A wider pelvis does not increase locomotor cost in humans, with implications for the evolution of childbirth." PloS one 10 (3): e0118903.
Warrener, Anna G. 2017. "Hominin hip biomechanics: changing perspectives." The Anatomical Record 300 (5): 932-945.


61
Weaver, Timothy D. 2003. "The shape of the Neandertal femur is primarily the consequence of a hyperpolar body form." Proceedings of the National Academy of Sciences 100 (12): 6926-6929.
Weinbaum, S., S. C. Cowin, and Yu Zeng. 1994. "A model for the excitation of osteocytes by mechanical loading-induced bone fluid shear stresses." Journal of Biomechanics 27 (3): 339-360.
Wetzsteon, R. J., B. S. Zemel, J. Shults, K. M. Howard, L. W. Kibe, and M. B. Leonard. 2011. “Mechanical loads and cortical bone geometry in healthy children and young adults.” Bone 48 (5): 1103-1108.
WoldeGabriel, Giday, Stanley H. Ambrose, Doris Barboni, Raymonde Bonnefille, Laurent Bremond, Brian Currie, David DeGusta et al. 2009. "The geological, isotopic, botanical, invertebrate, and lower vertebrate surroundings of Ardipithecus ramidus." Science 326 (5949): 65-65e5.
Wolff, J. 1986. The Law of Bone Remodelling. Translated by P. Maquetand R. Furlong. A. Hirschwald, Berlin. First published 1892.
Woo, S. L., S. C. Kuei, D. Amiel, M. A. Gomez, W. C. Hayes, F. C. White, and W. H. Akeson. 1981. “The effect of prolonged physical training on the properties of long bone: a study of Wolff's Law.” JBJS 63 (5): 780-787.
Yu, Huiying, Murad Alaqtash, Eric Spier, and T. Sarkodie-Gyan. 2010. “Analysis of muscle activity during gait cycle using fuzzy rule-based reasoning.” Measurement 43 (9): 1106-14.
Zipfel, Bernhard, Jeremy M. DeSilva, Roberts. Kidd, Kristian J. Carlson, Steven E. Churchill, and Lee R. Berger. 2011. "The foot and ankle of Australopithecus sediba." Science 1202703


Full Text

PAGE 1

SKELETAL ANALYSIS OF THE FORMATION OF THE BICONDYLAR ANGLE by MARY SHIRLEY MITCHELL A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Arts Anthropology Program 2018

PAGE 2

ii © 2018 MARY SHIRLEY MITCHELL ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Arts degree by Mary Shirley Mitchell has been approved for the Anthropology Program by Anna Warrener, Chair Caley Orr Charles Musiba Date: December 15, 2018

PAGE 4

iv Shirley Mitchell, Mary (M.A., Anthropology) Skeletal Analysis of the Formation of the Bicondylar Angle Thesis directed by Assistant Professor Anna Warrener ABSTRACT The femoral bicondylar angle is often associated with bipedality in the fossil record because it forms during ontogeny as the result of biomechanical stress. However, the exact mechanisms responsible for its forma tion remain unknown. Two dominant theories based on static models are assumed in the literature, one focused on skeletal architecture at the proximal femur and one on architecture at the distal femur. The proposed relationships have never been statistical ly analyzed. This study aims to investigate several of the skeletal measurements at the proximal and distal femur often associated with the theories on the formation of the bicondylar angle, in order to determine if correlations do exist between them and t he degree of bicondylar angle in adulthood. The form and content of this abstract are approved. I recommend its publication. Approved: Anna. G. Warrener

PAGE 5

v ACKNOWLEDGEMENTS This process began under the always wry and compelling tutelage of Dr. Caley Orr, whos edits along the way kept me laughing even when I wanted to cry. Thank you for everything. I would never have finished it if not for the patient efforts of Dr. Anna Warrener, whose firm hand and empathetic ear guided me as I stumbled down an unknown pa th. I am wanted to hear. To my Will, your support has meant the world to me and I could ask for nothing more in a teammate. My parents Richard and Jackie are, as always, my two anchors in the stormiest of seas.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION 1 Background 2 Wolff 2 Anatomical Description of the Pelvis and Femur 5 Biomechanics of the Lower Limb During Locomotion 6 Competing Models for the Forma tion of the Bicondylar Angle 13 Proximal Femur 13 Distal Femur 16 The Fossil Record 17 Hypotheses 24 II. MATERIALS AND METHODS 26 Anatomical Measurements 26 Analysis 30 Results 31 III. DISCUSSION 43 Conclusion 47 References 48

PAGE 7

1 CHAPTER I INTRODUCTION Bipedality is considered one of humanity's defining traits. Its evolution marked a turning point in our species by changing the way we moved across the landscape and freeing the hands of our ancestors, eventually allowing them to cre ate and effectively wield tools, (Marzke et al . 1988) and increasing their capacity to carry supplies over long distances (Carvalho et al . 2012). Many osteological changes were necessary to habitually balance on two legs, and the development of the bicondy lar angle of the femur has long been associated with remodeling of et al . 2002, pg. 765), the bicondylar ang le brings the body cent er of mass over the foot allowing for a smoother and more stable gait (Tardieu and Trinkaus 1994). Because it forms as a response to mechanical stress during ontogeny, bicondylar angle is useful for identifying bipedality in hominins . The formation of this angle is thought to result from mechanical strain during locomotion at the proximal and distal ends of the femur. At the proximal end, femoral neck shaft length and angle interact with pelvic width, influencing magnitude of bicondyl ar angle. At the distal end of the femur, stress at the knee joint, particularly during early development, results in the asymmetrical growth of the lateral and medial condyles leading to an offset in femoral shaft (Tardieu and Trinkaus 1994). The mechani sms by which bone responds to mechanical stress to form the bicondylar angle are complex. At birth, babies are genu varum , or bow legged, with a bicondylar angle of 0°. However as they begin to walk the angle forms in response to the new strains being plac ed on the proximal and distal epiphysis at the growth plates and the bicondylar angle is thought to reach stasis at around age seven (Tardieu and Trinkaus 1994; Shefelbine et al . 2002). The fossil record suggests that this pattern of ontogenetic bone formation was the same in extinct populations of hominins (Tardieu and Trinkaus 1994). Comparative studies of human and extant ape locomotion allow a better understanding of the mechanical stresses that influence bone

PAGE 8

2 shape during development. This information can be applied to the fossil record allowing greater understanding of the mobility strategies practiced throughout our evolution. Femoral length, biacetabular width, neck shaft length and angle, and femoral condylar size all have been presumed to be determina nts of bicondylar angle by various researchers (Ruff 1995, Shefelbine et al . 2002, Lovejoy et al . 2002, Tardieu et al . 2006), but no studies have systematically assessed patterns of correlation between these traits and many of these bone features which are not fully developed by the age of seven when bicondylar angle if thought to reach stasis. The goal of this study is to better understand the relationship between key features of the femur and pelvis and to test the assumptions of how mechanical stress at the proximal and distal ends of the femur are responsible for the formation of the bicondylar angle in humans. In the current model, an increase in biacetabular width or a decrease in neck shaft angle should lead to a greater bicondylar angle by pushing the pr oximal femur out laterally. A larger medial condyle should also correlate with larger bicondylar angles by creating more asymmetry at the growth plate during ontogeny, causing increased asymmetrical growth in the femoral shaft. Longer femoral length should correlate with a lesser degree of bicondylar angle as it will decrease the adducting moment at the knee and place less stress on the femoral condyles. The results of this analysis may be useful for understanding the plasticity of bicondylar angle beyond t he juvenile period if traits like femoral length and biacetabular width, which are still developing well into adolescence, are strongly correlated with its magnitude. Additionally, a better understanding of osteological features associated with bicondylar angle could help elucidate locomotor behavior in extinct hominins given the mosaic nature of the postrcranial elements currently known in the fossil record (Lovejoy, 2005). Background nges within bone are the result of load bearing, and that altered mechanical forces result in bone modification during life (Wolff, 1892). Wolff

PAGE 9

3 also proposed the Trajectorial Theory, describing mathematically the lines of compression and tension within in ternal structure of the bone exposed to physical stressors. The credibility of and the accuracy of the mathematical models of stress. However, the concept of skeleta l remodeling in response to stress is generally accepted and has been supported by decades of observational and experimental data (Chamay and Tschantz, 1972; Woo et al ., 1981; Mullender and Huiskes, 1995; Jang and Kim, 2008; Barak et al ., 2011). In particu frequently been used in analyses of the proximal femur to understand it s plasticity during ontogeny (Hammer 2015; Hammer 2016). To understand how remodeling of the femur due to mechanical stress occurs, the processes of long bone form ation and growth will briefly be described. Long bone growth is the result of two different but homologous types of bone growth called endochondral ossification and appositional growth. Endochondral ossification is responsible for the longitudinal growth o f the bone. In utero, mesenchymal cells begin to clump together to form an early cartilage structure after which cellular death leaves a lattice for bone building cells to begin the process of ossification and transformation into bone. The process conclude s with the calcification of all the cartilage save that on the articular surfaces (Mackie et al . 2008). The primary site of ossification is the bone collar around the middle of the shaft, and the secondary site of ossification is at the epiphyses. This lea ves the pliable growth plate with proliferating chondrocytes at both ends of the shaft where bone growth continues until vertical growth has ceased and ossification concludes (Kronenberg 2003). Appositional growth determines the thickness of the bone. The process is the same as endochondral ossification, except the chondrocyte cells secrete extracellular matrix on the peripheral surface of the developing bone beneath the perichondrium. Appositional bone growth continues throughout development and persists u ntil age 20 to maintain an optimal cortical bone size and structure relative to mechanical stress (Bronner et al ., 2010).

PAGE 10

4 Skeletal remodeling in adulthood is a continuous process regulated by three types of bone dwelling cells, namely the osteoclasts, oste oblasts, and osteocytes. (Sims and Vrahnas, 2014). Osteoclasts break down existing bone by demineralizing sections of bone through secreted enzymes, breaking down organic elements like collagen and inorganic elements like calcium and phosphorus (Teitelbaum , 2000). Osteoblasts build new bone by constructing an organic matrix which is mineralized by laying down hydroxyapatite which hardens the bone (Capulli et al ., 2014). Osteocytes are osteoblasts that have been entrapped within mineralized bone and become t he living bone cell. Osteocytes use small channels within the bone called canaliculi to carry communicative chemicals to osteoclasts, osteoblasts, and other osteocytes relaying information about the state of the bone and regulating bone destruction and for mation (Bonewald, 2011; Capulli et al ., 2014). The process by which the cells translate mechanical loads into electrochemical activity is called mechanotransduction and is most often attributed to osteocyte function (Wang et al ., 2007; Bonewald, 2014). Me chanical loading causes tissue deformation, but also prompts fluid flow within the physical passages of the bone utilized by the osteocytes for communication. This fluid movement is believed to stimulate the highly reactive osteocytes into releasing signal ing molecules to osteoclasts and osteoblasts (Ehrlich and Lanyon, 2002; Bonewald, 2014). The result is localized bone growth dependent on the type and locus of the loading pressures (Meakin et al ., 2014). Load bearing functional related pressures from birt h to adolescence have a significant effect on the development of cortical bone in the femur, influencing the strength of the bone throughout life (Ruff 2002; Tanck et al . 2006). Long bones respond to various levels of physical activity through structural c hanges, regardless of age, sex, or ethnicity (Wetzsteon et al . 2011), and are most reactive to dynamic mechanical loading, as opposed to static mechanical loading ( Nordström e t al . 1998; Meakin et al . 2014). Particularly important are the dynamic loads

PAGE 11

5 exp erienced in bipedality as the mechanical strains of walking and running are transmitted through the long bones of the legs and distributed through the pelvis. Anatomical Description of the Pelvis and Femur The bony pelvic femoral complex is formed by the r ight and left os coxae and femora. The pelvis consists of two os coxae, jointed at the sacrum posteriorly and the pubic symphysis anteriorly. The pelvic brim separates the greater pelvis from the lesser pelvis, and defines the pelvic inlet. The greater pel vis is made of the iliac blades, and the lesser pelvis consist of ischium, os pubis, and sacrum. The hip joints are ball and socket joints where the acetabulum articulates with the round femur head. The bony pelvis functions as an anchor for 45 muscles tha t allow movement of the trunk and lower limbs (Gruss and Schmitt 2015), and as a mechanical structure that absorbs and transmits loads during standing, locomotion, and childbirth (Levin 1997). The femur is the longest and heaviest bone in the body. Its ep iphyses are composed of spongy trabecular bone encased in a superficial sheath of dense cortical bone. The proximal terminus includes the round head which articulates with the acetabulum, a thinner neck shaft, and the greater and lesser trochanters which a re the main attachment sites for muscles originating on the pelvis and trunk. The distal end consists of the lateral and medial condyles which articulate with the tibia to form the knee joint. The intercondylar fossa serves as the site of attachment for mo st of the ligaments of the knee joint, and the patella sits between the condyles on the anterior surface within the patellofemoral groove. The femoral shaft is comprised of a cortical outer table of bone that contains a medullary cavity housing bone marrow . The human femur is triangular in cross section with a distinctive posterior projection called the linea aspera. In healthy adults, the femoral shaft exhibits an anterior bowing. The femur serves as the attachment point or origination site of 23 different muscles which exert force on the hip and knee during locomotion. In humans, the femur helps transmit and absorb ground reaction and

PAGE 12

6 gravitational forces during bipedality (Warrener 2017) and aids in stability by bringing the center of mass underneath the body (Lovejoy 2005). Several major muscle groups are involved in gait at the hip joint. The gluteus muscles (minimus, medius, and maximus) originate on the outer ilium and attach at the greater trochanter and gluteal tuberosity and serve to extend and abd uct the hip as well as stabilizing the body in the mediolateral plane. The adductor muscles (magnus, brevis, longus) originate at the pubis and attach at the linea aspera along the femoral shaft, and are responsible for hip adduction and internal rotation. The iliopsoas originates at the lower spine and inner ilium and attaches at the lesser trochanter. The iliopsoas is the strongest hip flexor. The rectus femoris originates at the anterior inferior iliac spine and attaches at the base of the patella; it is part of the quadricep group and acts as a knee extensor. Finally, the hamstring muscles ( semimembranosus, semitendinosus and biceps femoris ) originate at the ischial tuberosity and attach posteriorly to the knee joint and act as hip extensors and knee fle xors. Biomechanics of the Lower Limb During Locomotion Biomechanical stress and bone loading is described in terms of the mechanical forces acting on the bone at any one time (Fig. 1). The human gait cycle in walking begins at heel strike with one foot an d ends when the same foot heel strikes again, and is therefore made of two steps. During walking it consists of 60% stance phase in which the foot is on the ground (20% of that is double support where both feet are on the ground) and 40% swing phase in whi ch the foot is in the air during limb advancement (Yu et al . 2010). During running forces are greater and gait cycle consists of approximately 40% stance phase and 60% swing phase (including about 20% double float in which both feet are off the ground) dep endent on speed (Dugan and Baht 2005). Stance phase is where most loading occurs in the femur, with mainly compressive force at the head, tensile forces at the greater trochanter, and both compressive and tensile forces at the lesser trochanter (Taylor et al . 1996; Duda et al . 1997).

PAGE 13

7 Figure 1 . Biomechanical stress as it relates to bone. Republished from hardest structures. Shelton, Michael (2015) [PowerPoint Slides] http://slidepl ayer.com/slide/4698157/ . In the study of human bipedality, the gluteal muscles have been of particular interest because of their role in pelvic equilibrium (Stern and Susman 1981, Lovejoy 2005, Al Imam 2017, Warrener 2017). The gluteus minimus and medius serve to keep the hip upright, preventing a drop and lateral shift in the center of gravity, while the gluteus maximus is the primary stabilizer of the trunk over the lower limb (Lovejoy 2005). These muscles function in great apes as powerful hip extensor s during quadrupedal locomotion, but their demand in bipedal locomotion as hip abductors would have changed the shape and orientation of the muscles and ilium (Stern and Susman 1981, Lovejoy 1988, Kozma et al . 2018, discussed below ). Joint Reaction Force (JRF) is the force generated within a joint as a response to the forces acting upon it, and in the hip it is the result of balancing the moments of body weight and abductor tension. The equilibrium formula for the hip joint states that the abductor force t imes the moment arm to the axis of rotation (roughly center of the femoral head) will equal the body weight times the moment arm from the centerline of the body to the axis of rotation. Anything that affects the length of the moment arm will impact the JRF . If we increase or decrease a

PAGE 14

8 measurement on one side of the formula, a reactionary increase or decrease of a related measurement must occur in order to maintain biomechanical stability in the joint. For instance, an increase in biacetabular width would serve to increase the body weight moment arm from the axis of rotation to the center of the body, which would increase the moment on the hip. To balance the equilibrium formula, there can be three reactionary changes. Internally, the body weight would need to decrease to maintain equilibrium. Externally, either a decrease in neck shaft angle would lengthen the external moment arm or the abductor muscles would have to generate increased force (Fig. 2). These longer moment arms would also increase the distanc e from the center of the body to the proximal femur, creating a more lateral orientation, greater valgus at the knee and thereby angulation of the femur, and greater bicondylar angles (Lovejoy et al . 2002). The need to balance hip JRF is what many research ers consider the driving force in pelvic/femoral morphology. A .

PAGE 15

9 B . C .

PAGE 16

10 D . Figure 2 . Hip joint reaction force and equilibrium formula reactions. Panel A : Balanced equilibrium formula. Panel B : Increased body weight moment arm and reactionary decreased body weight. Panel C: Increased body weight moment arm and reactionary increased abductor moment arm. Panel D : Increased body weight moment arm and reactionary increased abductor force. Many researchers have suggested that mechanical stress is reflected i n the architecture of trabecular bone inside the femoral head and neck to better absorb compressive, shear, and tensile strain (Bousson et al . 2006; Baum et al . 2010; Metzger et al . 2015). Lang et al . (1997) found a direct correlation between trabecular bo ne density and bone strength, and Weinbaum et al . (1994) found that shear stress in bovine humeri followed the lines of trabecular density. Stresses experienced by bipeds would most likely be very different, but trabecular bone has shown to be highly respo nsive to mechanical stress in humans (Duncan and Turner 2995; Metzger et al . 2015; Roberts et al. 2017). Studies of this kind for the knee joint in humans are few, but Roberts et al . (2017) reported responsive changes in the trabecular bone of the tibia in response to pathological gait patterns in those suffering from knee osteoarthritis. However, Nawathe et al . (2015) argue that trabecular and cortical bone share loads, and that cortical bone supports up to 90% of the frontal plane bending moment in stanc e loading at

PAGE 17

11 the proximal femur. Holzer et al . (2009) found that the contribution of trabecular bone to overall bone strength was < 10%. The load bearing properties of trabecular vs. cortical bone under various strains in the femur is still a topic of deba te (Nawathe et al . 2015), but one aspect to the understanding of bicondylar angle development as both proximal and distal epiphysis are composed of trabeculae. The bicondylar angle (Fig. 3) is typically measured in the coronal plane, and it is the angle fr om which the line through the center of the femur shaft meets the line of the infracondylar plane, or the line on which the condyles sit flat (Shefelbine et al . 2002). This angle adducts the distal femur (Fig. 4), which lessens the adduction moment in the knee joint reducing transverse shear stress and allowing the trunk of the body to sit dir ectly over the center of mass (Tardieu and Trinkaus 1994; Javois et al . 2009). Hypothetically this angle serves to reduce lifetime wear of the knee joint and stabilize s it during the stance phase of locomotion.

PAGE 18

12 Figure 3. Bicondylar angle = . Republished from "Ontogeny of the knee joint in humans, great apes and fossil hominids: pelvi femoral relationships during postnatal growth in humans." Tardieu, Christine, and H olger Preuschoft (1996) Folia Primatologica Academic Press. Figure 4 . The effect of a higher and lower bicondylar angle on the adduct ing moment arm of the Relationship between foot function and medial knee joint loading in Levinger et al . (2013) Journal of Foot and Ankle Research. bicondylar angle forms during childhood in response to the mechanical stressors of bipedality (Tardieu 1996, Tardieu and Damsin 1997; Tardieu 2010), and a study by Tardieu (1996) suggests that it reaches stasis around the age of seven. Modern human babies display no bicondylar angle, and it increases as he/she begins walking bipedality degrees) (Tardieu and Trinkaus 1994). This epigenetic phenomenon is most impo rtant during youth and adolescence while the femoral growth plate is still unossified. However, levels of

PAGE 19

13 bipedal mobility impact the degree of angle throughout a person's life, and the angle will decrease in adults who lose mobility or increase later if m obility is gained (Tardieu 2010). How plastic the bicondylar angle is during adulthood is of interest because many of the skeletal components credited with determining the degree of angle are not fully formed by age seven. Postcranial skeletal development continues into adulthood, with the femur fusing at the epiphysis a nd reaching stasis around age seventeen (Tanner 1990) and pelvic bon es reaching statis around age twenty one (Coleman 1969; Gonzalez et al . 2009). If skeletal architecture is a causal factor in the degree of bicondylar angle, its continued growth past the accepted age of stasis for bicondylar angle may suggest that bicondylar angle formation is more plastic than previously assumed, or that skeletal architecture does not influence bicondylar angle to the ex tent that has been accepted within the current model. The degree of bicondylar angle in modern populations is often cited as being between 8 11 degrees (Tardieu and Trinkaus 1994), however there may be some question as to the variation around that mean in modern adult populations . Igbigbi and Sharrif (2005) found in a study of modern Malawians a range of between 1.5 and 12 degrees of bicondylar angle in both sexes and Pandya et al. (2008) reported bicondylar angles between 2 and 14 degrees in adults of the Gujarat region of India. This study produced a range of between 3 and 13 degrees . If this wide variation is indicative of the population at large, our cited means and comparisons against the fossil populations for whom we have a limited sample may not be as meani ngful as previously assumed. Competing Models for the Formation of Bicondylar Angle Proximal Femur Proximally, a key biomechanical factor is the femoral neck shaft angle (NSA). NSA and neck length are thought to be determinants of bicondylar angle by incre asing or decreasing the moment arm of the hip abductors at the hip joint (Lovejoy et al . 2002). While neck length is genetically regulated, neck shaft angle changes during ontogeny due to mechanical stress

PAGE 20

14 (Houston and Zaleski 1967; Trinkaus 1993). Neck an gles at birth are around 150° and slowly decrease over time, eventually reaching stasis in adolescence with a great degree of variability but usually between 140° and 120° (Henriksson 1980; Trinkaus 1993). Many influencing factors have been suggested for t he development of NSA, but one of the more debated in the literature has been latitude. Climate reliant body types have been well documented in all animal species as a result s will display shorter, stockier builds accompanied by shorter limbs, while bodies evolved in warmer climates will display longer, leaner builds with protracted limbs. This is a phenomena observable in the animal kingdom and demonstrable in both fossil and modern human populations (Ruff 1994; Holliday and Falsetti 1995; Porter 1999; Ruff 2002). Weaver (2003) and Gillian et al . (2013) reported what they interpreted to be a clear correlation between modern human neck shaft angle and latitude. They argue that wider, stockier builds result in greater biacetabular distances, which would place the greater biomechanical strain on the proximal femur during ontogeny, decreasing NSA. Ruff also postulates that greater biacetabular width would specifically increase the mediolateral bending of the neck shaft, forcing a reduction in NSA (Ruff 1995). Femoral neck structure is the result of the arcuate nature of the trabeculae and asymmetry of the cortical bone (Bonneau et al . 2012). Its development remains mysterious in pa rt because of its wide variation (Anderson and Trinkaus 1998), and researchers have long attempted to explain it as a means of increasing femoral rotation at the hip (Kay et al. 2000) crease in its neck shaft angle, acts to reduce the moment at the hip joint, tending to sublux the femoral head and However, the suggestion that NSA might be greater in warmer ada pted builds and vice versa was approached and dismissed by Trinkaus (1993; 1994; Anderson and Trinkaus 1998)

PAGE 21

15 in various studies of modern human populations, whose fluctuation he ascribed to differences in mobility rather than body shape. He argued that pop ulations with greater mobility would place greater mechanical strain on the head of the femur, therefore lowering NSA. Further complicating the variable of neck shaft angle is its antiversion, or the position of the neck anterior to the shaft in the corona l plane (Bulagouda et al . 2014). To solve for this, many (but not all) researchers measure NSA in the coronal plane rather than three dimensions, especially when working with scans over material remains (Bonneau et al . 2012). Some discrepancies in the rese arch may stem from the lack of a standardized measurement protocol for NSA (Bonneau et al . 2012), but much of the debate stems from the biomechanical cause and effect of strain during ontogeny. The lack of consensus in the literature is indicative of the c omplexity involved in NSA development. Presumably this theory of NSA being related to latitude would link smaller NSAs with shorter overall femurs and wider biacetabular widths and larger NSAs with longer femurs and smaller biacetabular widths, but more s tudies would be needed to demonstrate a correlation between NSA and climate adapted body types. Some manner of relationship might exist as a function of a working pelvic femoral complex, but the extent to which they influence one another remains unknown. I f NSA and bicondylar angle prove to be correlated, the nature of that association with climate reliant body types would require much more investigation than what is currently available in the literature. The current model of biacetabular width and its rela tionship to bicondylar angle is problematic in part because it is static, presuming that biomechanical stressors primarily act vertically through the body. A dynamic model would need to take into account the biomechanical forces at work in the joint from a ll directions and during all phases of the gait cycle. Under the static model an increase in biacetabular width would simply result in an increase in hip abductor force or decrease in NSA to balance the moments. As an example, higher bicondylar angles in w omen over men are widely accepted in both modern human and

PAGE 22

16 ancestral populations (Lovejoy 2005). This sexual dimorphism is tied to the obstetrical dilemma, which theorizes that the expanded pelvic dimensions associated with birthing increasingly encephaliz ed neonates during Homo evolution was constrained by mechanical aspects of a bipedal gait (Ruff 2017). In a purely static model, the wider hips needed to give birth would result in larger biacetabular widths and therefore higher degrees of bicondylar angle . Greater biacetabular distances in humans do not develop in females over males until they reach puberty as the result of hormonal changes (Coleman 1969). Should bicondylar angle prove to reach a majority stasis by the age of seven as Tardieu suggests (19 96), then purely skeletal arguments do not explain the discrepancy between higher bicondylar angles in adult women over men. A more dynamic model of biomechanical stress would be needed to satisfactorily describe the forces behind the sexual dimorphism see n in bicondylar angles. However, this static model based on skeletal architecture remains the predominantly used model in fossil analysis (although this may be changing, see Warrener et al . 2015). Distal Femur An alternative model for the formation of bico ndylar angle is that the compressive stress produced at the knee joint during adduction in the stance phase when a ch ild begins to walk bipedally is the driving force behind the development of the bicondylar angle, compelling a response from the bone with in the medial condyle to grow larger to reduce the moment at the knee (Lovejoy and Heiple 1973; Pauwels 1980; Lovejoy et al . 2002). Given the responsive nature of trabecular bone as discussed earlier, and the fact that both the amount and thickness of trab eculae as well as the density of cortical bone increase as children begin to w alk (Ryan and Kro vitz 2006), this theory has gained traction. Furthermore, medial condyles are demonstrably larger proximodistally than lateral condyles in humans (Lovejoy 20 05). However, it is not just the condyles that are asymmetrical, but the long bone axis itself. Tardieu and Preuschoft (1996) suggest that the bicondylar angle seen in orangutans (discussed below) is due only to medial condylar height, but that in humans t he uneven growth pattern

PAGE 23

17 takes place in the growth plate at the medial and lateral side of the metaphysis (Tardieu and Preuschoft 1996; Tardieu et al . 2006). She argues that compressive force on the medial condyle in ontogeny causes it to grow larger than the lateral condyle, resulting in uneven growth at the medial metaphysis growth plate. This uneven growth then translates to anterorposterior widening of the lateral growth plate, and an anterorposterior elongation of the lateral shaft. The resulting oval shape of the lateral condyle serves to stabilize the knee by providing a greater radius of curvature for the tibia to articulate with at full extension of the knee (Tardieu et al . 2006). She further postulates that the prominence of the lateral lip of the femoral trochlea in humans (while genetically determined now) initially formed to stabilize the patella after the development of the bicondylar angle in ontogeny, and eventually became genetically coded. That the bicondylar angle was not similarly assimila ted genetically has never been fully addressed, and no studies have yet been concluded on the lateral growth of the femoral shaft in children. Support for the distal condyle model is found in rates of medial compartment osteoarthritis in the knee joint, wh ich are extremely common (Kerrigan et al . 2002). It has been shown that compressive force in adulthood is most often experienced at the medial side of the knee joint at up to 2.5 times the compressive force experienced in the lateral side (Kerrigan et al . 2002). This medial com pression is widely considered the result of the varus torque experienced during walking (Kerrigan et al . 2002, Purcell 2013), and the varus knee displayed during early childhood would create compressive stress on the medial condyle . The higher the valgus in the knee, the more this compressive force moves to the lateral side of the knee joint which can be seen in the high rates of lateral compartment knee osteoarthritis in adults who possess a valgus knee deformity (Felson et al . 201 3). The Fossil Record While the earliest hominins show adaptations to bipedal locomotion, fossil discoveries over the past decade have revealed the mosaic nature of adaptations to this form of locomotion.

PAGE 24

18 While humans are obligate bipeds, all extant homin in s including gorillas, orangutans, chimps, and bonobos engage in forms of situational bipedalism (Kimura 1996; Videan and McGrew 2002; Thorpe and Crompton 2006). A comparative approach allows greater interpretation of the hominin fossil record with regar ds to the variety of morphological features that appear to be associated with bipedalism but also those that indicate different forms of bipedal progression in extinct hominins. The anatomy of extant apes and monkeys used for comparative analysis of homini n fossils is common practice (Cartmill and Milton 1977; Fleagle et al . 1981; Gebo 1996; Ruff 2002; Harcourt Smith et al . 2004; Locke et al . 2011). While imperfect in that extant species will never be an exact proxy for extinct species, comparative analysi s allows the identification of osteological traits associated with specific biomechanical strains incurred during various mobility strategies. Those traits can then be identified on fossil elements to indicate similar forms of mobility. For instance, argum ents on the mobility strategies of the last common ancestor (LCA) find support in the various locomotive strategies of modern primates. Richmond et al . (2001) summarizes the LCA hypotheses that have gained the most traction amongst paleoanthropologists in the last decades into four categories. Arboreal quadrupeds would have used above branch pronograde quadrupedalism like most extant anthropoids. Terrestrial quadrupeds would have walked on land quadrupedally using various hand positioning like digitigrady ( weight on the toe), palmigrady (weight on the flat palm), fist walking (weight on the back of the proximal phalanges), and knuckle walking (weight on the back of the middle phalanges). Finally, antipronograde and hylobatian climbers would have practiced an orthograde arboreal locomotion with significant fore and hindlimb flexibility. Among modern great apes, only the orangutan is fully arboreal (Locke et al. 2011). Eastern gorillas, while occasionally arboreal, employ mainly quadrupedal terrestrial knuckl e walking (Doran 1997) while lowland gorillas are much more combination arboreal and terrestrial (Remis 1995). Chimpanzees and bonobos engage in both suspensory arboreal and terrestrial

PAGE 25

19 quadruped knuckle walking behavior at different stages of ontogeny, of ten practicing more forelimb dominant arboreal orthogrady as infants and moving to hindlimb dominant quadrupedalism during adolescence and adulthood (Pontzer et al . 2014). This wide width of locomotive strategies practiced by our closest genetic relatives allows for in depth study of those strategies on bone architecture, and specifically in pelvic and femoral morphology. Orangutans are the only other great ape that display a bicondylar angle, around 6° (Tardieu and Preuschoft 1996). Tardieu and Preuschoft (1 996) argue that the development of this angle differs than that of humans in that it is solely the result of a superoinferior lengthening of the medial condyle rather than uneven growth in the femoral shaft. However, if the antipronograde and/or hylobatian hypotheses are considered and bipedalism arose from an arboreal context rather than a terrestrial one as some researchers have suggested (Senut 2006; Thorpe et al. 2007; Thorpe et al . 2014) it stands to reason that the ontogenetic formation of the bicondy lar angle in orangutans may be related to the biomechanical stress of their selective bipedality. No studies yet exist on the growth and formation of the bicondylar angle in orangutans. Orrorin tugenensis is the first femoral fossil we have from the homini n clade, and it dates to 5.7 6 mya (Richmond and Jungers 2008). The samples consist of three proximal femoral fragments, one of which still has an intact head. While no pelvic fossils currently exist for Orrorin tugenensis , the femoral fragments have sever al markers that suggest bipedality. According to Pickford et al . (2002) the neck is elongated and on par with both Australopithecus and modern humans over apes. Further, the femoral head is large compared to shaft diameter which is closer to modern humans than either Australopithecus or great apes, and the presence of the Obturator externus groove, where the Obturator externus muscle inserts and acts as a lateral rotator of the hip during bipedal movement in humans and which is not present in extant apes. Most re levant to this study, O. tugenensis displays a pronounced gluteal tuberosity suggestive of heightened gluteal activity and extension of the acetabulofemoral joint, which brings the thigh in

PAGE 26

20 line with the body during a bipedal stance . There is no mention of b icondylar angle in O. tugenensis , presumably due to the lack of a full shaft. Some researchers argue that O. tugenensis femoral morphology is actually closer to that of modern humans over Australopithecus , and therefore may be a closer relative to Homo and exclude Australopithecus as a direct relative (Pickford et al . 2002). Other researchers suggest that O. tugenensis morphology shows intermediate architecture indicative of a basal hominin directly related to both Australopithecus and Homo (Richmond and Ju ngers 2008). Still others propose that the mosaic of arboreal phalenx and bipedal femoral morphology support s the hypothesis of an arboreal origin to hominin bipedality, whenever that may have occurred (Senut 2006). Without the pelvic fossils the full locom otive strategies of O. tugenensis may never be known, but by the outlined biomechanical model the longer neck and larger femoral head suggest a bicondylar angle may have been present. Ardipithecus ramidus is temporally the next hominin fossil with both pel vic and femoral elements available in Homininae, dating to 4.4 mya (WoldeGabriel et al . 2009). Two proximal femora have been recovered, but with no head, neck, or greater trochanter and as with O. tugenensis there is no mention of a possible bicondylar ang le (Lovejoy et al . 2009). Lovejoy et a l. (2009) describes the pelvic fossils as already undergoing the repositioning of the gluteal muscles indicated by the shape and flare of the ilium as well as the appearance of the anterior inferior iliac spine, unique to hominins (see also Kozma et al . 2018). These features of the false pelvis in Ar. ramdus are often discussed as derived for bipedality, while the true pelvis of Ar. ramdus is described as being more archaic with a long ischial ramus seen in African apes and associated with hamstring usage in active climbing. Lovejoy et al . (2009) argues that this is indicative of a combined mobility strategy employed by Ar. ramdus . However, the mosaic of derived features in the upper pelvis and primitive features of the lower pelvis continues into Australopithecus and is not only related to locomotion. A ustralopithecines have several surviving femoral and pelvic fossils both whole and partial, and

PAGE 27

21 most follow a pattern of morphology that has been described as: large and l aterally flared illia, long ischium, extremely wide interacetabular widths, short femur lengths, large bicondylar angles, long femoral necks with lower neck shaft angles, small femoral heads, and an elliptical lateral condyle in the sagittal plane (Lovejoy et al . 1970; Berge 1994; Stern 2000; Lovejoy et al . 2002; Haile Selassie et al . 2010; DeSilva et al . 2013). Many of these features have been tied to some form of bipedality, yet they must also be considered in the context of parturition. Much of the obst etrical information about Australopithecus comes from two fossils: AL 288 1, an Au. afarensis also known as Lucy, which has a preserved innominate bone including a pubic ramus and the entire sacrum allowing a reconstruction of the key measurement of the pe lvic inlet (Berge et al . 1984) and Sts 14, a Au. africanus pelvis from Sterkfontein whose reconstruction has been a topic of contention due to extreme taphonomic compression (Berge and Goularas 2010). The classic form of the Australo pithecus pelvis, supported by these and other partial pelvic fossils, is platypelloid (although see Kibii et al . 2011 for Au. sediba pelvic dimensions, which may differ) (Tague and Lovejoy 1986; Häusler and Schmid 1995; Berge and Goularas 2010). This pelvi c shape presents with wide biacetabular widths and comparatively short femurs, creating extreme lateral placement of the proximal femur and large bicondylar angles in comparison with modern Homo (Fig. 5). The platypelloid pelvis and great interacetabular d istances have been argued to be the result of bipedality free of the restraints of birthing encephalized infants (Kibii et al . 2011). Ruff (1998; 2017) postulates that these proportions would have caused the longer neck shafts and a ustr alopithecines . His study suggests that this would have required they walk with more lateral deviation of their centers of gravity to reduce the joint reaction forces of the hips. However, Lovejoy (2005) has argued that longer neck lengths would have served to balance the joint reaction forces at the hip and allowed for a more human like bipedal gait. Warrener et al . (2015) conducted a study suggesting that interacetabular distance was not

PAGE 28

22 correlated with external torque which the hip abductors must oppose, and therefore hip width and abductor mechanics are not related in the way the current static model postulates. In response, Ruff (2017) argued that locomotor costs are not the only factor in determining pelvic and femoral architecture and more study was re quired. Early Homo is characterized by an overall increase in body size, but generally retains the mosaic morphology characteristic of a ustralopithecines including the mediolaterally wide pelvic shape (Churchill and Vansickle 2017). The Gona pelvis dated t o 0.9 1.4 mya and attributed to Homo erectus is the exception, as it is the only early Homo pelvic fossil available that displays the posterior/anterior widening of the pelvic midplane and inlet often associated with infant encephalization (Simpson et al . 2008). However, the Gona pelvis shows other morphological anomalies from Homo erectus , most importantly a small body size closer to that of Australopithecus , which has led some researchers to question its taxonomy and the selective pressures resultant in the oval pelvic shape (Ruff 2010). The fossil record consistently begins to display anterior/posterior pelvic widening during the middle Pleistocene in fossils attributed to Homo heidelbergensis and Homo neanderthalensis , although many remain mediolateral y wide (Rak and Arensberg 1987; Arsuaga et al . 1999; Rosenberg et al . 2006; Bonmatí et al . 2010). Modern human ratios were not dependability reached until the emergence of Homo sapien (Franciscus 2009). Complete pelvic fossils to the degree that dimensions can be obtained are relatively rare ( Bonmatí et al . 2010), so there remains a possibility that the anterior/posterior pelvic widening may have evolved earlier in Homo . This widening also correlates in the fossil record with a reduction in overall femoral bicondylar angle, which researchers argue lends support to the theory that the rounder pelvic inlets developed in the middle Pleistocene. Posterior/anterior widening in the pelvis results in smaller biacetabular distances, which in turn would theoreticall y produce smaller femoral bicondylar angles (Tardieu 2010). This reduction in angle is aided by the increase of body size seen in Homo , which theoretically

PAGE 29

23 lengthens the femur and reduces stress at the proximal and distal joints. These effects can be seen in the fossil femur of a hominin from the early Pleistocene in Dmanisi, Georgia designated early Homo . Although lacki ng pelvic elements, the fossils bicondylar angle falls within the accepted range of Australopithecus (Lordkipanidze et al . 2007). In contr ast we see the bicondylar angles of Neanderthals begin to approach those seen in modern human populations (Fig. 5), even given their shorter stature (Duarte et al . 1999; Arsuaga et al . 2007). Figue 5 Christine and Erik Trinkaus (1994) American Journal of Physical Anthropology . If biomechanical stress acting upon skeletal architecture is a driving factor behind bicondylar angle formation, the proposed anteroposterior expan sion and correlating reduction in bicondylar

PAGE 30

24 angle make sense and should be reflected in modern populations . However, if the bulk of the formation lies at the stress and architecture of the distal femur, this correlation may not have any causal relationshi p. If neither prove driving factors behind bicondylar angle, a deeper investigation of skeletal fossil elements and modern biomechanical strain and bone relationships may be needed in order to draw any conclusion about the mobility patterns of fossil homin ins. Hypotheses Two distinct theories dominate the literature on bicondylar angle formation . O ne is based on the dimensions at the pelvis and proximal femur and is often used in fossil analysis , and one is based on the dimensions of the distal femur to explain growth during ontogeny . These models suggest that b iomechanical strain actin g upon and constrained by these skeletal dimensions is the ultimate driver behind the formation and degree of bicondylar angle seen in adults. The proposed c orrelations have been widely accepted , but never statistically analyzed . This study aims to investigate the correlations between the skeletal relationships that have been proposed to dictate the degree of femoral bicondylar angle under the two distinct hypotheses of its formation. Neck shaft angle (NSA) will be lower a nd the bicondylar angle will be higher in individuals that display a wider biacetabular width. A wider width would increase the moment at the hip, causing the neck shaft angle to lower during ontogeny to stabilize the joint. This would in turn push the pro ximal femur out laterally and create a greater bicondylar angle. Height and area of the medial condyle will correlate with greater bicondylar angles by creating more asymmetry in the knee joint, increasing the adduction of the distal femur thereby increasi ng bicondylar angle. The biomechanical model s described above suggest the need to control for some variables, namely sex, femur length, and body mass. If the proximal hypothesis proves true, we should

PAGE 31

25 see in the data a negative correlation between NSA an d bicondylar angle and a positive correlation between biacetabular width and bicondylar angle. If the distal model proves true, we expect to see a positive correlation between medial condylar ridge height and area, and bicondylar angle.

PAGE 32

26 CHAPTER II MATERI ALS AND METHODS The data set for this study consists of 28 magnetic resonance images (MRIs) originally recreation al runner between the ages of twenty one and thirty three . The lowe r body of each was scanned from the fourth lumbar vertebrae to the mid metatarsals, and four overlapping sections were scanned isotropically at 1.7mm resolution from pelvis to feet then appended. The lower limbs were kept in anatomical orientation with a l eg b oard and dividers. Each subject s feet were positioned in dorsiflexion and immobilized using a footboard. MRIs were analyzed using the open source imaging software Image J with Fiji plugin version 2.0.0 rc 65/1.51s (Schindelin et al . 2012). Cropped an d appended images were inspected in coronal, sagittal, and transverse planes in order to record landmarks in 3D X (medial/lateral ), Y (vertical ), and Z (posterior/anterior) coordinates. Each plane was assessed independently, however the X, Y, Z coordinate grid was consistent throughout all three. All distances were calculated in 3D, however angles were obtained in both 3D and planar measurements in order to make them comparable to the existing literature. Distance measurements were calculated as: 2 2 2 using an online calculator (Furey 2018). Anatomical Measurements The measurements of the pelvis taken were biacetabular width, bicristal width, biishial diameter, and obstetrical conjugate. Biacetabular width was m easured in two ways: from one center of the femur head to the other (method of collection below ), and from the first showing of the bone table in the sagittal plane after moving through the femur head. Bicristal width was also measured in the sagittal plan e from the first showing of one iliac crest to the other. Bi ishial

PAGE 33

27 diameter was taken in the transverse plane from the first showing of each ischium moving cranially. The obstetrical conjugate was measured from the sacral promontory to the deepest showing of the pubic symphysis. The measurements of the femur taken for both right and left were femoral length, femoral head diameter, neck length, neck angle, bicondylar angle, height of the media l condyle, and area of the medial condyle . Femur length was measured from the first showing of the superior aspect of the femoral head in the transverse plane to the last showing of the inferior aspect of the medial condyle (McHenry and Corruccini 1978; Doyle and Windsor 2011). Femoral head diameter was found by drawing a best fit circle using the circle tool in all three (Athapattu et al . 2013). Figure 6. Best fit circle around the femoral hea d. Both neck length and neck angle were calculated by using the circle tool to find the centroid of the femoral head in all three planes, then averaging them to get the X, Y , and Z of the center of the femoral head. The line through the femoral shaft was drawn by finding the transverse slice at 20% and 80% of the shaft length, then finding the X, Y, and Z coordinates of each centroid through the shaft using the polygon tool. Finally, a line was drawn from edge to edge through the thinnest part of the neck in the coronal plane to find a center point through which to draw the line from the center of the femur head to the intersect at the line through the

PAGE 34

28 shaft. The length of this line is the true neck length and the angle at the neck shaft intersect is the ne ck angle. Mechanical neck length was the distance calculated from the first showing of the femoral head in the transverse plane to the first showing of the greater trochanter in the sagittal plane moving medially (Lovejoy 1973). Bicondylar angle was calcu lated by finding the X, Y, and Z coordinates of the first inferior showing of both the lateral and medial condyles in the transverse plane, then drawing a line from one point to the other in the coronal plane to represent the infracondylar plane (Fig. 7). Wher e the infracondylar plane and shaft line meet is the bicondylar intersect (BI), an d BI 90° = bicondylar angle (a dapted for MRI from Shefelbine et al . 2002). Figure 7. Plotted points and resulting lines through the neck shaft angle and bicondylar angl e. The height of the medial condyle, or condylar ridge, was calculated by taking the Y coordinate of the crest at the medial condyle in the coronal plane and calculating the distance from the Y of the inferior showing of the condyle. The area of the media l condyle was taken in

PAGE 35

29 the transverse plane using the polygon tool at the slice just before medial and lateral condyle join together (Fig. 8). Figure 8. Area of the medial condyle in the transverse plane. Table 1. Variable definitions and source materi al. Measurement Method Source Biacetabular Width (acetabulum) Distance between first showing of each acetabulum in the sagittal plane. Tague 1989 Biacetabular Width (femoral heads) Distance between the center of the femoral heads. Ruff 1995 Bicristal wi dth Distance between first showing of each iliac crests in the sagittal plane. Tague 1989 Bi ischial diameter Distance between first showing of each ischium in the transverse plane. Tague 1989 Obstetrical conjugate Distance between the sacral promontory and deepest point of pubic symphysis. Tague 1989 Femoral length Measured from the first showing of the femur head to the last showing of the medial condyle. McHenry and Corruccini 1978; Doyle and Windsor 2011

PAGE 36

30 . Femoral head diameter Avera ged out from best fit circle in all three planes and calculated from diameter. Athapattu et al . 2013 True femoral neck length Distance from the center of the femoral head to intersect of shaft. Michelotti and Clark 1999. Mechanical neck length Distance from the first showing of the greater trochanter in the sagittal plane to the first showing of the femoral head in transverse plane. Lovejoy 1973 Femoral neck angle Angle between center of the femoral head, neck shaft intersect, and infracondylar plane. M ichelotti and Clark 1999. Medial condylar height Distance from the medial condylar crest to the last showing in the transverse plane. A/P lateral and medial condyles Anterior to posterior measurement of the thickest part of the condyles in the sagittal plane. Bicondylar angle Angle between the last showing of the medial condyle, the infracondylar plane, and the neck/shaft intersect 90°. Adapted from Shefelbine et al . 2002 Analysis To test for normality of the data, histograms, skewness, and kurtos is were run, and a Shapiro Wilk test was conducted for all variables. Residuals scatter plots were built for each relationship to test for homoscedasticity. T tests were performed to compare left and right side measurements as well as 3D and 2D angle measu rements. tailed

PAGE 37

31 test for the control variable of sex were conducted against the independent and dependent variables to assess which were significa nt. Multiple regression analyses were then conducted on all model relationships with control variables when appropriate. All analyses were performed in IBM Corp. Released 2017. IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp. A sensitivity power analysis was conducted using G*Power Version 3.1 in order to establish the effect size (See F igure 1). Type I error was set to 0.05 and power was set to 0.8 error. Results Histograms showed some visual deviation from the classic bell curve, but all values for skewness were within the 2 and +2 cutoff for normal univariate distribution. All values for kurtosis also fell within this cut off except true neck length of the right femurs which returned a value of 2.031. The Shapiro Wilk test, shown to be the mos t powerful test for normality (Razali and Wah 2011), showed all values as failing to reject normal distributions over the 0.05 cut off except for the true neck length of the right femurs, which produced a result of .037. Residuals scatter plot s each display ed the random distribution suggestive of a linear relationship. alone (table 2). None showed any significant relationship except obstetrical conjugate with left bicondyla r angle ( r =0.386, p =0.042) and right bicondylar angle ( r =0.374, p =0.05). These values only barely reach the p =0.05 cut off for significance and show only weak correlation in r values of 0.3. In order to ensure no variables were distorting the results, cont rol variables needed to be identified and a multiple regression analysis conducted.

PAGE 38

32 Table 2 . Variables Tested r = Sig. p = Biacetabular Width (Acetabulum)/Left Bicondylar Angle 0.068 0.733 Biac etabular Width (Acetabulum)/Right Bicondylar Angle 0.174 0.376 Biacetabular Width (Femur Heads)/Left Bicondylar Angle 0.132 0.503 Biacetabular Width (Femur Heads)/Right Bicondylar Angle 0.079 0.689 Left Femoral Length/Left Bicondylar Angle 0.05 0.801 Right Femoral Length/Right Bicondylar Angle 0.074 0.707 Left Mechanical Neck Length/Left Bicondylar Angle 0.02 0.918 Right Mechanical Neck Length/Right Bicondylar Angle 0.262 0.178 Left Condyle Area /Left Bicondylar Angle 0.072 0.716 Right Condyle Area /Right Bicondylar Angle 0.069 0.727 Left Neck Angle/Left Bicondylar Angle 0.106 0.59 Right Neck Angle/Right Bicondylar Angle 0.25 0.199 Bicristal Distance/Left Bicondylar Angle 0.25 0.2 Bicristal Distance/Right Bicondylar Angle 0.109 0.581 Bitubero us Diameter/Left Bicondylar Angle 0.063 0.751 Bitubrous Diameter/Right Bicondylar Angle 0.217 0.267 Obstetrical Conjugate/Left Bicondylar Angle 0.386 0.042 Obstetrical Conjugate/Right Bicondylar Angle 0.374 0.05

PAGE 39

33 Paired sample t tests were conducted t o determine which measurements were statistically different from one another and if one side could serve as proxy for both. If the p values fell above the p = 0.05 significance level, the null hypothesis could not be rejected and the measurements were stat istically the same. There was a significant difference between right neck angle in 3D ( M =132.158, SD =4.476) and left neck angle in 3D ( M =136.141, SD =4.574 ); t =4.54, p =<0.001. There was also a significant difference between right neck angle in 2D ( M =134.13 2, SD =4.447) and left neck angle in 2D ( M =137.68, SD =4.279); t =4.271, p =<0.001. Further, there was a significant difference between left neck angle in 3D ( M =136.141, SD =4.574) and left neck angle in 2D ( M =137.68, SD =4.279); t = 6.489, p =<0.001 as well as ri ght angle in 3D ( M =132.158, SD =4.476) and right neck angle in 2D ( M =134.132, SD =4.447); t = 6.827, p =<0.001. These differences require that all variables tested against these measurements must be done in both 2D and 3D, on both the left and right side indiv idually. Left medial condylar ridge ( M =37.407, SD =3.381) and right medial condylar ridge ( M =36.376, SD =3.696) were also significantly different from one another; t =2.812, p =0.009 requiring they be tested independently. All other results showed no significa nt difference between both sides or 2D and 3D angles (table 3) and could therefore use one representative side or angle measurement. Table 3. Paired T test variable comparison results. Variable Pair M = SD = t = Sig. p = Right Bicondylar Angle 2D Left Bic ondylar Angle 2D 7.051 7.79 2.085 2.656 1.837 0.077 Left Bicondylar Angle 2D Left Bicondylar Angle 3D 7.79 7.606 2.656 2.445 1.823 0.079 Right Bicondylar Angle 2D Right Bicondylar Angle 3D 7.051 7.375 2.085 3.076 0.866 0.394

PAGE 40

34 . Left Mech anical Neck Length Right Mechanical Neck Length 80.927 82.243 7.555 8.069 1.024 0.315 Left True Neck Length Right True Neck Length 52.55 50.74 6.58 5.294 2.024 0.053 Left Neck Angle 3D Right Neck Angle 3D 136.141 132.158 4.574 4.476 4.54 <0.001 Left Ne ck Angle 2D Right Neck Angle 2D 137.68 134.132 4.279 4.447 4.271 <0.001 Left Neck Angle 2D Left Neck Angle 3D 137.68 136.141 4.279 4.574 6.489 <0.001 Right Neck Angle 2D Right Neck Angle 3D 134.132 132.158 4.447 4.476 6.827 <0.001 Left Femoral Length Right Femoral Length 458.673 458.682 4.575 4.682 0.015 0.988 Left Medial Condylar Ridge Right Medial Condylar Ridge 37.405 36.376 3.381 3.696 2.812 0.009 Area Left Medial Condyle Area Right Medial Condyle 1,1119.95 1,091.67 179.085 193.053 1.531 0.138 significantly correlated with mechanical neck length ( r =0.55, p =0.002), right neck angle in 3D ( r = 0.513, p =0.005) and 2D ( r = 0.404, p =0.033), femur length ( r =0.525, p =0.004 ), left medial condylar ridge ( r =0.406, p =0.032) and right medial condylar ridge ( r =0.404, p =0.032), and area

PAGE 41

35 of the medial condyle ( r =0.711, p (table 5) showed it was significantly correlated w ith bicristal distance ( r =0.57, p =0.002), true neck length ( r =0.402, p =0.035) and mechanical neck length ( r =0.637, p =<0.001), left medial condylar ridge ( r =0.451, p =0.016) and right medial condylar ridge ( r =0.385, p =0.043), and area of the medial condyle ( r =0.654, p test for the control variable of sex (table 6) showed it was significantly correlated with biacetabular width from the acetabulum ( t = 2.121, p =0.044), bituberous diameter ( t = 3.662, p =0.001), mechanical neck length ( t =4.005, p =<0.001), femur length ( t =3.292, p =0.003), left medial condylar ridge ( t =3.213, p =0.003) and right medial condylar ridge ( t =2.862, p =0.003), and area of the medial condyle ( t =3.287, p =0.003). All significantly correlated control variables were then intro duced to the multiple regression analysis when their correlated variables were tested. Table 4. Variable Body Mass/ r = Sig. p = Biacetabular Width (acetabulum) 0.258 0.185 Biacetabular Width (femur heads) 0.1 21 0.539 True Neck Length 0.113 0.567 Mechanical Neck Length 0.55 0.002 Left Neck Angle 3D 0.009 0.965 Right Neck Angle 3D 0.513 0.005 Femoral Length 0.525 0.004 Left Medial Condylar Ridge 0.406 0.032 Right Medial Condylar Ridge 0.404 0.033 Area Medial Condyle 0.711 <0.001

PAGE 42

36 Left Neck Angle 2D 0.063 0.748 Right Neck Angle 2D 0.404 0.033 Bicondylar Angle 0.124 0.529 Bituberous Diameter 0.23 0.24 Obstetrical Conjugate 0.175 0.372 Bicristal Distance 0.373 0.051 Table 5. Pea Variable Femur Length/ r = Sig. p = Biacetabular Width (from acetabulum) 0.334 0.082 Biacetabular Width (from femur heads) 0.201 0.304 True Neck Length 0.401 0.035 Mechanical Neck Length 0.637 <0.001 Left Neck Angle 3D 0.112 0.572 Right Neck Angle 3D 0.228 0.242 Left Medial Condylar Ridge 0.451 0.016 Right Medial Condylar Ridge 0.385 0.043 Area Medial Condyle 0.654 <0.001 Left Neck Angle 2D 0.14 0.478 Right Neck Angle 2D 0.15 0.446

PAGE 43

37 d. Bicondylar Angle 0.05 0.801 Bituberous Diameter 0.294 0.129 Obstetrical Conjugate 0.063 0.751 Bicristal Distance 0.57 0.002 Table 6. test for control variable sex. Variables M = SD = t = Sig. p = Biacetabular Width (acetabulum) Male Female 119.396 124.606 6.478 6.489 2.121 0.044 Biacetabular Width (femur heads) Male Female 175.155 176.884 4.614 6.235 0.841 0.408 True Neck Length Male Female 52.185 52.971 6.887 6.461 0.31 0.759 Mechanical Neck Length Male Female 85.193 76.004 6 .201 5.879 4.005 <0.001 Left Neck Angle 3D Male Female 135.051 137.399 4.646 4.322 1.377 0.18 Right Neck Angle 3D Male Female 131.306 133.141 5.243 3.324 1.086 0.288 Left Neck Angle 2D Male Female 136.739 138.766 4.36 4.079 1.264 0.218 Right Neck A ngle 2D Male Female 133.676 134.658 5.114 3.665 0.576 0.57

PAGE 44

38 . Femoral Length Male Female 470.675 444.825 23.102 17.55 3.292 0.003 Left Medial Condylar Ridge Male Female 39.053 35.504 3.588 1.842 3.213 0.003 Right Medial Condylar Ridge Male Female 38.03 34.469 3.843 2.472 2.862 0.008 Area Medial Condyle Male Female 1217.79 1022.11 146.029 157.322 3.287 0.003 Bicondylar Angle Male Female 7.672 7.925 3.195 1.981 0.247 0.807 Bituberous Diameter Male Female 90.235 107.107 11.328 13.06 3.662 0.001 Bicristal Distance Male Female 266.195 263.433 16.318 13.116 0.488 0.629 Obstetrical Conjugate Male Female 129.054 134.809 10.139 10.275 1.489 0.149 Once control variables were identified for each relationship, an individual multiple regressio n analysis was run on each proposed relationship (table 7, table 8). For the proximal model, two measurements of biacetabular width, true and mechanical neck length, and both 2D and 3D neck shaft angle were tested against bicondylar angle. In the distal mo del, medial condylar height and area were tested against bicondylar angle. None of the multiple regressions showed a significant relationship between any variables.

PAGE 45

39 Table 7. Multiple Regression Analysis: Proximal Model. Length measurements in mm, mass in kg, angles in degrees. Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 6.358 24.597 0.258 0.779 Biacetabular Width (acetabulum) 0.029 0.091 0.074 0.313 0.757 Sex 0.328 1.564 0.063 0.21 0.836 Left Femoral Length 0.018 0.032 0.161 0.547 0.59 Body Mass 0.06 0.074 0.218 0.817 0.423 Left Mechanical Neck Length 0.037 0.114 0.104 0.322 0.751 Left Neck Angle 3D 0.023 0.132 0.04 0.175 0.863 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 6 21 0.174 0.047 0.981 Left Bicondyla r Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 5.577 25.419 0.219 0.828 Biacetabular Width (acetabulum) 0.000 0.09 0.001 0.005 0.996 Sex 0.318 1.445 0.061 0.22 0.828 Left Femoral Length 0.002 0.035 0.016 0.052 0.959 Body Mass 0 .05 0.07 0.179 0.712 0.484 Left True Neck Length 0.128 0.102 0.317 1.257 0.223 Left Neck Angle 3D 0.001 0.127 0.002 0.008 0.993 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 6 21 0.431 0.11 0.85 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 1.347 27.192 0.05 0.961 Biacetabular Width (acetabulum) 6.573 0.09 0.00 0.001 0.999 Sex 0.355 1.43 0.68 0.249 0.806 Left Femoral Length 0.001 0.035 0.006 0.019 0.985 Body Mass 0.051 0.069 0.183 0.733 0.472 Left True Neck Length 0.123 0.103 0.305 1.196 0.245

PAGE 46

40 Left Neck Angle 2D 0.031 0.136 0.049 0.226 0.824 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 6 21 0.441 0.112 0.843 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 11.1 25.955 0.428 0.673 Biacetabular Width (acetabulum) 0.027 0.091 0.071 0.301 0.766 Sex 0.234 1.545 0.045 0.152 0.881 Left Femoral Length 0.019 0.032 0.172 0.582 0.567 Body Mass 0.059 0.074 0.215 0.809 0.428 Left Mechanical Neck Length 0.029 0.115 0.082 0.252 0.803 Left Neck Angle 2D 0.06 0.14 0.096 0.426 0.675 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 6 21 0.2 0.054 0.973 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 4.901 24.829 0.197 0.845 Biacetabul ar Width (femur heads) 0.014 0.108 0.028 0.127 0.9 Left Neck Angle 3D 0.014 0.121 0.025 0.118 0.907 Left True Neck Length 0.127 0.096 0.315 1.325 0.198 Left Femoral Length 0.009 0.024 0.085 0.379 0.708 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 4 23 0.559 0.089 0.695 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 0.003 26.919 0.000 1.00 Biacetabular Width (femur heads) 0.012 0.107 0.023 0.107 0.915 Left Neck Angle 2D 0.021 0.13 0.034 0.164 0.871 Left True Ne ck Length 0.121 0.097 0.299 1.244 0.226 Left Femoral Length 0.008 0.025 0.07 0.311 0.759 ANOVA and Model Summary df1 df2 F R 2 Sig. p =

PAGE 47

41 . 4 23 0.562 0.089 0.692 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant ) 15.27 26.552 0.575 0.571 Biacetabular Width (femur heads) 0.067 0.116 0.136 0.581 0.567 Left Femoral Length 0.011 0.034 0.102 0.327 0.747 Left Neck Angle 2D 0.053 0.14 0.086 0.381 0.707 Left Mechanical Neck Length 0.038 0.115 0.107 0.327 0.747 Se x 0.099 1.559 0.019 0.063 0.95 Body Mass 0.066 0.074 0.239 0.895 0.381 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 6 21 0.244 0.065 0.957 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 10.396 25.048 0.415 0.682 Bi acetabular Width (femur heads) 0.07 0.117 0.142 0.601 0.555 Left Femoral Length 0.01 0.034 0.088 0.283 0.78 Left Neck Angle 3D 0.014 0.133 0.023 0.103 0.919 Left Mechanical Neck Length 0.046 0.115 0.131 0.402 0.691 Sex 0.201 1.573 0.038 0.128 0.9 Body Mass 0.067 0.074 0.242 0.903 0.377 Table 8. Multiple Regression Analysis: Distal Model. Length measurements in mm, mass in kg, angles in degrees. Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 4.461 14.281 0.312 0.758 Left Medial Condylar Ridge 0.107 0.196 0.136 0.544 0.592 Body Mass 0.048 0.069 0.174 0.698 0.492 Sex 0.053 1.391 0.01 0.038 0.97 Left Femoral Length 0.023 0.029 0.208 0.786 0.44

PAGE 48

42 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 4 23 0.293 0.048 0.88 Left Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 2.678 14.214 0.188 0.852 Area Left Medial Condyle 0.005 0.005 0.329 0.929 0.363 Body Mass 0.099 0.083 0.356 1.191 0.247 Sex 0.523 1.383 0.098 0.378 0.709 Left Femoral Length 0.012 0.033 0.101 0.356 0.725 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 4 21 0.455 0.08 0.768 Right Bicondylar Angle Coefficients B Std. Error Beta t Sig. p = (Constant) 3.844 10.512 0.366 0.718 Right Medial Condylar Ridge 0.058 0.13 0.103 0.446 0.659 Body Mass 0.081 0.052 0.371 1.55 0.135 Sex 0.138 1.033 0.034 0.134 0.895 Right Femoral Length 0.024 0.021 0.283 1.148 0.263 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 4 23 0.84 0.128 0.514 Right Bicondylar Ang le Coefficients B Std. Error Beta t Sig. p = (Constant) 0.279 10.352 0.027 0.979 Area Right Medial Condyle 0.002 0.004 0.198 0.588 0.563 Body Mass 0.11 0.065 0.504 1.699 0.104 Sex 0.14 1.055 0.033 0.133 0.895 Right Femoral Length 0.025 0.023 0.28 1.092 0.287 ANOVA and Model Summary df1 df2 F R 2 Sig. p = 4 21 0.938 0.152 0.461

PAGE 49

43 CHAPTER III DISCUSSION T he assumption that genetically determined elements of skeletal structure combined with mobility result s in the degree of bicondylar angle have long been accepted as cause and effect. While Tardieu's work (1996; 2010) has proven that mobility is a driving fa ctor in the formation of the bicondylar angle, this study indicates that the architectural elements long assumed to be a restraining parameter may in fact have less of an impact than previously presumed. In the proximal hypothesis of this study, the most discussed relationship in the context of the fossil record has been that of biacetabular width and bicondylar angle, often as it pertains to a ustralopithecines (Lovejoy et al . 2002). Two measurements for biacetabular width were analyzed , both from the cen ter of the femur heads and from the first showing of the bone table in the acetabulum. Using both measurement s , the scatter plots below (Fig. 9 and Fig. 10) k of a clear linear relationship between these two elements suggests that wider hips may not be able to account for the extreme bicondylar angles found in fossil hominin populations, and instead they may only be explicable through questions of mobility str ategies, i.e. an altered form of bipedal gait, or a combination of bipedality and some other form of locomotion.

PAGE 50

44 Figure 9 : Biacetabular width as taken from the bone table of the acetabulum, and left bicondylar angle (left) and right bicondylar angle (right). Measurements in mm, an gles in degrees. Figure 10 : Biacetabular width as taken from the center of the femur heads, and left bicondylar angle (left) and right bicondylar angle (right). Measure ments in mm, angles in degrees. The condylar ridge and medial condyle area also showed a clear lack of relationship on biomechanical stress in the condyles is insufficient to explain the role of condylar involvement in bicondylar angle formation in its entirety. Presumably plasticity at the condyles is a driving force behind the degree of angle, but so too might the str ess in the shaft itself at the growth plate play a larger than anticipated role if each of these major theories fail to produce significance in the measurements of this study.

PAGE 51

45 Figure 11: Condylar ridge and left bicondylar a ngle (left) and right bicondylar angle (right). Measurements in mm, angles in degrees. Figure 12 . Medial condyle area and left bicondylar angle (left) and right bicondylar area (right). Measure ments in mm, angles in degr ees. Many researchers h ave suggested that Australopithecus engaged in a vastly differ ent form of bipedality than is practiced by modern humans, including the possibility of an obligate upright bent hip, bent knee gait similar to that occasionally implemented by modern gorillas and chimpanzees (Stern and Susman 1983; Kullmer et a l. 2003; So ckol et al . 2007; Raichlan et al . 2010; Zipfel et al . 2011). If true, the development of a femur exposed to disparate biomechanical pressures during ontogeny than seen in modern populations cannot be predicted, and may result in extreme bicondylar angles d ue to some differing form of biomechanical stress at play in the architecture. However, the significant results between bicondylar angle and obstetrical conjugate do suggest a relationship between pelvic dimension and bicondylar angle formation. Althoug h

PAGE 52

46 these results are only barely within the significance cut off (p = 0.042 and p = 0.05), they suggest that pelvic morphology correlate with degree of bicondylar angle through an unexpected pelvic dimension. The results show that expansion of the pelvis in the posterioranterior axis increase the degree of bicondylar angle largely determined in the mediolateral axis. This is the opposite of previous assumptions that this posterioanterior pelvic expansion in Homo led to smaller biacetabular widths and lower b icondylar angles in the fossil record. O ne possibility is that posterioanterio r expansion of the pelvis would also move the acetabulum anterior, creating a greater degree of anteversion at the femoral head. In order to retain the ~ 16° anteversion requir ed to avoid hip dysplasia (Sugano et al . 1998) and shaft fractures ( Bråten et al . 1993), the trochanter of the femur would be forced outward laterally, which could result in a greater degree of bicondylar angle. It is also possible that this skeletal eleme nt is associated with some other pelvic measurement not taken for this study that would increase bicondylar angle. A third explanation for the correlation between bicondylar angle and obstetrical conjugate may lie in the method of data collection. Angles for this study were calculated originally in 3D coordinates using X, Y, and Z points, and were then recalculated in the coronal plane in order to make them comparable to existing literature (Lovejoy and Heiple 1970; Tardieu and Damsin 1997; Shefelbine et a l . 2002). This recalculation did shift the ranges by several degrees (i.e. the largest bicondylar angle went from 19.7 to 13.44). While this recalculation from n throughout the literature may change if measurement s were taken in 3D coordinates. Furthermore, the effect size as calculated by the power analysis was 0.37, indicating that that power of this study was small. Further research with a larger dataset may y ield different results. However, the most likely inference is that the static model is inadequate to predict biomechanical loading of the hip and femur, and therefore to explain femoral bicondylar angle

PAGE 53

47 development. The results of this study suggest that s keletal architecture is not a driving force behind the degree of bicondylar angle seen in adults. Therefore, a dynamic model of biomechanical forces and a deeper understanding of osteological reaction to those forces is needed if we are to understand how mo bility is reflected in bone. Conclusion The presence of a bicondylar angle is a vital indicator of bipedality throughout the fossil record, and yet the process by which it develops remains undetermined. Without a thorough understanding of how and why this angle forms, the modes of bipedality practiced by our ancestors in comparison to our own can never be fully ascertained. The fossil record is incomplete, and the equivalence of osteological elements to various kinds of mobility requires an intimate unders tanding of the formation of those elements during ontogeny. The current model of bicondylar angle formation rests on a static view of the interplay within the pelvic femoral complex. The results of this study suggest that it has insufficient explanatory po wer to determine the mobility strategies practiced by our hominin ancestors or by current human populations. Understanding the biomechanical stresses that result in the formation of the bicondylar angle requires a more complex model that integrates a dynam ic approach to mobility, as well as a deeper understanding of the forces at work in ontogeny and throughout adulthood. If skeletal architecture plus Newtonian forces alone cannot explain bicondylar angle formation, as is suggested by the results of this st udy, a more composite and integrative model of formation is required to understand the circumstances that lead to a bicondylar angle.

PAGE 54

48 References Aiello, Leslie, and Christopher Dean. 1990. An introduction to human evolutionary anatomy . Academic Pres s. Al Imam, A. 2017. "Anatomical Adaptations for Obligate Bipedal Locomotion in Humans." International Journal of Anatomy and Physiology 3 (1): 63 68. Anderson, John Y., and Erik Trinkaus. 1998. "Patterns of sexual, bilateral and interpopulational variat ion in human femoral neck shaft angles." The Journal of Anatomy 192 (2): 279 285. Arsuaga, Juan Luis, Carlos Lorenzo, José Miguel Carretero, Ana Gracia, Ignacio Martínez, Nuria García, José María Bermúdez de Castro, and Eudald Carbonell. 1999. "A complete human pelvis from the Middle Pleistocene of Spain." Nature 399 (6733): 255. Arsuaga, J. L., V. Villaverde, R. Quam, I. Martínez, J. M. Carretero, C. Lorenzo, and A. Gracia. 2007. "New Neandertal remains from Cova Negra (Valencia, Spain)." Journal of Huma n Evolution 52 (1): 31 58. Athapattu, Meuru, Amir Hossein Saveh, Jamaluddin Mahmud, Mohammad Alrashidi, Bin Wang, and Mahmoud Chizari. 2013. "Accuracy of Measuring Methods on the Femoral Head." Procedia Engineering 68: 83 87. Barak, M. M., D. E. Lieberma Bone 49 (6): 1141 1151. Baum, T., J. Carballido Gamio, MB Huber, D. Muller, R. Monetti, C. Rath, F. Eckstein, et al. 2 prediction of Osteoporosis International 21 (9): 1553 64. Berge, Christine, Rosine Orban Segebarth, and Peter Schmid. 1984. "Obstetrical int erpretation of the australopithecine pelvic cavity." Journal of Human Evolution 13 (7): 573 587. Berge, Christine. 1994. "How did the australopithecines walk? A biomechanical study of the hip and thigh of Australopithecus afarensis ." Journal of Human Evol ution 26 (4): 259 273.

PAGE 55

49 Berge, Christine, and Dionysis Goularas. 2010. "A new reconstruction of Sts 14 pelvis ( Australopithecus africanus ) from computed tomography and three dimensional modeling techniques." Journal of Human Evolution 58 (3): 262 272. Bon Journal of Bone and Mineral Research 26 (2): 229 238. Osteoporosis International 25 (S3). Bonmatí, Alejandro, Asier Gómez Olive ncia, Juan Luis Arsuaga, José Miguel Carretero, Ana Gracia, Ignacio Martínez, Carlos Lorenzo, José María Bérmudez de Castro, and Eudald Carbonell. 2010. "Middle Pleistocene lower back and pelvis from an aged human individual from the Sima de los Huesos sit e, Spain." Proceedings of the National Academy of Sciences 107 (43): 18386 18391. Bonneau, Noémie, Paul Antoine Libourel, Caroline Simonis, Laurent Puymerail, Michel Baylac, Christine Tardieu, and Olivier Gagey. 2012. "A three dimensional axis for the study of femoral neck orientation." Journal of Anatomy 221 (5): 465 476. Bousson, Val erie, Anthony Bras, F. Roqueplan, Yan Kang, David Mitton, Sami Kolta, C. Bergot, quantitative computed tomography of the proximal femur: Relationships linking geometri c and Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA. 17: 85 5 64. The Journal of Bone and Joint Surgery. British volume. 75 (5):799 803. Bronner, F., M. C. Farach Carson, H. I. Roach, & SpringerLink (Online service). 2010. Bone and Development. London: Springer Verlag. Bulagouda, R. S., S. D. Desai, D. G. Gannur, and U. C. Nuchhi. 2014. "A Study on the Neck Shaft Angle of Femur in 10 0 Macerated Bones." Indian Journal of Forensic Medicine & Toxicology 8 (1): 116.

PAGE 56

50 Archives of Biochemistry and Biophysics 561 (3 12). Cartmill, Matt, and Katha rine Milton. 1977. "The lorisiform wrist joint and the evolution of American Journal of Physical Anthropology 47 (2): 249 272. Carvalho, Susana, Dora Biro, Eugénia Cunha, Kimberley Hockings, William C. McGrew, Brian G. Richmond, and Tetsuro Matsuzawa. 2012. "Chimpanzee carrying behaviour and the origins of human bipedality." Current Biology 22 (6): R180 R181. research on Wol Journal of Biomechanics 5 (2): 173 180. Churchill, Steven Emilio, and Caroline Vansickle. 2017. "Pelvic morphology in Homo erectus and early Homo." The Anatomical Record 300 (5): 964 977. Cohen, Jacob. 1988. "Statistical power analysis for the behavioral sciences. 2nd." Coleman, William H. 1969. "Sex differences in the growth of the human bony pelvis." American Journal of Physical Anthropology 31 (2): 125 151. DeSilva, Jeremy M., Kenneth G. Holt, Steven E. Churchill, Kristian J. Carlson, Chri stopher S. Walker, Bernhard Zipfel, and Lee R. Berger. 2013. "The lower limb and mechanics of walking in Australopithecus sediba ." Science 340 (6129): 1232999. Doran, D. M. 1997. "Ontogeny of locomotion in mountain gorillas and chimpanzees." Journal of Hu man Evolution 32 (4): 323 344. Doyle, Anthony J., and Sandra Winsor. 2011. "Magnetic resonance imaging (MRI) lower limb length measurement." Journal of Medical Imaging and Radiation Oncology 55 (2): 191 194. Duarte, Cidália, João Maurício, Paul B. Pettit t, Pedro Souto, Erik Trinkaus, Hans Van Der Plicht, and João Zilhão. 1999. "The early Upper Paleolithic human skeleton from the Abrigo do Lagar Velho (Portugal) and modern human emergence in Iberia." Proceedings of the National Academy of Sciences 96 (13): 7604 7609.

PAGE 57

51 Duda, Georg N., Erich Schneider, and Edmund YS Chao. 1997. "Internal forces and moments in the femur during walking." Journal of Biomechanics 30 (9): 933 941. Dugan, Sheila A., and Krishna P. Bhat. 2005. "Biomechanics and analysis of running gait." Physical Medicine and Rehabilitation Clinics 16 (3): 603 621. Duncan, R. L., and C. H. Turner. 1995. "Mechanotransduction and the functional response of bone to mechanical strain." Calcified Tissue International 57 (5): 344 358. Ehrlich, P. J., an Osteoporosis International 13 (9): 688 700. Felson, David T., Jingbo Niu, K. Douglas Gross, Martin Englund, Leena Sharma, T. Derek V. Cooke, Ali Guermazi et al. 2013. "Valgus mala lignment is a risk factor for lateral knee osteoarthritis incidence and progression: findings from the Multicenter Osteoarthritis Study and the Osteoarthritis Initiative." Arthritis & Rheumatism 65 (2): 355 362. Fleagle, John G., Jack T. Stern, William L. Jungers, Randall L. Susman, Andrea K. Vangor, and James P. Wells. 1981. "Climbing: a biomechanical link with brachiation and with bipedalism." Symposia of the Zoological Society of London 48: 359 375. Franciscus, Robert G. 2009."When did the modern human pattern of childbirth arise? New insights from an old Neandertal pelvis." Proceedings of the National Academy of Sciences 106 (23): 9125 9126. Furey, Edward. 2017. " 3D Distance Calculator "; from https://www.calculatorsoup.com Online Calculator Resource. Gebo, Daniel L. 1996. "Climbing, brachiation, and terrestrial quadrupedalism: historical precursors of hominid bipedalism." American Journal of Physical Anthropology 101 (1): 55 92. Femoral neck shaft Journal of Anatomy 223 (2): 133 51.

PAGE 58

52 Gonzalez, Paula N., Valeria Bernal, and S. Ivan Perez. 2009. "Geometric morphometric approach to sex estimation of human pelvis." Forensic Science International 189 (1 3): 68 74. Gruss, Laura Tobias, and Daniel Schmitt. 2015. "The evolution of the human pelvis: changing adaptations to bipedalism, obstetrics and thermoregulation." Philosophical Transactions of the Royal Society 370 (1663): 20140063. Irish Journal of Medical Science 184 (1): 13 22. doi:10.1007/s11845 014 1070 y Irish Journal of Medical Science (1971 ) 186 (4): 939 946. Haile Selassie, Yohannes, Bruce M. Latimer, Mulugeta Alene, Alan L. Deino, Luis Gibert, Stephanie M. Melillo, Beverly Z. Saylor, Gary R. Scott, and C. Owen Lovejoy. 2010. "An early Australopithecus afarensis postcranium from Woranso M ille, Ethiopia." Proceedings of the National Academy of Sciences 107 (27): 12121 12126. Harcourt Journal of Anatomy 204 (5): 403 416. Häusler, Martin, and Peter Schmid. 1995. "Comparison of the pelves of Sts 14 and AL288 1: implications for b irth and sexual dimorphism in australopithecines." Journal of Human Evolution 29 (4): 363 383. Henriksson, Lars. 1980. "Measurement of femoral neck anteversion and inclination: a radiographic study in children." Acta Orthopaedica Scandinavica 51 (186): 1 59. Journal of Human Evolution 29: 141 153. Holzer, Gerold, Gobert von Skrbensky, Lukas A. Holzer, and Wolfgang Pichl. 20 09. "Hip fractures and the contribution of cortical versus trabecular bone to femoral neck strength." Journal of Bone and Mineral Research 24 (3): 468 474. Houston, C. Stuart, and Witold A. Zaleski. 1967. "The shape of vertebral bodies and femoral necks i n relation to activity." Radiology 89 (1): 59 66.

PAGE 59

53 Igbigbi, P. S., and M. Sharrif. 2005. "The bicondylar angle of adult Malawians." American Journal of Orthopedics 34 (6): 291 294. trabecular architecture Journal of Biomechanics 41 (11): 2353 2361. Javois, C., C. Tardieu, B. Lebel, R. Seil, C. Hulet, Soc Francaise Arthroscopie, and Société parative anatomy of the knee joint: Effects on the lateral Orthopaedics & Traumatology: Surgery & Research 95 (8): 49 59. Kay, Robert M., Kai A. Jaki, and David L. Skaggs. 2000. "The effect of femoral rotation on the projected femoral neck shaf t angle." Journal of Pediatric Orthopaedics 20 (6): 736 739. Kerrigan, D. Casey, Jennifer L. Lelas, Joyce Goggins, Greg J. Merriman, Robert J. Kaplan, and David T. Felson. 2002. "Effectiveness of a lateral wedge insole on knee varus torque in patients wit h knee osteoarthritis." Archives of Physical Medicine and Rehabilitation 83 (7): 889 893. Kibii, Job M., Steven E. Churchill, Peter Schmid, Kristian J. Carlson, Nichelle D. Reed, Darryl J. De Ruiter, and Lee R. Berger. 2011. "A partial pelvis of Australop ithecus sediba ." Science 333 (6048): 1407 1411. Kimura, Tasuku. 1996. "Centre of gravity of the body during the ontogeny of chimpanzee bipedal walking." Folia Primatologica 66 (1 4): 126 136. Kozma, Elaine E., Nicole M. Webb, William EH Harcourt Smith, D avid A. Raichlen, Kristiaan D'Août, Mary H. Brown, Emma M. Finestone, Stephen R. Ross, Peter Aerts, and Herman Pontzer. 2018. "Hip extensor mechanics and the evolution of walking and climbing capabilities in humans, apes, and fossil hominins." Proceedings of the National Academy of Sciences 201715120. Nature 423 (6937): 332. image analysis of ape, hominid and human foot Courier Forsch Senkenberg 243: 85 91.

PAGE 60

54 Lang, T. F., J. H. Keyak, M. W. Heitz, P. Augat, Y. Lu, A. Mathur, and H. K. Genant. 1997. "Volumetric quantitative computed tomography of the proximal femur: precision and relation to bone strength." Bone 21 (1): 101 108. Levin, S. A. M. 1997. "10. A different approach to the mechanics of the human pelvis: tensegrity." In Movement, Stability, and Lower Back Pain , edited by Vleeming at al, 157 167. Churchill Livingstone, London. Levinger, P., H. B. Menz, A. D. Morrow, J. R. Bartlett, J. A. Feller, and N. R. Bergman. 2013. Journal of Foot and Ankle Research 6 (1): 33. Locke, D. P., L. W. H Nature 469: 529 533. Lordkipanidze, David, Tea Jashashvili, Abesalom Vekua, Marcia S. Ponce de León, Christoph PE Zollikofer, G. Philip Rightmire, Herman Pon tzer et al. 2007. "Postcranial evidence from early Homo from Dmanisi, Georgia." Nature 449 (7160): 305. Lovejoy, C. Owen, and Kingsbury G. Heiple. 1970. "A reconstruction of the femur of Australopithecus africanus ." American Journal of Physical Anthropolo gy 32 (1): 33 40. Lovejoy, C. Owen, Kingsbury G. Heiple, and Albert H. Burstein. 1973. "The gait of Australopithecus ." American Journal of Physical Anthropology 38 (3): 757 779. Lovejoy, C. Owen. 1988. "Evolution of human walking." Scientific American 25 9 (5): 118 125 . Lovejoy, C. Owen, Richard S. Meindl, James C. Ohman, Kingsbury G. Heiple, and Tim D. White. 2002. "The Maka femur and its bearing on the antiquity of human walking: applying contemporary concepts of morphogenesis to the human fossil record ." American Journal of Physical Anthropology 119 (2): 97 133. Lovejoy, C. Owen. 2005. "The natural history of human gait and posture: part 2. Hip and thigh." Gait & Posture 21 (1): 113 124. Lovejoy, C. Owen. 2007. "The natural history of human gait and p osture: Part 3. The knee." Gait & Posture 25 (3): 325 341.

PAGE 61

55 Lovejoy, C. Owen, Gen Suwa, Scott W. Simpson, Jay H. Matternes, and Tim D. White. 2009. "The great divides: Ardipithecus ramidus reveals the postcrania of our last common ancestors with African ap es." Science 326 (5949): 73 106. Journal of Biochemistry and Cell Biology 40(1): 46 62. Marzke, Mary W., Julie M. Longhill, and Stanley A. Rasmussen.1988. "Gluteus maximus muscle function and the origin of hominid bipedality." American Journal of Physical Anthropology 77 (4): 519 528. McHenry, Henry M., and Robert S. Corrucci ni. 1978. "The femur in early human evolution." American Journal of Physical Anthropology 49 (4): 473 487. models to understanding the mechanisms of adaptation t Frontiers in Endocrinology 5. Metzger, Thomas A., Stephen A. Schwaner, Anthony J. LaNeve, Tyler C. Kreipke, and Glen L. Journal o f Biomechanics 48 (12): 3035 43. Journal of Bone and Mineral Research 14 (10): 1714 1720. Mirza, Saqeb B., Douglas G. Dunlop, Sukhmeet S. Panesar, Syed G. Naqvi, Shafat Gang oo, and Saif Salih. 2010. "Basic science considerations in primary total hip replacement arthroplasty." The Open Orthopaedics Journal 4: 169. Journal of Orthop aedic Research 13 (4): 503 512. Nawathe, S., B. P. Nguyen, N. Barzanian, H. Akhlaghpour, M. L. Bouxsein, and T. M. Keaveny. Journal of Biomechanics 48 (5): 816 822.

PAGE 62

56 Nordström, P., Pe strength, and pubertal stage as determinants of bone mineral density and bone area in Journal of Bone and Mineral Research 13 (7): 1141 1148. Pauwels, Friedrich. 19 80. Biomechanics of the locomotor apparatus: contributions on the functional anatomy of the locomotor apparatus . Berlin: Springer Verlag. Pandya, A. M., T. C. Singel, M. M. Patel, and D. V. Gohil. 2008. "A study of the femoral bicondylar angle in the Guja rat region." J ournal of the Anat omical Soc iety of India 57 (2 ): 131 134. Pickford, Martin, Brigitte Senut, Dominique Gommery, and Jacques Treil. 2002. "Bipedalism in Orrorin tugenensis revealed by its femora." Comptes Rendus Palevol 1 (4): 191 203. Pontz er, Herman, David A. Raichlen, and Peter S. Rodman. 2014. "Bipedal and quadrupedal locomotion in chimpanzees." Journal of Human Evolution 66: 64 82. Porter, A. M. W. 1999. "Modern human, early modern human and Neanderthal limb proportions." International Journal of Osteoarchaeology 9 (1): 54 67. Purcell, Maureen W. 2013. "Sex Differences in the Femur and Acetabulum: Biomechanical Analysis with Forensic Significance." Raichlen, David A., Adam D. Gordon, William EH Harcourt Smith, Adam D. Foster, and Wm Ra ndall Haas Jr. 2010. "Laetoli footprints preserve earliest direct evidence of human like bipedal biomechanics." PLoS One 5 (3): e9769. Rak, Yoel, and Baruch Arensburg. 1987. "Kebara 2 Neanderthal pelvis: first look at a complete inlet." American Journal o f Physical Anthropology 73 (2): 227 231. Razali, Nornadiah Mohd, and Yap Bee Wah. 2011. "Power comparisons of shapiro wilk, kolmogorov smirnov, lilliefors and anderson darling tests." Journal of Statistical Modeling and Analytics 2 (1): 21 33. Remis, Mel issa. 1995. "Effects of body size and social context on the arboreal activities of lowland gorillas in the Central African Republic." American Journal of Physical Anthropology 97 (4): 413 433.

PAGE 63

57 Richmond, Brian G., David R. Begun, and David S. Strait. 2001. "Origin of human bipedalism: The knuckle walking hypothesis revisited." American Journal of Physical Anthropology 116 (S33): 70 105. Richmond, Brian G., and William L. Jungers. 2008. " Orrorin tugenensis femoral morphology and the evolution of hominin bip edalism." Science 319 (5870): 1662 1665. Roberts, B. C., L. B. Solomon, G. Mercer, K. J. Reynolds, D. Thewlis, and E. Perilli. 2017. "Joint loading and proximal tibia subchondral trabecular bone microarchitecture differ with walking gait patterns in end s tage knee osteoarthritis." Osteoarthritis and Cartilage 25 (10): 1623 1632. Rosenberg, Karen R., Lü Zuné, and Christopher B. Ruff. 2006. "Body size, body proportions, and encephalization in a Middle Pleistocene archaic human from northern China." Proceedi ngs of the National Academy of Sciences of the United States of America 103 (10): 3552 3556. Ruff, Christopher B. 1994. "Morphological adaptation to climate in modern and fossil hominids." American Journal of Physical Anthropology 37 (S19): 65 107. Ruff, Christopher B. 1995. "Biomechanics of the hip and birth in early Homo." American Journal of Physical Anthropology 98 (4): 527 574. Ruff, Christopher. 1998. "Evolution of the hominid hip." In Primate locomotion , pp. 449 469. Springer, Boston, MA. Ruff, C Annual Review of Anthropology 31: 211 232. Ruff, Christopher B. 2002. "Long bone articular and diaphyseal structure in Old World monkeys and apes. I: locomotor effects." American Journal of Phy sical Anthropology 119 (4): 305 342. Ruff, Christopher. 2010. "Body size and body shape in early hominins implications of the Gona pelvis." Journal of Human Evolution 58 (2): 166 178. Ruff, Christopher. 2017. "Mechanical constraints on the hominin pelvis The Anatomical Record 300 (5): 946 955.

PAGE 64

58 Ryan, Timothy M., and Gail E. Krovitz. 2006. "Trabecular bone ontogeny in the human proximal femur." Journal of Human Evolution 51 (6): 591 602. Schindelin, Johannes, Ignacio Argand a Carreras, Erwin Frise, Verena Kaynig, Mark Longair, Tobias Pietzsch, Stephan Preibisch et al. 2012. "Fiji: an open source platform for biological image analysis." Nature Methods 9 (7): 676. Senut, Brigitte. 2006. "Arboreal origin of bipedalism." In Huma n Origins and Environmental Backgrounds , pp. 199 208. Springer, Boston, MA. Bone 30 (5): 765 770. Shelton, Michael. 2015. One of t [PowerPoint Slides] Retrieved from http://slideplayer.com/slide/4698157/ . Simpson, Scott W., Jay Quade, Naomi E. Levin, Robert Butler, Guillaume Dupont Nivet, Melanie E verett, and Sileshi Semaw. 2008. "A female Homo erectus pelvis from Gona, Ethiopia." Science 322 (5904): 1089 1092. communication between osteoblasts, osteocytes and osteo Archives of Biochemistry and Biophysics 561: 22 28. Sockol, Michael D., David A. Raichlen, and Herman Pontzer. 2007. "Chimpanzee locomotor energetics and the origin of human bipedalism." Proceedings of the National Academy of Sciences 104 (30): 12265 12269. Stern, Jack T., and Randall L. Susman. 1981. "Electromyography of the gluteal muscles in Hylobates, Pongo, and Pan: implications for the evolution of hominid bipedality." American Journal of Physical Anthropology 55 (2): 153 166. Stern Jr., Jack T., and Randall L. Susman. 1983. "The locomotor anatomy of Australopithecus afarensis ." American Journal of Physical Anthropology 60 (3): 279 317. Stern, Jack T. 2000. "Climbing to the top: a personal memoir of Australopithecus afarensis ." Evolutiona ry Anthropology: Issues, News, and Reviews 9 (3): 113 133.

PAGE 65

59 Sugano, Nobuhiko, Philip C. Noble, Emir Kamaric, Joseph K. Salama, Takahiro Ochi, and Hugh S. Tullos. 1998. "The morphology of the femur in developmental dysplasia of the hip." The Journal of Bone and Joint Surgery. British Volume 80 (4): 711 719. Tague, Robert G., and C. Owen Lovejoy. 1986. "The obstetric pelvis of AL 288 1 (Lucy)." Journal of Human Evolution 15 (4): 237 255. Tague, Robert G. 1989. "Variation in pelvic size between males and fem ales." American Journal of Physical Anthropology 80 (1): 59 71. Tanck, E. J. M., G. J. Hannink, R. Ruimerman, P. Buma, E. H. Burger, and R. Huiskes. 2006. Journal of Anatomy 208 (1): 73 79. Human Biology edited by G. A. Harisson, J. M. Tanner, D. R. Pilbeam, and P. T. Baker. Oxford: Oxford University Press, pp. 320 360. Tardieu, Christine, and Erik Trinkaus. 1994. "Early ontogeny of the human femoral bicondylar angle." American Journal of Physical Anthropology 95 (2): 183 195. Tardieu, Christine, and Holger Preuschoft. 1996. "Ontogeny of the knee joint in humans, great apes and fossil hominids: pelvi femoral relat ionships during postnatal growth in humans." Folia Primatologica 66 (1 4): 68 81. during growth Surgical and Radiologic Anatomy 19 (2): 91 97 . Tardieu, Christine, Yann Glard, Emmanuelle Garron, Christophe Boulay, J L. Jouve, Olivier Dutour, Gilles Boëtsch, and Gérard Bollini. 2006. "Relationship between formation of the femoral bicondylar angle and trochlear shape: independence of diaphyseal a nd epiphyseal growth." American Journal of Physical Anthropology 130 (4): 491 500. Taylor, M. E., K. E. Tanner, M. A. R. Freeman, and A. L. Yettram. 1996. "Stress and strain distribution within the intact femur: compression or bending?." Medical Engineeri ng & Physics 18 (2): 122 131.

PAGE 66

60 Science 289 (5484): 1504 1508. Thorpe, Susannah KS, and Robin H. Crompton. 2006. "Orangutan positional behavior and the nature of arboreal locomotion in Hominoidea ." American Journal of Physical Anthropology 131 (3): 384 401. Thorpe, Susannah KS, Roger L. Holder, and Robin H. Crompton. 2007. "Origin of human bipedalism as an adaptation for locomotion on flexible branches." Science 316 (5829): 1328 1331. Thorpe, Susa nnah KS, Juliet M. McClymont, and Robin H. Crompton. 2014. "The arboreal origins of human bipedalism." Antiquity 88 (341): 906 914. Trinkaus, Erik. 1993. "Femoral neck shaft angles of the Qafzeh Skhul early modern humans, and activity levels among immature Near Eastern Middle Paleolithic hominids." Journal of Human Evolution 25 (5): 393 416. Videan, Elaine N., and W. C. McGrew. 2002. "Bipedality in chimpanzee ( Pan troglodytes ) and bonobo ( Pan paniscus ): testing hypotheses on the evolution of bipedalism." American Journal of Physical Anthropology 118 (2): 184 190. el for the role of Proceedings of the National Academy of Sciences 104 (40): 15941 15946. Warrener, Anna. 2011. "Pelvic shape, hip abductor mechanics and locomotor energetics in extinct hominin s and modern humans." Warrener, Anna G., Kristi L. Lewton, Herman Pontzer, and Daniel E. Lieberman. 2015. "A wider pelvis does not increase locomotor cost in humans, with implications for the evolution of childbirth." PloS one 10 (3): e0118903. Warrener, Anna G. 2017. "Hominin hip biomechanics: changing perspectives." The Anatomical Record 300 (5): 932 945.

PAGE 67

61 Weaver, Timothy D. 2003. "The shape of the Neandertal femur is primarily the consequence of a hyperpolar body form." Proceedings of the National Acad emy of Sciences 100 (12): 6926 6929. Weinbaum, S., S. C. Cowin, and Yu Zeng. 1994. "A model for the excitation of osteocytes by mechanical loading induced bone fluid shear stresses." Journal of Biomechanics 27 (3): 339 360. Wetzsteon, R. J., B. S. Zemel, J. Shults, K. M. Howard, L. W. Kibe, and M. B. Leonard. 2011. Bone 48 (5): 1103 1108. WoldeGabriel, Giday, Stanley H. Ambrose, Doris Barboni, Raymonde Bonnefille, Lauren t Bremond, Brian Currie, David DeGusta et al. 2009. "The geological, isotopic, botanical, invertebrate, and lower vertebrate surroundings of Ardipithecus ramidus." Science 326 (5949): 65 65e5. Wolff, J. 1986 . The Law of Bone Remodelling . Translated by P. Maquet and R. Furlong. A. Hirschwald, Berlin. First published 1892. Woo, S. L., S. C. Kuei, D. Amiel, M. A. Gomez, W. C. Hayes, F. C. White, and W. H. Akeson. Wolf JBJS 63 (5): 780 787. Yu, Huiying, Murad Alaqtash, Eric Spier, and T. Sarkodie activity during gait cycle using fuzzy rule Measurement 43 (9): 1106 14. Zipfel, Bernhard, Jeremy M. DeSilva, Rober t S. Kidd, Kristian J. Carlson, Steven E. Churchill, and Lee R. Berger. 2011. "The foot and ankle of Australopithecus sediba ." Science 1202703