HOMO Vol. 53/3, pp.

255–262
© 2003 Urban & Fischer Verlag HOMO
http://www.urbanfischer.de/journals/homo

Sexually dimorphic pelvic morphology

in South African whites and blacks

M. L. PATRIQUIN, S. R. LOTH *, M. STEYN

Department of Anatomy, Faculty of Medicine, University of Pretoria, PO Box 2034, Pretoria 0001,
South Africa

Summary
It is well known that there is metric and morphologic variation in the expression of sexual dimorphism
between racial phenotypes and populations. Therefore, the purpose of this research is to assess morpho-
logic sex differences in the pelves of South African whites and blacks. Results will be used to improve
the identification of human skeletal remains by producing group specific standards. Morphologic fea-
tures of both left and right os coxae were studied in a sample of 400 known sex/race individuals from
the Pretoria and Dart collections. Bones that were clearly pathologic or visibly deformed were excluded
from the study. Data were subjected to SPSS analysis.
Results indicated that overall, pubic bone shape was the easiest to assess and was the most consis-
tently reliable morphological indicator of sex in both sexes and population groups. At 88% average
accuracy, the most discriminating traits in whites were pubic bone shape and subpubic concavity form.
In blacks, greater sciatic notch form allowed the highest separation, averaging 87.5%, followed by
pubic shape at 84.5%. Important findings included the fact that there were significant differences in the
accuracy of sex determination from pelvic morphology between both males and females and whites and
blacks.
In conclusion, this study provides quantification of individual morphological traits in the os coxae of
white and black South Africans that will be of value in forensic and archaeological analyses, especially
when dealing with fragmentary remains. It also demonstrates that population differences affect the
expression of sexual dimorphism and must be accounted for to develop the most effective methods of
analysis.

Introduction
Considerable research has been conducted to assess sexually dimorphic morpho-
logical characteristics of the pelvis (Phenice 1969, Houghton 1974, Kelley 1978,
İşcan & Derrick 1984, Suri & Tandon 1987, Lovell 1989, Anderson 1990,
MacLaughlin & Bruce 1990, Sutherland & Suchey 1991, Fernandez Camacho et
al 1993, Rogers & Saunders 1994, Luo 1995). Tables compiled from numerous
studies visually and metrically assessing sex from the pelvis can be found in
major texts and reference books (eg, Krogman & İşcan 1986, İşcan & Kennedy
1989).

* 2. May 1949 – 23. September 2002

0018-442X/03/53/03–255/$ 15.00/0
256 M. L. Patriquin, S. R. Loth, M. Steyn

Although often ignored, race and even population differences can affect the
expression of sexual dimorphism, especially metrically and even morphologically.
No one can deny that there are biological differences among the three major racial
phenotypes, Caucasoid, Mongoloid and Negroid (Todd & Lindala 1928, Cobb
1934, 1942, Lewis 1942, Holliday & Falsetti 1999, Loth & İşcan 2000, İşcan et al
2000). Morphological differences between race groups are readily observable,
especially in the skull (Keen 1950, Giles & Elliot 1962, Birkby 1966, DeVilliers
1968, Krogman & İşcan 1986, St. Hoyme & İşcan 1989, Gill & Gilbert 1990,
Brooks et al 1990, Rhine 1990, Curran 1990). The skulls of Negroids are com-
monly long, narrow and low (dolichocephalic) as opposed to the rounder, broader,
(brachycephalic) higher heads of Caucasoids and Mongoloids. Since the skull must
pass through the pelvis during birth, cranial shape differences have also been asso-
ciated with pelvic configuration (Schultz 1930, İşcan 1983). Moreover, postcranial
skeletal form and proportions have evolved to adapt to such factors as climate (eg,
Harrison et al 1988). Todd (1929) described the brachypellic white pelvis as a wide
basin for a broad torso, and the dolichopellic black pelvis as a pedestal for a nar-
row torso. Considerable variation has been demonstrated in the pelves of various
populations of blacks and whites (Letterman 1941, Davivongs 1963, Kajanoja
1966, İşcan 1981, 1983). Therefore, the purpose of this research is to assess tradi-
tional morphologic sex differences in the pelves of South African whites and
blacks.

Materials and methods

The skeletal material used in this research consisted of 400 pairs of adult os coxae,
evenly distributed between whites and blacks, and males and females. All speci-
mens had documentation of age (sometimes estimated), sex, and race. Since these
data were recorded prior to assessment, this was not a strictly blind study. The
skeletons for this project were obtained from anatomical dissecting room samples
housed in the Pretoria Collection (Department of Anatomy, Faculty of Medicine,
University of Pretoria) and Dart Collection (Department of Anatomical Sciences,
University of the Witwatersrand). In general, the skeletons are from individuals
who died in and around the Pretoria and Johannesburg areas, but both collections
include several black South African groups. The whites are predominantly of west-
ern European origin. Thus, it is important to emphasize that these South African
samples allow the assessment of variation at both racial and population levels since
these groups may differ from the whites and blacks from whose pelves morphology
was originally described.
Five morphological characteristics were assessed on each os coxa:
Pubic bone shape. Rectangular (female), triangular (male) or indeterminate as
determined by observing the pubic bone from the ventral aspect (Rogers & Saun-
ders 1994).
Subpubic concavity. Determined by observing the ventral aspect of the ischiopu-
bic ramus immediately inferior to the pubic bone. The presence of a subpubic con-
cavity is associated with females, its absence with males. If not clear the specimen
was recorded as «indeterminate» (modified from Phenice 1969).
 Sexually dimorphic pelvic morphology 257

Ischiopubic ramus form. Assess presence (male), absence (female), or indetermi-

nate expression of roughness and eversion associated with muscle attachment
(MacLaughlin & Bruce 1990).
Orientation of ischial tuberosity: To determine if the tuberosity is oriented lat-
eroanteriorly or lateroposteriorly, view os coxa ventrally with the pubic symphysis
in a vertical plane. Position thumb on the centre of the lateral aspect of the ischial
tuberosity. If 3/4 or more of the thumbnail is visible, the ischial tuberosity is consid-
ered visible anteriorly (female form). If only 1/2 or less of the nail is visible then the
orientation is rated as not visible from the anterior (male form). Intermediate val-
ues are recorded as indeterminate (see Gray’s Anatomy 1989, figures 3.222,
3.223).
Greater sciatic notch shape. Observations of the notch for the lateral aspect are
divided into 5 classifications: wide and asymmetrical (female), wide and symmetri-
cal (female), narrow and asymmetrical (male), narrow and symmetrical (male), or
indeterminate. A thumb is inserted into the notch; if there is little space it is consid-
ered narrow, but if there is plenty of room, wide. Symmetry was determined by
locating the point of greatest depth and noting if the distance on either side was
approximately equal (symmetrical) or clearly unequal (asymmetrical).

Results
The frequency distribution of each of the pelvic traits appears in table 1 and their
calculated sexing accuracy is in table 2. Side differences (ranging from 0% to 6%)
were not significant, thus only values for the left side are included here.
For all traits tested here, there was a disparity in diagnostic accuracy between
both the sexes and races. Females of both racial phenotypes were most accurately
classified by pubic bone shape, but whites had a higher rate (96%) than blacks
(88%). This part was less effective in males reaching only 80% in whites and
81% in blacks. Sciatic notch form in females (96% in whites, 84% in blacks) was
second best. This trait was also good on black males (91%), but poor in white
males (33%). The best indicators in males, ischial tuberosity orientation (96% in
whites, 92% in blacks) and ischiopubic ramus form (93% in both), were the
worst in females ranging from only 8% to 40% (table 2). Despite the low accu-
racy for some traits, all were significantly different between the sexes (p < 0.001)
(table 3).
Chi2 tests also indicated significant differences in the effectiveness of all sex indi-
cators between the racial phenotypes (table 4). The most diagnostic features in
whites (pubic bone shape and subpubic concavity form) both averaged 88% accu-
racy. In blacks, highest separation (averaging 87.5%) was obtained from sciatic
notch form. While this trait was excellent in white females (96%), it averaged only
64.5% in white males (33%).
There were relatively few bones in which some of the traits were not easily clas-
sifiable (table 5). These «indeterminates» (N = 24) were predominantly male (N =
16) and black (N = 14). Ischiopubic ramus form was by far the most frequent inde-
terminate (N = 15) followed by pubic bone shape (N = 6) and subpubic concavity
(N = 3).
258 M. L. Patriquin, S. R. Loth, M. Steyn

Table 1: Frequency distribution of morphological traits for each sex/race group (N = 100 for each group).

Trait Sex Race Total

–––––––––––––––––––––––––
White Black

Shape of greater sciatic notch

Wide Asymmetrical Male 26 6 32
Female 13 26 39
Wide Symmetrical Male 41 3 44
Female 83 58 141
Narrow Asymmetrical Male 18 73 91
Female 2 11 13
Narrow Symmetrical Male 15 18 33
Female 2 5 7
Subpubic concavity
Present Male 6 6 12
Female 84 74 158
Absent Male 92 94 186
Female 15 26 41
Indeterminate Male 2 0 2
Female 1 0 1
Ischiopubic ramus roughness
Present Male 93 93 186
Female 90 78 168
Absent Male 3 1 4
Female 8 19 27
Indeterminate Male 4 6 10
Female 2 3 5
Orientation of ischial tuberosity
Visible Anteriorly Male 4 8 12
Female 39 40 79
Little/Not visible Anteriorly Male 96 92 188
Female 61 60 121
Pubic bone shape
Rectangle Male 21 15 36
Female 96 98 184
Triangle Male 79 81 160
Female 3 11 14
Indeterminate Male 0 4 4
Female 1 1 2

Table 2: Percent of correctly assigned males and females based on morphological characters for the left
side.

Character White Black

–––––––––––––––––––––––––––––––––––– ––––––––––––––––––––––––––––––––––––
Males Females Average Males Females Average

Shape of Sciatic Notch 33.0 96.0 64.5 91.0 84.0 87.5

Subpubic Concavity 92.0 84.0 88.0 94.0 74.0 84.0
Ischiopubic Ramus Form 93.0 8.0 50.5 93.0 19.0 56.0
Ischial Tuberosity 96.0 39.0 67.5 92.0 40.0 66.0
Pubic Shape 80.0 96.0 88.0 81.0 88.0 84.5
 Sexually dimorphic pelvic morphology 259

Discussion
The results of this study indicate that while some traditionally applied morphologic
traits in the pelvis are relatively effective sex indicators, there is significant varia-
tion in their accuracy both by sex and race, as well as population. In fact, a major
reason this study was undertaken was that SRL and MS have noted that expected
morphologic sex difference in the pelves of South African blacks were not nearly as
clear as those in Caucasoids and Mongoloids as well as other populations of blacks
in the USA and elsewhere in Africa. In other words, there is significant disparity at
both the racial and population levels. Some of these traits may not be usable at all
in specific populations. Although many researchers have studied morphological sex
differences in the pelvis (eg, Phenice 1969, Kelley 1978, MacLaughlin & Bruce
1986, Lovell 1989, Anderson 1990, Sutherland & Suchey 1991, Rogers & Saun-
ders1994), little attention has been paid to the racial and population variation of
these characteristics.
Chi2 tests revealed highly significant differences (p < 0.001) between white and
black males in this sample (table 4). The most striking example was found in the
greater sciatic notch (GSN). The male form of the GSN was described as narrow,
deep and asymmetrical (Krogman & İşcan 1986). The present observations reveal
that only 33% of South African white males followed the pattern seen elsewhere in
both width and asymmetry. This was quite a contrast to their black counterparts,
91% of whom exhibited a narrow GSN (table 1). In addition, the notches of just
over half (54.5%) of white males were asymmetrical. While females showed an
overall higher percentage of the expected GSN form, population difference is sig-
nificant here, too, with many more white females (96%) than their black counter-
parts (84%) endowed with a wide notch. It is possible that the South African
whites, especially the males, tend to have wider notches to accommodate their
stature. The «founders» of this group were a relatively small subgroup of Euro-
peans whose descendants are larger than their counterparts in America (Steyn &
İşcan 1999). Thus, sciatic notch form exhibits both racial (white/black) and popu-
lation (groups within racial phenotypes) variation.

Table 3: Chi2 test of significance of morphologic sex differences between white males
and females and black males and females.

Characteristic Chi2 Deg. of freedom Signif.

Whites
Arc shape of sciatic notch 151.72 3 0.001
Subpubic concavity 93.37 2 0.001
Ischiopubic ramus form 304.69 2 0.001
Orientation of ischial tuberosity 64.98 1 0.001
Pubic bone shape 106.21 2 0.001
Blacks
Arc shape of sciatic notch 46.60 3 0.001
Subpubic concavity 8.00 1 0.005
Ischiopubic ramus form 245.83 2 0.001
Orientation of ischial tuberosity 54.08 1 0.001
Pubic bone shape 86.47 2 0.001
260 M. L. Patriquin, S. R. Loth, M. Steyn

Subpubic concavity is usually attributed to females. In over 90% of males this

trait was absent. However, it was not observed in many females either. Again, there
were significant differences in its expression in whites (84%) and blacks (74%).
The lesser frequency of the subpubic concavity in black females may be related to
the relatively shorter pubic bone length in blacks (Patriquin 2001, Patriquin et al
nd) and could possibly play a role in the high prevalence of problems with child-
birth in this population.
Greater roughness and eversion on the ischiopubic ramus ridge is associated
with the attachment of the crus penis in males. As would be expected, the corre-
sponding attachment in females for the clitoris is poorly developed (Williams &
Warwick 1989). However, the form of the ridge was not consistent in females in
these populations. One problem may be that it is necessary to define this trait more
precisely, since the ramus was noticeably roughened in many females, especially the
whites (table 2).
Overall, pubic bone shape was the easiest to assess and was the most consis-
tently reliable morphological indicator of sex in both sexes and populations tested
here. At 88% average accuracy, the most discriminating traits in whites were pubic
bone shape and subpubic concavity form. In blacks, greater sciatic notch allowed
the highest separation (averaging 87.5%) followed by pubic shape (84.5%). These

Table 4: Chi2 test of significance of morphologic race differences between white and
black males and white and black females.

Characteristic Chi2 Deg. of freedom Signif.

Males
Arc shape of sciatic notch 46.60 3 0.001
Subpubic concavity 321.16 2 0.001
Ischiopubic ramus form 320.68 2 0.001
Orientation of ischial tuberosity 154.88 1 0.001
Pubic bone shape 203.68 2 0.001
Females
Arc shape of sciatic notch 232.40 3 0.001
Subpubic concavity 199.69 2 0.001
Ischiopubic ramus form 234.67 2 0.001
Orientation of ischial tuberosity 8.82 1 0.003
Pubic bone shape 310.84 2 0.001

Table 5: Distribution of indeterminates of traits by sex and race.

Trait Sex Race Total

–––––––––––––––––––––––
White Black

Pubic Bone Shape Male 0 4 4

Female 1 1 2
Ischiopubic Ramus Form Male 4 6 10
Female 2 3 5
Subpubic Concavity Male 2 0 2
Female 1 0 1
 Sexually dimorphic pelvic morphology 261

results for individual traits should be particularly helpful when examining frag-
mented remains.
In conclusion, this research confirms that while pelvic morphology is a good
indicator of sex differences, some parts of the bone are more reliable than others.
More importantly, the present work has clearly demonstrated differences, associ-
ated with both race and population, in the expression and classification accuracy
of sexually dimorphic morphology in the os coxae. This work exemplifies the
importance of race and population specific standards to ensure the highest possible
accuracy when identifying human skeletal remains.

Acknowledgements
The authors would like to thank the Dept. of Anatomical Sciences, University of the Witwatersrand for
access to the Raymond Dart Collection. This study was funded by an NRF grant to MS and SRL.

References
Anderson BE (1990) Ventral arc of the os pubis: anatomical and developmental considerations. Am J
Phys Anthrop 83: 449–458.
Birkby WH (1966) An evaluation of race and sex identification from cranial measurements. Am J Phys
Anthrop 24: 21–28.
Brooks S, Brooks RH, France D (1990) Alveolar prognathism contour, an aspect of racial identification.
In Gill GW, Rhine S (eds) Skeletal Attributions of Race Methods for Forensic Anthropology.
Maxwell Museum of Anthropology, Albuquerque, New Mexico, 41–46.
Cobb WM (1934) The physical constitution of the American Negro. J Negro Ed 3, 340–388.
Cobb WM (1942) Physical anthropology of the American negro. Am J Phys Anthrop 29: 113–223.
Curran BK (1990) The application of measures of midfacial projection for racial classification. In Gill
GW, Rhine S (eds) Skeletal Attributions of Race Methods for Forensic Anthropology. Maxwell
Museum of Anthropology, Albuquerque, New Mexico, 55–57.
Davivongs V (1963). The pelvic girdle of the Australian Aborigine; sex differences and sex determina-
tion. Am J Phys Anthrop 21: 443–445.
DeVilliers H (1968). Sexual dimorphism of the skull of the South African Bantu-speaking negro. S Afr J
Sci 64: 118–124.
Fernandez Camacho FJ, Gomez Pellico L, Fernandez-Valencia R (1993) Osteometry of the human iliac
crest: patterns of normality and its utility in sexing human remains. J Forensic Sci 38: 779–787.
Giles E, Elliot O (1962) Race identification from cranial measurements. J Forensic Sci 7: 147–157.
Gill GW, Gilbert BM (1990) Race identification from the midfacial skeleton: American blacks and
whites. In Gill GW, Rhine S (eds) Skeletal Attributions of Race Methods for Forensic Anthropology.
Maxwell Museum of Anthropology, Albuquerque, New Mexico, 47–53.
Harrison GA, Tanner JM, Pilbeam DR, Baker PT (1988) Human Biology. Oxford Scientific Publica-
tions, Oxford.
Holliday TW, Falsetti AB (1999) A new method for discriminating African-American from European-
American skeletons using postcranial osteometrics reflective of body shape. J Forensic Sci 44: 926–930.
Houghton P (1974) The relationship of the pre-auricular groove of the ilium to pregnancy. Am J Phys
Anthrop 41: 381–390.
İşcan MY (1981) Race determination from the pelvis. OSSA 8: 95–100.
İşcan MY (1983) Assessment of race from the pelvis. Am J Phys Anthrop 62: 205–208.
İşcan MY, Derrick K (1984) Determination of sex from sacroiliac joint: a visual assessment technique.
Fl Sci 47: 94–98.
İşcan MY, Kennedy KAR (1989) Reconstruction of Life from the Skeleton. Alan R Liss, New York
İşcan MY, Loth SR, Steyn M (2000) Determination of racial affinity. In Siegel JA, Saukko PJ, Knupfer
GC (eds) Encyclopedia of Forensic Sciences. Academic Press, London, 227–235.
262 M. L. Patriquin, S. R. Loth, M. Steyn

Kajanoja P (1966) Sex determination of Finnish crania by discriminant function analysis. Am J Phys
Anthrop 24: 29–33.
Keen JA (1950). A study of the differences between male and female skulls. Am J Phys Anthrop 8:
64–80.
Kelley MA (1978) Phenice’s visual sexing technique for the os pubis: a critique. Am J Phys Anthrop 48:
121–122.
Krogman WM, İşcan MY (1986) The Human Skeleton in Forensic Medicine. Charles C. Thomas,
Springfield, Illinois.
Letterman GS (1941) The greater sciatic notch in American whites and negroes. Am J Phys Anthrop 28:
99–116.
Lewis JH (1942) The Biology of the Negro. University of Chicago Press, Chicago.
Loth SR, İşcan MY (2000) Sex determination. In Siegel JA, Saukko PJ, Knupfer GC (eds): Encyclopedia
of Forensic Sciences. Academic Press, London, 252–260.
Lovell NC (1989) Test of Phenice’s technique for determining sex from the os pubis. Am J Phys Anthrop
79: 117–120.
Luo YC (1995) Sex determination from the pubis by discriminant function analysis. Forensic Sci Int 74:
89–98.
MacLaughlin SM, Bruce MF (1986) Population variation in sexual dimorphism in the human innomi-
nate. J Hum Evol 1: 221–231.
MacLaughlin SM, Bruce MF (1990) The accuracy of sex identification in European skeletal remains
using Phenice characters. J Forensic Sci 35: 1384–1392.
Patriquin ML (2001) A comparative analysis of differences in the pelves of South African blacks and
whites. Unpublished MSc thesis, University of Pretoria.
Patriquin ML, Steyn M, Loth SR (n.d.) Metric analysis of sex differences in South African black and
white pelves. Submitted to Forensic Science International.
Phenice TW (1969) A newly developed visual method of sexing the os pubis. Am J Phys Anthrop 30:
297–302.
Rhine S (1990) Non-metric skull sexing. In Gill GW, Rhine S (eds) Skeletal Attributions of Race Meth-
ods for Forensic Anthropology. Maxwell Museum of Anthropology, Albuquerque, New Mexico,
9–18.
Rogers T, Saunders S (1994) Accuracy of sex determination using morphological traits of the human
pelvis. J Forensic Sci 39: 1047–1056.
Schultz A (1930) The skeleton of the trunk and limbs of higher primates. Hum Biol 2: 303–456.
St. Hoyme LE, İşcan MY (1989) Determination of sex and race: accuracy and assumptions. In İşcan
MY, Kennedy KAR (eds) Reconstruction of Life from the Skeleton. Alan R Liss, New York, 53–93.
Steyn M, İşcan MY (1999) Osteometric variation in the humerus: sexual dimorphism in South Africans.
Forensic Sci Int 106: 77–85.
Suri RK, Tandon JK (1987) Determination of sex from the pubic bone. Med Sci Law 27: 294–296.
Sutherland LD, Suchey JM (1991) Use of the ventral arc in pubic sex determination. J Forensic Sci 36:
501–511.
Todd TG, Lindala LA (1928) Dimensions of the body: Whites and American negroes of both sexes. Am
J Phys Anthrop 12: 35–119.
Todd TG (1929) Entrenched negro physical features. Hum Biol 1: 57–69.
Williams PL, Warwick PR (1989) Grey’s Anatomy. 37th ed, Churchill Livingstone, Edinburgh.

Corresponding address: Prof. Dr. MARYNA STEYN, Department of Anatomy, Faculty of Medicine,
University of Pretoria, PO Box 2034, Pretoria 0001, South Africa. Phone 27-12-319-2917, Fax 27-12-
319-2240, e-mail steyn@medic.up.ac.za.

Ms received 25.5.01, accepted 6.9.01, resubmitted 8.12.01