Content uploaded by Muhammad A Spocter
Author content
All content in this area was uploaded by Muhammad A Spocter on Feb 08, 2022
Content may be subject to copyright.
uncorrected proof
Anthropol. Anz. Article
J. Biol. Clin. Anthropol.
Published online February 2022
© 2022 E. Schweizerbart’sche Verlagsbuchhandlung, 70176 Stuttgart, Germany www.schweizerbart.de
DOI: 10.1127/anthranz/2022/1560 0003-5548/2022/1560 $ 0.00
Sex estimation from dimensions of the base of the skull in
Black South Africans
Manisha R. Dayal1,*, Brendon K. Billings2, Desiré Brits2, Abdallah Abdallah1,
Muhammad A. Spocter2,3, Mubarak A. Bidmos4
1 School of Science, Western Sydney University, Penrith, New South Wales, Australia
2 Human Variation and Identication Research Unit (HVIRU), School of Anatomical Sciences, Faculty of Health Sciences,
University of the Witwatersrand, Johannesburg, South Africa
3 Department of Anatomy, Des Moines University, Iowa, USA
4 College of Medicine, Qatar University, QU Health, Doha, Qatar
* Corresponding author: m.dayal@westernsydney.edu.au
With 2 gures and 6 tables
Abstract: The pelvis and the skull are the two most utilised skeletal elements to estimate sex from skeletonised remains
due to their sexually dimorphic traits. However, as increasingly more fragmented remains have been presented for analyses,
other bones and their fragments have now been subjected to analyses for sex estimation. In the skull particularly, the base
has shown to survive harsh conditions. In this study the foramen magnum region was explored in Black South Africans to
estimate sex during forensic analyses. Seven measurements of the foramen magnum and surrounding areas were measured
in 120 male and female crania and subjected to discriminant function analyses. The average accuracies for the stepwise
discriminant functions ranged from 60–71% whilst the average accuracies for the direct discriminant functions ranged from
63–69%. The average accuracies obtained in this study are similar to other studies performed using the foramen magnum.
However, these average accuracies are much lower than other skeletal elements that have been used for sex estimation in
South Africans. Thus, the equations in this study should be used with caution and only in the absence of more accurate ele-
ments. The cranial base has always shown to have a low to moderate expression of sexual dimorphism. The cranial base of
Black South Africans is no dierent.
Keywords: Sex estimation; foramen magnum; population standards; Black South Africans; forensic anthropology popula-
tion data
Introduction
One of the key components of the scope of practice for
forensic anthropologists is the estimation of sex. This can be
performed by visual assessment of sexually dimorphic mor-
phological traits of the skull and postcranial skeleton. The
main consequence of sex estimation from skeletal remains is
the reduction of the number of possible matches of individu-
als by fty percent (Loth & İşcan 2000). However, forensic
anthropologists do not always have the opportunity to evalu-
ate complete skeletons in forensic cases and as such they
rely on available skeletal elements for the development of
the biological prole of the victim(s). Consequently, there
have been increased eorts in the past few years on the part
of researchers in the eld of forensic anthropology to derive
population-specic equations for the estimation of sex from
various bones of the human skeleton in dierent parts of the
world (İşcan & Steyn 2013). This is due to the presence of
signicant variation in the expression of sexual dimorphism
between population groups, which is related to observed dif-
ferences in both human growth and development (Ubelaker
& DeGaglia 2017).
The pelvic bone remains the most sexually dimorphic
bone (Bruzek 2002; Patriquin et al. 2003; Steyn & İşcan 2008;
Gómez-Valdés et al. 2011) because of its design for parturi-
tion (İşcan & Steyn 2013). In the absence of the pelvic bone,
sex estimation can be performed using the skull and other
bones of the postcranial skeleton. The skull is considered by
some authors (Zaafrane et al. 2018) as the second-best sex
estimator compared to the pelvic bone in which an average
accuracy of 90% was achieved. However, other researchers
(Spradley & Jantz 2011) presented contrary results to this in
which they observed that long bones are better estimators of
sex compared to the skull. This observation notwithstanding,
eschweizerbart_xxx
uncorrected proof
2 M. R. Dayal et al.
there has been a plethora of studies on the estimation of sex
from measurements of the intact skull in dierent parts of
the world (Kranioti et al. 2008; Franklin et al. 2013; Ogawa
et al. 2013; Krüger et al. 2015; Mahakkanukrauh et al. 2015;
Ekizoglu et al. 2016; Small et al. 2018) with average accura-
cies in correct sex classication as high as 98% (Abdel Fatah
et al. 2014). In addition, morphological features of the skull
such as the brow ridge, shape and size of the mastoid pro-
cess, shape of the mandible, presence of occipital hook, and
size of the foramen magnum have also been used to augment
the results of metrical assessment of sex from the intact skull
(İşcan & Steyn 2013).
The skull may be recovered in forensic cases in incom-
plete or fragmentary states due to the fragility of the facial
skeleton (Gapert et al. 2009). The fragmentation may result
from trauma, explosion, crashes, and attempts to conceal
identication of the victim (Gapert et al. 2009). It may also
be a consequence of natural taphonomic processes (Ubelaker
1997). The base of the skull, however, has proved more likely
to withstand destruction, a fact attributed to its fortuitous
position and relative thickness in relation to the rest of the
skull (Gapert et al. 2009). This, therefore, makes this region
a valuable candidate for the investigation of sex estimation
in a forensic setting. The base of the skull consists of three
depressions or fossae which are placed at relatively lower
levels. These include the anterior, middle and posterior cra-
nial fossae. The largest and the deepest is known as the pos-
terior fossa and it consists of the basilar part of the sphenoid
bone, the petrous and mastoid parts of the temporal bone, the
parietal bones and the occipital bone. Located at the centre
of the posterior cranial fossa is the foramen magnum, which
serves as the point of continuation of the medulla oblongata
as the spinal cord. It also serves as a passage for the vertebral
arteries, anterior spinal artery, and spinal accessory nerve
(Drake et al. 2015). The basilar portion of the occipital bone
joins with the body of the sphenoid bone in the young skull
by a synchondrosis anterior to the foramen magnum which
is also a valuable age estimation site.
It is not surprising that the foramen magnum is consid-
ered to be an important anatomical feature of the base of
the skull, which over the years has attracted the attention of
forensic anthropologists with the view to assess its potential
in the estimation of sex as part of the other metrical features
of the base of the skull. Keen (1950) conducted one of the
earliest studies on sexual dimorphism of the base of the skull
and reported that the length of the foramen magnum from a
sample of South African ‘Coloured’ (Mixed-ancestry) skulls
can provide a 85% correct sex classication rate. Holland
(1986) subjected nine measurements of the base of the skull
to regression analysis, which produced models with predic-
tion accuracies that ranged from 70% to 90% which again
emphasised the utility of this region and of the foramen mag-
num in particular for sex estimation. Other studies have also
shown that measurements of the base of the skull in general
and the foramen magnum in particular are useful in the esti-
mation of sex with moderate to high average accuracies in
correct sex classication (Uysal et al. 2005; Raghavendra
Babu et al. 2012; Amores-Ampuero 2017; Toneva et al.
2018; González-Colmenares et al. 2019).
However, other researchers (Gapert et al. 2009; Uthman
et al. 2012; Edwards et al. 2013; Saini et al. 2014; Tambawala
et al. 2016; Madadin et al. 2017; Lopez-Capp et al. 2018)
have reported contrary ndings in which dimensions of the
foramen magnum presented with low to average accuracies.
For example, in 2009, Gapert et al. (2009) using both a dis-
criminant function and regression approach on the foramen
magnum dimensions of a historical British sample, obtained
percentage accuracies in the range of 66 and 71% thereby
questioning the forensic utility of this feature of the base of
the skull in the assignment of sex. Since discriminant func-
tion and logistic regression equations are population-spe-
cic, it is the aim of this paper to explore the degree of sexual
dimorphism in the cranial base of Black South Africans. In
addition, our specic aim is to derive useable discriminant
function equations based on linear dimensions of the fora-
men magnum and other features of the base of the skull of
Black South Africans for the accurate estimation of sex.
Material and methods
The study sample consisted of 120 (59 males, 61 females) ran-
domly selected skulls of Black South Africans whose skele-
tal remains are housed in the Raymond A. Dart Collection of
Modern Human Skeletons, School of Anatomical Sciences,
University of the Witwatersrand, Johannesburg, South
Africa. This collection was established in 1923 by Professor
Raymond Arthur Dart and is mainly derived from cadavers
that were used as part of the dissection programme in Human
Anatomy courses by medical and allied health sciences stu-
dents (Dayal et al. 2009). The age at death of individuals
whose crania were selected for the present study ranged from
18 to 90 years old. This distribution is illustrated in Table 1.
The mean age at death for males and females were 48.3 and
44.5 years respectively. These individuals died between the
years 1927–1992. South Africa has four major self-iden-
tied population groupings which are classied as South
African Black or African (81.2%), South African Coloured
(8.7%), South African White (7.6%) and South African
Indian or Asian (2.5%) according to the most recent popu-
lation estimates (Statistics South Africa, 2020). The South
African Black population consists of nine dierent indige-
nous ethnic groupings which include Zulu, Xhosa, Ndebele,
Swazi, Tswana, Shangaan (Tsonga), Pedi (Northern Sotho),
Southern Sotho and Venda (Garenne 2016). Hence the cur-
rent study had a focus on the largest South African popula-
tion group, namely the South African Black population that
was mainly from the Zulu ethnic group. Ethical clearance
was granted under the School of Anatomical Sciences ethics
waiver: W-CJ-140604-1.
eschweizerbart_xxx
uncorrected proof
Sex estimation from dimensions of the base of the skull 3
Skulls with signs of fractures, severe damage, congenital
abnormalities and other forms of pathologies were excluded
from this study. All osteological measurements as described
below were taken with the use of digital sliding calipers
(Mitutoyo Corporation, calibrated to 0.01 mm) based on
the denitions of Gapert et al. (2013) and Singh & Talwar
(2013):
1. Maximum length of the foramen magnum (FML):
Measured from the basion, which is the most anterior
aspect of the foramen magnum, to the most posterior
margin of the foramen magnum (Fig. 1).
2. Maximum width of foramen magnum (FMW):
Measured between the most lateral margins of the fora-
men magnum, perpendicular to the length of the fora-
men magnum (Fig. 1).
3. Maximum length of condyle (MLC): This is the
maximum length measured along the long axis of the
condylar facet, between the most anterior and posterior
projecting margins of both the left and right occipital
condyles (Fig. 1).
4. Maximum width of condyle (MWC): This is the
widest part of both condylar facets and was measured
perpendicular to the maximum length of the condyle
(MLC) (Fig. 1).
5. Maximum bicondylar breadth (BCB): This is the lin-
ear distance measured between the most laterally pro-
jecting margins of the articular facets of the occipital
condyles (Fig. 2).
6. Minimum distance between condyles (MnD): This
measurement was slightly modied and was taken as
the narrowest distance on the antero-medial aspect of
the occipital condyles (Fig. 2).
7. Maximum interior distance between condyles
(MxID): This measurement was also slightly modied
and was measured as the maximum distance between
the medial aspects of the condyles, approximately at
the intersection of the occipital condyle and the fora-
men magnum (Fig. 2).
Two of the authors (DB and BB) collected the data and
took all measurements to the nearest millimetre. The bilat-
eral variables (MLC and MWC) were measured on both the
left and right sides and kept as two separate measurements.
An assessment of intra-observer and inter-observer error of
repeatability was performed on data collected from twenty
randomly selected skulls. All measurements as described
above were taken twice (test and retest) about 48 hours apart,
which is an adequate time interval for the test measurement
to have been forgotten by all investigators. Subsequently,
intra- and inter-observer variability of test and retest mea-
surements was assessed using a Lin’s concordance correla-
tion coecient of reproducibility (Pc) (Lin 1989). This index
provides objective evidence of the accuracy of the method
used in the acquisition of data for the present study. The for-
mula of Pc is as shown below:
Pc = 1 – [1/n Σ (y1 – y2)2 / (sdy12 + sdy22 + (my1 – my2)2],
where n = sample size, y1 = test measurement, y2 = retest
measurement, sdy1 = standard deviation of the test measure-
ments, sdy2 = standard deviation of the retest measurements,
my1 = mean of the test measurements and my2 = mean of the
retest measurements.
All the data were analysed using the IBM SPSS (version
24) program from which descriptive statistics, including
means and standard deviations, were obtained for each of the
measurements. The Shapiro-Wilk test of normality was per-
formed together with a Mann-Whitney U test. Sexual dimor-
phism was assessed using a student t-test. Subsequently,
stepwise and direct discriminant function analyses were per-
formed on the data after the establishment of the existence
of a signicant dierence (p ≤ 0.05) between the male and
female mean measurements.
Results
The results for the repeatability of measured variables are
as shown in Table 2. Substantial to moderate reproducibility
(Pc > 0.9) (McBride (2005) was noted for all observers for
the following measurements: maximum length of the fora-
men magnum (FML), maximum width of foramen magnum
(FMW), maximum bicondylar breadth (BCB) and maximum
length of condyle (MLC). Similar results were also obtained
for the inter-observer error of reproducibility for the afore-
mentioned variables (Table 2). In addition, poor repeatabil-
ity (Pc = 0.70–0.89) was noted for the minimum distance
between condyles (MnD), maximum distance between
condyles (MxID) and width of condyle (MWC). However,
maximum interior distance between condyles (MxID) and
width of condyle (MWC) showed the least (Pc ≤ 0.69) intra-
observer and inter-observer error of repeatability (Table 2).
The descriptive statistics for all measured variables for
both sexes are as shown in Table 3. All male mean measure-
ments showed statistically signicant larger values at p < 0.05
compared with female mean measurements except MWC(l)
Table 1. Sample according to age.
Age (years) Males Females
18–28 11 7
29–38 19 27
39–48 22 14
49–58 16 9
59–68 13 12
69–78 68
79–90 5 2
eschweizerbart_xxx
uncorrected proof
4 M. R. Dayal et al.
Fig. 1. Illustration of foramen magnum (FML – maximum length of the foramen magnum and FMW – maximum width of foramen
magnum) and occipital condyle (MLC – maximum length of condyle and MWC – maximum width of condyle) measurements.
Fig. 2. Illustration of maximum bicondylar breadth (BCB), minimum distance between condyles (MnD) and maximum interior distance
between condyles (MxID) measurements.
eschweizerbart_xxx
uncorrected proof
Sex estimation from dimensions of the base of the skull 5
and MWC(r). These two measurements were subsequently
removed from further discriminant function analysis and are
therefore not a component of any of the derived functions.
In the stepwise analysis of all variables, two measure-
ments were selected namely FML and MnD (Function 1,
Table 4). A combination of these two variables can correctly
estimate sex with an average accuracy of 71%. A similar
analysis of length measurements selected FML and MLC(r)
with an average correct classication rate of 68% (Function
2, Table 4). The function with the lowest average correct
classication accuracy (60%) was obtained from the step-
wise analysis of all breadth measurements in which FMW
was the only selected measurement (Function 3, Table 4).
Table 5 shows the functions, arranged in descend-
ing order of average correct classication rate, that were
obtained from direct discriminant function analyses. The
function with the highest average accuracy in correct sex
classication (69%) was obtained from the direct analysis
of all variables (Function 1, Table 5). Individual measure-
ments with the highest average accuracies in correct sex
classication were FML (67%: Function 4, Table 5), MnD
(65%: Function 5, Table 5) and MLC(r) (63%: Function 7,
Table 5). A combination of these three variables yielded an
average accuracy of 67% in correct sex estimation (Function
2, Table 5). Function 6 of Table 5 was obtained from direct
analysis of the length and width of the foramen magnum
with an average accuracy in correct sex estimation of 64%.
The average accuracy in correct sex classication for most
of the presented functions for both stepwise and direct analy-
ses remained unchanged after the cross-validation procedure
using drop-one out classication. The drop in average accu-
racy for a few of the functions ranged between 0 and 5%
(Tables 4 and 5) which indicates the validity of the presented
functions.
Discussion
Forensic anthropologists across the globe are increasingly
faced with the challenges of human identication from skel-
etal remains amidst a backdrop of ever-changing criminal
behaviour such as dismemberment of the remains of the
victim (Steyn & İşcan 1997; Di Nunno et al. 2006). This
is usually done in order to thwart the eorts of forensic
anthropologists through the concealment of the identity of
the victim through fragmentation and dispersal of the burnt
remains of the victim (Di Nunno et al. 2006). It is therefore
essential that sex should be estimated from fragmentary and
incomplete skeletal elements as this remains a critical step in
routine biological proling. Consequently, researchers in the
eld of forensic and physical anthropology have investigated
the possibility of other methods of human identication
Table 3. Descriptive statistics of measurements of foramen magnum (in mm).
Variables Males Females p-value
No Mean SD No Mean SD
FML 59 37.3 2.3 61 35.5 2.4 0.000
FMW 59 29.4 2.3 61 28.2 1.9 0.002
BCB 59 49.1 3.0 61 47.9 3.1 0.036
MnD 59 15.3 2.5 61 13.5 2.8 0.000
MxID 59 31.4 2.1 61 30.4 2.2 0.006
MLC (L) 59 23.4 2.6 61 22.5 2.1 0.044
MLC (R) 59 23.1 2.2 61 21.9 2.0 0.002
MWC (L) 59 13.8 1.6 61 14.2 2.3 0.264*
MWC (R) 59 13.5 1.5 61 13.2 2.1 0.528*
* p > 0.05
Table 2. Lin’s concordance correlation of reproducibility.
Measurement Intra-observer 1 Intra-observer 2 Inter-observers
FML 0.992 0.995 0.984
FMW 0.989 0.995 0.987
BCB 0.971 0.912 0.951
MnD 0.761 0.538 0.828
MxID 0.735 0.495 0.735
MLC (L) 0.974 0.940 0.969
MWC (L) 0.758 0.758 0.543
eschweizerbart_xxx
uncorrected proof
6 M. R. Dayal et al.
Table 5. Direct discriminant function coecients, constants and classication rates.
Function Variables Unstandardized
coecient Centroids Sectioning
point
Average Accuracy (%)
Original
classication
Cross
validation
1
BCB –0.036 F = –0.517 0.009 69 64
MnD 0.18 M = 0.534
MxID 0.004
FML 0.197
FMW 0.095
MLC (L) –0.075
MLC (R) 0.302
constant –15.907
2
MLC (R) 0.226 F = –0.504 0.009 67 67
MnD 0.185 M = 0.522
FML 0.237
constant –16.364
3
BCB 0.026 F = –0.364 0.007 67 63
MnD 0.283 M = 0.377
MxID 0.192
constant –11.256
4FML 0.428 F = –0.372 0.007 67 67
constant –15.556 M = 0.385
5MnD 0.374 F = –0.326 0.006 65 65
constant –5.379 M = 0.337
6
FML 0.349 F = –0.382 0.007 64 63
FMW 0.132 M = 0.395
constant –16.479
7MLC (R) 0.475 F = –0.287 0.005 63 63
constant –10.666 M = 0.297
Table 4. Stepwise discriminant function coecients, constants and classication rates.
Function Variables Unstandardized
coecient Centroids Sectioning
point
Average Accuracy (%)
Original
classication
Cross
validation
1
FML 0.300 F = –0.445 0.008 71 70
MnD 0.211 M = 0.460
constant –13.940
2
FML 0.328 M = 0.457 0.008 68 68
MLC (R) 0.259 F = –0.442
constant –17.760
3FMW 0.475 F = –0.283 0.005 60 60
constant –13.658 M = 0.293
eschweizerbart_xxx
uncorrected proof
Sex estimation from dimensions of the base of the skull 7
from fragmentary and incomplete skeletal elements such as
the skull with reasonable degrees of accuracy (Wescott &
Moore-Jansen 2001; Uysal et al. 2005; Raghavendra Babu
et al. 2012; Uthman et al. 2012; Madadin et al. 2017; Toneva
et al. 2018; González-Colmenares et al. 2019).
Several studies on sexual dimorphism of the cranial base
have included the foramen magnum, its linear or circumfer-
ential dimensions have been subjected to various statistical
analyses including and not limited to logistic and discriminant
function analyses. In the current study, seven linear measure-
ments at the cranial base including the maximum width and
length of the foramen magnum were taken. Most of the mea-
surements showed moderate to high intra- and inter-observer
repeatability (Table 2) whereas a low Pc value was noted for
the maximum width of condyle (MWC). Wescott & Moore-
Jansen (2001) also found it dicult to measure this param-
eter of the cranial base as they reported high intra-observer
error (5.4%). They (Wescott & Moore-Jansen 2001) noted
that the long interval (about two years) between their test
and retest measurements could have been responsible for the
low degree of repeatability that was observed. However, the
interval between both measurements in the current study was
48 hours and could not explain the low Pc observed. Some
possible reasons for the low degree of repeatability observed
in MWC in the current study can be attributed to diculties
encountered in distinguishing between the occipital condyle
margins, margins of the articular facets of the occipital con-
dyles and also the margins of the foramen magnum. Due to
the lack of signicant dierence between males and females,
this measurement was however not used in the development
of any of the discriminant functions.
In this study, all male mean measurements were signi-
cantly higher than female mean measurements except for
maximum width of both right and left condyles. We attribute
the apparent lack of sexual dimorphism of these measure-
ments to possible diculty in repeating these measurements
as shown by the low intra- and inter-observer concordance
correlation of reproducibility (Table 2). Consequently, these
two measurements (width of right and left condyles) were
eliminated from subsequent discriminant function analyses
performed in this study. The observed sexual dimorphism
displayed by most of the measurements of the base of the
skull of black South Africans is in support of ndings from
previous studies (Franklin et al. 2005; Dayal et al. 2008;
Gapert et al. 2009; Raghavendra Babu et al. 2012; Franklin
et al. 2013; Ogawa et al. 2013; Abdel Fatah et al. 2014; Saini
et al. 2014; Mahakkanukrauh et al. 2015; Ekizoglu et al.
2016; Amores-Ampuero 2017; González-Colmenares et al.
2019). The most consistently measured variables in all these
studies are the length (FML) and breadth (FMW) of the fora-
men magnum which are also referred to as A-P and trans-
verse diameters. A comparison of the means and the variation
in the degree of sexual dimorphism of these measurements
is an indication of the generally held view regarding the
existence of osteometric variation in dierent population
groups. The mean values for FML and FMW for both sexes
fall within the average for other population groups (Table 6).
Gruber et al. (2009) and Toneva et al. (2018) observed that
even though osteometric dierences exist in the dimensions
of the foramen magnum, the magnitude of the dierences is
not statistically signicant enough to be of forensic utility
in the estimation of population anity as shown in Table 6.
The highest average accuracy of 71% in correct sex clas-
sication from the present study was obtained from the step-
wise analysis of all measured variables (Function 1, Table 4).
The other functions present with a range of average accura-
cies of between 63% and 69% (Tables 4 and 5). This nd-
ing is in agreement with what has been reported for other
population groups in which the highest average classication
rate was around 70% (Gapert et al. 2009; Singh & Talwar
2013; Saini et al. 2014; Madadin et al. 2017). While some
authors reported lower average accuracies in correct sex
classication (Wescott & Moore-Jansen 2001; Uthman et al.
2012; Edwards et al. 2013; Tambawala et al. 2016; Lopez-
Capp et al. 2018), others obtained slightly higher classi-
cation rates but not exceeding 76% (Amores-Ampuero
2017; Chovalopoulou & Bertsatos 2017; Seifert et al. 2017;
Toneva et al. 2018). A few other studies presented what some
researchers considered to be a reasonably high average accu-
racy of between 80 and 90% (Holland 1986; Uysal et al.
2005; Raghavendra Babu et al. 2012; González-Colmenares
et al. 2019; Tomaszewska et al. 2020). It is worth mention-
ing that there is still no general consensus amongst practi-
tioners of forensic anthropology on an acceptable level of
accuracy for sex estimation using osteometric dimensions
(Toneva et al. 2018). The general trend in Table 6 shows that
the majority of studies reported average accuracies of less
than 80% for discriminant function equations while none of
these studies reported an average accuracy of 95%, which is
regarded as the cut o point for variables that are considered
as primary sex estimators (Toneva et al. 2018). However,
common practice in forensic anthropology is to accept accu-
racies that are 50% better than chance (i.e. 75% accuracy
for distinguishing between two groups such as in the case of
sex) (Liebenberg et al. 2019). The discriminant function and
logistic regression equations utilizing measurements of the
foramen magnum and the base of the skull are not of great
forensic utility and should be used with caution only in the
absence of intact skull or other skeletal elements with vari-
ables which are primary sex estimators. It can therefore be
inferred that dimensions of the foramen magnum and the rest
of the cranial base cannot be used with high certainty in a
medico legal context for the estimation of sex due to the lack
of high degree of sexual dimorphism on this part of the skull.
A number of hypotheses have been postulated to explain
this apparent lack of high degree of sexual dimorphism on
the base of the skull. An irrefutable fact is that the most
useful measurable traits or features of bones that provide
high classication rates in correct sex estimation are usu-
ally the ones that show the existence of functional dier-
eschweizerbart_xxx
uncorrected proof
8 M. R. Dayal et al.
Table 6. Comparison of dimensions of the foramen magnum and average accuracies in correct sex classication from previous studies and present study.
Study Year Skeletal element Population
group Data source
Measurements (in mm)
Highest
average
accuracy
FML FMW
Males Females Males Females
Mean SD Mean SD Mean SD Mean SD
Holland 1986 Base of the skull American Whites Skeletal collection:
Terry 38.0 2.5 34.9 2.7 31.8 2.0 30.4 2.3 90.0
Holland 1986 Base of the skull American Blacks Skeletal collection:
Terry 37.1 2.2 34.6 3.1 31.1 1.8 28.4 2.6 90.0
Tomaszewska et al. 2020 Foramen magnum Polish Skeletal collection 37.4 2.2 32.1 1.5 33.3 1.7 29.9 1.6 90.0
Raghavendra Babu
et al. 2012 Foramen magnum Indian Skeletal collection 35.7 1.8 28.9 1.6 32.6 2.1 28.2 1.8 88.0
González-
Colmenares et al. 2019 Base of the skull Colombian Radiographs 41.0 2.2 36.9 2.1 35.5 1.9 33.6 2.2 87.9
Uysal et al. 2005 Foramen magnum Turkish CT scans 37.1 1.9 34.9 2.6 30.8 2.0 28.9 2.4 81.0
Wescott &
Moore-Jansen 2001 Foramen magnum American Whites Dry skull 36.7 2.5 34.6 2.3 31.6 2.4 29.8 2.0 76.4
Amores-Ampuero 2017 Base of the skull Spanish Skeletal collection 35.9 2.6 33.7 2.7 30.4 2.3 28.4 2.2 75.7
Toneva et al. 2018 Foramen magnum Bulgaria CT scans 36.6 2.9 35.2 2.3 31.5 2.4 29.3 2.1 74.9
Chovalopoulou &
Bertsatos 2017 Foramen magnum Greeks Skeletal collection 36.7 2.5 34.9 2.4 32.5 2.7 30.6 2.2 74.0
Seifert et al. 2017 Foramen magnum French CT scans 36.8 2.8 35.0 2.2 31.0 2.1 29.3 2.3 73.4
Saini et al. 2014 Cranial base North Indians Forensic medical
collection 34.2 2.3 33.0 1.9 29.0 1.8 27.8 1.7 71.0
Madadin et al. 2017 Foramen magnum Sauid Arabians CT scans 37.2 2.2 36.1 2.7 31.7 2.3 30.6 2.5 71.0
Current study Foramen magnum South African
Blacks Skeletal collection 37.3 2.3 35.5 2.4 29.4 2.3 28.2 1.9 71.0
Gapert et al. 2009 Foramen magnum British Skeletal collection:
18th and 19th century 35.9 2.4 34.7 1.9 30.5 1.8 29.4 2.0 70.3
Singh & Talwar 2013 Foramen magnum Indians Skeletal collection 33.5 2.8 32.3 3.2 27.8 2.1 27.2 3.0 70.0
Uthman et al. 2012 Foramen magnum Iraq CT scans 34.9 2.0 32.9 2.0 29.5 2.5 27.3 2.2 69.3
Edwards et al. 2013 Foramen magnum Swiss CT scans 38.2 2.7 36.7 2.3 33.1 2.6 31.3 2.2 67.2
Tambawala et al. 2016 Foramen magnum Indian CBCT 36.2 2.3 34.5 2.4 30.8 2.5 29.2 2.5 66.4
Lopez-Capp et al. 2018 Foramen magnum Brazilian Skeletal collection 32.4 3.3 30.8 3.3 33.7 3.6 32.1 3.0 65.0
Wescott &
Moore-Jansen 2001 Foramen magnum American Blacks Dry skull 36.0 2.9 34.8 2.5 29.8 2.3 28.4 2.3 63.6
Murshed 2003 Foramen magnum Turkish CT scans 37.2 3.4 34.6 3.2 31.6 3.0 29.3 2.2
eschweizerbart_xxx
uncorrected proof
Sex estimation from dimensions of the base of the skull 9
ences between males and females (İşcan & Steyn 2013). The
female pelvis for example is designed for parturition and
consequently exhibits traits that are morphologically dier-
ent from the male pelvis. Measurements of these traits pro-
vide a more objective evaluation of the dierences between
both sexes. But such functionally distinct traits which reect
sexual dimorphism are not always present in all bones of
the human skeleton. The human skull is one of such bones
and any expression of sexual dimorphism from this part of
the skeleton is considered to be a reection of the quantita-
tive dierences between the sexes with respect to the size
and the shape of the bone (Steyn & İşcan 1998). The most
studied trait of the base of the skull is the foramen magnum
and it has been suggested that it is highly unlikely that it can
express the same level of sexual dimorphism as the entire
skull (Seifert et al. 2017).
Another factor that can be used in the explanation of the
low degree of sexual dimorphism reported for variables of
foramen magnum is developmental (Gapert et al. 2009).
Since the foramen magnum reaches its adult size early in
childhood, it is hypothesized that its nal size is not aected
to any extent by secondary sexual changes (Gapert et al.
2009). In addition, it has also been postulated that since the
foramen magnum only acts as a passage for structures from
and into the skull that it lacks muscle attachment which may
inuence its size in males due to increase muscle mass, the
degree of sexual dimorphism is therefore less compared to
other parts of the skull (Gapert et al. 2009).
Conclusion
Measurements of the foramen magnum in particular and the
cranial base in general are sexually dimorphic. The range
of average accuracies in correct sex classication from this
study (63–71%) is not reasonably high compared to other
skeletal elements that have been previously used for sex
estimation in South African populations. Therefore, the
equations presented in this study should only be used in the
absence of an intact skull, pelvis or long bones as supportive
evidence in forensic cases. This study is a further conrma-
tion of the low to moderate expression of sexual dimorphism
of the cranial base.
Acknowledgements: The authors would like to thank the School of
Anatomical Sciences for allowing access to the Raymond A. Dart
Collection of Modern Human Skeletons. More signicantly, we
would like to acknowledge the individuals who have generously
donated their bodily remains for teaching and research without
which this study would not have been possible. The authors would
like to acknowledge the former curator of the Raymond Dart, Mr.
Elijah Mofokeng for his help with sourcing a subset of the speci-
mens used in these analyses.
References
Abdel Fatah, E. E., Shirley, N. R., Jantz, R. L., & Mahfouz, M. R.
(2014). Improving sex estimation from crania using a novel
three-dimensional quantitative method. Journal of Forensic
Sciences, 59(3), 590–600. https://doi.org/10.1111/1556-4029.
12379 PMID:24502609
Amores-Ampuero, A. (2017). Sexual dimorphism in base of
skull. Anthropologischer Anzeiger, 74(1), 9–14. https://doi.org/
10.1127/anthranz/2017/0603 PMID:28362021
Bruzek, J. (2002). A method for visual determination of sex, using
the human hip bone. American Journal of Physical Anthro-
pology, 117(2), 157–168. https://doi.org/10.1002/ajpa.10012
PMID:11815949
Chovalopoulou, M., & Bertsatos, A. (2017). Estimating sex of
modern Greeks based on the foramen magnum region. Journal
of Anthropology, 2017, 1–7. https://doi.org/10.1155/2017/
9129801
Dayal, M. R., Spocter, M. A., & Bidmos, M. A. (2008). An assess-
ment of sex using the skull of black South Africans by discrimi-
nant function analysis. Homo, 59(3), 209–221. https://doi.org/
10.1016/j.jchb.2007.01.001 PMID:18439606
Dayal, M. R., Kegley, A. D. T., Štrkalj, G., Bidmos, M. A., &
Kuykendall, K. L. (2009). The history and composition of
the Raymond A. Dart Collection of Human Skeletons at the
University of the Witwatersrand, Johannesburg, South Africa.
American Journal of Physical Anthropology, 140(2), 324–335.
https://doi.org/10.1002/ajpa.21072 PMID:19382178
Di Nunno, N., Costantinides, F., Vacca, M., & Di Nunno, C.
(2006). Dismemberment: A review of the literature and descrip-
tion of 3 cases. The American Journal of Forensic Medicine
and Pathology, 27(4), 307–312. https://doi.org/10.1097/01.paf.
0000188170.55342.69 PMID:17133026
Drake, R., Vogl, A. W., & Mitchell, A. W. M. (2015). Gray’s
Anatomy for students (3rd ed., pp. 1–868). Philadelphia: Elsevier.
Edwards, K., Viner, M. D., Schweitzer, W., & Thali, M. J. (2013).
Sex determination from the foramen magnum. Journal of
Forensic Radiology and Imaging, 1(4), 186–192. https://doi.
org/10.1016/j.jofri.2013.06.004
Ekizoglu, O., Hocaoglu, E., Inci, E., Can, I. O., Solmaz, D.,
Aksoy, S., … Sayin, I. (2016). Assessment of sex in a modern
Turkish population using cranial anthropometric parameters.
Legal Medicine (Tokyo), 21, 45–52. https://doi.org/10.1016/j.
legalmed.2016.06.001 PMID:27497333
Franklin, D., Freedman, L., & Milne, N. (2005). Sexual dimor-
phism and discriminant function sexing in indigenous South
African crania. Homo, 55(3), 213–228. https://doi.org/10.1016/j.
jchb.2004.08.001 PMID:15803767
Franklin, D., Cardini, A., Flavel, A., & Kuliukas, A. (2013).
Estimation of sex from cranial measurements in a Western
Australian population. Forensic Science International, 229(1-
3), 158.e1–158.e8. https://doi.org/10.1016/j.forsciint.2013.03.
005 PMID:23537716
Gapert, R., Black, S., & Last, J. (2009). Sex determination from
the foramen magnum: Discriminant function analysis in an
eighteenth and nineteenth century British sample. International
Journal of Legal Medicine, 123(1), 25–33. https://doi.org/
10.1007/s00414-008-0256-0 PMID:18553095
Gapert, R., Black, S., & Last, J. (2013). Test of age-related variation
in the craniometry of the adult human foramen magnum region:
eschweizerbart_xxx
uncorrected proof
10 M. R. Dayal et al.
Implications for sex determination methods. Forensic Science,
Medicine, and Pathology, 9(4), 478–488. https://doi.org/
10.1007/s12024-013-9437-3 PMID:23584959
Garenne, M. (2016). A century of nuptiality decline in South Africa:
A longitudinal analysis of census data. African Population
Studies, 30(2). https://doi.org/10.11564/30-2-846
Gómez-Valdés, J. A., Torres Ramírez, G., Báez Molgado, S.,
Herrera Sain-Leu, P., Castrejón Caballero, J. L., & Sánchez-
Mejorada, G. (2011). Discriminant function analysis for sex
assessment in pelvic girdle bones: Sample from the contempo-
rary Mexican population. Journal of Forensic Sciences, 56(2),
297–301. https://doi.org/10.1111/j.1556-4029.2010.01663.x
PMID:21265837
González-Colmenares, G., Sanabria Medina, C., Rojas-Sánchez,
M. P., León, K., & Malpud, A. (2019). Sex estimation from skull
base radiographs in a contemporary Colombian population.
Journal of Forensic and Legal Medicine, 62, 77–81. https://doi.
org/10.1016/j.jm.2019.01.006 PMID:30684829
Gruber, P., Henneberg, M., Böni, T., & Rühli, F. J. (2009).
Variability of human foramen magnum size. Anatomical Record
(Hoboken, N.J.), 292(11), 1713–1719. https://doi.org/10.1002/
ar.21005 PMID:19777568
Holland, T. D. (1986). Sex determination of fragmentary crania
by analysis of the cranial base. American Journal of Physical
Anthropology, 70(2), 203–208. https://doi.org/10.1002/ajpa.
1330700207 PMID:3740247
İşcan, M. Y., & Steyn, M. (2013). The Human skeleton in foren-
sic medicine (3rd ed., p. 494). Springeld, Illinois: Charles C.
Thomas.
Keen, J. A. (1950). A study of the dierences between male and
female skulls. American Journal of Physical Anthropology,
8(1), 65–79. https://doi.org/10.1002/ajpa.1330080113 PMID:
15410864
Kranioti, E. F., Işcan, M. Y., & Michalodimitrakis, M. (2008).
Craniometric analysis of the modern Cretan population. Forensic
Science International, 180(2-3), 110.e1–110.e5. https://doi.
org/10.1016/j.forsciint.2008.06.018 PMID:18718728
Krüger, G. C., L’Abbé, E. N., Stull, K. E., & Kenyhercz, M. W.
(2015). Sexual dimorphism in cranial morphology among mod-
ern South Africans. International Journal of Legal Medicine,
129(4), 869–875. https://doi.org/10.1007/s00414-014-1111-0
PMID:25394745
Liebenberg, L., Krüger, G. C., L’Abbé, E. N., & Stull, K. E. (2019).
Postcraniometric sex and ancestry estimation in South Africa:
A validation study. International Journal of Legal Medicine,
133(1), 289–296. https://doi.org/10.1007/s00414-018-1865-x
PMID:29797281
Lin, L. I. (1989). A concordance correlation coecient to evalu-
ate reproducibility. Biometrics, 45(1), 255–268. https://doi.org/
10.2307/2532051 PMID:2720055
Lopez-Capp, T. T., Rynn, C., Wilkinson, C., Paiva, L. A. S., Michel-
Crosato, E., & Biazevic, M. G. H. (2018). Sexing the cranium
from the foramen magnum using discriminant analysis in a
Brazilian sample. Brazilian Dental Journal, 29(6), 592–598.
https://doi.org/10.1590/0103-6440201802087 PMID:30517483
Loth, S., & İşcan, M. (2000). Sex determination. In J. A. Siegel,
P. J. Saukko, & G. C. Knupfer (Eds.), Encyclopedia of Forensic
Sciences (pp. 252–260). London: Academic Press.
Madadin, M., Menezes, R. G., Al Saif, H. S., Abu Alola, H., Al
Muhanna, A., Gullenpet, A. H., … Al Dhafery, B. (2017).
Morphometric evaluation of the foramen magnum for sex
determination: A study from Saudi Arabia. Journal of Forensic
and Legal Medicine, 46, 66–71. https://doi.org/10.1016/j.jm.
2017.01.001 PMID:28157592
Mahakkanukrauh, P., Sinthubua, A., Prasitwattanaseree, S.,
Ruengdit, S., Singsuwan, P., Praneatpolgrang, S., & Duangto,
P. (2015). Craniometric study for sex determination in a Thai
population. Anatomy & Cell Biology, 48(4), 275–283. https://
doi.org/10.5115/acb.2015.48.4.275 PMID:26770879
McBride, G. B. (2005). A proposal for strength-of-agreement cri-
teria for Lin’s concordance correlation coecient. NIWA client
report: HAM2005-062, 62.
Murshed, K. A., Çiçekcibaşi, A. E., & Tuncer, I. (2003).
Morphometric evaluation of the foramen magnum and varia-
tions in its shape: A study on computerized tomographic images
of normal adults. Turkish Journal of Medical Sciences, 33(5),
301–306.
Ogawa, Y., Imaizumi, K., Miyasaka, S., & Yoshino, M. (2013).
Discriminant functions for sex estimation of modern Japanese
skulls. Journal of Forensic and Legal Medicine, 20(4), 234–
238. https://doi.org/10.1016/j.jm.2012.09.023 PMID:23622466
Patriquin, M. L., Loth, S. R., & Steyn, M. (2003). Sexually dimor-
phic pelvic morphology in South African whites and blacks.
Homo, 53(3), 255–262. https://doi.org/10.1078/0018-442X-
00049 PMID:12733399
Raghavendra Babu, Y. P., Kanchan, T., Attiku, Y., Dixit, P. N., &
Kotian, M. S. (2012). Sex estimation from foramen magnum
dimensions in an Indian population. Journal of Forensic and
Legal Medicine, 19(3), 162–167. https://doi.org/10.1016/j.jm.
2011.12.019 PMID:22391003
Saini, V., Srivastava, R., Shamal, S. N., Singh, T. B., Kumar, V.,
Kumar, P., & Tripathi, S. K. (2014). Temporal variations in basi-
cranium dimorphism of North Indians. International Journal
of Legal Medicine, 128(4), 699–707. https://doi.org/10.1007/
s00414-013-0957-x PMID:24374986
Seifert, Z., Friedl, L., Chaumoitre, K., & Brůžek, J. (2017).
Applicability and limitations of sex assessment based on
foramen magnum. Forensic Science International, 271, 126.
e1–126.e9. https://doi.org/10.1016/j.forsciint.2016.11.044
PMID:28024921
Singh, G., & Talwar, G. (2013). Morphometric analysis of foramen
magnum in human skull for sex determination. Human Biology
Review, 2(1), 29–41.
Small, C., Schepartz, L., Hemingway, J., & Brits, D. (2018).
Three-dimensionally derived interlandmark distances for sex
estimation in intact and fragmentary crania. Forensic Science
International, 287, 127–135. https://doi.org/10.1016/j.forsciint.
2018.02.012 PMID:29655098
Spradley, M. K., & Jantz, R. L. (2011). Sex estimation in foren-
sic anthropology: Skull versus postcranial elements. Journal
of Forensic Sciences, 56(2), 289–296. https://doi.org/10.1111/
j.1556-4029.2010.01635.x PMID:21210801
STATISTICS SA. (2020). Mid-year population estimates. 2019.
Pretoria: Statistics SA.
Steyn, M., & Işcan, M. Y. (1997). Sex determination from the femur
and tibia in South African whites. Forensic Science International,
90(1-2), 111–119. https://doi.org/10.1016/S0379-0738(97)00156-
4 PMID:9438371
Steyn, M., & Işcan, M. Y. (1998). Sexual dimorphism in the cra-
nia and mandibles of South African whites. Forensic Science
International, 98(1-2), 9–16. https://doi.org/10.1016/S0379-
0738(98)00120-0 PMID:10036755
eschweizerbart_xxx
uncorrected proof
Sex estimation from dimensions of the base of the skull 11
Steyn, M., & Işcan, M. Y. (2008). Metric sex determination from
the pelvis in modern Greeks. Forensic Science International,
179(1), 86.e1–86.e6. https://doi.org/10.1016/j.forsciint.2008.
04.022 PMID:18554832
Tambawala, S. S., Karjodkar, F. R., Sansare, K., Prakash, N., &
Dora, A. C. (2016). Sexual dimorphism of foramen magnum
using Cone Beam Computed Tomography. Journal of Forensic
and Legal Medicine, 44, 29–34. https://doi.org/10.1016/j.
jm.2016.08.005 PMID:27591340
Tomaszewska, A., Psonak, D., Maślińska, P., & Kwiatkowska,
B. (2020). Sex determination from fragmented human
remains – hierarchy of the foramen magnum dimensions.
Homo, 71(1), 9–17. https://doi.org/10.1127/homo/2019/1139
PMID:31651930
Toneva, D., Nikolova, S., Harizanov, S., Georgiev, I., Zlatareva,
D., Hadjidekov, V., … Lazarov, N. (2018). Sex estimation by
size and shape of foramen magnum based on CT imaging.
Legal Medicine (Tokyo), 35, 50–60. https://doi.org/10.1016/j.
legalmed.2018.09.009 PMID:30268691
Ubelaker, D. H. (1997). Taphonomic applications in forensic
anthropology. In W. D. Haglund & M. H. Sorg (Eds.), Forensic
taphonomy: the postmortem fate of human remains (pp. 77–92).
Boca Raton: CRC Press.
Ubelaker, D. H., & DeGaglia, C. M. (2017). Population variation
in skeletal sexual dimorphism. Forensic Science International,
278, 407.e1–407.e7. https://doi.org/10.1016/j.forsciint.2017.
06.012 PMID:28698063
Uthman, A. T., Al-Rawi, N. H., & Al-Timimi, J. F. (2012). Evaluation
of foramen magnum in gender determination using helical CT
scanning. Dento Maxillo Facial Radiology, 41(3), 197–202.
https://doi.org/10.1259/dmfr/21276789 PMID:22116135
Uysal, S., Gokharman, D., Kacar, M., Tuncbilek, I., & Kosa, U.
(2005). Estimation of sex by 3D CT measurements of the fora-
men magnum. Journal of Forensic Sciences, 50(6), 1310–1314.
PMID:16382824
Wescott, D. J., & Moore-Jansen, P. H. (2001). Metric variation in
the human occipital bone: Forensic anthropological applica-
tions. Journal of Forensic Sciences, 46(5), 1159–1163. https://
doi.org/10.1520/JFS15115J PMID:11569559
Zaafrane, M., Ben Khelil, M., Naccache, I., Ezzedine, E., Savall,
F., Telmon, N., … Hamdoun, M. (2018). Sex determination
of a Tunisian population by CT scan analysis of the skull.
International Journal of Legal Medicine, 132(3), 853–862.
https://doi.org/10.1007/s00414-017-1688-1 PMID:28936605
Manuscript received: 29 July 2021
Revisions requested: 12 October 2021
Revised version received: 06 December 2021
Manuscript accepted: 06 December 2021
eschweizerbart_xxx
