Vous êtes sur la page 1sur 18

This article was downloaded by: [KU Leuven]

On: 21 September 2010


Access details: Access Details: [subscription number 915637588]
Publisher Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-
41 Mortimer Street, London W1T 3JH, UK

British Poultry Science


Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713408216

Eggshell strength: A relationship between the mechanism of failure and


the ultrastructural organisation of the mammillary layer
M. M. Baina
a
Poultry Research Group, Dept. Veterinary Anatomy, University of Glasgow, Bearsden, Glasgow,
Scotland

To cite this Article Bain, M. M.(1992) 'Eggshell strength: A relationship between the mechanism of failure and the
ultrastructural organisation of the mammillary layer', British Poultry Science, 33: 2, 303 — 319
To link to this Article: DOI: 10.1080/00071669208417469
URL: http://dx.doi.org/10.1080/00071669208417469

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
British Poultry Science (1992) 33: 303-319

EGGSHELL STRENGTH: A RELATIONSHIP BETWEEN THE


MECHANISM OF FAILURE AND THE ULTRASTRUCTURAL
ORGANISATION OF THE MAMMILLARY LAYER
M. M. BAIN
Poultry Research Group, Dept. Veterinary Anatomy, University of Glasgow, Garscube
Estate, Bearsden, Glasgow G61 1QH, Scotland
Received for publication 5th January 1991

Abstract 1. Stress analysis using modern computer modelling tech-


Downloaded By: [KU Leuven] At: 14:17 21 September 2010

niques indicated that the shell should fail in tension as a result of the
accumulation of stress at its inner most surface with the characteristic
notched appearance of the mammillary layer providing the ideal crack
initiation site.
2. Practical observations however indicated that crack initiation does
not inevitably lead to catastrophic failure. This partly explains why
previous attempts to predict shell failure from classical engineering
theory underestimated the actual force required to break an egg.
3. Crack initiation should now be denned as localised trauma, charac-
terised by a series of radial cracks which propagate outwards from the
load point at first in a stable manner. As the interaction between the
shell and its loading environment continues, one or more of these
minor cracks become unstable, thus giving rise to catastrophic failure.
4. The fracture toughness of the eggshell is a measure of its resistance
to unstable crack growth and in this respect there appears to be a link
between the latter and the structural organisation of the mammillary
layer.

INTRODUCTION
As the economics of commercial egg production become more critical,
then so too does the relative importance of the downgraded egg. Although the
majority of cracked eggs arise as a result of insult, as one egg collides with
another egg or part of the collecting machinery (Anderson et al., 1970;
Hamilton et al., 1979), it is the evaluation and improvement of shell strength
which has posed the more difficult problem. On average between 6 and 8% of
all eggs produced in the United Kingdom are downgraded annually.
The strength of eggshells has been the subject of several extensive reviews
(Tyler, 1961; Wells, 1968; Voisey and Hunt, 1974; Hamilton, 1982; Hunton,
1985, 1989). In general, two types of tests can be identified: those that
measure shell strength directly and those that measure some indirectly related
303
304 M. M. BAIN
physical property. Traditionally, specific gravity and non-destructive deforma-
tion are used to monitor shell quality in the field. The relationship between
these indirect measures and shell performance, however, remains to be clearly
defined (Thompson and Hamilton, 1986).
Eggshells can generally support large forces, provided these are distri-
buted over their surface (Rehkugler, 1963). However, because contact is
usually with a non-matching surface, the initial contact can be assumed to be
confined to a single point (Voisey and Hunt, 1967). When a force is applied to
any object counteracting internal forces, in terms of stresses and strains, are
generated. According to classical engineering theories, failure should occur
when one or more of the principal stresses attains a critical level. Voisey and
Hunt (1967) concluded that the tensile stresses were at a maximum at the inner
surface of the eggshell (directly beneath the load) and predicted that failure
would occur at this site when the stress reached a critical level. These authors,
nevertheless, found that this approach failed to predict accurately the force
required to break eggshells in subsequent quasi-static compression tests.
Voisey and Hamilton (1975), later suggested that the eggshell flattens
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

following the first initial contact and as a result the load is not confined to a
single point as had previously been implied. In other words, the stress at the
inner surface of the shell is less than that prescribed for any given load.
According to Gordon (1976), however, the most important criteria of failure in
modern engineering applications are not necessarily those associated with how
stiff or strong a material is, but how tough or resistant it is to the propagation
of pre-existing cracks.
It is now generally accepted that most materials contain cracks or flaws,
introduced during manufacture or initiated early in their life. Perhaps the
eggshell could be considered as the end product of a genetically-controlled
manufacturing process, which is subject to modification depending upon an
individual bird's harmony with its environment. For example, the greatest
variation in the ultrastructural organisation of the eggshell has been shown to
take place in the mammillary layer (Bunk and Balloun, 1978; Watt, 1985;
Solomon, 1985, 1990), which is characterised by the presence of many natu-
rally occurring fissures where adjacent mammillae fuse. Such 'intermammillary
spaces' could be regarded as defects as they vary both in depth, size and
distribution (Van Toledo et al., 1982).
Nowadays, fracture mechanics problems deal largely with the influence of
defects on the ability of structures to resist loads. A complete understanding of
the eggshells' reaction to external insult, therefore, requires a knowledge of the
stress distribution and the effect on this of naturally occurring flaws in
ultras tructure.
Sophisticated high precision computer programmes are routinely used by
engineers to analyse the theoretical stress induced in complex multilayered
structures under specific loads. The finite element (FE) method (Ross, 1985;
Gallagher, 1987; Balderes, 1987) is possibly the most general and powerful
technique currently available, and in this paper the theoretical distribution of
stress within the eggshell under load is re-assessed using this technology.
Information from this study is then compared to observations of the behaviour
EGGSHELLS: THE MECHANISM OF FAILURE 305
of eggshells during quasi-static compression tests with particular reference to
pre- and post-fracture behaviour. A new theory is subsequently described in
which the mechanism of failure is related to the ultrastructural organisation of
the mammillary layer.

MATERIALS AND METHODS


Computer modelling
The concept behind the FE method is that a geometric representation of
any complex structure can be made using commercially available computer
graphic programmes which are capable of creating planes and surfaces from
three dimensional coordinates. The resulting geometry is then subdivided into
a number of discrete elements or blocks (the finite elements). This process of
subdivision makes it feasible to manipulate the resulting model in such a way
that specific elements can be subjected to load or their properties changed.
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

The latter information together with experimentally derived data relating to


the thickness and the material properties, or the numbers of layers to be
included in the model, are then translated into a series of numerical codes. The
analysis subsequently determines the theoretical stresses induced throughout
the model, using these specifications.
A FE model representation of a 'standard' shaped egg (see Fig. 1) was
generated with the aid of the commercially available FE mesh generating code
PATRAN, and analysed using the FE code ABAQUS (Hibbitt, Karlsson and
Sorenson Incorporated, 35 South Angell Street, Providence, Rhode Island
02906, USA) available at the time of writing on a VAX 11/750 computer.
The geometric boundaries of the model were defined by assuming that the
shape of an egg can be mathematically described in terms of a sheared
ellipse.
[X+Ftan0+a 2 ] | Y2
a2 b2

Here a is equal to 5 length, b is equal to 5 breadth, tan 0 is equal to


[x—a]/b, (x being the distance between the maximum breadth and the mid-
point of the egg), and X and Y are the cartesian co-ordinates in the X and Y
planes, respectively. Measurements of a, b and x were obtained from a number
of eggs using a shadowgraph machine. The profile of the 'standard' egg-shaped
FE model was based on their mean values. Further details of the modelling
procedure are given elsewhere (Bain, 1990).
Traditionally eggshell strength tests are performed equatorially because
the egg is regarded as being weakest at this point (Voisey and Hunt, 1964). In
order to standardise the sample site it was, therefore, decided that in the egg-
shaped FE models, the load should also be equatorially applied.
The strength or the loading conditions required to induce structural failure,
cannot be directly obtained from FE analysis. The latter can only indicate those
regions of the structure that are most highly stressed under the prescribed
306 M. M. BAIN
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 1.—(a) FE representation of a 'standard' shaped egg (see text). Only quarter of the total
geometry is illustrated, (b) The 'standard' egg shaped model subdivided into 3306 individual
triangular finite elements (type STRI 3). In order to obtain accurate solutions (see Bain, 1990)
the elements have been directed towards the load site (arrowed).

loading conditions. In the current analyses, the principal stresses resulting


from the application of an equatorial point load (L) were determined at three
points through the thickness of the 'standard' egg-shaped FE model, defined in
terms of Teffective and EsheU (after Bain, 1990). A crack of variable length was then
incorporated into the model by creating a discontinuous limiting surface
between varying numbers of elements within the vicinity of the loadpoint
(Fig. 2). Each 'cracked' model was subsequently analysed under point load
conditions as before. The relationship between the applied force, the resulting
deformation, the geometry and the material properties of each cracked
model was then compared to that obtained in the uncracked model described
above.
EGGSHELLS: THE MECHANISM OF FAILURE 307

Quasi-static Compression Tests [QSC]


Each egg was candled to eliminate those with damaged shells. Of the
remaining eggs the position of translucent patches, spots or lines were clearly
outlined in pencil. Each egg was then compressed at a precisely controlled
constant displacement rate of 0-5 cm/min between two flat steel plates (fine
machine finish) by means of a screw-driven tensile testing machine (T20K
Series, J J . Lloyd Instruments Ltd., Brook Avenue, Warsash, Southampton,
England). In each case the egg was placed with its major axis perpendicular to
the compression surfaces, and supported by hand until both loading surfaces
had made contact with the shell. Thereafter, the force exerted on the shell was
detected by a 100N load cell which supported the upper compression surface.
Deformation was recorded by means of a linear variable differential transfor-
mer (adjusted sensitivity 0 to 0-5 mm) mounted rigidly to the upper compres-
sion surface. The position of the latter, however, was adjustable to accommo-
date for different sized eggs. Both the resulting applied force and deformation
were subsequently recorded and displayed throughout the experiment by
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

means of an X Y chart recorder connected to the output terminals of the test


machine. The extent of distortion in the compression equipment under the test
forces used was assumed to be negligible.
In the majority of cases the egg was compressed until failure, as character-
ised by a sharp drop in the force vs deformation curve and a clearly audible
cracking noise. At this point, the cross-head of the machine was immediately
reversed. The area of shell which had been in contact with the upper moving
plate was subsequently marked in pencil.
On a few occasions, the pressure on the egg and the recordings of the
force vs displacement was continued beyond the point of failure in an attempt
to characterise post-fracture behaviour. Alternatively, the cross-head was re-
versed before obvious damage had occurred. The latter non-destructive tests
were used to investigate the possibility of crack initiation preceding cata-
strophic failure.
After each experiment, the eggs were left for several hours to permit
diffusion of moisture from the egg contents into the crack space, and then they
were re-candled. In each case, photographs were taken of the visible damage
within the 'crack initiation zone' and the eggs classified according to crack type,
i.e. multiple or mainline. The direction of the longest mainline crack was
recorded using a numerical rating procedure similar to that described by Hunt
and Voisey (1966). Evidence of damage induced at the opposite loading point
was also noted.

Examination of crack sites by scanning electron microscopy (SEM)


'Crack initiation sites', and damaged areas from the opposite load site were
carefully removed using a diamond-tipped circular drill piece. Because of the
brittle nature of eggshell (Voisey and Hunt, 1974), a large binding surface was
removed along with these areas in an attempt to preserve the experimentally
induced crack lines. Samples were also selected for analysis by scanning
308 M. M. BAIN

crack
(length variable)
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 2.—Finite element idealisation of an equatorially cracked egg. A crack was incorporated
into the model by simply defining a discontinuous boundary or surface between some of the
triangular elements near the load point.

electron microscope at positions around the egg at, and away from, the major
crack line. Each sample was then prepared for SEM.
The bulk of the shell membranes was carefully removed by first soaking in
water. The loosely adhering membranes were then gently peeled from the edge
of the sample inwards. To remove any remaining membrane fibres, each
sample was then subjected to the non-destructive technique of plasma etching
for four hours (Reid, 1983).
Samples from both load sites, which had remained intact following these
preparative treatments, were mounted innerside uppermost on aluminium
stubs, and coated with gold/palladium for 4 minutes in an Emscope sputter
coater. Each sample was then examined using a Philips 50IB SEM at 15 kv for
evidence of internal damage. External damage was also assessed by mounting
unplasmolysed samples on stubs with their cuticular surface outermost. The
latter were coated with gold/palladium as above.
Samples selected at points along the mainline cracks tended to separate
into two halves following plasma treatment. However, the mainline crack could
be reconstructed by careful mounting procedures. These specimens were
examined to determine if the crack direction was influenced by the structural
organisation of the mammillary layer using the criteria described first by Reid
(1984) and later modified by Watt (1985, 1989) and Solomon (1990).

RESULTS AND DISCUSSION


Under point load conditions, the principal stresses at the outer surface of
the 'standard' egg-shaped FE model were found to be compressive in nature,
EGGSHELLS: THE MECHANISM OF FAILURE 309
while the inner surface of the model was found to be in tension. The
distribution of tensional stress at the inner surface of the model is illustrated in
Fig. 3. In both cases the stresses were at a maximum directly beneath the point
at which the load was applied.

LOAD
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 3.—The maximum principal stresses at the inner surface of the 'standard' egg shaped model
are tensile in nature and give rise to a typical star shaped configuration, (only quarter of the total
geometry is illustrated). The numbers on the contour lines represent increasing levels of stress
from 1 to 10.

A typical compression test force vs deformation plot is given in Fig. 4. The


eggshell failed in a typical brittle elastic manner in which the force and
deformation increased linearly up to the point at which failure occurred. Most
brittle materials characteristically fail in tension (Gordon, 1976). These results,
therefore, seem to support the findings of Voisey and Hunt (1967), that failure
is initiated at the inner surface of the eggshell just below where the palisade
columns fuse as a result of the development of tensile stress concentrations.
Post-fracture behaviour of eggshells is illustrated in Fig. 5. After the
ultimate failure point had been reached, the shell re-established a degree of
resistance to the applied load. That the shell did not completely collapse after
the first relaxation suggests that the mechanism of failure in eggshells is
dependent on the resistance of the shell to crack growth rather than by the
attainment of the ultimate tensile stress.
Further loading of the shell was characterised by a second audible crack
and a second relaxation period. This cycle of resistance and failure was
repeated several times, until the eggshell eventually imploded.
The typical type of damage to the shell resulting from compressive loading
is illustrated in Fig. 6. Similar types of shell damage have been recorded under
dynamic forces (Tyler and Moore, 1965) and have also been observed in the
field (Carter, 1970; Belyavin and Boorman, 1982). The major crack lines (m)
were visible to the naked eye. The small radiating cracks within the 'crack
310 M. M. BAIN

o
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

Deformation

FIG. 4.—The eggshell fails in a typical brittle elastic manner in which the force and deformation
increase linearly (a) up to the point at which failure occurs (b). The latter is characterised by a
large relaxation in the force vs deformation relationship (c).

Deformation

FIG. 5.—Post-fracture behaviour of eggshells is characterised by the re-establishment of the


shells' resistance to crack growth (d). This is then followed by a series of secondary failures
(arrowed) and relaxations until the shell eventually implodes (*).

initiation zone' were only visible when the fractured shells were candled.
Candling also revealed the presence of a translucent patch at the opposite load
EGGSHELLS: THE MECHANISM OF FAILURE 311
site (Fig. 7) when the eggs were left to stand for several hours after testing.
This translucent area had not previously existed.
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 6.—Typical type of damage induced to the shell in quasi-static compression tests. A series of
radiating cracks (arrowed) were found within the 'crack initiation zone' and these were accom-
panied usually by one or more major cracks (m).

FIG. 7.—A translucent patch (arrowed) was found on each egg opposite the 'crack initiation
zone'. Similar areas were also found at each load site in those eggs tested non-destructively.

Similar translucent patches were found at each load site in the eggs tested
using non-destructive loads (15N<Force<25N). When these sites were ana-
lysed by SEM, microscopic radiating cracks on the inner and outer surfaces of
the shell were revealed (Figs 8 and 9). This type of damage is typically found in
structures at stress intensities below those normally associated with catastrophic
failure, but is quickly superseded by a more rapid phase of crack growth if the
interactions between the structure and its loading environment continue.
The effect of incorporating a crack of varying lengths into a FE model of
the egg is illustrated in Fig. 10. As the crack expanded beyond a critical length,
the relationship between the applied force, the resulting deformation, the
geometry and the material properties of the model (non-dimensionalised
compliance) increased exponentially. Photographic evidence suggested that for
the eggshell, crack growth became unstable when one or more of the minor
radiating tracks within the vicinity of the loadpoint attained a length of
between 5 and 7 mm. Thereafter, trie latter gave rise to mainline cracking and
hence catastrophic failure.
Occasionally circular cracks similar to those described by Tyler and Moore
(1965) were also found in conjunction with the radial cracks within the 'crack
312 M. M. BAIN

FIG. 8.—SEM revealed the presence of radiating cracks (arrowed) at the inner surface of
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

the eggshell directly beneath the load point in those eggs tested non-destructively (X 90).
Bar=100/zm.

FIG. 9.—Damage induced to the outer cuticular surface of the eggshell as a result of non-
destructive quasi-static compression tests (X90). Bar=100 film.

initiation zone' (Fig. 11). This phenomenon, however, was more pronounced in
those eggs tested beyond the ultimate failure point (see Fig. 5). The fact that
the radial cracks passed through these circular cracks, suggests that the latter
formed sometime after the ultimate failure point had been reached. This is in
agreement with Carter (1970) who proposed that these secondary circular
cracks are caused by bending moments which are set up as the shell flattens to
conform with the loading plate.
EGGSHELLS: THE MECHANISM OF FAILURE 313
80-|
Cnd (r) = 0-754 42-388X + 14-957X*
8 70-

0:0 0-5 \-0 1-5 2-0 2-5


Cracklength <t> [a/R]

FIG. 10.—The effect of crack length (0) on the non-dimensionalised compliance (Cnd) or
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

stiffness characteristics of the 'standard' egg-shaped FE model, (a = Crack length in mm,


R=radius of the egg at its equator in mm).

FIG. 11.—Occasionally circular cracking (arrowed) was also found within the 'crack initiation
zone'.

Thus, it can be speculated that the mechanism of failure in eggshells under


compressive forces, and presumably under dynamic forces, consist of several
phases: upon the application of a load to the surface of the shell, counteracting
internal stresses and strains are generated througout its thickness (Fig. 12).
Stress concentrations however soon build up at the inner surface of the shell,
314 M. M. BAIN

fension
FIG. 12.—The failure mechanism in eggshells begins with the accumulation of tensional stress (s)
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

where adjacent mammillae fuse. A crack then quickly propagates through the shell wall towards
the outer surface (*).
where adjacent mammillae fuse. As a result, a crack quickly propagates through
the shell wall towards the outer surface. A series of radial cracks then form as
the interactions between the shell and its loading environment continue to
increase in magnitude (Figs 7 to 9, and 13). At a critical level of stress one or
more of these radiating cracks becomes unstable and this is accompanied by
the characteristic relaxation in the force deformation curve normally associated
with shell failure (Fig. 4). Post-fracture behaviour of the eggshell is character-
ised by the re-establishment of resistance to crack growth. Thereafter, circular
cracking (Fig. 11) is induced as the shell conforms to the loading plate. This is
then accompanied by a series of secondary failures (Fig. 5) until the shell
eventually collapses.

advancing crack face

FIG. 13.—On reaching the surface this crack then proceeds to grow outwards in a radial fashion,
at first in a stable manner, but as the interactions between the shell and its loading environment
continue, unstable growth of one or more of these radical cracks soon follows.
EGGSHELLS: THE MECHANISM OF FAILURE 315
TABLE 1

The direction and frequency of fractures induced by quasi-static compressive loads. (System
of enumerating taken from Hunt and Voisey, 1966)
Crack direction Description Frequency of occurrence
1 Equator-Blunt pole 88 (25%)
2 45° Equator/Blunt pole 20 (6%)
3 Equatorial Direction 96 (27%)
4 45° Equator/pointed pole 37 (11%)
5 Equator-pointed pole 74 (21%)
x Multiple cracking 35 (10%)
Total Eggs 350

From Table 1, it would appear that it was those radial cracks running
along the major axis of the egg, which most frequently gave rise to unstable
crack growth. Hunt and Voisey (1966) suggested that the direction of crack
growth maybe influenced by the curvature of the shell and variations in shell
thickness, but could find no evidence to support this. Current SEM observa-
tions, however, suggest that the structural organisation of the mammillae
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

within the 'crack initiation zone' plays a role in this scheme of events. For
example, the small radiating cracks within this zone tended to move more
freely between the columns where individual mammillae occurred in alignment
(Fig. 14). In contrast, mainline cracks were found to be essentially unaffected
by the structural arrangement of mammillae. Nevertheless, there was some
evidence to suggest that while confluent areas were avoided (Fig. 15), pitted or
poorly structured areas were favoured by the advancing crack (Figs 16 and 17).

FIG. 14.—Cracks (arrowed) tended to select the path of least resistance within the crack initiation
zone (X180). Bar=100/ttn.

In total, Solomon (1990) has described twelve variations in structure at the


level of the mammillary layer in eggshells from commercial laying flocks. From
what is now known about the mechanism of failure in eggshells, it is possible to
316 M. M. BAIN

TABLE 2

Categorisation of the mammillary layer abnormalities described by Solomon (1990) into those which increase, and those
which decrease an eggshells' fracture toughness. Those remaining are of the basal cap to membrane variety. The latter
have a direct influence on shell thickness (modified from Bain, 1990)
Structural variations which increase the resistance of eggshells to unstable fracture:
early fusion
, cuffing
confluent mammillae
a low mammillary density
Structural variations which decrease the resistance of eggshells to unstable fracture:
late fusion
type B's=open framework
aragonite=open framework
pitting: depressions, erosions, pin holes
alignment of mammillae
a high mammillary density
Others affecting the relationship between the organic and inorganic components of the shell:
changed membrane
cap to cone contact
type A's
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 15.—Confluent areas (arrowed) were avoided by the advancing crack (X 360). Bar= 100 fim.

categorise the latter into those structural variants which increase, or decrease
the shells' resistance (or toughness) to unstable crack growth (Table 2). As a
consequence of this work, research is now being undertaken to establish if the
ultrastructural organisation of the mammillary layer can be manipulated (either
genetically, through diet or by changing management practices), in such a way
as to increase the shells resistance to trauma and subsequent breakage.
EGGSHELLS: THE MECHANISM OF FAILURE 317

FIG. 16.—Pitted areas (arrowed) were vulnerable to cracking (X90). Bar=100 ,


Downloaded By: [KU Leuven] At: 14:17 21 September 2010

FIG. 17.—Poorly structured areas were favoured by the advancing crack (X360). Bar=100 jim.

ACKNOWLEDGEMENTS
The author would like to acknowledge Hibbitt, Karlsson and Sorenson
Incorporated, for the provision of ABAQUS, and the Departments of Veteri-
nary Anatomy and Mechanical Engineering, Glasgow University for technical
assistance. Dr S. E. Solomon and Professor J. W. Hancock are also thanked for
their advice and encouragement. This work was supported financially by the
Department of Agriculture and Fisheries for Scotland.
318 M. M. BAIN
REFERENCES
ANDERSON, G.B., CARTER, T.C. & MORLEY-JONES, R. (1970) Some factors affecting the incidence
of cracks in hen's eggshells. British Poultry Science, 11: 103-116.
BAIN, M.M. (1990) Eggshell strength: a mechanical/ultrastructural evaluation. Ph.D Thesis,
University of Glasgow.
BALDERES, T. (1987) Finite element method, in: PARKER, S. (Ed.), McGraw-Hill Encyclopedia ofScience
and Technology, 6th edn, pp. 113-116 (New York, London, McGraw-Hill Book Company).
BELYAVIN, C.G. & BOORMAN, K.N. (1982) Crack type and severity as indications of shell quality.
Poultry Science, 61: 591-594.
BUNK, M.J. & BALLOUN, S.L. (1978) Ultrastructure of the mammillary region of low puncture
strength avian eggshells. Poultry Science, 57: 639-647.
CARTER, T.C. (1970) Why do eggshells crack? World's Poultry Science Journal, 26: 549-561.
GALLAGHER, R.H. (1987) Structural analysis, finite elements, in: MEYERS, R.A. (Ed.), Encyclopedia
of Physical Science and Technology, Vol. 13, pp. 457-465 (Orlando & London, Academic
Press).
GORDON, J.E. (1976) The New Science of Strong Materials (or Why you Don't Fall Through the Floor),
2nd edn (Middlesex, Penguin Books Ltd).
HAMILTON, R.M.G. (1982) Methods and factors that affect the measurement of eggshell quality.
Poultry Science, 61: 2022-2039.
Downloaded By: [KU Leuven] At: 14:17 21 September 2010

HAMILTON, R.M.G, HOLLANDS, K.G., VOISEY, P.W. & GRUNDER, A.A. (1979) Relationship between
eggshell quality and shell breakage and factors that affect breakage in the field: A review.
World's Poultry Science Journal, 35: 177-190.
HUNT, J.R. & VOISEY, P.W. (1966) Physical properties of eggshells: 1. Relationships of resistance
to compression and force at failure of eggshells. Poultry Science, 45: 1398-1404.
HUNTON, P. (1985) Laboratory evaluations of egg quality, in: WELLS, C.G. & BELYAVIN, C.G.
(Eds), Egg Quality—Current Problems and Recent Advances, pp. 87-102 (London, Butter-
worths).
HUNTON, P. (1989) Egg shell quality: Evaluation criteria for different industry participants. 3rd
European WPSA Symposium on Egg Quality, pp. 336-344 (Stuttgart, Germany, World's
Poultry Science Association).
REHKUGLER, G.E. (1963) Modulus of elasticity and ultimate strength in the hen's egg shell.
Journal of Agriculture, Engineering, Research, 8: 352-354.
REID, J. (1983) The use of the plasma chemistry unit as an aid to the scanning electron
microscope study of avian eggshell structure. British Poultry Science, 24: 233-235.
REID, J. (1984) The effect of diet on certain aspects of the eggshell of the domestic fowl (Gallus
domesticus). Ph.D. Thesis, University of Glasgow.
Ross, C.T.E. (1985) Finite Element Methods in Structural Mechanics, Ellis Horwood
Series—Engineering Science (London, John Wiley & Sons).
SOLOMON, S.E. (1985) Eggshell quality: A structural evaluation. Poultry International, Vol 24, No
11: 58-62.
SOLOMON, S.E. (1990) Egg and Eggshell Quality (London, Wolfe Publications Limited).
THOMPSON, B.K. & HAMILTON, R.M.G. (1986) Relationships between laboratory measures of
eggshell strength and breakage of eggs collected at a commercial grading station. Poultry
Science, 65: 1877-1885.
TYLER, C. (1961) Shell strength: It's measurement and it's relationship to other factors. British
Poultry Science, 2: 3-18.
TYLER, C. & MOORE, D. (1965) Types of damage caused by various cracking and crushing
methods used for measuring eggshell strength. British Poultry Science, 6: 175-182.
VAN TOLEDO, B., PARSONS, A.H. & COMBS, G.F. (1982) Role of ultrastructure in determining
eggshell strength. Poultry Science, 61: 569-572.
VOISEY, P.W. & HAMILTON, R.M.G. (1975) Behaviour of eggshell under compression in relation
to deformation measurements. British Poultry Science, 16: 461-470.
VOISEY, P.W. & HUNT, J.R. (1964) A technique for determining approximate fracture propaga-
tion rates of eggshells. Canadian Journal of Animal Science, 44: 347-350.
VOISEY, P.W. & HUNT, J.R. (1967) Physical properties of eggshells: 4. Stress distribution in the
shell. British Poultry Science, 8: 263-271.
EGGSHELLS: THE MECHANISM OF FAILURE 319

VOISEY, P.W. & HUNT, J.R. (1974) Review paper: Measurement of eggshell strength. Journal of
Texture Studies, 5: 135-182.
WATT, J.M. (1985) An ultrastructural evaluation of eggshell quality in the domestic fowl. M.Sc.
Thesis, University of Glasgow.
WATT, J.M. (1989) The effect of stress on the reproductive tract of the domestic fowl. Ph.D.
Thesis, University of Glasgow.
WELLS, R.G. (1968) The measurement of certain egg quality characteristics: A review, in: CARTER,
T.C. (Ed.), Egg Quality—A Study of the Hen's Egg, pp. 214-225 (Edinburgh, Oliver and
Boyd).
Downloaded By: [KU Leuven] At: 14:17 21 September 2010