close

Вход

Забыли?

вход по аккаунту

?

Costal process of the first sacral vertebra Sexual dimorphism and obstetrical adaptation.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 132:395?405 (2007)
Costal Process of the First Sacral Vertebra: Sexual
Dimorphism and Obstetrical Adaptation
Robert G. Tague
Department of Geography and Anthropology, Louisiana State University, Baton Rouge, LA 70803-4105
KEY WORDS
australopithecine; obstetrics; pelvis; sacrum
ABSTRACT
The human sacrum is sexually dimorphic,
with males being larger than females in most dimensions.
Previous studies, though, suggest that females may have
a longer costal process of the ?rst sacral vertebra (S1)
than males. However, these studies neither quanti?ed nor
tested statistically the costal process of S1. This study
compares S1 with the ?ve lumbar vertebrae (L1 to L5) for
a number of metric dimensions, including costal process
length. Four issues are addressed, the: 1) hypothesis that
females have a longer costal process of S1 than males;
2) hypothesis that homologous structures (i.e., costal processes of L1 to S1) differ in their direction of sexual
dimorphism; 3) importance of the costal process of S1
to the obstetrical capacity of the pelvis; and 4) evolution of sexual dimorphism in costal process length of
S1. One hundred ninety-seven individuals, including
males and females of American blacks and whites, from the
Hamann?Todd and Terry Collections were studied. Results
show that males are signi?cantly larger than females for
most vertebral measurements, except that females have a
signi?cantly longer costal process of S1 than males. Costal
process length of S1 is positively correlated with the transverse diameter and circumference of the pelvic inlet. The
magnitude of sexual dimorphism in costal process length
of S1 ranks this measure among the most highly dimorphic of the pelvis. Compared with the humans in this
study, australopithecines have a relatively long costal process of S1, but their broad sacrum was not associated with
obstetrical imperatives. Am J Phys Anthropol 132:395?
405, 2007. V 2006 Wiley-Liss, Inc.
The human sacrum is sexually dimorphic in some
dimensions but not others. For example, males are
larger than females in curved length and maximum
depth along the ventral surface of the sacrum, and the
anteroposterior and transverse diameters of the body of
the ?rst vertebra (S1), but the sexes are monomorphic
in breadth of the sacrum at the plane of the pelvic
inlet and height of S1 (Paterson, 1893; Trotter, 1926;
Stra?dalova?, 1974; Flander, 1978; Tague, 1992; Fang
et al., 1994). Studies by Fawcett (1938) and Flander
(1978) are pertinent to this study. Fawcett (1938) showed
that males have a higher corporo-basal index (computed
as, breadth of the body of S1/breadth of the sacrum at
the level of S1) than females. Flander (1978) demonstrated that the body of S1 is signi?cantly wider in
males than females, but the sexes are not signi?cantly
different from one another in breadth of the sacrum. As
sacral breadth at S1 is comprised of the breadth of the
body of S1 and length of the costal processes of S1, the
implication of these two studies is that females have a
longer costal process of S1 than males. However, the costal process of S1 has neither been quanti?ed nor tested
statistically in previous studies. This study addresses
four issues. First, the hypothesis is tested that females
have a longer costal process of S1 than males. Second,
the hypothesis is tested that homologous structures (i.e.,
costal processes of the ?ve lumbar vertebrae (L1 to L5)
and of S1) differ in their direction of sexual dimorphism,
with the costal processes of the lumbar vertebrae being
longer in males, and that of S1 being longer in females.
Third, the contribution of the costal process of S1 to the
obstetrical capacity of the pelvic inlet is determined.
Fourth, the evolution of sexual dimorphism in costal process length of S1 is considered by evaluation of Australopithecus and earlier Homo.
Understanding sexual dimorphism of the costal process
of S1 is important from obstetrical and developmental perspectives. Obstetrically, the pelvic inlet can be considered a
closed ring of bone, consisting of lengths of the ilia, pubes,
and costal processes of S1, and breadth of the body of S1.
The circumference of the inlet during parturition is modi?ed minimally by hormonal ??relaxation?? of pelvic ligaments (Tague, 1992). Therefore, obstetrical adequacy of the
inlet is ensured by having one or more of its components
lengthened relative to that of males. Females have a longer
ilium and pubis than males, though the sexes are monomorphic in pubic length in some populations (Straus, 1927;
Washburn, 1948, 1949; Segebarth-Orban, 1980; Rosenberg,
1988). As males are larger than females in breadth of the
body of S1 (Flander, 1978), the costal process of S1 is the
?nal skeletal element that can contribute to an enlarged
inlet in females. As males are typically larger than females
in pre-sacral vertebral anatomy (MacLaughlin and Oldale,
1992; Fang et al., 1994; Kim et al., 1994; Liguoro et al.,
1994; Tatarek, 2005) and in the ribs (see below), a longer
costal process of S1 in females than males would imply
both a distinctive developmental biology for S1 and that
this sexual dimorphism is an obstetrical adaptation.
Costal processes develop in most, if not all, vertebrae
(Last, 1978; O?Rahilly et al., 1990b). The costal elements
C 2006
V
WILEY-LISS, INC.
C
Correspondence to: Robert G. Tague, Department of Geography
and Anthropology, Louisiana State University, Baton Rouge, Louisiana 70803-4105, USA. E-mail: rtague@lsu.edu
Received 15 February 2006; accepted 12 October 2006
DOI 10.1002/ajpa.20531
Published online 18 December 2006 in Wiley InterScience
(www.interscience.wiley.com).
396
R.G. TAGUE
??are presumed to arise from ontogenetically comparable
primordia?? (O?Rahilly et al., 1990b). In humans, the costal processes of the thoracic vertebrae (i.e., ribs) form synovial joints with the bodies1 and transverse processes
of these vertebrae (ribs 11 and 12 do not articulate with
the transverse processes of the corresponding vertebrae).
Among the other vertebrae, the costal process fuses with
the transverse process, and fails to develop distally. This
study compares the sexes in the length of the costal
processes of L1 to S1. Most researchers consider the descriptive ??transverse process?? of a lumbar vertebra to be
principally the costal process (Nutter, 1914; Harris,
1933; Mitchell, 1936?1937; O?Rahilly et al., 1990a;
Standring, 2005; but see Fawcett, 1907). For the sacrum,
most researchers consider the anterior two-thirds of the
ala to be the costal process, and the posterior one-third
the transverse process (Fawcett, 1907; Mitchell, 1936?
1937; Standring, 2005). Figure 1 shows homologous
parts of adult thoracic, lumbar, and sacral vertebrae,
including the costal process. O?Rahilly et al. (1990b),
however, argued that the sacral ala is ontogenetically
different from the costal processes of other vertebrae,
and is new to the sacrum. The costal process of the sacrum, according to O?Rahilly et al. (1990b), is more dorsally placed, near the transverse process.
This researcher is not aware of any study comparing
the sexes for the costal processes of the cervical or lumbar vertebrae. However, the ribs are sexually dimorphic.
Males have longer ?rst and twelfth ribs than females,
and, more generally for ribs, males are larger than
females for heights of the head and neck, anteroposterior
diameter of the head, straight length from the articular
facet of the tubercle to the posterior angle, and maximum and minimum diameters of the shaft (Lanier, 1944;
I?s?can, 1985; Dupras and Pfeiffer, 1996; C?o?log?lu et al.,
1998; Owers and Pastor, 2005). This study tests the hypothesis that homologous structures (i.e., costal processes of L1 to S1) are dissimilar in their sexual dimorphism (but see O?Rahilly et al., 1990b, on whether the
sacral ala is homologous with the costal processes of
lumbar vertebrae). As males have longer ribs than
females (Lanier, 1944), but as females may have a longer
costal process of S1 than males (see above), this study
considers whether the costal processes of L1 to L5 show
a uniform difference with S1 in the direction of sexual
dimorphism or a gradient of sexual dimorphism. If the
lumbar vertebrae show a uniform difference with S1 in
the direction of sexual dimorphism, males will have longer costal processes than females for the lumbar vertebrae, and S1 will be unique with females having a longer
costal process than males. If the lumbar vertebrae show
a gradient of sexual dimorphism, males will have longer
costal processes than females in the upper lumbar vertebrae, and the lower lumbar vertebrae will show either
no sexual difference in costal process length or females
will have longer costal processes than males.
MATERIALS AND METHODS
One hundred ninety-seven individuals from the
Hamann?Todd and Terry skeletal collections were used:
51 black males, 50 black females, 52 white males, and
1
O?Rahilly et al. (1980:566) stated that the head of the rib ??articulates with the neural arch only, not with the centrum. . . . In the
adult . . . the head of the rib articulates with that part of the body of
the vertebra that has ossi?ed from the neural arch.?? Also see Figure
1.
Fig. 1. Homologies of adult thoracic, lumbar, and ?rst sacral
vertebrae, derived from embryonic centrum, neural arch, and
costal process. Superior view of left half of vertebrae. Drawing
based on Standring (2005:794, Fig. 47.7B).
44 white females. All individuals were between the ages
at death of 25 and 49. The age limit of 49 was chosen to
have the samples consisting of individuals with little or
no osteophytic growth on the vertebrae. Records at the
Cleveland Museum of Natural History and the National
Museum of Natural History, Smithsonian Institution
were used for information on sex, ethnicity, and age at
death. Individuals were selected for study if they had
the modal number of vertebrae for the thoracic, lumbar,
and sacral segments of the vertebral column?12, 5, and
5, respectively (Standring, 2005).
American Journal of Physical Anthropology?DOI 10.1002/ajpa
COSTAL PROCESS OF SACRUM: SEXUAL DIMORPHISM
Measurements of the vertebrae were as follows: maximum anteroposterior (Fig. 2a,b, A-B) and transverse
diameters (Fig. 2a,b, C-D) on the superior surface of the
body (L1 to S1); maximum transverse diameter of the
neural canal (L1 to S1) (Fig. 2a,b, E-F); minimum width
of the right and left pedicles (L1 to L5) (Fig. 2a, E-G,
F-H); and maximum diameter between the costal processes (L1 to S1) (Fig. 2a,b, I-J). Measurements of the
hipbones and pelvis included: curved length of the linea
terminalis, from the point where the linea terminalis
meets the auricular surface of the ilium to the superomedial point of the pubic symphysis (Fig. 2c, K-L); straight
length of the pubis, from the most anteromedial point of
the lunate surface of the acetabulum to the superomedial point of the pubic symphysis (Fig. 2d, L-M); and
maximum transverse diameter of the pelvic inlet, measured between the linea terminales and visually oriented
to be perpendicular to the anteroposterior diameter of
the pelvic inlet (not shown) (Fig. 2e, N-O). To measure
the transverse diameter of the inlet, the hipbones and
sacrum were articulated with rubber bands. The pubes
touched in the midline; no compensation was made for a
symphyseal disk.
Three variables were computed2:
costal process length М №maximum
diameter between costal processes maximum
transverse diameter of bodyо=2;
397
When the measurement was between calibrated units of
the curvometer, this researcher estimated the distance
between the units as 25, 50, and 75%. This procedure of
estimation is accurate to within 0.13 mm (Tague, 1995).
For variables for which measurements of left and right
sides were available, the mean of the measurements for
an individual was used; if only one side of the bone could
be measured for an individual, that single value was
used.
The 29 measurements of the vertebrae were repeated
on 10 individuals several days after the original measurements were taken. Measurements of the transverse
diameter of the inlet, and lengths of the linea terminalis
and pubis were repeated on 36, 36, and 20 individuals,
respectively, from several days to several years after the
original measurements were taken. With the exception
of width of the pedicle of L5, measurement precision was
high for all variables. Measurement precision for pedicle
width of L5 was 94.2% (mean) and 96.1% (median). For
all other variables, measurement precision ranged from
98.5% (mean) and 99.0% (median) for linea terminalis
length to 100% (both mean and median) for maximum
diameter between costal processes of L1.
Measurement precision М №1 №joriginal measurement
repeat measurementj=original measurementоо100%:
№1о
relative costal process length М №2№costal process lengthо
100о=maximum diameter between costal processes;
№2о
and
pelvic inlet circumference М maximum transverse
diameter of body of S1 ў 2№costal process length of
S1 ў linea terminalis length). №3о
Sliding calipers and a curvometer were used to take
linear and curvilinear measurements, respectively. With
the exception of pubic length, all linear measurements
were recorded to the nearest 0.1 mm. For pubic length,
113 individuals were measured to the nearest 0.l mm
and 75 to the nearest mm. Although the curvometer is
calibrated in units of 0.05 inches, these measurements
were converted subsequently to the nearest 0.1 mm.
2
The costal process of L1 to L4 arises near the junction of the pedicle and lamina, and that of L5 arises from the ??whole of the pedicle
and encroach(es) on the body?? (Standring, 2005:749). For this study,
the body of the vertebra [Eq. (1), text] rather than the summation
of the diameter of the neural canal and pedicle width was used in
computation of costal process length because the pedicle could not
be measured for S1. For L1 to L5, the alternative computation for
costal process length would be:
costal process length М (maximum diameter between costal processes (maximum transverse diameter of neural canal ў 2 (width
of pedicle)))/2.
Pearson?s correlation coef?cients between these two computations
of costal process length are high, positive, and signi?cant using the
combined sample of individuals in this study; the level of probability
for all ?ve coef?cients is <0.001: 0.905 (L1, n М 146), 0.895 (L2, n М
147), 0.922 (L3, n М 139), 0.868 (L4, n М 91), and 0.651 (L5, n М
104). Moreover, the results of this study are essentially the same
regardless of which computation is used for costal process length.
№4о
Statistical procedures, using SPSS software (1992,
2001), included Pearson?s and Spearman?s correlation
coef?cient analysis, Kruskal?Wallis test, Student?s t-test,
and Wilcoxon?Mann?Whitney test. The nonparametric
statistical tests (Spearman?s correlation coef?cient analysis, Kruskal?Wallis test, and Wilcoxon?Mann?Whitney
test) were used if n < 25 for one of the samples being
evaluated. Two-tailed tests of signi?cance were used,
with the level of signi?cance set at P 0.05.
The samples of blacks and whites were analyzed separately for two reasons. First, blacks and whites in both
sexes have been shown to differ signi?cantly for some of
the measurements used in this study: breadth of the sacrum at the level of S1 and transverse diameter of the
body of S1 (Flander, 1978), transverse diameter of the
pelvic inlet (I?s?can, 1983; I?s?can and Cotton, 1985), and
lengths of the pubis and linea terminalis (unpublished
results by author, but are available upon request).
Lanier (1939) also ??suggest(ed a) signi?cant difference??
between black and white males in the anteroposterior diameter of vertebral bodies. Second, sample sizes of
blacks and whites in this study differ due to variable
preservation of the bones.
Individuals from the Hamann?Todd collection were
compared with those from the Terry collection to determine if pooling the individuals from the two collections
was appropriate. For the suite of 45 vertebral and pelvic
variables in this study, there was no signi?cant difference between the collections for any variable for black
and white males. For white females, only the transverse
diameter of the pelvic inlet showed a signi?cant difference, with the Hamann?Todd sample being larger (Wilcoxon?Mann?Whitney test; results not presented, but
are available from author upon request). Black females
were not included in this analysis because all but one
individual was from the Terry collection. Therefore,
individuals from the two collections were pooled for all
analyses.
American Journal of Physical Anthropology?DOI 10.1002/ajpa
398
R.G. TAGUE
Fig. 2. Measurements of lumbar and sacral vertebrae and of
hipbone and pelvis. Superior
views of lumbar vertebra (a),
?rst sacral vertebra (b), and pelvis (e); medial and anterior views
of left hipbone (c and d, respectively). Dotted line represents
curved length (c, K-L); all other
lines represent straight length.
Points E and F (on ??a??) for measurement of maximum transverse diameter of neural canal
and minimum width of pedicle
may not have been the same on
individual specimens, but are
shown here as the same for ease
of illustration.
RESULTS
Ericksen (1976, 1978a, 1978b) reported that the bodies
of the lumbar vertebrae increase in breadth with
advancing age among adults; she studied individuals
whose ages at death ranged from 20 to 90 years. As vertebral body breadth is used to compute costal process
length in this study [Eq. (1)], the samples of black males,
black females, white males, and white females were
tested to determine if skeletal size varied with age. For
the suite of 45 vertebral and pelvic variables, there is
only one signi?cant difference in a comparison of age
groups 25?29, 30?39, and 40?49?the oldest age group
in black males has a signi?cantly shorter diameter
between the costal processes of L5 than the younger two
age groups (Kruskal?Wallis test; results not presented,
but are available from author upon request).
Table 1 presents summary statistics of the four samples. Results of the Student?s t-test and Wilcoxon?
Mann?Whitney test (Table 2) show a number of signi?cant differences between black and white males and
between black and white females. Whites are signi?cantly larger than blacks in both sexes for the following:
anteroposterior diameters of the bodies of L1 to L3;
transverse diameter of the neural canal for L1 and L2;
diameter between the costal processes for L1 and S1;
costal process length for L1 and S1; relative costal process length for S1; lengths of linea terminalis and pubis;
and transverse diameter and circumference of the pelvic
inlet. Blacks are signi?cantly larger than whites in both
American Journal of Physical Anthropology?DOI 10.1002/ajpa
399
COSTAL PROCESS OF SACRUM: SEXUAL DIMORPHISM
TABLE 1. Summary statistics of vertebral and pelvic measurements for black males (BM), black females (BF), white males (WM),
and white females (WF); all variables in mm, except relative length of costal process in %a
BM
Variables
Anteroposterior diameter of body
L1
L2
L3
L4
L5
S1
Transverse diameter of body
L1
L2
L3
L4
L5
S1
Transverse diameter of neural canal
L1
L2
L3
L4
L5
S1
Width of pedicle
L1
L2
L3
L4
L5
Diameter between costal processes
L1
L2
L3
L4
L5
S1
Length of costal process
L1
L2
L3
L4
L5
S1
Relative length of costal process
L1
L2
L3
L4
L5
S1
Length of linea terminalis
Length of pubis
Transverse diameter of pelvic inlet
Circumference of pelvic inlet
a
BF
WM
WF
x
s
n
x
s
n
x
s
n
x
s
n
29.2
30.9
33.1
34.0
35.1
32.1
1.9
2.2
2.2
2.3
2.4
2.5
50
47
49
45
50
50
25.6
26.9
28.5
29.5
30.5
28.6
1.8
2.0
2.1
2.0
2.1
2.1
50
50
49
47
50
49
32.6
33.5
34.3
34.6
34.8
32.4
2.6
2.5
2.6
2.6
2.4
2.6
49
48
44
45
45
46
27.2
28.5
29.9
30.3
31.1
28.2
2.1
2.2
2.0
2.1
2.0
1.9
44
41
41
42
44
37
45.4
47.1
49.2
51.4
54.1
53.2
2.9
2.7
2.9
2.8
3.3
3.8
50
48
48
48
50
50
39.1
41.4
43.7
45.9
48.5
46.2
2.1
2.0
2.3
2.6
2.9
3.6
50
50
50
49
50
49
46.7
48.5
50.9
52.5
54.2
53.1
3.0
2.9
2.7
3.0
3.5
3.8
51
47
45
47
49
47
39.8
41.9
44.1
46.1
48.1
45.7
3.0
2.9
3.1
3.0
3.5
4.2
42
42
41
43
42
40
21.9
22.4
23.4
24.4
26.8
31.3
1.8
1.8
2.1
1.9
2.2
2.9
50
51
50
50
50
51
20.2
20.6
21.3
22.3
25.2
28.9
1.6
1.5
1.7
1.8
2.7
2.8
50
50
50
50
50
49
23.4
23.3
23.3
23.5
26.5
32.3
1.8
1.6
1.6
1.9
2.4
2.6
52
52
52
52
51
49
22.3
22.3
22.5
23.0
26.2
31.6
1.9
2.0
1.9
2.1
2.0
2.7
44
43
44
44
44
42
8.7
8.9
10.2
11.1
16.3
1.7
1.5
1.6
1.6
3.0
50
51
50
50
49
6.7
7.1
8.4
9.6
14.3
1.3
1.2
1.3
1.3
2.2
50
50
50
50
50
8.0
8.2
9.8
11.1
17.4
1.7
1.6
1.7
1.7
2.9
52
52
52
52
50
6.1
6.5
8.1
9.2
14.3
1.4
1.1
1.2
1.1
2.6
44
43
44
44
44
70.9
84.6
94.2
89.1
90.3
111.6
7.3
6.3
6.6
5.9
6.3
6.1
42
43
45
30
36
49
64.8
73.6
81.3
78.4
84.0
110.8
7.2
5.0
7.1
5.9
5.8
6.8
45
48
46
38
40
50
75.8
85.0
93.8
90.4
93.5
117.5
8.0
5.0
5.9
5.1
6.4
5.5
35
36
33
15
19
51
68.8
74.5
81.5
79.2
85.9
116.3
5.2
4.9
5.5
4.6
5.6
7.9
27
24
22
13
19
40
12.7
18.7
22.3
19.2
18.4
29.1
3.4
2.9
3.5
2.7
3.2
2.8
41
41
43
28
36
48
12.9
16.1
18.8
16.4
17.8
32.3
3.2
2.2
3.3
2.8
2.3
2.9
45
48
46
37
40
49
14.6
17.9
21.1
19.0
19.6
32.2
3.8
2.5
2.8
2.7
3.0
2.8
35
35
30
14
17
46
14.4
16.4
19.0
16.9
18.8
35.3
2.3
2.2
2.4
2.5
2.6
3.7
26
23
21
12
18
37
35.3
44.1
47.2
42.5
40.4
52.2
130.5
66.6
116.5
372.6
7.6
4.2
4.5
4.2
4.7
3.4
7.3
4.1
7.5
19.3
41
41
43
28
36
48
49
49
48
47
39.2
43.6
45.9
41.6
42.2
58.2
136.0
67.6
122.8
382.7
6.0
3.6
4.5
4.7
3.3
2.9
8.8
3.8
7.7
22.4
45
48
46
37
40
49
49
49
49
49
37.8
42.1
45.1
41.8
41.9
54.7
137.2
69.5
126.2
392.1
7.0
3.9
3.6
3.9
4.2
3.1
6.4
3.6
5.9
15.2
35
35
30
14
17
46
51
52
49
50
41.4
43.7
46.2
42.4
43.5
60.5
148.0
72.5
134.6
413.3
4.6
3.6
3.9
4.3
4.3
3.5
10.0
5.0
10.0
24.8
26
23
21
12
18
37
37
38
37
36
Abbreviations: L1 to L5 М ?rst to ?fth lumbar vertebrae; S1 М ?rst sacral vertebra.
sexes for width of the pedicle of L1 and L2. There are
several other variables in which one ethnic group is
larger than the other for one sex, but not the other.
Comparison between black males and females and
between white males and females shows that in both
blacks and whites (Table 2): 1) males are signi?cantly
larger than females for the anteroposterior and transverse
diameters of the bodies of L1 to S1, transverse diameter of
the neural canal for L1 to L3, width of the pedicle and diameter between the costal processes for L1 to L5, and costal process length for L2 and L3; 2) females are signi?cantly larger than males for costal process length of S1,
relative costal process length of L1 and S1, linea terminalis length, and transverse diameter and circumference of
the pelvic inlet; and 3) the sexes are not signi?cantly different for the diameter between the costal processes of S1,
costal process length of L1 and L5, and relative costal process length of L2 to L5. There are several other variables
in which one sex is larger than the other for one ethnic
group, but not the other.
Correlation coef?cient analysis (Table 3) shows that,
with the exception of the transverse diameter of the body
of L3 in black males, costal process length is not signi?cantly associated with the transverse diameter of the body
American Journal of Physical Anthropology?DOI 10.1002/ajpa
400
R.G. TAGUE
TABLE 2. Results of Student?s t-test and Wilcoxon?Mann?Whitney test for pairwise comparisons of black males (BM), black females
(BF), white males (WM), and white females (WF)a
Variables
Anteroposterior diameter of body
L1
L2
L3
L4
L5
S1
Transverse diameter of body
L1
L2
L3
L4
L5
S1
Transverse diameter of neural canal
L1
L2
L3
L4
L5
S1
Width of pedicle
L1
L2
L3
L4
L5
Diameter between costal processes
L1
L2
L3
L4
L5
S1
Length of costal process
L1
L2
L3
L4
L5
S1
Relative length of costal process
L1
L2
L3
L4
L5
S1
Length of linea terminalis
Length of pubis
Transverse diameter of pelvic inlet
Circumference of pelvic inlet
BM vs. WM
P
BF vs. WF
P
BM vs. BF
P
WM vs. WF
P
WM
WM
WM
ns
ns
ns
<0.001
<0.001
0.016
0.194
0.625
0.482
WF
WF
WF
ns
ns
ns
<0.001
<0.001
0.003
0.054
0.172
0.365
BM
BM
BM
BM
BM
BM
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
WM
WM
WM
WM
WM
WM
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
WM
WM
WM
WM
ns
ns
0.021
0.019
0.004
0.050
0.934
0.914
ns
ns
ns
ns
ns
ns
0.253
0.305
0.489
0.709
0.606
0.527
BM
BM
BM
BM
BM
BM
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
WM
WM
WM
WM
WM
WM
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
WM
WM
ns
BM
ns
ns
<0.001
0.012
0.928
0.034
0.442
0.068
WF
WF
WF
ns
WF
WF
<0.001
<0.001
0.001
0.061
0.030
<0.001
BM
BM
BM
BM
BM
BM
<0.001
<0.001
<0.001
<0.001
0.001
<0.001
WM
WM
WM
ns
ns
ns
0.004
0.008
0.024
0.206
0.602
0.195
BM
BM
ns
ns
ns
0.028
0.023
0.215
0.989
0.063
BF
BF
ns
ns
ns
0.017
0.026
0.241
0.136
0.995
BM
BM
BM
BM
BM
<0.001
<0.001
<0.001
<0.001
<0.001
WM
WM
WM
WM
WM
<0.001
<0.001
<0.001
<0.001
<0.001
WM
ns
ns
ns
ns
WM
0.007
0.727
0.777
0.588b
0.117b
<0.001
WF
ns
ns
ns
ns
WF
0.013
0.546b
0.854b
0.517b
0.394b
0.001
BM
BM
BM
BM
BM
ns
<0.001
<0.001
<0.001
<0.001
<0.001
0.527
WM
WM
WM
WM
WM
ns
<0.001
<0.001b
<0.001b
<0.001b
0.001b
0.418
WM
ns
ns
ns
ns
WM
0.031
0.212
0.131
0.729b
0.156b
<0.001
WF
ns
ns
ns
ns
WF
0.040
0.527b
0.909b
0.585b
0.076b
<0.001
ns
BM
BM
BM
ns
BF
0.844
<0.001
<0.001
<0.001
0.352
<0.001
ns
WM
WM
ns
ns
WF
0.846
0.034b
0.015b
0.060b
0.448b
<0.001
ns
BM
BM
ns
ns
WM
WM
WM
WM
WM
0.145
0.038
0.044
0.455b
0.182b
0.001
<0.001
<0.001
<0.001
<0.001
ns
ns
ns
ns
ns
WF
WF
WF
WF
WF
0.104
0.712b
0.989b
0.642b
0.139b
0.001
<0.001
<0.001
<0.001
<0.001
BF
ns
ns
ns
ns
BF
BF
ns
BF
BF
0.010
0.580
0.183
0.423
0.063
<0.001
0.001
0.171
<0.001
0.021
WF
ns
ns
ns
ns
WF
WF
WF
WF
WF
0.024
0.129b
0.358b
0.681b
0.262b
<0.001
<0.001
0.003
<0.001
<0.001
In each comparison, the sample that is signi?cantly larger is listed with the level of probability. Abbreviations: L1 to L5 М ?rst to
?fth lumbar vertebrae; S1 М ?rst sacral vertebra; ns М not signi?cant.
b
Wilcoxon?Mann?Whitney results presented because n < 25 for one group in the comparison (Table 1).
a
for L1 to S1. The transverse diameter of the pelvic inlet is
signi?cantly, positively correlated with costal process
length of S1 and the diameter between the costal processes of S1 in all four samples, but with the transverse
diameter of the body of S1 only in black males. The circumference of the pelvic inlet is signi?cantly, positively
correlated with costal process length of S1, transverse diameter of the body of S1, and diameter between the costal
processes of S1 in all four samples. Finally, the correlations among the three components of inlet circumference
show that: 1) costal process length of S1 is not signi?cantly associated with the transverse diameter of the body
of S1 in any of the four samples (see above); 2) costal process length of S1 is signi?cantly, positively associated with
linea terminalis length in three samples, but the correlation coef?cient is not signi?cantly different from zero in
white males; and 3) the transverse diameter of the body of
S1 is signi?cantly, positively associated with linea terminalis length in black males and females, but the correlation coef?cients are not signi?cantly different from zero in
white males and females.
Absolute and relative costal process length of S1 can be
computed for three australopithecines (Table 4): AL 288-1
[assigned to Australopithecus afarensis (Johanson and
American Journal of Physical Anthropology?DOI 10.1002/ajpa
401
COSTAL PROCESS OF SACRUM: SEXUAL DIMORPHISM
TABLE 3. Pearson?s (r) and Spearman?s correlation coef?cients involving costal process length and transverse diameter of body
for L1 to S1, diameter between costal processes of S1, transverse diameter of pelvic inlet, circumference of pelvic inlet, and linea
terminalis length for black males (BM), black females (BF), white males (WM), and white females (WF)a
BM
Set of correlations
r
P
BF
n
r
WM
P
1. Correlations between costal process length and transverse diameter
L1
0.004
0.982 41
0.206
L2
0.037
0.819 41
0.042
L3
0.356
0.019 43
0.115
L4
0.328
0.089 28 0.078
L5
0.279
0.099 36
0.127
S1
0.272
0.061 48 0.012
of body
0.175
0.776
0.445
0.647
0.433
0.934
n
r
45
48
46
37
40
49
0.034
0.267
0.163
0.246b
0.103b
0.271
P
WF
N
r
0.845
0.122
0.389
0.396
0.694
0.069
35
35
30
14
17
46
0.181
0.024b
0.411b
0.382b
0.209b
0.115
P
n
0.378
0.914
0.064
0.221
0.404
0.499
26
23
21
12
18
37
2. Correlations between transverse diameter of pelvic inlet and
costal process length of S1
0.466
0.001 45
0.775
transverse diameter of body of S1
0.302
0.039 47
0.262
diameter between costal
0.679 <0.001 46
0.793
processes of S1
<0.001
0.072
<0.001
48
48
49
0.468
0.198
0.569
0.001
0.199
<0.001
44
44
49
0.664
0.261
0.765
<0.001
0.136
<0.001
34
34
37
3. Correlations between circumference
costal process length of S1
transverse diameter of body of S1
diameter between costal
processes of S1
<0.001
<0.001
<0.001
48
48
49
0.428
0.341
0.612
0.003
0.022
<0.001
45
45
50
0.554
0.388
0.701
0.001
0.023
<0.001
34
34
36
0.001
48
0.166
0.277
45
0.341
0.049
34
0.002
48
0.283
0.056
46
0.322
0.063
34
of pelvic inlet and
0.513 <0.001 46
0.473
0.001 46
0.796 <0.001 47
0.611
0.507
0.787
4. Correlations among components of pelvic inlet circumference
costal process length of S1 and
0.341
0.020 46
0.453
linea terminalis length
transverse diameter of body of S1
0.439
0.002 48
0.441
and linea terminalis length
a
b
Abbreviations: L1 to L5 М ?rst to ?fth lumbar vertebrae; S1 М ?rst sacral vertebra.
Spearman?s correlation coef?cient presented because n < 25.
TABLE 4. Absolute and relative costal process length of ?rst sacral vertebra (S1) in three australopithecinesa
Specimens
AL 288-1
Sts 14
Stw 431
Breadth of
sacrum at
S1 (mm)
Transverse
diameter of
body of S1 (mm)
Costal process
length of
S1 (mm)
Relative costal
process length
of S1 (%)
91.3
76.0
89.5
34.3
27.0
37.5
28.5
24.5
26.0
62.4
64.5
58.1
a
Costal process length of S1 М (breadth of sacrum at S1transverse diameter of body of S1)/2; relative costal process length of S1 М
(2(costal process length of S1)100)/breadth of sacrum at S1. Data on breadth of sacrum at S1 and transverse diameter of body of S1 are
from Robinson (1972), Johanson et al. (1982), and Toussaint et al. (2003). The term ??breadth of sacrum at S1?? is used here rather than
??maximum diameter between costal processes at S1?? [see Eq. (1) in text] because Robinson (1972), Johanson et al. (1982), and Toussaint
et al. (2003) are not speci?c as to whether the measurement was on the costal or transverse process part of the ala.
White, 1979), and considered to be female (Johanson and
Edey, 1981; Berge et al., 1984; Tague and Lovejoy, 1986,
1998; McHenry, 1992; but see Ha?usler and Schmid, 1995,
1997)]; Sts 14 [assigned to A. africanus (Howell, 1978; but
see Robinson, 1972), and considered to be female (Leutenegger, 1972; Robinson, 1972; McHenry, 1992)]; and Stw
431 [assigned to A. africanus, and considered to be male
(Toussaint et al., 2003)]. Although Stw 431 has an absolutely
longer costal process of S1 than Sts 14, the two female specimens have relatively longer costal processes of S1 than the
male: Sts 14, 64.5%; AL 288-1, 62.4%; and Stw 431, 58.1%.
Compared with the combined sample of blacks and whites
in this study, Sts 14?s index of relative costal process length
of S1 is higher than that of 100% of males and 91.9% of
females; AL 288-1?s index is higher than that of 98.9% of
males and 81.4% of females; and Stw 431?s index is higher
than that of 91.5% of males and 38.4% of females.
DISCUSSION
The results show a decreasing gradient of difference
between blacks and whites from L1 to L5. For the 7 variables for each lumbar vertebra, blacks differ signi?cantly
from whites in both sexes for 5 variables for L1, 3 for
L2, 1 for L3, and zero for L4 and L5. For S1, the two
ethnic groups differ signi?cantly for three of six variables. For costal process length, whites are signi?cantly
larger than blacks in both sexes for L1 and S1, whereas
the two ethnic groups are not signi?cantly different from
one another for L2 to L5. Whites are larger than blacks
for all four pelvic dimensions. The results of the comparison of blacks and whites in this study are principally in agreement with those of previous studies
(see above), with the exception that Flander (1978)
reported that blacks have a wider body of S1 than
whites, whereas the two ethnic groups in this study do
not differ signi?cantly. Interpreting the obstetrical signi?cance of differences between black and white females
in pelvic size is beyond the scope of this study. However,
of notable interest is that about half of the difference
between these two groups of females in the transverse
diameter of the inlet (11.8 mm) is due to the difference
in length of the costal processes of S1 (6 mm). As blacks
and whites do not differ signi?cantly from one another
in breadth of the body of S1, the remaining difference
between the two groups of females in the transverse di-
American Journal of Physical Anthropology?DOI 10.1002/ajpa
402
R.G. TAGUE
ameter of the inlet must be due to a difference in lateral
?are of the ilia.
Males generally have larger lumbar and ?rst sacral vertebrae than females. This study shows that males are
larger than females in: 1) anteroposterior and transverse
diameters of the bodies of L1 to S1 (also see Flander, 1978;
MacLaughlin and Oldale, 1992; Fang et al., 1994); 2) diameter between costal processes and width of the pedicle of
L1 to L5 (but see Kim et al., 1994 for lack of sexual dimorphism in pedicle width of L1, L2, and L4 in Koreans); and
3) costal process length of L2 and L3. Fang et al. (1994)
also reported that males are larger than females in the
height of the posterior aspect of the bodies L1 to L5.
Among the vertebral metric measures in this study, costal
process length of S1 is unique?females are larger than
males. Females also have relatively longer costal processes
than males for L1 and S1. Therefore, sexual dimorphism
in costal process length of the lumbar vertebrae shows neither a gradient from upper to lower vertebrae nor a uniform direction of dimorphism different from that of S1.
Rather, males have a longer costal process than females
for L2 and L3, but the sexes are not different from one
another for L1 and L5. Sexual monomorphism in costal
process length of L1 is unexpected for two reasons. First,
though both the ?rst thoracic vertebra (T1) and L1 are the
leading vertebrae in their respective anatomical series (i.e.,
thoracic and lumbar segments), costal process length of T1
differs from that of L1 by being sexually dimorphic, with
males having a longer process than females (Lanier, 1944).
Second, the costal processes for the twelfth thoracic vertebra (T12; Lanier, 1944) and L2 are also sexually dimorphic,
with males having a longer process than females. Moreover, the sexual monomorphism for the costal process of
L1, interposed between the sexual dimorphisms for the
costal processes of T12 and L2, fails to support an interpretation that the sexual monomorphism for the costal process
of L5 represents a transition in dimorphism for the costal
process from males being larger than females for L2 and
L3, to females being larger than males for S1. Finally,
though the sacral ala (i.e., costal process of S1) is unique
among the vertebral measures in this study in that
females are larger than males, this result does not resolve
the issue of whether the ala is homologous with the costal
processes of pre-sacral vertebrae or is new to the sacrum.
As the ?rst sacral vertebra is part of the true pelvis,
its long costal process in females implies that this sexual
dimorphism is related to obstetrics. The costal process of
S1 contributes to both the transverse diameter and circumference of the pelvic inlet3. The sacrum undeniably
3
The anatomical correlates of the costal process of S1 with other
dimensions of the pelvis may be limited to the inlet. For this study,
the only measurements taken of the pelvis were those of the inlet.
However, costal process length of S1 is highly, positively correlated
with the maximum diameter between costal processes (i.e., sacral
breadth). Based on data of blacks and whites (ages 25?49) collected
by this researcher for previous studies, sacral breadth is not signi?cantly correlated with the transverse diameter of the midplane (i.e.,
diameter between the ischial spines), and the correlation with the
transverse diameter of the outlet (i.e., diameter between the ischial
tuberosities) is signi?cant only in white females. The correlations of
sacral breadth with the transverse diameters of the midplane and outlet are as follows (r М Pearson?s correlation coef?cient, and rs М Spearman?s correlation coef?cient): black males, rs М 0.319, P М 0.184, n М
19, and r М 0.250, P М 0.120, n М 40, respectively; black females, r М
0.122, P М 0.471, n М 37, and r М 0.174, P М 0.128, n М 78, respectively; white males, rs М 0.432, P М 0.333, n М 7, and r М 0.190, P М
0.281, n М 34, respectively; and white females, rs М 0.453, P М 0.090,
n М 15, and r М 0.335, P М 0.004, n М 73, respectively.
contributes to the transverse diameter of the inlet by
separating the ilia. Agenesis of the sacrum is associated
with close approximation of the posterior aspects of the
ilia and narrowing of the transverse diameter of the
inlet (Hamsa, 1935; Del Duca et al., 1951; Katz, 1953;
Dassel, 1961; Russell and Aitken, 1963). This study
shows that for S1, both the costal process and the diameter between the costal processes (i.e., sacral breadth) are
positively correlated with the transverse diameter of the
pelvic inlet. Although the transverse diameter of the
body of S1 is a component of sacral breadth4 (Fig. 2b), it
generally is not signi?cantly correlated with the transverse diameter of the inlet. In fact, costal process
length of S1 and the transverse diameter of the body of
S1 are statistically independent from one another (and,
more generally, costal process length and transverse diameter of the body are independent from one another
for L1 to L5). For S1, the sexual dimorphisms are in
the opposite direction for costal process length and
transverse diameter of the body ? longer in females for
the former, and longer in males for the latter. The
result is that the sexes are not signi?cantly different
from one another in sacral breadth. However, this sexual monomorphism in sacral breadth belies the obstetrically important sexual dimorphism in costal process
length of S1.
The obstetrical importance of the costal process of S1
is implied by its high index of sexual dimorphism (computed as, female mean(100)/male mean): 111.0 in blacks
and 109.6 in whites. The magnitude of this index
shows the costal process of S1 to be one of the most
highly dimorphic measures of the pelvis (Tague, 1992).
By way of comparison, absolute or relative pubic length
is often celebrated as the metric trait or index that
most effectively discriminates the sexes from one
another and represents a principal obstetrical adaptation (Washburn, 1942, 1948, 1949; Dunmire, 1955;
Lovell, 1965; Black, 1970; Gingerich, 1972; Leutenegger, 1973; Ridley, 1995). For example, Last (1978;
author?s emphasis) stated that, ??The surest single feature (to determine sex) is . . . that the distance from the
pubic tubercle to the acetabular margin is greater than
the diameter of the acetabulum in the female, equal or
less in the male bone,?? and Ridley (1995) asserted that,
??the length of the pubis divided by the length of the
ischium . . . shows the greatest sexual dimorphism of
all measures of the pelvis: it is always larger in
females than in males.?? However, the index of sexual
dimorphism for costal process length of S1 is higher
than that for pubic length in both blacks (101.5) and
whites (104.3).
In addition to the transverse diameter of the inlet, the
costal process of S1 contributes to the overall capacity of
the inlet (e.g., circumference and area), which may be
the inlet?s most important obstetrical dimension (Ince
and Young, 1940; Allen, 1947; Mengert, 1948). In this
study, females are larger than males in pelvic inlet circumference by 10.1 mm in blacks and 21.2 mm in
4
The contribution of the body of S1 to the diameter between the
costal processes (i.e., sacral breadth) is seen in the following index:
relative breadth of body of S1 М maximum transverse diameter of
body of S1(100)/maximum diameter between costal processes of S1.
The mean indices for the samples in this study are 47.8 in black
males, 41.8 in black females, 45.3 in white males, and 39.5 in white
females.
American Journal of Physical Anthropology?DOI 10.1002/ajpa
COSTAL PROCESS OF SACRUM: SEXUAL DIMORPHISM
5
whites . The right and left costal processes of S1 contribute 6.4 and 6.2 mm to this sexual difference in inlet circumference in blacks and whites, respectively. Importantly, the components of inlet circumference?costal
process of S1, breadth of body of S1, and linea terminalis?are relatively independent from one another in their
sizes. For the females in this study, the coef?cients of
determination (r2, which represents the proportion of
variation in one measure that is associated with variation in another measure; see correlation coef?cients in
Table 3, with r2 М 0 when r is not signi?cant) are: 1)
zero for both blacks and whites for costal process of S1
and breadth of the body of S1; 2) 0.21 in blacks and 0.12
in whites for costal process of S1 and the linea terminalis; and 3) 0.19 in blacks and zero in whites for breadth
of the body of S1 and the linea terminalis. The consequence of this relative independence among these measures is that variability in inlet circumference is low.
Tague (1992) explained the mathematical basis for this
relationship, and he showed that inlet circumference is
the least variable dimension among females in a suite of
24 pelvic measurements. The importance of this low
variability is that obstetrical insuf?ciency in inlet circumference should be an infrequent occurrence within a
population (this inference is based on the reasonable
assumption that the mean or median inlet circumference
among females is obstetrically suf?cient).
Although the developmental biology of sexual dimorphism in costal process length of S1 is not known, it
likely differs, in part, from that of the pubis and linea
terminalis. First, females are larger than males for
lengths of the costal process of S1 and pubis (whites, not
blacks, show sexual dimorphism of the pubis in this
study, but see Washburn (1948)). Sexual dimorphism in
absolute size results from sexual differences in rate and/
or duration of growth. Ossi?cation and fusion of the auricular epiphysis of the sacrum, with fusion precluding
further growth of the costal process of S1, begin at
an earlier age in females than males (Johnston, 1961;
Bollow et al., 1997). Therefore, sexual dimorphism in
the costal process of S1 must be due to a higher growth
rate in females than males. Sexual dimorphism in pubic
length, however, is due to both a higher growth rate
(Coleman, 1969; LaVelle, 1995) and a longer period of
growth (Tague, 1994; but see Fuller, 1998) in females
than males. Second, lengths of the pubis (Rosenberg,
1988) and linea terminalis (Tague, 2000) are positively
associated with femoral head diameter in both sexes
and, by inference, with body mass because femoral head
5
The difference between blacks and whites in magnitude of sexual
dimorphism in pelvic inlet circumference is not unusual; it is also
seen in other comparisons (Tague, 1992). The principal basis for this
difference in the present study is that sexual dimorphism in curved
length of the linea terminalis is greater in whites than blacks (10.8
and 5.5 mm, respectively, with females being larger than males).
The linea terminalis is comprised of iliac and pubic components,
with the iliopubic eminence representing the area of fusion of the
ilium with the pubis. However, this eminence is not a discrete point
on the linea terminalis permitting exact delineation between pubis
and ilium. Nevertheless in whites, females have a longer straight
length of the pubis than males by 3.0 mm; in blacks, the sexes are
not signi?cantly different in pubic length. Inferentially, females
have a longer curved length of the ilium than males in both blacks
and whites, with the magnitude of sexual dimorphism being higher
in whites than blacks. Straus (1927) showed that females have a
longer straight length of the iliac part of the linea terminalis than
males, with the sexual difference being 5.4 mm in blacks and 7.2
mm in whites.
403
diameter is positively correlated with body mass (Ruff
et al., 1991; McHenry, 1992; Walrath and Glantz, 1996).
However, costal process length of S1 may not be associated with body mass. McHenry (1992) showed that body
mass is positively associated with the product of the
anteroposterior and transverse diameters of the superior
aspect of the body of S1. However in the present study,
costal process length of S1 is not signi?cantly associated
with the product of these two diameters: black males
(r М 0.156, P М 0.294, n М 47); black females (r М 0.030,
P М 0.838, n М 48); white males (r М 0.170, P М 0.271,
n М 44); and white females (r М 0.180, P М 0.332, n М
31). Therefore, though Rosenberg (1988) demonstrated a
positive association among populations between the magnitudes of sexual dimorphism in pubic length and body
mass, the present study suggests no such relationship
among populations between dimorphisms in the costal
process of S1 and body mass.
Evolution of sexual dimorphism in costal process
length of S1 in humans is speculative. Toussaint et al.
(2003:221) noted a number of differences among three
australopithecine sacra?AL 288-1, Sts 14, and Stw
431?and stated that ??it is possible that most of the sacral variations . . . may be accounted for by sexual dimorphism.?? Toussiant et al. (2003:221) compared the australopithecines with respect to the ??transverse process element comprising the dorsal part of the ala,?? but they did
not discuss the costal process of S1. Stw 431 has an
absolutely longer costal process of S1 than Sts 14. However, Sts 14 may have been a sub-adult, and its sacrum
may have become broader with advancing age (Berge
and Gommery, 1999). AL 288-1 and Sts 14 do have relatively longer costal processes of S1 than Stw 431. This
sexual difference in relative costal process length of S1
may be an artifact of small sample size because A. afarensis and A. africanus were likely sexually monomorphic
in this index. Tague (1991) demonstrated that the broad
sacrum in australopithecines was not requisite for
obstetrical success and, therefore, there would not have
been selection for differential elongation of the costal
process of S1 in females. Nevertheless, AL 288-1, Sts 14,
and Stw 431 do have high indices of relative costal process length of S1 compared with humans. Tague and
Lovejoy (1986) argued that the functional signi?cance of
the broad sacrum of australopithecines was to maintain
capacity of the false pelvis concomitant with other pelvic
changes associated with bipedality. As australopithecines
have a relatively small body of S1 (McHenry, 1992;
Haeusler, 2002), elongation of the costal processes of S1
in both sexes may have been the anatomical solution to
develop a broad sacrum.
Homo is distinguished from Australopithecus in having larger adult cranial capacity. This difference in adult
cranial capacity likely was coincident with a difference
in fetal cranial capacity, though the two differences may
not have been proportional. An increase in fetal cranial
size in Homo would have led to selection for enlarged
adult female pelvic capacity, including elongation of the
costal process of S1. For earlier Homo, there are two
sacra of adult males: Pelvis 1 from Sima de los Huesos,
Spain (dated to the Middle Pleistocene and likely ancestral to Neandertals; Arsuaga et al., 1999), and Kebara 2
from Kebara, Israel (dated to the Late Pleistocene and a
Neandertal; Rak and Arensburg, 1987). The breadth of
both sacra (measured at the plane of the linea terminalis, which is highly, positively correlated with breadth at
the level of the superior aspect of S1 as in the present
American Journal of Physical Anthropology?DOI 10.1002/ajpa
404
R.G. TAGUE
study) is within the one standard deviation range of
modern human males (Tague, 1992; Arsuaga et al., 1999).
Based on overall pelvic size, Arsuaga et al. (1999) concluded that Pelvis 1?s ??birth canal could be easily negotiated by a fetus of modern human dimensions at term.?? As
these European Middle Pleistocene hominins had smaller
adult cranial capacities than Neandertals and modern
humans (Arsuaga et al., 1999) (and, perhaps, smaller fetal
cranial capacities), females in this Middle Pleistocene population may have had similarly spacious pelves as the
males, and there may not have been selection for females
to have a longer costal process of S1 than males. As ??Neanderthal encephalization was greater than in their (European Middle Pleistocene) ancestors?? (Arsuaga et al.,
1999), and as Pelvis 1 and Kebara 2 are essentially identical in their pelvic inlet circumferences (405 and 406 mm,
respectively; Tague, 1992; Arsuaga et al., 1999), Neandertal females may have been subject to stronger selection
than their ancestors for enlargement of their pelves,
including differential elongation of their costal process of
S1 compared with males. Nevertheless, the current fossil
record is insuf?cient to identify which species of Homo
may have been the ?rst to show sexual dimorphism in the
costal process of S1 as a result of selection for obstetrical
success.
CONCLUSIONS
This study shows that the costal process of S1 is longer in females than males in modern blacks and whites.
This sexual dimorphism differs from that for the costal
processes of lumbar vertebrae in which either the sexes
are not signi?cantly different (L1 and L5) or males are
signi?cantly larger than females (L2 and L3). For other
vertebral measures (e.g., anteroposterior and transverse
diameters of bodies and diameters between costal processes), males typically are larger than females. The
uniquely long costal process of S1 in females likely is
associated with selection for obstetrical suf?ciency of the
pelvis. The costal process of S1 contributes to the transverse diameter and circumference of the pelvic inlet.
Among earlier hominins, differential elongation of the
costal process of S1 in females compared with males is
likely to be found in Homo, not Australopithecus.
ACKNOWLEDGMENTS
The author thanks the following individuals and institutions for allowing him to study skeletal material in
their care: David Hunt, Department of Anthropology,
National Museum of Natural History, Smithsonian Institution; and Bruce Latimer, Yohannes Haile-Selassie, and
Lyman Jellema, Department of Physical Anthropology,
Cleveland Museum of Natural History. The author
thanks Mary Lee Eggart, Department of Geography and
Anthropology, Louisiana State University, for drawing
the ?gure illustrations.
LITERATURE CITED
Allen EP. 1947. Standardised radiological pelvimetry IV. Interpretation of pelvimetry. Br J Radiol 20:205?218.
Arsuaga J-L, Lorenzo C, Carretero J-M, Gracia A, Martinez I,
Garcia N, de Castro J-MB, Carbonell E. 1999. A complete
human pelvis from the Middle Pleistocene of Spain. Nature
399:255?258.
Berge C, Gommery D. 1999. Le sacrum de Sterkfontein Sts 14Q
(Australopithecus africanus): Nouvelles donne?es sur la crois-
sance et sur l?a?ge osseux du spe?cimen (homage a? R. Broom et
J.T. Robinson). CR Acad Sci Ser IIa Sci Terre Planetes
329:227?232.
Berge C, Orban-Segebarth R, Schmid P. 1984. Obstetrical interpretation of the australopithecine pelvic cavity. J Hum Evol
13:573?587.
Black ES. 1970. Sexual dimorphism in the ischium and pubis of
three species of South American monkeys. J Mammal 51:794?
796.
Bollow M, Braun J, Kannenberg J, Biedermann T, SchauerPetrowskaja C, Paris S, Mutze S, Hamm B. 1997. Normal
morphology of sacroiliac joints in children: magnetic resonance studies related to age and sex. Skeletal Radiol 26:697?
704.
Coleman WH. 1969. Sex differences in the growth of the human
bony pelvis. Am J Phys Anthropol 31:125?151.
C?o?log?lu AS, I?s?can MY, Yavuz MF, Sari H. 1998. Sex determination from the ribs of contemporary Turks. J Forensic Sci
43:273?276.
Dassel PM. 1961. Agenesis of the sacrum and coccyx. Am J
Radiol 85:697?700.
Del Duca V, Davis EV, Barroway JN. 1951. Congenital absence
of the sacrum and coccyx. Report of two cases. J Bone Joint
Surg A 33:248?253.
Dunmire WW. 1955. Sex dimorphism in the pelvis of rodents.
J Mammal 36:356?361.
Dupras TL, Pfeiffer SK. 1996. Determination of sex from adult
human ribs. Can Soc Forensic Sci J 29:221?231.
Ericksen MF. 1976. Some aspects of aging in the lumbar spine.
Am J Phys Anthropol 45:575?580.
Ericksen MF. 1978a. Aging in the lumbar spine. II. L1 and L2.
Am J Phys Anthropol 48:241?245.
Ericksen MF. 1978b. Aging in the lumbar spine. III. L5. Am J
Phys Anthropol 48:247?250.
Fang D, Cheung KMC, Ruan D, Chan FL. 1994. Computed tomographic osteometry of the Asian lumbar spine. J Spinal
Disord 7:307?316.
Fawcett E. 1907. On the completion of ossi?cation of the human
sacrum. Anat Anz 30:414?421.
Fawcett E. 1938. The sexing of the human sacrum. J Anat
72:633.
Flander LB. 1978. Univariate and multivariate methods for sexing the sacrum. Am J Phys Anthropol 49:103?110.
Fuller K. 1998. Adult females and pubic bone growth. Am J
Phys Anthropol 106:323?328.
Gingerich PD. 1972. The development of sexual dimorphism in
the bony pelvis of the squirrel monkey. Anat Rec 172:589?
595.
Haeusler M. 2002. New insights into the locomotion of Australopithecus africanus based on the pelvis. Evol Anthropol 11
(Suppl 1):53?57.
Hamsa WR. 1935. Congenital absence of the sacrum. Arch Surg
30:657?666.
Harris HA. 1933. Ossi?cation in the lumbo-sacral region. Br J
Radiol 6:685?688.
Ha?usler M, Schmid P. 1995. Comparison of the pelvis of Sts 14
and AL 288-1: implications for birth and sexual dimorphism
in australopithecines. J Hum Evol 29:363?383.
Ha?usler M and Schmid P. 1997. Assessing the pelvis of AL 2881: a reply to Wood and Quinney. J Hum Evol 32:99?102.
Howell FC. 1978. Hominidae. In: Maglio VJ, Cooke HBS, editors. Evolution of African mammals. Cambridge, Massachusetts: Harvard University Press. p 154?248.
Ince JGH, Young M. 1940. The bony pelvis and its in?uence on
labour: A radiological and clinical study of 500 women.
J Obstet Gynaecol Br Emp 47:130?190.
I?s?can MY. 1983. Assessment of race from the pelvis. Am J Phys
Anthropol 62:205?208.
I?s?can MY. 1985. Osteometric analysis of sexual dimorphism in
the sternal end of the rib. J Forensic Sci 30:1090?1099.
I?s?can MY, Cotton TS. 1985. The effect of age on the determination of race from the pelvis. J Hum Evol 14:275?282.
Johanson DC and Edey MA. 1981. Lucy: the beginnings of
humankind. New York: Simon and Schuster.
American Journal of Physical Anthropology?DOI 10.1002/ajpa
COSTAL PROCESS OF SACRUM: SEXUAL DIMORPHISM
Johanson DC, Lovejoy CO, Kimbel WH, White TD, Ward SC,
Bush ME, Latimer BM, Coppens Y. 1982. Morphology of the
Pliocene partial hominid skeleton (A.L. 288-1) from the Hadar
Formation, Ethiopia. Am J Phys Anthropol 57:403?451.
Johanson DC, White TD. 1979. A systematic assessment of
early African hominids. Science 202:321?330.
Johnston FE. 1961. Sequence of epiphyseal union in a prehistoric Kentucky population from Indian Knoll. Hum Biol 33:
66?81.
Katz JF. 1953. Congenital absence of the sacrum and coccyx.
J Bone Joint Surg A 35:398?402.
Kim N-H, Lee H-M, Chung I-H, Kim H-J, Kim S-J. 1994. Morphometric study of the pedicles of thoracic and lumbar vertebrae in Koreans. Spine 19:1390?1394.
Lanier RR Jr. 1939. The presacral vertebrae of American white
and negro males. Am J Phys Anthropol 25:341?420.
Lanier RR Jr. 1944. Length of ?rst, twelfth, and accessory ribs
in American whites and negroes; their relationship to certain
vertebral variations. Am J Phys Anthropol 2(ns):137?146.
Last RJ. 1978. Anatomy: regional and applied, 6th ed. Edinburgh: Churchill Livingstone.
LaVelle M. 1995. Natural selection and developmental sexual variation in the human pelvis. Am J Phys Anthropol 98:59?72.
Leutenegger W. 1972. Newborn size and pelvic dimensions of
Australopithecus. Nature 240:568?569.
Leutenegger W. 1973. Functional aspects of pelvic morphology
in simian primates. J Hum Evol 3:207?222.
Liguoro D, Vandermeersch B, Gue?rin J. 1994. Dimensions of
cervical vertebral bodies according to age and sex. Surg
Radiol Anat 16:149?155.
Lovell AP. 1965. Bony pelvic sexual dimorphism in rabbits.
Anat Rec 151:462 (abstract).
MacLaughlin SM, Oldale KNM. 1992. Vertebral body diameters
and sex prediction. Ann Hum Biol 19:285?292.
McHenry HM. 1992. Body size and proportions in early hominids. Am J Phys Anthropol 87:407?431.
Mengert WF. 1948. Estimation of pelvic capacity. JAMA 138:
169?174.
Mitchell GAG. 1936?1937. The signi?cance of lumbosacral transitional vertebrae. Br J Surg 24:147?158.
Nutter JA. 1914. Congenital anomalies of the ?fth lumbar vertebra and their consequences. J Anat Physiol 48:24?36.
O?Rahilly R, Muller F, Meyer DB. 1980. The human vertebral
column at the end of the embryonic period proper. I. The column as a whole. J Anat 131:565?575.
O?Rahilly R, Mu?ller F, Meyer DB. 1990a. The human vertebral
column at the end of the embryonic period proper. III. The
thoracicolumbar region. J Anat 168:81?93.
O?Rahilly R, Mu?ller F, Meyer DB. 1990b. The human vertebral
column at the end of the embryonic period proper. IV. The
sacrococcygeal region. J Anat 168:95?111.
Owers SK, Pastor RF. 2005. Analysis of quantitative methods
for rib seriation using the Spital?elds documented skeletal
collection. Am J Phys Anthropol 127:210?218.
Paterson AM. 1893. The human sacrum. Sci Trans R Dubl Soc
5:123?204.
Rak Y, Arensburg B. 1987. Kebara 2 Neanderthal pelvis: ?rst
look at a complete inlet. Am J Phys Anthropol 73:227?231.
405
Ridley M. 1995. Brief communication: pelvic sexual dimorphism
and relative neonatal brain size really are related. Am J Phys
Anthropol 97:197?200.
Robinson JT. 1972. Early hominid posture and locomotion. Chicago: University of Chicago Press.
Rosenberg KR. 1988. The functional signi?cance of Neandertal
pubic length. Curr Anthropol 29:595?617.
Ruff CB, Scott WW, Liu Ay-C. 1991. Articular and diaphyseal
remodeling of the proximal femur with changes in body mass
in adults. Am J Phys Anthropol 86:397?413.
Russell HE, Aitken GT. 1963. Congenital absence of the sacrum
and lumbar vertebrae with prosthetic management. A survey
of the literature and presentation of ?ve cases. J Bone Joint
Surg A 45:501?508.
Segebarth-Orban R. 1980. An evaluation of the sexual dimorphism of the human innominate bone. J Hum Evol 9:601?607.
SPSS. 1992. SPSS/PCў. Version 5.0. Chicago: SPSS.
SPSS. 2001. SPSS for Windows. Version 11.0. Chicago: SPSS.
Standring S, editor. 2005. Gray?s anatomy: The anatomical basis
of clinical practice, 39th ed. Edinburgh: Elsevier.
Stra?dalova? V. 1974. Determination of sex from metrial characteristics of the sacrum. Folia Morphol (Praha) 22:408?412.
Straus WL Jr. 1927. The human ilium: Sex and stock. Am J
Phys Anthropol 11:1?28.
Tague RG. 1991. Commonalities in dimorphism and variability
in the anthropoid pelvis, with implications for the fossil record. J Hum Evol 21:153?176.
Tague RG. 1992. Sexual dimorphism in the human bony pelvis,
with a consideration of the Neandertal pelvis from Kebara
Cave, Israel. Am J Phys Anthropol 88:1?21.
Tague RG. 1994. Maternal mortality or prolonged growth: age
at death and pelvic size in three prehistoric Amerindian populations. Am J Phys Anthropol 95:27?40.
Tague RG. 1995. Variation in pelvic size between males and
females in nonhuman anthropoids. Am J Phys Anthropol
97:213?233.
Tague RG. 2000. Do big females have big pelves? Am J Phys
Anthropol 112:377?393.
Tague RG, Lovejoy CO. 1986. The obstetric pelvis of A.L. 288-1
(Lucy). J Hum Evol 15:237?255.
Tague RG, Lovejoy CO. 1998. AL 288-1?Lucy or Lucifer: gender confusion in the Pliocene. J Hum Evol 35:75?94.
Tatarek NE. 2005. Variation in the human cervical neural
canal. Spine J 5:623?631.
Toussaint M, Macho GA, Tobias PV, Partridge TC, Hughes AR.
2003. The third partial skeleton of a late Pliocene hominin (Stw
431) from Sterkfontein, South Africa. S Afr J Sci 99:215?223.
Trotter M. 1926. The sacrum and sex. Am J Phys Anthropol
9:445?450.
Walrath DE, Glantz MM. 1996. Sexual dimorphism in the pelvic
midplane and its relationship to Neandertal reproductive patterns. Am J Phys Anthropol 100:89?100.
Washburn SL. 1942. Skeletal proportions of adult langurs and
macaques. Hum Biol 14:444?472.
Washburn SL. 1948. Sex differences in the pubic bone. Am J
Phys Anthropol 6(ns):199?207.
Washburn SL. 1949. Sex differences in the pubic bone of Bantu
and Bushman. Am J Phys Anthropol 7(ns):425?432.
American Journal of Physical Anthropology?DOI 10.1002/ajpa
Документ
Категория
Без категории
Просмотров
2
Размер файла
333 Кб
Теги
process, dimorphic, sacra, first, sexual, vertebrate, obstetrics, adaptation, costas
1/--страниц
Пожаловаться на содержимое документа