1

This is a modified version of a paper published in the International Journal of Osteopathic Medicine 11 (2008) 80-89

Received 20 September 2007; received in revised form 5 January 2008; accepted 27 March 2008

A model of the cranial vault as a tensegrity structure, and its

significance to normal and abnormal cranial development

Graham Scarr
60 Edward Street, Stapleford, Nottinghamshire NG9 8FJ, UK

Abstract
Background: Traditional views of the human cranial vault are facing challenges as researchers find
that the complex details of its development do not always match previous opinions that it is a
relatively passive structure. In particular, that stability of the vault is dependant on an underlying
brain; and sutural patency merely facilitates cranial expansion. The influence of mechanical forces
on the development and maintenance of cranial sutures is well-established, but the details of how
they regulate the balance between sutural patency and fusion remain unclear. Previous research
shows that mechanical tensional forces can influence intracellular chemical signalling cascades and
switch cell function; and that tensional forces within the dura mater affect cell populations within the
suture and cause fusion.
Understanding the developmental mechanisms is considered important to the prevention and
treatment of premature sutural fusion - synostosis - which causes skull deformity in approximately
0.05% of live births. In addition, the physiological processes underlying deformational plagiocephaly
and the maintenance of sutural patency beyond early childhood require further elucidation.
Method: Using a disarticulated plastic replica of an adult human skull, a model of the cranial vault
as a tensegrity structure which could address some of these issues is presented.
Conclusions: The tensegrity model is a novel approach for understanding how the cranial vault
could retain its stability without relying on an expansive force from an underlying brain, a position
currently unresolved. Tensional forces in the dura mater have the effect of pushing the bones apart,
whilst at the same time integrating them into a single functional unit. Sutural patency depends on
the separation of cranial bones throughout normal development, and the model describes how
tension in the dura mater achieves this, and influences sutural phenotype. Cells of the dura mater
respond to brain expansion and influence bone growth, allowing the cranium to match the spatial
requirements of the developing brain, whilst remaining one step ahead and retaining a certain
amount of autonomy. The model is compatible with current understandings of normal and abnormal
cranial physiology, and has a contribution to make to a hierarchical systems approach to whole body
biomechanics.

Keywords: Craniofacial; Craniosynostosis; Cranium; Dura mater; Mechanotransduction; Morphogenesis; Plagiocephaly; Skull; Sutures; Tensegrity
 2

INTRODUCTION

For many years it has been widely accepted that the cranial vault expands through an outward pushing

pressure from the growing brain, with the sutures merely accommodating its growth and fusing in the third

decade of life.1,2 However, recent data suggests that daily brain growth is too small to induce sutural

osteogenesis, and that in any case, substantial growth is over before the completion of sutural growth.3,4,5,6

Human facial sutures normally remain patent until at least the seventh or eighth decade, whereas the timing

of sutural fusion in the cranial vault is extremely variable and unreliable forensically.7,8 Many factors affect

cranial enlargement - some are genetic while others are epigenetic.

Understanding the developmental mechanisms of the cranium is considered important to the prevention

and treatment of the pathologies affecting the neonatal cranium. Craniosynostosis is the premature fusion of

one or more of the cranial sutures resulting in skull deformity, and occurs in roughly 1 in 2000 live births.4 It

may be associated with specific genetic syndromes or occur sporadically, and any cranial suture may be

involved, although with differing frequencies.2,9,10 Premature fusion results in arrested bone growth perpen-

dicular to the synostosed suture, with subsequent abnormal compensatory growth in the patent sutures.1,2,9,11

Another skull deformity, not due to synostosis, is positional moulding or deformational plagiocephaly. When

present at birth it is the result of in-utero or intrapartum molding, often associated with multiple births, forceps

or vacuum-assisted delivery; or post-natally resulting from a static supine positioning.12 One of the difficulties

during this period is differentiating premature fusion from abnormal moulding. By the time children are

diagnosed with craniosynostosis, the suture has already fused and the associated dysmorphology well

established. Surgical intervention may then be necessary for neurological or cosmetic reasons.

The adult skeleton is mostly capable of healing defects and deficiencies via the formation of new bone.

However, while children under the age of 2 years maintain the capacity to heal large calvarial defects, adults

are incapable of healing the smallest of injuries. The coordinating mechanisms behind normal and abnormal

development are currently incomplete,10,13 and the model to follow presents a novel approach to furthering

our understanding of the processes involved. Although many readers will have an extensive knowledge of the

cranium, others may be unfamiliar with the details which underlie the significance of this model, and a brief

overview follows.
 3

The Cranial Vault or calvarium: The cranial vault, or calvarium, surrounds and encloses the brain, and is

formed from several plates of bone which meet at sutural joints, unique to the skull, and which display a

variety of morphologies specific to each suture.2,7,11,14,15 The high compressive and tensile strength of bone

provides mechanical protection for the underlying brain, while the sutural joints provide a soft interface and

accommodate brain growth.10 The vault bones are the frontal, parietals and upper parts of the occiput,

temporals and sphenoid (Fig 4&11). Inferior to the vault is the cranial base, or chondrocranium, which is

made up of the lower parts of the occiput and temporals, the ethmoid and the majority of the sphenoid. In the

embryo, the vault bones develop through ossification of the ectomeninx - the outer membranous layer

surrounding the brain; while the cranial base develops through an additional cartilaginous stage, the signifi-

cance of which will be discussed later. (Individual bones spanning both regions fuse at a later stage).

Enlargement of the neurocranium occurs through ossification of sutural mesenchyme at the bone edges, and

an increase in bone growth around their perimeters.1,15 During this process, the ectomeninx becomes

separated by the intervening bones into an outer periosteum and internal dura mater. By the time of full term

birth, the growth of the different bones has progressed sufficiently so that they are in close apposition, only

separated by the sutures which intersect at the fontanelles (Figure 1). At full-term birth, sutural bone growth

is progressing at about 100microns/day, but this rate rapidly decreases after this. Maintenance of sutural

patency is essential throughout for normal development of the brain and craniofacial features.2,4,10 The brain

has usually reached adult size by the age of 7 years but the sutures normally persist long after this - until at

least 20 years of age. Even after this, there is considerable variation in the pattern and timing of sutural fusion

in the human adult throughout life.2,7,8,16 Animal studies of the cranial vault clearly demonstrate
sagittal
sutural patency throughout.2,16
coronal
Anterior
fontanelle

metopic
Parietal bone

Frontal bone

lambdoidal

squamous

Fig.1. Full-term foetal skull showing vault sutures and anterior fontanelle.
 4

The Dura Mater: The dura mater is the outer one of three membranes surrounding the brain (figure 2). Its

outer surface – the endosteal layer, is loosely attached to most of the inner bone surface, particularly in

children, but more firmly attached around the bone margins, the base of the skull and foramen magnum. The

inner meningeal layer of the dura mater continues down through the foramen magnum and surrounds the

spinal cord as far as the sacrum. This layer also reduplicates inwards as four sheets which partially divide the

cranial cavity and unite along the straight sinus - the falx cerebri, falx cerebellum and bilateral tentorium

cerebelli (Figure 2&11).

The internal structure of the dura mater consists of inner and outer elastic networks and integumentary

layers, and a collagen layer; although abrupt boundaries between these ‘layers’ cannot be distinguished

histologically.17 The collagen layer occupies over 90% of its thickness, with collagen fibres arranged in

parallel bundles and differing orientations - varying from highly aligned to apparently random, and arranged

in lamellae.18 Typically, with age, the dura mater thickness changes from 0.3 to 0.8 mm.17,18 Collagen has the

strongest mechanical properties of the different structural proteins, and fibre orientation has been observed to

coincide with the direction of tensile stress.9,18,19,20

The Sutures: Adjacent cranial vault bones are linked through fibrous mesenchymal tissue, referred to as the

sutural ligament (figure 2).15 The two layers which derive from the embryonic ectomeninx – the periosteum

and dura mater, continue across the suture, and also unite around the bone edges.15 In the cranial base,

ossification occurs through cartilage precursors, some of which fuse together in the foetus or early childhood.
Periosteum Osteogenic bone front Bone cortex

Bone Sutural Bone

ligament

Superior
Endosteal sagittal Arachnoid
dura mater sinus mater
Meningeal
dura mater

Pia mater
Cerebral Cerebral
hemisphere hemisphere
Falx cerebri

Figure 2. Schematic diagram of the superior sagittal suture in coronal section. Note the
continuity between periosteum and dura mater; the firmer adherence of the endosteal
‘layer’ of dura mater at the bone margins; and reduplication of the meningeal ‘layer’
of dura mater between the cerebral hemispheres that form the falx cerebri.
(Drawing not to scale)
 5

The synchondroses are the intervening cartilages between the bones of the cranial base. The spheno-basilar

synchondrosis normally ossifies in the third decade, and the petro-occipital fissure (synchondrosis) in the

seventh.21 The cranial base is relatively stable during development, with the greatest size changes taking place

in the vault.

Morphogenesis and phenotypic maintenance of the sutures is a result of intrinsic differences within the

dura mater.1,5,10,16,20,22 The significant factors in this are cellular differentiation, intercellular signals and

mechanical signals.23

(1) Cells of the dura mater beneath the suture undergo epithelial-mesenchymal transitions - a mechanism for

diversifying cells found in complex tissues, and migrate into the suture as distinct cell populations (Fig

10).23,24,25 Fibroblast-like cells in the centre produce collagen and maintain suture patency. Those with an

osteoblast lineage also produce a collagen matrix, but lead onto bone formation at the suture margins,

causing the cranial bones to expand around their perimeters.13 Osteoclast mediated bone resorption may be

necessary for changes in the complex morphological characteristics at the sutures edges.26 A complex

coupling between fibroblast, osteoblast and osteoclast populations determines the actual position and rate

of sutural development.5,10,26,27 In addition, a critical mass of apoptotic cells within the suture is essential

to maintaining the balance between sutural patency and new bone formation.10,14

(2) Intercellular signalling influences cellular function through the production and interactions of soluble

cytokines such as the ‘fibroblast growth factors’ and ‘transforming growth factors’.23,25 The cells at the

approximating edges of the bones, either side of the suture (bone fronts), set up a gradient of growth factor

signalling which regulates the sequential gene expression of other cells, and causes changes in the spatial

and temporal development of different cell populations.10,13,22,28

(3) Mechanical signals. The morphology of the suture also reflects the intrinsic tensional forces in the dura

mater, in the order of nano or pico Newtons.1,3,27,28 Regional differentials in this tension create mechanical

stresses which interact and exert their effects on the cells, stimulating them to differentiate and produce

different cell populations.4,20,23,27,28 The sensitivity of the cellular cytoskeleton to tensional forces, and the

particular pattern of stress application, has been shown to be crucial in determining the cellular response

through a process of mechanotransduction.2,28-34 Given that the cytoskeleton is attached to the surrounding

extracellular matrix through mechano-receptors in the cell membrane, a mechanical force transfer between
 6

them can produce global changes within the cell by altering the cytoskeletal tension. Multiple chemical

signalling pathways are activated within the cell as a result, and together with intercellular chemical signals,

provides multiplexed switching between different functional states such as differentiation, proliferation and

cell death.29,30,32

It is actually not an essential requirement for a spherical tensional structure to be maintained through an

expansive force (such as a growing brain) in order to remain stable.3,35 The proposal here is that the calvarium

of the neonate could be such a structure which maintains its shape through other mechanisms, being

influenced by the expanding brain as a secondary factor.

THE TENSEGRITY MODEL

The concepts of tensegrity have become increasingly recognized over the last thirty years as a model for

understanding some of the structural properties of living organisms.29,30,35-42 This appreciation follows from

investigations in the 1940s by the sculptor Kenneth Snelson, and the architect Buckminster Fuller, into novel

structures in free standing sculpture and building design.35,41 Although Snelson actually coined the term, and

has used it to great effect in his sculptures, it was Fuller who defined the basic mathematics. The word

‘tensegrity’ is derived from the words ‘tension’ and ‘integrity’ and describes structures which are inherently

stable as a result of their particular geometry.

Fuller described two basic types of tensegrity structure - geodesic and prestressed, and found the

icosahedron a useful shape for modelling them - itself a geodesic structure.35,36 The outstanding feature of

geodesic structures is that they have a rigid external frame maintaining their shape, based on a repeating

pattern of simple geometry (figure 3a). In the human body, this type of structure is found in the cytoskeletal

cortex of most cells;43 and in the erythrocyte, the geodesic structure is considered a primary contributor to the

functionality of its peculiar shape.44 Prestressed structures have been well described by Ingber in the inner

cytoskeletons of cells;29,30 and Levin in the shoulder, pelvis and spine,36-40 suggesting their ubiquity through-

out the organism.

In development of the model, the geodesic icosahedron is converted into a prestressed tensegrity structure

by using six new compression members to traverse the inside, connecting opposite vertices and pushing them

apart (figure 3b). Replacing the edges with cables now results in the outside being entirely under isometric

tension. The inward pull of the cables is balanced by the outward push of the struts, providing structural
 7

integrity so that the compression elements appear to float within the tension network. A load applied to this

structure causes a uniform change in tension around all the edges (cables), and distributes compression evenly

to the six internal rods, which remain distinct from each other and do not touch.35 (Some of the edges of the

geodesic structure (figure 3a) have disappeared in the transition to prestressed tensegrity (figure 3b) because

they now serve no structural purpose and are redundant.) Replacing the straight rods in figure 3b with curved

ones (figure 3c) maintains the same stability, but they now surround a central space. In the same way, the

curved rods can be replaced with curved plates (not shown) and the structure still retains its inherent stability.

(a (b) (c)

Figure 3. Diagram showing how a geodesic tensegrity icosahedron (a), can be converted into a
prestressed tensegrity structure using straight internal struts (b). The straight struts can be replaced with
curved struts without altering the tensegrity principle (c). Curved plates (not shown) could be fixed to
the curved struts so that they do not contact each other directly, and the structure would still remain
stable. See text for details.

The plastic adult skull model illustrated in figures 4 - 7 shows curved plates of cranial bone - representing

the compression struts, apparently ‘floating’ in the dura mater - shown here as elastic tension cords. The bones

do not make actual contact with each other at any point. As this paper essentially concerns the cranial vault,

the facial bones have not been separated. Bones of the cranial base are shown here as part of a prestressed

structure in spite of the synchondroses being under a certain amount of compression in vivo. Their develop-

ment in the early embryo could serve as a prestressed tensegrity, only changing to a more geodesic structure

as the cartilage growth plates replace membrane between the bones. They are shown as they are in order to

demonstrate the potential of the prestressed tensegrity principle through all stages of cranial development.

Substituting the tension cords of these model sutures with a compression union would not alter that principle

in the vault. The spheno-basilar synchondrosis (figure 7) has been distracted in order to display the isolation

of each bone within the dura mater more clearly. (It also supports the unbalanced weight of the face pending

further modelling.) Internal cranial structures have been omitted for the sake of clarity.
 8

frontal
parietal

sphenoid
Ethmoid temporal
crista gal-
li

Figure 4. Tensegrity skull model, antero-lateral view.

parietal

7b
7a
occiput
temporal

Figure 5. Tensegrity skull model, postero-lateral view.

To frontal
parietal frontal

sphenoid

tempo-
ral crista
galli

to sphenoid

Figure 6. Tensegrity skull model, right orbital view.

 9

parietal
frontal

sphenoid

Left temporal
Figure 6b Left infero-lateral
maxilla

SB occiput
S
Right temporal
Figure 7. Tensegrity skull model, left infero-lateral view. (SBS – spheno-basilar synchondrosis)

A fundamental characteristic of prestressed tensegrity structures is, as Fuller described it, “...continuous

tension and discontinuous compression”.35 These concepts are illustrated in figure 8a which shows a schemat-

ic diagram of the bones spread out in two dimensions. The bones are the compression elements which are

being pulled by dural tension (only a small number of tension forces pulling in one general direction are

shown in this diagram). Here they remain distinct from each other and do not make contact with each other at

any point - ‘discontinuous compression’. This contrasts with figure 8c, which shows the compressive load of

a stone wall bearing down through the keystone and both sides of the arch - the compression force here is

continuous.

Figure 8. (a) Schematic diagram of the vault bones spread out to illustrate how the
compression elements (bone) of the prestressed tensegrity structure remain discrete, and do
not make direct contact with each other (‘discontinuous compression’). A tension force
pulling on one side of a bone (small arrows), will pass through that bone and generate another (c) Diagram showing compression
tension force on the opposite side. This pulls on the adjacent bone through the dura mater bearing down on the keystone of a
and the process repeats itself. Large arrows show the direction of the resultant tensional stone arch and distributing itself
force. Many of these resultants (not shown) will be acting in different directions in the model. either side through continuous
(b) Diagram of the bones arranged in a circular anatomical sequence, showing the same contact.
resultant tensional force moving around the sequence in both directions (‘continuous ten-
sion’).
 10

Returning to figure 8a, the tension cords are pulling in different directions, but a resultant tensional force

develops (large arrows) which is dependent on the size and direction of the contributing tensions (the

‘parallelogram of forces’ in mechanics terminology). Starting with the left temporal: the tension pulls the left

parietal (indirectly here) towards the left temporal in the direction of the resultant force. At the same time, the

left parietal is pulling on the right parietal through the same mechanism, and this in turn is pulling on the right

temporal. The consequence of all this is brought together in figure 8b, showing the same bones arranged in a

circular anatomical sequence, the resultant tension pulling on each bone in turn, passing around the circle, and

ultimately pulling on itself – ‘continuous tension’. Before running away with thoughts of perpetual motion, it

must be pointed out from Newtons third law of motion that -‘action and reaction are equal and opposite’. So,

an equal and opposite tensional force will also be pulling in the opposite direction with the effect of – zero –

nothing happens! This same isometric tension is acting across all the sutures in 3 dimensions, and because it

is a prestressed tensegrity, the consequence is that all the tensional forces are balanced, the bones appear to

float, and unless acted upon by another force (Newtons first law of motion) the structure will remain as it is

forever. The precise placement and directions of the tensions is extremely important if the structure is to

maintain itself as described, and is detailed later. While the simple 6-strut model is useful for demonstrating

prestressed tensegrities, such structures can be made using any number of compression struts from two

upwards, with the compression members remaining distinct from each other.45

The model was constructed from a full size plastic adult skull obtained from a medical suppliers and cut

into the individual bones using a fine coping saw, with the exceptions of the facial bones which remained as

a unit with the sphenoid. Although the intricacies of the serrate sutures could not be followed exactly,

comparison with a real bone skull confirmed their essential similarities for the purpose described. Holes

drilled at the bone perimeters were threaded with an elastic cord, as used in textile manufacture. The tension

cords are positioned so that they illustrate the nature of the tensegrity structure and do not necessarily follow

any particular anatomic structure. However, during positioning of the attachment holes, it became apparent

that they should be as close to the edge as possible in order for the structure to work effectively. It was also

evident that the various curves of the bone edges, in all three dimensions, facilitated a separation of the bones

by making alternate attachments between the peaks of opposing bone edge convexities (Figure 9a).
 11

DISCUSSION

One of the difficulties found in constructing this model was the unexpected vault shape changes caused by

adjusting individual cord tensions. Prestressed tensegrities have viscoelastic properties similar to biological

structures, and this can cause them to behave unpredictably because of a non-linear relationship between

stress and strain.9,35,46 A summary of some of the significant mechanical aspects of prestressed tensegrity

design and how they apply to the human skull follows:

3.1. Stability

Stability is achieved through the configuration of the whole network, and not because of the individual

components. The model describes a mechanism whereby the calvarial shape could be maintained independ-

ently of any outward-pushing pressure from the brain within,1-6 a position currently unresolved. The sutures

remain under tension (tension being necessary for regulating bone growth), while the bones remain mechan-

ically distinct from the brain, being influenced through cells of the dura mater to expand. It is likely that the

vault shape of the early foetus would be reliant on the expanding brain pushing outwards on the

ectomeninx,2,4,10 but prestressed tensegrity could become a significant factor after 8 weeks, as ossification

stiffens the membranous tissue and transfers tensional stresses across the developing bone (figure 8a).23

Chondrification would transform the base into a more geodesic structure with greater stability, and reorient

certain vectors of growth influencing the greater expansion of the vault.1,2

During construction of the model, it became evident that it would only work effectively if the tension cords

were attached near the edges of the bone. In children, the strongest attachments of the dura mater are also

around the bone margins, suggesting that this may be significant and congruent with the mechanism being

modelled.20 Continuity of dural tension is thus maintained beneath the bone and may affect intercellular

signalling from one side to another. Firmer attachments of dura mater in the centre of adult bone would not

affect the tensegrity principle, but implies a change in that signalling, and may influence the lack of bone

healing capability in the skull after early childhood.

It must be emphasized that this model describes a structural mechanism which may be functioning in living

tissue. It would not work in preserved skulls or cadavers where sutural and dural tissues have lost their elastic

resiliency, and the structure becomes fixed under continuous compression (Figure 8c).
 12

3.2. Balance

The tension and compression components are balanced mechanically throughout the entire structure, which

will optimize automatically so as to remain inherently stable. The various curves of the bone edges, in all three

dimensions, facilitate a separation of the bones through alternate tension attachments between opposing bone

edge convexities (Figure 9a). The attachments on either side naturally settle along the tension line. Conse-

quently, if those attachments are at the peak of each convexity, the bones will be pushed apart in a direction

perpendicular to the tension force, and held there. Directional tensile stresses in the dura mater and collagen

fibre orientations have been found.9,18-20 For example, symmetrical fibre orientations in the temporal regions

were observed to be 6.3 degrees +/- 0.8 degrees in respect to the sagittal suture.18 At a different size scale,

figure 9b demonstrates the same principle in a serrated suture. The serrate sutures increase the surface area

between adjacent bones because of their interlocking projections, but the tension attachments holding the

bones apart, as described above, would also decrease the potential for sutural compression in this model. In

figures 9a and 9b the tension cords are causing the bones to be pushed apart. This is strange behaviour indeed,

considering that tension is generally noted for pulling, and not pushing. It underlines how the non-linear

relationship between stress and strain in prestressed tensegrity and biological structures could be brought

about. Conflicting forces resolve themselves by taking the paths of least resistance, eventually settling into a

stable and balanced state of minimal energy. However, a living organism has a field of force dynamics which

are in a continuous state of flux, so that stability and balance are constantly changing (if that is not a

contradiction in terms).32,47,48

Figure 9. Schematic diagram of part of the cranial vault

which shows how the bones could be kept apart through a
prestressed tensegrity mechanism. (a) - An internal view
showing stylised curves found at sutural edges, and tensile
stresses in the dura mater (dotted lines). These tensile stress-
es create a secondary force (arrows) at their attachments that
pushes the bones apart in a perpendicular direction. (b) - A
serrated suture as found in the sagittal suture, with the same
effect seen at a smaller scale.
 13

Cells of the dura mater respond to brain expansion and influence bone growth, allowing the cranium to

match the spatial requirements of the developing brain, whilst remaining one step ahead and retaining a

certain autonomy.1-6 This position renders the vault more adaptable to other functional requirements, such as

the demands of external musculo-tendinous and fascial attachments.7,21 A tensegrity cranium balances its

stability through all stages of development, by allowing small and incremental changes compatible with the

mechanical demands of all connected structures.

3.3. Energetically efficient

Energetically efficient means it has maximum stability for a given mass of material. The geodesic dome

can enclose a greater volume for minimal surface area, with less material than any other type of structure apart

from a sphere. When the diameter of a sphere doubles, the surface area increases 4 fold and the volume

increases 8 fold. The entire structure of the model neurocranium resembles a sphere-like tensegrity icosahe-

dron, with a dural ‘skin’ under tension and bones enmeshed as an endoskeleton. In mechanical terms, a

prestressed tensegrity structure cannot be anything other than in a balanced state of minimal energy

throughout.35,45

3.4. Integration

In a prestressed tensegrity structure, a change in any one tension or compression element causes the whole

shape to alter and distort, through reciprocal tension, distributing the stresses to all other points of

attachment.29,30,32,35-41 In this model, the occiput is fixed at the condyles whilst the sphenoid exerts an elastic

compression through the spheno-basilar synchondrosis. Apart from this, the frontal, ethmoid, sphenoid,

occiput, temporals and parietals do not make direct contact with each other at any point (‘discontinuous

compression’), and are suspended all around (‘continuous tension’) (figure 8). It has been known for a long

time that cranial base dysmorphology may be fundamental to the aetiology of premature suture closure.1,2 The

cartilage growth plates in the chondrocranium have been shown to respond to mechanical stresses, although

normally the spheno-basilar region is the only one to remain metabolically active for very long after birth, and

remains so until adolescence.6,49,50 The dural sheets connecting across the neurocranium short cut mechanical

stresses from one part to the other1- the falx cerebri/cerebelli linking the ethmoid, frontal, parietals and
 14

occiput; and the tentorium cerebelli linking the sphenoid, temporals and occiput with the falx along the

straight sinus.

The icosahedron has several attributes that are advantageous for modelling biological structures.35,36 A full

account is beyond the scope of this paper, but a few significant points are worth mentioning. It is fully

triangulated, which is the most stable of truss configurations (figure 3a); it comes closest to being spherical,

with the largest volume to surface area ratio of all the regular polyhedrons - making it materially efficient; its

surfaces can be divided equally into smaller triangles and the structure scaled up into higher frequencies -

making it even more energetically efficient;43-45 it provides a link between close-packing in 2 and 3 dimen-

sions; and as a fractal generator, it can polymerize into a sheet, stack in a column or helix, and create complex

patterns and shapes. Fractal analysis is commonly applied to natural structures. Their formal definition is

rather obtuse for the purposes of this paper, but a working definition could be: ‘A shape or pattern which

evolves as it changes, reappearing in a hierarchy of different size scales’. Although the frequencies and

amplitudes of the ‘wave’ curvatures seen at the bone edges in figures 9a and 9b vary, they are both examples

of a fractal nature – with a similar pattern appearing at different size scales.51 Fractals are probably relevant

to linking structural hierarchies throughout the body,2,32,35,36 thus making the icosahedron particularly versa-

tile, because it also gives rise to structures with geodesic and prestressed tensegrity properties.

As the vault bones approximate each other, a sort of hybrid geodesic/prestressed tensegrity structure would

provide the required rigidity for brain protection, but with the facility for micro mobility at the sutures.1,2,15

Prestressed tensegrity in the cranium allows for flexibility during development, and whatever other functions

that patent sutures might serve beyond cranial expansion.4,7,15,21 The cranial base naturally develops a geodesic

structure and provides a platform from which the vault bones could expand, through prestressed tensegrity, to

accommodate brain growth. If the transfer of tensional forces in the dura mater, and the suggested mechanisms

illustrated in figure 9 really do form an essential part of sutural patency, an aberration in this system which

leads to compressive bone contact at any point could be one step towards a rigid geodesic cranium.1,5,15 This

may explain why cartilage sometimes appears in sutural joints.1,14

A local tensional stress generated within the cellular cytoskeleton could transfer to the extracellular matrix

of the dura mater and produce effects on other cells at some distance, with structural rearrangements

throughout the network. Long-distance transfer of mechanical forces between different tissues could contrib-
 15

ute to dural development, and be responsible for spatially orchestrating bone growth and

expansion.3,28,29,30,32,34,47,49,52 Similarly, an ‘aberrant’ tensile stress from elsewhere in the cranium could exert

its effects on sutures some distance away, and contribute to a change in interactions between the dura mater,

bones and brain, ultimately leading to premature synostosis.1,2

CONCLUSION
The tensegrity model is a novel approach to understanding how the cranial vault could retain its stability

without relying on an expansive force from an underlying brain, a situation currently unresolved.1-6 Tensional

forces in the dura mater have the effect of pushing the bones apart, whilst at the same time integrating them

into a single functional unit. Sutural patency depends on the separation of cranial bones throughout normal

development, and the model describes how tension in the dura mater achieves this, and influences sutural

phenotype. Cells of the dura mater respond to brain expansion and influence bone growth, allowing the

cranium to match the spatial requirements of the developing brain, whilst remaining one step ahead and

retaining a certain amount of autonomy. Tensegrity may also be an integrating mechanism in a hierarchical

structure that extends from the cell to the whole organism, with complex 3D patterns the outcome of a

network of interactions which feedback on each other.2,29,30,32,36-40,47,52 This provides a context for this model

and could indicate a new approach to understanding the pathologies seen in the neonate.

One of the most significant aspects of biology is the efficiency with which it packs multiple functions into

minimal space. This presents a conundrum in physical modelling, as any structure will inevitably be limited

in its behaviour if it is incomplete or in isolation. It must be emphasized that much of the supporting evidence

for this model is circumstantial, and more research is needed to verify it, but it is compatible with current

understandings of cranial physiology, and has a contribution to make to a hierarchical systems approach to

whole body biomechanics.

Acknowledgement

I wish to express my sincere appreciation to Nic Woodhead, Chris Stapleton and Andrea Rippe, for their

contributions and thoughts during discussions in the preparation of this paper.

 16

Supplementary material figures 10 and 11

Right parietal

Right squamous
Right sphenoid

Right parietal
frontal occiput

Left parietal

frontal

Falx cerebri

Straight
sinus

sphenoid

Left squamous
ethmoid
Left parietal
sphenoid

Left squamous

Left
mastoid
 17

References

1. Moss ML. The pathogenesis of premature cranial synostosis in man. Acta Anat 1959; 37:351–70.

2. Richtsmeier JT, Aldridge K, Deleon VB, Panchal J, Kane AK, Marsh JL et al; Phenotypic Integration of

Neurocranium and Brain. J Exper Zool 2006;306B:360–78.

3. Henderson JH, Longaker MT, Carter DR. Sutural bone deposition rate and strain magnitude during cranial

development. Bone 2004;34;(2):271-80.

4. Mao JJ. Calvarial development: cells and mechanics. Current Opinion in Orthopaedics 2005;16;(5):331-7.

[Review]

5. Yu JC, McClintock JS, Gannon F, Gao XX, Mobasser JP, Sharawy M. Regional differences of dura

osteoinduction: squamous dura induces osteogenesis, sutural dural induces chondrogenesis and osteogenesis.

Plastic and Reconstructive Surgery 1997;100;(1):23-31.

6. Jeffery N and Spoor F. Brain size and the human cranial base: a prenatal perspective. Am J Phys Anthro-

pol 2002;118:324-40.

7. Sabini RC and Elkowitz DE. Significance of differences in patency among cranial sutures. J Am Ost Assoc

2006;106;(10):600-4.

8. Cotton F, Rozzi FR, Valee B, Pachai C, Hermier M, Guihard-Costa AM, Froment JC. Cranial sutures and

craniometric points detected on MRI. Surg Radiol Anat 2005;27:64-70.

9. McLaughlin E, Zhang Y, Pashley D, Borke J, Yu J. The load-displacement characteristics of neonatal rat

cranial sutures. The Cleft Palate Craniofacial J 2000;37;(6):590-5.

10. Opperman LA. Cranial sutures as intramembranous bone growth sites. Developmental Dynamics

2000;219:472-85. [Review]

11. Mommaerts MY, Staels PFJ, Casselman JW. The faith of a coronal suture grafted onto midline

synostosis inducing dura and deprived from tensile stress. The Cleft Palate Craniofacial J 2001;38;(5):533-7.

12. Persing J, James H, Swanson J, Kattwinkel J. Prevention and management of positional skull deformi-

ties in infants. Pediatrics 2003;112;(1):199-202.

 18

13. Cowan CM, Quarto N, Warren SM, Salim A, Longaker MT. Age related changes in the biomolecular

mechanisms of calvarial osteoblast biology affects Fibroblast Growth Factor-2 signaling and osteogenesis. J

Biol Chem 2006;278;(34):32005-13.

14. Hall BK and Miyake T. All for one and one for all: condensations and the initiation of skeletal

development. Bioessays 2000;22;(2):138-147. [Review]

15. Pritchard JH, Scott JH, Girgis FG. The structure and development of cranial and facial sutures. J Anat

1956;90;(1):73-86.

16. Bradley JP, Levine JP, Blewett C, Krummel T, McCarthy GJ, Longaker MT. Studies in cranial suture

biology: in vitro cranial suture fusion. The cleft palate craniofacial J 1996;33;(2):150-6.

17. Bashkatov AN, Genina EA, Sinichkin YP, Kochubey VI, Lakodina NA, Tuchin VV. Glucose and

mannitol diffusion in human dura mater. Biophys J 2003; 85:3310-8.

18. Jimenez-Hamann MC, Sacks MS, Malinin TI. Quantification of the collagen fibre architecture of human

cranial dura mater. J Anat 1998;192:99-106.

19. Runza M, Pietrabissa R, Mantero S, Albani A, Quaglini V, Contro R. Lumbar dura mater biomechanics:

experimental characterization and scanning electron microscopy observations. Anesth Analg 1999;88:1317.

20. Henderson JH, Nacamuli RP, ZhaoB, Longaker MT, Carter DR. Age-dependent residual tensile strains

are present in the dura mater of rats. J Royal Soc Interface 2005;2:159–167.

21. Balboni AL, Estenson TL, Reidenberg JS, Bergemann AD, Laitman JT. Assessing age-related ossifica-

tion of the petro-occipital fissure: laying the foundation for understanding the clinicopathologies of the cra-

nial base. The Anatomical Record part A 2005;282A:38-48.

22. Greenwald JA, Mehrora BJ. Spector JA,Warren SM. Crisera FE, Fagenholz PJ. et al Regional differenti-

ation of cranial suture-associated dura mater in vivo and in vitro: implications for suture fusion and patency.

J Bone Miner Res 2000;15:2413-30.

23. Ogle RC, Tholpady SS, McGlynn KA, Ogle RA. Regulation of cranial suture morphogenesis. Cells

Tissues Organs 2004;176:54-66.

 19

24. Kallun R and Neilson EG. Epithelial-mesenchymal transition and its implications for fibrosis. J Clin

Invest 2003;112:1176-84.

25. Hay ED. The mesenchymal cell, its role in the embryo, and the remarkable signaling mechanisms that

create it. Developmental Dynamics 2005;233;(3):706-20. [Review]

26. Byron D. Role of the osteoclast in cranial suture waveform patterning. The Anatomical Record part A:

Discoveries in molecular, cellular and evolutionary biology. 2006;288A,(5):552-63.

27. Vij K and Mao JJ Geometry and cell density of rat craniofacial sutures during early postnatal development

and upon in vivo cyclic loading. Bone 2006;38:722-30.

28. Mao JJ. Mechanobiology of craniofacial sutures. J Dent Res 2002;81;(12):810-6.[Review]

29. Ingber DE. Tensegrity I. Cell structure and hierarchical systems biology. J Cell Sci 2003;116:1157-73.

30. Ingber DE. Tensegrity II. How structural networks influence cellular information processing networks. J

Cell Sci 2003;116:1397-408.

31. Sarasa-Renedo AS and Chiquet M. Mechanical signals regulating extracellular matrix gene expression in

fibroblasts. Scand J Med Sci Sports 2005;15;(4):223-30.

32. Ingber DE. Cellular mechanotransduction: putting all the pieces together again. The FASEB J

2006;20:811-27.

33. Kumar S, Maxwell IZ, Heisterkamp A, Polte TR, Lele TP, Salanga M et al Viscoelastic retraction of sin-

gle living stress fibres and its impact on cell shape, cytoskeletal organization, and extracellular matrix me-

chanics. Biophys J 2005;90:3762-73.

34. LeDuc PR and Bellin RB. Nanoscale intracellular organization and functional architecture mediating

cellular behavior. Ann BioMed Eng 2006;34;(1):102-13. [Review]

35. Fuller BB. Synergetics, explorations in the geometry of thinking. Macmillan; 1975.

36. Levin S. The importance of soft tissues for structural support of the body. Spine: state of the art reviews.

Dorman,T. ed. 1995;9;(2). [see ref. 40]

 20

37. Levin S. Putting the shoulder to the wheel: a new biomechanical model for the shoulder girdle. J Biomed

Sci Instrum 1997;33. [see ref.40]

38. Levin S. The tensegrity truss as a model for spine mechanics: biotensegrity. J Mech Med Biol 2002;2:375-

88. [see ref.40]

39. Levin S. A suspensory system for the sacrum in pelvic mechanics: biotensegrity. In: Vleeming A, Mooney

V, Stoeckart R. editors. Chap 15: Movement, stability and lumbopelvic pain. Churchill Livingstone 2007. [see

ref.40]

40. Levin S. www.biotensegrity.com Includes references: 36-39.

41. Snelson K. www.kennethsnelson.net

42. Stamenovi D and Coughlin MF. A prestressed cable network model of the adherent cell cytoskeleton.

Biophysical Journal 2003;84:1328-36.

43. Weinbaum S. Mechanotransduction and flow across the endothelial glycocalyx. PNAS

2003;100;(13):7988-95.

44. Li J, Dao M, Lim CT, Suresh S. Spectrin-level modeling of the cytoskeleton and optical tweezers

stretching of the erythrocyte. Biophys J 2005;88;(5):3707-19.

45. Connelly R & Back A. Mathematics and tensegrity. American Scientist 1998; 86:142-51.

46. Gordon JE. Structures, or why things don’t fall down. Penguin 1978.

47. Ingber DE. Mechanical control of tissue morphogenesis during embryological development. Int J Dev Biol

2006;50:255-66.

48. Van Workum K and Douglas JF. Symmetry, equivalence, and molecular self-assembly. Physical Reviews

2006;73:031502.

49. Wang X and Mao JJ. Chondrocyte proliferation of the cranial base cartilage upon in vivo mechanical

stresses. Dent Res 2002;81;(10):701-5.

50. Tang M and Mao JJ. Matrix and gene expression in the rat cranial base growth plate. Cell Tissue Res

2006;324:467-74.
 21

51. Yu JC, Wright RL, Williamson MA, Braselton JP, Abell ML. A fractal analysis of human cranial sutures.

The cleft palate craniofacial J 2003;40;(4):409-15.

52. Nelson CM, Jean RP, Tan JL, Liu WF, Sniadecki NJ, Spector AA et al (2005) Emergent patterns of growth

controlled by multicellular form and mechanics. PNAS 2005;102;(33):11594-9.