Development of the fetal ilium – challenging concepts of bipedality

Abstract

Macroradiographs of 30 human fetal and neonatal ilia were analysed to investigate the early pattern of trabecular bone organization prior to the influences of direct weight-bearing locomotion. Consistent and well-defined patterns of internal organization were identified within the fetal and neonatal ilium, which correspond with previously recognized regions that have been attributed directly to forces associated with bipedal locomotion. This study proposes that patterns previously attributed to weight-bearing locomotive responses are present in the earliest stages of the development of this bone. It is suggested that the rudimentary scaffold seen in the fetal and neonatal ilium could indicate a predetermined template upon which locomotive influences may be superimposed and perhaps reinforced at a later age. Alternatively, this early pattern may mimic the adult form due to the effects of in-uterolimb movement activity even though it is not weight bearing. This is a preliminary study that will be supported in a further communication with three-dimensional micro-computed trabecular analysis.

Introduction

In recent studies, and indeed in the historical literature, much work has concentrated on the investigation of trabecular bone architecture with the aim of elucidating its mechanical role within the skeletal system (Evans, 1973; Lanyon, 1974; Carter & Hayes, 1977; Gibson, 1985; Goldstein, 1987; Frost, 1990; Ciarelli et al. 1991; Biewener et al. 1996; Keaveny et al. 2001; Jee, 2005; Liu et al. 2006). Consistently, these investigative studies have made particular and continued reference to long bones and most specifically to the proximal femur (Ward, 1838; Osborne et al. 1980; Wolff, 1986; Carter et al. 1989; Drapeau & Streeter, 2006; Rudman et al. 2006; Ryan & Krovitz, 2006; Skedros & Baucom, 2007). More recent investigations have extended into other previously neglected areas of the skeleton, notably those that are subjected to significant tensile and compressive forces, including the calcaneus (Maga et al. 2006; Rupprecht et al. 2006), talus (Pal & Routal, 1998) and vertebral column (Jensen et al. 1990; Haidekker et al. 1999; Rapillard et al. 2006).

Despite an increasing trend towards investigation of trabecular bone dynamics, surprisingly little is documented regarding the changes to the internal structure of the developing human pelvis (Dalstra et al. 1993). An isolated study by Dalstra & Huiskes (1995) documented the load transfer across the innominate using finite element analysis and built upon previous observations that the internal structure of the pelvic bone was of a sandwich construction, with a core of trabecular bone covered by an outer and an inner thin shell of compact bone (Dalstra et al. 1993). They concluded that a major proportion of the load passing through the adult pelvis is primarily transferred through the cortical shell with stress distribution observed principally in the sacro-iliac region. It was also highlighted that the highest trabecular bone stresses were situated in the central region of the ilium.

The pelvis should be a significant locus for developmental and biomechanical studies as it represents a junctional complex where the transfer of weight from the upper body to the lower extremity is redirected through a non-linear route. As changes associated with body size and locomotor behaviour are reflected in trabecular bone architecture, the pelvic complex should be a key focus for such studies (Oxnard & Yang, 1981; Thomason, 1985; Carter et al. 1989; Galichon & Thackeray, 1997; Macchiarelli et al. 1999; Rook et al. 1999; Fajardo & Müller, 2001; MacLatchy & Müller, 2002; Ryan & Ketcham, 2002; Lai et al. 2005; Ryan & Krovitz, 2006; Fajardo et al. 2007). However, previous studies have tended to consider isolated temporal characteristics relating to a particular bipedal behaviour and have a more phylogenetic rather than ontogenetic emphasis (Macchiarelli et al. 1999; Rook et al. 1999; Marchal, 2000; Martinon-Torres, 2003).

The available information on the internal architecture of the adult human ilium is sparse and what is known is summarized here to permit an understanding of the patterns shown in the younger individuals in this study. The ilium is purported to consist of mainly low-density trabecular bone covered by inner and outer layers of thin cortical bone forming a strong, low-weight structure that is well suited to accommodate high loads and is commonly referred to as a ‘sandwich construction’ by engineers (Dalstra & Huiskes, 1995). Pelvic trabeculae are reported to be predominantly plate-like and oriented perpendicular to the cortical shell, affording the trabecular structure an optimal mechanical orientation for accommodating predominantly shear-loading modes (Dalstra et al. 1993). In the adult ilium (Fig. 1), the most strongly represented cortical/trabecular pattern is located posteriorly in what is commonly referred to as a compressive trajectory, which is directed antero-inferiorly from the sacro-iliac region towards the upper area of the acetabulum. A further structural ray is located anteriorly and is represented by a well-defined and reported tensile trajectory, which passes postero-inferiorly from the anterior superior region towards the upper area of the acetabulum. These two trajectories are believed to absorb and distribute loads that are generated during a striding gait and converge upon a structurally significant central region, termed the trabecular chiasma, which is located in the widest region of the bone (Macchiarelli et al. 1999; Rook et al. 1999). The trabecular trajectories form the boundaries of a poorly represented structural region supero-medially, which is clearly defined in all normal adult innominates and represents an area of the iliac blade where the gluteal (outer) and pelvic (inner) shells of compact bone may fuse without any intervening cancellous bone. Surprisingly, Dalstra & Huiskes (1995), suggested that this is an area of maximum stress distribution within the trabecular bone. Posteriorly in the regions of the greater sciatic notch and acetabular roof there are additional areas of dense structural organization. Each of these regions labelled in Fig. 1 has been explained by various authors as reflecting the interaction of biomechanical forces acting on the pelvic complex during a bipedal stance and bipedal locomotion.

Fig. 1.

Conventional radiograph of an adult innominate. Principle features are labelled: posterior trajectory (pt), anterior trajectory (at), superior medial region (sm), trabecular chiasma region (tc), sciatic notch region (sn) and acetabular roof (ar).

As the primary task of the pelvis is to support the weight of the upper body and transfer it to the lower extremities (Dalstra & Huiskes, 1995), distinct function-related density representations are alleged to be produced. Kapandji (1987) proposed that there are three principle trabecular pathways within the pelvis: two arising from the auricular region that extend towards the acetabulum for weight transfer to the lower limb during bipedal locomotion and one passing to the ischium for weight transfer in sitting (Aiello & Dean, 1990). This architectural representation is said to be produced by the site-specific magnitude and direction of the locomotion-related peak strains imposed on the growing bone (Gibson, 1985; Rook et al. 1999).

During growth, bone structure responds to loading and, in particular, the trabecular architecture is reported to respond to the magnitude and direction of loading reflected in the relative densities and structural organisation of trabecular pathways. These pathways or trajectories are observed to be of high density in regions of higher stress and of conversely low density in regions of lower stress (Lanyon, 1974; Currey, 1986; Turner, 1992). As such, the forces involved in bipedal locomotion are reported to be reflected in the pattern of trabecular trajectories observed in the adult human innominate (Macchiarelli et al. 1999). The density representation in the ilium has been further investigated radiographically with trabecular patterning being documented in both the human and non-human primate (Latarjet & Gallois, 1910; Macchiarelli et al. 1999; Rook et al. 1999; Martinon-Torres, 2003). Studies have confirmed characteristic features of gait-related trabecular systems in the human ilium, with the presence of distinctive trabecular bundles that intersect at the well-defined trabecular chiasma (Macchiarelli et al. 1999).

To date, there appears to be no single study that has investigated the full human ontogenetic spectrum of the development of the internal architecture of the pelvis, encompassing all locomotive stages of development that may influence the morphology of the internal architecture. The lack of information may be attributed to a number of factors, including the paucity of appropriate samples upon which to base such an investigation. This preliminary study utilizes the unique Scheuer collection (University of Dundee, Dundee, Scotland) of juvenile skeletal remains and examines the internal architecture of the ilium via macroradiography to establish the pattern of early-stage trabecular maturation prior to load-bearing locomotive influences.

At the commencement of this study we adopted the null hypothesis that the pattern of fetal and neonatal iliac trabecular organization would be one of random distribution between the gluteal and pelvic cortical shells acting only as a rudimentary spacer for subsequent ontogenetic remodelling.

These results form the first part of a larger study that will also examine changes to the internal architecture upon attainment of different stages of locomotor maturation. Through computed tomography, three-dimensional reconstructions will evaluate remodelling in relation to attainment of locomotor maturation. Therefore, it is the aim of this preliminary part of the study to document the qualitative, gross trabecular organization present in the earliest developmental ranges of the human ilium and to compare this with the well-established, although ill-understood, adult pattern described in the literature.

Materials and methods

Thirty ilia (18 fetal and 12 neonatal) representing the entirety of the prenatal component of the Scheuer collection were analysed (Table 1). This collection is an active osteological repository for human juvenile skeletal remains and consists of over 100 subadult individuals amassed from a variety of archaeological, historical and forensic sources. Individuals in this collection range from early fetal age through to the final stages of adolescence. For this study, if the age of the specimen was not known, it was assigned following the metric evaluations described by Fazekas & Kosa (1978). Ilium length (maximum distance between the anterior and posterior superior iliac spines) and ilium width (maximum distance between the mid-point of the iliac crest and the convexity of the acetabular extremity) were measured. Specimens were then assigned to one of four age groups (Table 1).

Table 1.

Scheuer collection specimen age ranges and numbers used in this study

Age (weeks)	n
18–22	2
23–30	5
31–39	11
40+ (term)	12

Each specimen was subjected to macroradiography using a Multix Tube & Table (Siemens). The exposure factors used were 47 kV, 2 mA, Fine Focus, with an Agfa film screen combination. A focus film distance of 140 cm and an object film distance of 30 cm were applied. Macroradiography is a radiographic imaging technique used to increase the size of the image relative to that of the object. It is based upon purposely increasing the object film distance in relation to a fixed focal film distance (Clark et al. 1984; Davidson & Bowman, 2002).

The pattern of trabeculae was analysed by transferring the resulting image to a software package capable of differential mapping of grey levels. As macroradiographs were produced as hard copies, they had to be converted into a digitized format to facilitate manipulation (Fig. 2). A flat-bed scanner with a superiorly-mounted radiographic lamp was used to obtain high-resolution images of each radiograph. The scan resolution was set to a consistently high dpi of 1200, which optimized visualization and file size. Each image was saved as a TIFF file without compression and was comprised of a standard 256 grey levels. Scans were opened in the Adobe PhotoShop CS2 application and auto-coloured, relative to their grey levels, using the gradient map tool. The ‘gradient map’ command maps the equivalent greyscale range of an image to the colours of a specified gradient fill. A four-colour gradient fill was selected to assist in the evaluation of gross trabecular organization. Grey values in the image were mapped to one of the endpoint colours of the gradient fill, dependent on their predefined percentage value. This technique employs a colour look-up table function, a tool that converts the logical colour numbers stored in each pixel into physical colours. In using this software application a representation of the two-dimensional radiographic structural organization can be obtained and depicted in colour. This process was repeated for each grey level range, ultimately producing a colour-coded map of graded structural ‘densities’ (Fig. 3).

Fig. 2.

Macroradiograph and gradient map of a neonatal ilium. Three distinct grades of differing density are observed within the ilium as represented by the three colours.

Fig. 3.

Macroradiograph (a) and gradient map (b) of fetal specimen (18–22 weeks). Areas outlined include density regions associated with the greater sciatic notch (sn), anterior trajectory (at) and posterior trajectory (pt). Macroradiograph (c) and gradient map (d) of fetal specimen (23–30 weeks). Areas outlined include density regions associated with the greater sciatic notch (sn), acetabular roof (ar), anterior (at) and posterior (pt) trajectories and trabecular chiasma (tc). Macroradiograph (e) and gradient map (f) of fetal specimen (31–39 weeks). Areas outlined include density regions associated with the greater sciatic notch (sn), acetabular roof (ar), anterior (at) and posterior (pt) trajectories and the trabecular chiasma (tc). Macroradiograph (g) and gradient map (h) of neonatal specimen (40+ weeks). Areas outlined include density regions associated with the greater sciatic notch (sn), acetabular roof (ar), anterior (at) and posterior (pt) trajectories and the trabecular chiasma (tc).

Macroradiographic gradient analysis

Gradient analysis divided the ilia into a colour map composed of four colours representing differing apparent ‘densities’ (Fig. 3). The colours selected were: blue, representing the highest density regions of 71–100% opacity; orange, representing densities of 51–70% opacity; violet, representing densities of 31–50% opacity; and yellow, representing the lowest density regions of 0–30% opacity, which almost exclusively represented background levels of film exposure.

Results

An incremental and progressive pattern of developing internal bone architecture was observed between each of the age groups. Within each age cohort the extent of structural maturity was ranked from least to most mature so that the progressive pattern of structural maturation could be examined and any variability in the pattern could be highlighted. Passing from left to right, Fig. 4 illustrates the least, average and most mature specimens within each of the four age cohorts.

Fig. 4.

Gradient maps of multiple specimens within each developmental group. Those to the left represent the least mature within each age group and those to the right are the most mature, whereas those in the middle represent the modal appearance of specimens in each age group. (a) 18–22 weeks. (b) 23–30 weeks. (c) 31–39 weeks. (d) 40+ weeks. The scale bar provided is relevant to each image.

18–22 weeks (n = 2)

In the youngest fetal specimens (Figs 3a and b, and 4a), radiographs were difficult to analyse due to the small dimensions of the pelvic complex and the low level of mineralization. This resulted in a somewhat distorted resolution as individual trabeculae can be as small as 45 µm (Salle et al. 2002; Nuzzo et al. 2003). The radiographic representation was also obscured as specimens possessed a small degree of retained but mummified soft tissue. As such, the ilium, ischium and pubis were conjoined in an orientation that prevented optimal flat-plate radiographic imaging. Due to the importance of the Scheuer collection, specimens could not be macerated as this would result in irreversible damage to this irreplaceable material. Although the gradient representation was somewhat obscured, a basic pattern of internal architecture could still be observed. As such, interpretations from this age group were treated with caution as being loosely supportive of the more mature pattern witnessed in the other age groups. Both specimens showed faint patterns of increased density in the region of the greater sciatic notch and in the central region of the ilium with identifiable trajectories observed extending into both anterior and posterior regions of the bone.

23–30 weeks (n = 5)

A consistent pattern was observed in each specimen (Fig. 3c and d) that was progressive to the primitive pattern identified in the 18–22 week group. The most significant differences were observed in the region of the greater sciatic notch, which was represented by a more clearly defined region of increased density. The anterior and posterior trabecular fanning were also more clearly delineated and could clearly be seen emanating from the area of the trabecular chiasma. An area of markedly increased density was also observed in the region of the acetabular roof.

31–39 weeks (n = 11)

The specimens from this more mature period of fetal development (Fig. 3e and f) displayed a progressive pattern from the previous age cohort. More refined details relating to both the external and internal characteristics were observed. This was primarily due to the increased size of the ilia, allowing for greater macroradiographic resolution. Trajectory lines were clearly visible passing in rays from the well-defined trabecular chiasma to both the posterior area of the ilium and the anterior superior iliac spine region. These trajectories could be seen to form the boundaries of a well-defined area of low density in the body of the iliac blade. There was also a more clearly defined radio-opaque region at the greater sciatic notch and acetabular roof.

40+ weeks (n = 12)

In the neonatal age range (40+ weeks) (Fig. 3g and h), radiographs presented markedly improved visual representation with greater definition and delineation of all internal structures that had been identified in the younger age cohorts. Regions of highest structural density, depicted in blue in Fig. 5, presented in three distinct areas of interest: (1) trabecular chiasma, represented by the intersection of two dense flanges, the first passing antero-inferior to postero-superior and another passing postero-inferior to antero-superior; (2) acetabular region, represented in the area of the future acetabular roof; and (3) sciatic notch, represented as a wedge of well-defined dense bone associated with the position of the greater sciatic notch.

Fig. 5.

Gradient map of a neonatal ilium illustrating regions of density. Acetabular roof (ar), anterior (at) and posterior (pt) trajectories with central trabecular chiasma (tc) and greater sciatic notch region (sn).

Discussion

The radiographic representation of all specimens in this study demonstrated a progressive but consistent pattern of trabecular alignment that was comparable with the template identified in more developmentally mature individuals. Density patterns, which have previously been attributed to locomotive response in the adult, were observed in both the fetal and neonatal samples prior to the possibility of any significant weight-bearing locomotive influences.

Specific areas of internal architectural interest have been identified in the adult human ilium and previous literature has suggested a strong relationship with locomotion and bipedal gait (Thomason, 1985; Carter et al. 1989; Galichon & Thackeray, 1997; Macchiarelli et al. 1999; Rook et al. 1999; Fajardo & Müller, 2001; MacLatchy & Müller, 2002; Ryan & Ketcham, 2002; Lai et al. 2005; Ryan & Krovitz, 2006; Fajardo et al. 2007). The primary areas of functional interest in the adult include two distinct flanges of structural significance present within the anterior and posterior portions of the ilium. The first and most pronounced of these is a structural ray that is located in the posterior aspect of the ilium. Some previous literature has attributed this to biomechanical compressive forces passing from the sacro-iliac joint through the body of the ilium towards either the acetabulum in a bipedal stance or towards the ischium in a sitting posture (Kapandji, 1987; Aiello & Dean, 1990; Scheuer & Black, 2000). The second consistent structural ray is located anteriorly within the ilium and has often been attributed to the tensile forces that are set up in this region to act as a counterbalance to the compressive forces from the posterior flange to prevent buckling of the bone under the pressures of bipedal locomotion (Macchiarelli et al. 1999). These two defined rays of structural significance, previously attributed solely to forces associated with bipedal locomotor abilities, converge upon a structurally distinct region within the ilium termed the trabecular chiasma. This structural chiasma has been described in terms of bipedal gait as being the locus where loads derived from the sacro-iliac joint and those arising from the acetabulum are distributed and absorbed (Macchiarelli et al. 1999; Rook et al. 1999). Furthermore, it has been suggested that the well-developed arcuate bundles in the posterior and anterior flanges, which intersect into the high-density trabecular chiasma, are related to striding gait (Correnti, 1957; Dalstra & Huiskes, 1995; Rook et al. 1999). Further areas of internal architectural interest have been identified in the acetabular roof and sciatic notch of the ilium. Previous literature has emphasized that these density representations are also gait related with a maintained emphasis on the forces that pass through the sacrum into the ilium via the sacro-iliac joint (Dalstra & Huiskes, 1995). These forces are reported to set up distinct trabecular bundles that pass posteriorly in a sacro-ischial trajectory, contributing to the sciatic notch density representation, and through the acetabulum in the superior and inferior auriculo-acetabular trajectories, contributing to the density representation observed in the acetabular component of the ilium (Kapandji, 1987; Aiello & Dean, 1990; Rook et al. 1999). Additionally, the iliac part of the acetabulum has been noted to be particularly robust in the adult, due to alterations in bony trabeculation caused by buttressing in the weight-bearing line of the pelvis (Johnstone et al. 1982). Finally, the superior medial region of the ilium is observed to be structurally redundant in the adult due to the proximity of the gluteal and pelvic cortices and the absence of any significant trabecular structure. Surprisingly, this observation of maintained structural redundancy is in contrast to observations by Dalstra & Huiskes (1995) who suggested that this region was an area of maximal stress distribution within the trabecular bone. This suggestion is confusing as it would be expected that trabecular regions experiencing high stresses would adapt in morphology to a structural conformation capable of accommodating the prevailing stresses rather than simply regressing and, in some individuals, disappearing completely.

Each of these internal architectural regions of interest has been well documented for mature individuals and has been explained in terms of locomotion and bipedal gait. However, in this preliminary study, the distinct structural regions of the adult representation have been unequivocally identified in the fetus and neonate, which cannot be influenced by direct stance-related weight transfer as the pelvis is not weight bearing in utero (Walker, 1991).

Therefore, an alternative suite of forces and influences must be considered during the earliest stages of development of the ilium to cause and influence the maintenance of this distinctive pattern. Consideration has been given to potential genetic influences on the bone structure and resultant density representation present in the earliest of developmental stages. The relative influences of mechanical and genetic factors are still a question of great debate with the full contribution of genetic influences to bone development not fully understood (Bertram & Swartz, 1991; Huiskes, 2000; Huiskes et al. 2000; Lovejoy et al. 2002; Pearson & Lieberman, 2004; Ruff et al. 2006). However, the role of genetic and epigenetic influences on bone development have been investigated extensively in recent years by employing experimental models where paralysis has been artificially induced or is inherent in congenital conditions such as cerebral palsy (Hall & Herring, 1990; Hosseini & Hogg, 1991; Germiller & Goldstein, 1997; Bobroff et al. 1999; Henderson et al. 2005). In using such models the influences of muscular contraction on bone development can effectively be eliminated allowing for a greater insight into genetic contributions to skeletal form. These studies have strongly suggested that bone form and internal structure are in part genetically determined as, although skeletal growth is reduced (Hall & Herring, 1990), bone structure is not significantly altered in the paralysed state (Hall & Herring, 1990; Hosseini & Hogg, 1991; Germiller & Goldstein, 1997; Gilbert et al. 2004; Henderson et al. 2005; Sawamura et al. 2006).

It is further suggested that, coupled with genetic influences, mechanical stimuli in utero and in later development may serve to reinforce bone shape and structure (Ruff et al. 2006; Skedros et al. 2007). The evidence that the final structure of bone results from a combination of genetic, epigenetic and extragenetic factors is now widely accepted (Skedros et al. 2007). In the case of the developing ilium, the regions of increased density may well represent initial genetic patterning upon which subsequent in-uteroloading may then direct the remodelling of this primary structure into the form observed in this study.

During fetal development the primary forces acting on the ilium may include those induced by primitive reflex muscular action. These reflex contractions are known to be responsible for initiating intramembranous ossification of the ilium (Laurenson, 1964; Delaere et al. 1992) and recent studies have shown that early limb movements induced by random neurological firing are pivotal to the normal development of the locomotor apparatus (Pitsillides, 2006).

As a function of muscular attachment, movement induced by limb musculature will have an influence on all three components of the innominate (ilium, ischium and pubis) as the ilium maintains cartilaginous continuity with both the ischium and pubis via the acetabulum from its earliest embryological formation (Laurenson, 1963; Scheuer & Black, 2000; Lee & Eberson, 2006). Therefore, it is highly likely that any forces acting on these developing bones, both independently and in unison, may be transferred across the primitive acetabulum into the ilium. As bone is highly responsive to biomechanical forces (Lanyon, 1974; Currey, 1986; Linde et al. 1991; Turner, 1992; Huiskes et al. 2000), the patterns that are observed passing in well-defined trajectories through the ilium may represent some element of force distribution from these sources. The posterior trajectory passes in a direct line from the acetabular site of articulation between the ilium and pubis, whereas the anterior trajectory passes in a direct line from the acetabular articulation between the ilium and ischium. It is possible that forces instigated by early reflexive limb movement, which originate through muscle attachment to both the ischial and pubic components of the developing pelvis, might be transferred across the acetabular complex to manifest in the ilium. The concept of ossification and bone induction by early reflexive movement of musculature has been discussed in other areas of the skeleton including neural arch ossification in response to the gasp reflex (Bagnall et al. 1977) and, as such, reinforces the possibility that the early internal architecture of the human ilium could mirror the final adult pattern as both reflect a form of force transfer across the acetabulum in relation to the lower limb movement. Laurenson (1964) indicated that the position of the nutrient foramen, which lies over the region of the trabecular chiasma, is the first region of the ilium to commence ossification and there is evidence in the literature to suggest that areas of intersection of differential stress patterns can act as an initiator for bone deposition and as an angiogenic attractant (Carter & Beaupre, 2001). Although it is reported that the compact shells of the ilium ossify through intramembranous ossification resulting from the attachment of the iliacus muscle on the inner surface and the gluteal muscles on the pelvic surface (Scheuer & Black, 2000), the maintenance and development of the internal architecture could also be influenced by forces generated through muscle contraction and limb movement associated with muscular groups attaching to the ischium (hamstrings) and pubis (adductors).

This study has contradicted this null hypothesis by highlighting that the iliac trabecular architecture is organized in well-defined and regular patterns from a very early stage of fetal development. This precocious development mirrors the more mature pattern, which has been attributed to biomechanical forces associated with bipedalism. If there is indeed a basic genetic internal and external form to the ilium then it is possible that the patterns of internal architecture seen in this study may represent a maintenance and reinforcement of that preliminary genetic scaffold by the forces instigated by in-utero reflexive limb movement.

Conclusion

The results of this preliminary study revealed that a recognizable internal cancellous architectural pattern is established at a very early age in the human ilium. This pattern mirrors the template seen in the adult, which has been attributed in the past to a biomechanical response to weight-bearing bipedal locomotion. As the fetal and neonatal representation is free from direct stance-related weight transfer, further investigation is required to elucidate the origins and progression of structured bone patterning in the ilium. It is proposed that the structural observations made in this study in non-weight-bearing fetal and neonatal pelves may be a preliminary response to early limb movement perhaps augmenting a pre-existing genetic template. Further work will examine the three-dimensional pattern of the trabecular system of the juvenile ilium via micro-computerized tomography imaging and finite element modelling. This will aid the evaluation of changes that occur in this primitive pattern as more mature influences of locomotion manifest in older juveniles as the child moves from sitting, to crawling and finally to a mature bipedal gait.