- Forensic Science International

Forensic Science International 253 (2015) 134.e1–134.e7
Contents lists available at ScienceDirect
Forensic Science International
journal homepage: www.elsevier.com/locate/forsciint
Forensic anthropology population data
Relevance of discrete traits in forensic anthropology: From the first
cervical vertebra to the pelvic girdle
Emeline Verna a,*, Marie-Dominique Piercecchi-Marti a,b, Kathia Chaumoitre a,c,
Pascal Adalian a
a
b
c
Aix-Marseille Université/EFS/CNRS/UMR 7268 ADES, 13916 Marseille, France
Service de Médecine Légale, CHU Timone, 264 Rue Saint-Pierre, 13385 Marseille Cedex 5, France
Service d’Imagerie Médicale, Hôpital Nord-CHU Marseille, Chemin des Bourrely, 13915 Marseille Cedex 20, France
A R T I C L E I N F O
A B S T R A C T
Article history:
Available online 14 May 2015
In forensic anthropology, identification begins by determining the sex, age, ancestry and stature of the
individuals. Asymptomatic variations present on the skeleton, known as discrete traits, can be useful to
identify individuals, or at least contribute to complete their biological profile.
We decided to focus our work on the upper part of the skeleton, from the first vertebra to the pelvic
girdle, and we chose to present 8 discrete traits (spina bifida occulta, butterfly vertebra, supraclavicular
nerve foramen, coracoclavicular joint, os acromiale, suprascapular foramen, manubrium foramen and
pubic spine), because they show a frequency lower than 10%.
We examined 502 anonymous CT scans from polytraumatized individuals, aged 15 to 65 years, in
order to detect the selected discrete traits. Age and sex were known for each subject. Thin sections in the
axial, coronal and sagittal planes and 3D volume rendering images were created and examined for the
visualization of the selected discrete traits.
Supraclavicular foramina were found only in males and only on the left clavicle. Coracoclavicular
joints were observed only in males. The majority of individuals with a suprascapular foramen were older
than 50 years of age. Pubic spines were observed mostly in females. Other traits did not present
significant association with sex, age and laterality. No association between traits was highlighted.
Better knowledge of human skeletal variations will help anthropologists come closer to a positive
identification, especially if these variations are rare, therefore making them more discriminant.
ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords:
Discrete traits
Forensic anthropology population data
Identification
Bones
1. Introduction
In forensic anthropology, identification begins by determining
sex, age, ancestry and stature of individuals. However, some
individuals may share the same biological profile. When this
occurs, bone discrete traits can be useful to identify individuals, or
at least contribute to completing their biological profile [1],
especially if they show a frequency lower than 10%. In this case
discrete traits are qualified as rare [1,2], i.e. few individuals have
them. Individualizing traits must also be easily observable through
imaging, on CT-scans and radiographies, in order to allow ante and
post mortem data comparison. Post-mortem data can consist of
medical images or dry bones.
* Corresponding author. Tel.: +33 491698870.
E-mail address: [email protected] (E. Verna).
http://dx.doi.org/10.1016/j.forsciint.2015.05.005
0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.
Bone discrete traits are asymptomatic anatomical variations [3]
and can be found on many bones of the human skeleton. They have
a specific location on the skeleton and can be classified as either
present or absent [3]. The exact etiology of most discrete traits is
unknown but genetic, epigenetic and environmental factors are
probably involved [4]. Discrete traits are generally an incidental
finding of medical imaging.
Cranial discrete traits [5] have been studied more commonly
than postcranial ones. Studies on cranial traits have been more
extensive and some authors have tested the association between
them [6]. This hasn’t been done for postcranial traits. An
association between two traits can bring additional information
for identifying an individual or for completing the biological
profile.
This study was conducted on a sample of French population.
Some discrete traits are population dependent, so the goal of the
study is to provide frequencies for French population and compare
them with other populations. This allows us to see if the selected
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
134.e2
traits are population dependent or if they have the same frequency
in all populations. In the latter case, the traits could be useful and
observable universally, particularly since the ancestry of the
analyzed individual is unknown.
In this study, we decided to focus our work on the postcranial
skeleton; from the first vertebra to the pelvic girdle. Using 3D
volume rendering from CT scans, we documented the frequency of
the eight selected discrete traits and their association with sex,
laterality and age in a modern population of Southern France. The
selected discrete traits present genuine relevance for forensic
identification.
2. Materials and methods
2.1. Sample
CT scans from polytraumatized individuals were carried out in
the medical imaging department of our institution. Age and sex
were known for each subject. We chose to study individuals
between 15 and 65 years of age in order to avoid degenerative bone
problems or insufficient ossification that could interfere with the
visualization of the discrete traits. Good quality scans of
individuals without bone disease and of the required age range
were selected. This provided a final sample of 502 subjects
composed of 344 males and 158 females (Fig. 1).
2.2. Method
The CT scans were performed with a 64-row multidetector CT
(Somatom Sensation 64, Siemens1, Erlangen, Germany). Scanning
parameters were as follows: 120 kV, 50–150 mA s, thickness = 0.6 mm. Scanning extended from the level of the first
cervical vertebra to the pelvis. Most scans were obtained after
administration of an intravenous contrast media.
Volume reconstruction was carried out on CT scan images
(DICOM images) with Avizo version 7 Software. Thin sections in
the axial, coronal and sagittal planes and 3D volume rendering
images created were examined for detecting the presence or
absence of selected discrete traits by a single observer. 3D
reconstruction allowed us to easily observe the selected discrete
traits, even without a special training in CT imaging.
2.3. Definition of the selected discrete traits
Four regions were examined: the scapular girdle, the thorax, the
vertebral spine and the pelvic girdle. Discrete traits were selected
from literature. At the beginning of our study, all discrete traits
present on our region of interest were selected and rated. The pool
of discrete traits was composed of 40 different traits. Out of these
40 traits, around 30 presented a frequency inferior or equal to 10%
but not all of them were considered relevant for forensic
identification. We kept only eight discrete traits with a frequency
lower than 10% (i.e. corresponding to a rare trait), easily observable
for a non-expert on medical images and were not subject to interand intra-observer errors as reported previously [2], and were
considered useful in forensic anthropology for identification
purposes. The selected eight traits are spread out on all the bones
examined.
We have selected two variations for the spine: spina bifida
occulta [2] and butterfly vertebra [7], which can affect all vertebrae.
Spina bifida occulta (SBO), an asymptomatic form of spina bifida, is
due to an ossification failure of the laminae during the fusion of the
neural arch [2]. This incomplete fusion of the neural arch of the
vertebra is usually observed in the lumbosacral region. We
regrouped SBO in two categories to facilitate its study: SBO of
the spine (from the first cervical vertebra to the fifth lumbar one)
and SBO of the sacrum.
The origin of the butterfly vertebra is a developmental anomaly.
The division of the center of the vertebra body along the sagittal
plane into two laterally wedge-shaped halves, often asymmetrical
in form, with decreased or absent anterior portions gives them a
butterfly wings-like shape observable on radiographs [7].
Discrete traits selected for the scapular girdle were as follows:
supraclavicular nerve foramen [8], coracoclavicular joint [9], os
acromiale [10] and suprascapular foramen [11].
The supraclavicular nerve foramen is a foramen situated on the
superior face of the mid-shaft of the clavicle [2]. This foramen
allows the passage of the supraclavicular nerve. The coracoclavicular joint is a diarthrotic synovial joint between the conoid
tubercle of the clavicle and the superior surface of the horizontal
part of the coracoid process of the scapula [12]. The os acromiale is
an accessory bone resulting from the failure of the acromial
apophysis to fuse with the scapula during adolescence [13]. The
acromial apophysis typically begins to fuse by age 17 or 18 and
completes its union by the age of 20 years [14]. Therefore, we only
examined individuals aged 20 years and more for this trait. The
suprascapular foramen is a foramen situated on the lateral part of
the superior border of the scapula, running medial to the coracoid
process. It is due to the ossification of the superior transverse
scapular ligament [15,16].
We selected the manubrium foramen [17] for the sternum and
the pubic spine for the pelvic girdle.
The manubrium foramen is a round defect resulting from a
failure of fusion between the two ossification centers of the bone.
Generally, the manubrium is composed of one ossification center
[14] but in some cases another center can be observed.
The pubic spine is a small osseous spur located on the pubic
branch. The length of the spine can vary between individuals.
2.4. Statistical analysis
Fig. 1. Distribution of our sample according to sex and age.
For each discrete trait, the prevalence was calculated in our
population according to sex, laterality and age. We also looked for
significant associations with sex, laterality or age using the chisquare test [18]. When the conditions of chi-square test were not
met, we used Fisher’s exact test. Statistical analyses were carried
out with the R Software (version 3.0.1), with a threshold level of
significance of 5%. In order to facilitate statistical tests, the age of
the individuals was categorized as follows: under 20 years, 20–29
years, 30–39 years, 40–49 years, 50–59 years, and over 60 years.
We evaluated the association between discrete traits using
Cramer’s V coefficient. It is a measure of association between two
nominal variables, giving a value between 0 (very poor association
between variables) and 1 (very high association between variables)
included [19].
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
134.e3
Table 1
Number and prevalence of vertebral discrete traits in our sample.
3. Results
3.1. Spine
Table 1 gives the prevalence found for the discrete traits
selected for the spine.
Spina bifida occulta (Fig. 2) was found only on the first cervical
vertebra with a prevalence of 4.6% (23 individuals). The other
cervical vertebrae were not affected in our population. For the
thoracic vertebrae only the first and second vertebrae displayed
the trait, with a frequency of 0.4% (2 individuals) and 0.2% (one
individual) respectively. For lumbar vertebrae, only the fifth was
concerned with a prevalence of 1.6% (8 individuals). Other
vertebrae did not show this trait.
The sacrum was also affected by spina bifida occulta. All the
sacral vertebrae can display this anomaly but we chose to present
the rarest type of spina bifida occulta, the sacral hiatus. It involves
all sacral vertebrae (from S1 to S5) and in our sample the frequency
was 1.8% (9 individuals). No statistically significant difference was
found according to sex or age (p > 0.05).
The butterfly vertebra trait (Fig. 3) affected only two vertebrae
in our whole sample. One individual showed this trait on the fourth
lumbar vertebra (0.2%) and 5 individuals displayed a butterfly
vertebra on the first sacral vertebra (1%). No statistically significant
difference was found according to sex or age (p > 0.05).
3.2. Scapular girdle and thorax
Tables 2 and 3 present the prevalence of discrete traits selected
on the scapular girdle and thorax.
The supraclavicular nerve foramen (Fig. 4) was found for
3 individuals (0.6%), only in males and only on the left clavicle. No
statistically significant difference was found by age (p > 0.05).
The coracoclavicular joint (Fig. 5) was observed for 3 individuals
(0.6%) and in males only. This trait was present one right clavicle
Trait
Vertebra
Total
n
%
n
%
n
%
Spina bifida occulta
C1
T1
T2
L5
S1–S5
L4
S1
23
2
1
8
9
1
5
4.6
0.4
0.2
1.6
1.8
0.2
1
18
1
0
7
7
1
1
5.2
0.3
0.0
2.0
2.0
0.3
0.3
5
1
1
1
2
0
5
3
1
1
1
1.3
0
3
Butterfly vertebra
Males
Females
and 3 left clavicles. One case of a bilateral coracoclavicular joint
was found in our sample. No statistically significant difference was
found according to laterality or age (p > 0.05).
The os acromiale (Fig. 6) occurred in 13 individuals (3%). This
trait was observed on 8 left scapulae and on 9 right ones. Bilateral
cases were found for 4 individuals. No statistically significant
difference was found according to sex, laterality, or age (p > 0.05).
The suprascapular foramen (Fig. 7) was found for 14 individuals
(2.8%). This trait was observed on 10 left scapulae and on 9 right
ones. Five individuals had bilateral suprascapular foramina. No
statistically significant difference was found according to sex or
laterality (p > 0.05). However, eight of individuals with a
suprascapular foramen were more than 50 years of age (pvalue = 0.0002).
A manubrium foramen (Fig. 8) was found for one male only
(0.2%).
3.3. Pelvic girdle
A pubic spine (Fig. 9) was observed for 15 individuals (3%),
2 males and 13 females (Tables 2 and 3). A statistically significant
difference (p-value = <0.0001) was found according to sex. The
Fig. 2. Spina bifida occulta on the first thoracic vertebra (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
Fig. 3. Butterfly vertebra (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
134.e4
Table 2
Number and prevalence of scapular and pelvic girdles and thoracic discrete traits in our sample.
Bone
Trait
Clavicle
Suprascapular nerve foramen
Coracoclavicular joint
Os acromiale
Suprascapular foramen
Manubrium foramen
Pubic spine
Scapula
Sternum
Pelvis
Total
Males
Females
n
%
n
%
n
%
3
3
13
14
1
15
0.6
0.6
3.0
2.8
0.2
3.0
3
3
9
7
1
2
0.9
0.9
3.1
2.1
0.1
0.6
0
0
4
7
0
13
0
0
3.0
4.5
0
8.4
Table 3
Number and prevalence of bilateral traits.
Bone
Trait
Left
Clavicle
Suprascapular nerve foramen
Coracoclavicular joint
Scapula
Pelvis
Right
%
n
n
Unilateral
Bilateral
%
n
%
n
%
3
1
0
0.2
0
3
0
0.6
3
2
0.6
0.4
0
1
0
0.2
Os acromiale
Suprascapular foramen
8
10
1.8
2.0
9
9
2.0
1.8
9
9
2.0
1.8
4
5
0.9
1.0
Pubic spine
14
2.8
11
2.2
5
1.0
10
2.0
pubic spine was more frequently present in females. Pubic spines
were found on 14 left and 11 right pubic bones bilateral cases were
found for 10 individuals. No statistically significant difference was
found according to age or laterality (p > 0.05).
3.4. Association between traits
Table 4 presents Cramer’s V coefficient. No association between
traits was observed. The highest Cramer’s V was found between the
Fig. 4. Supraclavicular nerve foramen of the left clavicle (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
Fig. 5. Coracoclavicular joint on the left clavicle (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
134.e5
Fig. 6. Suprascapular foramen of the right scapula (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
pubic spine and the suprascapular foramen, with a value of 0.191,
indicating no significant association.
4. Discussion
Discrete traits are useful in forensic anthropology for adding
individualising information to a biological profile, and thus help
with the identification or exclusion of an individual. We
documented the prevalence of eight traits on the upper part of
the post-cranial skeleton in French, compared their prevalence
with other populations, and assessed the association of these
traits with sex, age, side and each other. These findings provide
more knowledge about traits and how to use them in forensic
anthropology, by selecting the most informative traits. The
thorax is the most X-rayed part of the skeleton in medical
facilities, so the possibility of obtaining ante-mortem data for
Fig. 7. Os acromiale of the right scapula (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
Fig. 8. Manubrium foramen (white arrow). (A) DICOM image of CT scan (B) 3D volume rendering done with Avizo1.
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
134.e6
than 50 years of age. Therefore, in cases when ante-mortem
imaging was undertaken in an individual before the age of 50, the
trait may be absent, precluding the comparison. For future
research, it will be important to assess the association of more
discrete traits with age in order to better understand the
limitations of comparing ante- and post-mortem data. The
knowledge of the occurrence and morphology of discrete traits
also helps us to distinguish these normal skeletal variants from
pathological or traumatic conditions. For example, the manubrium
foramen, though extremely rare, should not be confused with a
gunshot wound.
4.1. Frequency of discrete traits
Fig. 9. Pubic spine on both pubic branches (white arrow). (A) DICOM image of CT
scan (B) 3D volume rendering done with Avizo1.
comparisons with post-mortem findings is high for this anatomical region
To be considered ‘‘a good discrete trait’’ for identification, an
anatomical variation has to be easily visible both on imaging and
dry bones and have a frequency lower than 10% [20]. Furthermore,
if it is related to sex or laterality, it gives additional information to
complete and refine the biological profile of the individual.
It is important to know which discrete traits are associated with
age in a population to narrow down the age range for identification
purposes. The process of identification is based on ante- and postmortem data comparison, therefore with age-related traits the age
when comparative ante-mortem imaging was obtained with
relation to age at death is of great importance. In our study, the
majority of individuals with a suprascapular foramen were more
The thoracic area rarely display spina bifida occulta (SBO); for all
thoracic vertebrae combined, Saunders [2] found a frequency of
1.7% in the American population. In this study we observed SBO
only on the first and second thoracic vertebrae with a frequency of
0.4% and 0.2% respectively.
The lumbar area is the most affected region after the sacrum
[21]. SBO was present only on the fifth vertebra with a frequency of
1.6%. Saluja [22] has found only one case of SBO at the same level in
a population from London.
In the sacral area, we took into account only the sacral hiatus
(SBO involving all sacral vertebrae), since the frequency of
occurrence of only one or two sacral vertebrae affected was found
to be higher than 10%, being not in agreement with the definition of
a rare trait. In our population, the frequency of the presence of
sacral hiatus was 1.8% compared to 1.1–2.1% in other populations
(North American and English populations) [2,22].
SBO was present only for the first vertebra in the cervical region,
with a frequency of 4.6%. In other populations (North American and
Polish), the frequency observed was lower important, ranging from
0.3 to 3.2% [2,23].
In our sample, SBO was most commonly present on the first
cervical vertebrae, followed by the fifth lumbar vertebrae and the
thoracic region. This is in contrast to the findings presented by
Albano [21] for US soldiers, in whom SBO occurred most commonly
on the first sacral vertebrae, followed by the fifth lumbar vertebrae,
and the cervical region [21].
Butterfly vertebra was found only in one case on the fourth
lumbar vertebra (1%). This location is the best known and most
commonly described in the literature [25–27]. Individual cases of
butterfly vertebrae were reported on dry bones or imagery, but no
population frequency was given, except for the Inuit population
who showed a frequency of 8% [24]. This relatively high frequency
seems to reflect the influence of genetic factors, reinforced through
isolation.
The supraclavicular nerve foramen can be observed in juvenile
individuals and fetuses, so a hypothesis of genetic origin can be
advanced. No study has yet confirmed this hypothesis [28]. In our
Table 4
Results of Cramers’V for association between traits.
SBO
SBO of sacrum
Butterfly vertebra
Suprascapular nerve foramen
Coracoclavicular joint
Os acromiale
Suprascapular foramen
Manubrium foramen
Pubic spine
0.022*
0.033
0.021
0.022
0.164
0.047
0.012
0.043
SBO = spina bifida occulta.
*
V of Cramer. comprised between 0 and 1.
SBO of
sacrum
Butterfly
vertebra
Suprascapular
nerve foramen
Coracoclavicular
joint
0.016
0
0.0
0.023
0.023
0.006
0.064
0.009
0.01
0.022
0.021
0.005
0.021
0
0.015
0.013
0.003
0.014
0.016
0.143
0.004
0.012
Os acromiale
0.047
0
0.031
Suprascapular
foramen
Manubrium
foramen
0.008
0.191
0.008
E. Verna et al. / Forensic Science International 253 (2015) 134.e1–134.e7
sample, we found a frequency of 0.6%. Other populations (North
American and European) [2,8,28] reported a frequency lower than
3% for supraclavicular nerve foramen.
The scapular facet of the coracoclavicular was not found in our
population as in the literature [28], and only the clavicular facet is
observed [28]. Frequencies found in literature extend from 0.6 to
21% [12] and our findings are similar.
The prevalence of the os acromiale varied from 0.3 to 20%
[2,11,29] in the North American populations. In our population the
frequency was 3%.
The prevalence of suprascapular foramen in other populations
(North American, European, and Asian) varied from 3.7 to 13.6%
[15,30] whereas our population showed a slightly lower prevalence of 2.8%.
It is very rare to observe a manubrium foramen: only one case
has been found in the literature [17]. We also found only one case
was found in our population.
For the pubic spine, no previous studies have reported a
frequency of occurrence. We found a frequency of 3% in our sample.
4.2. Association of discrete traits with sex, laterality and age
Out of the eight discrete traits selected and presented here,
some are related to sex, laterality and/or age.
Wysocki [23] found an association with the female sex for SBO
in a Polish population (on dry bones and radiographies). In our
population, no statistically significant difference was observed.
For the supraclavicular nerve foramen only the left clavicle was
affected in our population and Voisin [28] reported the same
tendency in his review. We observed this trait only in males. To our
knowledge, no previous publications presented sex-specific
results.
For the coracoclavicular joint we didn’t observe any difference
according to age, contrary to Gumina [12] who studied dry bones
from an Italian population, and Saunders [2] who studied dry
bones from a North American population. Both of them showed the
frequency of this trait increased with age.
Individuals aged older than 50 years showed the highest
frequency of the supraclavicular foramen in our population. This is
in contrary to the literature [2], which shows no difference
according to age.
The pubic spine was more frequently present in females (8.4%).
We found no other studies presenting results for this trait.
4.3. Association between selected traits
In general, no strong association between traits was found. It
may be useful to assess also associations with other discrete traits
present on the upper skeleton (those visible on CT-scans).
According to the literature, spondylolysis and SBO can be
associated, particularly on the fifth lumbar vertebra [31]. So in a
future study, we plan to test this association in our population,
including associations between other postcranial discrete traits
not only the eight presented here.
5. Conclusion
In conclusion, the eight postcranial discrete traits selected and
studied here had a frequency lower than 5% in our population and
were easily visible both on imaging and on dry bones. Some traits
were significantly associated with sex, such as the supraclavicular
nerve foramen, the coracoclavicular joint and the pubic spine. One
trait was significantly associated with age: the suprascapular
foramen.
The study and knowledge of skeletal discrete traits and their
possible association with other skeletal and physiological features
134.e7
provides a useful addition for the construction of the biological
profile of individuals for the purpose of identification. Ultimately,
better knowledge of human skeletal variations will help anthropologists come closer to a positive identification.
References
[1] E. Cunha, Pathology as a factor of personal identity in forensic anthropology, in: A.
Schmitt, E. Cunha, J. Pineiro (Eds.), Forensic Anthropology and Medicine: Complementary Sciences from Recovery to Cause of Death, Humana Press Inc, Totowa,
NJ, 2006, pp. 33–358.
[2] S.R. Saunders, The Development and Distribution of Discontinuous Morphological
Variation of the Human Infracranial Skeleton, University of Toronto, Toronto,
Canada, 1978p. 534p (Doctoral Thesis).
[3] I. Gemmerich Pfister, Création d’une collection anthropologique de référence et
application des caractères discrets dans le cas de généalogies connues, University
Department of Anthropology and Ecology, Geneva, Switzerland, 1999 (Doctoral
Thesis).
[4] S.R. Saunders, Non-metric skeletal variation, in: M.Y. Iscan, K.A.R. Kennedy
(Eds.), Reconstruction of Life from the Skeleton, Wiley-Liss, New York, NY,
1989, pp. 95–108.
[5] G. Hauser, G.F. De Stefano, Epigenetic variants of the human skull, in: E. Schweizerbart’sche Verlagsbuchhandlung, 1989 (Nägele u. Obermiller).
[6] E. Crubézy, N. Telmon, A. Sevin, J. Picard, D. Rougé, G. Larrouy, et al., Microévolution d’une population historique: Etude des caractères discrets d’une population Missiminia (Soudan, III-IVe siècle), Bull. Mém. Soc. Anthropol. Paris 11
(1999) 1–213.
[7] E. Barnes, Atlas of Developmental Field Anomalies of the Human Skeleton: A
Paleopathology Perspective, second ed., John Wiley & Sons Inc, Hokoben, NJ,
2012p. 232 p.
[8] G. Olivier, Anthropologie de la clavicule, Bull. Mém. Soc. Anthropol. Paris 2 (1951)
121–157.
[9] N. Haramati, R.A. Cook, B. Raphael, T.S. McNamara, R.B. Staron, F. Feldman, Coracoclavicular joint—normal variant in humans—a radiographic demonstration in the
human and nonhuman primate, Skeletal Radiol. 23 (1994) 117–119.
[10] L. Capasso, K.A.R. Kennedy, C.A. Wilczak, Atlas of Occupational Markers on Human
Remains, Edigraphital SpA, Teramo, Italy, 1999.
[11] M. Finnegan, Nonmetric variation of infra-cranial skeleton, J. Anat. 125 (1978)
23–37.
[12] S. Gumina, M. Salvatore, P. De Santis, L. Orsina, F. Postacchini, Coracoclavicular
joint: osteologic study of 1020 human clavicles, J. Anat. 201 (2002) 513–519.
[13] D. Case, S. Burnett, T. Nielsen, Os acromiale: population differences and their
etiological significance, HOMO—J. Comp. Human Biol. 57 (2005) 1–18.
[14] L. Scheuer, S. Black, The Juvenile Skeleton, Academic Press, London, 2004.
[15] K. Natsis, T. Totlis, P. Tsikaras, H.J. Appell, P. Skandalakis, J. Koebke, Proposal for
classification of the suprascapular notch: a study on 423 dried scapulas, Clin.
Anat. 20 (2007) 135–139.
[16] D. Gray, Variations in the human scapulae, Am. J. Phys. Anthropol. 29 (1942)
57–72.
[17] P.D. Cooper, J.H. Stewart, W.F. McCormick, Development and morphology of the
sternal foramen, Am. J. Forensic Med. Pathol. 9 (1988) 342–347.
[18] D.A. Donlon, The value of infracranial nonmetric variation in studies of modern
Homo sapiens: an Australian focus, Am. J. Phys. Anthropol. 113 (2000) 349–368.
[19] H. Cramér, Mathematical Methods of Statistics, Princeton University Press,
United States, 1946.
[20] E. Verna, P. Adalian, K. Chaumoitre, Y. Ardagna, G. Leonetti, M.D. Piercecchi-Marti,
L’intérêt des caractères discrets en identification médicolégale, La Revue De
Médecine Légale 4 (2013) 8–15.
[21] J.P. Albano, S.G. Shannon, N.M. Alem, K.T. Mason, Injury risk for research subjects
with spina bifida occulta in a repeated impact study: a case review, Aviat. Space.
Environ. Med. 67 (1996) 767–769.
[22] P.G. Saluja, The incidence of spina bifida occulta in a historic and a modern London
population, J. Anat. 158 (1988) 91–93.
[23] J. Wysocki, M. Bubrowski, J. Reymond, J. Kwiatkowski, Anatomical variants of the
cervical vertebrae and the first thoracic vertebra in man, Folia Morphol. 62 (2003)
357–363.
[24] C.F. Merbs, Sagittal clefting of the body and other vertebral development errors in
Canadian Inuit skeletons, Am. J. Phys. Anthropol. 123 (2004) 236–249.
[25] T. Anderson, A medieval example of a sagittal cleft or ‘butterfly’ vertebra, Int. J.
Osteoarchaeol. 13 (2003) 352–357.
[26] P. Brasili, B. Bonfiglioli, A.R. Ventrella, A case of ‘butterfly’ vertebra from Sardinia,
Int. J. Osteoarchaeol. 12 (2012) 415–419.
[27] B. Sonel, P. Yalcin, E.A. Ozturk, I. Bokesoy, Butterfly vertebra: a case report, Clin.
Imaging 25 (2001) 206–208.
[28] J.L. Voisin, Les caractères discrets des membres supérieurs: un essai de synthèse
des données, Bull. Mém. Soc. Anthropol. Paris 24 (2012) 107–130.
[29] K.A.R. Kennedy, Skeletal markers of occupational stress, in: M.Y. Iscan, K.A.R.
Kennedy (Eds.), Reconstruction of Life from the Skeleton, Wiley-Liss, New York,
NY, 1989, pp. 129–161.
[30] M. Polguj, K. Jedrzejewski, M. Podgorski, M. Topol, Morphometric study of the
suprascapular notch: proposal of classification, Surg. Radiol. Anat. 33 (2011)
781–787.
[31] T. Waldron, A case-referent study of spondylolysis and spina bifida and transitional vertebrae in human skeletal remains, Int. J. Osteoarcheol. 3 (1993) 55–57.