Jeanne Lynch-Aird Thesis June 2016

Estimation of Post-Mortem Interval
Using Decomposition Scales for

Hanging Bodies
by
Jeanne Elizabeth Lynch-Aird
A thesis submitted in partial fulfilment for the requirements of

the degree of Doctor of Philosophy at the University of Central
Lancashire
June 2016
STUDENT DECLARATION FORM
Concurrent registration for two or more academic awards
I declare that while registered as a candidate for the research degree, I have not been a
registered candidate or enrolled student for another award of the University or other
academic or professional institution
______________________________________________________________________________________________________
Material submitted for another award
I declare that no material contained in the thesis has been used in any other submission
for an academic award and is solely my own work
Signature of Candidate
______________________________________________________
Type of Award Doctor of Philosophy
School Forensic and Applied Sciences
i
Abstract
The extent of decomposition of a body can be used, in conjunction with
accumulated degree days (ADD), to provide an estimate of the post-mortem
interval (PMI). PMI estimations are important in aiding police to narrow down the
possible identity of a body, and to include or exclude suspects, and also to establish
the order of death for inheritance purposes when two or more potential
beneficiaries die at around the same time. Previous studies have shown the
decomposition pattern in hanging bodies to be different from that of a body on the
ground, but the sample sizes used have been small.
This study presents the results of a series of decomposition studies on hanging
bodies in a variety of situations; clothed and unclothed, and fully or partially
suspended. The study used domestic pigs (Sus scrofa) which enabled large enough
sample sizes for statistical robustness. Pigs lying on the ground were used as
controls. The pattern of decomposition in hanging pigs was found to differ
sufficiently from that of pigs lying on the ground to require the creation of a novel
decomposition scoring scale, which was used successfully to score both clothed
and unclothed fully suspended bodies, as well as the upper, suspended, part of
partially suspended bodies.
The presence of loose, lightweight clothing, which did not impede insect access,
was found to affect both the pattern and rate of decomposition in hanging pigs,
with the clothed bodies decomposing faster than the unclothed bodies (p < 0.05,
F2, 477 = 1238).
ii
The variations in the start weights of the pigs used for these studies was found to
have a statistically significant effect on the rate of decomposition for both the
hanging bodies and those on the ground (p < 0.05, F5, 714 = 1962) but the effect was
so small as to make no practical difference across the range of start weights
encountered. The effect of variation in start weight may be of greater concern,
however, in scoring very heavy, obese, bodies and may be exacerbated by the
increased fat-to-muscle ratios encountered in such bodies.
Finally a set of ADD prediction tables were produced for the hanging and surface
pigs. Further work is needed to establish to what extent these tables can be used
for humans and, in light of the growing obesity problems in humans, to investigate
the effect of weight and increased fat-to-muscle ratios on the pattern and rate of
decomposition.
iii
Table of Contents
Abstract .....................................................................................................................................ii
Table of Contents...................................................................................................................iv
List of Figures ...................................................................................................................... viii
List of Tables...........................................................................................................................xi
Acknowledgements ........................................................................................................... xiii
Abbreviations ...................................................................................................................... xiv
1 Introduction ...................................................................................................................1
1.1 The Research Study ........................................................................................................... 2
1.2 Aims of this Study............................................................................................................... 2
1.3 Outline of Thesis................................................................................................................. 3
2 Literature Review.........................................................................................................5
2.1 Hanging.................................................................................................................................. 5
2.2 Estimating Post-Mortem Interval................................................................................. 9
2.3 Accumulated Degree Days ............................................................................................10
2.4 Decomposition Scoring ..................................................................................................11
2.5 Process and Patterns of Decomposition ..................................................................13
2.6 Effects of Insects ...............................................................................................................14
2.7 Effects of Clothing.............................................................................................................16
2.8 Decomposition of Hanging Bodies .............................................................................19
2.9 The Use of Pigs as Human Analogues........................................................................21
3 General Materials and Methods ........................................................................... 24

3.1 The Field Research Site..................................................................................................24
3.2 Animals ................................................................................................................................25
3.3 Hanging Frames and Surface Enclosures.................................................................26
3.4 Electronic Data Logging .................................................................................................29
3.5 Observations......................................................................................................................30
3.6 Insect Data Collection .....................................................................................................31
3.7 Disposal of Pig Carcasses...............................................................................................31
4 EXPERIMENT 1 – Comparing Decomposition Patterns and Rates

Between Hanging Pigs and Pigs Laid on the Ground .................................... 32
iv
4.1 Materials and Methods...................................................................................................33
4.1.1 Sex and Weight of Pigs............................................................................................................ 35
4.1.2 Measuring Percentage Weight Loss of Hanging Pigs ................................................. 36
4.1.3 Creation of a Scale for Hanging Body Score ................................................................... 37
4.1.4 Statistical Analysis.................................................................................................................... 38
4.2 Results..................................................................................................................................39
4.2.1 General Pattern of Decomposition..................................................................................... 39
4.2.2 Patterns of Decomposition in Male and Female Hanging Pigs............................... 42
4.2.3 Timing of First Beetle Activity............................................................................................. 43
4.3 Total Body Score for Hanging Bodies (TBShang) .....................................................49
4.4 Analysis................................................................................................................................52
4.5 Discussion ...........................................................................................................................55
4.5.1 Decomposition Scoring Scale for Hanging Bodies....................................................... 56
4.5.2 Mummification........................................................................................................................... 57
4.5.3 Initial Weight of the Pigs........................................................................................................ 58
4.5.4 Rate of Decomposition in Hanging and Control Pigs.................................................. 59
4.5.5 Comparison of Weight Loss in the Male and Female Hanging Pigs...................... 60
4.5.6 Beetles ........................................................................................................................................... 61
4.5.7 Internal Temperature Measurements.............................................................................. 63
4.5.8 Effects of Heavy Rainfall......................................................................................................... 63
5 EXPERIMENT 2 – The Effect of Clothing on the Decomposition Pattern

and Rate in Hanging and Surface Pigs................................................................ 64
5.1.1 Sex and Mean Weight of Pigs................................................................................................ 66
5.1.2 Hanging Pigs................................................................................................................................ 66
5.1.3 Clothing the Pigs........................................................................................................................ 67
5.1.4 Data Collection Intervals........................................................................................................ 67
5.1.5 Data Collection ........................................................................................................................... 68
5.1.6 Decomposition Scoring........................................................................................................... 69
5.1.7 Statistical Analysis.................................................................................................................... 69
5.2 Results..................................................................................................................................71
5.2.1 General Pattern of Decomposition..................................................................................... 71
5.2.2 Timing of the First Beetle Activity ..................................................................................... 80
5.3 Analysis................................................................................................................................85
5.4 Discussion ...........................................................................................................................88
5.4.1 Hanging Pigs................................................................................................................................ 89
5.4.2 Surface Pigs ................................................................................................................................. 93
v
6 EXPERIMENT 3 – Comparing the Decomposition of Partially Suspended
(Semi-Recumbent) Pigs with Fully Suspended Hanging pigs and with
Fully Recumbent Pigs in Direct Contact with the Ground........................... 96
6.1.1 Sex and Mean Weight of Pigs................................................................................................ 99
6.1.2 Data Loggers ............................................................................................................................... 99
6.1.3 Data Collection and Collection Intervals ...................................................................... 100
6.1.4 Decomposition Scoring........................................................................................................ 100
6.1.5 Statistical Analysis................................................................................................................. 101
6.2 Results............................................................................................................................... 103
6.2.1 General Pattern of Decomposition.................................................................................. 103
6.3 Statistical Analysis........................................................................................................ 123
6.4 Discussion ........................................................................................................................ 126
6.4.1 Semi-Recumbent Upper Section (SRU)......................................................................... 128
6.4.2 Semi-Recumbent Lower Body (SRL).............................................................................. 131
7 Analysis of the Combined Data Set and ADD Prediction Tables for
Hanging and Surface Bodies ................................................................................134
7.1 Introduction .................................................................................................................... 134
7.2 Method .............................................................................................................................. 135
7.2.1 Statistical Analysis................................................................................................................. 136
7.2.1.1 Data Pre-Treatment and Transformations....................................................................... 136
7.2.1.2 Inter-Year Consistency Test.................................................................................................... 139
7.2.1.3 Effects of Start Weight ............................................................................................................... 139
7.2.1.4 Prediction Tables ......................................................................................................................... 139
7.3 Results............................................................................................................................... 140
7.3.1 Consistency Across Successive Years ............................................................................ 140
7.3.2 Sensitivity of TBS responses ............................................................................................. 141
7.3.2.1 Effect of Increase in Start Weight on Decomposition Scoring ................................. 142
7.3.2.2 Effect of Increase in ADD on Decomposition Scoring.................................................. 142
7.3.3 ADD Prediction Tables......................................................................................................... 143
7.3.3.1 Linear Regression Lines Used for the ADD Prediction Tables................................. 147
7.4 Discussion ........................................................................................................................ 150
7.4.1 Comparison Between Years............................................................................................... 150
7.4.2 Effect of Start Weight on Decomposition ..................................................................... 150
7.4.3 Pig Weight Differences and this Study .......................................................................... 151
7.4.4 Obesity........................................................................................................................................ 151
7.4.5 ADD Prediction Tables......................................................................................................... 153
vi
8 Discussion and Conclusions.................................................................................157
8.1 Areas for Further Work............................................................................................... 166
8.1.1 Validation of PMI Prediction Tables for Use With Humans.................................. 166
8.1.2 Use of Pigs as Human Analogues for Decomposition Studies.............................. 167
8.1.3 Use of Accumulated Degree Days (ADD)...................................................................... 168
8.1.4 Rainfall and Hanging Bodies.............................................................................................. 169
8.1.5 Indoor Versus Outdoor Decomposition........................................................................ 170
8.2 Conclusion........................................................................................................................ 170
9 References .................................................................................................................172
10 Appendices ................................................................................................................186
10.1 Appendix 1: Megyesi et al.’s (2005) Decomposition Scoring Scales (Used for
Scoring Pigs Lying on the Ground).......................................................................... 186
10.2 Appendix 2: Example ADD Estimate Calculations for a Hanging Pig .......... 190
10.3 Appendix 3: Start Weight, Sex, and Age Tables for the Pigs Used in the
Three Experiments ....................................................................................................... 192
10.3.1 Experiment 1............................................................................................................................ 192
10.3.2 Experiment 2............................................................................................................................ 193
10.3.3 Experiment 3............................................................................................................................ 195
11 Published Papers.....................................................................................................196
vii
List of Figures
Figure Page #
3.1 Diagrammatic representation of the scaffolding A-frames used 26

for the hanging pigs showing the bird proof netting, chicken
wire and approximate dimensions. The entry point was via a
section of detachable wire in the centre of one end. Some
sections of wire and netting have been left out for clarity.
3.2 Hanging pigs in one of the scaffolding A-frames at the start of 27

the experiment.
3.3 The scavenger proof cages used to cover the surface pigs. The 28
cage in the foreground shows one of the clothed pigs from
Experiment 2.
4.1 Insertion of thermocouple for attachment to temperature data 34

logger.
4.2 Decomposition sequence for one hanging (H6, female) and one 44
control pig (C9, male).
4.3 Showing the ribs in the torso of a hanging pig viewed through 48
the breached axilla at 591 ADD.
4.4 Hanging male at 512 ADD. The arrow points to the anal 48
opening extending beneath the body.
4.5 Hanging female at 600 ADD. Arrow A points to the pouch of 48

skin hanging below the anal opening (arrow B).
4.6 Hanging female pig at the close of the experiment. The pouch of 48
skin was cut open and the arrow points towards the bones
which had collected in the pouch.
4.7 Differences in percentage weight loss between male and female 53

hanging pigs to 900 ADD. The data points show the values for
each sex at each value of ADD. The model lines are the
polynomial regression (here fourth order) fitted values. Some
jitter has been added to the ADD values of the data points for
clarity.
4.8 Rate of decomposition showing the difference between the 54

male and female pigs using Total Body Scores for Hanging
(TBShang) to 900 ADD. Jitter has been added to ADD for clarity.
viii
Figure Page #
4.9 Comparison of decomposition rates of hanging and grounded 55

pigs using Total Body Scores for hanging (TBShang) and surface
(TBSsurf) pigs, respectively, to 900 ADD. Jitter has been added to
ADD for clarity.
5.1 Data loggers sealed in 50 ml conical centrifuge tubes with wires 67

for use with the hanging pigs. The wires were attached to the
hanging ropes once the loggers had been inserted into the pigs.
5.2 Group pairings used for TBS comparisons: a) the S/C and H/U 70
groups were compared to the S/U group; b) the S/C and H/U
groups were compared to the H/C group; c) the latter enabled
the H/C group to be compared to the H/U group, and also to the
S/C group.
5.3 Decomposition sequence for one H/C (HC1, male) and one H/U 81
pig (H1, female).
5.4 Decomposition sequence for one S/C (CC3, female) and one S/U 83
pig (C6, male).
5.5 Differences in the rate of decomposition of the S/C and H/U 87

groups, compared to the S/U group as the reference. The
results were derived from 10,000 simulations per group based
upon linear regression models. Some jitter has been added to
the ADD values at the data points to add clarity and the error
bars show 95% confidence intervals.
5.6 Differences in the rate of decomposition of the S/C and H/U 88

groups, compared to the H/C group as the reference. The
results were derived from 10,000 simulations per group based
upon linear regression models. Some jitter has been added to
the ADD values at the data points to add clarity and the error
bars show 95% confidence intervals.
6.1 The position of a semi-recumbent pig, supported from a 98

scaffolding A-frame.
6.2 Decomposition sequence for one hanging, one semi-recumbent, 114

and one control pig.
6.3 One of the semi-recumbent pigs showed a large mass of 120

Necrodes littoralis (Silphidae) beetle larvae, these were present
on both sides of the pig, at 674 ADD.
6.4 Bones could be seen adhering to the mummified skin of the 121
hanging pigs at 1078 ADD.
ix
Figure Page #
6.5 Bones could be seen adhering to the mummified skin of the 121
hanging pigs at 1078 ADD.
6.6 The mummified upper torso of the semi-recumbent pigs had 122
separated, and could be lifted away from, the decomposed
lower torso (1078 ADD).
6.7 Total Body Scores for Hanging (TBShang) for the upper sections 123
of the semi-recumbent pigs and the hanging pigs scored using
the hanging scoring scale from Experiment 1. Some offset has
been added to the ADD values at the data points to add clarity.
6.8 Partial Body Scores (PBST + PBSL), lower limbs and torso, for 124
the lower sections of the semi-recumbent pigs, and the limbs
and torso of the control pigs scored using the modified Megyesi
et al. (2005) scoring scale. Some offset has been added to the
ADD values at the data points to add clarity.
6.9 The mummified upper torso of the semi-recumbent pigs had 129
separated, and could be lifted away from, the decomposed
lower torso (1078 ADD).
7.1 Total Body Score plotted against Accumulated Degree Days for 140
all the control pigs for years 2011, 2012 and 2013.
7.2 Total Body Score plotted against Accumulated Degree Days for 141
all the hanging pigs in years 2011, 2012 and 2013.
7.3 Total Body Score TBSsurf against Accumulated Degree Days 146
(ADD), for the control pigs, showing estimates and inverse
prediction intervals from the linear model overlaid.
7.4 Total Body Score TBShang, of hanging pigs against Accumulated 147
Degree Days (ADD) showing estimates and inverse prediction
intervals from the linear model overlaid. The kinks in the
curves result from applying an offset to the TBS values as
described in the Method section.
x
List of Tables
Table Page #
4.1 Stages of hanging decomposition for the head and neck. 50
4.2 Stages of hanging decomposition for the torso. 51
4.3 Stages of hanging decomposition for the limbs. 52
6.1 Sensitivities of the TBS responses to changes in the start weight 125
and ADD around a reference point of 282 ADD and a start weight
of 54.0 kg. The Semi-Recumbent Lower and Control Lower have
partial body scores (PBS) out of a possible 20 points using the
amended Megyesi et al. (2005) scoring scale. The Semi-
Recumbent Upper and Hanging are TBS scored on the Hanging
scale from a possible 30 points.
7.1 Sensitivities of the TBS responses to changes in the start weight 143
and ADD around a reference point of 282 ADD and a start weight
of 40.4 kg.
7.2 Showing the estimated Accumulated Degree Days (ADD) in °C 144

and prediction confidence intervals for Total Body Scores,
TBSsurf, for the control pigs, on the surface, using the statistical
analysis described in the Method section above. Note: the greyed
out areas show negative TBS values which could not occur in
practice but which are predicted from the data ‘spread’ and have
been included purely for completion of the table.
7.3 Showing the estimated Accumulated Degree Days (ADD) in °C 145

and prediction confidence intervals for Total Body Scores,
TBShang, for the hanging pigs using the statistical analysis
described in the Method section above. Note: the greyed out
areas show negative TBS values which could not occur in
practice but which are predicted from the data ‘spread’ and have
been included purely for completion of the table.
8.1 Average monthly rainfall data (mm) recorded at the nearest 165
available Meteorological Office weather station (Bingley 2,
53.81° N, 1.87° W, 262 m) during the three years of the study.
10.1 Megyesi et al.’s (2005) adjusted decomposition scoring scale for 187
the head and neck.
xi
Table Page #
the torso.
the limbs.
10.4 The start weights and sex of the pigs used in Experiment 1. 192
xii
Acknowledgements
I would like to thank my supervisory team: Prof. Tal Simmons, Dr. Colin Moffatt,
Dr. William Goodwin, and Dr. Vicki Cummings for all their support and guidance
over the course of this study.
In particular I would like to thank Dr. Colin Moffatt for his unending patience
during my struggles to get to grips with R, and for his suggestions and
programming fixes.
Thank you to Peter Cross for lifting and holding my pigs whenever it was needed,
regardless of their state of decomposition.
Thank you to Dr. Viv Heaton for her support, coffee, and for forging the way
through the Ph.D. process ahead of me.
Special thanks go to my husband, Dr. Nicolas Lynch-Aird, for his support, supplying
endless cups of strong coffee, and fixing everything from my dying computers to
the endless spelling mistakes and proof reading. Chocolate is owed.
xiii
Abbreviations
ADD Accumulated degree days
ADH Accumulated degree hours
ANOVA Analysis of Variance
ARF Anthropological Research Facility (Tennessee, USA)
ATP Adenosine Tri-Phosphate
BMI Body Mass Index
CRISPR Clustered Regularly Interspaced Short Palindromic Repeats
CPSC Consumer Product Safety Commission (USA)
DEFRA Department for Environment, Food & Rural Affairs (UK)
PBS Partial Body Score
PBSH Partial Body Score for the Head
PBSHhang Partial Body Score for the Head of a hanging body
PBSHsurf Partial Body Score for the Head of a body on the surface/ground
PBSL Partial Body Score for the Limbs
PBSLhang Partial Body Score for the Limbs of a hanging body
PBSLsurf Partial Body Score for the Limbs of a body on the surface/ground
PBST Partial Body Score for the Torso
PBSThang Partial Body Score for the Torso of a hanging body
PBSTsurf Partial Body Score for the Torso of a body on the surface/ground
PMI Post-Mortem Interval
RoSPA The Royal Society for the Prevention of Accidents (UK)
SRL Semi-Recumbent Lower
SRU Semi-Recumbent Upper
xiv
TBS Total Body Score
TBShang Total Body Score for a hanging body
TBSsurf Total Body Score for a body on the surface/ground
TRACES Taphonomic Research in Anthropology: Centre for Experimental
Studies (University of Central Lancashire, UK)
WHO World Health Organization
xv
1 Introduction
It is estimated that globally by 2020 there will be a death by suicide every 20
seconds, or 4320 suicides daily (Dogan et al., 2015). Hanging is the most common
method of suicide. Hanging deaths also occur as the result of accidents, including in
autoerotic activities, where there is a malfunction of the fail-safe mechanism, and
the solitary and secretive nature of these activities means that bodies are not
always found immediately and identification may be difficult.
Post-mortem interval (PMI) estimations are important in aiding police to narrow
down the possible identity of a body, and to include or exclude suspects (Moffatt et
al., 2016), and also to establish the order of death for inheritance purposes when
two or more potential beneficiaries die at around the same time. One method of
estimating post-mortem interval uses decomposition scores, to assess the degree
of decomposition of the body, in conjunction with accumulated degree days (ADD)
to provide a timeline. Studies have shown that patterns of decomposition vary in
different situations, thus one decomposition scoring scale will not suffice; there is
no “one size fits all” decomposition model (Parks, 2011). Various decomposition
scoring scales have been produced including those by Megyesi et al. (2005) for
bodies on a surface, Heaton et al. (2010) for submerged bodies, and Gruenthal et al.
(2012) for charred bodies. Each of these scales reflects the decomposition patterns
specific to the situation.
1
1.1 The Research Study
Studies have shown the decomposition pattern in hanging to be different from that
of a body on the ground. Prior to this present study, however, there have been no
decomposition studies of hanging bodies with large enough sample sizes (Chong
Chin et al., 2010; Shalaby et al., 2000) to determine if a separate decomposition
scoring scale, specific to hanging bodies, would provide greater accuracy in
estimating PMI.
This study attempted to fill this gap through a series of decomposition studies on
hanging bodies in a variety of situations; clothed and unclothed, and fully and
partially suspended. The study used domestic pigs (Sus scrofa) to enable large
enough sample sizes for statistical robustness. Pigs are the closest human
analogues for this kind of study, and the results produced can be compared with
known human hangings to assess if they are useful enough to be applied in
estimating the PMI for hanged human bodies.
1.2 Aims of this Study
The principal aims of this study were to further the understanding of
decomposition in hanging bodies and to investigate how the patterns and rates of
decomposition may be used to more accurately estimate time since death.
A series of field studies using pigs in different hanging conditions (with pigs on the
ground as controls) were carried out with the following aims:
2
• To investigate whether there were differences in the rates and patterns of
decomposition between the hanging and control pigs.
• To produce an amended or new decomposition scoring scale for hanging
pigs if necessary.
• To investigate whether the sex of the pig affected the rates and patterns of
decomposition.
• To investigate if there were significant differences in the decomposition
rate and pattern for clothed hanging bodies.
• To investigate what difference it makes to the decomposition process if a
body is partially suspended.
The specific objectives of the various field studies are given in the introduction to
each experiment.
1.3 Outline of Thesis
This introductory section is followed by a review of the literature relating to
hanging; the process of decomposition and decomposition patterns; the
importance of accurate post-mortem interval estimation, and the use of
decomposition scoring and accumulated degree days in making these estimations;
the effect of insects and clothing on decomposition; and the use of pigs as human
analogues.
3
Chapter 3, to avoid repetition in each experiment, describes the materials and
methods that were common to all the experiments and the experimental design,
including the use of hanging frames, scavenger proof cages, and data loggers for
temperature collection. The reader may wish to refer back to this chapter when
reading the materials and methods sections of the chapters describing the field
experiments.
Chapters 4, 5, and 6 describe the field experiments, with each chapter following the
same format of introduction; materials and methods, describing any amendments
or specific changes from the general approach that were needed for that
experiment; results; and a discussion section specific to that experiment.
Chapter 7 contains analyses of the combined data on hanging and control pigs
longitudinally across the three years of the field experiments. This includes the
effect of start weight on the decomposition score to ADD relationship, and the
consistency of the decomposition score to ADD relationship over the period.
Finally, Chapter 8 provides a general discussion of the research with conclusions
and suggestions for future studies.
4
2 Literature Review
2.1 Hanging
World wide, hanging is one of the most commonly used ways of committing suicide
(Thomas and Gunnell, 2010), and in many countries it is the leading method. It is
the most common method in England and Wales, accounting for about 2000 deaths
a year (Bennewith et al., 2005; Gunnell et al., 2005), with a clear pattern showing
the number of men to be consistently higher than women by a ratio of
approximately 5 to 1 (Thomas and Gunnell, 2010). To exclude foul play the
circumstances of a hanging should be carefully examined, particularly in the case
of females where this is an uncommon method of suicide (Bowen; 1982). The
frequency of use of hanging and its reported increase in incidence may be the
result of a mix of factors. It is seen as easy, inexpensive, readily available and
reliable, while being clean and avoiding disfigurement (Biddle et al., 2012; Thomas
and Gunnell, 2010; Thomas et al., 2013). Possibly one of the biggest factors
resulting in its prominence in suicide is its success rate. Hanging is effective; the
case fatality rate for hanging has been reported as 70% (Gunnell et al., 2005) and
between 69% and 84% in an American study from 2000 to 2010 (Baker et al.,
2013). This is probably due to the mechanism of death in hanging.
Hanging is a form of ligature strangulation in which the force applied to the neck is
derived from the gravitational pull of the weight of the body or part of the body
(Ferris, 2000; Turvey, 2000). A weight of 2 kg is all that is needed to block the
jugular vein and venous return to the heart (Gunnell et al., 2005; Sikary et al.,
5
2016; Turvey, 2000). The body does not need to be fully suspended for death to
occur (Bennewith et al., 2005; Ferris, 2000) and may be found with the feet
touching the ground. In a study of 162 deaths by hanging, occurring in 24 English
Coroners’ jurisdictions within a six month period, nearly 50% of hanging bodies
were found to be not fully suspended, and had feet touching the ground or were
kneeling or semi-recumbent (Bennewith et al., 2005). Gunnell et al. (2005) looked
at 2251 hangings from a review of 15 suicide studies, covering the period 1964-94
and eight countries, noting that around 50% of hangings were not fully suspended
and had the ligature attachment below head height.
Although most deaths from hanging are the result of suicide (Ferris, 2000; Sikary
et al., 2016), this is not always the case; they can also be the result of homicide or
accident. Ferris (2000) states that “Homicidal hanging, with the relatively rare
exception of lynching, is very rare.” Accidental hanging may happen with small
children, particularly toddlers, as their heads weigh proportionately more than
their bodies compared to adults, and their muscular control and trachea are not
fully developed, so death occurs more quickly if the neck is constricted (RoSPA,
2015). Children’s cots are often near windows and blinds where they can become
entangled in the blind cords, resulting in approximately one death a month in the
USA (CPSC, n.d.). Children may also become caught up in ropes, their own clothing
or cot attachments, although this is relatively uncommon (Bowen, 1982; Ferris,
2000). Children and young adults may also die as the result of a ‘choking game’
where they induce non-sexual euphoria through cerebral hypoxia (Byard and
Winskog, 2012). Anthropologists describe Eskimo children and South American
Yahgans inducing partial strangulation for exhilaration (Ueno et al., 2003).
6
It should be noted that children do commit suicide. It is the 2nd or 3rd most
common cause of death in children and adolescents worldwide, and hanging is the
most common method used (Mendes et al., 2015; Pakis et al., 2010). A study of
child suicide in Portugal (Mendes et al., 2015) noted that, whilst child suicide was
rare, hanging was the most common method employed. Austin et al. (2011), in two
studies over two and five years, found that while overall suicide rates in
adolescents were decreasing, the incidence of hanging was increasing.
Accidental death in autoerotic practices results when there is a failure in the safety
release mechanism which usually forms part of the routines. This is usually a
secretive and solitary activity (Ueno et al., 2003) practiced in secure and secluded
locations (Byard and Winskog, 2012), which may lead to a delay in discovery
(Komar et al., 1999) – most outdoor hangings occur in rural woodland. There are
an estimated 500 to 1000 autoerotic deaths per annum in the U.S.A., and this figure
is believed to be conservative (Sauvageau and Racette, 2006; Turvey, 2000).
Relatives will often remove any potentially embarrassing items, such as
pornography, before calling the police (Byard and Winskog, 2012; Ueno et al.,
2003). Autoerotic hangings occur almost exclusively in men (Bowen, 1982; Byard
and Winskog, 2012; Ueno et al., 2003). However, certain criteria must be fulfilled
before the hanging is classed as autoerotic, and this may lead to an
underestimation of female deaths as they don’t fit the criteria used, and are usually
naked and without elaborate props (Byard and Winskog, 2012; Turvey, 2000;
Ueno et al., 2003). Most deaths from autoerotic asphyxia are the result of hanging
or the use of a ligature (Sauvageau and Racette, 2006).
7
Among countries for which data are available, hanging is known to be the most
common method of committing suicide in Galicia (in northwest Spain), Saudi
Arabia, Hungary, Belgium, Norway, England and Wales, Turkey, Lithuania,
Germany, Japan (Ambade et al., 2015), and Australia (Austin et al., 2011), and is
the second most frequently used method in Portugal, Sri Lanka, and the U.S.A. In
the case of the U.S.A., firearms remain the most used method except, paradoxically,
in military bases or onboard ship where hanging is most common as personnel do
not have access to firearms unless they are in training or on guard duty (Clark and
Kerr, 1986). Hanging is also one of the most common suicide methods in India
(Ambade et al., 2015) and China (Sun et al., 2012) although it is not the primary
method.
One of the reasons hanging is so ‘easy’ to carry out is the ubiquity of the items
needed. Whilst the methods used to commit suicide have changed over time, a
recent study by Thomas et al. (2013) concluded that this was due to the physical
availability of the method. This is borne out in a study of physician suicide (Austin
et al., 2013) in which 88.9% of the cases examined (8 out of 9) died from lethal
drug self administration, contrasting with the general public.
Much has been made of the influence of the internet and social media in possible
copycat suicides, but there is no rigorous empirical evidence of the impact of the
internet (Biddle et al., 2012). However, whilst steps could be taken to minimize the
availability of ligatures and suspension points in institutions such as hospitals and
prisons (Baker et al., 2013; Bennewith et al., 2005; Gunnell et al., 2005), nothing
can be done to limit access to these in the general community. In a bid to prevent
8
copycat events the reporting of suicides in public places may be restricted by local
authorities (Dogan et al., 2015).
Finally, it is estimated that by 2020 there will be one suicide globally every 20
seconds (Dogan et al., 2015). Clearly, as hanging is the most common form of
suicide worldwide, it can be expected that the incidence of hangings will only
increase.
2.2 Estimating Post-Mortem Interval
For forensic anthropologists and the police, the estimation of post-mortem interval
(PMI) is an important step in the investigation of any death and the identification
of unknown bodies, narrowing down the pool of missing people to whom the
individual body could belong, and providing a time frame from which suspects may
be excluded or included (Moffatt et al., 2016).
The estimation of how long a person has been dead is far from simple, the progress
of decomposition being a complex process of inextricably interrelated variables
(Mann et al., 1990). Whilst the process of decomposition has become no less
complex, the understanding of its progress has improved. There are many methods
of estimating PMI, including entomology, biochemical markers, cadavaric signs and
putrefaction, but the determination of PMI remains a difficult and complex task
(Tracqui; 2000). The process of decomposition in humans is now well understood
and one method by which the PMI can now be estimated is to use decomposition
scoring in conjunction with the accumulated average temperature (Megyesi et al.,
2005).
9
2.3 Accumulated Degree Days
Accumulated Degree Days (ADD) is measured by taking the cumulative average
daily temperature from the time of death to the time of final measurement. Thus at
an average daily temperature of 10 °C it would take 10 days to reach 100 ADD. The
same ADD figure would result from 5 days at 20 °C. ADD signifies the accumulation
of thermal energy. When a given amount of thermal energy is put into a carcass the
same amount of reaction should occur (Simmons et al., 2010a), such that the same
body score for decomposition will be seen. Current work indicates that this may
not be quite as straightforward as previously accepted, and adjustments for season
of death may need to be factored in (Bates and Wescott, 2016; Dautartas et al.,
2016; Simmons et al., 2016).
Because the use of ADD automatically incorporates adjustments for yearly,
seasonal, and within season variation, it has many advantages over a model based
on a count of days alone. The use of ADD allows for prediction models to be
developed (Michaud and Moreau, 2011). Vass et al. (1992) predict human
skeletonisation will have occurred by 1285 ±10 ADD based on the absence of
volatile fatty acids, produced from the breakdown of tissues during decomposition,
in the soil below a body. To estimate PMI, the ADD is calculated using temperature
readings from the nearest weather stations. This is not without problems as the
nearest weather station may not reflect the temperature at the site (Dabbs, 2010,
2015). To compensate for this, data loggers are left for a number of days at the site
where the body was found. These readings are then compared with those from the
weather stations so that a correction factor can be applied to the weather station
data recorded prior to the body being found.
10
2.4 Decomposition Scoring
Using systematic observation of decomposition rates and patterns of bodies within
a given environment, the patterns of decomposition can be seriated to allow the
production of a scale of decomposition descriptions specific to that environment
Decomposition scoring is used to give a value to each milestone of the
decomposition process. This score then forms part of the calculations used to
estimate PMI. Discolouration, bloating, liquefaction, and advanced decay or
skeletonisation are all decomposition phases following a coded timeline, and these
phases are markers for indicating the elapsed time since death (Campobasso et al.,
2001). However, the process of decomposition is not made up of discrete stages,
but is a continuous process with the stages merging and overlapping. Thus, the
determination of decomposition stages is a somewhat subjective decision,
depending on the perception of the observer (Michaud and Moreau, 2011), with
different people defining many decomposition stages. Mégnin (1894) at the end of
the 19th century, produced possibly the first table describing eight waves of
arthropod invasion on human cadavers, mostly using bodies from inside closed
rooms, linked to the decomposition changes of the corpse and noted that these
arthropods may differ with season and site. More recently Adlam and Simmons
(2007) and Oliviera and Vasconcelos (2010) described four stages of
decomposition: fresh, bloat, decay, and dry or skeletonised. Payne (1965), in his
study of summer carrion of the baby pig, had two separate sets of stages in his
study; using six stages where insects had access: fresh, bloated, active decay,
advanced decay, dry, and remains; and five stages for the insect-free pigs: fresh,
bloating and decomposition, flaccidity and dehydration, mummy stage, and
11
desiccation and disintegration. Archer (2004), whilst acknowledging that the
stages observed were similar to those seen by Payne, produced five different stage
descriptors.
Various factors affect the decomposition process which can differ from body to
body, from environment to environment, and even from one part of the same
corpse to another (Campobasso et al., 2001). Differential decomposition within a
body was noted in the scale produced by Megyesi et al. (2005), which allows for
this by dividing the body into three areas and scoring these areas separately. The
Partial Body Scores (PBS) for each region are added together to give a Total Body
Score (TBS) for the observed decomposition.
Parks (2011) noted that a “one size fits all” decomposition model is unrealistic.
This can be extended to include the manner of death. A scale for scoring drowned
bodies was produced by Heaton et al. (2010), who noted in their study that the
decomposition pattern for bodies in water was different from that described by
Megyesi et al. (2005), who did not include any bodies recovered from water. A
scale for scoring charred bodies was produced by Gruenthal et al. (2012) to
encompass the different patterns found in burnt bodies.
When scoring the state of decomposition of a body, or body region, the observer
must select the scale stage that best fits the largest number of scoring criteria. The
scoring scale that best matches the environmental circumstances of a given body
should, in turn, provide the most relevant and appropriate scoring criteria from
which to choose.
12
2.5 Process and Patterns of Decomposition
Decomposition starts within minutes of death as the body’s cells start to self-
destruct by autolysis, the result of enzymatic self digestion caused as the cells
become anoxic. This occurs first in the cells that are most metabolically active with
high rates of Adenosine Tri-Phosphate (ATP) production such as the liver,
intestines, stomach and digestive organs (Gill-King, 1997; Vass, 2001). Algor mortis
occurs as the body temperature changes to match the ambient temperature. This
usually involves the body cooling, but in hot environments the body’s temperature
may increase. Pallor mortis or paling of the skin occurs immediately after death,
particularly in white or pale skins, as gravity causes the still fluid blood to pool to
the lowest point leading to livor mortis (lividity). Initially this discolouration is not
fixed and pressure will result in a white area which then re-fills with blood when
the pressure is removed. As the blood solidifies and clots the discolouration
remains and pale or white areas are left within the lividity where pressure has
been applied. This can be important forensically in showing the position of the
body after death, and if it has been moved since death; this may also give rise to the
possibility that the body has been moved from one geographical location to
another which could make ADD calculations less accurate and thus affect the PMI
estimation.
Rigor mortis sets in usually within 2 to 6 hours of death as loss of ATP results in
muscular stiffening, and cellular cytoplasm gelling (Gill-King, 1997; Vass, 2001).
The rigidity wears off, usually within 18 hours, as decomposition causes the
muscle structure to break down. The process of rigor mortis is sensitive to
temperature, progressing rapidly with hot temperatures and potentially taking
13
days when temperatures are below 4 °C. It has also been noted (Tracqui, 2000)
that the onset of rigidity may be delayed in hanging.
As the body moves into anaerobic decomposition, or putrefaction, the
carbohydrates, proteins and fats break down to produce alcohol and toxic gases
including ammonia, hydrogen sulphide, methane, pyruvic acid, cadaverine and
putriscine (Gill-King, 1997). This results in bloating of the body. The characteristic
smell is caused primarily by the cadaverine and putriscine. As bilirubin and
biliverdin from the liver break down in the caecum in the lower right section of the
abdomen, which is high in anaerobic bacteria, this produces hydrogen sulphide
which reacts with the haemoglobin to form sulphaemaglobin. As this and the
oxidation of bile pigments continue, there is a gradual change in colour to green
moving through purple to black. This discoloration in the superficial vessels causes
the marbling often seen on the abdomen. Gas is produced in the intestinal tract
and, in males, gas from the peritoneal space may be pushed through the inguinal
canal into the scrotum, causing swelling.
The contents of the intestinal tract and the lining itself break down to form purge
fluid which, with the gas in the intestines, purges from the mouth, nose and
rectum. The quantity of fluid purged is not great and after purging active decay
begins and insect activity is the main destructive influence.
2.6 Effects of Insects
Previous studies have demonstrated that the two factors most influencing
decomposition are temperature and insect activity (Cross and Simmons, 2010;
14
Simmons et al., 2010a). Bodies left exposed for any length of time become more
vulnerable to consumption by scavengers. The primary scavengers fall into the
groups of insects, birds, rodents, plus larger carnivores and, in water, aquatic
animals (Rodriguez, 1997). All of these have an impact on the decomposition rate
(Payne, 1965; Simmons et al., 2010a). Particularly in the early stages of
decomposition, the processes, including insect activity, are dependent on many
environmental conditions including temperature, humidity, and the specific
scavengers present.
As bodies decompose the gut bacteria produce large volumes of gas and the release
of this gas attracts flies, particularly some blow flies (Diptera: Calliphoridae), which
begin laying eggs on exposed areas (Campobasso et al., 2001). After the eggs hatch
and the maggot masses increase in size, their own metabolic activity causes an
increase in the internal temperature of the animal. This can fluctuate between 5 to
10 °C above the ambient temperature (Heaton et al., 2014; Simmons et al., 2010b).
Temperature has an effect on the speed of biochemical processes and on insect
activity; it is the most important variable influencing maggot development
(Campobasso et al., 2001). Increase in temperature increases the rate of
decomposition such that time between death and skeletonisation is much shorter
in tropical than temperate climates (Oliviera and Vasconcelos, 2010). Conversely
very low temperatures can inhibit or stop decomposition where maggot masses
have not become established within the body where they can remain active
(Heaton et al., 2014; Huntington et al., 2007).
15
Forensic entomologists use insect data such as developmental stage and species
succession to calculate the estimated PMI. However, the use of insects to estimate
PMI is subject to many potential sources of error (Moffatt et al., 2016). Maggots
migrate away from the body at times specific to their species, thus the oldest
insects may not be on the body (Anderson, 2011). Minimum PMI can be estimated
from the insects found but the longer the PMI, the less precise the PMI estimate
becomes (Amendt et al., 2007). Temperature data obtained from the nearest
weather stations may not reflect the actual microclimate (Dabbs, 2010, 2015),
which can be particularly important if the body is inside a building (Amendt et al.,
2007).
2.7 Effects of Clothing
While many decomposition studies are carried out on naked bodies, making
observation easier, clothing is commonly present on the bodies of forensic cases
(Dillon and Anderson, 1995; Komar, 1998). This is also true of hangings, where
bodies may be found fully clothed (Clark and Kerr, 1986; Hejna and Bohnert, 2013)
or, as most autoerotic deaths are due to hanging, partially clothed, usually in
female underwear (Byard and Winskog, 2012), or naked. Insect activity is
important in the process of decomposition and clothing may hamper arthropod
access to the body, delaying oviposition and maggot activity. It is important to
understand the effects that clothing may have on the rate and pattern of
decomposition, as these could affect the techniques used for PMI estimation.
Experiments and studies on the effect that clothing and wrapping bodies has on
oviposition and the rate and pattern of decomposition have produced conflicting
16
results. Sample sizes have ranged from one or two (Dillon and Anderson, 1995;
Goff, 1992) to Komar’s review (1998) of 16 carcasses and Card et al’s (2015) study
using 20. Both pigs (Anderson, 2011; Card et al., 2015; Dillon and Anderson, 1995;
Dillon and Anderson, 1996; Kelly, 2006) and human cadavers (Dautartas, 2009;
Miller, 2002) have been used. Additionally, retrospective studies by Campobasso et
al. (2001), Haglund (1997), and Mann et al. (1990) have provided further
information.
Experiments comparing the rate of decomposition between clothed and unclothed
bodies carried out by Dautartas (2009) and Miller (2002) using cadavers, and by
Card et al. (2015) and Kelly (2006) using pigs, showed that whilst differences
between the bodies were observed, the presence of clothing had no statistically
significant effect on the rate of decomposition.
The largest study by Mann et al. (1990), a retrospective study of 150 bodies from
the Anthropological Research Facility (ARF) in Tennessee, produced a table of
variables affecting the decay rate of the human body. Scoring was subjective and
used a scale of 1 to 5 with 5 having the greatest impact. The scores do not indicate
if the effect is positive or negative, only its level of impact. Clothing received a low
impact score of 2 and the report concluded that the presence of clothing aided in
speeding up decay by providing maggots with protection from sunlight. The report,
however, gives no indication of how many of the 150 bodies in the study were
clothed and contributed to this finding. The same conclusion, that clothing sped up
the process of decomposition, was reached by Anderson (2010) and Dillon and
Anderson (1995), but the authors’ reasoning related to clothing providing more
locations for ovipositing.
17
Card et al. (2015), in a study of 10 clothed and 10 unclothed pigs, also observed
that clothing provided additional ovipositing sites, but did not agree with the
finding that the decomposition rate was increased. A retrospective study of twenty
individuals in states of advanced decomposition or partial skeletonisation by
Komar (1998) concluded, contrary to Mann et al.’s (1990) findings and in
agreement with Haglund’s (1997), that clothing was found to protect the
underlying tissues from animals and weather, resulting in delayed decomposition
rates. However, as this was a retrospective study no information was available on
the progression of decomposition within individuals, so it was not possible to
determine when the difference in decomposition rate occurred.
Goff’s (1992) study, using a pig carcass wrapped in two blankets rather than
clothed, found that wrapping impeded oviposition and delayed decomposition by
2.5 days. Initial observations were from an adult human who had been wrapped in
two blankets. Following this an experiment was carried out treating a single pig in
the same manner, which showed the same 2.5 day delay in oviposition. Delays in
ovipositing have not been reported in other studies. Kelly et al. (2009), using three
groups of two pigs – loosely wrapped, clothed, and unclothed – found no difference
in the decomposition time, with insects ovipositing on all six carcasses on the first
day. Studies by Grassberger and Frank (2004) and Marchenko (2001) also found
no difference in the oviposition timing, but did report a delay in decomposition.
Work by Aturaliya and Lukasewycz (1999), in which they maintained a flow of dry
air across the carcasses, found that the presence of clothing helped to desiccate
carcasses by wicking fluid away from the skin and, through evaporation, to
18
accelerate the rate of mummification. The subjects of this experiment were adult
mice, making its application to human bodies equivocal.
In conclusion, controlled studies with larger sample sizes appear to indicate that
clothing has no significant effect on the rate of decomposition; retrospective
studies and small sample size studies report an effect – interestingly, both to
increase and to decrease the rate of decomposition.
2.8 Decomposition of Hanging Bodies
Whilst many individuals are found relatively quickly if the hanging occurred in the
home, often as a result of suicide or autoerotic activities (Byard and Winskog,
2012), when the hanging has occurred outdoors or in a remote location bodies
may remain undiscovered for longer periods (Komar et al., 1999).
It might be expected that hanging bodies may display a different pattern of
decomposition, and thus might also benefit from a specific decomposition scoring
scale to reflect this when scoring decomposition to calculate PMI. Shalaby et al.
(2000), using a hanging pig and a pig on the ground, studied the differences in
rates and patterns of decomposition, finding significant differences in the rates of
decomposition of the two bodies. The rate at which biomass reduced was slower in
the hanging carcass during the bloat and decay stages. There are also reasons to
suspect that the species of insects attracted to a hanging body may differ if the
carcass is not touching the ground (Chinery, 1976). A sample size of one individual
does not, however, provide validated data, and further experiments using larger
numbers are needed to see if the results are repeatable and consistent.
19
When a body is hanging there is the potential for the whole body surface to be
exposed to the drying effect of winds, which could increase the possibility of
mummification or have a greater effect on the mummification rate than for a
carcass on the ground. In addition, female Diptera recognise mummified skin as an
inhospitable substrate for larval stages and do not lay there (Campobasso et al.,
2001). Where rapid dehydration occurs, typical of dry windy environments, this
reasonably explains the low external maggot activity observed (Campobasso et al.,
2001). De Jong et al. (2011) noted that mummified skin may act as insulation, so
there is also the possibility that internal temperatures may be affected, changing
the environment for any insects and impacting internal ADD calculations.
Disturbance of carcasses has been shown to have a significant effect on the rate of
biomass loss during decomposition (Adlam and Simmons, 2007; Cross and
Simmons, 2010) due to the displacement of necrophagous insects. Unless a
hanging body is in partial contact with the ground, crawling insects and maggots
are unlikely to return to the body once they fall into the drip zone, the area directly
below the body (Shalaby et al., 2000). Once in this drip zone, and unable to return
to the carcass, the insects may continue to feed on material falling from the body.
Thus the maggot mass within a hanging animal may be less than that within an
animal on the ground, where the maggots remain within the body cavity. A smaller
maggot mass within the hanging body may affect the rate of biomass reduction
mediated by insects. There may be changes in the internal temperatures as the
maggot masses are disturbed (Adlam and Simmons, 2007) and as maggots are lost
through gravity and movement. In a study using rabbit carcasses it was found that,
whilst individual elements of the decomposition process were affected by repeated
disturbance of a carcass, over time the decomposition was not significantly altered
20
(Adlam and Simmons, 2007). In a study using rat carcasses it was reported that the
repeated weighing of carcasses did not statistically affect the rate of biomass loss
(De Jong et al., 2011). The masses involved were, however, very small (grams and
milligrams) and the authors noted that a further study with larger animals such as
pigs, and with more intensive sampling during decomposition stages, could be
carried out to investigate the validity of the technique used (De Jong et al., 2011).
In a study carried out using pigs it was found that tissue loss in disturbed carcasses
was indeed adversely affected, and was slower than in the undisturbed control
group (Cross and Simmons, 2010).
2.9 The Use of Pigs as Human Analogues
For many taphonomic studies the use of human cadavers may be impractical or
impossible. Some countries do not allow experimentation on human cadavers and,
where research on human bodies is allowed, the use of human tissue is strictly
controlled. In the UK this is governed by the Human Tissue Act (2004). There may
also be restrictions in the availability of donated human cadavers and the existence
of suitable research facilities (Simmons and Cross, 2013). However, the need for
experiments to be carried out with sufficiently large sample sizes, and to be readily
replicable, means suitable human analogues are required. The analogues must
closely approximate human decomposition patterns, be readily available in
sufficient numbers, not too expensive, and unlikely to arouse public objection
(Catts and Goff, 1992).
Domestic pigs, Sus scrofa, are the most widely used human cadaver analogues in
taphonomic experiments (Forbes et al., 2005; Myburgh et al., 2013; Paczkowski et
21
al., 2015; Payne, 1965; Roberts and Dabbs, 2015; Schotsmans et al., 2012;
Schotsmans et al., 2014a; Schotsmans et al., 2014b; Swindle et al., 2012; Wilson et
al., 2007). Pigs are omnivores with a similar gut bacteria to humans (Anderson and
Van Laerhoven, 1996). They have a body mass and muscle-to-fat ratio that is
similar to a human and, like humans, are largely hairless (Anderson, 2010;
Schoenly et al., 2006). In adipocere formation studies, porcine adipose tissue is the
nearest and most reliable model for human adipose (Forbes et al., 2005). The skin
has similar dermal collagen, its elastic content is more similar to humans than
other animals, and porcine skin has been used for many years in the treatment of
burns victims, the study of dermal healing, and for skin grafts (Dillon and
Anderson, 1995, 1996). Additionally, the use of complete porcine heart valves for
valve replacement in humans is common. Pigs, as quadrupeds do, however, have
different limb proportions to bi-pedal humans.
Pigs are also considered reliable substitutes in entomological studies (Dillon and
Anderson, 1995; Tabor et al., 2004). The activity of insects is one of the most
important drivers of decomposition. Schoenly et al. (2007) found there to be
sufficient overlap in the arthropod fauna found on pigs and humans to recommend
the use of pigs as substitutes for humans in research. Based on 15 years of
entomology studies (1987-2002) the use of a domestic pig (Sus scrofa) weighing
approximately 23 kg (50 lb) was found to be a reliable model of a human corpse
(Schoenly and Hall, 2002; Schoenly et al., 2006; Schoenly et al., 2007). Additionally
the density, succession, and variety of arthropod species on a 23 kg (50 lb) pig
were found to be comparable to those on a 68 kg (150 lb) pig (Catts and Goff, 1992;
Schoenly and Hall, 2002; Schoenly et al., 2007).
22
In a study by Stokes et al. (2013), using soil microcosms, samples of human muscle
were compared with muscle samples from pig, cow, and sheep. It was found that
for the measure they were using the ovine muscle was the most similar to human,
but that they all “afford close approximation in decomposition dynamics”.
Whilst the value of using animals as human analogues for research, and how
applicable the results are to human decomposition, may be open to debate
(Simmons and Cross, 2013), there are many decomposition experiments which
could be carried out on animals but which might cause ethical objections if carried
out on human cadavers, e.g. gun shots, burning, hanging, etc.
23
3 General Materials and Methods
This Research was carried out during the months of May through August over a
period of three consecutive years and consisted of a series of field experiments
each with a specific set of objectives. To maximise consistency and repeatability,
and to maintain scientific rigour, where possible the experimental design and
methods used within the individual experiments were the same. This applied
particularly to the equipment used and the methods of data collection, recording,
and analysis. Some of the methods used were necessary to meet the requirements
to minimize biohazards and comply with UK regulations.
All experiments described in subsequent chapters followed the same general
experimental protocol outlined below. Where differences occurred these are noted
in the subsequent chapters concerning each specific experiment.
This chapter describes the field site; the equipment used and the general
preparation of the pigs; the methods of data collection; and the treatment of the
animals from death to disposal at the end of the experiments.
3.1 The Field Research Site
The research was carried out at the University of Central Lancashire’s Taphonomic
Research in Anthropology: Centre for Experimental Studies (TRACES) facility
(Cross et al., 2009). The site is situated near Burnley (53.77° N, 2.24° W) in the
northwest of England. TRACES is a 13 acre outdoor site of sloping rough pasture
land with areas of young trees, consisting of rowan, alder, birch, oak and pine, and
24
is surrounded by fencing outside of which is sheep pasture and rough fell side. The
site is 169 m above average sea level and exposed to southwesterly winds
(Lapworth and McGregor, 2008). Rainfall is high in this part of the country,
averaging 1294 mm per year (Met Office, 2010), and summer temperatures during
the experimental period average 17 °C or 63 °F. Insects commonly found on the
site that were of direct interest to these experiments are blow flies and beetles;
particularly Silphidae, Histeridae, and Staphylinidae.
3.2 Animals
Domestic pig (Sus scrofa) carcasses were used in all of these experiments. The use
of pigs as human analogues is a common practice (Anderson and Van Laerhoven,
1996; Catts and Goff, 1992; Huntington et al., 2007; Payne, 1965; Roberts and
Dabbs, 2015; Spicka et al., 2011; Stokes et al., 2013) due to practical and ethical
problems in sourcing sufficient human bodies for such studies. The pigs were bred
for the domestic meat market, not for experimentation, and were killed on the
farms by a licensed professional. To ensure biosecurity and prevent any
inadvertent spread of disease during transportation from farm to the experimental
site, each animal was placed into a body bag, which was sealed, before being
transported in a vehicle with appropriate signage.
On arrival at the site, as each pig was removed from the body bag it was given a
unique numbered ID label, reflecting its allocated group, which was attached to a
hind limb using a cable tie. Each pig was weighed using a hanging scale from
which the pig was suspended via a rope noose around a hind limb. The weight was
25
recorded (to the nearest 5 g) and the sex was also recorded. All pigs were in
experimental position within a few hours of this procedure.
3.3 Hanging Frames and Surface Enclosures
Hanging structures were constructed from scaffolding poles, with a pair of A-
frames each supporting a horizontal ridge pole 2.5 m above the ground. Figure 3.1
shows a diagram of the A-frame layout, while Figure 3.2 shows one of the A-frames
in use.
Figure 3.1 Diagrammatic representation of the scaffolding A-frames used for the
hanging pigs showing the bird proof netting, chicken wire and approximate
dimensions. The entry point was via a section of detachable wire in the centre of
one end. Some sections of wire and netting have been left out for clarity.
26
Figure 3.2 Hanging pigs in one of the scaffolding A-frames at the start of the
experiment.
These structures were covered in a scavenger proof chicken wire mesh to a height
of 1 m, above which was attached bird-proof plastic netting. Thus, the entire
structure was covered. Entrance into the frame was achieved via a detachable
section of chicken wire. Neither mesh nor netting impeded insect access. On each
ridge pole, spaced approximately 80 cm apart, five pig carcasses were hung by the
neck using nylon rope tied in a noose. Nylon rope was used to avoid the problems,
common to natural fibres, of rope stretching and shrinking with changes in
moisture. Each noose was attached to the ridge pole using a butchers’ hook, which
suspended the carcass so that the hind limbs hung a minimum of 60 cm off the
ground. It was anticipated that there would be some initial untwisting and
stretching of the nylon rope and that the animals themselves might elongate, due
to gravity, as decomposition progressed.
27
Each experiment had a group of pigs that lay directly on the ground. These are
referred to below as control or surface pigs, with the term ‘control group’ referring
specifically to a group of naked, fully non-suspended pigs, positioned so that they
were lying on their sides directly on the ground surface. These surface pigs were
placed 30 m from the hanging frames and spaced about five metres apart. Cages
covered in scavenger proof mesh, which did not impede insect access, were placed
over the surface pigs. A label identical to the one on the pig was attached to the
scavenger proof cage (Figure 3.3).
Figure 3.3 The scavenger proof cages used to cover the surface pigs. The cage in
the foreground shows one of the clothed pigs from Experiment 2.
28
3.4 Electronic Data Logging
Internal and external temperatures were recorded using EL-USB-1 Lascar
Electronics self-contained data loggers (Lascar Electronics Ltd, Salisbury, U.K.).
Each data logger was tested as follows: The loggers were programmed to collect
data every 3 minutes for a period of 15 minutes. They were then left on the lab
bench for 5 minutes; transferred to a fridge for 5 minutes; and then back into the
lab. The recorded data were downloaded and checked to make sure the loggers
were operating correctly. The loggers were fitted with new batteries and pre-
programmed to record temperature at 6-hour intervals throughout each
experiment. The casings of the data loggers were sealed with strips of parafilm
around the joins to help prevent decomposition fluid or precipitation reaching the
electronics. The data loggers remained in place until completion of the experiment
when the data were retrieved. Any loggers expulsed naturally prior to this were
returned to the lab and their data retrieved.
A data logger in a film-sealed case was inserted rectally into each surface pig and
pushed internally using a bamboo rod to a distance of approximately 40 cm. The
data loggers were placed internally to enable comparisons to be made between the
core temperatures of the pigs and the external temperature, as it is known that
maggots generate their own heat (Gallagher et al., 2010; Heaton et al., 2014;
Huntington et al., 2007; Marchenko, 2001; Simmons et al., 2010b). The data logger
arrangements for the hanging pigs differed for each experiment, due to experience
gained throughout the study, and are described in the Materials and Methods
section for each experiment. Each A-frame had an additional data logger, attached
at a height of 1.5 m above ground, to record the ambient temperature.
29
The data loggers were set up to start recording temperatures every six hours from
07.00 hrs on the day the pigs were killed. Each daily average temperature was
calculated from the 6-hourly recordings for the 24 hour period from 07.00 to 07.00
(as the average of the four readings taken at 13.00, 19.00, 01.00, and 07.00 hrs).
The daily average temperature values were used to calculate the Accumulated
Degree Days (ADD) values for the periods from the start of the study until each
observation day. The ambient ADD was calculated from the external loggers
attached to the hanging frames, and the internal ADD from the internally sited data
loggers. The study periods commenced after the threat of frosts had passed, so no
adjustments for temperatures at or below 0 °C were required in the ADD
calculations (Megyesi et al., 2005). All ADD figures were calculated using
temperatures in degrees Celsius.
3.5 Observations
Throughout the experiments the pigs were examined for visible decomposition
and this was recorded for three regions: head and neck, torso, and limbs. These
observations included changes in colour, size, presence of maggots,
decompositional fluids, changes in skin, and visibility of bone. For the hanging pigs
the area directly below the pig, the drip zone (Shalaby et al., 2000), was also
examined for any maggots, intestines or bones that had fallen from them.
Each pig was photographed at each visit with the hanging pigs being photographed
from front and side view. The photographs and written descriptions were used to
enable decomposition scores to be given to each animal at each visit.
30
3.6 Insect Data Collection
At the same time as making observations of the decomposition features, the
animals were checked for the presence of blow fly eggs and maggots, and larval
and adult beetles. The dates at which beetle larvae were first seen on the surface
pigs and on the hanging pigs were recorded. When beetles appeared on the pigs a
sample was collected and identified. The numbers of beetles were not recorded.
Blow fly eggs, which are small rice shaped and white or yellowy in colour, are
generally seen in small clusters or patches on the animals, particularly around the
natural orifices in the head, the anus, and in the hanging pigs around the noose.
Beetle larvae do not look like adult beetles and are easily distinguished from them.
The larvae of the beetles most commonly found on the TRACES site and associated
with the carcasses are elongated and flattened in appearance with legs and
antennae.
3.7 Disposal of Pig Carcasses
At the completion of each experiment the pig remains were sprayed with the food
dye ‘Brilliant Black BN’ (E151) to comply with the Animal By-Product
(Identification and Staining) Regulations 1995 and regulation amendments. The
pig remains were then placed in leak-proof bags which were sealed for disposal at
a DEFRA approved and licensed animal by-product renderer.
31
4 EXPERIMENT 1 – Comparing Decomposition Patterns
and Rates Between Hanging Pigs and Pigs Laid on the
Ground1
The aim of this first experiment was to compare the patterns and rates of
decomposition in hanging bodies with those of bodies in direct contact with the
ground, and to determine whether a different decomposition scoring scale is
required for hanging bodies to enable more accurate PMI estimation. To achieve
this, an experiment was carried out in the field using hanging pigs, fully suspended
off the ground, and control pigs placed on the ground, with large enough sample
sizes to ensure scientific rigour. Previously reported work that has been carried
out on hanging body decomposition has been on small samples, and has been
reported in case study format (Chong Chin et al., 2010; Shalaby et al., 2000), but
would indicate that the pattern of decomposition is different from that observed in
surface decomposition. The objectives of the experiment were to:
• record and compare the patterns and rates of decomposition of the hanging
pigs and the control pigs lying on the ground;
• record the weights of the hanging pigs each week to enable comparison of
the percentage weight loss between the males and females;
1 The results of this experiment have been published:
Lynch-Aird, J., Moffatt, C., Simmons, T. (2015). Decomposition Rate and Pattern in Hanging
Pigs. Journal of Forensic Sciences, 60(5):1155-1163.
32
• record the arrival of the first beetle or larva on the control pigs and on the
hanging pigs, or in the drip zones beneath the hanging pigs.
4.1 Materials and Methods
The general experimental method is described in Chapter 3. Twenty pigs (Sus
scrofa) were used for this experiment. The pigs were killed and collected on the
same day, and transported to the test site in a single delivery run. Several
experiments using pigs were being run at this time. To ensure the allocation of
pigs was random with no allocation by sex, size or other features, as they were
unloaded from the transport, the pigs were allocated into groups, one group after
another, and labelled uniquely to reflect their group.
For the hanging pigs, in addition to the rectally inserted data loggers,
thermocouples were inserted into the thorax via the oesophagus, using a metal
tube as a guide which was then removed (Figure 4.1). Once the pigs had been hung,
the thermocouples were each attached to additional pre-programmed data loggers.
To prevent rainwater damaging the data loggers the connection between the
thermocouple and the data logger was wrapped in parafilm. The logger and
junction were placed in a plastic bag sealed with duct tape, around the exit point
for the leads, and the bag was then attached to the hanging rope above the head of
the pig.
All pigs were put into position within an hour of each other on 29 May 2011 with
the exception of one of the surface pigs which, due to lack of availability on the day,
was killed and positioned one day later. No adjustments were made for the delay in
33
setting out this pig when calculating the results, as the average ADD for the first
day was less than 15, which represented only 1.6% of the full ADD range of 932 at
the completion of this experiment, and the TBS scores for this pig were within the
range of TBS scores for the other pigs within this group.
Figure 4.1 Insertion of thermocouple for attachment to temperature data logger.
Data collection took place every Monday and Thursday, until the end of the study
(25 July 2011) for a total of 17 observation days. Average daily temperatures at
this site during summer are around 17 °C, so visual changes associated with
decomposition do not occur rapidly. At each visit, the physical state of the surface
pigs was closely observed, recorded and photographed.
For scoring, each pig was treated as three distinct regions: head and neck, torso,
and limbs. For pigs lying on the surface, scores for each of three regions were
recorded using the scales developed by Megyesi et al. (2005) to produce a total
34
body score (TBSsurf). Megyesi et al.’s system was adjusted to a baseline of zero by
subtracting 1 from each of the regional scores; thus three from the overall TBSsurf
value. Megyesi et al.’s (2005) decomposition scoring tables, with the adjusted
scores, are shown in Appendix 1. Each scored part of the body thus starts with a
score of zero when fresh. This approach better reflects the fact that, in a fresh
condition, the body exhibits no visible decomposition. It also supports the fitting of
regression lines through the origin and allows for an easier conversion between
similar scales of different lengths (Gruenthal et al., 2012).
Photographs were taken and detailed descriptions recorded for the hanging pigs to
enable seriation of the decomposition process and development of a comparable
scoring system. Any differences in appearance which might have been sex-specific
were also noted. The ground directly beneath the hanging pigs, termed the “drip
zone” (Shalaby et al., 2000), was checked for any insects and examined for parts of
the pigs that may have fallen, and these findings were recorded.
4.1.1 Sex and Weight of Pigs
The hanging group contained five females and five males, and the control group
four females and six males.
The average weight of the hanging pigs at the start of the experiment was 33.15 kg
with a standard deviation of 6.26 kg, and the average weight of the control pigs
was 47.9 kg with standard deviation 8.8 kg. The difference in the weights between
the two groups was unintentional and unforeseen. The expectation was that the
pigs would all have been obtained from the same farm and would have been the
same age and about the same weight; thus controlling the age and weight variables
35
and not compromising the validity of the experimental results. On this occasion the
pigs were bought from two different farms and, although killed and collected on
the same day, the possibility of the pigs from different farms having greatly varying
weights had not been considered. The animals for slaughter were chosen by the
farmer and should have been of the same age, and weight, from each farm. They
were killed and immediately placed in bags and in the transport as described in
Chapter 3.
When several experiments are being carried out simultaneously at the site the
control pigs are under observation by several researchers and on this occasion the
controls were the last pigs to be unloaded from the transport. Consequently they
were probably amongst the first loaded and from the first farm, whilst the hanging
pigs were the first to be unloaded and would have been from the second farm.
There was no reason to believe the pigs were of different weights when collecting
them. The effects of the differing pig weights are considered in the Discussion
section below.
4.1.2 Measuring Percentage Weight Loss of Hanging Pigs
At weekly intervals the weights of the hanging pigs were recorded using the
suspended weighing scale on which they were originally weighed. This was hung
next to the pig by a hook over the horizontal scaffold bar of the A-frame. The pig
was moved carefully from the S hook it was hanging on onto the S hook of the
weighing scale, weighed, and then returned to the hanging frame hook, ensuring
minimal disturbance to the carcass.
36
4.1.3 Creation of a Scale for Hanging Body Score
Decomposition is a mosaic process, where different parts of the body decompose
at different rates, which is why the system defined by Megyesi et al. (2005) scores
the degree of decomposition present in the head and neck, torso, and limbs
separately. The hanging bodies were scored over the same three regions.
However, Megyesi et al.’s scoring scales were developed using data that contained
only bodies lying on a substrate. The hanging pigs did not display the same
decomposition features and hence could not be scored on this scale.
Using the characteristics of decomposition displayed by each pig at each ADD
interval, a seriation was produced reflecting the most typical pattern over a known
timeline, i.e. from known time of death. Scores were then allocated to each pig
based upon a review of the written description and photographs recorded at each
observation. Independent scores were given to the three regions: head and neck,
torso, and limbs. These scores were added to give a total body score for hanging
(TBShang) on this novel scale.
All scores were integers, with a baseline of zero equating to no visible
decomposition. As the scale records scores of visible decomposition, a fresh body
which does not show any such decomposition receives a score of zero points.
Where body sections did not show the same levels of decomposition over the
whole section, for example across all four limbs, then the score was an average of
the two extremes, to the nearest integer.
37
4.1.4 Statistical Analysis
All data analyses were performed using the software R (R Development Core
Team, version 3.1.1) and its LME4 package for mixed effects models (Bates et al.,
2015).
An attempt was made to model the relationship between the decomposition scores
and the ADD using linear regression. Various simple transformations of the ADD
values and decomposition scores were tried but none produced a sufficiently
linear relationship.
Higher order polynomial regression models were therefore tried of the form:
TBS = a 0 + a1 . ADD + a 2 . ADD 2 + ... + a n . ADD n
Inevitably there is a compromise to be made between the complexity (order) of the
regression equation and the accuracy of the fit to the data (mean-square error). A
chi-squared test was used to assess the significance of the fitting errors obtained
with different order models. For each test, the order of the regression model used
was increased until an adequate fit to the data was obtained, and within the
constraint that the version of R being used only supported regression models up to
order 5.
The fitted curves should not be extrapolated beyond the range of the recorded
ADD values, and thus any suggestion that the decomposition scores might start to
fall again at higher ADDs, outside the recorded range, should be ignored.
38
The rate of decomposition can be found by taking the first derivative of the fitted
curve with respect to ADD:
d (TBS )
= a1 + 2a 2 . ADD + ... + n.a n . ADD n −1
d ( ADD)
It is then a straightforward matter to determine both the maximum rate of
decomposition and the time (ADD) at which it occurred.
4.2 Results
The final observation date corresponded to 932 ADD. Data from the internal
thermocouples in the hanging pigs could not be used as a measure of internal ADD
since the majority of the loggers failed due to the ingress of moisture despite the
efforts made to seal them. Therefore, ADD was calculated based on the
temperatures recorded by the ambient data loggers attached to the hanging
frames.
4.2.1 General Pattern of Decomposition
Figure 4.2 shows a sequence of photographs taken throughout the study for one
hanging (H6, female) and one control pig (C9, male), while the paragraphs below
provide a more comprehensive description of the observed patterns of
decomposition in the hanging and control groups of pigs.
Initially the appearances of the hanging and control pigs were similar. On the first
day of the study, as the pigs were removed from the bags, and as they were set out
in the field, blow flies were observed landing on them. The flies continued to land
39
on the carcasses as they were hung for weighing suggesting that colonisation
would have occurred on the hanging carcasses in the same way as those on the
ground had the carcass been hanging at time of death. The flies oviposited in the
natural orifices in the head, around the anus, and in the skin creases, ostensibly
showing no preference for either group of pigs. Eggs were also laid around the
noose on the hanging pigs.
Once larvae had hatched and all the pigs had started to purge decomposition fluids
at 147 ADD, the patterns observed in the two groups diverged. The bowels had
prolapsed in all the hanging pigs with some of the data loggers being avulsed; this
did not occur in the control pigs. The testicles of the male hanging pigs were
considerably more distended than those of the controls which never reached this
level of swelling.
229 ADD: The maggots in the surface group remained visible and active, with wet
decomposition destroying the throat tissue and opening the carcass. Beetle adults
and larvae were also present. In contrast, maggots were no longer visible in the
hanging pigs and no beetles or their larvae were seen. Drip zones had formed
beneath the hanging pigs. These contained maggots which had fallen from the pigs
in quantities from a few to volumes of approximately 240 ml.
279 ADD: All the pigs’ heads had been largely consumed; 80% of the control pigs
had part of the maxilla or mandible exposed whilst only 10% of the hanging pigs
were showing any bone. Differences in the pattern of decomposition were evident.
The hanging pigs were mummified with little visible external insect activity.
40
Maggots were evident only when they fell to the drip zone, at which point they
effectively had no further impact on the hanging body.
385 ADD: The appearance of the hanging and control pigs was now very different.
The control pigs could be seen to be decomposing, evincing clear moist
decomposition with parts of the torso breached and bones visible. From this point
there was little breaching of the hanging pigs and they became hanging ‘empty
pigs’, with the pig appearing externally to be complete but with the internal organs
decomposing or decomposed. The bowels of the hanging pigs had now prolapsed
fully with some fallen to the ground.
459 ADD: Externally the hanging pigs displayed deflation and lengthening. Lifting
the forelimbs showed active maggot activity in some of the axillae and some with
breached skin. Two of the pigs had hind trotters just touching the ground so were
raised back off the ground. No adult or larval beetles were observed on the pigs or
in the drip zones. The control pigs had foaming maggot masses and bones visible,
including mandibles, limbs, ribs, scapulae and vertebrae.
512 ADD: The intestines of the majority of the hanging pigs had dropped to the
drip zones. All the hanging males had lost their intestines, as had 20% of the
females. In the male pigs the scrotal sac opened and joined with the anal hole. In
the controls there were few or no visible maggot masses on the skin, but these
were still present beneath the skin.
591 ADD: In the hanging pigs the torso had become leathery. The stomach and lots
of fat were in the drip zones and the first bones were found in the drip zones: ribs,
41
scapulae and vertebra. Figure 4.3 shows exposed rib bones within the torso of one
of the hanging pigs, seen through the axilla. The controls all had their head and
some limb bones exposed.
631 ADD: The torsos of the hanging pigs had become mummified. The long bones
and innominates were found in the drip zone. High rainfall had left the control pigs
lying in pools of water and all of them had soft swollen skin and were fatty.
767 ADD: The hanging pigs were elongated, mummified and hard. In the control
pigs all the heads were skeletonised and most of them had more than 50% of
bones visible in the limbs and torso.
828 ADD: Prolonged heavy rainfall left all the pigs looking fatty and white where
rehydrated. The control pigs were lying in rain water.
932 ADD: The hanging pigs were mummified and all remained hanging. All the
controls now had more than 50% bone exposure on the head, limbs and torso.
4.2.2 Patterns of Decomposition in Male and Female Hanging Pigs
The pattern of decomposition displayed two main differences between the hanging
male and female pigs: in overall torso shape and in the position of the final opening
in the lower torso. The control pigs did not show the corresponding differences in
shape.
At initial bloat, all the hanging pigs assumed an evenly swollen appearance with
the limbs swollen and held out from the body. As the bloat subsided, the males
42
deflated first on the upper torso, remaining swollen around the mid-torso with the
greatest width around the umbilicus and penis. The scrotum also remained
swollen. This shape was retained until fully deflated even though the penis and
umbilicus had become open holes. The female pigs maintained an evenly swollen
shape, appearing rectangular in frontal view and with no protrusion around the
umbilical area. This shape was retained until fully deflated, leaving a pouch of skin
drooping on the lower abdomen.
In both males and females the anus enlarged to around 10 cm in diameter early in
the decomposition. In a hanging position the anus of a pig is not at the lowest part
of the torso but higher and dorsal. In the males, as the decomposition progressed,
the back of the scrotal sac decomposed joining with the anus and creating a much
larger opening directly below the body (Figure 4.4) enabling bones to drop out. In
the females the anus remained open only at the back with the pouch of stretched
skin remaining from the bloating stage left hanging lower than the anus (Figure
4.5). Some bones collected in this pouch, below the anus, (Figure 4.6) and were still
there at the end of the study period.
4.2.3 Timing of First Beetle Activity
The first beetles were observed on the control pigs at 57 ADD. The first beetles
were observed on a hanging pig at 869 ADD with a beetle larva having been seen in
the drip zone of a pig at 591 ADD.
43
15 ADD 57 ADD 147 ADD 229 ADD
All pigs: Both the hanging and All pigs: Bloated and displaying All pigs: Bloated and purging All pigs: Maggot masses in mouth.
control pigs showed effects of blood colour changes with clear marbling decomposition fluids; maggots on Hanging: Few other maggot masses
pooling. in some animals. and in the head. visible, but drip zones had formed
Hanging pigs: All had prolapsed beneath the pigs.
bowels and some of the anal data Controls: Throats open and moist
loggers had already been avulsed. with visible decomposition; maggot
The male pigs’ testicles were masses visible on the underside.
considerably more swollen than in
the control pigs.
Figure 4.2 Decomposition sequence for one hanging (H6, female) and one control pig (C9, male).
44
279 ADD 338 ADD 385 ADD 459 ADD
Hanging: One hanging pig showed Hanging: Little visible external Hanging: Appeared externally to be Hanging: Deflating and lengthening;
exposed bone. maggot activity; skin slippage but no complete but internal organs were active maggot activity in some of the
Controls: 80% of the control pigs breaches of skin. decomposing; bowels fully prolapsed axillae and in some the skin was
had part of the maxilla or mandible Controls: Throat and chest open with some fallen to the ground. breached.
exposed. with visible maggot activity; some Controls: Could be seen to be Controls: Foaming maggot masses;
ribs visible. decomposing; evincing clear moist bones visible, including mandibles,
decomposition with parts of the limbs, ribs, scapulae and vertebrae.
torso breached and bones visible.
Figure 4.2 (cont.) Decomposition sequence for one hanging (H6, female) and one control pig (C9, male).
45
512 ADD 591 ADD 631 ADD 714 ADD
Hanging: In male pigs the scrotal sac Hanging: Skin leathery. The stomach Hanging: All torsos mummified; long Hanging: Maggot fall continuing into
opened and joined with the anal and lots of fat were in the drip zone. bones and innominates were found drip zone; torsos very deflated.
hole. In 100% males & 20% females First bones found in the drip zone: in the drip zone. Controls: More bones exposed.
the intestines had dropped into drip ribs, scapulae and vertebra. Controls: High rainfall had left the
zone. Controls: All heads and some limb pigs lying in pools of water and all of
Controls: Few or no visible maggot bones exposed. them had soft swollen skin and were
masses on the skin but maggots still fatty.
present beneath the skin.
46
767 ADD 828 ADD 869 ADD 932 ADD
Hanging: Torsos were elongated, All pigs: Prolonged heavy rainfall All pigs: Drying out after previous Hanging: Pigs were mummified and
hard and mummified. left all the pigs looking fatty and wet period. all remained hanging.
Controls: All heads skeletonised and white where re-hydrated. Controls: All the controls now had
most of them had more than 50% of Controls: The control pigs were more than 50% bone exposure on
bones visible in the limbs and torso. lying in rain water. the head, limbs and torso.
47
B
Figure 4.3 Showing the Figure 4.4 Hanging male at Figure 4.5 Hanging female Figure 4.6 Hanging female
ribs in the torso of a 512 ADD. The arrow at 600 ADD. Arrow A pig at the close of the
hanging pig viewed points to the anal opening points to the pouch of skin experiment. The pouch of
through the breached axilla extending beneath the hanging below the anal skin was cut open and the
at 591 ADD. body. opening (arrow B). arrow points towards the
bones which had collected
in the pouch.
48
4.3 Total Body Score for Hanging Bodies (TBShang)
The new scoring scale for hanging bodies was aligned with Megyesi et al.’s scale
spanning the same decomposition range from fresh to dry bone and scoring the
same body regions. The scale produced for scoring the hanging pigs is based on a
score for the head and neck between 0 and 12 points to give a Partial Body Score
Head (PBSHhang); a score for the torso between 0 and 11 points to give a Partial
Body Score Torso (PBSThang); and a score for the limbs between 0 and 9 points to
give a Partial Body Score Limbs (PBSLhang). Total Body Score for Hanging (TBShang)
was a sum of the three region scores and provided an overall measure of
decomposition for the hanging body with a minimum score of 0 and a maximum of
32 points. Tables 4.1 to 4.3 list the decomposition scoring scales used for the three
regions of the hanging pigs.
49
Table 4.1 Stages of hanging decomposition for the head and neck.
Score Description
0 Fresh, no discolouration.
1 Swelling of head and neck.
Purging of decomposition fluids, darkening of skin to black, sinking in

2
around eyes, some drying of snout and lips, most flesh still fresh.
3 Skin shrinking back from jaw, skin slippage.
4 Flesh sinking in around jaw and throat, drying of skin and lips.
5 Hair loss, drying shrinking snout.
6 Leathery skin, less than 10% bone visible.
7 Holes in or tearing of the skin.
8 Complete mummification, ears dry inflexible.
9 Partial mask formed where some of the skin is lifting from skull.
Skin like parchment, very thin and translucent with holes and tears
10
forming.
Bone visible over more than 50% of skull, or detached mask for more than
11
75% of skull.
Dry bone which may still be hanging if attached by mummified tissue or

12
may have fallen to the ground.
50
Table 4.2 Stages of hanging decomposition for the torso.
Score Description
1 Bloating, red to green discolouration, swelling of males’ testicles.
2 Colour changing from green to grey to dark grey / black.
3 Prolapsed bowel, purging of decomposition fluid.
Lower torso more swollen than upper. Penis and umbilicus are open holes
4
on males.
5 Body elongated with some caving in of upper torso.
Anus open to diameter of 10 cm or greater, intestines protruding. Scrotum

6
soft and shrivelling in males.
7 Drying and browning of skin, becoming leathery on upper torso.
Intestines dropped out of body, skin mummified. Scrotal sac open at back
8
into anal opening on males.
Stomach dropped from body, ribs, scapulae and vertebrae dropped, flanks
9
sunken, skin very greasy to touch.
Large loss of fat onto drip zone, torso drum like, innominates and long
10
bones dropped.
Dry bone below carcass or visible beneath mummified skin, carcass may
11
still be hanging.
51
Table 4.3 Stages of hanging decomposition for the limbs.
Score Description
Reddening of skin, becoming dark grey, limbs swollen and extended out
1
from the body.
2 Some skin slippage, drying of limbs at extremities.
3 Grey becoming black, forelimbs hanging in usual position.
4 Drooping forelimbs hanging loosely, disarticulated at shoulder.
Hind limbs hanging loosely, disarticulated at pelvis, loss of hind hooves,

5
hair loss.
Mummification of lower end of limbs, skin colour dark brown, tears and
6
holes in skin.
Bones in hanging bags of skin. Mummified with distal ends of limbs rock
7
hard, skin stuck to bones.
Bones dropping through torso, scapulae first then humerus. Limb surface
8
very greasy.
Dry bone below carcass or visible beneath mummified skin, carcass may
9
still be hanging.
4.4 Analysis
In hanging pigs, the sexes showed significantly different patterns of loss of mass
(χ2 = 24.4, df = 3, p < 0.001), with females losing mass more slowly and retaining
more mass throughout (Figure 4.7). The maximum rate of loss for males was
52
0.17% loss per ADD (at ADD = 483), and for females 0.13% loss per ADD (at
ADD = 503).
Figure 4.7 Differences in percentage weight loss between male and female hanging
pigs to 900 ADD. The data points show the values for each sex at each value of
ADD. The model lines are the polynomial regression (here fourth order) fitted
values. Some jitter has been added to the ADD values of the data points for clarity.
When using the Total Body Score as the response variable, there was again a
difference between sexes (χ2 = 15.6, df = 2, p < 0.001), though this was not as large,
and again, males showed a greater rate of decomposition (Figure 4.8).
This curve is a quadratic, and, as can be seen from the plot, maximum slope is
found where ADD = 0; being 0.062 TBShang per ADD for females, and 0.068 TBShang
53
per ADD for males. Despite being statistically significantly different, the differences
are therefore small.
Figure 4.8 Rate of decomposition showing the difference between the male and
female pigs using Total Body Scores for Hanging (TBShang) to 900 ADD. Jitter has
been added to ADD for clarity.
Using the Total Body Score and Total Body Score for Hanging (TBSsurf and TBShang)
to compare the rate of decomposition of surface and hanging pigs, a highly
significant difference was evident (χ2 = 104, df = 5, p < 0.001), with hanging pigs
reaching later stages of decomposition more quickly (Figure 4.9).
The behaviour in this case is more complicated (requiring a fifth order
polynomial). In relatively early decomposition, e.g. around 150 ADD,
54
decomposition rates are steeper and similar (controls: 0.051 TBSsurf/ADD, hanging:
0.060 TBShang/ADD); whereas later, when decomposition slows on the ground,
hanging slows even more (e.g. at 650 ADD controls: 0.020 TBSsurf/ADD, hanging:
0.014 TBShang/ADD).
Figure 4.9 Comparison of decomposition rates of hanging and grounded pigs using
Total Body Scores for hanging (TBShang) and surface (TBSsurf) pigs, respectively, to
900 ADD. Jitter has been added to ADD for clarity.
4.5 Discussion
This study showed the pattern of decomposition in hanging pigs differed from that
in pigs lying on the ground to the extent that the construction of a new
55
decomposition scoring scale was needed for the hanging pigs. The comparison of
percentage weight loss between the male and female hanging pigs showed a clear
difference between the sexes, which may be the result of the different shapes
assumed by the male and female pigs as they decomposed. The corresponding
decomposition scores for the two sexes did not, however, show a similar
difference. These findings and other observations are discussed below.
4.5.1 Decomposition Scoring Scale for Hanging Bodies
The Megyesi et al. (2005) Total Body Score scale was produced using a
retrospective cross sectional study of humans which, by its nature, provided a
single snapshot of the stage of decomposition at which each individual was found,
rather than a longitudinal study observing progressive changes in the same
individuals over the whole decomposition process. Although Megyesi et al. studied
many individuals covering much of the range of decomposition stages, there were
no instances of hanging bodies. It appears from the findings of the hanging studies,
including the present longitudinal study, that the pattern of decomposition in
hanging bodies is markedly different, making Megyesi et al.’s scale unsuitable for
scoring these bodies.
For their study, Megyesi et al. calculated ADD values using the daily average
temperature records from the nearest national weather station for each body,
while the PMI in each case was determined either from insect evidence or the
associated police investigation (Megyesi et al., 2005). This approach would have
led to uncertainties in the accuracy of both the ADD (Dabbs, 2010, 2015) and PMI
values being used. For the present study the ADD was calculated from two
temperature data loggers situated within 10 m of the pigs, with measurements at
56
6-hour intervals, while the PMI was known to within one hour. Both the ADD and
PMI values for the present study can therefore be regarded as accurate with a high
degree of confidence.
Decomposition scores generated using the new hanging scale cannot be directly
compared with those from Megyesi et al.’s (2005) scale as they describe different
patterns of decomposition; there is, currently at least, no established absolute
scale of decomposition which could be used to link them. However, both scales
span the same range of decomposition and, once adjusted for a zero baseline,
assess the same body areas using the same point ranges. Furthermore, the new
scale was constructed as far as possible so that a given score (for each body
region) corresponded to a similar overall level of decomposition. Where the
hanging and control groups did display common features these were generally
awarded the same scores in both scales. Consequently, while it would be
imprudent to read much into the detailed differences displayed in Figure 4.9, it is
encouraging that the total body score plots for both the hanging and control
groups show a broadly similar progression against ADD, and that the differences
between the two groups matches the observed behaviour.
4.5.2 Mummification
The new scoring scale produced for hanging pigs includes mummification in the
description of what is seen during the decomposition. Whilst it could be argued
that it should not be in the scale as it is not a feature common to all decomposition,
it was found to occur in all 20 hanging pigs of the present 2011 study; was also
observed in a previous trial in 2010; and has been reported in case studies of
57
human hangings (Komar et al., 1999). This would indicate that mummification is a
feature of hanging, much like adipocere is a feature of many immersed bodies.
Micozzi (1991) describes mummification as referring to all natural and artificial
processes that bring about preservation of the body or its parts. Mummification
may occur naturally as the result of a number of factors including heat, humidity,
cold, wind, or the absence of air. Natural mummification may be deliberately
enhanced by the way the body is treated or placed; for example, burying in hot dry
sand, which is known to enhance the removal of water and speed up desiccation
(Aturaliya and Lukasewycz, 1999).
The most common form of natural or spontaneous mummification is desiccation.
Aufderheide (2011) says: “The interface of the skin surface with the body’s
environment is the principle determining variable for establishing the rate of a
corpse water loss.” This loss of water may result in desiccation and may lead to
mummification of the skin and the preservation of some or all of the soft tissues.
In these experiments the action of the wind on the fully exposed hanging pig
bodies is the most probable cause of their desiccation and mummification. The rate
of desiccation was such that the internal decay continued unabated leaving the
carcasses ‘hollow’ (Figure 4.6). The same feature may not be found where wind is
not a relevant environmental factor or where humid conditions prevail.
4.5.3 Initial Weight of the Pigs
The average weight of the hanging pigs and the control pigs at the start of the
experiment was significantly different: using a T test gave t = 4.32, df = 18 and
58
p = 0.0004 at the 5% significance level. Although unintentional, this difference
resulted from the way in which the pigs were allocated to the two groups and
introduced an unexpected shortcoming into the experimental design where failing
to control the weight variable may have undermined the validity of the results. To
strengthen this part of the experimental design for future experiments it would be
desirable to ensure all the pigs come from the same farm where it could be
expected that they would be of the same age and roughly the same weight. The
hanging decomposition scoring scale produced in this experiment should also be
tested on pigs of different weights.
Chapter 7, which reports on results obtained using the combined data from all the
experiments carried out in this study, covering a much wider range of pig weights,
shows that the difference in weight would not in fact have made a difference to the
production of the decomposition scoring scale.
4.5.4 Rate of Decomposition in Hanging and Control Pigs
Accepting that the two scales were comparable, and differences in scores between
hanging and surface carcasses really did reflect the differences in decomposition,
the difference in the rate of decomposition shown in the current study, with the
hanging pigs lagging behind the controls in the early stages of decomposition, is in
agreement with Shalaby et al. (2000) and the previous 2010 study. This difference
may be attributed to the differences in levels of insect activity between the two
groups. Insect access, particularly for Coleoptera (beetles), to the hanging pigs was
limited and this may explain the delay in decomposition of the hanging pigs.
Additionally, the action of gravity and movement disturbed the maggot masses
causing maggots to fall from the hanging pigs to the drip zones below, thus
59
decreasing the internal maggot masses. This has implications for calculating the
PMI of hanging bodies as many are found with the feet in contact with the ground.
If this has occurred at the time of hanging or rapidly after, the insect access is likely
to be similar to that of a body on the ground.
4.5.5 Comparison of Weight Loss in the Male and Female Hanging Pigs
The differences in shape between the hanging males and females during much of
their decomposition may be explained by the positioning of the urethral opening.
In the males this is situated just posterior to the umbilicus and may thus provide
another escape route for the gases produced during decomposition, making this
point the widest as the gases gather there. The escaping gases may also attract
insects, particularly Diptera, as the umbilicus and penis of the males soon became
full of larvae and progressed to open holes.
This did not occur in the females where the urethral opening is just ventral to the
anus, indicating that it is the urethral opening that is the weak point rather than
the umbilicus. Without this escape route higher on the torso, the females retained
a rectangular shape and, as they deflated, the skin which had remained stretched
and swollen now formed a pouch of loose skin at the lower torso. This pouch of
skin hung lower than the anal opening, so that disarticulated bones dropping down
within the female pigs could become trapped here rather than falling to the
ground. This contrasted with the males where, as the tissue of the swollen scrotum
decomposed it became part of the anal hole. A single large opening directly
beneath the pig was produced and this allowed bones to fall through more easily.
Whilst of note in this study, this feature is unlikely to be of any significance when
observing humans as, when hanging, the anal openings of quadrupeds and bipeds
60
are not in the same place. In humans the anal opening would be directly below the
body and the positioning of the human penis is also low down, not at the umbilical
level, so the shape differences between the male and female pigs in late bloat and
deflation may not be present in humans.
The weight loss started slowly, increased, and then slowed down toward the end of
the decomposition as would be expected (Shalaby et al., 2000). The difference in
decomposition pattern of the male and female pigs may account for the difference
in percentage weight loss during decomposition, as bones were lost from the males
through the anal opening. Whilst the rate of decomposition between the males and
female pigs was shown to be statistically significant the actual difference was very
small. The hanging pigs in this experiment were weighed weekly and in doing so it
is inevitable that the maggot masses would have been disturbed. Adlam and
Simmons (2007), suggest that the physical disturbance of a carcass may disrupt
the feeding of the maggots and retard the rate of decomposition, but any such
effect would be expected to be the same for both the males and females. Weight
loss in soft tissues has been used by some authors (Payne, 1965; Payne et al., 1968;
Shalaby et al., 2000), as a way of assessing decomposition. This study has shown,
however, that care is required in extending this principal to whole body weight
loss. Past studies (Adlam and Simmons, 2007; Payne, 1965) have shown a
correlation between decomposition rate and weight loss. This may not be the case
where part of the body has been lost or dispersed.
4.5.6 Beetles
In a pilot study (unpublished) previously carried out at TRACES the rate of
decomposition of hanging pigs was significantly different (faster) than in this
61
study. A noticeable difference was the presence of beetles on the carcasses at an
early stage thus increasing the diversity of insects. The number of species of
insects recovered from hanging cases tends to be lower than for bodies lying on
the ground (Goff and Lord, 1994). It has been reported that the greater the
diversity of insects, the greater the rate of decomposition (Campobasso et al.,
2001) whilst a study by Chong Chin et al. (2010) noted that hanging altered the
insects that colonised by excluding the ground living taxa.
Nine of the pilot study’s hanging pigs had their hind limbs in contact with the
ground for some or all of the time between 54 ADD and 159 ADD before being
raised up. During this time, effectively, these pigs were not hanging. Whilst the
beetles most commonly found on the pigs at the study site, Silphidae, Histeridae
and Staphylinidae, are all strong fliers, Chinery (1976) states “Beetles can and do
fly well, but relatively little time is spent in flight; beetles are very much insects of
the ground and low vegetation”. The pigs’ hind limbs resting on the ground would
have provided the beetles with an easy means to reach the hanging pigs and the
pilot study suggested that they may have taken advantage of this route.
Additionally, between 54 ADD and 200 ADD five of the pilot study pigs herniated
the intestine providing another site for the release of volatile gases, attracting flies
and providing a moist access to the carcass for oviposition. It was noted that four
of these five pigs were male. In pigs, unlike humans, the intestines do not usually
lie above the inguinal canal, but in the hanging position the gravity and pressure of
gases may have acted on them pushing into the inguinal canal and causing them to
herniate.
62
4.5.7 Internal Temperature Measurements
This experiment intended to compare the internal and external temperatures of
the pigs using internal data loggers and thermocouples, but these failed to record
over the duration of the study period or were deactivated by decomposition fluid
and/or rain. The experience gained did, however, lead to changes and
improvements in the way that the data loggers were prepared for the subsequent
experiments.
4.5.8 Effects of Heavy Rainfall
Also noted was the change to the appearance of bodies on the ground when they
had become sodden and waterlogged following prolonged rain. This rehydrating of
the tissues and fats gave the bodies the appearance of being ‘fresher’ than prior to
the rain. This may give the impression that the progression of decomposition is
less advanced and lead to a lower TBSsurf which, in turn, will impact the PMI
calculations possibly providing an inadvertently lowered PMI. An accurate PMI is
important in establishing the events surrounding a death and creating a timeframe
when searching for missing persons. It may, therefore, be prudent to check the
amount of rainfall and duration in the days prior to a body being found if it does
look particularly fatty, and make allowances for the apparent retardation of
decomposition. To ensure that PMI calculations are not inadvertently lowered it
may be necessary to increase the range of dates within which the PMI falls.
63
5 EXPERIMENT 2 – The Effect of Clothing on the
Decomposition Pattern and Rate in Hanging and
Surface Pigs
Hanged bodies2 are found in various stages of dress and undress ranging from fully
clothed through semi-clothed to naked; the latter two being common in autoerotic
hangings where death has resulted from a failure of the safety release mechanism.
In Experiment 1 (Chapter 4) it was shown how the decomposition patterns
differed over time, using Accumulated Degree Days (ADD), for ‘naked’ hanging and
surface bodies. Results from other experiments (Anderson, 2010; Aturaliya and
Lukasewycz, 1999; Campobasso, et al., 2001; Card et al., 2015; Dautartas, 2009;
Goff, 1992; Haglund, 1997; Kelly, 2006; Komar, 1998; Mann et al., 1990) with
clothed, wrapped and unclothed bodies have reported different effects on the
decomposition rates and the patterns. As Total Body Score (TBS) scores are
obtained from the decomposition pattern, and these are used with the ADD to
calculate the estimated PMI, it is important that any effects due to the presence or
absence of clothing are known so that, if necessary, adjustments can be made to
the calculations.
The aim of this experiment was to investigate the effect of clothing on the
decomposition pattern and rate of hanging pigs and pigs in direct contact with the
ground (surface pigs). The objectives were to:
2 In British English people are ‘hanged’ and animals are ‘hung’.

64
• record the patterns and rates of decomposition of clothed and unclothed
hanging and surface pigs;
• compare the rates of decomposition between the clothed and unclothed
hanging pigs, and between the clothed and unclothed surface pigs; and
• compare the patterns and rates of decomposition between the hanging
clothed pigs and the clothed pigs lying on the ground.
The experiment also provided more decomposition data to add to those already
collected at the same site for (unclothed) hanging and surface pigs in previous
years.
This experiment used 40 pigs (Sus scrofa), killed as described in the General
Materials and Methods (Chapter 3), which were allocated randomly to four equal
groups: hanging unclothed (H/U), hanging clothed (H/C), surface unclothed (S/U)
and surface clothed (S/C), and labelled uniquely to reflect their group. The H/U
group was therefore equivalent to the hanging group in the other two experiments,
while the S/U group was equivalent to the control group.
As in Experiment 1, all the surface pigs, clothed and unclothed, had pre-
programmed data loggers inserted into the rectum as described in General
Materials and Methods.
65
5.1.1 Sex and Mean Weight of Pigs
The S/C group contained six females and four males, while the other three groups
each had five males and five females. The mean weight of the hanging pigs, H/C
and H/U, was 26.29 kg, with a standard deviation of 2.36 kg; and for the surface
pigs, S/C and S/U, the average weight was 27.28 kg, with a standard deviation of
5.02 kg. A t-test showed that there was no significant difference in the weights of
the pigs between the two groups (t = 1.75, df = 38, p = 0.08).
5.1.2 Hanging Pigs
Because of the failure of the thermocouples used in the hanging pigs and the early
avulsion of the anally inserted data loggers in Experiment 1, changes were made to
enable measurement of internal temperature in the hanging pigs.
The wires connecting the thermocouples to the data loggers were fragile and may
have been one of the causes of their failure to record in the first experiment. To
avoid this happening again the pre-programmed data loggers were sealed with
parafilm then placed inside leak-proof tubes, the openings of which were smeared
with petroleum jelly. The lids were fitted and the junction sealed with duct tape.
Wire was wrapped around the top of each tube with a length of about one metre of
free wire which, once the pig was hung, was attached to the hanging rope to
prevent the tube becoming lost into the pig, or dropping out of it, as decomposition
progressed (Figure 5.1). The data loggers were inserted into the thorax using the
same techniques as for Experiment 1.
66
Figure 5.1 Data loggers sealed in 50 ml conical centrifuge tubes with wires for use
with the hanging pigs. The wires were attached to the hanging ropes once the
loggers had been inserted into the pigs.
5.1.3 Clothing the Pigs
Immediately after the anal data loggers had been inserted, the pigs in the S/C
group were dressed in identical white cotton tee shirts, black cotton boxer shorts,
and a webbing belt. This was to ensure that they were left unclothed for as short a
time as possible. As pigs have short legs, boxer shorts were used to mimic trousers.
A loosely tied belt was included as it is common to wear one with trousers and this
may result in restriction during bloat. The H/C pigs were clothed in the same way
and had the data loggers inserted into the thorax after being dressed.
5.1.4 Data Collection Intervals
All the pigs were put into position in a two-hour period on 29 May 2012. Data
collection occurred every other day until 635 ADD, when decomposition had
67
slowed, and then at four day intervals until the end of the study at 1034 ADD (29
July 2012), for a total of 24 observation days. However, on two occasions prior to
635 ADD, the data collections took place at four day intervals, rather than every
other day, corresponding to an interval of approximately 50 ADD. The first
occasion followed heavy rainfall of 11.5 cm in a 24 hour period before data
collection was due, and the second occasion was to comply with safety regulations
which did not allow researchers to visit the site alone (staff illness meant it was
not possible to have the required minimum of two people on the site).
5.1.5 Data Collection
Data collection for the unclothed pigs was carried out and recorded as described in
the General Materials and Methods chapter. The decomposition pattern was
relatively easy to observe in the unclothed pigs. However, the presence of clothing,
which covered most of the clothed pigs, made the observation impossible if the
clothing was not moved.
To make comparisons between the clothed and unclothed pigs the same areas
needed recording but with as little disturbance of the pigs and insects and larvae
as possible. To achieve this, the observations of the clothed pigs were made
initially with the clothes in place. The bottom end of the shorts legs were then
lifted carefully to observe any changes that could be seen beneath the shorts. The
bottom of the tee shirt was carefully lifted to observe changes beneath this and
around the belted area. Finally, where possible, the sleeve of the tee shirt was lifted
to observe changes in the axilla and on the fore limb. Great care was taken to lift
the clothing gently and replace it as quickly as possible to minimize the
disturbance of the maggots or insects beneath the clothing.
68
5.1.6 Decomposition Scoring
Visible decomposition of the S/C and S/U pigs was scored using Megyesi et al.’s
system (2005) adjusted to a baseline of zero to give TBSsurf. The H/C and H/U pigs
were scored using the new hanging scale developed in Experiment 1 to give
TBShang.
For the analysis, data for the H/U and S/U pigs were taken from all three years’
experiments. Since the data were not collected at the same ADD values each year,
this provided TBS scores for a larger number of actual ADD data points. The
combined group analysis in Chapter 7 showed that the results for H/U and S/U
pigs were consistent across all three experiments.
Experiment 1 and the analysis in Chapter 7 further showed that the sex of the pigs
and their initial weight did not affect the TBS, so weight and sex variables were not
considered in this experiment. All data analysis was performed using the software
R (R Development Core Team, version 3.1.1) and its ‘arm’ package (Gelman and Su,
2015). Linear regression of TBS versus ADD was carried out for each treatment
group and, using the summary table from this, simulations were performed
(Gelman and Hill, 2006) enabling comparisons to be made at the same selected set
of ADD values, with a fixed interval of 100 ADD, despite the data having been
collected at different ADD values during each year’s experiment.
The linear regression model produced a line of best fit, but the summary report
gave the errors for the intercept and slope. The simulation then used the error
structure in the model to give a large number of possible TBS values at each ADD;
69
with errors from which a mean and standard deviation were produced. Graphs
produced using this method are clear and include errors making them easier to
interpret than one showing the regression model directly (Moffatt and Simmons,
2014).
S/C
S/U
a) H/U
S/C
H/C
b) H/U
S/C
H/C
c)
H/U
Figure 5.2 Group pairings used for TBS comparisons: a) the S/C and H/U groups
were compared to the S/U group; b) the S/C and H/U groups were compared to
the H/C group; c) the latter enabled the H/C group to be compared to the H/U
group, and also to the S/C group.
The TBS against ADD responses were compared for the four groups as illustrated
in Figure 5.2. The S/C and H/U groups were compared against the S/U group
(Figure 5.2a), and the S/C and H/U groups were compared to the H/C group
(Figure 5.2b); with the latter enabling the comparison of the H/C to the H/U group,
and the comparison of the H/C to the S/C group (Figure 5.2c) without the need for
a large number of graphs. 10,000 simulations were performed for each of the four
70
group-pair comparisons, with the simulations generating the TBS difference values
between the two groups in each case.
5.2 Results
The two existing decomposition scoring scales used in Experiment 1 proved to be
sufficient and appropriate for scoring the clothed groups: the S/C group and the
S/U group were scored successfully using the scale developed by Megyesi et al.
(2005), again adjusted for a baseline of zero, while the H/C group was scored using
the same scale as the H/U group, developed during Experiment 1.
As the pigs were being removed from the body bags, flies were seen to land on
them, but no eggs were seen on any of the animals before clothing them. Flies
continued to land on the pigs once they had been placed in the field. Over the
following few days, eggs were observed around the hanging nooses, and in the
natural orifices around the anus and head in all four sets of pigs. Additionally, eggs
were laid in the creases of the clothing and beneath the tee shirts, particularly
around the waistband of the shorts.
The H/U pigs displayed a different pattern and rate of decomposition from the S/U
pigs. Furthermore the H/C and H/U pigs differed from each other in their rate and
pattern of decomposition. However, there were no significant differences between
the S/C and S/U pigs in their pattern and rate of decomposition. Descriptions of
the main differences between the groups, for a range of ADD values, are given
below.
71
Figure 5.3 and Figure 5.4 show the progression for one pig from each group
through the decomposition process. Figure 5.3 compares one of the H/C pigs with
one of the H/U pigs; while Figure 5.4 does the same for an S/C pig and an S/U pig.
Not all the sampled ADDs are represented since on some occasions no obvious
changes had occurred between successive visits.
110 ADD: Hanging Pigs: In the H/U pigs, the bowels were starting to protrude and
in the males the umbilicus and testes were swelling. These changes were not seen
in the H/C pigs. (Observation was via the leg of the shorts.)
Surface Pigs: Maggots were present and visibly active on the heads and in the
mouths of all the surface pigs. There were no noticeable differences between the
two groups.
167 ADD: Hanging Pigs: Both groups of pigs had protruding bowels, drying snouts
and visible skin slip on the head and torso. In all the male pigs the umbilicus/penis
area was swollen. A drip zone, containing a few maggots, had formed beneath two
of the pigs (one clothed, one unclothed).
Surface Pigs: All animals had maggot masses around the head, and the males had
swollen umbilicus/penis with maggots present (the same as in the hanging group).
The only obvious difference between the two groups was the presence of maggots
just under the epidermis in the S/C pigs, and eggs around the waistband of the
shorts beneath the tee shirts.
197 ADD: Hanging Pigs: Differences in appearance were now apparent between
the H/C and H/U pigs. All the H/C pigs displayed swelling and discolouration of the
72
lower abdomen on the right side, with four of this group having small holes in the
skin and larger skin breaches forming in this area. One H/C pig had an open hole
on the left lower side with a maggot mass clearly visible. The skin was
deteriorating beneath the wet clothing and sticking to the shorts. Beneath the wet
clothing, maggot activity was clearly visible on and beneath the epidermis. During
observation the shorts were not detached from the skin where sticking had
occurred.
In the H/U pigs all the bowels had now prolapsed. Maggots were not visible on the
unclothed pigs and were seen only where they had fallen into the drip zone. Drip
zones had formed below the pigs with most containing a small quantity of
maggots, from a teaspoonful to about 100 ml.
Surface Pigs: Most of the pigs had some bone visible on the head. Differences in
appearance were not great, the most obvious being the hind limbs which in the S/C
pigs remained relatively fresh looking whilst the S/U pig limbs showed browning
and drying.
224 ADD: Hanging Pigs: Little change was evident in the appearance of the H/U
pigs. In the H/C group the skin beneath the shorts was wet and disintegrating with
maggots clearly visible. The skin of the lower abdomen, on the right-hand side, was
breached in five of the pigs. The prolapsed bowel of one of the pigs was caught in
the shorts. Maggots were active and visible in the axilla of the H/C pigs and four
pigs had clear breaches of the skin.
73
Surface Pigs: The main difference between the two groups was the degree of
decomposition around the lower abdomen. This was greater in the S/C pigs with
marked increase in maggot activity and tissue removal under the shorts and
around the waistband. The skin around the breached areas was brown and drying.
284 ADD: Hanging Pigs: There was a marked difference in the destruction of the
axilla with 90% of the H/C group having breached axilla compared with 40% of the
H/U group. The bodies of the H/U pigs were elongating and the lower limbs were
disarticulated in 60% of the pigs compared to 30% of the H/C group. In all of the
pigs the anus had become a large open hole.
The shorts from one of the pigs had slipped from the body into the drip zone. The
intestine of this pig had fallen through the shorts two days earlier and the whole of
the lower abdomen, anus and area beneath the pig (in its hanging position) had
become one open hole. The drip zone for this pig contained the shorts, maggots
and a bone.
Surface Pigs: Visible differences were still not marked. The abdomens were open in
four of the S/U pigs and six of the S/C pigs with bones visible in two of the latter.
The most obvious difference was the presence of small holes around the areas
where the waist band of the shorts was in contact with the skin and where there
were maggots on the skin beneath the tee shirt. It should be noted that the weather
had been very wet and as a result all the clothes were wet or damp.
326 ADD: Hanging Pigs: All of the H/C pigs had breached open lower abdomens on
the right side. Three of these had completely open lower abdomens from front to
74
back and the shorts had fallen into the drip zone. The lower section of the pigs,
covered by the shorts, remained wet, soft and maggoty with more eggs being laid
under the tee shirts.
In the H/U pigs the skin was drying and hardening. The appearance of the H/U pigs
had not changed greatly although their shoulders and upper backs were dry and
hard. This drying and browning was also occurring in the H/C pigs where the
shoulders were not covered by the tee shirts.
Surface Pigs: For both groups the heads of most pigs had over 50% of the bone
visible. The S/C group had bones visible in four of the animal torsos with none
visible in the S/U pigs. One S/C pig had bones visible on one of the hind limbs
where it had been covered by the shorts.
361 ADD: Hanging Pigs: All the animals were very, very, wet; it had been raining
almost constantly.
Surface Pigs: All the animals were very, very, wet; it had been raining almost
constantly. Bones were visible in the torsos of three S/U pigs and in the limbs of
five. The S/C pigs had bones visible in the limbs of three pigs. Only two of the S/C
pig torsos had visible bone. On the previous visit bone could be seen in four S/C
animals but maggot masses were covering these in three of the four cases noted
previously. One of the two S/C pigs where bone was observed this time had shown
no bone previously.
418 ADD: NOTE: a month’s worth of rain, 114 mm, had fallen in a 24 hour period
since the previous visit and all the animals were very wet.
75
Hanging Pigs: The limbs of all the pigs were mummified. All of the H/C pigs had
vertebrae and ribs present in their drip zones or visible in their breached bodies.
The guts of this group had either fallen, or were protruding from the lower
abdomen at the front of the torso and/or below the pig where the anus had opened
through. Most of the H/C group also had areas in the upper torso and axilla where
the skin was tearing or had opened into large holes. Only one H/U pig had any
bone in the drip zone.
Surface Pigs: There was very little difference in the numbers of pigs within each
group showing bone in the torso; and all of these had less than 50% of bone
showing. All pigs showed areas of dry mummifying skin. The weather was dry and
although the S/U pigs had dry skin on their upper surfaces this was not the case
with the S/C pigs, where the clothing remained wet.
457 ADD: Hanging Pigs: No great changes in appearance.
Surface Pigs: 90% of the S/U pigs had bones visible in the torso with the remaining
skin browning and drying. 40% of the S/C pigs had bone visible in their torso,
although for two pigs where bone had been visible previously none could be seen
as active maggot masses, beneath the clothing, were covering the previously seen
bones.
497 ADD: Hanging Pigs: 90% of the H/U pigs were now mummified with hardened
shoulders and backs. There were clear differences in the drip zones: 80% of the
H/C pig drip zones contained vertebrae and 70% contained ribs, whilst for the H/U
pigs 20% contained ribs and vertebrae. Shorts now remained on only three pigs,
76
the rest of the shorts having slipped off the pigs and fallen into the drip zones
along with four of the tee shirts.
Surface Pigs: All the S/U pigs had exposed bones in the torso with the skin on the
upper surface of all but one of these pigs dry and mummifying. Maggot activity was
present in some animals where the tissue on the ground remained wet and sodden.
In the S/C pigs bone was visible in all but one of the torsos. The clothing on the
pigs remained sodden and the skin in the abdominal area was wet with maggot
activity. The skin on the backs of these pigs, despite being beneath wet clothing,
was brown and drying on most of the pigs.
559 ADD: Hanging Pigs: The skin on both the H/C and H/U pigs was becoming
greasy and fatty. The lower abdomen was open in the entire H/C group, with the
lower half of one of the bodies now detached and in the drip zone. All the H/C pigs
had ribs and vertebrae in their drip zones. All the H/C group had open abdomens
and much of the rest of the torsos had tears and holes, with bones caught or clearly
visible. Four of the H/U group had bones in the drip zone and three of the pigs had
tears and holes in the upper torso.
Surface Pigs: All pigs were on very wet ground and had rehydrated fat. The S/C
pigs appeared to have retained a more rounded pig shape than the S/U; looking
more intact. However, the amount of bone visible in both groups was very similar.
599 ADD: Hanging Pigs: All of the H/C pigs had vertebrae and ribs in their drip
zones, but vertebrae and ribs were only found in 30% of the H/U pig drip zones.
Long bones and innominates were present in the drip zones of both groups.
77
Surface Pigs: The S/C pigs still retained a more rounded pig shape with the S/U pig
bodies having a collapsed appearance. There was little difference in the amount of
bone visible between the two groups and both groups had active maggots in some
of the pigs.
634 ADD: Hanging Pigs: All the H/C and H/U pigs were mummified. The H/U
groups had much greater skin destruction with more than 50% of the bones
visible, either on the ground or through the holes and tears in the skin. Three of the
pigs in the H/C group had the lower part of their bodies, from about mid torso
down, in the drip zone with the bones clearly visible where the tissue has been
destroyed. A fourth pig in the H/C group had the lower portion of the body
attached only by threads of skin. One of the H/C pigs had vertebrae and ribs falling
out in one piece.
Surface Pigs: All the pigs were very wet with hydrated fat and softened skin. The
skin on the upper surface of the S/U pigs was detached from the bones and lay
blanket-like on the bones. The skin of the S/C pigs was wet and maggot masses
were visible, surrounding bones, beneath the shorts in the fat of two pigs.
698 ADD: Hanging Pigs: All the pigs were mummified with little change in the
appearance of the H/U pigs. The H/C pigs had continued to lose limbs.
Surface Pigs: S/U pigs: More than 50% of the skeleton was visible either through
the very thin covering of mummified skin lying on the uppermost surface or visible
when the detached skin blanket was lifted. The skin covering the S/C pigs
appeared to be more intact; it was attached to fat and could not be lifted to view
78
beneath it. Where the skin was not attached in this way the bones were clearly
visible and not covered in tissue. The clothing was wet.
770 ADD: Hanging Pigs: The skin on some of the H/U pig torsos showed tearing
and holes forming, through which bones were falling or could be seen. This
occurred mainly on the front of the torso. The H/C pig torsos had many tears and
holes in both the back and the front of the animal. Where the lower bodies of the
H/C pigs had fallen into the drip zones the bones were visible and largely clean of
any tissue.
Surface Pigs: The skin on the S/U pigs was mummified. Where skin was present on
the torso it was detached and when lifted the bones beneath were visible and clean
of tissue, although there was fat on the ground around them. This was not the case
with the S/C pigs where skin remained wet and attached to fat, and occasionally
the clothing, and could not be easily moved. However, bones that were visible were
clean without tissue adhering. There was still some maggot activity in the fat.
867 ADD: Hanging Pigs: All pigs were mummified. The pigs which had been
clothed (H/C) were clearly distinguishable from the H/U animals, the former
having torsos with many tears and holes and, for many, the lower portions
detached from around the area of the shorts waistband. The H/U pigs still, largely,
retained the appearance of an elongated deflated pig, although the skin was tearing
on many of them.
Surface Pigs: The clothes remained on the pigs, and the clothing and the skin
beneath it remained wet, throughout the experiment. Skin and fat remained on and
79
around the S/C pigs although the bones were clearly visible through this and did
not have tissue adhering to them. The S/U pigs had little skin remaining or it was a
mummified detached blanket which could be lifted to expose the bones beneath.
1033 ADD: Hanging Pigs and Surface Pigs: There was no real change in the
appearance of the pigs. However, the weather had been very wet and some
maggots were seen on the surface of the pigs. These were few in numbers; in the
tens rather than hundreds or thousands.
5.2.2 Timing of the First Beetle Activity
An adult beetle was first observed on one of the surface pigs at 110 ADD, but this
evaded capture and so was not identified. At 326 ADD the first adult beetle,
Staphylinidae, was seen in the drip zone of a H/C pig, on fallen gut. This beetle also
evaded capture and could not be further identified, but the most commonly found
Staphylinidae at this site are Creophilus maxillosus. The first adult beetle was seen
on the hanging body of a pig at 457 ADD, on the outside of the tee shirt of one the
H/C pigs. This beetle was identified as Silphidae Thanatophilus rugosus.
80
110 ADD. 167 ADD.
224 ADD. 284 ADD.
326 ADD. 361 ADD.
Figure 5.3 Decomposition sequence for one H/C (HC1, male) and one H/U pig (H1,
female).
81
418 ADD. 457 ADD.
497 ADD. 559 ADD.
634 ADD. 698 ADD.
Figure 5.3 (cont.) Decomposition sequence for one H/C (HC1, male) and one H/U
pig (H1, female).
82
770 ADD. 867 ADD.
Figure 5.3 (cont.) Decomposition sequence for one H/C (HC1, male) and one H/U
pig (H1, female).
110 ADD. 167 ADD. 224 ADD.
Figure 5.4 Decomposition sequence for one S/C (CC3, female) and one S/U pig (C6,
male).
83
284 ADD. 326 ADD. 361 ADD.
418 ADD. 457 ADD. 497 ADD.
Figure 5.4 (cont.) Decomposition sequence for one S/C (CC3, female) and one S/U
pig (C6, male).
84
559 ADD. 634 ADD. 698 ADD.
770 ADD. 867 ADD.
Figure 5.4 (cont.) Decomposition sequence for one S/C (CC3, female) and one S/U
pig (C6, male).
5.3 Analysis
The rate of decomposition did not differ significantly between the S/C and S/U
pigs, but did differ from that of the hanging pigs. Figure 5.5 shows the difference in
rate during the decomposition between these groups.
85
The S/U group was taken as the reference for the y-axis zero in the graph. As
described in the Materials and Methods section, the analysis generated the
differences in TBS, at each ADD, between the S/C group and this reference, and
between the H/U group and the reference. Negative values on the graph indicate a
lower TBS than the S/U for that ADD value and positive values a TBS higher than
the S/U at that ADD value.
Figure 5.5 shows that there is no significant difference in the TBS values between
the two surface groups, with the S/C TBS points or error bars falling on the S/U
(reference) line, i.e. overlapping and not differing from it. The H/U group, however,
is clearly different with a lower TBS than both surface groups at the start, but with
a steady increase in (relative) TBS as ADD increases, and reaching the later stages
of decomposition (higher TBS) faster.
Figure 5.6 illustrates the difference in rate between the H/C group (taken as the
reference) and the H/U and S/C groups. As before the H/C group is shown with a
TBS value of zero across the ADD range, and the differences in the TBS values of
the H/U and S/C groups are plotted. The difference in TBS between the groups
show the H/U pigs reaching later stages of decomposition earlier than the S/C pigs,
but slower than the H/C pigs. Only one group of surface pigs (S/C) is used in Figure
5.6 since Figure 5.5 shows there to be no significant difference between the
decomposition rate of the S/C and S/U groups. The hanging pigs decomposed
faster than the surface pigs and the H/C pigs decomposed at a faster rate than all of
them.
86
Figure 5.5 Differences in the rate of decomposition of the S/C and H/U groups,
compared to the S/U group as the reference. The results were derived from 10,000
simulations per group based upon linear regression models. Some jitter has been
added to the ADD values at the data points to add clarity and the error bars show
95% confidence intervals.
87
Figure 5.6 Differences in the rate of decomposition of the S/C and H/U groups,
compared to the H/C group as the reference. The results were derived from 10,000
simulations per group based upon linear regression models. Some jitter has been
added to the ADD values at the data points to add clarity and the error bars show
95% confidence intervals.
5.4 Discussion
The experiment showed that the same scoring scales that were used previously for
H/U and S/U groups could be applied to clothed pigs in similar situations.
As in Experiment 1, the results from this experiment showed that the hanging pigs
(H/U) displayed a different pattern and rate of decomposition from the control
88
(S/U) pigs. Furthermore, the H/C and H/U pigs differed from each other in their
rate and pattern of decomposition. There were no significant differences, however,
between the S/C and S/U pigs in their pattern and rate of decomposition. This is
described in more detail below.
It was also found that the rate of decomposition in all four groups of pigs, at the
chosen test site and time of year, was sufficiently slow that performing the data
collections every other day was unnecessary; it would have been sufficient to
examine the pigs at three or four day intervals.
5.4.1 Hanging Pigs
The most likely explanation for the differences in decomposition between the H/C
and H/U pigs is that the presence of the clothing provided a more favourable
environment for the maggots. The ovipositing flies showed no preference for H/C
or H/U pigs. This may have been because the clothing, which was of a light
material and loose fitting, did not impede access to the usual ovipositing sites on
the head, which was not covered, or around the genitalia and anus which could be
reached via the legs of the shorts. This indifference to the presence of clothing
when ovipositing was also noted by Kelly et al. (2009).
However, additional eggs were laid on the outside of the clothing, in the creases in
the material, but these did not appear to hatch, with few if any maggots seen in the
area and eggs remaining in the creases. Whilst eggs were laid beneath the tee
shirts along the waistband of the shorts, eggs were not seen on the torsos of the
H/C pigs other than in the usual crease areas of the axilla and groin; the waistband
of the shorts appeared to be treated as another damp crease. This laying of eggs
89
and subsequent hatching around the waistband and shorts was also noted by Kelly
et al. (2009).
Once the eggs had hatched the clothing made a noticeable difference to the
behaviour of the maggots. The H/U pigs were fully exposed to the weather with the
wind and, infrequently, sun, which quickly dried the skin and any maggots thereon.
The maggots rapidly moved from the laying sites to inside the head and torso
where they were protected from the weather. Maggots prefer to colonise areas
which are sheltered and warm (Anderson, 2010; Campobasso et al., 2001; Mann et
al., 1990). The clothing provided a microclimate on the body of the pig with
protection from the weather and maggots were able to remain beneath the
clothing on the surface of the pig. The weather during this experiment was
particularly wet and the clothes on the pigs were always wet or damp, which
meant that the skin beneath them remained damp and flexible for most of the
experiment. In some areas, particularly around the waist band, the skin beneath
the clothing became very wet and thin with the later stage instars appearing to be
able to eat through the flexible skin leaving small holes in it. It may be that in drier
conditions this would not have been the case.
“The interface of the skin surface with the body’s environment is the principle
determining variable for establishing the rate of a corpse water loss” (Aufderheide,
2011). Where the skin is exposed to the air, and particularly to draughts, wind or
dry air, the transfer of water from the underlying tissues, via the skin to the air,
may result in desiccation and mummifying of the skin. The rate of desiccation may
be altered by the presence of clothing which can act to increase or impede the
water loss. Where clothing is loosely in contact with the skin, it tends to act as a
90
wick, drawing water away from the skin and increasing the rate of desiccation.
This desiccation is further accelerated if there is airflow (Aturaliya and
Lukasewycz, 1999). When the clothing remains wet, as was the case in this
experiment, it impedes the removal of water and delays or prevents the process of
dehydration. This may lead to faster breakdown of the skin, as it is kept wet and
pliable, accelerating its loss of integrity and the rate of destruction.
A major difference in the patterns of decomposition was in the lower torso from
the waist band of the shorts down. There are a number of reasons that could
account for this. The shorts were wet because of the rain and because of the
decomposition fluid which was released from the bowel, and in the case of the
males the umbilicus / penis, wicking along the material. This would have kept the
shorts damp for longer even without rain. The damp shorts, in turn, kept the skin
beneath them very damp and flexible. The experiment found that in all of the H/C
pigs the lower right portion of the abdomen became swollen and had active
maggot activity quickly rupturing through the skin, which continued to
decompose. The breached area contained gut and faeces. It may be that the extra
bacterial action in the gut produced more gas and pressure on the soft flexible
tissue and caused the breach. The area around this breach became more rapidly
destroyed. This did not happen in the H/U pigs whose skin was not so wet or
flexible. It is, however, improbable that this feature would be significant in the
decomposition pattern of hanging humans as the positioning of the intestines
differs in this area between humans and pigs. In the human the large intestine
loops around the outer portion of the abdomen but in the pig the large intestine is
found in tight spirals in the left hand side of the abdomen and the small intestine,
91
where most of the bacterial digestion takes place, lies in the right hand side
(Swindle et al., 2012).
When the bowel prolapsed in the H/C pigs it was contained, at least initially,
within the shorts keeping the gut close to the groin of the pig, and adding to the
moist conditions, rather than falling from the pig as in the H/U pigs. One of the
features of hanging animals is that once a maggot falls from the body it is unable to
gain access again and thus has no further impact on the hanging animal, although it
may continue to feed on material in the drip zone. Feeding maggot masses are very
active with maggots constantly writhing around within the mass as part of their
feeding and thermoregulatory behaviour (Charabidze et al., 2011; Heaton et al.,
2014). The shorts provided a physical barrier to the maggots falling and in the
experiment it could be seen, clearly, that the numbers of maggot below the H/U
pigs were far greater than beneath the H/C pigs. From this it would be reasonable
to assume that the maggot masses within the H/C pigs remained larger and this
would have helped to consume the pig more quickly.
The presence of maggots on the torso of the H/C pigs and the holes in the skin that
they made destroyed the integrity of the skin. The formation of tears in the skin as
the bodies elongated, stretching the skin, and the large numbers of maggots
around the waistband of the shorts resulted, for some of the pigs, in the lower
section of the torso becoming detached from the upper torso and falling from the
animals into the drip zone.
Active maggot masses are known to move bones, and dentures in the case of
humans, some distance from the body (Haskell et al., 1997) and to move clothing
92
(Kelly et al., 2009; Komar, 1998). Such movement of bones and clothing was
observed in this experiment, although the distances moved were not great. In one
of the hanging pigs the lower torso became detached and dropped to the ground
early in the decomposition process when the feeding maggot masses would have
been very active. This section of pig was found to have been moved some feet from
the drip zone. There was no evidence of any large scavengers having been in the
cage. This is worth noting as it may make a difference to how far away from the
hanging remains articles of clothing and bones from within them may be found.
Bones were moved around beneath the H/U pigs but generally not far from the
drip zone; usually by the time they fell to the ground there was little in the way of
visible tissue adhering to them and the numbers of maggots in the drip zone would
not have been as great as an active mass falling with part of the clothed torso.
5.4.2 Surface Pigs
There was no significant difference in the rate or in the patterns of decomposition
between the S/C and S/U pigs. All the surface pigs were laid directly upon the
ground which, because of very high levels of rainfall, remained wet throughout the
entire experiment. This meant that the lower surfaces of the pigs were always wet
and clothing soaked up water from the ground. Despite being constantly wet the
clothing showed little deterioration. These results may be different in different
weather conditions where clothing could provide protection from drying sun and
wind.
Reports on the effects of clothing on the rate of decomposition vary from showing
accelerated decay, via no effect, to delaying the rate of decomposition (Mann et al.,
1990). A retrospective study of the work carried out at the Anthropology Research
93
Facility Knoxville concluded that clothing accelerated the rate of decomposition.
This study, whilst having the advantage of a large number of human subjects (150
bodies had been observed over 10 years by students and faculty members), was
not a controlled experiment and the data reported for each of the case studies
would not be expected to be consistent in content or format. The conclusion
reached was that the clothing provided protection from sunlight for the maggots.
Campobasso et al. (2001), also making a retrospective study, found clothing could
slow down the post-mortem cooling, favouring putrefaction, and additionally
provided protection, from inclement weather, for the larvae.
Work by Dautartas (2009), Kelly (2006), and Miller (2002) found that, whilst
clothing had a noticeable effect on the decomposition, this was not borne out by
the statistics which showed no significant differences between the rate for clothed
or unclothed carcasses. This experiment also found that there was no statistically
significant difference in the rate of decomposition for the surface groups.
Delay in decomposition was reported by Haglund (1997) and Komar (1998) with
clothing cited as protecting the tissue by restricting the access to the carcass by
scavengers. Delay was also reported by Goff (1992) whose experiment involved
wrapping a carcass in cotton blankets and securing them, and who found that
insect access was delayed by 2.5 days. Kelly (2006) also had a wrapped carcass but
this carcass was loosely wrapped and the insects were able to access it at the same
time as they accessed the unwrapped and clothed carcasses. Whilst the last two
experiments were on wrapped rather than clothed carcasses, they show that the
level of access to the carcass by insects is important in dictating the rate of
decomposition.
94
Work by Aturaliya and Lukasewycz (1999) found that the presence of clothing
helped to desiccate carcasses by wicking fluid away from the skin and, through
evaporation, to accelerate the rate of mummification. This was not found to be the
case in this experiment, but the constant rain and high humidity would not have
been conducive to evaporation at the rate seen by Aturaliya and Lukasewycz as
they maintained a flow of dry air over the carcasses during their experiment.
The current experiment used light, loose clothing and insects were able to access
the normal ovipositing sites without great obstruction. However, clothing used in
winter and inclement weather may consist of multiple layers of heavy materials or
even include balaclavas, obstructing the mouth and nostrils, and these may show
results more aligned with those of Goff.
The results of the current experiment, with clear differences in the decomposition
of the hanging pigs but not the surface pigs due to the presence of clothing,
coupled with the range of results reported elsewhere, indicate that there may be
more factors that need to be taken into consideration. The presence, quantity and
type of clothing may act together with the weather conditions to alter the
decomposition rate. These are factors which need to be taken into consideration
when estimating the post-mortem interval of a clothed body.
95
6 EXPERIMENT 3 – Comparing the Decomposition of
Partially Suspended (Semi-Recumbent) Pigs with Fully
Suspended Hanging pigs and with Fully Recumbent
Pigs in Direct Contact with the Ground
The aim of this experiment was to investigate whether a semi-recumbent body
displays the same decomposition patterns as a body in direct contact with the
ground, or a fully suspended hanging body, or whether it displays some
intermediate pattern. The objectives of the experiment were to:
• record and compare the patterns and rates of decomposition of fully
suspended hanging pigs, partially suspended semi-recumbent pigs, and
control pigs lying directly on the ground;
• determine whether the decomposition scoring scale for bodies on the
ground or the scale for hanging could be used with the partially suspended
pigs to determine PMI, or whether an intermediate scale would be needed.
The results of the experiment should help to increase the accuracy of determining
the PMI in semi-recumbent bodies.
Additionally the observations of the hanging and control pigs provided the
opportunity to add to the known data on decomposition in hanging pigs and pigs
lying on the ground, with data already collected from the same site in previous
years. This increased the overall sample size available for any statistical analysis,
96
and could be used for looking at the effects, if any, of the changing weather
conditions at that site.
Thirty pigs (Sus scrofa), were allocated randomly into three groups: hanging,
control, and semi-recumbent. Killing, weighing, and sequential labelling of the pigs
followed the approach described in the General Materials and Methods chapter, as
did the pre-programming of data loggers and the preparation and placement of the
hanging pigs and the control pigs in direct contact with the ground. The semi-
recumbent pigs were hung by the neck in the same manner as the hanging pigs
except the lower portion of their bodies, the rump, and the hind legs were left
resting on the ground (Figure 6.1).
The semi-recumbent pigs were placed in separate A-frames from the fully
suspended hanging pigs. The frames were set out in a line close to each other so
that conditions would be as similar as possible in terms of sun and wind exposure
and tree shade. The hanging and semi-recumbent pigs were placed in alternate
frames. They were not mixed in the frames as previous experiments had shown
beetles appearing early in the animals in contact with the ground, while beetles
were not found on the hanging pigs and it was thought the lower portions of the
semi-recumbent pigs might also attract beetles earlier and introduce a close
population of beetles that had not been present with the hanging pigs in previous
experiments. It is improbable that a fully suspended human body would be found
alongside a semi-recumbent hanging body as the nature of hanging tends to be
97
solitary. A final consideration was that the hanging pigs were much heavier than
previously and it was not clear whether they might stretch to touch the ground.
All the pigs were put into position within a 2 hour period on 22 May 2013. The
ground beneath the control and semi-recumbent pigs was very wet with some
surface water.
Figure 6.1 The position of a semi-recumbent pig, supported from a scaffolding A-

frame.
98
6.1.1 Sex and Mean Weight of Pigs
The hanging group contained seven males and three females; the semi-recumbent
group contained five females and five males; and all ten pigs in the control group
were male. This resulted from the random allocation of the pigs to their respective
groups. However, the results from Experiment 1 showed that the sex of the pig had
no statistically significant effect upon the decomposition rate in the control or
hanging pigs.
The average weight of the hanging and semi-recumbent groups was 57.7 kg, with a
standard deviation of 5.7 kg, while that of the control group was 46.6 kg with a
standard deviation of 6.6 kg. The pigs weighed almost twice as much as the pigs in
the previous year’s experiment and, although the pigs were allocated as they were
unloaded from the transport, the controls were lighter than the hanging pigs.
6.1.2 Data Loggers
The hanging and semi-recumbent pigs had pre-programmed data loggers inserted
into the thorax, using a metal tube as a guide which was then removed. The data
loggers were sealed and placed inside leak-proof centrifuge tubes which were,
additionally, sealed with duct tape. A length of wire was wrapped around the top of
each tube and attached to the hanging rope to prevent the tube becoming lost into,
or dropping out of, the pig.
Two pre-programmed data loggers were also attached to each of the four hanging
A-frames. Two data loggers were used in each cage to allow one to be removed
from each cage and downloaded during the experiment to examine the
progression of ADD up to that point.

99
6.1.3 Data Collection and Collection Intervals
Data collection took place every Monday and Thursday until 22 July, giving a total
of 19 observation days, and was carried out and recorded in the same manner as
described in the General Materials and Methods chapter.
6.1.4 Decomposition Scoring
Visible decomposition of the control pigs was scored using Megyesi et al.’s system
(2005) adjusted to a baseline of zero to give TBSsurf. The hanging, fully-suspended
pigs were scored using the new hanging scale developed in Experiment 1 to give
TBShang.
Initially the semi-recumbent pigs were scored using both scales to assess the
whole body until it became clear the upper and lower portions of the animals were
displaying different decomposition patterns such that there was no longer a
consistent surface or hanging decomposition pattern across the whole pig. From
this point on, as the decomposition progressed, the pigs were scored as two
separate sections. The portion resting on the ground and part way up the body, i.e.
the lower limbs and lower torso, was designated as semi-recumbent lower (SRL)
and scored using the same scoring scale as the control pigs. The upper body
portion not resting on the ground, i.e. the upper torso, upper limbs, and the head
and neck, was designated as semi-recumbent upper (SRU) and scored using the
same scoring scale as the fully suspended pigs. Both scales were applied to both
sections of the semi-recumbent pigs until it became clear they were displaying
different decomposition patterns.
100
The SRL scores constituted a Partial Body Score (PBS), being the sum of the scores
for the lower torso and limbs (PBST + PBSL), i.e. without a score for the head and
neck, and so varied between 0 and 20, rather than 0 and 32 for the whole body
scores. Therefore, for comparison purposes, the torso and limb scores of the
control pigs were similarly added to give the corresponding partial body score
PBSsurf = PBSTsurf + PBSLsurf which also varied between 0 and 20 and is referred to
below as the Control Lower score.
The TBS versus ADD was plotted for the hanging pigs and the SRU scores, and for
the Control Lower and SRL scores, to provide an initial visual comparison.
Because the effect of ADD on TBS is not the same across the ADD range there is no
simple ratio of increase in TBS for increase in ADD and when plotted the
relationship was a curve. Prior to running the linear regression analysis and
ANOVA, a simple transformation of the ADD and TBS values was performed to
improve the linearity of the response, between the transformed variables, thereby
improving the fit of the model generated by the linear regression analysis and
increasing the validity and relevance of the sensitivity analysis based on the
response gradients of the model. Following some investigation the data were pre-
treated by excluding those data points having ADD ≤ 35 or ADD ≥ 625. These ADD
values were excluded as at ADD ≤ 35 very little change had occurred in the
appearance of the pigs since death, and in humans the individual should still be
recognisable, and at ADD ≥ 625 there was again very little change in the
appearance of the pigs. The square root of the remaining ADD values was then
101
used. These transformations increased the linearity of the response prior to
running the linear regression analysis.
The linear regression analysis generated an output table showing the gradients for
the fitted model, and the TBS values at the zero intercepts on the √ADD and start
weight axes. Having an intercept for a start weight of zero, an impossible weight,
would not have provided a useful reference point, since the corresponding TBS
would be zero, independent of ADD. The mean start weight of 54.0 kg was used as
the reference instead, and this value was subtracted from each of the individual
start weights (making 54 kg the new intercept point). This did not affect the
gradient of the fitted curve, nor the p value, but it did mean the intercept would be
at a weight of 54 kg.
Similarly at zero ADD the TBS would always be zero, independent of the start
weight. The mean ADD of the data points, with 35 < ADD < 625, across all three
experiments was 282 ADD. The corresponding square root is 16.8, and this value
was subtracted from each of the individual √ADD values (making 282 ADD the
new intercept point). Again this did not affect the fitted gradient or p value, but
gave an intercept value that was more relevant; in this case at 282 ADD
The last two data adjustments meant that, when using the output table from the
linear regression analysis to calculate what effect a change in start weight of 1 kg,
or a change of 1 ADD, would make to TBS, this would be for an initial start weight
of 54 kg and at 282 ADD.
102
6.2 Results
Flies were seen landing on the bodies of the pigs as they were being removed from
the body bags, and continued to land on them when they were placed into position
on the field. The flies were observed, over the following few days, laying eggs in
the natural orifices and around the head and anus in all of the pigs. In addition eggs
were laid around the nooses of the hanging and semi-recumbent pigs.
The hanging pigs and the upper portions of the semi-recumbent pigs were scored
using the scales developed in Experiment 1 (Chapter 4), and the control pigs and
the lower portions of the semi-recumbent pigs were scored using the scoring scale
developed by Megyesi et al. (2005) adjusted for a zero baseline. The
decomposition patterns of the upper and lower sections of the semi-recumbent
pigs followed the same stages and sequences as described by these scoring scales,
making them both sufficient and appropriate for scoring the animals in this
experiment.
All three groups of pigs showed similar features for the first few days
(discolouration and bloat), and this is reflected in the initial stages of both scoring
scales.
From 197 ADD onwards the patterns of decomposition differed between the
hanging and control pigs, while the semi-recumbent pigs started to display
features of both the hanging and the control pigs. Within 253 ADD the
decomposition pattern of the semi-recumbent pigs had started to become more
103
similar to the hanging pigs for the upper portion of the body which was not in
contact with the ground, while the portion of the body in contact with the ground
looked similar to that of the control pigs. These differences between the groups
are described below for a range of the ADD values.
Figure 6.2 shows the differences and similarities between the groups at the same
ADD values. The photographs are set out with the semi-recumbent pig displayed
between the hanging, above it, and the control, beneath it, to make comparing the
similarities between them easier.
97 ADD: Most of the hanging pigs needed to be re-hung as their weights had
caused the ropes to stretch and the metal S hooks to stretch. The hooks had to be
replaced with ropes pulling the support loop, previously hung from the S hook,
closer to the horizontal bar of the A-frame. All the pigs appeared remarkably fresh
and, despite having been out for 9 days, only one pig in each group had visible
maggots. The weather had been wet and windy with the coldest spring since the
early 1960s, which may have affected the flies ovipositing.
165 ADD: Hanging: All the pigs had a swollen head, torso and limbs and the
majority were starting to prolapse the bowel. The males had swollen testes and
umbilicus/penis. More than half of the animals had started to purge. A few had
drying skin on the snout and neck.
Semi-Recumbent: All had swollen heads, limbs and torso and all were purging
decomposition fluid. 60% had drying skin on the snout and the neck around the
noose rope.
104
Control: All had a swollen head with four pigs having maggot masses visible in the
mouth. All torsos were swollen with 60% showing small patches of drying skin on
the abdomen. 90% were purging decomposition fluid.
226 ADD: Hanging: All had drying snouts with lips drying and retracting. Maggots
were visible in the mouths of 50% of the pigs. All pigs had swollen limbs
protruding from a swollen torso with the bowel prolapsing. All had maggots visible
on the torso and showed evidence of purging. In the males the umbilicus/penis
area and the testes were swollen. Four pigs had skin slip on the lower abdomen.
Semi-Recumbent: All animals had a swollen head with drying snouts and lips.
Some pigs had brown drying skin on the necks particularly around the noose. They
all had a swollen torso with 50% showing skin slip on the lower abdomen. One
animal had a breached lower left abdomen/groin. 50% of the animals had maggots
and/or eggs in the crease between the lower limb and abdomen; in these pigs the
abdomen was almost resting on the lower limbs. All limbs were swollen with
upper limbs extended away from the body. All pigs had been purging.
Control: All animals had a swollen head which had started changing colour to dark
grey or black. Their torsos were swollen although some of these were no longer
rock hard but soft. The colour had changed to dark grey or black in 60% of the
pigs. The limbs remained swollen with a couple of animals displaying skin slip and
some hair loss.
The first beetle was seen on a control pig but was too quick to catch!
105
302 ADD: Hanging: The skin was drying on all of the pigs’ heads, three of the
heads had a small amount (less than 10%) of bone showing. The skin on the
shoulders and backs had also started to dry. The torso of each pig remained
bloated but no longer taut and the chest area had become sunken in some of the
pigs.
The lower abdomens had become dark grey and all had prolapsed bowels. All the
male pigs had swollen testes, and in three of the males the umbilicus and penis had
formed two separate holes. Both holes had combined to form one hole in one of the
other males, with a loop of gut protruding from the left side of the lower abdomen.
The limbs of all the pigs were still extended but some had started to droop.
A drip zone had formed below one of the pigs; this contained only maggots and
faeces.
Semi-Recumbent: All the pigs had dry skin on their heads which had started to
harden and mummify. The throats and necks were also drying and turning brown.
Three of the pigs had dry mummifying skin on their backs. The upper and lower
torso had started to look different with the lower torso having maggots in the
groin abdomen crease area of eight pigs. This area was breached and open in four
of the pigs with active maggot masses visible. The upper torsos were not breached
but some had patches of dry hard skin.
The upper limbs had started to droop in most of the animals and in all the pigs the
lower limbs were wet, with maggots in the groin creases even if they were not
breached.
106
Control: All showed darkening and drying of parts of the head. Four pigs had
foaming maggot masses in their mouths and breached throats. The ground side of
the head and throat was wet in most of the pigs. Three had maggots in the testes
and anus; the testes in the other seven were drying and blackening. Four pigs had
the ground side of the torso open with maggots present. Most of the pigs had
maggots in the axillae and groin.
353 ADD: Hanging: The pigs had elongated as the bloat decreased. This,
combined with stretching of the rope, suggested some of the animals would soon
be very close to touching the ground, with little hope of being able to lift them
higher because of the height of the suspension bar. All pigs had dry snouts and
most had sunken throats with the lips retracting and some skin slip. Only one pig
still had maggots visible in the mouth. In all pigs the torso had started to deflate
and become dark grey or black; all the bowels had prolapsed. The axillae of the
upper limbs were breached in more than half the pigs. Most pigs had some skin
slippage. Drip zones had formed beneath nine of the pigs and all contained
maggots in quantities from a teaspoonful to about 100 ml.
Semi-Recumbent: The decomposition pattern was clearly different between the
upper and lower sections of the animals requiring the use of two scoring scales.
From this point on the hanging scale was more suited to the upper section of the
body, and the amended Megyesi et al. (2005) scale was used for the lower section.
All pigs had drying snouts, caved in throats, and drying skin on the head. Less than
half had active maggot masses in the mouth. Three pigs had drying skin around the
shoulders and top of the back. The lower abdomen was wet and soft with maggots
in the groin and leg creases. In the males maggots were present in the
107
umbilicus/penis and the testes were breached. Small holes were seen in the skin of
the lower torso.
Control: All pigs had some maggot activity around the mouth and/or throat. In the
majority the throat had ‘caved in’. Moist decomposition could be seen. The
majority of the pigs had breached axillae with maggots visible there and in the
open anus and testes. Maggots were also present in the penis/umbilicus of three
pigs. The limbs were darkening with maggots visible in the groin of most pigs, with
breaches of the groins visible in some. Small holes were noted in the skin.
410 ADD: Hanging: Three pigs needed raising to prevent them touching the
ground and in all cases it was possible to do so using an additional noose.
All the heads were dry and mummifying with the skin becoming leathery and little
or no maggot activity visible. All the torsos were elongating post bloat with the
skin becoming dry and leathery on the shoulders and backs. The limbs were dark
coloured and drooping. The anuses were open, with a diameter greater than 10 cm.
Semi-Recumbent: All heads were dry and mummifying. The skin on the shoulders
and back of the upper torso was dark and mummifying. The upper limbs were
disarticulated. In contrast the lower torso had wet breached skin and visible
maggot activity. Half of the pigs had visible bone in the groin and or lower limbs. In
one of the animals a trail of maggots could be seen in a mass 1 m away from the
animal and crawling through the grass.
Control: All heads had moist decomposition in the throat and ground side, with
some mummifying of the skin on the upper side of the head. All torsos had areas of
108
drying skin with maggot activity visible where the skin was breached. Bone was
visible in the torso and limbs of four pigs.
469 ADD: Hanging: All heads were mummified with bone visible in most of the
heads and/or in the mouths.
Semi-Recumbent: All heads were mummified with bone visible in 90%. All upper
torsos were mummified. Lower torsos were breached with active maggots and
disintegration of the skin leaving lower limb bones visible in 90% of the pigs.
Control: Maggot activity visible in all pigs, skin on upper surface drying to form a
mummified blanket on most pigs. Bone was visible in the torsos of five pigs with
limb bones visible in 90% of the pigs.
525 ADD: Hanging: All pigs were mummified with bone visible on the head and in
one pig visible within the torso, through a hole in the skin. Drip zones contained
maggots with two containing hooves.
Semi-Recumbent: No change in the head and upper torso; all were mummified and
hard with some having holes or tears. This changed where the torso was in contact
with the ground. Here the skin was black and disintegrating or destroyed. The area
below and in front of the pig, where the lower limbs lay, was fatty and contained
visible bones including some ribs. Some animals had a few maggots still visible.
Ribs were visible in the fat. One pig remained almost intact in appearance: the
upper torso was mummified and there were maggots in the groin but little other
visible activity.
109
Control: All pigs had bone visible in the torso, on the head and in the limbs. The
skin was dry and mummified on the upper surface of most of the pigs. Most had
maggot activity under the skin where the body had become sodden in the rain.
575 ADD: Hanging: Two pigs were touching the grass and were raised up, no
beetles or beetle larvae were visible on them. All pigs were mummified. Vertebrae,
ribs, innominates and intestines were seen in some of the drip zones. In the male
pigs the anus and scrotum now formed one hole.
Semi-Recumbent: All pigs had mummified heads with bone visible and
parchment-like skin. The upper torsos were mummified. Two pigs remained
swollen but with hard dry skin. Lower torsos consisted of disintegrating skin with
visible bones in all but one pig. Limb bones were visible or could be felt in the
mummified skin.
Control: All pigs had a covering of mummified skin, some of which was still
attached to the head or the limbs. Beneath this bones were visible with some fat
and a few maggots. No skin visible on the groundside of the pig; bones were
visible.
632 ADD: Hanging: No change in appearance overall. However, in three of the
males the testicles were visible. Drip zone beneath one pig contained vertebrae,
long bones, scapula, ribs and an innominate.
Semi-Recumbent: Upper torso and head were dry and mummified. Lower torso
had bone visible within a fatty mass and mummified skin. No identifiable torso left.
110
One exception to this was the pig which remained swollen: the lowest part of the
lower torso was covered in hundreds of beetle larvae.
Control: The skin on the upper part of the pigs was mummified and in most cases
could be lifted to show bones beneath. Most bones were still fatty, although some
bones were clean. Maggot masses could still be seen in the axillae or anus of a few
pigs.
674 ADD: Hanging: Four pigs had small holes in the skin of the torso and three
pigs had their dried intestine dropped through the anus or in the drip zone.
Semi-Recumbent: Upper torso and head were mummified. Lower torso consisted
of bones and fat with small amount of mummified skin. One pig still had an
identifiable lower torso with bones visible and hundreds of beetle larvae (Figure
6.3).
Control: Dry mummified skin on top of bones and fat. Some bones were clean and
largely fat free and some remained in fat where the pig was on very wet ground.
The skin could be lifted from all the pigs.
761 ADD: Hanging: All pigs were mummified, the torsos were hard and hollow
sounding and bones could be felt through the skin. Vertebrae, ribs, innominates,
long bones and scapula were found in the drip zones. Bones could be seen caught
in the congealed fat in the anus. The first beetle larvae were found in the drip zone.
Semi-Recumbent: The upper torso could be lifted from the lower torso from which
it was completely separate. There was no change in the lower torso. One pig
111
remained soft around the lower abdomen where beetles were present. Ribs were
seen in the bones on the ground below the upper torso mingled with the lower
limb bones. Lots of maggot pupae could be seen inside the torso. Beetle larvae
were present in the fat of two pigs.
Control: All the skin was mummified and could be detached; the bones beneath
were largely clean with only a small amount of desiccated tissue or fat adhering.
829 ADD: Hanging: All pigs were mummified with some holes and tears in the
skin. Some pigs appeared elongated and empty looking. Maggots were seen in the
axilla of one pig.
Semi-Recumbent: No change in appearance, upper torsos were hard and hollow.
Control: Beneath the mummified skin the bones were largely clean and greasy.
923 ADD: Hanging: The pigs were hollow sounding and light, moving easily if
touched. The skin was fatty.
Semi-Recumbent: The upper torsos resembled the hanging pigs, being dry, hard
and sounding hollow. They could also be moved easily.
Control: Beneath the mummified skin some of the bones were in ‘hydrated’ fat, a
few had dried tissue attached, most were clean and greasy. The bones of the head
were clean and dry.
112
989 ADD: Hanging: No change in appearance of the pigs. Large amounts of fat
could be seen in the drip zones.
Semi-Recumbent: No change in appearance although there was a lot of fat on the
ground and ribs and scapula had fallen from some of the upper torsos.
Control: All pigs covered by some amount of detached mummified skin beneath
which were greasy and clean dry bones.
1078 ADD: The data loggers were extracted and the pigs removed for disposal. (A
few hanging pigs were left to see what would happen over the next few months.)
Hanging: When removing the data loggers some of the pigs were opened up
completely down the front. Bones could be seen adhering to the mummified skin
(Figures 6.4 and 6.5).
Semi-Recumbent: No change in appearance. Figure 6.6 shows how the mummified
upper torso had separated, and could be lifted away from, the lower torso. At the
end of the experiment the upper torsos were found to contain numerous pupae
cases.
Control: No visible changes.
113
2 ADD 13 ADD 97 ADD
Figure 6.2 Decomposition sequence for one hanging, one semi-recumbent, and one
control pig.
114
165 ADD 226 ADD 353 ADD
Figure 6.2 (cont.) Decomposition sequence for one hanging, one semi-recumbent,
115
410 ADD 469 ADD 525 ADD
116
575 ADD 632 ADD 674 ADD
117
761 ADD 829 ADD 923 ADD
118
989 ADD 1078 ADD
119
Figure 6.3 One of the semi-recumbent pigs showed a large mass of Necrodes
littoralis (Silphidae) beetle larvae, these were present on both sides of the pig, at
674 ADD.
120
Figure 6.4 and Figure 6.5 Bones could be seen adhering to the mummified skin of
the hanging pigs at 1078 ADD.
121
Figure 6.6 The mummified upper torso of the semi-recumbent pigs had separated,
and could be lifted away from, the decomposed lower torso (1078 ADD).
122
6.3 Statistical Analysis
When the TBS values for the hanging pigs and the TBS scores for the upper section
of the semi-recumbent pig were plotted against the square root of ADD, the slopes
for both were very similar (Figure 6.7).
Figure 6.7 Total Body Scores for Hanging (TBShang) for the upper sections of the
semi-recumbent pigs and the hanging pigs scored using the hanging scoring scale
from Experiment 1. Some offset has been added to the ADD values at the data
points to add clarity.
123
Plotting the PBSsurf = PBSTsurf + PBSLsurf for the control pigs, and PBS = PBST +
PBSL for the semi-recumbent pigs, the PBS values can be seen to be overlapping
(Figure 6.8).
Figure 6.8 Partial Body Scores (PBST + PBSL), lower limbs and torso, for the lower
sections of the semi-recumbent pigs, and the limbs and torso of the control pigs
scored using the modified Megyesi et al. (2005) scoring scale. Some offset has been
added to the ADD values at the data points to add clarity.
Comparing the TBS versus ADD responses for the hanging pigs and the upper
section of the semi-recumbent pigs, ANOVA produced a p value of p = 0.53, F2, 197 =
1402. Thus there is no statistically significant difference between the two groups
in their TBS versus ADD responses.
124
When the scores of combined ( PBSTsurf + PBSLsurf ) response to ADD was tested for
the Control Lower and SRL groups there was no statistically significant difference
with ANOVA giving a p value of p = 0.8, F2, 197 = 1157.
Treatment was shown to have a statistically significant effect on the TBS when
start weight was included in the linear regression. For both SRL and Control Lower
p < 0.001, F5, 194 = 487.8, and for both Hanging and SRU p < 0.001, F5, 194 = 576.9.
The effects of this are shown in Table 6.1 below.
Table 6.1 Sensitivities of the TBS responses to changes in the start weight and
ADD around a reference point of 282 ADD and a start weight of 54.0 kg. The Semi-
Recumbent Lower and Control Lower have partial body scores (PBS) out of a
possible 20 points using the amended Megyesi et al. (2005) scoring scale. The
Semi-Recumbent Upper and Hanging are TBS scored on the Hanging scale from a
possible 30 points.
Semi-
Control Semi-
Recumbent Hanging
Scenario Lower (total Recumbent
Lower (total Pigs
score of 20) Upper Body
score of 20)
Start weight = 54.0 kg

8.10 PBS 8.00 PBS 11.95 TBS 11.87 TBS
& ADD = 282
To gain an increase of +24.05

+56.3 ADD +43.63 ADD +29.93 ADD
1 TBS ADD
Increase of 1 kg in the
+0.01 TBS +0.01 TBS −0.03 TBS −0.03 TBS
start weight
125
6.4 Discussion
This experiment showed that the semi-recumbent pigs did not require a new scale
in order to score their decomposition, but that the scales used to score the hanging
pigs and the control pigs were sufficient. The hanging and control pigs showed
different patterns of decomposition between their groups, as in the previous two
experiments, with the semi-recumbent animals displaying similar decomposition
patterns to both the hanging and control pigs. The upper sections of the pigs could
be scored using the hanging pig decomposition scoring scale from Experiment 1,
and the lower sections using the amended Megyesi et al. (2005) scoring scale, with
a clear delineation between the two sections. If the entire body were present,
either scoring scale could be used but the hanging scale could be used to score all
three body sections, while the Megyesi et al. scale could only be applied to the
lower torso and limbs.
Decomposition is a complex and continuous process making the determination of
discrete stages difficult and subjective, leading to different researchers identifying
different decomposition stages (Adlam and Simmons, 2007; Archer, 2004;
Campobasso et al., 2001; Oliviera and Vasconcelos, 2010; Payne, 1965). In 2011,
Parks noted that it was unrealistic to have a “one size fits all” decomposition
model. Other researchers produced scales for scoring decomposition in
circumstances where the decomposition patterns were specific to the conditions of
death or discovery. Megyesi et al. (2005) produced a scale for bodies on the
surface based on Galloway et al.’s (1989) method; Heaton et al. (2010) for
submerged bodies; Gruenthal et al. (2012) for charred bodies. Where
decomposition patterns are not sufficiently different from current scoring scales a
126
new scale may not be warranted, even if the location and situation of the body is
different.
After death, the body is the subject of a battle between decomposition and
desiccation (Micozzi, 1991), the outcome of which depends on the body’s
immediate environment and factors such as the temperature, humidity, and insect
access. The outer surface of the body loses water to the air with the rate of loss
dictated by various factors including heat, aridity, and air flow.
Natural desiccation occurs in desert regions such as north Africa, Australia, costal
zones of Chile and parts of the United States, and northern Mexico (Micozzi, 1991).
External desiccation may not reach to the internal organs, where the decay is
largely driven by bacteria and enzymes and may continue whilst the outer skin
surface mummifies (Aufderheide, 2003). However, the rate of desiccation may be
very fast in hyper-arid regions such as the Sahara, Gobi, and the Atacama, where, if
desiccation occurs quickly enough before the onset of decay, the decay may be
arrested and mummification may be complete, leaving internal organs largely
intact (Aufderheide 2003, 2011; Micozzi, 1991). It is probable that the smaller the
body the more likely this is, as the surface area to volume ratio is higher with a
relatively larger surface area for the loss of water to occur over.
Mummification can take place in conditions other than hot dry deserts, as can be
seen in the mummified hanging pigs from this experiment, where desiccation is the
result of almost constant airflow over the exposed bodies. Mummified humans
have been found in the remote mountainous region of Laguna de los Momias in
Peru (Aufderheide, 2003) where the climate is predominantly rain and fog. At this
127
site more than 200 mummies were found which appeared to be the result of
natural mummification. There was no evidence of enhanced mummification, nor
was assistance such as evisceration seen where it was possible to view the
perineum, abdomen or chest.
Most naturally occurring mummies have no internal organs or soft tissues left as
these have decomposed. Examples of more visible differential decomposition can
be seen where bodies have been found indoors in bed. Here the body beneath the
covers may be decayed or decaying whilst hands and arms outside the bed covers
have mummified. A body of an 18 year old girl, who had been buried in a shallow
grave, was found to have some areas of the body reduced to skeletal elements, the
upper limbs mummified, and the lower limbs converted to adipocere, a fatty
substance usually occurring in very wet or submerged bodies. This grave, which
was in clayish soil, had been subjected to large quantities of rain and, clearly,
different body parts had undergone different forms of decay and mummification
(Cunha and Pinheiro, 2007). The semi-recumbent pigs in this experiment also
displayed distinctly differential decomposition between the mummified upper,
suspended, and skeletonised lower, unsuspended, sections of the bodies.
6.4.1 Semi-Recumbent Upper Section (SRU)
The suspended portion of the semi-recumbent pigs followed the same pattern of
decomposition and mummification as the fully suspended pigs, with both exposed
to the wind and sun in the same way. This mummification continued to the point
where the lower body, in contact with the ground, had been destroyed; Figure 6.9
shows this clear demarcation. A few maggots can be seen on the outside of the
mummified skin although it was more usual to see none.
128
Figure 6.9 Showing the delineation between the mummified upper body and the
decomposed lower body of one of the semi-recumbent pigs.
The mummified skin probably acted in the same way as in the fully suspended
pigs, providing protection from sun and adverse weather conditions for the
maggots (Campobasso et al., 2001) which rapidly move inside the head and off the
exposed surface of the torso. As the decomposition progressed the lower section
and upper section of the pig became detached from each other. As a result of this
separation, combined with the action of maggot mass writhing and gravity,
129
maggots would have fallen through the open torso onto the decomposing carcass
beneath, effectively forming the hanging drip zone. In the case of a fully suspended
pig, once fallen, the maggots would have difficulty regaining access to the carcass
and would be dependent on food fall from the body above. In the semi-recumbent
case the maggots did not need to move back since a food supply was readily
available in the carcass beneath. The maggots did not appear to move back up into
the upper section and none were seen on the outside of the pigs.
The experiment using clothed hanging pigs (Chapter 5) found the quantity of
maggots in the drip zone beneath the unclothed pigs was greater than beneath the
clothed pigs, where maggot loss was impeded as they were caught in the clothing.
Since the opening beneath the semi-recumbent upper bodies was much larger than
in a fully suspended pig, the maggot loss would be greater with less to prevent the
fall. This did not, however, slow the decomposition of the upper bodies.
Despite large numbers of adult beetles and beetle larvae on the lower sections of
the pigs, none were seen on the outside of the upper sections or at the transition
area, although access to the upper section would not have been difficult.
Presumably there was sufficient food supply in the lower sections.
As the mummification progressed, the upper torso became stretched and took on
the appearance of an empty coat with the bottom resting on the ground. This could
be lifted easily and would move in the strong winds.
130
6.4.2 Semi-Recumbent Lower Body (SRL)
The lower sections of the pigs were resting directly on the ground, which remained
damp or wet for much of the experiment, and their decomposition pattern was the
same as the control pigs. The skin around the groin area was rapidly breached and
active foaming maggot masses were visible. These maggots, although present on
the outside of the lower abdominal and flank areas, were not seen to move up onto
to the upper sections of the pigs. Where the upper torso was an open hole, the
maggots dropped onto the lower section adding to the already present maggot
masses. These maggots were not observed moving back to the upper torso,
although this was still in contact with the lower section for much of the time,
presumably continuing to feed on the lower section.
A blanket of mummified skin formed on the upper exposed surface of the control
pigs and remained there, albeit detached from the bones beneath. In the semi-
recumbent pigs the skin on the lower section did not form a mummified covering
but remained wet and pliable possibly aided by the maggot foam and fall, and
decomposition fluid from the upper section. This soft skin would be more
attractive to ovipositing flies and easier for the maggots to eat. The protection
afforded to the maggots on the control pigs from the mummified skin could have
been provided by the mummified upper torso which maintained its shape
providing a tent which reached down to the ground or just above it. The presence
of numerous pupae cases within the upper torsos at the end of the experiment
would indicate that these provided a suitable environment for the post-feeding
maggots to pupate.
131
Adult beetles and beetle larvae were present on the lower section of most of the
pigs but were not seen on the upper section. The Silphidae are some of the most
commonly found beetles on this site. One pig had large numbers of beetle larvae
Necrodes littoralis (Silphidae) and some adult beetles on the lower section,
particularly the flanks, with no maggots observed on it. There was no obvious
explanation for this.
The decomposition of the semi-recumbent lower section followed the same
pattern as the controls and could be scored using Megyesi et al.’s (2005) scoring
scale. However, as only the limbs and torso can be used the maximum score
possible is 20 and this would make it difficult to use the prediction equations
created for a full body score to calculate the PMI. The process of decomposition
across the body is not uniform (Campobasso et al., 2001; Megyesi et al., 2005) and
a simple scaling up of the decomposition score would not reflect the level of
decomposition that would have occurred in the head at that stage, thus decreasing
the accuracy of the PMI estimate for that individual. Comparing the partial body
scores from the 30 control pigs used across the entire study showed that the heads
were further decomposed than the torso and limbs at the same ADD. The ADD data
points examined were 279 ADD, 282 ADD and 303 ADD these being the closest
points, for each year, to the 282 ADD used for the calculations in this experiment.
While the results of this experiment found that the effect of start weight on TBS
was statistically significant, in practice it doesn’t make a large enough difference to
the TBS values. Although the semi-recumbent lower section of the pig and the
truncated scoring for the control pigs showed that the TBS would increase with an
increase in start weight, an increase of 100 kg would be required to see an increase
132
of 1 TBS. Why this should show as an increase is not clear. However, these pigs
were scored out of a maximum possible 20 points since the heads could not be
included. This makes it difficult to compare the trends with those of the full body
score. The effect of start weight on TBS is examined more fully in Chapter 7 using
results from across all three experiments.
Within these results it can be seen that the hanging and semi-recumbent upper
bodies show the same levels of TBS response to ADD, and that the lower sections
of the semi-recumbent bodies and controls are different to the upper body groups,
but show the same level of TBS response, to ADD, as each other.
While the results may not be directly applicable to scoring semi-recumbent
hangings for humans, there is no reason to believe that the decomposition pattern
of the upper body would not follow that of a fully suspended hanging body, or that
the lower section would not follow the pattern of a body on the surface. This being
the case the use of the hanging scale would provide scores for all three body
sections and should be easier to use than the partial score provided by the lower
section when estimating the post-mortem interval.
133
7 Analysis of the Combined Data Set and ADD Prediction
Tables for Hanging and Surface Bodies
7.1 Introduction
After completion of the experiments, and following the statistical analysis for the
first experiment, it became clear that using the largest possible set of data to carry
out some parts of the analysis would provide a clearer picture of what was
occurring. In each of the three experiments carried out in this study, groups of
hanging and control pigs were studied at the same geographical site and the same
time of year over three consecutive years. The decomposition of all the pigs was
assessed by the same observer using the same observation and data collection
techniques. This ensured there were no inter-observer errors, and reduced the
likelihood of variations in the method used.
This chapter presents an analysis of the combined data set from all three
experiments to determine:
• The degree of consistency in the TBS versus ADD responses for the 30
hanging pigs and 30 control pigs from the three experiments.
• The effect of the initial weight of the pigs on their rate of decomposition
using data from 60 pigs, comprising the 30 hanging and 30 control pigs.
134
• A set of ADD prediction tables for hanging pigs and for control pigs using
TBS data from the 30 hanging pigs and the 30 control pigs from all three
experiments.
Regression analysis can be applied to the observed TBS and ADD values to produce
an equation giving the expected TBS for a given ADD, with ADD as the explanatory
variable. Depending on the complexity and order of the regression equation, it may
be possible to invert it to obtain an expression giving the expected ADD
corresponding to an observed TBS. In general, however, it will not be possible to
obtain confidence intervals for the ADD values in this way, but a PMI estimate
must always be given as a range of ADD values, or the corresponding dates, with
the accompanying confidence level.
To obtain the required confidence intervals a form of inverse prediction can be
used, such as that employed by Moffatt et al. (2016), to produce prediction tables
that give the expected ADD value, and the ADD ranges for various confidence
levels, for each possible TBS value. The tables given below for the control and
hanging pigs will enable these results to be compared with those from human
cases to assess the validity of using pigs as human analogues in these two
situations.
7.2 Method
Data from a total of 60 pigs were used: 10 fully suspended hanging pigs and 10
control pigs laid directly onto the ground for each of the three experiments
conducted over consecutive years. The decomposition of the two groups of pigs
135
was scored in the same way for each experiment, as described in the General
Materials and Methods chapter (Chapter 3), to provide Total Body Scores (TBS)
and Accumulated Degree Days (ADD) values for each observation day. All the
control pigs were scored using the scale developed by Megyesi et al. (2005),
adjusted to range from 0 to 32, as a score of zero more logically quantifies the state
of no decomposition described by Megyesi et al. (2005) as fresh (Moffatt et al.,
2016); whilst the new scale developed in Experiment 1 (Chapter 4) was used for
all the hanging pigs. The temperature data for the ADD calculations were collected
from the data loggers attached to the hanging frames as described in the General
Materials and Methods chapter.
7.2.1.1 Data Pre-Treatment and Transformations
The various data analyses all required the use of linear regression analysis as a
first step. However, as can be seen from Figures 7.1 and 7.2, the TBS versus ADD
responses for the control and hanging pigs do not follow a straight line, so the data
was pre-treated prior to each test.
For the inter-year consistency and the effect of start weight tests, the data was first
filtered to remove the data points having ADD ≤ 35 or ADD ≥ 625 (the same range
of ADD values examined in the decomposition analysis undertaken for Experiment
3). At very low ADD values, as the decomposition is first getting underway, and at
high ADD values, as the TBS reaches the mid-20’s and skeletonisation occurs, it
would be expected that the relationship between TBS and ADD may show a greater
degree of variability than during the period when the majority of decomposition
136
occurs. Filtering the data in this way removed these tail regions and enabled the
analysis to focus on the main period of decomposition.
The aim of the inter-year consistency test was to investigate to what extent the
relationship between TBS and ADD varied across the different years, rather than
trying to explore the nature of that relationship per se, so no further pre-treatment
of the data was required.
To explore the effects of the pig start weight a more linear relationship between
the (transformed) TBS and ADD values was required so that the gradients of the
fitted linear models could be used to examine the rates of variation (sensitivities)
in the TBS responses (as the dependent variable). Having pre-filtered the data as
described above, it was found that the (transformed) TBS versus ADD response
could be made fairly linear by taking the square root of the ADD values.
Offsets were also applied to the start weight and √ADD values. This did not affect
the fitted response gradients produced by the linear regression analysis, but was
done for convenience to make interpreting the results simpler. A start weight of
zero is not possible, thus the mean start weight of 40.4 kg was used as the
reference instead, and this value was subtracted from each of the individual start
weights. Similarly, at zero ADD the TBS would always be zero, independent of the
start weight. The mean ADD of the remaining data points, with 35 < ADD < 625,
was 282 ADD. The corresponding square root is 16.79944, and this value was
subtracted from each of the individual √ADD values.
137
To produce the ADD prediction tables it was necessary to use the full set of
available data, to cover the full range of TBS and ADD values. The method used to
calculate the tables, however, required the (transformed) TBS versus ADD
response to be as linear as possible, and simply taking the square root of the ADD
values (equivalent to squaring the TBS values) was no longer good enough.
For the control pigs it was found to be sufficient to raise the power of the TBS
values by an appropriate power factor. The power factor used was chosen using an
iterative manual search to maximise the corresponding coefficient of
determination between the transformed TBS values and the (untransformed) ADD
values.
For the hanging pigs a more complex transformation was required due to the S-
curve running through the untransformed TBS versus ADD response (Figure 7.2).
In this case an offset was first subtracted from the TBS values, with the offset value
chosen to line up with the inflection point of the (approximate) S-curve. The
overall TBS versus ADD response could then be linearised by raising the offset TBS
values by an appropriate power factor. It was found that a single power factor
could be used for both ‘halves’ of the response, on either side of the inflection
point. The offset and the power factor used were again both chosen using an
iterative manual search to maximise the corresponding coefficient of
determination between the transformed TBS values and the (untransformed) ADD
values.
138
7.2.1.2 Inter-Year Consistency Test
To test the consistency of the relationship between the TBS and ADD data across
the three experiments for the hanging and control groups, the data for each group
were pre-treated as described above, and then linear regression analysis was
carried out followed by ANOVA.
7.2.1.3 Effects of Start Weight
To investigate the effect of the start weight of the pigs, and using ADD as the
explanatory (x axis) variable, the data for each group were pre-treated as
described above. A least squares linear regression was then carried out between
TBS and √ADD and an ANOVA carried out to determine whether start weight had a
significant effect on the rate of decomposition.
7.2.1.4 Prediction Tables
To produce the prediction tables all the values of TBSsurf and TBShang were used to
produce least squares linear regression models for the control and hanging groups
against ADD as the explanatory variable. Prior to running the regression analyses
the TBS values were transformed as described above to provide a more linear
response against ADD. It was not found to be necessary to transform the ADD
values.
Regression analysis was then used to fit linear models to the transformed TBS
versus ADD responses. Finally the Inverse Prediction approach described by
Moffatt et al. (2016) was used to create ADD prediction tables for the hanging data
and the control data, showing the expected ADD estimates and the corresponding
139
ADD ranges for various confidence intervals across the range of possible TBS
values.
7.3 Results
7.3.1 Consistency Across Successive Years
For 35 < ADD < 625 the ANOVA showed the year had no statistically significant
effect on the TBS to ADD relationship returning a p value of p = 0.953,
F5, 714 = 1261.
Figure 7.1 Total Body Score plotted against Accumulated Degree Days for all the
control pigs for years 2011, 2012 and 2013.
140
Figure 7.2 Total Body Score plotted against Accumulated Degree Days for all the
hanging pigs in years 2011, 2012 and 2013.
Figures 7.1 and 7.2 provide a visual comparison of the TBS versus ADD data sets
from the three experiments for each of the two groups, hanging and control. These
show that there is indeed a high degree of consistency between the three years.
7.3.2 Sensitivity of TBS responses
From the linear regression analysis, for a pig with a start weight of 40.4 kg at
282 ADD (square of 16.8):
141
TBS for a control pig: 14.30 with a 95% confidence interval of ±0.20
TBS for a hanging pig: 13.74 with a 95% confidence interval of ±0.49
At this point (282 ADD) the control pigs, as a group, had decomposed faster than
the hanging pigs. (In practice the TBS values would always be integers, but non-
integer values are useful here for exploring the sensitivities in the TBS responses
to changes in the independent variables.)
Table 7.1 provides a summary of the sensitivities of the TBS responses to changes
in start weight and ADD.
7.3.2.1 Effect of Increase in Start Weight on Decomposition Scoring
The analysis showed that the pig start weight had a statistically significant effect
on the rate of decomposition measured by TBS and ADD, for both the hanging pigs
and the control pigs, with a p value of p < 0.05, F5, 714 = 1962.
For an increase of 1 kg in the start weight (40.4 kg), at 282 ADD the rate of
decomposition for the hanging pigs, as measured by the total body score, slowed
more than that of the control pigs (Table 7.1).
7.3.2.2 Effect of Increase in ADD on Decomposition Scoring
For an increase in the √ADD of 1 at 282 ADD, and a start weight of 40.4 kg, there
was an increase in the TBS for both the control and hanging pigs. At this point the
hanging pigs were decomposing at a faster rate than the control pigs. For an
increase of 1 in the √ADD the increase in TBS was 8.5% for the control pigs and
142
11% for the hanging pigs. A higher change in ADD was needed to produce an
increase of 1 in the TBS for the control pigs than for the hanging pigs (Table 7.1).
Table 7.1 Sensitivities of the TBS responses to changes in the start weight and
ADD around a reference point of 282 ADD and a start weight of 40.4 kg.
Scenario Control Pigs Hanging Pigs
Start weight = 40.4 kg & ADD = 282 14.30 TBS 13.74 TBS
To gain an increase of 1 TBS +28.1 ADD +22.5 ADD
Increase of 1 kg in the start weight −0.09 TBS −0.10 TBS
7.3.3 ADD Prediction Tables
Tables 7.2 and 7.3 show the estimated Accumulated Degree Days (ADD) in °C and
various prediction intervals for the possible range of Total Body Scores for the
control pigs, on the surface, (TBSsurf) and for the hanging pigs (TBShang) using the
statistical analysis given in the Method section above. The grey cells show negative
TBS values, which could not occur in practice, but which are predicted from the
data ‘spread’ and have been included purely for completion of the table. In practice
these values should be read as zeros. Figures 7.3. and 7.4 display the same data in a
graphical format. The ‘wiggles’ in Figure 7.4 are a consequence of applying the
offset when transforming the TBS values.
143
Table 7.2 Showing the estimated Accumulated Degree Days (ADD) in °C and
prediction confidence intervals for Total Body Scores, TBSsurf, for the control pigs,
on the surface, using the statistical analysis described in the Method section above.
Note: the greyed out areas show negative TBS values which could not occur in
practice but which are predicted from the data ‘spread’ and have been included
purely for completion of the table.
Lower Limit ADD Upper Limit

TBSsurf
95% 75% 50% Estimate 50% 75% 95%
0 -148 -83 -46 7 60 98 162
1 -145 -81 -43 10 63 101 165
2 -139 -74 -37 16 69 107 171
3 -129 -65 -28 25 79 116 180
4 -118 -54 -16 37 90 128 192
5 -104 -40 -2 51 104 142 206
6 -88 -23 14 67 120 158 222
7 -69 -5 32 86 139 176 240
8 -49 15 53 106 159 196 260
9 -27 37 75 128 181 218 282
10 -3 61 99 152 205 242 306
11 23 87 124 178 231 268 332
12 50 114 152 205 258 296 359
13 80 144 181 234 287 325 389
14 110 174 212 265 318 356 419
15 143 207 244 298 351 388 452
16 177 241 279 332 385 422 486
17 213 277 314 367 420 458 522
18 250 314 351 404 457 495 559
19 289 353 390 443 496 534 598
20 329 393 430 483 536 574 638
21 371 434 472 525 578 616 679
22 414 478 515 568 621 659 723
23 458 522 560 613 666 703 767
24 504 568 606 659 712 749 813
25 552 616 653 706 759 797 861
26 600 664 702 755 808 845 910
27 650 714 752 805 858 896 960
28 702 766 803 856 910 947 1011
29 755 819 856 909 962 1000 1064
30 809 873 910 963 1017 1054 1118
31 864 928 966 1019 1072 1110 1174
32 921 985 1022 1076 1129 1167 1231
144
Table 7.3 Showing the estimated Accumulated Degree Days (ADD) in °C and
prediction confidence intervals for Total Body Scores, TBShang, for the hanging pigs
using the statistical analysis described in the Method section above. Note: the
greyed out areas show negative TBS values which could not occur in practice but
which are predicted from the data ‘spread’ and have been included purely for
completion of the table.
Lower Limit ADD Upper Limit

TBShang
95% 75% 50% Estimate 50% 75% 95%
0 -224 -149 -104 -41 21 66 141
1 -187 -112 -67 -4 58 103 178
2 -152 -76 -32 31 93 138 213
3 -119 -43 1 64 127 171 246
4 -88 -12 32 95 158 202 278
5 -59 17 61 124 187 231 306
6 -32 43 88 151 213 258 333
7 -8 67 112 175 237 281 357
8 13 89 133 196 258 303 378
9 31 106 151 213 276 320 396
10 44 120 164 227 290 334 409
11 52 127 172 234 297 341 417
12 59 135 179 242 304 349 424
13 73 148 193 255 318 362 438
14 90 166 210 273 336 380 456
15 112 187 231 294 357 401 477
16 136 211 255 318 381 425 501
17 162 238 282 345 407 452 527
18 191 267 311 374 436 481 556
19 222 298 342 405 467 512 587
20 255 331 375 438 501 545 620
21 291 366 410 473 536 580 656
22 328 403 447 510 573 617 693
23 366 442 486 549 611 656 731
24 407 482 526 589 652 696 772
25 449 524 568 631 694 738 814
26 492 568 612 675 737 782 857
27 537 613 657 720 783 827 903
28 584 659 704 766 829 873 949
29 632 707 752 814 877 921 997
30 681 757 801 864 926 971 1047
31 732 807 851 914 977 1022 1097
32 784 859 903 966 1029 1074 1149
145
Figure 7.3 Total Body Score TBSsurf against Accumulated Degree Days (ADD), for
the control pigs, showing estimates and inverse prediction intervals from the
linear model overlaid.
146
Figure 7.4 Total Body Score TBShang, of hanging pigs against Accumulated Degree
Days (ADD) showing estimates and inverse prediction intervals from the linear
model overlaid. The kinks in the curves result from applying an offset to the TBS
values as described in the Method section.
7.3.3.1 Linear Regression Lines Used for the ADD Prediction Tables
The central “ADD Estimate” values shown in Tables 7.2 and 7.3 were produced by
inverting the fitted line equations, for TBS versus ADD, generated by the linear
regression analysis.
For the control pigs the TBSsurf values were raised to the power of 1.72 with the
fitted regression line for TBSsurf1.72 versus ADD giving a coefficient of
147
determination of r2 = 0.94. The high value of r2 indicates that the regression line
was a relatively good fit.
The regression line equation for the control pigs was:
TBSsurf1.72 = 0.36317 × ADD − 2.64
which can be inverted to give:
ADDest = (2.64 + TBSsurf 1.72) / 0.36317
⇒ ADDest = 2.64 / 0.36317 + TBSsurf 1.72 / 0.36317
⇒ ADDest = 7.27 + 2.754 × TBSsurf 1.72
As an example, for an observed TBSsurf value of 18 the corresponding ADD estimate
(as shown in Table 7.2) is:
ADDest = 7.27 + 2.754 × 181.72 = 404
For the hanging pigs the TBS values were first offset by an adjustment of 11, so
that an original TBShang of 11 would become 0 and any TBShang values below 11
would become negative. For convenience these offset TBS values are represented
below using B = TBShang – 11. The resulting offset TBS values, B, were then raised
to the power of 1.51. However, where the original TBShang was less than 11, making
the offset TBS value B negative, this negative argument was first changed to
positive, then raised to the required power and the result reverted back to
148
negative to get the required value for the transformed TBS. The fitted line for the
transformed TBS versus ADD had a coefficient of determination of r2 = 0.92, again
indicating that the regression line was a relatively good fit.
For the hanging pigs, the transformed TBS values can be represented as:
sgn(B) × | B |1.51
where B = TBShang – 11; |B| represents the absolute value of B, such that |B| = B
when B ≥ 0, and |B| = –B when B < 0; and sgn(B) represents the sign of B, such that
sgn(B) = +1 when B ≥ 0, and sgn(B) = –1 when B < 0.
The regression line equation was:
sgn(B) × | B |1.51 = 0.13555 × ADD − 31.765
which can be inverted to give:
ADDest = (31.765 + sgn(B) × | B |1.51) / 0.13555
⇒ ADDest = 31.765 / 0.13555 + sgn(B) × | B |1.51 / 0.13555
⇒ ADDest = 234.3 + 7.38 × sgn(B) × | B |1.51
Example calculations for both positive and negative values of B are given in
Appendix 2.
149
(Note: sufficient decimal places have been shown that the ADD estimate values
shown in Tables 7.2 and 7.3 can be reproduced to within ±0.5 ADD.)
7.4 Discussion
7.4.1 Comparison Between Years
For the hanging and control groups of pigs the relationship between TBS and ADD
is consistent across the three years’ experiments.
7.4.2 Effect of Start Weight on Decomposition
Whilst the start weight of the pig has a statistically significant effect on the rate of
decomposition, because of the large increase in weight needed to result in a
measurable change in TBS it may make very little difference to the actual score,
especially given the scoring is restricted to whole numbers (integers). In order to
effect a decrease of 1 to the total body score for a hanging pig at 282 ADD the
weight would have to increase by 11.6 kg, and for a control pig by 10.5 kg. The pigs
used in this study ranged in weight from 19.5 kg to 69.2 kg. The TBS scoring tables
used to calculate PMI range from 0 for a fresh body to 32 for total skeletonisation
(clean skeleton) with TBS score being the sum of the score for three areas, the
head and neck, the torso, and the limbs. As the TBS scores are assigned to the state
of visible decomposition it is clear that large weight increases would be necessary
before visible changes to the decomposition were evident.
150
7.4.3 Pig Weight Differences and this Study
The pigs used for these experiments were not bred for experimentation but bought
from farmers breeding for the consumer market. Growth rate in pigs bred for
eating is rapid and an increase in age of a few weeks makes a great difference to
the size and weight of the pigs. Ideally, to remove or at least minimize the variables
possibly acting on the decomposition rate, the pigs used in all the experiments
would have been of approximately the same size and weight. Because the
experiments were carried out at roughly the same time each year (to enable
sharing of the control pigs with other researchers and thus minimize the number
of animals that needed to be slaughtered), and the farmers could not guarantee the
pigs would be ready for killing at the same weight each year, this meant that it was
not possible to ensure that the pigs were all approximately the same size year on
year or, indeed, in the same year, particularly when the pigs came from several
farms. Consequently, the dependence on the farmers’ decision on when and which
pigs would be ready for slaughter when we required them resulted in a large
variation in pig weight.
7.4.4 Obesity
The current, worldwide, increase in obesity may mean that effects on TBS brought
about by increased body weight will have to be taken into account when
calculating the PMI estimate. “The worldwide prevalence of obesity more than
doubled between 1980 and 2014” (WHO, 2015). In adult humans weight ranges
are delineated using the BMI (Body Mass Index) which is calculated by dividing the
body weight in kilograms by the square of the height in metres, although this
measure does not take into account the known differences in the amount of normal
151
fat deposition between males and females, or between races, or for athletes. BMI
tables are readily available. For an adult 1.75 m tall the ‘normal’ BMI range of 18.5
to 24.9 would equate to a weight range of 56.7 kg to 76.3 kg. The pigs used in these
experiments ranged from 19.5 kg to 69.2 kg, the heaviest being towards the upper
end of the normal BMI range for a 1.75 m adult human.
The use of pigs as human analogues is well established (Forbes et al., 2005;
Myburgh et al., 2013; Paczkowski et al., 2015; Roberts and Dabbs, 2015;
Schotsmans et al., 2012; Wilson et al., 2007). Pigs have similar gut bacteria to
humans (Anderson and Van Laerhoven, 1996). They are largely hairless, and their
body mass and muscle-to-fat ratios are similar to adult humans (Anderson, 2010;
Catts and Goff, 1992). Furthermore, the arthropod species density is similar on a
decomposing adult human and comparable sized pig (Anderson, 2010; Catts and
Goff, 1992; Schoenly and Hall, 2002; Rodriguez and Bass, 1983; Schoenly et al.,
2006). Hence they are widely used as human analogues.
Worldwide, in 2014, over 1.9 billion adults (39%) were overweight and 600
million (13%) were obese (WHO, 2015), so the reality of having bodies with a start
weight in the 90 kg to 140 kg range is increasing. At these increased weights the
effect on TBS, seen in these experiments, may become a problem. If TBS is lower
than currently expected at a given ADD this leads to the prospect that accuracy of
PMI estimation is compromised with the time since death being under estimated.
Further studies with carcasses in the obese and morbidly obese weight ranges may
be necessary to quantify the effects of higher start weights on TBS. Additionally, as
excess weight gain in humans tends to be composed of 60% to 80% fat and 20% to
40% lean mass, the use of pigs as human analogues (Anderson, 2010; Catts, 1992;
152
Schoenly et al., 2006; Swindle et al., 2012) may not be appropriate for
decomposition studies on obese bodies, and the results obtained may not be valid.
The pigs used in this study were sub-adult age range, at around 16-20 weeks, and
as such the weight differences between them were due mainly to increase in
muscle, in contrast to the weight gained in obese humans which is largely fat.
These pigs were produced as pork pigs (with lean meat) and would normally be
slaughtered young. Large heavy pigs could be used and those kept for bacon may
weigh between 175 and 220 kg. These are, however, still largely lean meat. There
are pigs bred for their high lard content but it is not clear whether these would be
suitable analogues for obese adults.
It cannot be assumed that the decomposition rate and pattern in obese individuals
or high lard pigs will remain the same as in these experiments. The presence of
high levels of fat may act to decrease the rate at which a body cools by acting as an
insulator (Tracqui, 2000) and the consumption of the carcass by necrophagous
insects and maggots may alter if it has a high fat content. Fat contains less water
than lean muscle.
It should also be noted that obesity levels in children are also rising, with WHO
figures for 2013 showing 42 million children under 5 to be overweight or obese – a
problem which may further complicate the issues of calculating PMI for juveniles.
7.4.5 ADD Prediction Tables
Work by Megyesi et al. (2005) used a scoring scale, based on an earlier scale by
Galloway et al. (1989), to score decomposition in human cadavers. The resulting
153
TBS values were combined with accumulated degree days (Vass et al., 1992) to
estimate post-mortem interval using statistical analysis and a more scientific
approach than was previously in use. Later use of this work identified flaws in the
formula produced, the application of which can lead to inaccurate estimates of
time since death. These flaws have been addressed in a paper by Moffatt et al.
(2016) which included prediction tables for estimating ADD, and confidence
intervals, from a given TBS value.
The prediction tables produced in this study use the same method. For both the
hanging pig and control pig data, the fitted regression line equation can be
inverted to estimate the ADD value most likely to be associated with a given TBS
value. As that ADD value increases or decreases the TBS value is less likely to be
associated with it, but the corresponding confidence intervals cannot be obtained
directly. The use of the inverse prediction method employed by Moffatt et al.
(2016), and reapplied here, allows the appropriate ADD confidence intervals to be
calculated as a function of the observed TBS.
The resulting tables provide a quick and simple way to determine the most likely
ADD value for any given TBS value, and the confidence intervals around the ADD.
When combined with the local temperature data, the estimated time since death
can thus be calculated, together with the prediction intervals giving the range of
time on either side of the central estimate.
The regression line equation and the resulting ADD prediction table (Table 7.2) for
the control pigs can be compared with that produced by Moffatt et al. (2016) for a
subset of the human cadavers originally studied by Megyesi et al. (2005). The ADD
154
prediction table produced by Moffatt et al. has a dramatically wider range of ADD
values than those shown in Table 7.2, extending as far as 16,830 for the 95% upper
confidence limit for the maximum TBSsurf value of 32. Moffatt et al. raised the
TBSsurf values to the power of 1.6, rather than the value of 1.72 used here, but by
far the biggest difference in the two analyses is that Moffatt et al. took the base-10
logarithm of the ADD values. While this resulted in a high value for the coefficient
of determination of r2 = 0.91 in the linear regression analysis, the use of the
logarithm function makes the estimated ADD values at higher TBS values
extremely sensitive to any errors in the original data set; and the data set was
somewhat sparse with only four data points above 280 ADD. In fairness the
authors did draw attention to these limitations, and their revised analysis is still a
significant improvement on the original results of Megyesi et al. (2005).
In contrast, the prediction tables presented here have been prepared using a large
data set which, generally speaking, should increase the accuracy of the statistical
analysis on which they are based. Each of the 30 pigs involved in this study was
observed and scored regularly and repeatedly throughout the decomposition
process, rather than the single point observations that were used in Megyesi et al.’s
original study (being based on photographs taken after each of the bodies was
found). Furthermore, the ADD values for the present study were derived from
temperature measurements made in the immediate vicinity of the pigs, while
Megyesi et al. had to rely on data from nearby weather stations, which
nevertheless could still differ significantly from the conditions local to each body
(Dabbs, 2010, 2015).
155
It should be borne in mind, however, that the ADD prediction tables presented
here have been prepared using data from pigs and may not be directly applicable
to humans. To get the same level of accuracy for humans would require data from
a large number of individuals for whom the time of death was known and for
whom observations started at death. This coupled with accurate temperature data
for the surrounding area (Dabbs, 2010, 2015), plus any necessary site-specific
correction factor, would provide a data set equivalent to that used in the
production of the tables from this study.
The existence of ‘body farms’ and taphonomic research sites using human
cadavers makes this more feasible but would require many bodies and much
cooperation. In the UK, human cadavers are not used for this type of research thus
pigs, which are the best human analogues available, have been used instead, with
the advantage that larger numbers could be studied at the same location over a
period of several years and differing weather conditions. Consequently there was
less chance of obtaining ‘odd’ one-off results.
It is perhaps interesting to note that the power factor of 1.51, which was used in
transforming the TBShang values prior to the linear regression analysis, is very close
to the ratio of 3:2 between the volume and surface area of a three-dimensional
object. It may be tempting to ask if this relates in some way to the progression of
the decomposition of a hanging body, or its consumption by maggots, and
furthermore whether the difference between this power factor and the value of
1.72 used with the control pigs relates somehow to the collapsing of the bodies
lying on the ground affecting the available surface area. At this point, this can only
be speculation but it may indicate another line of future investigation.
156
8 Discussion and Conclusions
This was the first thorough, robust and qualitative study into the decomposition of
hanging bodies using large numbers of subjects. The primary aim was to develop a
method for estimating the post-mortem interval (PMI) of hanging bodies based on
decomposition scoring and accumulated degree days (ADD). The findings showed
distinct differences between the patterns of decomposition in hanging and surface
pigs, leading to the production of a novel decomposition scoring scale for hanging
bodies.
The new scale scored the same body regions over the same scoring ranges as the
scale produced by Megyesi et al. (2005) for bodies lying on a substrate, but started
with a score of zero for a fresh body making it more intuitive to use (as opposed to
a score of 3 in the original Megyesi et al. scale). The new scoring scale also shared
many of the same decomposition descriptors as the Megyesi et al. scale, and was
constructed as far as possible so that a given score (for each body region)
corresponded to a similar overall level of decomposition. Consequently, the
differences in the total body score (TBS) versus ADD responses for the hanging
and control pigs did appear to match the observed differences in the overall rates
of decomposition. Amendments were made to some of the decomposition
descriptors to reflect observed behaviour that was specific to pigs. This can be
seen in the scoring sheet for the torso, where the change in shape of the male pigs
and the opening of the penis and umbilicus to one hole are scored. This reflects a
difference in the pig anatomy that would not be seen in humans. These
amendments would need to be removed, as would the use of fore and hind limbs
157
and snout, and appropriate changes made, in order to use the new scoring scale
with hanging human bodies. Such differences are obvious and should not cause
difficulties.
The new decomposition scale for hanging bodies was found to be sufficient and
appropriate for scoring both fully suspended bodies and the upper, suspended,
part of partially suspended bodies, as well as both naked and clothed bodies. The
decomposition scale produced by Megyesi et al. (2005) was found to be similarly
sufficient and appropriate for scoring clothed and unclothed bodies lying fully on
the ground, and also the lower, unsuspended, portions of partially suspended
bodies. Semi-recumbent bodies could, therefore, be scored using either scale but
the hanging scale is expected to give better results due to the ability to include the
head in the decomposition assessment.
The presence of loose, lightweight clothing, which did not impede insect access,
was not found to significantly affect the rate of decomposition in the bodies lying
on the ground. For the hanging pigs, however, the presence of clothing was found
to affect both the pattern and rate of decomposition, with the clothed bodies
decomposing faster than the unclothed bodies. Studies into insect access to bodies
which have been wrapped, rather than clothed, have reported delays in blowfly
arrival from 1 to 13 days (Ahmad et al., 2011; Goff, 1992). As some autoerotic
practices include wrapping of the body, although the face may be unobstructed,
this delay should be borne in mind.
Whilst the study was robust and carried out over a number of years, it was
conducted with the intention of increasing the accuracy of PMI estimation for
158
human bodies, rather than pigs, for use in aiding victim identification. Clearly it
would be preferable to use humans for such a study, but this would require a large
number of humans for whom the time and cause of death were known. Preferably
the bodies should be of the same size, age and weight, and should be studied at the
same time and throughout the whole decomposition process from fresh through to
skeletonisation. This would be a difficult if not impossible set of criteria to fulfil. A
further complication is that some countries, such as the UK, do not allow these
sorts of experiments to be carried out on humans. It is therefore considerably
easier, and often necessary, for human analogues, such as the domestic pig (Sus
scrofa) to be used (Forbes et al., 2005; Myburgh et al., 2013; Paczkowski et al.,
2015; Roberts and Dabbs, 2015; Schotsmans et al., 2012). Pigs have the advantages
of being easily obtainable and similar to humans in their decomposition process
(Catts and Goff, 1992). It should usually be possible to use animals of the same size,
weight, and age, with known times and methods of death, and which are unlikely
to be suffering from any illnesses. Their diet, environment, and possibly even
parents will have been the same, thus decreasing the variables that might affect the
decomposition rate or pattern.
However, the fact remains that pigs are not humans and the degree to which the
decomposition rates observed in this study, expressed as TBS against ADD and
converted into the ADD prediction tables, apply to humans is currently unknown.
The first step towards validating the ADD prediction tables for hanging bodies
would be to collect decomposition TBShang scores from hanged bodies where the
time of death is known or fairly confidently known, using the hanging body
decomposition scoring scale presented here, and to compare the associated ADD
with the corresponding ADD predictions for the observed TBShang. An assessment
159
could then be made as to whether any differences in the actual and predicted ADD
values fall within an acceptable confidence interval that would allow the tables to
be used, or whether there is a consistent difference which would allow the ADD
prediction tables to be used in conjunction with some form of correction factor.
This would not, however, replace the need for eventual human testing – consistent
results obtained from testing chloroform on dogs would never have given the same
results as when chloroform was tested on humans.
An unintentional shortcoming in the experimental design led to a wide range of
weights in the pigs used. This did, however, provide an unplanned opportunity to
investigate the effect of start weight on the decomposition rate and showed that a
large increase in weight, of about 10 kg, was required to produce a change in value
of 1 in TBS. Within the scope of this study this meant that the pig start weight had
an effect that would make only a negligible difference to PMI estimates.
The shortcoming in experimental design which led to the wide range of pig start
weights could have been avoided if the pigs had been sourced from the same farm
at the same age each year. More pragmatically, if it had been known prior to
allocation into groups and weighing, i.e. if it had been noticed on the farms that the
pigs differed by such a great amount, the allocation into groups could have been
done by weight, ensuring this study had animals of similar weight over the three
years. To minimise the number of pigs slaughtered and maximise their use,
taphonomic experiments at the study site are coordinated to ensure the control
pigs (on the ground) are shared by multiple researchers and experiments. This
meant that, while the start of experiments was usually at the end of May, the
160
sourcing of pigs and exact start dates were determined by the site manager and
could, therefore, not be controlled across the three years of this study.
The opportunity to explore the effect of start weight on decomposition has also
presented another set of, as yet unanswered, questions and suggested areas for
future work: For reasons given earlier the use of humans for decomposition
studies is not always possible, and the domestic pig is considered the best available
substitute for decomposition studies because of its similarities to humans,
including skin, diet, digestive fauna, and muscle-to-fat ratio (Anderson, 2010;
Anderson and Van Laerhoven, 1996; Catts and Goff, 1992). Changes in the human
diet, which is not as well controlled as that of a farm-bred pig, have resulted in
humans becoming increasingly fatter, with the rate of increase in obesity levels
showing little evidence of slowing down (WHO, 2015). This presents a potential
problem with the use of pigs in future studies because of the difference in the way
in which humans and pigs gain weight within the normal to morbidly obese range
for humans. As humans gain excess weight the ratio of fat to muscle changes with
the percentage of fat increasing. The pigs within the normal human adult weight
range are juveniles of 16 to 20 weeks (20 week old pigs used in this study could
weigh as much as 69 kg). As pigs continue to gain weight to their adult size this
weight is gained mostly as muscle. While an adult pig may weigh the same as a
morbidly obese adult human, their muscle-to-fat ratios will be very different. The
likelihood of having human bodies in the 90 kg to 140 kg range is increasing. This
raises the following questions: Does a morbidly obese adult human decompose
differently from a normal weight adult and, if so, does this mean that the pig is no
longer a good substitute at these weights? Will the rate and pattern of
decomposition differ between the trunks of morbidly obese male and female
161
humans, as females tend to lay down fat between the skin and torso muscles whilst
males tend to accumulate the fat beneath the muscle layer and around the organs?
Do insects differentiate between the fat and muscle? If the muscle-to-fat ratio does
make a difference, then an additional complication is that scientists are now using
clustered regularly interspaced short palindromic repeats (CRISPR) to breed pigs
with higher muscle mass (Hall, 2016).
The use of ADD as the explanatory variable for the progress of decomposition is
based on the accepted, and seemingly logical, premise that, if a given amount of
thermal energy is put into a carcass the same amount of reaction should occur,
such that the same body score for decomposition will be seen, regardless of the
timescale over which this takes place. On this basis the same degree of
decomposition is expected after, say, 200 ADD whether it was as the result of 20
days at 10 °C, or 10 days a 20 °C, or 100 days at 2 °C. The inter-year consistency
test conducted as part of this study appears to support this approach, but these
experiments were all conducted at a location and time of year when the average
daily temperature was always above freezing.
The two most important influences on the rate of decomposition are temperature
and insect activity (Cross and Simmons, 2010; Mann et al., 1990; Simmons et al.,
2010a) and the above use of ADD assumes that insect activity is unaffected within
the temperature range used. Forensic entomologists use insects to determine the
minimum time since death based on known species succession to the corpse, the
temperatures required for the insect eggs to hatch, and the known length of time
for larval development through the instar stages – all of which differ from insect to
insect – to calculate minimum ADH (accumulated degree hours). In calculating the
162
ADH for any insect the accumulation of degree hours does not start until the
critical hatching temperature for that insect has been reached.
Forensic anthropologists, however, calculate ADD values by accumulating the
degree days for any day, or part day, when the temperature is above freezing
(Megyesi et al., 2005). But if insects do not hatch before, for example, 4 °C (and it is
not clear if this is for all the eggs, any eggs at all, or a critical mass of eggs) then the
insect activity may still not have occurred by a given ADD value if the temperature
did not rise above the critical threshold on any of those days.
Conversely, if there are established maggot masses present which are generating
their own heat, and the temperature drops below 0 °C (halting the accumulation
for ADD), the decomposition may continue unabated or at a slower rate, but the
ADD value would be lower than might be expected from the observed TBS.
Huntington et al. (2007) reported that maggots continued to feed and grow whilst
stored in a morgue at a temperature of 4 °C. Marchenko (2001) provides a table
giving the lower development temperature of Protophormia terraenova as 7.8 °C
(this is usually the most abundant species found at the study site) and Calliphora
vicina as 2 °C. If maggots had hatched at an expected temperature of, say, 4 °C,
what would be the effect on them of several days at below this temperature if they
had not yet established masses? The experiments in this study did not encounter
these issues, being conducted during the summer in a temperate climate where the
temperatures were neither high nor very low. Elsewhere in the world, or at
different times in the year, the situation would be different and this may impact
the use of ADD. Studies, reported this year (Bates and Wescott, 2016; Dautartas et
al., 2016; Simmons et al., 2016), are looking at whether the accepted use of ADD as
163
a means of levelling out seasonality and geographical location is as straightforward
as has previously been accepted.
It is recognised that the cause of death and the environment in which a body
decomposes have an effect on the pattern of decomposition. To address this
decomposition scoring tables, which better reflect these differences, have been
produced for specific situations including charred bodies (Gruenthal et al., 2012),
submerged bodies (Heaton et al., 2010), and now hanging bodies. Work has also
been carried out to determine the differences for bodies in mass graves, where the
situation is further complicated by the surrounding soil types and conditions. It
would seem that if the reliance on ADD having compensated for seasonal and
geographical differences is not as straightforward as has been accepted, then
specific decomposition scoring tables may be needed not only for different causes
of death and physical decomposition sites, but also to reflect the seasons. There
may be significant weather changes between time of death and discovery.
Temperatures were collected during the experiments to enable ADD to be
accurately calculated. It was not intended, as part of the experimental design, to
record rainfall as part of the general data collection. However, at the end of
Experiment 2, as the weather had been extremely wet throughout the whole
experiment, attempts were made to collect rainfall figures from the on-site
weather station. This had not been used for collecting temperature data since
temperature data loggers were attached to the hanging frames. Unfortunately, the
on-site weather station was found to have failed and no data was available. After
completion of this study rainfall data was requested, and received, from the
Meteorological Office. The nearest available weather station was Bingley 2
164
(53.81° N, 1.87° W, 262 m). Using this data the average rainfall, in millimetres, for
the months of June and July were calculated for the three years (Table 8.1).
Table 8.1 Average monthly rainfall data (mm) recorded at the nearest available
Meteorological Office weather station (Bingley 2, 53.81° N, 1.87° W, 262 m) during
the three years of the study.
Average Monthly
2011 2012 2013
Rainfall (mm)
June 60.2 141.4 24.0
July 52.2 145.8 97.8
It can be seen that the rainfall figures for 2012 were significantly higher than the
other two years. It must, however, be borne in mind that this data is from the
nearest weather station to the experimental site, but is still some 15.9 miles away
and at an altitude of 262 m above average sea level compared to 169 m for
TRACES. It is unlikely that the rainfall would be exactly the same and, as confirmed
in conversation with the Met Office, the differences in rainfall from fell side to fell
side can be significantly different over small distances. This data does show,
though, that for the general area the rainfall over these three years was not
consistent.
Lastly, these experiments took place outdoors and most hangings occur indoors,
which leads to the question: Would the results have been the same if they had been
carried out indoors, where insect access to the bodies may have been impeded?
Indoor decomposition rates tend to be slower than those outdoors and this is
attributed to delayed insect access (Bhadra et al., 2014; Catts and Goff, 1992). Both
Charabidze et al. (2015) and Reibe and Madea (2010) urge caution in quantifying
165
the pre-appearance time of insects found indoors. Decomposition experiments
with hanging bodies should therefore be repeated indoors, where the results may
be very different from outdoors where the temperatures vary and the insects have
free access. Houses may maintain a relatively high temperature, 18 °C is
considered comfortable, suitable for insect egg hatching but if the temperatures
outside are sub-zero for extended periods would this impact on the insect arrival
times? One of the features of hanging decomposition found in all the hanging pigs
in this study was mummification, due to the result of wind drying, and it has been
added to the decomposition scoring scale. Would the rate of mummification be
sped up or slowed down indoors by the presence of central heating and air
conditioning, or would it not occur at all, requiring amendments to the hanging
decomposition scoring tables?
8.1 Areas for Further Work
During the course of this research a number of questions arose, which could not be
addressed within the scope of these experiments, but which highlight areas in
which further research could be conducted.
8.1.1 Validation of PMI Prediction Tables for Use With Humans
Because of restrictions in the UK on the use of human cadavers, as is the case in
many other countries, all this work was carried out using the domestic pig Sus
scrofa which is widely considered to be the most suitable human analogue for
decomposition studies. These experiments were carried out, however, to increase
the accuracy with which post-mortem interval estimations for hanging human
bodies can be calculated for use in forensic case work. The PMI prediction tables
166
produced as part of this study for hanging bodies and bodies on the ground need to
be validated for use with human cadavers.
Ideally longitudinal studies of hanging human bodies would provide the
information required to determine if the PMI values given in the tables are within
an acceptable range of accuracy to be used as they are; or if they could be used
with a correction factor; or if they require recalculating from human observational
data. Initial studies could be carried out using decomposition scoring data taken
from bodies found in forensic cases and for whom the time of death is known with
some accuracy.
8.1.2 Use of Pigs as Human Analogues for Decomposition Studies
Pigs are used as human analogues in decomposition studies for many reasons
including legality, the numbers of pigs that can be used, cost, acceptability, and the
ability to control for weight and size, and to decrease the number of variables
within an experiment. One of the main reasons the pig is considered such a good
human analogue is the ratio of fat to muscle, which is similar to that of a human at
the same weight. However, as obesity becomes more prevalent in humans, the
similarity in muscle-to-fat ratio changes and becomes very different; bringing into
question whether the pig will continue to be a suitable analogue.
Further research is needed to establish whether this change in muscle-to-fat ratio
has a significant effect on the decomposition pattern and rate for humans in the
obese and morbidly obese ranges, and whether it impacts the decomposition
scoring scales and PMI prediction tables used to estimate time since death.
167
Note should be taken to observe for any differences between males and females, as
the pattern of fat distribution is different between the sexes and may have an effect
on the pattern of decomposition and insect activity.
8.1.3 Use of Accumulated Degree Days (ADD)
ADD has been used throughout these experiments and is used by forensic
anthropologists in calculating PMI estimates because it incorporates adjustments
for seasonal, inter-seasonal and annual temperature variations. However, recent
work indicates that it may not be as straightforward as has been previously
accepted.
ADD accumulates temperature only when it is above 0 °C, and assumes that insect
activity is unaffected within the temperature ranges used. Work is needed to look
at:
• The effects of long periods of low temperature, above 0 °C but below the
laying and hatching temperature thresholds of flies associated with
decomposition.
• The effects of periods of low temperature, 0 °C and below, on the insect
activity once maggot masses have been established, when activity may be
continuing but ADD scores are not accumulating.
• The effects of periods of high temperature at the start of decomposition and
when maggot masses are established. Maggot masses may not be able to
establish or maintain viability if temperatures are too high and the mass
168
temperature cannot be controlled through the maggots’ normal
thermoregulatory behaviour.
• Whether the speed with which temperatures rise or fall has an effect on
insect activity and the ability of the maggot mass thermoregulatory
behaviour to compensate for the temperature changes.
This work could be of use in refining the accuracy of ADD calculations which are
currently critical to the calculation of PMI estimates.
8.1.4 Rainfall and Hanging Bodies
A feature of decomposition in the hanging bodies in this experiment has been the
desiccation, caused by the wind, and mummification of the naked hanging bodies.
In the experiment with clothed hanging pigs this mummification did not take place
in the same way. Throughout the experiment rainfall was high and the clothing
worn by the pigs never dried out. Loose clothing, in contact with the skin, would
normally be expected to have a wicking action and, where there was airflow or
wind, this should accelerate the water loss from the epidermis, desiccating the
body faster than would be the case for a naked body. Wet clothing in contact with
the skin has the effect of slowing the desiccation and, in the case of constantly wet
clothing, of stopping it.
These experiments with hanging clothed pigs need to be repeated in conditions
where the clothing is not constantly wet. Different results may also be obtained if
the hanging bodies are clothed in heavier or tighter clothing.
169
8.1.5 Indoor Versus Outdoor Decomposition
Mummification was a common feature of the hanging bodies studied in these
experiments, where the bodies were exposed to the drying effects of wind. If the
degree of mummification of a body hanging indoors is more advanced relative to
the rate at which insects are able to gain access to the body, this may impede the
decomposition, relative to a body hanging outdoors, since desiccated skin is an
inhospitable environment for oviposition. Conversely, if mummification is not a
common feature of bodies hanging indoors, then decomposition may advance
more quickly indoors than outdoors. Decomposition experiments with hanging
bodies should be repeated indoors to investigate:
• Whether mummification occurs at all or at a different rate, and the effects
that central heating and air conditioning have on the rate of mummification.
• The effects of differential air temperatures between the indoor and outdoor
environments on insect arrival and activity.
• How the interplay between changes to the rate of mummification and any
delays in insect access affect the overall rate of decomposition.
8.2 Conclusion
This study provided a robust and quantitative investigation into the decomposition
pattern and rate in hanging bodies with the intention that the results would be of
use in increasing the accuracy of PMI estimation in hanged humans. The study,
which used Sus scrofa as human analogues, showed very clear differences in the
170
rate and pattern of decomposition between hanging pigs and controls on the
ground. It is expected that equally clear differences will be found between humans
who have been hanged and others decomposing on the ground. The ADD
prediction tables produced provide a quick and simple way of predicting the ADD
from observed TBS and further work should be carried out to use human results to
test and validate them. Whilst this study has answered the questions it set out to
address, it has raised a number of additional questions for further investigation.
171
9 References
Adlam, R.E. and Simmons, T. (2007). The Effect of Repeated Physical Disturbance
on Soft Tissue Decomposition – Are Taphonomic Studies an Accurate Reflection of
Decomposition? Journal of Forensic Sciences, 52(5):1007-1014.
Ahmad, A., Ahmad, A.H., Dieng, H., Satho, T., Ahmad, H., Aziz, A.T., Boots, M. (2011).
Cadaver Wrapping and Arrival Performance of Adult Flies in an Oil Palm Plantation
in Northern Peninsular Malaysia. Journal of Medical Entomology, 48(6):1236–1246.
Ambade, V.N., Kolpe, D., Tumram, N., Meshram, S., Pawar, M., Kukde, H. (2015)
Characteristic Features of Hanging: A Study in Rural District of Central India.
Journal of Forensic Sciences, 60(5):1216-1223.
Amendt, J., Campobasso, C.P., Gaudry, E., Reiter, C., LeBlanc, H.N., Hall, M.J. (2007).
Best Practice in Forensic Entomology - Standards and Guidelines. International
Journal of Legal Medicine, 121(2):90-104.
Anderson, G.S. (2010). Factors that Influence Insect Succession on Carrion. In:
Byrd, J.H. and Castner, J.L., eds. Forensic Entomology: The Utility of Arthropods in
Legal Investigations. 2nd ed. Boca Raton, Florida: CRC Press. Ch.5.
Anderson, G.S. (2011). Comparison of Decomposition Rates and Faunal

Colonization of Carrion in Indoor and Outdoor Environments. Journal of Forensic
Sciences, 56(1):136-142.
Anderson, G.S. and VanLaerhoven, S.L. (1996). Initial Studies on Insect Succession
on Carrion in Southwestern British Columbia. Journal of Forensic Sciences,
41(4):617-625.
Archer, M.S. (2004). Rainfall and Temperature Effects on the Decomposition Rate
of Exposed Neonatal Remains. Science & Justice, 44(1):35-41.
172
Aturaliya, S. and Lukasewycz, A. (1999). Experimental Forensic and
Bioanthropological Aspects of Soft Tissue Taphonomy: 1. Factors Influencing
Postmortem Tissue Desiccation Rate. Journal of Forensic Sciences, 44(5):893-896.
Aufderheide, A.C. (2003). The Scientific Study of Mummies. New York: Cambridge
University Press.
Aufderheide, A.C. (2011). Soft Tissue Taphonomy: A Paleopathology Perspective.

International Journal of Paleopathology, 1(2):75-80.
Austin, A.E., van den Heuvel, C., Byard, R.W. (2011). Cluster Hanging Suicides in the
Young in South Australia. Journal of Forensic Sciences, 56(6):1528-1530.
Austin, A.E., van den Heuvel, C., Byard, R.W. (2013). Physician Suicide. Journal of
Forensic Sciences, 58(S1):S91-S93.
Baker, S.P., Hu, G., Wilcox, H.C., Baker, T.D. (2013). Increase in Suicide by
Hanging/Suffocation in the U.S., 2000–2010. American Journal of Preventive
Medicine, 44(2):146-149.
Bates, D., Maechler, M., Bolker, B., Walker, S. (2015). Fitting Linear Mixed-Effects
Models Using lme4. Journal of Statistical Software, 67(1):1-48
Bates, L.N. and Wescott, D.J. (2016). Not all Degree Days are Equal in the Rate of
Decomposition: The Effect of Season of Death on the Relationship Between Gross
Postmortem Decomposition and Accumulated Degree Days. In: Proceedings of the
American Academy of Forensic Sciences 68th Annual Scientific Meeting, Las Vegas,
Nevada. Colorado Springs, Colorado: AAFS. A117, p.178.
Bennewith, O., Gunnell, D., Kapur, N., Turnbull, P., Simkin, S., Sutton, L., Hawton, K.
(2005). Suicide by Hanging: Multicentre Study Based on Coroners’ Records in
England. The British Journal of Psychiatry, (186):260-261.
173
Bhadra, P., Hart, A.J. & Hall, M.J.R. (2014). Factors Affecting Accessibility to
Blowflies of Bodies Disposed in Suitcases. Forensic Science International, 239:62–
72.
Biddle, L., Gunnell, D., Owen-Smith, A., Potokar, J., Longson, D., Hawton, K., Kapur,
N., Donovan J. (2012). Information Sources Used by the Suicidal to Inform Choice
of Method. Journal of Affective Disorders, 136(3):702-709.
Bowen, D.A.L.L. (1982). Hanging – A Review. Forensic Science International,

20(3):247-249.
Byard, R.W. and Winskog, C. (2012). Autoerotic Death: Incidence and Age of
Victims – A Population-Based Study. Journal of Forensic Sciences, 57(1):129-131.
Campobasso, C.P., Di Vella, G., Introna, F. (2001). Factors Affecting Decomposition

and Diptera Colonization. Forensic Science International, 120(1-2):18-27.
Card, A., Cross, P., Moffatt, C., Simmons, T. (2015). The Effect of Clothing on the
Rate of Decomposition and Diptera Colonization on Sus scrofa Carcasses, Journal of
Forensic Sciences, 60(4):979-982.
Catts, E.P. (1992). Problems in Estimating the Postmortem Interval in Death

Investigations. Journal of Agricultural Entomology, 9(4):245-255.
Catts, E.P. and Goff, M.L. (1992). Forensic Entomology in Criminal Investigations.
Annual Review of Entomology, 37:253-272.
Charabidze, D., Bourel, B., Gosset, D. (2011). Larval-Mass Effect: Characterisation of

Heat Emission by Necrophagous Blowflies (Diptera: Calliphoridae) Larval
Aggregates. Forensic Science International, 211(1-3):61-66.
Charabidze, D., Hedouin, V., Gosset, D. (2015). An Experimental Investigation into

the Colonization of Concealed Cadavers by Necrophagous Blowflies. Journal of
Insect Science, 15(1):149.
174
Chinery, M. (1976). A Field Guide to the Insects of Britain and Northern Europe. 2nd
ed. London: Collins.
Chong Chin, H., Sulaiman, S., Othman, H., Jeffrey, J., Kurahashi, H., Omar, B. (2010).
Insect Succession Associated with a Hanging Pig Carcass Placed in an Oil Palm
Plantation in Malaysia. Sains Malaysiana, 39(6):921-926.
Clark, M. A. and Kerr, F. C. (1986). Unusual Hanging Deaths. Journal of Forensic

Sciences, 31(2):747-755.
Cross, P. and Simmons, T. (2010). The Influence of Penetrative Trauma on the Rate
of Decomposition. Journal of Forensic Sciences, 55(2):295-301.
Cross, P., Simmons, T., Cunliffe, R., Chatfield, L. (2009). Establishing a Taphonomic
Research Facility in the United Kingdom. Forensic Science Policy & Management: An
International Journal, 1(4):187-191.
CPSC. (no date). Are Your Window Coverings Safe? [Online] Available at:
http://www.cpsc.gov/PageFiles/121510/5009a.pdf [Accessed 26 March 2016].
Cunha, E. and Pinheiro, J. (2007). Forensic Anthropology in Portugal: From Current

Practice to Future Challenges. In: Brickley, M.B. and Ferllini, R., eds. Forensic
Anthropology: Case Studies from Europe. Springfield, Illinois: Charles C Thomas. Ch.
3.
Dabbs, G.R. (2010). Caution! All Data are Not Created Equal: The Hazards of Using
National Weather Service Data for Calculating Accumulated Degree Days. Forensic
Science International, 202(1–3):e49-e52.
Dabbs, G.R. (2015). How Should Forensic Anthropologists Correct National

Weather Service Temperature Data for Use in Estimating the Postmortem Interval?
175
Dautartas, A.M. (2009). The Effect of Various Coverings on the Rate of Human
Decomposition. Master’s Thesis, University of Tennessee. [Online] Available at:
http://trace.tennessee.edu/utk_gradthes/69 [Accessed 19 March 2016].
Dautartas, A.M., Steadman, D.W., Mundorff, A.Z., Jantz, L.M., Vidoli, G.M. (2016). A
Comparison of Seasonal Decomposition Patterns Between Human and Non-Human
Animal Models. In: Proceedings of the American Academy of Forensic Sciences 68th
Annual Scientific Meeting, Las Vegas, Nevada. Colorado Springs, Colorado: AAFS.
A118, p.180.
De Jong, G.D., Hoback, W.W., Higley, L.G. (2011). Effect of Investigator Disturbance
in Experimental Forensic Entomology: Carcass Biomass Loss and Temperature.
Journal of Forensic Sciences, 56(1):143–149.
Dillon, L. and Anderson, G.S. (1995). Forensic Entomology: The Use of Insects in
Death Investigations to Determine Elapsed Time Since Death. Technical Report –
Canadian Police Research Centre, TR-05-95. [Online] Available at:
http://publications.gc.ca/collections/collection_2008/ps-sp/PS63-2-1995-5E.pdf
[Accessed 19 March 2016].
Dillon, L. and Anderson, G.S. (1996). Forensic Entomology: The Use of Insects in
Death Investigations to Determine Elapsed Time Since Death in Interior and
Northern British Columbia Regions. Technical Report – Canadian Police Research
Centre, TR-03-96. [Online] Available at:
http://publications.gc.ca/collections/collection_2008/ps-sp/PS63-2-1996-3E.pdf
Dogan, K.H., Demirci, S., Deniz, I. (2015). Why Do People Hang Themselves on
Trees? An Evaluation of Suicidal Hangings on Trees in Konya, Turkey, between
2001 and 2008. Journal of Forensic Sciences, 60(S1):S87-S92.
Ferris, J.A.J. (2000). Causes of Death: Asphyctic Deaths. In: Siegel, J.A., Saukko, P.J.,
Knupfer, G.C., eds. Encyclopedia of Forensic Sciences. London: Academic Press, 308-
316.
176
Forbes, S.L., Stuart, B.H., Dent, B.B. (2005). The Effect of the Burial Environment on
Adipocere Formation. Forensic Science International, 154(1):24–34.
Gallagher, M.B., Sandhu, S., Kimsey, R. (2010). Variation in Developmental Time for
Geographically Distinct Populations of the Common Green Bottle Fly, Lucilia
sericata (Meigen). Journal of Forensic Sciences, 55(2):438-442.
Galloway, A., Birkby, W.H., Jones, A.M., Henry, T.E., Parks, B.O. (1989). Decay Rates
of Human Remains in an Arid Environment. Journal of Forensic Sciences, 34(3):607-
616.
Gellman, A. and Hill, J. (2006). Data Analysis Using Regression and

Multilevel/Hierarchical Models. Cambridge: Cambridge University Press.
Gellman, A. and Su, Y-S. (2015). arm: Data Analysis Using Regression and
Multilevel/Hierarchical Models. Vienna, Austria: The R Foundation for Statistical
Computing. [Online] Available at: http://cran.r-
project.org/web/packages/arm/index.html [Accessed 20 March 2016].
Gill-King, H. (1997). Chemical and Ultrastructural Aspects of Decomposition. In:

Haglund, W.D. and Sorg, M.H., eds. Forensic Taphonomy: The Postmortem Fate of
Human Remains. Boca Raton, Florida: CRC Press. Ch. 6.
Goff, M.L. (1992). Problems in Estimation of Postmortem Interval Resulting from

Wrapping of the Corpse: A Case Study from Hawaii. Journal of Agricultural
Entomology, 9(4):237-243.
Goff, M.L. and Lord, W.D. (1994). Hanging Out at the Sixteenth hole: Problems in
Estimation of Postmortem Interval Using Entomological Techniques in Cases of
Death by Hanging. 46th American Academy of Forensic Sciences Annual Meeting, San
Antonio, Texas.
177
Grassberger, M. and Frank, C. (2004). Initial Study of Arthropod Succession on Pig
Carrion in a Central European Urban Habitat. Journal of Medical Entomology,
41(3):511-523.
Gruenthal, A., Moffatt, C., Simmons, T. (2012). Differential Decomposition Patterns

in Charred Versus Un-Charred Remains. Journal of Forensic Sciences, 57(1):12-18.
Gunnell, D., Bennewith, O., Hawton, K., Simkin, S., Kapur, N. (2005). The
Epidemiology and Prevention of Suicide by Hanging: a Systematic Review.
International Journal of Epidemiology, 34(2):433–442.
Haglund, W.D. (1997). Dogs and Coyotes: Postmortem Involvement with Human
Remains. In: Haglund, W.D. and Sorg, M.H., eds. Forensic Taphonomy: The
Postmortem Fate of Human Remains. Boca Raton, Florida: CRC Press. Ch. 23.
Hall, S.S. (2016). Editing the Mushroom. Scientific American, 314(3):49-55.
Haskell, N.H., Hall, R.D., Cervenka, V.J., Clark, M.A. (1997). On the Body: Insects' Life
Stage Presence, Their Postmortem Artifacts. In: Haglund, W.D. and Sorg, M.H., eds.
Forensic Taphonomy: The Postmortem Fate of Human Remains. Boca Raton, Florida:
CRC Press. Ch. 27.
Heaton, V., Lagden, A., Moffatt, C., Simmons, T. (2010). Predicting the Postmortem
Submersion Interval for Human Remains Recovered from U.K. Waterways. Journal
of Forensic Sciences, 55(2):302-307.
Heaton, V., Moffatt, C., Simmons, T. (2014). Quantifying the Temperature of Maggot
Masses and its Relationship to Decomposition. Journal of Forensic Sciences,
59(3):676-682.
Hejna, P. and Bohnert, M. (2013). Decapitation in Suicidal Hanging – Vital Reaction

Patterns. Journal of Forensic Sciences, 58(S1):S270-S277.
Human Tissue Act 2004. (c.30), London: HMSO.
178
Huntington, T.E., Higley, L.G., Baxendale, F.P. (2007). Maggot Development During
Morgue Storage and its Effect on Estimating the Post-Mortem Interval. Journal of
Kelly, J.A. (2006). The Influence of Clothing, Wrapping and Physical Trauma on
Carcass Decomposition and Arthropod Succession in Central South Africa. Ph.D
Thesis, University of the Free State. [Online] Available at:
http://scholar.ufs.ac.za:8080/xmlui/bitstream/handle/11660/1069/KellyJA.pdf?s
equence=1 [Accessed 20 March 2016].
Kelly, J.A., van der Linde, T.C., Anderson, G.S. (2009). The Influence of Clothing and
Wrapping on Carcass Decomposition and Arthropod Succession During the
Warmer Seasons in Central South Africa. Journal of Forensic Sciences, 54(5):1105-
1112.
Komar, D.A. (1998). Decay Rates in a Cold Climate Region: A Review of Cases
Involving Advanced Decomposition from the Medical Examiner’s Office in
Edmonton, Alberta. Journal of Forensic Sciences, 43(1):57-61.
Komar, D., Beattie, O., Dowling, G., Bannach, B. (1999). Hangings in Alberta, with
Special Reference to Outdoor Hangings with Decomposition. Canadian Society of
Forensic Science Journal, 32(2-3):85-96.
Lapworth, A. and McGregor, J. (2008). Seasonal Variation of the Prevailing Wind

Direction in Britain. Weather, 63(12):365-368.
Mann, R.W., Bass, W.M., Meadows, L. (1990). Time Since Death and Decomposition
of the Human Body: Variables and Observations in Case and Experimental Field
Studies. Journal of Forensic Sciences, 35(1):103-111.
Marchenko, M.I. (2001). Medicolegal Relevance of Cadaver Entomofauna for the

Determination of Time of Death. Forensic Science International, 120(1-2):89-109.
179
Mégnin, P. (1894). La Faune des Cadavres : Application de l’Entomologie à la
Médecine Légale. Paris: Gauthier-Villars et Fils.
Megyesi, M.S., Nawrocki, S.P., Haskell, N.H. (2005). Using Accumulated Degree-Days
to Estimate the Postmortem Interval from Decomposed Human Remains. Journal of
Mendes, R., Santos, S., Taveira, F., Dinis-Oliveira, R.J., Santos, A., Magalhães, T.
(2015). Child Suicide in the North of Portugal. Journal of Forensic Sciences,
60(2):471-475.
Met Office. (2010). Burnley Climate. [Online] Available at:

http://www.metoffice.gov.uk/public/weather/climate/gcw6h8p0c [Accessed 23
March 2016].
Michaud, J.P. and Moreau, G. (2011). A Statistical Approach Based on Accumulated

Degree-Days to Predict Decomposition-Related Processes in Forensic Studies.
Journal of Forensic Sciences, 56(1):229–232.
Micozzi, M.S. (1991). Postmortem Change in Human and Animal Remains.

Springfield, Illinois: Charles C Thomas.
Miller, R.A. (2002). The Affects [sic] of Clothing on Human Decomposition:

Implications for Estimating Time Since Death. Master’s Thesis, University of
Tennessee. [Online] Available at: http://trace.tennessee.edu/utk_gradthes/856
Moffatt, C. and Simmons, T. (2014). Forensic Entomology Research at the

University of Central Lancashire’s TRACES Decomposition Facility. Antenna, ECE
Special Edition: 114-115.
Moffatt, C., Simmons, T., Lynch-Aird, J. (2016). An Improved Equation for TBS and
ADD: Establishing a Reliable Postmortem Interval Framework for Casework and
Experimental Studies. Journal of Forensic Sciences, 61(S1):S201-S207.
180
Myburgh, J., L'Abbé, E.N., Steyn, M., Becker, P.J. (2013). Estimating the Postmortem
Interval (PMI) Using Accumulated Degree-Days (ADD) in a Temperate Region of
South Africa. Forensic Science International, 229(1-3):165.e1-165.e6.
Oliveira, T.C. and Vasconcelos, S.D. (2010). Insects (Diptera) Associated with
Cadavers at the Institute of Legal Medicine in Pernambuco, Brazil: Implications for
Forensic Entomology. Forensic Science International, 198(1-3):97-102.
Paczkowski, S., Nicke, S., Ziegenhagen, H., Schütz, S. (2015). Volatile Emission of
Decomposing Pig Carcasses (Sus scrofa domesticus L.) as an Indicator for the
Postmortem Interval. Journal of Forensic Sciences, 60(S1):S130-S137.
Pakis, I., Yayci, N., Karapirli, M., Yildiz, N., Gunce, E., Yilmaz, R., Polat, O. (2010).
Childhood Deaths Due to Suicide. Australian Journal of Forensic Sciences,
42(3):191-197.
Parks, C.L. (2011). A Study of the Human Decomposition Sequence in Central

Texas. Journal of Forensic Sciences, 56(1):19–22.
Payne, J.A. (1965). A Summer Carrion Study of the Baby Pig Sus Scrofa Linnaeus.
Ecology, 46(5):592-602.
Payne, J.A., King, E.W., Beinhart, G. (1968). Arthropod Succession and

Decomposition of Buried Pigs. Nature, 219:1180-1181.
R Development Core Team (2010). R: A Language and Environment for Statistical

Computing. Vienna, Austria: The R Foundation for Statistical Computing. ISBN: 3-
900051-07-0. [Online] Available at: http://www.r-project.org [Accessed 20 March
2016].
Reibe, S. and Madea, B. (2010). How Promptly do Blowflies Colonise Fresh

Carcasses? A Study Comparing Indoor with Outdoor Locations. Forensic Science
International, 195:52–57.
181
Roberts, L.G. and Dabbs, G.R. (2015). A Taphonomic Study Exploring the
Differences in Decomposition Rate and Manner between Frozen and Never Frozen
Domestic Pigs (Sus scrofa). Journal of Forensic Sciences, 60(3):588-594.
Rodriguez, W.C. (1997). Decomposition of Buried and Submerged Bodies. In:

Haglund, W.D. and Sorg, M.H., eds. Forensic Taphonomy: The Postmortem Fate of
Human Remains. Boca Raton, Florida: CRC Press. Ch. 29.
Rodriguez, W.C. and Bass, W.M. (1983). Insect Activity and its Relationship to
Decay Rates of Human Cadavers in East Tennessee. Journal of Forensic Sciences,
28(2):423-432.
RoSPA. (2015). Blind Cord Safety Campaign. [Online] Available at:

http://www.rospa.com/campaigns-fundraising/current/blind-cord/ [Accessed 26
March 2016].
Sauvageau, A. and Racette, S. (2006). Autoerotic Deaths in the Literature from

1954 to 2004: A Review. Journal of Forensic Sciences, 51(1):140-146.
Schoenly, K.G. and Hall, R.D. (2002). Testing Reliability of Animal Models in Research
and Training Programs in Forensic Entomology, Part II., Final Report. National
Criminal Justice Reference Service, [Online] Available at:
https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=192281 [Accessed 20
March 2016].
Schoenly, K.G., Haskell, N.H., Hall, R.D., Gbur, J.R. (2007) Comparative Performance
and Complementarity of Four Sampling Methods and Arthropod Preference Tests
from Human and Porcine Remains at the Forensic Anthropology Center in
Knoxville, Tennessee. Journal of Medical Entomology, 44(5):861-894.
Schoenly, K.G., Haskell, N.H., Mills, D.K., Bieme-Ndi, C., Larsen, K., Lee, Y. (2006).
Recreating Death’s Acre in the School Yard: Using Pig Carcasses as Model Corpses,
182
to Teach Concepts of Forensic Entomology & Ecological Succession. The American
Biology Teacher, 68(7):402-410.
Schotsmans, E.M.J., Denton, J., Dekeirsschieter, J., Ivaneanu, T., Leentjes, S.,
Janaway, R.C., Wilson, A.S. (2012). Effects of Hydrated Lime and Quicklime on the
Decay of Buried Human Remains Using Pig Cadavers as Human Body Analogues.
Forensic Science International, 217(1-3):50–59.
Schotsmans, E.M.J., Denton, J., Fletcher, J.N., Janaway, R.C., Wilson, A.S. (2014a).
Short-Term Effects of Hydrated Lime and Quicklime on the Decay of Human
Remains Using Pig Cadavers as Human Body Analogues: Laboratory Experiments.
Forensic Science International, 238:142.e1–142.e10.
Schotsmans, E.M.J., Fletcher, J.N., Denton, J., Janaway, R.C, Wilson, A.S. (2014b).
Long-Term Effects of Hydrated Lime and Quicklime on the Decay of Human
Remains Using Pig Cadavers as Human Body Analogues: Field Experiments.
Forensic Science International, 238:141.e1–141.e13.
Shalaby, O.A., deCarvalho, L.M., Goff, M.L. (2000). Comparison of Patterns of

Decomposition in a Hanging Carcass and a Carcass in Contact With Soil in a
Xerophytic Habitat on the Island of Oahu, Hawaii. Journal of Forensic Sciences,
45(6):1267-1273.
Sikary, A.K., Behera, C., Murty, O.P., Rautji, R. (2016). Hands Tied with Bag Full of
Books in Suicidal Hanging. Journal of Forensic Sciences, 61(S1):S265-S267.
Simmons, T., Adlam, R.E., Moffat, C. (2010a). Debugging Decomposition Data –

Comparative Taphonomic Studies and the Influence of Insect and Carcass Size on
Decomposition Rate. Journal of Forensic Sciences, 55(1):8-13.
Simmons, T. and Cross, P.A. (2013). Forensic Taphonomy. In: Siegel, J.A. and
Saukko, P.J., eds. Encyclopedia of Forensic Sciences. 2nd ed. Waltham: Academic
Press, vol. 1, pp. 12-17.
183
Simmons T., Cross P.A., Adlam R.E., Moffat C. (2010b). The Influence of Insects on
Decomposition Rate in Buried and Surface Remains. Journal of Forensic Sciences,
55(4):889-892.
Simmons T., Moffat C., Bulut, O., Rattanarungruang, N., Roe, A.L., Siwek, D.F. (2016).
Around the World in Accumulated Degree Days. In: Proceedings of the American
Academy of Forensic Sciences 68th Annual Scientific Meeting, Las Vegas, Nevada.
Colorado Springs, Colorado: AAFS. A119, p.181.
Spicka, A., Johnson, R., Bushing, J., Higley, L.G., Carter, D.O. (2011). Carcass Mass
Can Influence Rate of Decomposition and Release of Ninhydrin-Reactive Nitrogen
into Gravesoil. Forensic Science International, 209(1-3):80-85.
Stokes, K.L., Forbes, S.L., Tibbett, M. (2013). Human Versus Animal: Contrasting
Decomposition Dynamics of Mammalian Analogues in Experimental Taphonomy.
Sun, J., Guo, X., Zhang, J., Jia, C., Xu, A. (2012). Suicide Rates in Shandong, China,
1991-2010: Rapid Decrease in Rural Rates and Steady Increase in Male-Female
Ratio. Journal of Affective Disorders, 146(3):361-368.
Swindle, M.M., Makin, A., Herron, A.J., Clubb, F.J. Jr., Frazier, K.S. (2012). Swine as
Models in Biomedical Research and Toxicology Testing. Veterinary Pathology,
49(2):344-356.
Tabor, K.L., Brewster, C.C., Fell, R.D. (2004). Analysis of the Successional Patterns of
Insects on Carrion in Southwest Virginia. Journal of Medical Entomology,
41(4):785-795.
Thomas, K. and Gunnell, D. (2010). Suicide in England and Wales 1861-2007: A

Time-Trends Analysis. International Journal of Epidemiology. 39(6):1464-1475.
184
Thomas, K.H., Beech, E., Gunnell, D. (2013). Changes in Commonly Used Methods of
Suicide in England and Wales from 1901–1907 to 2001–2007. Journal of Affective
Disorders, 144(3):235-239.
Tracqui, A. (2000). Time Since Death. In: Siegel, J.A., Saukko, P.J., Knupfer, G.C., eds.
Encyclopedia of Forensic Sciences. London: Academic Press, 1357-1363.
Turvey, B.E. (2000). Autoerotic Death. In: Siegel, J.A., Saukko, P.J., Knupfer, G.C., eds.
Encyclopedia of Forensic Sciences. London: Academic Press, 290-295.
Ueno, Y., Asano, M., Nushida, H., Nakagawa, K., Adach, J., Nagasaki, Y. (2003).
Sexual Asphyxia by Hanging – a Case Report and a Review of the Literature. Legal
Medicine, 5(3):175-180.
Vass, A.A. (2001). Beyond the Grave – Understanding Human Decomposition.

Microbiology Today, 28:190-193.
Vass, A.A., Bass, W.M., Wolt, J.D., Foss, J.E., Ammons, J.T. (1992). Time Since Death
Determinations of Human Cadavers Using Soil Solution. Journal of Forensic
Sciences, 37(5):1236-1253.
WHO. (2015). Obesity and Overweight. [Online] Available at:

http://www.who.int/mediacentre/factsheets/fs311/en/ [Accessed 26 March
2016].
Wilson, A.S., Janaway,R.C., Holland, A.D., Dodson, H.I., Baran, E., Pollard, A.M., Tobin,
D.J. (2007). Modelling the Buried Human Body Environment in Upland Climes
Using Three Contrasting Field Sites. Forensic Science International, 169(1):6-18.
185
10 Appendices
10.1 Appendix 1: Megyesi et al.’s (2005) Decomposition Scoring
Scales (Used for Scoring Pigs Lying on the Ground)
The following tables show the decomposition scoring scales developed by Megyesi
et al. (2005) for scoring the decomposition of human bodies. Each of the scores
shown has been adjusted by subtracting 1 from the original scores given by
Megyesi et al., so that a fresh body receives a total decomposition score of zero.
186
Table 10.1 Megyesi et al.’s (2005) adjusted decomposition scoring scale for the
head and neck.
Score Description
Fresh
Early Decomposition
1 Pink-white appearance with skin slippage and some hair loss.
2 Gray to green discoloration: some flesh still relatively fresh.
Discoloration and/or brownish shades particularly at edges, drying of

3
nose, ears and lips.
Purging of decompositional fluids out of eyes, ears, nose, mouth, some

4
bloating of neck and face may be present.
5 Brown to black discoloration of flesh.
Advanced Decomposition
6 Caving in of the flesh and tissues of eyes and throat.
Moist decomposition with bone exposure less than one half that of the area
7
being scored.
Mummification with bone exposure less than one half that of the area
8
being scored.
Skeletonization
Bone exposure of more than half of the area being scored with greasy
9
substances and decomposed tissue.
Bone exposure of more than half the area being scored with desiccated or
10
mummified tissue.
11 Bones largely dry, but retaining some grease.
12 Dry bone.
187
torso.
Score Description
Fresh
Early Decomposition
1 Pink-white appearance with skin slippage and marbling present.
2 Gray to green discoloration: some flesh relatively fresh.
3 Bloating with green discoloration and purging of decompositional fluids.
Postbloating following release of the abdominal gases, with discoloration

4
changing from green to black.
Decomposition of tissue producing sagging of flesh; caving in of the

5
abdominal cavity.
6
being scored.
Mummification with bone exposure of less than one half that of the area
7
being scored.
Skeletonization
Bones with decomposed tissue, sometimes with body fluids and grease
8
still present.
Bones with desiccated or mummified tissue covering less than one half of
9
the area being scored.
11 Dry bone.
188
limbs.
Score Description
Fresh
Early Decomposition
1 Pink-white appearance with skin slippage of hands and/or feet.
2 Gray to green discoloration; marbling; some flesh still relatively fresh.
Discoloration and/or brownish shades particularly at edges, drying of

3
fingers, toes, and other projecting extremities.
4 Brown to black discoloration, skin having a leathery appearance.
5
being scored.
Mummification with bone exposure of less than one half that of the area
6
being scored.
Skeletonization
Bone exposure over one half the area being scored, some decomposed
7
tissue and body fluids remaining.
9 Dry bone.
189
10.2 Appendix 2: Example ADD Estimate Calculations for a
Hanging Pig
These examples are for the inverted TBShang versus ADD regression line equation
given at the end of section 7.3.3.1:
ADDest = 234.3 + 7.38 × sgn(B) × | B |1.51
where B = TBShang – 11; |B| represents the absolute value of B, such that |B| = B
when B ≥ 0, and |B| = –B when B < 0; and sgn(B) represents the sign of B, such that
sgn(B) = +1 when B ≥ 0, and sgn(B) = –1 when B < 0.
As an example in which a negative value for B is encountered: for an observed
TBShang value of 2, B = 2 – 11 = –9, and the corresponding ADD estimate (as shown
in Table 7.3) is:
ADDest = 234.3 + 7.38 × sgn(–9) × | –9 |1.51
⇒ ADDest = 234.3 + 7.38 × –1 × 91.51
⇒ ADDest = 31
Where the argument B does not return a negative value the calculation is simpler.
For example for TBShang = 22, B = 22 – 11 = 11, and the corresponding ADD
estimate (as shown in Table 7.3) is:
190
ADDest = 234.3 + 7.38 × sgn(11) × | 11 |1.51
⇒ ADDest = 234.3 + 7.38 × 1 × 111.51
⇒ ADDest = 510
191
10.3 Appendix 3: Start Weight, Sex, and Age Tables for the Pigs
Used in the Three Experiments
10.3.1 Experiment 1
20 pigs were used for Experiment 1; 10 hanging pigs and 10 control pigs. The pigs
for the two groups came from different farms, but all were expected to be about 16
weeks old.
Table 10.4 The start weights and sex of the pigs used in Experiment 1.
Hanging Group Control Group
Label Sex Weight (kg) Label Sex Weight (kg)
H1 M 31.00 C1 M 55.00
H2 F 19.50 C2 F 36.00
H3 F 35.50 C3 M 51.00
H4 F 37.00 C4 F 52.00
H5 F 37.00 C5 F 41.00
H6 F 42.50 C6 M 55.00
H7 M 35.50 C7 M 57.00
H8 M 28.00 C8 F 50.00
H9 M 34.50 C9 M 51.00
H10 M 31.00 C10 M 31.00
192
10.3.2 Experiment 2
40 pigs were used for Experiment 2, with 10 pigs in each of four groups: hanging
unclothed; hanging clothed; surface unclothed (equivalent to the control groups in
the other two experiments; and surface clothed. The pigs all came from the same
farm and were all about 16 weeks old.
Hanging Unclothed Group Hanging Clothed Group
H1 F 28.15 HC1 M 23.30
H2 F 24.00 HC2 F 28.70
H3 M 29.00 HC3 F 23.60
H4 M 26.90 HC4 F 24.75
H5 F 29.40 HC5 F 25.75
H6 M 24.55 HC6 M 28.40
H7 F 29.55 HC7 M 24.75
H8 M 25.85 HC8 F 23.20
H9 M 28.25 HC9 M 22.35
H10 F 25.15 HC10 M 28.25
193
Table 10.5 (cont.) The start weights and sex of the pigs used in Experiment 2.
Surface Unclothed Group Surface Clothed Group
C1 F 34.20 CC1 F 38.50
C2 M 22.10 CC2 M 28.50
C3 F 25.35 CC3 F 29.55
C4 M 31.95 CC4 M 30.85
C5 M 31.55 CC5 F 21.75
C6 M 32.10 CC6 M 31.80
C7 F 30.80 CC7 M 20.90
C8 F 24.50 CC8 F 25.25
C9 F 32.75 CC9 F 31.00
C10 M 21.25 CC10 F 22.80
194
10.3.3 Experiment 3
30 pigs were used for Experiment 3, with 10 pigs in each of three groups: hanging;
semi-recumbent; and control. The pigs all came from the same farm and were all
about 20 weeks old.
Hanging Group Semi-Recumbent Group Control Group
Label Sex Weight Label Sex Weight Label Sex Weight

(kg) (kg) (kg)
H1 M 56.70 HS1 M 49.20 C1 M 56.20
H2 F 60.40 HS2 F 64.40 C2 M 43.40
H3 F 63.90 HS3 F 54.90 C3 M 41.80
H4 M 64.20 HS4 M 51.90 C4 M 40.90
H5 M 56.40 HS5 F 62.00 C5 M 54.80
H6 F 60.70 HS6 M 51.10 C6 M 36.60
H7 M 69.20 HS7 M 57.10 C7 M 54.00
H8 M 48.70 HS8 F 61.70 C8 M 43.70
H9 M 52.40 HS9 M 60.10 C9 M 48.90
H10 M 55.60 HS10 F 53.00 C10 M 45.60
195
11 Published Papers
196
J Forensic Sci, September 2015, Vol. 60, No. 5
doi: 10.1111/1556-4029.12796
PAPER Available online at: onlinelibrary.wiley.com
ANTHROPOLOGY
Jeanne Lynch-Aird1, M.Sc.; Colin Moffatt1,2, Ph.D.; and Tal Simmons3, Ph.D.
Decomposition Rate and Pattern in Hanging

Pigs
ABSTRACT: Accurate prediction of the postmortem interval requires an understanding of the decomposition process and the factors acting
upon it. A controlled experiment, over 60 days at an outdoor site in the northwest of England, used 20 freshly killed pigs (Sus scrofa) as
human analogues to study decomposition rate and pattern. Ten pigs were hung off the ground and ten placed on the surface. Observed differ-
ences in the decomposition pattern required a new decomposition scoring scale to be produced for the hanging pigs to enable comparisons with
the surface pigs. The difference in the rate of decomposition between hanging and surface pigs was statistically significant (p = 0.001). Hang-
ing pigs reached advanced decomposition stages sooner, but lagged behind during the early stages. This delay is believed to result from lower
variety and quantity of insects, due to restricted beetle access to the aerial carcass, and/or writhing maggots falling from the carcass.
KEYWORDS: forensic science, forensic anthropology, hanging, decomposition, postmortem interval, accumulated degree days, pig car-
casses
Worldwide, hanging is one of the most commonly used ways on any material falling from the body. In addition, when a body
of committing suicide (1); it is the most common method in both is hanging outdoors, a greater surface area of the body will be
England and Wales, accounting for about 2000 deaths a year (2). exposed to the drying effect of wind, which could increase the
Many individuals are found relatively quickly if the hanging possibility and/or rate of mummification. Blow flies (Diptera;
occurred in the home, often as a result of suicide or auto-erotic Calliphoridae) are less likely to lay eggs on leathery or mummi-
activities (3,4); however, when the hanging has occurred outdoors fied skin (10). Physical disturbance of carcasses has also been
or in a remote location, bodies may remain undiscovered for shown to have a significant effect on the rate of mass loss dur-
longer periods. Although using decomposition scoring and accu- ing decomposition (11,12) due to the displacement of necropha-
mulated average temperature to calculate the postmortem interval gous insects. The maggot mass within a hanging animal may,
(PMI) is possible, Parks has noted that a “one size fits all” therefore, be smaller than that within an animal on the ground,
decomposition model is unrealistic (5). A scale for scoring sub- where the maggots are more likely to remain within the body
merged bodies and a predictive equation for calculating PMI was cavity. A smaller maggot mass inside a hanging body may affect
produced by Heaton et al. (6), who found that the decomposition the rate of weight loss mediated by insects.
pattern for bodies in water was different from that described by This study presents a comparison of pattern and rate of
Megyesi et al. (7) based solely on terrestrially deposited cases. decomposition in pigs (Sus scrofa) hung, fully suspended, by the
When comparing the rate and pattern of decomposition of a neck, with others on the ground surface. We use “rate” to mean
single hanging pig and one pig on the ground, Shalaby et al. (8) the change in decomposition with respect to accumulated degree
found that the rate at which mass decreased was slower in the days (ADD), which is a variable incorporating time. Based upon
hanging carcass and that each stage of decomposition was pro- the findings of Shalaby et al. (8), the hypothesis on which this
longed. In the absence of replication, conclusions from such a experiment was based was that decomposition in hanging car-
small study must be treated with caution; however, there are rea- casses would be slower than in those on the surface of the
sons to suspect that the species of insects attracted to a hanging ground. The research required the development of a new Total
body may differ if the carcass is not touching the ground (9). Body Score for Hanging (TBShang) scale, for scoring the decom-
Unless a hanging body is partially in contact with the ground, position of hanging bodies, to complement the existing Total
crawling insects would be unlikely to return to the carcass if Body Score (TBSsurf) scale for bodies in contact with a substrate
they fall into the drip zone (8), where they may continue to feed (7). Used in conjunction with ADD, this new scale may provide
a more accurate determination of PMI for hanging bodies.
1
School of Forensic and Investigative Sciences, University of Central Materials and Methods
Lancashire, Preston, PR1 2HE, U.K.
2
Zayed University, Dubai, United Arab Emirates. The research was carried out at the University of Central Lan-
3
Department of Forensic Science, Virginia Commonwealth University,
Richmond, VA 23284, U.S.A.
cashire’s Taphonomic Research in Anthropology Centre for
Received 19 Dec. 2013; and in revised form 14 May 2014; accepted 21
Experimental Studies (TRACES) facility (13) in the northwest of
Aug. 2014. England; TRACES is a 13-acre outdoor site consisting of slop-
© 2015 American Academy of Forensic Sciences 1155

1156 JOURNAL OF FORENSIC SCIENCES
ing rough pasture land with areas of young trees, consisting of specific were also noted. The ground directly beneath the hang-
rowan, alder, birch, oak, and pine. ing pigs, or the “drip zone” (8), was examined and sampled for
Twenty pigs (S. scrofa), killed within an hour before with a any insects and parts of the pig that may have fallen. At weekly
captive bolt gun, were numbered serially as they were removed intervals, the weight of each hanging pig was recorded, mea-
from body bags and allocated into two equal groups. The first sured using the weighing scale on which they were originally
ten removed were assigned to the “hanging” group and the sec- weighed. This scale was hung by a hook over the horizontal
ond ten to the “surface/control” group. The hanging group con- scaffold bar of the A-frame next to the pig, and the pig was
tained five females and five males and the control group four moved carefully from the S hook it was hanging on onto the
females and six males. As they were removed from the body weighing S hook and then returned to the hanging frame hook.
bags, they were weighed to the nearest 5 g using a rope noose Also noted were the dates at which blow fly eggs and maggots,
tied around a hind limb and an S hook attached to a hanging and larval and adult beetles were first observed on the carcasses
scale on a fixed frame. All pigs were put into position within an of both groups.
hour of each other on 28 May 2011 with the exception of one The ambient ADD was calculated from the external loggers
of the surface pigs which, due to lack of availability on the day, attached to the hanging frames, and the internal ADD was calcu-
was killed and positioned one day later. lated from the internally sited data loggers and thermocouples.
An EL-USB-1 Lascar Electronics self-contained data logger Data from the temperature data loggers were downloaded. The
(Lascar Electronics Ltd., Salisbury, U.K.; the authors have no daily average temperature was calculated from the 6-h record-
financial interest) in a film-sealed case was inserted into the anus ings, and the ADD was calculated for the periods from the start
of each pig and pushed internally using a bamboo rod to a dis- of the study until each observation day. The study period com-
tance of approximately 40 cm. The loggers had been prepro- menced after the threat of frosts had passed, so no adjustments
grammed to record temperature at 6-h intervals throughout the for temperatures at or below 0 °C were required in the ADD cal-
experiment. The data loggers remained in place until completion culations (14). All ADD figures were calculated using tempera-
of the experiment. The pigs in the hanging group also had ther- tures in degrees Celsius.
mocouples inserted into the thorax via the esophagus in the
event that the loggers in the anus were expulsed through gravita-
Hanging Body Score Scale
tional force prior to completion of the study. The data loggers
were placed internally to enable comparisons to be made Decomposition is a mosaic process, where different parts of the
between the core temperatures of the pigs and the external tem- body decompose at different rates, which is why the system
perature, as it is known that maggots generate their own heat. defined by Megyesi et al. (7) independently scores the degree of
This also facilitated comparisons between the internal tempera- decomposition present in the head/neck, torso, and limbs. The
tures of the hanging and surface pigs in the hopes that it might hanging bodies were scored in the same three regions. Using the
provide indirect information concerning maggot mass size differ- characteristics of decomposition displayed by each pig at each
ences in these groups. ADD interval, a seriation was produced reflecting the most typical
Two hanging structures were constructed from scaffolding pattern over a known timeline. Scores were then allocated to each
poles, where a pair of A-frames supported a horizontal ridge pig based upon a review of the written description and photo-
pole 2.5 m above the ground. These structures were covered in graphs recorded at each observation. Independent scores were
a scavenger proof chicken wire mesh to a height of 1 m, above given to the three regions: head and neck, torso, and limbs. These
which was attached bird-proof plastic netting. Thus, the entire scores were added to give a TBShang on this novel scale. All scores
structure was covered and researchers’ entrance into the frame were integers with a baseline of zero equating to no visible decom-
was achieved via a detachable section of chicken wire. Neither position. As the scale records scores of visible decomposition, a
mesh nor netting impeded insect access. On each ridge pole, fresh body does not show any such decomposition and receives a
spaced approximately 80 cm apart, five pig carcasses were hung score of zero points. Where body sections did not show the same
by the neck using nylon rope tied in a noose. Each noose was levels of decomposition over the whole section, for example,
attached to the ridge pole using a butchers’ hook, which sus- across all four limbs; then, the score was an average of the two
pended the carcasses so that the hind limbs were hung a mini- extremes, to the nearest integer. The surface pigs were scored
mum of 60 cm off the ground. Each hanging structure had an using Megyesi et al.’s system (7) which began at 3 for a fresh
additional data logger, attached at a height of 1.5 m above body, as 1 is the fresh score for each of the component body parts;
ground, to record ambient temperatures. The second group of 10 head, torso, and limb. This can cause some complication, as
pigs was placed 30 m distant on the ground spaced about 5 m regression lines are not through the origin and conversions
apart. The cages were covered in scavenger proof mesh which between scales are more convoluted (15). Therefore, the scales
did not impede insect access. have been converted to begin at zero (Tables 1–3) simply by the
The observation and data collection interval for this study was subtraction of 1 point from each of Megyesi’s head, torso, and
approximately every 50 ADD; because the average daily temper- limb scores (7). Total body scores were otherwise unaltered,
atures in the north of England are only 12–15 °C, this approxi- merely progressing from 0 to, for example, 12 for the head, rather
mates twice a week (on day 3 and day 7 of each week), until than from 1 to 13. We feel this better reflects the fact that, in a
the end of the study (25 July 2011) for a total of 17 observation fresh condition, the body exhibits no decomposition.
days. At each visit, the physical state of the surface pigs was
closely observed and photographed, with scores for each of three
Statistical Analysis
regions (head and neck, torso, and limbs) recorded to produce
TBSsurf (7). Photographs were taken and detailed descriptions Linear models were used to explore the relationship between
recorded for the hanging pigs to enable seriation of the decom- ADD and measures of decomposition. Such relationships rarely
position process and development of a comparable scoring sys- produce a straight line and, while a simple mathematical trans-
tem. Any differences in appearance which appeared to be sex formation of ADD is often adequate to produce one, in these
LYNCH-AIRD ET AL. . DECOMPOSITION IN HANGING PIGS 1157
TABLE 1––Stages of hanging decomposition for the head and neck cases, it was not. Instead, we used a polynomial regression
(PBSHhang). (Northwest England Summer 2011). which fits a more complex curve, of the form:
Score Description
decomp ¼ a0 þ a1 # ADD1 þ a2 # ADD2 . . . þ an # ADDn :
0 Fresh, no discoloration
1 Swelling of head and neck The maximum rate of decomposition can, therefore, be found
2 Purging of decompositional fluids, darkening of skin to
black, sinking in around eyes, some drying of snout and lips, by calculus. Extensions of the polynomial curves beyond the
most flesh still fresh data points may well suggest a reversal of decomposition, but
3 Skin shrinking back from jaw, skin slippage these should be ignored, as the models are applicable only
4 Flesh sinking in around jaw and throat, drying of skin and lips within the ADD range encountered in data collection.
5 Hair loss, drying shrinking snout
6 Leathery skin, less than 10% bone visible
A compromise between the fit of the curve and parsimony
7 Holes in or tearing of the skin was made using goodness of fit comparisons. Significance tests
8 Complete mummification, ears dry inflexible for decomposition curve differences were carried out by a com-
9 Partial mask formed where some of the skin is lifting from skull parison of models by means of a chi-squared test. All analyses
10 Skin like parchment, very thin and translucent with holes and tears were conducted using the software R (16), using the lme4 (17)
forming
11 Bone visible over more than 50% of skull, or detached mask for package for mixed-effects models.
more than 75% of skull
12 Dry bone which may still be hanging if attached by mummified
tissue or may have fallen to the ground Results
There were a total of 17 observation periods, the last of which
corresponded to 932 ADD. Data from the internal thermocouples
in the hanging pigs could not be used as a measure of internal
ADD as the majority of the loggers failed due to the ingress of
TABLE 2––Stages of hanging decomposition for the torso (PBSThang).
(Northwest England Summer 2011). moisture despite efforts made to seal them. Therefore, ADD was
calculated based upon the temperatures recorded by the ambient
Score Description data loggers attached to the hanging frames, and internal carcass
temperatures could not be compared between surface and hang-
1 Bloating, red to green discoloration, swelling of males’ testicles ing pigs.
2 Color changing from green to gray to dark gray/black Initially, the appearances of the hanging and control pigs were
3 Prolapsed bowel, purging of decompositional fluid similar. On the first day of the study as the pigs were removed
4 Lower torso more swollen than upper. Penis and umbilicus are from the bags, as well as when they were set out in the field,
open holes on males
5 Body elongated with some caving in of upper torso
blow flies were observed landing on them. These blow flies ovi-
6 Anus open to diameter of 10 cm or greater, intestines protruding. posited in the natural orifices in the head, around the anus, and
Scrotum soft and shriveling in males in the skin creases, showing no preference for either group of
7 Drying and browning of skin, becoming leathery on upper torso pigs. Eggs were also laid around the noose on the hanging pigs.
8 Intestines dropped out of body, skin mummified. Scrotal sac opens at Once larvae had hatched and all the pigs in both groups had
back into anal opening on males
9 Stomach dropped from body, ribs, scapulae and vertebrae dropped, started to purge decomposition fluids at 147 ADD, the patterns
flanks sunken, skin very greasy to touch observed in the two groups diverged. By 229 ADD, the maggots
10 Large loss of fat onto drip zone, torso drum like, innominates and in the surface group remained visible and active, with wet
long bones dropped decomposition destroying the throat tissue and opening the car-
11 Dry bone below carcass or visible beneath mummified skin, carcass
may still be hanging
cass. Beetle adults and larvae were also present in the surface
group. In contrast, maggots were no longer visible in the hang-
ing pigs, and no beetles or their larvae were seen. By 279 ADD,
all the surface pigs’ heads had been largely consumed, and dif-
ferences in the pattern of decomposition pattern were evident.
TABLE 3––Stages of hanging decomposition for the limbs (PBSLhang).
The greater surface area exposure of the hanging pigs to wind
(Northwest England Summer 2011). and sun led to drying and mummification, inhibiting insect
access and further oviposition. Hence, little external insect activ-
Score Description ity was visible; maggots were evident only when they fell to the
drip zone, at which point, they effectively had no further impact
1 Reddening of skin, becoming dark gray, limbs swollen, and on the hanging body. The general pattern of decomposition for
extended out from the body the control and hanging pigs is summarized by ADD in Table 4.
2 Some skin slippage, drying of limbs at extremities The shape of the hanging males and females differed as the
3 Gray becoming black, forelimbs hanging in usual position decomposition progressed. At initial bloat all the hanging pigs
4 Drooping forelimbs hanging loosely, disarticulated at shoulder
5 Hind limbs hanging loosely, disarticulated at pelvis, loss of hind assumed an evenly swollen appearance with the limbs swollen
hooves, hair loss and held out from the body. As the bloat subsided, the males
6 Mummification of lower end of limbs, skin color dark brown, deflated first on the upper torso, remaining swollen around the
tears and holes in skin midtorso with the greatest width around the umbilicus and penis.
7 Bones in hanging bags of skin. Mummified with distal ends of
limbs rock hard, skin stuck to bones
The scrotum also remained swollen. This shape was retained
8 Bones dropping through torso, scapulae first then humerus. Limb until fully deflated even though the penis and umbilicus had
surface very greasy become open holes. The female pigs maintained an evenly swol-
9 Dry bone below carcass or visible beneath mummified skin, carcass len shape, appearing rectangular in frontal view and with no pro-
may still be hanging trusion around the umbilical area. This shape was retained until
TABLE 4––General pattern of decomposition for the control and hanging pigs (Northwest England summer 2011).
ADD Hanging pigs Control pigs

147 Bodies were bloated and had started to purge decomposition fluids. Bodies were bloated and had started to purge decomposition fluids.
Maggots present on and in the head. Maggots present on and in the head.
All animals had prolapsed bowels and some of the anal data
loggers had already been avulsed.
The testicles of the male pigs were considerably more swollen than
in the control pigs.
229 Maggot masses visible in the mouth. The control pigs appeared to be decomposing faster.
Very few additional maggot masses visible. Maggot masses visible in the mouth.
Drip zones, as named by Megyesi et al. (7), had now The throats of the control pigs were open and moist with visible
formed beneath the pigs. decomposition.
The maggots falling onto the drip zones varied from a few to Maggot masses were visible on the underside of the control pigs
approximately 240 ml (2 cupfuls).
279 10% of the hanging pigs displayed exposed bone in the head region. 80% of the control pigs had part of the maxilla or mandible exposed.
385 From this point there was little breaching of the hanging pigs and they became The appearance of the control pigs was now very different from the
hanging “empty pigs,” with the pig appearing externally to hanging group.
be complete but with the internal organs decomposing or decomposed. The controls could be seen to be decomposing, evincing moist
The bowels had now fully prolapsed with some loops of decomposition with parts of the torso breached and bones visible.
intestine fallen to the ground.
459 Externally, the hanging pigs displayed deflation and lengthening. Foaming maggot masses and bones were visible, including
Lifting the forelimbs showed active maggot activity in some of the axillae mandibles, limb bones, ribs, scapulae, and vertebrae.
and in some the skin was breached.
Two of the pigs had hind trotters just touching the ground. The rope was
reknotted to raise them off the ground.
No beetles or larvae were observed on the pigs or in the drip zone.
512 The hanging pigs were starting to lose their intestines, which had Few or no visible maggot masses on the skin, but these were still
dropped to the drip zone. present beneath the skin.
In the male pigs, the scrotal sac was opened and joined with the anal hole.
All the hanging males had lost their intestines, as had 20% of the females.
591 Torsos had become dry but still flexible, feeling leathery. All animals displayed bone exposure of the head and some
The stomach and quantities of fat were in the drip zone. limb bones.
The first bones were found in the drip zone: these included ribs,
scapulae, and vertebra.
631 The torsos of the hanging pigs had become mummified. Long bones and High rainfall had left the control pigs lying in pools of water, and
innominates from either or both sides of the body were found all of them had soft swollen skin and were fatty.
in the drip zone.
767 The hanging pigs were elongated, mummified, and hard. All the heads were skeletonized.
The majority of animals had more than 50% of the bones visible
in the limbs and torso.
828 Prolonged heavy rainfall left all the pigs looking fatty and white Prolonged heavy rainfall left all the pigs looking fatty and white
where rehydrated. where rehydrated. The control pigs were lying in rain water.
932 The hanging pigs were mummified and all remained hanging. All the controls now had more than 50% bone exposure on the head,
limbs, and torso.
fully deflated, leaving a pouch of skin drooping on the lower give a Partial Body Score Limbs (PBSLhang). Total Body Score
abdomen. Figure 1 Illustrates progression of decomposition, in for Hanging (TBShang) was a sum of the three region scores and
one pig, throughout the study. provided an overall measure of decomposition for the hanging
In both males and females, the anus enlarged to around 10 cm body with a minimum score of 0 and a maximum of 32 points.
in diameter early in the decomposition. In a hanging position, Tables 2–4 list the decomposition scoring scales used for the
the anus of a pig is not at the lowest part of the torso but higher three regions of the hanging pigs.
and dorsal. In the males, as the decomposition progressed, the In hanging pigs, the sexes showed significantly different pat-
back of the scrotal sac decomposed joining with the anus and terns of weight loss (v2 = 24.4, df = 3, p < 0.001), with females
creating a much larger opening directly below the body (Fig. 2) losing weight more slowly and retaining more mass throughout
enabling bones to drop out. In the females, the anus remained (Fig. 5). This slope for males was 0.17% loss per ADD (at
open only at the back with the pouch of stretched skin remaining ADD = 483), and this slope for females was 0.13% loss per
from the bloating stage left hanging lower than the anus (Fig. 3). ADD (at ADD = 503).
Some bones collected in this pouch, below the anus (Fig. 4), When using “Total Body Score” as the response variable,
and were still there at the end of the study period. The first bee- there is again a difference between sexes (v2 = 15.6, df = 2,
tle larvae were observed on the hanging pigs at 869 ADD, some p < 0.001), although this is not as large, and again, males show
640 ADD later than in the surface group. a greater rate of decomposition, as illustrated in Fig. 6. This
curve is a quadratic, and, as can be seen from the plot, maxi-
mum slope is found where ADD = 0, which is 0.062 TBShang
Total Body Score for Hanging Bodies (TBShang)
per ADD for female, and 0.068 TBShang per ADD for male;
The scale produced for scoring the hanging pigs is based on a thus, despite being significantly different, the differences are
score for the head and neck between 0 and 12 points to give a small.
Partial Body Score Head (PBSHhang); a score for the torso Using the Total Body Score/Total Body Score for Hanging
between 0 and 11 points to give a Partial Body Score Torso (TBSsurf/TBShang) to compare the rate of decomposition of
(PBSThang); and a score for the limbs between 0 and 9 points to surface and hanging pigs, a highly significant difference was
FIG. 1––Hanging female pig showing progression of decomposition over the range of Accumulated Degree Days (ADD) (Northwest England Summer 2011).
FIG. 2––Hanging male pig at 512 ADD (Northwest England Summer FIG. 4––Hanging female pig at the close of the experiment, 950 ADD
2011). The arrow points to the anal opening extending beneath body. (Northwest England Summer 2011). The pouch of skin was cut open and the
arrow points toward the bones which had collected in the pouch.
model sex data

Female
Male
80
60
% Weight Loss
40
20
0
200 400 600 800
Accumulated Degree Days
FIG. 5––Differences in percentage weight loss between male and female

hanging pigs to 900 ADD (Northwest England Summer 2011). The model
FIG. 3––Hanging female pig at 600 ADD (Northwest England Summer points show mean values for each sex at each value of ADD. The model
2011). Arrow A points to the pouch of skin hanging below the below anal lines are the polynomial regression (here fourth order) fitted values. Some
opening (arrow B). jitter has been added to the ADD values of the data points for clarity.
evident (v2 = 104, df = 5, p < 0.001), with hanging pigs reach- The situation here is more complicated, requiring a fifth-order
ing later stages of decomposition more quickly (Fig. 7). Hence, polynomial. In relatively early decomposition, for example,
the results of the experiment did not support the hypothesis; around 150 ADD, decomposition rates are steeper and similar
hanging pigs, in fact, decomposed more rapidly in the latter (ground surface: 0.051 TBSsurf/ADD, hanging: 0.060 TBShang/
stages of decomposition than those on the surface of the ground. ADD); whereas later, when decomposition slows on the ground,
human cadavers. Therefore, it is suggested that Figs 6 and 7 are

30
Female
Male the most convenient way of communicating the relationship
between ADD and TBS in these cases.
25
Discussion
Total Hanging Body Score
20
Accumulated degree day constitutes the accumulation of ther-

mal energy (18), and when a given amount of thermal energy is
accrued by a carcass, a similar degree of decomposition should
15
be apparent, yielding comparable decomposition scores for

equivalent ADD. Accumulated degree days as calculated from
ambient temperature records provide the best measure of temper-
10
ature over time for decomposition studies, enabling comparison

of data from different seasons, years, and geographical areas. It
is well established that maggot activity produces heat and the
5
maggot mass-related internal carcass temperatures may be

5–10 °C above the ambient temperature during this time (19).
As increased temperature escalates the rate of decomposition,
0
factoring in the increased maggot generated heat when consider-

0 200 400 600 800
ing ADD may allow for more accuracy in predicting PMI. This
Accumulated Degree Days study hoped to compare the internal and external temperatures of
the pigs using internal data loggers and thermocouples, but these
FIG. 6––Rate of decomposition showing the difference between the male failed to record over the duration of the study period or were
and female pigs using Total Body Scores for Hanging (TBShang) to 900 ADD
deactivated by decomposition fluid and/or rain.
(Northwest England Summer 2011). Jitter has been added to ADD for clar-
ity. The difference in shape between the hanging males and
females during much of their decomposition may be explained
by the positioning of the urethral opening. In the males, this is
situated just posterior to the umbilicus and may thus provide
Hanging
30
Grounded
another escape route for the gases produced during decomposi-
tion, making this point the widest as the gases gather there.
25
The escaping gases would also attract insects, particularly Dip-

tera, with the umbilicus and penis of the males soon becoming
TBShang/TBSsurf
20
full of larvae and progressing to open holes. This did not occur
in the females where the urethral opening is just ventral to the
15
anus, indicating that it is the urethral opening that is the weak

point rather than the umbilicus. Without this escape route
10
higher on the torso, the females retained a rectangular shape

and, as they deflated, the skin which had remained stretched
5
and swollen now formed a pouch of loose skin at the lower

torso. This pouch of skin hung lower than the anal opening, so
0
that disarticulated bones dropping down within the female pigs

0 200 400 600 800 could become trapped here rather than falling to the ground.
Accumulated Degree Days This contrasted with the males where the tissue of the swollen
scrotum decomposed and it became part of the anal hole. A
FIG. 7––Comparison of decomposition rates of hanging and grounded pigs single large opening directly beneath the pig was produced and
using Total Body Scores for Hanging (TBShang) and Surface (TBSsurf) pigs,
respectively, to 900 ADD (Northwest England Summer 2011). Jitter has been this allowed bones to fall through more easily. While of note in
added to the ADD values for clarity. this study, this feature is unlikely to be of any significance
when observing humans as, when hanging, the anal openings
hanging slows even more (e.g. at 650 ADD grounded: 0.020 of quadrupeds and bipeds are not in the same place. In humans,
TBSsurf/ADD, hanging: 0.014 TBShang/ADD). the anal opening would be directly below the body and the
As the relationship between ADD and body score followed positioning of the human penis is also low down, not at the
curves when comparing sex and treatment, simple quantitation umbilical level, so the shape differences between the male and
of the patterns is challenging. It is not appropriate to give an female pigs in late bloat and deflation may not be present in
“average” rate of decomposition based upon a single point (e.g. humans.
a TBS of 10 at 200 ADD) as the relationship up until that point The weight loss started slowly, increased, and then slowed
is not a straight line. This means that rate of decomposition is down toward the end of the decomposition as would be expected
different for every value of ADD. An alternative would be to (8,20). The difference in decomposition pattern of the male and
present the formulae for the lines, but being fifth-order polyno- female pigs may account for the difference in percentage weight
mials, these are unwieldy and we believe unhelpful to most read- loss during decomposition, as bones were lost from the males
ers. Besides, overall patterns are of greater interest than through the anal opening. While the rate of decomposition
deterministic models where one would plug in a value of ADD between the males and female pigs was shown to be statistically
and get out a value of TBS. This would not be appropriate for a significant, the actual difference is very small. The hanging pigs
study such as this which uses pig carcasses as surrogates for in this experiment were weighed weekly, and in doing so, it is
inevitable that the maggot masses would have been disturbed. all of the time between 54 ADD and 159 ADD before being
Researchers (11,12) suggest that the physical disturbance of a raised up. During this time, effectively, these pigs were not
carcass may disrupt the feeding of the maggots and retard the hanging. Beetles can and do fly well, but relatively little time is
rate of decomposition, but any such effect would be expected to spent in flight; beetles are very much insects of the ground and
be the same for both the males and females. Weight loss in soft low vegetation (9). The pigs’ hind limbs resting on the ground
tissues is a true measure of decomposition. This study has would have provided the beetles with an easy means to reach
shown, however, that care is required in extending this principal the hanging pigs, and the pilot study demonstrated that they took
to whole body weight loss. Past studies (12,20) have shown a advantage of this route.
correlation between decomposition rate and weight loss. This Additionally, between 54 ADD and 200 ADD, five of the
may not be the case where part of the body has been lost or dis- pilot study pigs herniated the intestine providing another site for
persed, for example, in scavenging or in the case of the hanging the release of volatile gases, attracting flies and providing a
pigs in skeletal element loss via gravity. moist access to the carcass for oviposition. It was noted that four
The Megyesi et al.’s. (7) Total Body Score scale was proof these five pigs were male. In pigs, unlike humans, the intes-
duced using a retrospective cross-sectional study of humans tines do not usually lie above the inguinal canal, but in the hang-
which, by its nature, provided a single snapshot of the stage of ing position, the gravity and pressure of gases may have acted
decomposition at which each individual was found, rather than a on them pushing them into the inguinal canal and causing them
longitudinal study observing progressive changes in the same to herniate.
individuals over the whole decomposition process. Although The difference in the rate of decomposition, shown in this
Megyesi et al. studied many individuals covering the range of study with the hanging pigs lagging behind the controls in the
decomposition stages, there were no cases involving hanging early stages of decomposition, is in agreement with Shalaby
bodies. It appears from the findings of the hanging studies, et al. (8) and the pilot study. This difference may be attributed
including the present longitudinal study, that the pattern of to the differences in levels of insect activity between the two
decomposition in hanging bodies is markedly different, making groups. Insect access to the hanging pigs, particularly for Cole-
Megyesi et al.’s scale unsuitable for scoring these bodies. optera (beetles), was limited and this may explain the delay in
The new scoring scale produced for hanging pigs includes decomposition of the hanging pigs. While many egg masses
mummification in the description of what is seen during the were laid by blow flies around the noose, these desiccated
decomposition. While it could be argued that it should not be in unhatched and thus contributed to limiting the quantity of mag-
the scale as it is not a feature common to all decomposition, it gots on the hanging pigs. Additionally, the action of gravity and
was found to occur in all 20 hanging pigs of the present 2011 movement disturbed the maggot masses causing maggots to fall
study and was also observed in a previous trial in 2010, which from the hanging pigs to the drip zones below, thus decreasing
would indicate that it is a feature of hanging, much like adipo- the internal maggot masses. This has implications for calculating
cere is a feature of many immersed and buried bodies. The char- the PMI of hanging bodies as many are found with the feet in
acteristic mummification may be due to the higher surface area contact with the ground. If this has occurred at the time of hang-
of the carcass being exposed to sunshine and wind, which dry ing or rapidly after, the insect access is likely to be similar to
the body out quickly. The same feature may not be found where that of a body on the ground. Thus, the rate of decomposition,
wind is not a relevant environmental factor or where humid con- at least initially, may be less like that of a body suspended
ditions prevail. totally off the ground. Further investigative studies are currently
Decomposition scores generated using the new hanging scale underway.
cannot be directly compared with those from Megyesi et al.’s Also noted was the change to the appearance of bodies on the
scale as they describe different patterns of decomposition; there ground when they had become sodden and waterlogged follow-
is, currently at least, no established absolute scale of decomposi- ing prolonged rain. This rehydrating of the tissues and fats gave
tion which could be used to link them. However, both scales the bodies the appearance of being “fresher” than prior to the
span the same range of decomposition and, once adjusted for a rain. This may give the impression that the progression of
zero baseline, assess the same body areas using the same point decomposition is less advanced and lead to a lower TBSsurf
ranges. which, in turn, will impact the PMI calculations possibly provid-
For their study, Megyesi et al. calculated ADD values using ing an inadvertently lowered PMI. An accurate PMI is important
the daily average temperature records from the nearest national in the establishing the events surrounding a death and creating a
weather station for each body, while the PMI in each case was timeframe when searching for missing persons. It may, therefore,
determined either from insect evidence or the associated police be prudent to check the amount of rainfall and duration in the
investigation. For this study, the ADD was calculated from two days prior to a body being found if it does look particularly fatty
temperature data loggers situated within 10 m of the pigs (21), and make allowances for the apparent retardation of decomposi-
with measurements at 6-h intervals, while the PMI was known tion. To ensure that PMI calculations are not inadvertently low-
to within 1 h. Both the ADD and PMI values for this study ered, it may be necessary to increase the range of dates with in
should, therefore, be more accurate than those used in Megyesi which the PMI falls.
et al.’s study.
In a pilot study (unpublished data) at TRACES conducted in
Acknowledgments
the previous year, the rate of decomposition of hanging pigs was
significantly different (faster) than in this study. A noticeable The authors would like to thank Mary Megyesi for her
difference was the presence of beetles on the carcasses at an timely and helpful replies to our questions regarding her 2005
early stage thus increasing the diversity of insects. It has been publication and Peter Cross for his assistance at TRACES, in
demonstrated that the greater the diversity of insects, the greater particular, for holding the pigs while they were lifted on and
the rate of decomposition (10). Nine of the pilot study’s hanging off the weighing hook, in all their various stages of decomposi-
pigs had their hind limbs in contact with the ground for some or tion.
References 13. Cross P, Simmons T, Cunliffe R, Chatfield L. Establishing a taphonomic

research facility in the UK. Forensic Sci Policy Manage 2009;1:187–91.
1. Thomas K, Gunnell D. Suicide in England and Wales 1861–2007: a 14. Vass AA, Bass WM, Wolt JD, Foss JE, Ammons JT. Time since death
time-trends analysis. Int J Epidemiol 2010;39(6):1464–75. determinations of human cadavers using soil solution. J Forensic Sci
2. Bennewith O, Gunnell D, Kapur N, Turnbull P, Simkin S, Sutton L, 1992;37(5):1236–53.
et al. Suicide by hanging: multicentre study based on coroners’ records 15. Gruenthal A, Moffatt C, Simmons T. Differential decomposition pat-
in England. Br J Psychiatry 2005;March(186):260–1. terns in charred versus un-charred remains. J Forensic Sci 2012;57
3. Bayard RW, Winskog C. Autoerotic death: incidence and age of victims (1):12–8.
– a population-based study. J Forensic Sci 2012;57(1):129–31. 16. R Core Team. R: a language and environment for statistical computing.
4. Ueno Y, Asano M, Nushida H, Nakagawa K, Adach J, Nagasaki Y. Sex- Vienna, Austria: R Foundation for Statistical Computing, ISBN
ual asphyxia by hanging – a case report and a review of the literature. 3-900051-07-0, 2010; http://www.R-project.org/.
Leg Med 2003;5(3):175–80. 17. Bates D, Maechler M, Bolker B. lme4: linear mixed-effects models using
5. Parks CL. A Study of the human decomposition sequence in central S4 classes. R package version 0.999375-42, 2011; http://CRAN.R-
Texas. J Forensic Sci 2011;56(1):19–22. project.org/package=lme4.
6. Heaton V, Lagden A, Moffatt C, Simmons T. Predicting the post-mortem 18. Simmons T, Adlam RE, Moffatt C. Debugging decomposition data –
submersion interval for human remains recovered from UK waterways. J comparative taphonomic studies and the influence of insect and carcass
Forensic Sci 2010;55(2):302–7. size on decomposition rate. J Forensic Sci 2010;55(1):8–13.
7. Megyesi MS, Nawrocki SP, Haskell NH. Using accumulated degree-days 19. Simmons T, Cross P, Adlam R, Moffatt C. The influence of insects on
to estimate the postmortem interval from decomposed human remains. J decomposition rate in buried and surface remains. J Forensic Sci 2010;55
Forensic Sci 2005;50(3):618–26. (4):889–92.
8. Shalaby OA, deCarvalho LM, Goff ML. Comparison of patterns of 20. Payne JA. A Summer carrion study of the baby pig Sus scrofa linnaeus.
decomposition in a hanging carcass and a carcass in contact with soil in Ecology 1965;46(5):592–602.
a xerophytic habitat on the island of Oahu, Hawaii. J Forensic Sci 21. Dabbs GR. Caution! All data are not created equal: the hazards of using
2000;45(6):1267–73. National Weather Service data for calculating accumulated degree days.
9. Chinery MA. Field guide to the insects of Britain and northern Europe, Forensic Sci Int 2010;202(1–3):49–52.
2nd edn. London: Collins, 1976.
10. Campobasso CP, Di Vella G, Introna F. Factors affecting decomposition Additional information and reprint requests:
and diptera colonization. Forensic Sci Int 2001;120(1–2):18–27. Tal Simmons, Ph.D.
11. Cross P, Simmons T. The influence of penetrative trauma on the rate of P.O. Box 843079
decomposition. J Forensic Sci 2010;55(2):295–301. 1015 Floyd Avenue, Room 2015
12. Adlam RE, Simmons T. The effect of repeated physical disturbance on Richmond, VA 23284
soft tissue decomposition – are taphonomic studies an accurate reflection E-mail: [email protected]
of decomposition? J Forensic Sci 2007;52(5):1007–14.

Jeanne Lynch-Aird Thesis June 2016

Uploaded by

Copyright:

Available Formats

Jeanne Lynch-Aird Thesis June 2016

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jeanne Lynch-Aird Thesis June 2016

Uploaded by

Copyright:

Available Formats

Estimation of Post-Mortem Interval

Using Decomposition Scales for

Jeanne Elizabeth Lynch-Aird

A thesis submitted in partial fulfilment for the requirements of

Concurrent registration for two or more academic awards

Material submitted for another award

Type of Award Doctor of Philosophy

School Forensic and Applied Sciences

The extent of decomposition of a body can be used, in conjunction with

accumulated degree days (ADD), to provide an estimate of the post-mortem

decomposition pattern in hanging bodies to be different from that of a body on the

ground, but the sample sizes used have been small.

This study presents the results of a series of decomposition studies on hanging

bodies in a variety of situations; clothed and unclothed, and fully or partially

controls. The pattern of decomposition in hanging pigs was found to differ

partially suspended bodies.

F2, 477 = 1238).

so small as to make no practical difference across the range of start weights

encountered. The effect of variation in start weight may be of greater concern,

increased fat-to-muscle ratios encountered in such bodies.

List of Figures ...................................................................................................................... viii

Acknowledgements ........................................................................................................... xiii

Abbreviations ...................................................................................................................... xiv

3 General Materials and Methods ........................................................................... 24

4 EXPERIMENT 1 – Comparing Decomposition Patterns and Rates

5 EXPERIMENT 2 – The Effect of Clothing on the Decomposition Pattern

3.1 Diagrammatic representation of the scaffolding A-frames used 26

3.2 Hanging pigs in one of the scaffolding A-frames at the start of 27

4.1 Insertion of thermocouple for attachment to temperature data 34

4.5 Hanging female at 600 ADD. Arrow A points to the pouch of 48

4.7 Differences in percentage weight loss between male and female 53

4.8 Rate of decomposition showing the difference between the 54

4.9 Comparison of decomposition rates of hanging and grounded 55

5.1 Data loggers sealed in 50 ml conical centrifuge tubes with wires 67

5.5 Differences in the rate of decomposition of the S/C and H/U 87

5.6 Differences in the rate of decomposition of the S/C and H/U 88

6.1 The position of a semi-recumbent pig, supported from a 98

6.2 Decomposition sequence for one hanging, one semi-recumbent, 114

6.3 One of the semi-recumbent pigs showed a large mass of 120

4.1 Stages of hanging decomposition for the head and neck. 50

4.2 Stages of hanging decomposition for the torso. 51

4.3 Stages of hanging decomposition for the limbs. 52

7.2 Showing the estimated Accumulated Degree Days (ADD) in °C 144

7.3 Showing the estimated Accumulated Degree Days (ADD) in °C 145

over the course of this study.

regardless of their state of decomposition.

through the Ph.D. process ahead of me.

the endless spelling mistakes and proof reading. Chocolate is owed.

ADD Accumulated degree days

ADH Accumulated degree hours

ANOVA Analysis of Variance

ARF Anthropological Research Facility (Tennessee, USA)

ATP Adenosine Tri-Phosphate

BMI Body Mass Index

CRISPR Clustered Regularly Interspaced Short Palindromic Repeats

CPSC Consumer Product Safety Commission (USA)

DEFRA Department for Environment, Food & Rural Affairs (UK)

PBS Partial Body Score

PBSH Partial Body Score for the Head