Journal of Mammalogy, 97(2):503–517, 2016
DOI:10.1093/jmammal/gyv195
Published online December 24, 2015
Pelvic sexual dimorphism among species monomorphic in body
size: relationship to relative newborn body mass
Robert G. Tague*
Department of Geography and Anthropology, Louisiana State University, Baton Rouge, LA 70803-4105, USA
* Correspondent: [email protected]
Females have larger pelves than males among eutherians to mitigate obstetrical difficulty. This study addresses
3 issues concerning pelvic sexual dimorphism using 8 species that are sexually monomorphic in nonpelvic size:
Aotus azarae, Castor canadensis, Dasypus novemcinctus, Hylobates lar, Saguinus geoffroyi, Sciurus carolinensis,
Sylvilagus floridanus, and Urocyon cinereoargenteus. Using published data to compute the index of relative
newborn body mass (RNBM = [newborn body mass/adult female body mass]100%) for 266 eutherian species,
A. azarae, H. lar, and S. geoffroyi are characterized as giving birth to relatively large newborns and the other
5 species as giving birth to relatively small newborns. The 3 issues are, compared to species giving birth to
relatively small newborns, whether species that give birth to relatively large newborns have 1) higher magnitude
of pelvic sexual size dimorphism (SSD), 2) lower prevalence of pelvic joint fusion, and 3) dissociation between
pelvic and nonpelvic sizes. Nine measures of the pelvis were taken, and fusion of interpubic and sacroiliac joints
was observed. Species grouped by high and low RNBM do not differ significantly in magnitude of SSD of pelvic
inlet circumference. Species with high RNBM have significantly lower prevalence of interpubic joint fusion than
those with low RNBM. Sexes do not differ in their multiple correlation coefficients between inlet circumference
and nonpelvic body size in 7 of 8 species. Results suggest that 1) there are multiple anatomical pathways for
pelvic obstetrical sufficiency, 2) an unfused interpubic joint is obstetrically advantageous, and 3) relative newborn
size does not change the association between pelvic and nonpelvic size in females and males.
Key words: birth, newborn, obstetrics, pelvis
© 2015 American Society of Mammalogists, www.mammalogy.org
This study compares 8 species of eutherian mammals concerning 3 issues related to sexual dimorphism of the pelvis as
an obstetrical adaptation. Compared to metatherians, eutherians give birth to relatively large newborns (i.e., eutherians
have a higher index of newborn body mass as a percentage of
adult female body mass—Leitch et al. 1959:14–21, table 2;
Parker 1977:276, table 16.1; Tyndale-Biscoe and Renfree
1987:16–21, table 2.1). The implied obstetrical difficulty in
eutherians is expected to lead to pelvic sexual dimorphism,
with females having a larger pelvis than males as an adaptation for successful birth. Ridley (1995:197) asserted that,
“In most, probably all, mammals, the pelvis (scaled for body
size) is larger in females than males.” This study considers
the assertion of near universality of pelvic sexual dimorphism
among eutherians by comparing 1) females and males within
species and 2) females among species grouped by relative
newborn size to address 3 issues: 1) magnitude of sexual size
dimorphism (SSD) of pelvis, 2) prevalence of pelvic joint
fusion, and 3) association between pelvic size and nonpelvic
body size.
In a sample of 114 eutherian species from Leitch et al.
(1959), newborn body mass scales to the power of 0.83 with
adult female body mass; these species show a range in relative
newborn body mass (RNBM) from 0.23% for Ursus maritimus
(polar bear) to 34.5% for Rhinolophus ferrumequinum (greater
horseshoe bat). Several researchers have suggested that RNBM
is associated with degree of obstetrical difficulty and magnitude of pelvic SSD (Leutenegger 1974; St. Clair 2007; DeSilva
2011).
Eutherians also show a wide range in magnitude of nonpelvic SSD. For body mass, females are larger than males by 8%
in Plecotus auritus (brown long-eared bat) and Saccopteryx
leptura (lesser sac-winged bat), whereas males are larger than
females by 457% in Callorhinus ursinus (northern fur seal—
Ralls 1976; Weckerly 1998). Nonpelvic SSD can confound our
understanding of pelvic SSD as an obstetrical adaptation in 2
ways. First, among anthropoid primates, magnitude of nonpelvic SSD (i.e., males larger than females) is positively related to
magnitude of pelvic SSD (i.e., females larger than males). This
relationship is independent of RNBM (Schultz 1949; Tague
503
504
JOURNAL OF MAMMALOGY
2005). Second, identification of pelvic SSD can be problematic in species in which males are much larger than females in
nonpelvic size because females may have absolutely smaller,
but relatively larger, pelves than males. In Gorilla gorilla
(gorilla), for example, females are 14% smaller than males in
the transverse diameter of the pelvic inlet but are 9% larger
than males when this diameter is standardized by length and
head diameter of the femur (Tague 2005:396, 401, tables 1 and
3). Standardizing a pelvic variable by a measure of nonpelvic
size may not remove this ambiguity, however, because females
and males may differ in the allometric association between pelvic and nonpelvic measures (Schultz 1949; Tague 2000, 2005).
This study obviates the potentially confounding influence of
nonpelvic SSD on pelvic SSD by using only species that are
sexually monomorphic in nonpelvic size. Among these species,
a pelvic sexual dimorphism in which females are larger than
males can be logically interpreted as an obstetrical adaptation
(Tague 2003).
This study addresses 3 issues. The 1st issue concerns pelvic
SSD. Females are expected to have larger pelves than males
to facilitate birth. For example, females have an absolutely or
relatively longer pubis (relative length is based on the ischiopubic index, which is computed as pubic length/ischial length)
than males in a number of species of primates and rodents
(Schultz 1949; Dunmire 1955; Leutenegger 1974). A long
pubis increases the size of the pelvic inlet and, thereby, is
obstetrically advantageous. Ridley (1995:197) asserted that the
“ischio-pubic index . . . shows the greatest sexual dimorphism
of all measures of the pelvis: it is always larger in females than
in males” (emphasis added). Ridley (1995) also demonstrated
that magnitude of SSD in relative pubic length among primates
is positively associated with relative newborn cranial size.
Pubic length contributes to pelvic capacity, but it is not a
measure of capacity per se. Area and circumference are measures of pelvic capacity and are regarded as important indicators of pelvic obstetrical sufficiency (Ince and Young 1940;
Allen 1947; Mengert 1948; Bellows et al. 1971; Johnson
et al. 1988; Tague 1992). Area is estimated by multiplying the
anteroposterior diameter by the transverse diameter (Ince and
Young 1940; Allen 1947; Mengert 1948; Wiltbank and LeFever
1961); in humans, this computed area is “fairly close” in agreement with actual area (Young and Ince 1940:377). Pelvic inlet
circumference, however, can be measured directly rather than
estimated. Circumference is the focal pelvimetric in this study
because it is the actual space available to the fetus during delivery. Therefore, this study evaluates whether 1) species show
commonality in SSD of pelvic inlet circumference and 2) magnitude of SSD of pelvic inlet circumference is higher in species
that give birth to relatively large newborns compared to those
that give birth to relatively small newborns.
The 2nd issue concerns prevalence of pelvic joint fusion.
Pelvic joint mobility can be important for obstetrical success,
particularly for species that give birth to relatively large newborns (Tague 1988, 1990). Mobility of the interpubic joint
provides temporary increase in interpubic joint width and,
correspondingly, pelvic inlet circumference during delivery.
Among some species that give birth to relatively large newborns,
such as Cavia porcellus and Tadarida brasiliensis (guinea pig
and free-tailed bat, respectively), interpubic joint mobility is
requisite for successful delivery (Talmage 1947; Crelin 1969).
Mobility of the sacroiliac joint allows for sacral nutation, which
facilitates birth by increasing space in the lower pelvic planes.
Moreover, fusion of the interpubic joint precludes mobility of
the sacroiliac joint, and vice versa. Therefore, the expectations
in this study are that 1) females will have lower prevalence of
pelvic joint fusion than males within species and 2) among
females, species that give birth to relatively large newborns will
have lower prevalence of pelvic joint fusion than those that give
birth to relatively small newborns.
The 3rd issue concerns whether the sexes differ in the association between pelvic inlet circumference and nonpelvic body
size. Natural selection must be intense for the sexes to evolve
SSD (Lande 1980). For species that are either sexually monomorphic or dimorphic (with males larger than females) in nonpelvic body size, SSD of the pelvis (with females larger than
males) implies both 1) intense natural selection for obstetrical
sufficiency and 2) relative decoupling of pelvic size from nonpelvic body size. Humans give birth to big babies, and we show
these dual associations: females larger than males in pelvic
size, and low association between pelvic and nonpelvic sizes
(Tague 2000; Kurki 2011). Therefore, the expectation in this
study is that within species the multiple correlation coefficient
between pelvic inlet circumference and nonpelvic body size
will be lower in females than males.
Materials and Methods
Eight species were used in this study: Aotus azarae (Azara’s
night monkey), Castor canadensis (American beaver), Dasypus
novemcinctus (nine-banded armadillo), Hylobates lar (lar gibbon), Saguinus geoffroyi (Geoffroy’s tamarin), Sciurus carolinensis (eastern gray squirrel), Sylvilagus floridanus (eastern
cottontail), and Urocyon cinereoargenteus (gray fox). Three
criteria were used for selection of species: 1) species were
described in the literature as being sexually monomorphic
or almost monomorphic in nonpelvic size (e.g., total length,
length of long bones, or body mass, although body mass is
problematic because it can be seasonally variable; Supporting
Information S1); 2) hipbones, sacrum, or complete pelvis of
at least 10 females and 10 males were available; and 3) pelves
were large enough to be articulated and measured with calipers.
The literature is inconsistent on whether some of these species
are sexually monomorphic in nonpelvic size (e.g., U. cinereoargenteus; Supporting Information S1). Therefore, the samples of the 8 species were evaluated for sexual difference in 3
measures of nonpelvic skeletal size and body mass. Museum
records were used for information on species assignation, sex,
geographical locality, and body mass. These records did not
show that any specimen was held in captivity; by inference, all
specimens were wild caught. All specimens were adults, based
on fusion of long bone epiphyses. Table 1 shows sample sizes
for each species.
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS505
Table 1.—Sample sizes of females and males among 8 species.
Species
Aotus azarae
Castor canadensis
Dasypus novemcinctus
Hylobates lar
Saguinus geoffroyi
Sciurus carolinensis
Sylvilagus floridanus
Urocyon cinereoargenteus
Females
Males
15
27
29
29
28
40
37
33
19
20
33
28
30
39
44
36
Nominal and metrical data were collected. Nominal data
included presence or absence of fusion of the interpubic and
sacroiliac joints. Metrical data included 8 measurements of
the bony birth canal and 3 measurements of nonpelvic bones
(Fig. 1). One pelvic variable, circumference of inlet, was computed from these measurements, where Circinlet = 2(BD) + DD.
Here, Circinlet = circumference of inlet, BD = linea terminalis
length, and DD = sacral breadth (Fig. 1a). Within each species,
the index of SSD was computed for each pelvimetric variable,
where SSD = ((X F − X M ) / X M )100%. There are a number of
indices for characterizing SSD (Borgognini Tarli and Repetto
1986:56, table 2). The index used in this study expresses the
percentage deviation in size of females compared to males. This
index is appropriate for this study because the sexes are monomorphic for 4 measures of nonpelvic body size (see “Results”)
and, therefore, the sexes could be considered to be comparable
in pelvic size with each other in the absence of natural selection
on pelvic size for obstetrical sufficiency.
I articulated pelves to take measurements, applying strips
of adhesive tape or paraffin wax to the sacroiliac joints, and
encircling the pelvis with rubber bands. Pubic bones touched in
the midline; no compensation was made for an interpubic disk.
The full suite of measurements could not be taken for some
specimens due to damage to bones or ligamentous preparation.
Three pelvic measurements could not be taken for D. novemcinctus because the sacroiliac joint is fused in all specimens
(see “Results”). This fusion obliterated the boundary between
sacrum and ilium and, thereby, precluded measurement of
sacral breadth and lengths of linea terminalis and lower ilium
(Fig. 1a, D–D, B–D, and D–E, respectively). Although linea
terminalis and sacral breadth could not be measured separately
in D. novemcinctus, inlet circumference could be measured
directly.
Measurements of pelvic and nonpelvic bones were taken
with dial calipers and a curvometer. All linear measurements
were taken to the nearest 0.1 mm. The curvilinear measurement,
length of linea terminalis, was taken to the nearest 0.32 mm
(Tague 2005). Measurements were repeated on 38 specimens
(17 D. novemcinctus, 10 S. carolinensis, 7 S. floridanus, and 4
U. cinereoargenteus) from several days to several months after
the original measurements were taken. The range of the means
for measurement precision was from 0.985 for sacral breadth to
0.999 for femoral and humeral lengths, where Precision = 1 −
(|R − O|/O). Here, O = original measurement and R = repeat
measurement.
Data on body mass of newborns and adult females for 266
eutherian species, including the 8 species in this study, were
obtained from 3 sources: 1) primary literature, 2) original data
provided to me by a colleague (Supporting Information S2), and
3) museum records of specimens used in this study. Newborn
body mass in this study refers to body mass of a single individual on the day of birth to 2 days after birth. The predominance
of newborns was delivered in captivity, whereas all data on adult
female body mass are from noncaptive individuals. The index
RNBM was computed from these data (Supporting Information
S3), where RNBM = (NBM/AFBM)100%. Here, NBM = newborn body mass and AFBM = adult female body mass. Smith
and Jungers (1997:529) argued that there are “shortcomings of
some data sets” on adult body mass that are reported in the
literature. I used more than 2,900 articles and books that were
cited in compendia pertaining to reproductive biology or adult
body mass (e.g., Leitch et al. 1959; Hayssen et al. 1993; Silva
and Downing 1995; Jones et al. 2009) and found additional
references not cited in these compendia. However, I used only
primary literature in Supporting Information S3 (i.e., data from
that study and not from citation of a personal communication
or data from another study). If data on 2 or more individuals or
samples were reported in a literature source, then the median
value is used in this study. If data from 2 or more literature
sources were available for a species, then the median is used
in computation of RNBM. Median rather than mean is used to
minimize the influence of outliers. Outliers may occur for at
least 2 reasons: 1) reproductive status of adult females was not
always reported in studies, and body mass of pregnant females
would likely be higher than that for nonpregnant females; and
2) data were obtained from 769 studies, and researchers may
differ in method of data collection and reporting.
SPSS Statistics 22.0 (2013) was used for all analyses:
1) Student’s t-test to compare sexes for difference in means for
pelvic and nonpelvic variables; 2) Mann–Whitney test to compare species categorized by RNBM for difference in SSD of
pelvic variables and prevalence of pelvic joint fusion; 3) chisquare test with correction for continuity to compare sexes for
difference in prevalence of pelvic joint fusion; 4) Pearson’s
product-moment correlation coefficient to assess linear association between pelvic variables; 5) Fisher’s z-transformation
to compare females and males for difference in their multiple
correlation coefficients between pelvic inlet circumference
and 3 measures of nonpelvic body size, which are femoral and
humeral lengths and femoral head diameter (Olkin and Finn
1995; Zar 2010); body mass was not included in this analysis
because only 40% of specimens in this study (n = 193) have
data on all 4 nonpelvic variables, whereas 88% of specimens
(n = 428) have data on the other 3 nonpelvic variables; and
6) coefficient of multiple determination, adjusted for sample
size, which is the square of the multiple correlation coefficient
and which represents the proportion of variation in pelvic inlet
circumference associated with variation among 3 measures of
nonpelvic body size.
Level of significance was set at P ≤ 0.05, although Benjamini
and Hochberg’s (1995) procedure was used to adjust this level
506
JOURNAL OF MAMMALOGY
in a multiple comparison test (and this procedure is robust even
when the multiple tests are not independent of one another—
Benjamini and Yekutieli 2001). Two-tailed test of significance
was applied for analyses within species because some species may be subject to no selection pressure or weak pressure
for pelvic SSD, pelvic joint fusion, and relative dissociation
between pelvic size and nonpelvic body size. One-tailed tests
were used for analyses comparing species based on RNBM.
That is, compared to species giving birth to relatively small
newborns, those giving birth to relatively large newborns were
hypothesized to show 1) higher magnitude of pelvic SSD and
2) lower prevalence of pelvic joint fusion.
Results
Fig. 1.—Pelvis, humerus, and femur of Sylvilagus floridanus. Ventral
(a) and dorsal (b) views of articulated pelvis; medial view of left hip
bone articulated with sacrum (c); ventral views of humerus (d) and
femur (e). Dotted line represents curved length (a, B–D); all other lines
represent straight length. Measurements: 1) anteroposterior diameter
of inlet (a, A–B), sacral promontory to dorsomedial border of superior
aspect of pubic body; 2) anteroposterior diameter of outlet (c, G–H),
apex of last sacral vertebra (based on modal number of sacral vertebrae for each species) to dorsomedial border of inferior aspect of pubic
body for Castor canadensis, Dasypus novemcinctus, Sciurus carolinensis, and S. floridanus, and to dorsomedial border of superior aspect
of pubic body for Aotus azarae, Hylobates lar, Saguinus geoffroyi,
and Urocyon cinereoargentus (measurement to the inferior or superior
aspect of the pubis is based on whether the last sacral vertebra is inferior or superior to the superior border of pubis); 3) transverse diameter
of inlet (a, C–C), maximum distance between lineae terminales, with
this diameter being aligned visually to be perpendicular to the anteroposterior diameter of inlet; 4) transverse diameter of outlet (b, F–F),
between dorsomedial points of ischial tuberosities; 5) sacral breadth
(a, D–D), distance across ventral aspect of sacrum where sacrum met
linea terminales when pelvis was articulated; 6) linea terminalis (a,
B–D), dorsomedial border of superior aspect of pubic body to auricular surface of ilium; 7) pubic length (a, B–E), dorsomedial border of
superior aspect of pubic body to anterior point of lunate surface of
acetabulum; 8) lower iliac length (a, D–E), auricular surface of ilium
to anterior point of lunate surface of acetabulum; 9) maximum length
of humerus (d); 10) maximum length of femur between femoral head
and articular surface of condyle (e); and 11) maximum diameter of
femoral head (e, I–I).
Sexual monomorphism in nonpelvic body size.—I tested 8
species for differences in 4 measures of female and male nonpelvic body size (lengths of femur and humerus, femoral head
diameter, and body mass). Results show that the sexes are not
significantly different from each other for all nonpelvic variables in all species (sample sizes for body mass for A. azarae
and U. cinereoargenteus are too small to test for a difference
between the sexes; Table 2).
Relative newborn body mass.—Each of the 8 species in
this study is characterized as giving birth to relatively small
or large newborns based on their index of RNBM. Supporting
Information S3 presents RNBM for 266 species. For these
266 species, RNBM ranges from 0.3% for Ursus americanus
(American black bear) to 46.4% for Rhinolophus comutus (little Japanese horseshoe bat). When the frequency distribution is
subdivided by quartiles, which are 3.275%, 6.6%, and 10.1%,
5 species in this study are in the lower quartile (D. novemcinctus [1.7%], C. canadensis [1.8%], U. cinereoargenteus [2.4%],
S. floridanus [2.9%], and S. carolinensis [3.1%]), and 3 species are in the upper-middle quartile (S. geoffroyi [7.8%], H. lar
[8.6%], and A. azarae [9.6%]). Based on these results, 5 species (C. canadensis, D. novemcinctus, S. carolinensis, S. floridanus, and U. cinereoargentsus) are classified as giving birth
to relatively small newborns, and 3 species (A. azarae, H. lar,
and S. geoffroyi) are classified as giving birth to relatively large
newborns.
SSD of pelvis.—The 1st issue concerns whether 1) species
show commonality in SSD of the pelvis and 2) species that give
birth to relatively large newborns have higher magnitude of
SSD compared to those that give birth to relatively small newborns. Table 3 presents summary statistics and indices of SSD
for pelvic inlet circumference and the 4 measurements that contribute to inlet circumference in each species, sacral breadth,
linea terminalis, lower iliac length, and pubic length. As discussed above, inlet circumference is the focal variable for pelvimetric analysis of sexual dimorphism. However, pubic length
is also evaluated for sexual dimorphism because Ridley (1995)
asserted that it is absolutely or relatively longer in females than
males in all mammals. (Summary statistics and indices of SSD
for the anteroposterior and transverse diameters of the pelvic
inlet and outlet in each species are presented in Supporting
Information S4.) Results show that females are significantly
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS507
Table 2.—Summary statistics for females and males of 4 nonpelvic variables among 8 species; body mass in g, all other variables in mm;
Student’s t-test (t) used with 2-tailed test of significance.a
Species
Variables
Females
X
Aotus azarae
Castor canadensis
Dasypus novemcinctus
Hylobates lar
Saguinus geoffroyi
Sciurus carolinensis
Sylvilagus floridanus
Urocyon
cinereoargenteus
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
Femoral length
Humeral length
Femoral head diameter
Body mass
102.7
80.6
8.0
1,204.5
104.7
83.8
16.9
20,257.7
85.0
62.5
11.8
4,557.1
202.9
233.0
16.0
5,346.0
68.2
54.2
6.5
513.4
53.4
41.0
5.4
413.0
79.0
61.7
7.2
1,248.6
117.8
103.5
11.0
3,912.0
Males
SD
n
2.6
2.1
0.3
81.3
5.4
4.0
0.9
3,861.6
3.2
2.4
0.5
696.2
7.4
9.5
0.7
606.6
1.9
1.8
0.2
71.9
1.7
1.4
0.2
38.0
2.7
2.6
0.4
171.6
4.8
3.8
0.5
14
12
14
2
26
26
27
14
27
29
28
11
29
26
29
21
26
25
28
27
38
39
40
23
33
35
35
7
31
30
33
1
X
102.5
79.7
7.9
1,241.0
107.5
85.9
17.2
20,955.0
84.9
63.0
11.9
4,428.5
205.3
238.1
16.3
5,860.4
67.9
54.3
6.6
484.5
53.2
40.9
5.4
398.4
79.4
62.2
7.3
1,155.9
119.1
105.4
11.2
3,872.0
SD
n
2.9
1.5
0.3
48.1
6.8
4.4
1.0
2,951.6
3.3
2.7
0.7
706.3
8.5
10.8
0.7
734.0
2.1
1.9
0.3
35.0
1.8
1.6
0.3
55.7
3.2
2.6
0.5
130.0
5.7
4.8
0.6
469.3
17
17
19
3
19
19
19
12
29
33
32
13
26
27
26
25
27
25
30
30
38
39
39
16
40
41
43
10
35
35
35
6
t
P
0.124
1.419
0.837
0.902
0.167
0.409
−1.521
−1.625
−1.063
−0.510
0.050
−0.689
−0.477
0.447
−1.133
−1.806
−1.649
−2.559
0.492
−0.262
−1.249
1.961
0.531
0.514
0.370
0.977
−0.548
−0.804
−0.924
1.272
−1.051
−1.709
−1.556
0.136
0.111
0.294
0.615
0.960
0.494
0.635
0.659
0.262
0.077
0.105
0.014
0.625
0.795
0.217
0.055
0.597
0.608
0.713
0.335
0.585
0.424
0.359
0.223
0.297
0.092
0.125
Benjamini and Hochberg (1995) procedure used to adjust probability level for multiple comparison; probability level for body mass in H. lar is not significant,
with significant P ≤ 0.0125.
a
larger than males for inlet circumference for 7 of 8 species;
U. cinereoargenteus shows no significant difference between
the sexes. Females have a significantly longer pubis than males
in 6 species (C. canadensis, D. novemcinctus, H. lar, S. geoffroyi, S. carolinensis, and S. floridanus), but the sexes are not
significantly different from each other in 2 species (A. azarae
and U. cinereoargenteus). This study also evaluates whether
the 3 species that give birth to relatively large newborns have
higher indices of SSD of inlet circumference and pubic length
compared to the 5 species that give birth to relatively small
newborns. For both variables, results are in the expected direction but are not significant (rank 1 is assigned to the lowest
index of SSD in the Mann–Whitney tests): 1) inlet circumference, Mann–Whitney U = 5, P = 0.286, 1-tailed test of significance, with mean ranks being 5.33 and 4.00 for species that
give birth to relatively large newborns and those that give birth
to relatively small newborns, respectively; and 2) pubic length,
Mann–Whitney U = 6, P = 0.393, 1-tailed test of significance,
with mean ranks being 5.00 and 4.20 for species that give birth
to relatively large newborns and those that give birth to relatively small newborns, respectively.
Pelvic joint fusion.—The 2nd issue concerns whether 1) species show commonality in sexual difference in prevalence of
pelvic joint fusion (i.e., interpubic and sacroiliac joints) and
2) among females, species that give birth to relatively large
newborns have lower prevalence of pelvic joint fusion compared to those that give birth to relatively small newborns.
Comparison between sexes among the 8 species shows 1 significant difference for prevalence of pelvic joint fusion, wherein
females have significantly lower prevalence of interpubic joint
fusion than males in H. lar (Table 4). D. novemcinctus is unique
among the species in this study because all females and males
have fused interpubic and sacroiliac joints. In contrast, only 1
individual (1 female in U. cinereoargenteus) among the other
7 species (n = 418) shows fusion of the sacroiliac joint. For
the interpubic joint and with the exception of D. novemcinctus, prevalence of fusion in females ranges from 0% (A. azarae
and S. geoffroyi) to 80% (C. canadensis) and in males from
0% (A. azarae and S. geoffroyi) to 89% (U. cinereoargenteus).
Comparison among species shows that females that give birth
to relatively large newborns have significantly lower prevalence
of interpubic joint fusion compared to those that give birth to
508
JOURNAL OF MAMMALOGY
Table 3.—Summary statistics for females and males of 5 pelvic variables (in mm) among 8 species and index of sexual size dimorphism (SSD);
Student’s t-test (t) used in analysis of sexual difference for 2 variables with 2-tailed test of significance; significant results are given in bold.a
Species
Aotus azarae
Castor canadensis
Dasypus novemcinctus
Hylobates lar
Saguinus geoffroyi
Sciurus carolinensis
Sylvilagus floridanus
Urocyon
cinereoargenteus
Variables
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Circumference of inlet
Sacral breadth
Linea terminalis
Lower iliac length
Pubic length
Females
Males
X
SD
n
X
SD
n
100.2
23.4
38.4
24.9
17.3
250.2
56.5
96.8
35.3
65.7
181.8
3.6
1.5
1.6
1.7
0.9
16.5
4.3
6.6
3.7
3.8
8.7
14
14
14
14
15
25
25
26
27
26
29
93.5
22.2
35.6
24.0
16.6
239.0
54.6
91.7
36.3
61.1
177.0
4.1
1.5
1.6
1.0
1.0
16.6
4.1
6.8
3.0
4.4
9.2
18
19
18
19
18
18
19
19
20
20
32
44.6
207.3
36.9
85.2
58.0
34.6
75.3
18.9
28.2
18.2
13.1
56.0
13.3
21.3
13.6
11.7
74.1
19.9
26.9
17.0
15.5
110.5
27.7
41.4
27.4
18.3
1.9
7.1
2.4
3.2
3.2
1.6
3.8
1.0
1.5
0.9
0.5
2.9
0.9
1.1
0.8
0.7
4.4
1.4
1.9
0.7
1.1
6.2
1.7
2.4
1.7
0.9
29
29
29
29
29
28
28
28
28
28
28
34
35
36
40
37
31
31
37
36
36
31
31
33
33
33
43.2
195.0
33.7
80.6
55.5
33.0
72.8
18.2
27.3
18.2
12.6
53.6
13.0
20.2
12.9
11.4
68.6
18.5
25.0
16.7
14.4
109.6
27.4
41.1
27.7
18.1
2.2
8.9
2.3
4.0
2.9
2.1
2.9
0.9
1.1
1.0
0.5
2.4
0.8
1.0
0.6
0.6
4.8
1.4
1.8
1.0
1.1
6.0
2.1
2.2
1.7
1.2
33
28
28
28
28
27
30
30
30
30
30
33
33
37
39
37
37
38
42
44
42
34
34
35
36
35
t
P
SSD
4.813
< 0.001
1.861
2.192
0.072
0.034
3.744
2.086
0.001
0.041
7.2
5.4
7.9
3.8
4.2
4.7
3.5
5.6
−2.8
7.5
2.7
2.705
5.832
0.009
< 0.001
3.308
2.844
0.002
0.006
3.768
3.675
< 0.001
< 0.001
2.174
4.950
0.033
< 0.001
4.244
0.600
< 0.001
0.551
0.836
0.406
3.2
6.3
9.5
5.7
4.5
4.8
3.4
3.8
3.3
0.0
4.0
4.5
2.3
5.4
5.4
2.6
8.0
7.6
7.6
1.8
7.6
0.8
1.1
0.7
−1.1
1.1
Benjamini and Hochberg (1995) procedure used to adjust probability level for multiple comparison.
a
relatively small newborns (Mann–Whitney U = 14, P = 0.036,
1-tailed test of significance). The corresponding comparison
between relative newborn size and prevalence of sacroiliac
fusion is not performed because only 1 specimen in a species
other than D. novemcinctus shows this fusion.
Association between pelvic size and nonpelvic body size.—
The 3rd issue concerns whether females differ from males
within species in their multiple correlation coefficients between
pelvic inlet circumference and 3 measures of nonpelvic body
size, which are femoral and humeral lengths and femoral head
diameter. Results show 1 significant difference between the
sexes in their multiple correlation coefficients (S. geoffroyi;
Table 5). The adjusted coefficient of multiple determination
ranges from 0.000 to 0.433 in females and from 0.000 to 0.699
in males (S. floridanus and D. novemcinctus, respectively, for
both ranges; adjusted coefficient of multiple determination of
0.490 in A. azarae is not significantly different from 0).
Discussion
Limitations of study.—There are 3 limitations of this study.
First, neither the 8 species used in pelvic analysis nor the 266
species for which RNBM is computed are random samples
within Eutheria. The sample of 266 species represents my most
determined effort to obtain data from primary literature on both
newborn and adult female body mass for as many species as
possible; this sample represents 5.2% of all eutherian species
and 15 of the 21 orders (Wilson and Reeder 2005).
3.135
0.026
1.254
11.027
No value reported for χ2c when all specimens show either fused or unfused pelvic joint. Benjamini and Hochberg (1995) procedure used to adjust probability level for multiple comparison.
a
19 (100%)
20 (100%)
0 (0%)
27 (100%)
30 (100%)
39 (100%)
44 (100%)
34 (100%)
0 (0%)
0 (0%)
33 (100%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
15 (100%)
27 (100%)
0 (0%)
29 (100%)
28 (100%)
39 (100%)
37 (100%)
29 (97%)
1.00
0.968
1.00
0.001
1.00
0.077
0.872
0.263
14 (100%)
5 (20%)
0 (0%)
23 (79%)
28 (100%)
31 (82%)
24 (65%)
8 (24%)
Aotus azarae
Castor canadensis
Dasypus novemcinctus
Hylobates lar
Saguinus geoffroyi
Sciurus carolinensis
Sylvilagus floridanus
Urocyon
cinereoargenteus
0 (0%)
20 (80%)
29 (100%)
6 (21%)
0 (0%)
7 (18%)
13 (35%)
25 (76%)
0 (0%)
15 (75%)
33 (100%)
19 (69%)
0 (0%)
15 (39%)
12 (28%)
32 (89%)
Not fused (%)
Fused (%)
Not fused (%)
Fused (%)
Females
19 (100%)
5 (25%)
0 (0%)
9 (31%)
30 (100%)
23 (61%)
27 (72%)
4 (11%)
0.002
P
χc
Males
2
Interpubic joint
0 (0%)
0 (0%)
29 (100%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
1 (3%)
Not fused (%)
Fused (%)
Not fused (%)
Fused (%)
Males
Females
Sacroiliac joint
χ 2c
0.004
P
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.950
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS509
Species
Table 4.—Frequency and prevalence of pelvic joint fusion in females and males among 8 species; chi-square test with correction for continuity used with 2-tailed test of significance;
significant results are given in bold.a
Of the 8 species for pelvic analysis, 3 are primates, 2 are
rodents, and one each is a carnivore, cingulate, and lagomorph.
Phylogenetic association can influence statistical results and
increase the likelihood of Type I error (Rohlf 2006). The 2
analyses for which phylogenetic correction would pertain
are those comparing species that give birth to relatively large
versus relatively small newborns for SSD of inlet circumference and prevalence of interpubic joint fusion. Albeit recognizing the important influence of phylogenetic association in
statistical analysis, this study will nevertheless consider the 8
species to be independent samples for the following reasons.
Inlet circumference is not heritable. Rather, it is a composite
anatomy of 3 traits of different developmental modules: lower
iliac length, pubic length, and sacral breadth (Pomikal and
Streicher 2010; Lewton 2012). (Although the specific site of
fusion of iliac and pubic components of linea terminalis, which
is used in computation of inlet circumference [see above] cannot be determined on an adult hipbone, lower iliac length and
pubic length are close approximations of these components.
Pearson’s product-moment correlation coefficient between
linea terminalis length and summation of lower iliac length
and pubic length is 0.998, P < 0.001, 2-tailed test of significance, n = 409 for the combined sample of the 8 species in
this study.) Whereas traits in the same module are integrated
and coevolve, traits of different modules are relatively uncorrelated with one another (Grabowski et al. 2011; Lewton
2012). For primates and rodents in this study, therefore, phylogenetic correction would need to be performed separately for
lower iliac length, pubic length, and sacral breadth, and these
3 adjusted traits would then need to be combined to estimate
inlet circumference for the phylogenetic ancestor. This procedure would be problematic. For prevalence of interpubic joint
fusion, A. azarae and S. geoffroyi are New World monkeys, but
they are classified in different taxonomic families; A. azarae
is in Aotidae and S. geoffroyi is in Cebidae (both species are
classified in Infraorder Simiiformes—Wilson and Reeder
2005). However, caution is warranted in estimating prevalence of interpubic joint fusion in their phylogenetic ancestor
because prevalences may vary widely among descendant species. For example, Cercopithecidae (Old World monkeys) and
Hominidae (great apes and humans) are taxonomic families
also in Simiiformes. Within Cercopithecidae, Trachypithecus
cristatus (silvery lutung) and Presbytis rubicunda (maroon leafmonkey) show prevalences of interpubic joint fusion of 0% and
64%, respectively, and within Hominidae, G. gorilla and Pan
troglodytes (common chimpanzee) show prevalences of 0%
and 34%, respectively (species’ prevalence is the mean of sexspecific prevalences—Lovejoy et al. 1995:35, table 2).
Second, there are 2 concerns about RNBM as an index of
potential obstetrical difficulty: 1) the predominance of the indices of RNBM is not based on data for newborn:mother dyads,
and 2) newborn head size is more often used than newborn
body mass in obstetrical analyses of humans and other primates
(Leutenegger 1974; Tague and Lovejoy 1998; Cunningham
et al. 2001). However, newborn body mass is appropriate for
this study because not all species deliver with the fetus in a
510
JOURNAL OF MAMMALOGY
Table 5.—Comparison between sexes within species of multiple correlation coefficient (R) between pelvic inlet circumference and 3 measures
of nonpelvic body size (femoral and humeral lengths and femoral head diameter) based on Fisher’s z-transformation with 2-tailed test of significance; significant results are given in bold; Adj. R2 = adjusted coefficient of multiple determination.a
Species
Aotus azarae
Castor canadensis
Dasypus novemcinctus
Hylobates lar
Saguinus geoffroyi
Sciurus carolinensis
Sylvilagus floridanus
Urocyon cinereoargenteus
Female
Male
n
R
Adj. R
11
23
27
26
23
31
27
27
0.802
0.635
0.706
0.591
0.556
0.769
0.329
0.568
0.490
0.308
0.433
0.261
0.200
0.546
0.000
0.234
P
2
0.054
0.018
0.001
0.022
0.066
< 0.001
0.443
0.028
Female versus male
n
R
Adj. R
P
14
17
28
25
22
32
31
32
0.562
0.796
0.856
0.782
0.822
0.590
0.306
0.839
0.111
0.549
0.699
0.557
0.621
0.278
0.000
0.672
0.111
0.004
< 0.001
< 0.001
< 0.001
0.007
0.438
< 0.001
2
Z
P
−0.970
−1.396
−1.244
0.339
0.169
0.220
1.283
0.205
−2.076
0.043
Benjamini and Hochberg (1995) procedure used to adjust probability level for multiple comparison; probability level for difference between females and males
in U. cinereoargenteus is not significant, with significant P ≤ 0.010; Z not computed for S. geoffroyi because difference between sexes in R is significant based on
R being significant in males but not significant in females; females and males are not tested for difference in A. azarae and S. floridanus because both sex-specific
Rs are not significant. Adj. R2 < 0.000 is reported as 0.000 for both sexes in S. floridanus.
a
cephalic presentation (Hamilton 1941; Lang 1967; Gowda
1969; Kurta and Stewart 1990; Campagna et al. 1993). More
practically, there are far more data in the published literature on
newborn body mass than newborn head size and, at least in primates, neonatal body mass is closely associated with neonatal
brain mass (r = 0.98, n = 26 species, P < 0.001—Harvey and
Clutton-Brock 1985:562–566, table 1).
Third, Schutz et al. (2009) report that a single completed pregnancy and relative newborn mass, but not relative litter mass,
are significantly associated with size and shape of the pelvis in
Mus musculus (house mouse; relative newborn mass and relative litter mass are the percentage of individual newborn body
mass and total litter body mass to maternal body mass, respectively). Schutz et al. (2009:838–839) caution researchers that
“parturition may be an alternative or additional source of variation . . . in situations where there is unequal sampling of parous
and nulliparous females.” For females in the pelvimetric analysis of the present study, neither parity status nor newborn size
of parous females is fully known. Museum records report number of placental scars for some specimens, but that number may
not fully account for total reproductive history (Momberg and
Conaway 1956). Also, rarely is there information in museum
records on newborn size for specific adult females. Schutz et al.
(2009) did not address whether litter size is an intervening variable between relative newborn mass and maternal pelvic size
and shape. A relationship between litter size and newborn body
mass, however, may be species-specific. For example, among
11 species in the Superfamily Muroidea (Order: Rodentia),
2 species show a significant inverse relationship between litter size and newborn body mass, but the relationship between
these variables is not significant in 9 other species (Dorney
and Rusch 1953; Cowan and Arsenault 1954; Martin 1956;
Wojciechowska 1970; Ajayi 1975; Hasler and Banks 1975;
Myers and Master 1983; Kaufman and Kaufman 1987; McShea
and Madison 1989; Sikes 1995; Scharff et al. 1999). I am not
aware of research concerning whether litter size itself is a factor in selection for obstetrical sufficiency of the maternal pelvis (i.e., number of fetuses that must pass successfully through
the birth canal at the conclusion of 1 pregnancy). Finally, there
is both inter- and intraspecific variation in litter size. Among
births reported in Supporting Information S3, H. lar gives birth
to 1 offspring, S. geoffroyi to twins, and D. novemcinctus to
quadruplets; litter size in C. canadensis, however, ranges from
1 to 6.
Interpretation of results.—Ridley (1995) stated that most,
if not all, mammals show sexual dimorphism in absolute or
relative pelvic size, with females being larger than males. This
sexual dimorphism is inferred to be an obstetrical adaptation.
As species differ in their RNBM (Leitch et al. 1959; this study;
Supporting Information S3), they are expected to differ in magnitude of pelvic SSD. One problem of previous studies on pelvic SSD is how to control for nonpelvic SSD because species,
and sexes within a species, may differ in the allometric association between pelvic and nonpelvic sizes (Schultz 1949; Tague
2000, 2005). This study obviated the potentially confounding
matter of nonpelvic SSD by using 8 species that are shown here
to be sexually monomorphic for 4 measures of nonpelvic size,
which are lengths of femur and humerus, femoral head diameter, and body mass.
Metric data were collected for 9 pelvic variables (Table 3;
Supporting Information S4). However, this study focuses on
inlet circumference in analysis of SSD because it is the actual,
total space available for fetal entry into the bony birth canal.
Pubic length is also analyzed based on 1) prior documentation
that it is sexually dimorphic among primates and rodents, 2) its
ascription of being sexually dimorphic in all mammals, and
3) its inferred obstetrical importance (Dunmire 1955; Ridley
1995).
The 1st issue concerns hypotheses that 1) species will show
commonality in SSD of obstetrically important dimensions and
2) species that give birth to relatively large newborns will show
greater SSD of the pelvis compared to those giving birth to relatively small newborns. As support for the former hypothesis,
Ridley (1995) asserted that relative pubic length is significantly
longer in females than males in all mammals. Although this
study does show that species that give birth to relatively large
or relatively small newborns both show SSD in inlet circumference and pubic length, neither variable is sexually dimorphic in
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS511
all 8 species. Notable among the species that do show SSD in
inlet circumference is lack of commonality in the basis for that
dimorphism. Two examples illustrate this lack of commonality.
First, among the 3 species with the highest indices of SSD of
inlet circumference, the indices of SSD for sacral breadth and
linea terminalis length (which are the 2 components of inlet
circumference) are 1) equal in S. floridanus (7.6 for both pelvic
variables), 2) higher for linea terminalis than for sacral breadth
in A. azarae (7.9 and 5.4, respectively), and 3) lower for linea
terminalis than for sacral breadth in H. lar (5.7 and 9.5, respectively). Second, H. lar and S. geoffroyi have similar indices of
RNBM, 8.6 and 7.8, respectively, yet they differ in the relative
contributions of lower iliac length and pubic length to linea terminalis length. The indices of SSD for lower iliac length and
pubic length in H. lar are 4.5 and 4.8, respectively, but the indices are 0.0 and 4.0, respectively, in S. geoffroyi. These results
suggest that each species, or perhaps set of species related
phylogenetically, evolves its own suite of pelvic dimorphisms
in response to its obstetrical exigencies (Tague 1991). That is,
there are multiple anatomical pathways to achieve an obstetrically sufficient pelvis. Part of the pelvis that is enlarged in
females compared to males in 1 species to facilitate obstetrical
success may be constrained in its differential enlargement in
another species. There are 2 bases for constraint: 1) multiple
traits may be integrated in a developmental module, and selection for or against change in 1 trait leads to correlated change
in other traits; and 2) a trait may have more than 1 function,
which may lead to conflicting selection pressures for change
in that trait.
For example, selection for mechanical advantage in locomotion may constrain evolution of obstetrically advantageous pelvic SSD (and vice versa—Grabowski et al. 2011; Grabowski
2013). Lower iliac length approximates the lever arm between
sacroiliac and hip joints. Steudel (1984:552) explains the functional importance of this lever arm for a leaper: “The upward
forces exerted on the legs by the substrate will act on the pelvis
at the acetabulum, while the downward forces of the descending body will act on the pelvis at the sacro-iliac articulation.
The bony area through which these forces must be resisted
is the lower ilium. Shortening this region reduces the torques
produced by these two rotational forces.” Difference between
S. geoffroyi and H. lar in SSD of lower iliac length (0% and
4.5%, respectively, see above) may be due to locomotor difference. Both species are arboreal, but S. geoffroyi travels more
frequently by leaping than does H. lar. S. geoffroyi travels
principally by quadrupedal walking/running/bounding and by
leaping (43.3% and 41.5%, respectively, of locomotor activity—Garber 1991:222, table 2), whereas H. lar travels principally by brachiation and climbing (56% and 21%, respectively,
of locomotor activity), with leaping accounting for 15% of
travel (Fleagle 1976:260, table 7). The mechanical advantage
of a short lever arm between sacroiliac and hip joints in leaping
may constrain SSD in lower iliac length in S. geoffroyi (also see
Lewton 2015).
Support for the 2nd hypothesis of this issue, which is magnitude of pelvic SSD is related to relative newborn size, comes
from Ridley (1995), who showed that magnitude of SSD in
relative pubic length among primates is positively associated
with relative newborn cranial size. However, this study shows
that magnitude of SSD of inlet circumference (as well as pubic
length) is not significantly different between species that give
birth to relatively large newborns and those that give birth to
relatively small newborns. Two examples illustrate why this
hypothesized relationship is not significant. First, S. floridanus
has a low index of RNBM (2.9), yet it has the highest index
of SSD of inlet circumference (8.0). Second, S. geoffroyi gives
birth to relatively large newborns, but its index of SSD of
inlet circumference (3.4) is lower than that of 3 species that
give birth to relatively small newborns: S. carolinensis (4.5),
C. canadensis (4.7), and S. floridanus (8.0). Interpretation of
the latter result is presented conjointly with that for pelvic joint
fusion (see below).
The 2nd issue in this study concerns hypotheses about prevalence of pelvic joint fusion both between sexes within species
and among females in species giving birth to relatively large
and relatively small newborns. Unfused pelvic joints are obstetrically advantageous (Tague 1988, 1990). Pelvic joint mobility,
which is enhanced during pregnancy by the hormone relaxin,
provides temporary increase in pelvic capacity during delivery
(Hall and Newton 1947; Talmage 1947; Zarrow et al. 1961;
Zarrow and Wilson 1963; Hisaw and Hisaw 1964; O’Connor
et al. 1966; Bagna et al. 1991). Results of this study show a
dichotomy among species in sacroiliac joint fusion. Whereas
all D. novemcinctus females and males have sacroiliac joint
fusion (also see Galliari and Carlini 2015), only 0.2% (1 of
418) of individuals in the other 7 species shows this fusion.
These results suggest that sacroiliac joint fusion is rare among
species and that the universality of sacroiliac joint fusion in
D. novemcinctus may be functionally related to its fossorial
behavior (Hildebrand 1982); however, sacroiliac joint fusion
cannot be singularly ascribed to fossorial behavior because
sloths (Order: Pilosa) also show this fusion (Flower 1885),
but they are arboreal climbers. In contrast, interspecific variability in interpubic joint fusion may be obstetrically related.
For example, interpubic joint mobility is requisite for successful delivery in C. porcellus and T. brasiliensis (Talmage 1947;
Crelin 1969), which have RNBMs of 14.9 and 26.6, respectively (Supporting Information S3). Prevalence of interpubic
joint fusion in females among species in this study ranges from
0% to 100%, and this prevalence is related to RNBM, wherein
females among species that give birth to relatively large newborns have significantly lower prevalence of interpubic joint
fusion compared to those that give birth to relatively small newborns. Among the 3 species that give birth to relatively large
newborns, no female in A. azarae and S. geoffroyi has interpubic joint fusion, and H. lar females have significantly lower
prevalence of interpubic joint fusion than males. Perhaps the
reason that S. geoffroyi has a relatively low SSD of pelvic inlet
circumference (see above) is that this species has enhanced
interpubic joint mobility during parturition. However, data are
not available to test this speculation. The dimensions of the pelvis that are important for successful parturition are those that
are present at delivery. I measured pelvic size in the postmortem animal; this size, presumably, is close to the size in the
512
JOURNAL OF MAMMALOGY
antemortem, nonpregnant animal. Pelvic size reported in this
study does not account for enhanced interpubic and sacroiliac
mobility during parturition. This mobility can increase pelvic
capacity and facilitate birth. In heifers, for example, difficulty
of birth is significantly negatively associated with pelvic area
(Johnson et al. 1988); relaxin, however, leads to increase in pelvic area and easier delivery (Musah et al. 1986; Bagna et al.
1991). Whereas the sexual difference in interpubic joint fusion
in H. lar implies an obstetrical adaptation, a similar explanation for the absence of interpubic joint fusion in A. azarae
and S. geoffroyi is less certain. As neither females nor males
in A. azarae and S. geoffroyi show interpubic joint fusion, the
ultimate explanation for the unfused joint may not be related to
obstetrics. Furthermore, as both species are New World monkeys, their identity in absence of interpubic joint fusion could
be due to phylogenetic inheritance. Regardless of the ultimate
explanation for an unfused interpubic joint, it is obstetrically
advantageous.
The 3rd issue in this study concerns whether the sexes within
species differ in their multiple correlation coefficients between
inlet circumference and 3 measures of nonpelvic body size.
Only S. geoffroyi shows a significant difference between the
sexes among the 8 species. How could a species evolve a sexual
difference in inlet circumference if the sexes are monomorphic
in nonpelvic body size and if they do not differ in their multiple
correlation coefficients? The answer is that inlet circumference
is relatively dissociated in its size from nonpelvic body size. For
example, humans, who show marked pelvic SSD, also show a
low association between pelvic size and skeletal measures of
body size (Tague 2000; Kurki 2011). There can be selection
in females for increase in pelvic size without a corresponding
increase in nonpelvic body size. Among females in this study,
the adjusted coefficient of multiple determination ranges from
0.0% to 54.6% (Table 5). Notably, the 3 species that give birth
to relatively large newborns have “low” (≤ 33%—Tague 2000)
adjusted coefficients of multiple determination: A. azarae and
S. geoffroyi have coefficients not significantly different from
0, and H. lar’s coefficient is 26.1%. Therefore, selection for
pelvic obstetrical sufficiency may involve, in part, selection for
relative dissociation between pelvic size and nonpelvic body
size (also see Grabowski et al. 2011).
This study offers 2 cautions for other studies on pelvic sexual
dimorphism. First, for a pelvic trait that is sexually dimorphic
in some species, lack of dimorphism for that trait in another
species implies neither lack of obstetrical difficulty nor overall
sexual monomorphism of the pelvis for that species. Second,
conjecture about obstetrical difficulty from RNBM should
include consideration of both pelvic size and pelvic joint fusion.
Acknowledgments
I thank the following individuals and institutions for allowing
me to study skeletal material in their care: R. Baker, H. Garner,
and R. Monk, Natural Science Research Laboratory, Texas
Tech University; J. Chupasko and M. Rutzmoser, Museum of
Comparative Zoology, Harvard University; P. Freeman and
T. Labedz, University of Nebraska State Museum; M. Hafner
and R. Saunders, Museum of Natural Science, Louisiana State
University; C. Hood and N. Rios, Museum of Natural History,
Tulane University; N. Kraucunas, Milwaukee Public Museum;
B. Lundrigan, L. Abraczinskas, and C. Carmichael, Michigan
State University Museum; R. MacPhee and E. Westwig,
American Museum of Natural History; P. Myers, P. Tucker,
and S. Hinshaw, Museum of Zoology, University of Michigan;
E. Pillaert, Zoological Museum, University of WisconsinMadison; J. Purdue and T. Martin, Illinois State Museum;
E. Reitz, Georgia Museum of Natural History, University of
Georgia; K. Seymour, Royal Ontario Museum; F. Stangl,
Midwestern State University; W. Stanley, Field Museum of
Natural History; R. Thorington, Jr. and L. Gordon, National
Museum of Natural History, Smithsonian Institution; R. Timm
and T. Holmes, Natural History Museum, University of Kansas;
E. Vrba and P. Whitehead, Peabody Museum of Natural History,
Yale University; and J. Wible and S. McLaren, Carnegie
Museum of Natural History. L. Williams, M. E. Keeling Center
for Comparative Medicine and Research supported by National
Institutes of Health grant 8P40OD010938, University of Texas
MD Anderson Cancer Center, kindly provided data on newborn weights of Aotus azarae. I thank M. L. Eggart of the
Department of Geography and Anthropology, Louisiana State
University, for drawing Fig. 1. I also thank the 3 reviewers of
this manuscript for their helpful suggestions.
Supporting Information
Supporting Information S1.—Information concerning sexual
monomorphism in nonpelvic size among 8 species in this study.
Supporting Information S2.—Weight of newborn Aotus
azarae.
Supporting Information S3.—Relative newborn body mass
for 266 eutherian species.
Supporting Information S4.—Summary statistics for females
and males for 4 pelvic variables among 8 species in this study.
Supporting Information S5.—Literature Cited in Supporting
Information S1 and S3.
Literature Cited
Ajayi, S. S. 1975. Observations on the biology, domestication and
reproductive performance of the African giant rat Cricetomys gambianus Waterhouse in Nigeria. Mammalia 39:343–364.
Allen, E. P. 1947. Standardised radiological pelvimetry IV.
Interpretation of pelvimetry. British Journal of Radiology
20:205–218.
Bagna, B., C. Schwabe, and L. L. Anderson. 1991. Effect of relaxin on
facilitation of parturition in dairy heifers. Journal of Reproduction
and Fertility 91:605–615.
Bellows, R. A., R. E. Short, D. C. Anderson, B. W. Knapp, and
O. F. Pahnish. 1971. Cause and effect relationships associated with
calving difficulty and calf birth weight. Journal of Animal Science
33:407–415.
Benjamini, Y., and Y. Hochberg. 1995. Controlling the false discovery
rate: a practical and powerful approach to multiple testing. Journal
of the Royal Statistical Society B 57:289–300.
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS513
Benjamini, Y., and D. Yekutieli. 2001. The control of the false discovery rate in multiple testing under dependency. Annals of Statistics
29:1165–1188.
Borgognini Tarli, S. M., and E. Repetto. 1986. Methodological
considerations on the study of sexual dimorphism in past human
populations. Pp. 51–66 in Sexual dimorphism in living and fossil primates (M. Pickford and B. Chiarelli, eds.). Il Sedicesimo,
Florence, Italy.
Campagna, C., M. Lewis, and R. Baldi. 1993. Breeding biology of
southern elephant seals in Patagonia. Marine Mammal Science
9:34–47.
Cowan, I. McT., and M. G. Arsenault. 1954. Reproduction and
growth in the creeping vole, Microtus orgeoni serpens Merriam.
Canadian Journal of Zoology 32:198–208.
Crelin, E. S. 1969. Interpubic ligament: elasticity in pregnant freetailed bat. Science 164:81–82.
Cunningham, F. G., N. F. Gant, K. J. Leveno, L. C. Gilstrap III,
J. C. Hauth, and K. D. Wenstrom. 2001. Williams obstetrics. 21st
ed. McGraw-Hill, New York.
DeSilva, J. M. 2011. A shift toward birthing relatively large infants
early in human evolution. Proceedings of the National Academy of
Science, United States of America 108:1022–1027.
Dorney, R. S., and A. J. Rusch. 1953. Muskrat growth and litter production. Wisconsin Conservation Department, Technical Wildlife
Bulletin 8:1–32.
Dunmire, W. W. 1955. Sex determination in the pelvis of rodents.
Journal of Mammalogy 36:356–361.
Fleagle, J. G. 1976. Locomotion and posture of the Malayan siamang
and implications for hominoid evolution. Folia Primatologica
26:245–269.
Flower, W. H. 1885. An introduction to the osteology of the
Mammalia. 3rd ed. Macmillan, London, United Kingdom.
Galliari, F. C., and A. A. Carlini. 2015. Ontogenetic criteria to
distinguish vertebral types on the debated xenarthran synsacrum.
Journal of Morphology 276:494–502.
Garber, P. A. 1991. A comparative study of positional behavior in
three species of tamarin monkeys. Primates 32:219–230.
Gowda, C. D. K. 1969. Breeding the Great Indian rhinoceros
Rhinoceros unicornis at Mysore Zoo. International Zoo Yearbook
9:101–102.
Grabowski, M. W. 2013. Hominin obstetrics and the evolution of constraints. Evolutionary Biology 40:57–75.
Grabowski, M. W., J. D. Polk, and C. C. Roseman. 2011. Divergent
patterns of integration and reduced constraint in the human hip and
the origins of bipedalism. Evolution 65:1336–1356.
Hall, K., and W. H. Newton. 1947. The effect of oestrone and relaxin
on the x-ray appearance of the pelvis of the mouse. Journal of
Physiology 106:18–27.
Hamilton, W. J., Jr. 1941. Reproduction of the field mouse Microtus
pennsylvanicus (Ord). Cornell University Agricultural Experiment
Station Memoir 237:1–23.
Harvey, P. H., and T. H. Clutton-Brock. 1985. Life history variation
in primates. Evolution 39:559–581.
Hasler, J. F., and E. M. Banks. 1975. Reproductive performance and
growth in captive collared lemmings (Dicrostonyx groenlandicus).
Canadian Journal of Zoology 53:777–787.
Hayssen, V., A. van Tienhoven, and A. van Tienhoven. 1993. Asdell’s
patterns of mammalian reproduction: a compendium of speciesspecific data. Comstock Publishing, Ithaca, New York.
Hildebrand, M. 1982. Analysis of vertebrate structure. 2nd ed. John
Wiley & Sons, New York.
Hisaw, F. L., Jr., and F. L. Hisaw. 1964. Effect of relaxin on the uterus
of monkeys (Macaca mulatta) with observations on the cervix and
symphysis pubis. American Journal of Obstetrics and Gynecology
89:141–155.
Ince, J. G. H., and M. Young. 1940. The bony pelvis and its influence
on labour: a radiological and clinical study of 500 women. Journal
of Obstetrics and Gynaecology of the British Empire 47:130–190.
Johnson, S. K., G. H. Deutscher, and A. Parkhurst. 1988.
Relationships of pelvic structure, body measurements, pelvic area
and calving difficulty. Journal of Animal Science 66:1081–1088.
Jones, K. E., et al. 2009. PanTHERIA: a species-level database of
life history, ecology, and geography of extant and recently extinct
mammals. Ecology 90:2648.
Kaufman, D. W., and G. A. Kaufman. 1987. Reproduction by
Peromyscus polionotus: number, size, and survival of offspring.
Journal of Mammalogy 68:275–280.
Kurki, H. K. 2011. Pelvic dimorphism in relation to body size and
body size dimorphism in humans. Journal of Human Evolution
61:631–643.
Kurta, A., and M. E. Stewart. 1990. Parturition in the Silver-haired
Bat, Lasionycteris noctivagans, with a description of the neonates.
Canadian Field-Naturalist 104:598–600.
Lande, R. 1980. Sexual dimorphism, sexual selection, and adaptation
in polygenic characters. Evolution 34:292–305.
Lang, E. M. 1967. The birth of an African elephant Loxodonta africana at Basle Zoo. International Zoo Yearbook 7:154–157.
Leitch, I., F. E. Hytten, and W. Z. Billewicz. 1959. The maternal
and neonatal weights of some mammalia. Proceedings of the
Zoological Society, London 133:11–28.
Leutenegger, W. 1974. Functional aspects of pelvic morphology in
simian primates. Journal of Human Evolution 3:207–222.
Lewton, K. L. 2012. Evolvability of the primate pelvic girdle.
Evolutionary Biology 39:126–139.
Lewton, K. L. 2015. Allometric scaling and locomotor function in
the primate pelvis. American Journal of Physical Anthropology
156:511–530.
Lovejoy, C. O., R. S. Meindl, R. G. Tague, and B. Latimer. 1995. The
senescent biology of the hominoid pelvis. Its bearing on the pubic
symphysis and auricular surface as age-at-death indicators in the
human skeleton. Rivista di Antropologia 73:31–49.
Martin, E. P. 1956. A population study of the prairie vole (Microtus
ochrogaster) in northeastern Kansas. University of Kansas
Publications, Museum of Natural History 8:361–416.
McShea, W. J., and D. M. Madison. 1989. Measurements of reproductive traits in a field population of meadow voles. Journal of
Mammalogy 70:132–141.
Mengert, W. F. 1948. Estimation of pelvic capacity. Journal of the
American Medical Association 138:169–174.
Momberg, H., and C. Conaway. 1956. The distribution of placental scars of first and second pregnancies in the rat. Journal of
Embryology and Experimental Morphology 4:376–384.
Musah, A. I., C. Schwabe, R. L. Willham, and L. L. Anderson. 1986.
Relaxin on induction of parturition in beef heifers. Endocrinology
118:1476–1482.
Myers, P., and L. L. Master. 1983. Reproduction by Peromyscus
maniculatus: size and compromise. Journal of Mammalogy
64:1–18.
O’Connor, W. B., G. D. Cain, and M. X. Zarrow. 1966. Elongation
of the interpubic ligament in the little brown bat (Myotis lucifugus).
Proceedings of the Society for Experimental Biology and Medicine
123:935–937.
514
JOURNAL OF MAMMALOGY
Olkin, I., and J. D. Finn. 1995. Correlations redux. Psychological
Bulletin 118:155–164.
Parker, P. 1977. An ecological comparison of marsupial and placental patterns of reproduction. Pp. 273–286 in The biology of marsupials (B. Stonehouse and D. Gilmore, eds.). MacMillan Press,
London, United Kingdom.
Pomikal, C., and J. Streicher. 2010. 4D-analysis of early pelvic
girdle development in the mouse (Mus musculus). Journal of
Morphology 271:116–126.
Ralls, K. 1976. Mammals in which females are larger than males.
Quarterly Review of Biology 51:245–276.
Ridley, M. 1995. Brief communication: pelvic sexual dimorphism
and relative neonatal brain size really are related. American Journal
of Physical Anthropology 97:197–200.
Rohlf, F. J. 2006. A comment on phylogenetic correction. Evolution
60:1509–1515.
Scharff, A., S. Begall, O. Grütjen, and H. Burda. 1999.
Reproductive characteristics and growth of Zambian giant molerats, Cryptomys mechowi (Rodentia: Bathyergidae). Mammalia
63:217–230.
Schultz, A. H. 1949. Sex differences in the pelves of primates.
American Journal of Physical Anthropology 7:401–423.
Schutz, H., E. R. Donovan, and J. P. Hayes. 2009. Effects of parity on
pelvic size and shape dimorphism in Mus. Journal of Morphology
270:834–842.
Sikes, R. S. 1995. Costs of lactation and optimal litter size in northern grasshopper mice (Onychomys leucogaster). Journal of
Mammalogy 76:348–357.
Silva, M., and J. A. Downing. 1995. CRC handbook of mammalian
body masses. CRC Press, Boca Raton, Florida.
Smith, R. J., and W. L. Jungers. 1997. Body mass in comparative
primatology. Journal of Human Evolution 32:523–559.
SPSS Statistics 22.0. 2013. IBM, Armonk, New York.
St. Clair, E. M. 2007. Sexual dimorphism in the pelvis of Microcebus.
International Journal of Primatology 28:1109–1122.
Steudel, K. 1984. Patterns of allometry in the pelvis of higher primates. Journal of Human Evolution 13:545–554.
Tague, R. G. 1988. Bone resorption of the pubis and preauricular area
in humans and nonhuman mammals. American Journal of Physical
Anthropology 76:251–267.
Tague, R. G. 1990. Morphology of the pubis and preauricular area
in relation to parity and age at death in Macaca mulatta. American
Journal of Physical Anthropology 82:517–525.
Tague, R. G. 1991. Commonalities in dimorphism and variability
in the anthropoid pelvis, with implications for the fossil record.
Journal of Human Evolution 21:153–176.
Tague, R. G. 1992. Sexual dimorphism in the human bony pelvis, with a consideration of the Neandertal pelvis from Kebara
Cave, Israel. American Journal of Physical Anthropology
88:1–21.
Tague, R. G. 2000. Do big females have big pelves? American Journal
of Physical Anthropology 112:377–393.
Tague, R. G. 2003. Pelvic sexual dimorphism in a metatherian,
Didelphis virginiana: implications for eutherians. Journal of
Mammalogy 84:1464–1473.
Tague, R. G. 2005. Big-bodied males help us recognize that females
have big pelves. American Journal of Physical Anthropology
127:392–405.
Tague, R. G., and C. O. Lovejoy. 1998. AL 288-1—Lucy or Lucifer:
gender confusion in the Pliocene. Journal of Human Evolution
35:75–94.
Talmage, R. V. 1947. Changes produced in the symphysis pubis of
the guinea pig by the sex steroids and relaxin. Anatomical Record
99:91–113.
Tyndale-Biscoe, H., and M. Renfree. 1987. Reproductive physiology of marsupials. Cambridge University Press, Cambridge,
United Kingdom.
Weckerly, F. W. 1998. Sexual-size dimorphism: influence of mass
and mating systems in the most dimorphic mammals. Journal of
Mammalogy 79:33–52.
Wilson, D. E., and D. M. Reeder. 2005. Mammal species of the
world: a taxonomic and geographic reference. 3rd ed. Johns
Hopkins University Press, Baltimore, Maryland.
Wiltbank, J. N., and D. G. LeFever. 1961. Save more calves at birth.
Nebraska Experiment Station Quarterly 8:19.
Wojciechowska, B. 1970. The growth and net production in the common vole during postnatal period. Acta Theriologica 15:81–88.
Young, M., and J. G. H. Ince. 1940. A radiographic comparison of
the male and female pelvis. Journal of Anatomy 74:374–385.
Zar, J. H. 2010. Biostatistical analysis. 5th ed. Prentice Hall, Upper
Saddle River, New Jersey.
Zarrow, M. X., B. E. Eleftheriou, G. L. Whitecotten, and J. A. King.
1961. Separation of the pubic symphysis during pregnancy and
after treatment with relaxin in two subspecies of Peromyscus maniculatus. General and Comparative Endocrinology 1:386–391.
Zarrow, M. X., and E. D. Wilson. 1963. Hormonal control of the
pubic symphysis of the Skomer bank vole (Clethrionomys skomerensis). Journal of Endocrinology 28:103–106.
(Literature cited in supporting information files is presented in
Supporting Information S5.)
Submitted 15 April 2015. Accepted 20 November 2015.
Associate Editor was John Scheibe.
Appendix I
Specimens examined
The 487 specimens examined are listed by species, museum
abbreviation, specimen number, sex, Country, Department/
District/Province/State, and Province within Department for
Bolivia or County/Parish for United States of America (USA;
specific locality if Province within Department or County/
Parish not provided). Museums and their abbreviations are as
follows (Hafner et al. 1997): 1) AMNH, American Museum of
Natural History, 2) CM, Carnegie Museum of Natural History,
3) FMNH, Field Museum of Natural History, 4) ISM, Illinois
State Museum, 5) KU, University of Kansas, Natural History
Museum, 6) LSUMZ, Louisiana State University, Museum
of Natural Science, 7) MCZ, Harvard University, Museum of
Comparative Zoology, 8) MPM, Milwaukee Public Museum,
9) MSU, Michigan State University Museum, 10) MWSU,
Midwestern State University, Collection of Recent Mammals,
11) ROM, Royal Ontario Museum, 12) TTU, Texas Tech
University, Natural Science Research Laboratory, 13) TU,
Tulane University, Museum of Natural History, 14) UF, Florida
Museum of Natural History, 15) UGAMNH, University of
Georgia, Georgia Museum of Natural History, 16) UMMZ,
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS515
University of Michigan, Museum of Zoology, 17) UNSM,
University of Nebraska State Museum, 18) USNM, National
Museum of Natural History, Smithsonian Institution,
19) UWZM, University of Wisconsin-Madison, Zoological
Museum, and 20) YPM, Yale University, Peabody Museum of
Natural History.
Aotus azarae.—AMNH 211470, 211473, F, Bolivia, Beni,
Cercado; AMNH 211460, 211461, 211464, 211476, 211478,
211479, 211480, 215048, 215050, 215053, 215056, F, Bolivia,
Beni, Mamoré; AMNH 211466, 211484, F, Bolivia, Beni,
Yacuma; AMNH 211472, M, Bolivia, Beni, Cercado; AMNH
211458, 211459, 211462, 211463, 211475, 211482, 211483,
215051, 215052, 215054, 215057, 215059, M, Bolivia, Beni,
Mamoré; AMNH 211457, 211485, 211486, M, Bolivia, Beni,
Yacuma; AMNH 246659, M, Bolivia, Santa Cruz, Ichilo; CM
1982, M, Bolivia, Santa Cruz; UMMZ 124693, M, Paraguay
(Chaco Region).
Castor canadensis.—AMNH 143738, 204223, F, USA, New York
(Bear Mountain State Park); AMNH 145436, M, USA, Michigan,
Gladwin; AMNH 150136, M, USA, New York (Bear Mountain
State Park); CM 25282, F, USA, Pennsylvania, Crawford; CM
40370, M, USA, Pennsylvania, Beaver; ISM 683962, F, USA,
Illinois, Sangamon; ISM 687408, F, USA, Missouri, Jefferson;
KU 14389, F, USA, Kansas, Cloud; KU 43868, F, USA, Kansas,
Douglas; KU 43869, F, USA, Kansas, Leavenworth; KU 14403,
F, USA, Kansas, Pottawatomie; KU 14376, F, USA, Kansas,
Sherman; KU 27378, M, USA, Idaho, Custer; LSUMZ 3766, F,
USA, Louisiana, St. Helena; UMMZ 121012, F, USA, Mississippi,
Hinds; UMMZ 83971, M, USA, Alabama, Russell; MSU 7577,
M, USA, Michigan, Oakland; UGAMNH 460, M, USA, Georgia,
Columbia; USNM 484688, 484691, 484719, 484735, 484764,
F, USA, Alabama, Barbour; USNM 484537, 484540, 484670,
484731, F, USA, Alabama, Bullock; USNM 484698, 484715,
484744, F, USA, Alabama, Russell; USNM 484677, M, USA,
Alabama, Barbour; USNM 484545, 484732, M, USA, Alabama,
Bullock; USNM 484534, M, USA, Alabama, Pike; USNM
266378, 484727, 484748, M, USA, Alabama, Russell; UWZM
29801, F, USA, Wisconsin, Bayfield; UWZM 29810, F, USA,
Wisconsin, Marathon; UWZM 22800, F, USA, Wisconsin, Sauk;
UWZM 29802, M, USA, Wisconsin, Bayfield; UWZM 25673,
M, USA, Wisconsin, Dane; UWZM 29814, M, USA, Wisconsin,
Green; UWZM 22795, 29800, M, USA, Wisconsin, Marathon;
UWZM 27338, M, USA, Wisconsin, Walworth.
Dasypus novemcinctus.—AMNH 95128, F, Brazil, Pará
(Tauari); AMNH 205727, F, Uruguay, Lavalleja; AMNH
242663, F, USA, Florida, Highlands; AMNH 133357, M,
Brazil, Goyaz; ISM 687875, F, USA, Florida, Hillsborough;
KU 144598, F, USA, Arkansas, Franklin; KU 11557, F, USA,
Texas, Bee; KU 3374, F, USA, Texas, Colorado; KU 121890,
121891, 121892, F, USA, Texas, Gonzales; KU 143932,
M, USA, Kansas, Douglas; KU 143931, M, USA, Kansas,
Franklin; KU 14354, M, USA, Kansas, Sumner; KU 11588,
11559, M, USA, Texas, Bee; LSUMZ 15762, F, Costa Rica,
Alajuela; LSUMZ 28796, 29159, 29187, M, USA, Mississippi,
Hancock; MCZ 32368, M, Brazil (Tapajos River); MSU 2045,
2046, F, Belize, Belize (Crooked Tree); MSU 2047, M, Belize,
Belize (Crooked Tree); MWSU 1682, F, USA, Texas, Jack;
MWSU 1757, F, USA, Texas, Wichita; MWSU 1683, M,
USA, Texas, Archer; MWSU 1729, M, USA, Texas, Wichita;
TTU 40426, M, USA, Texas, Kimble; TU 41, 1585, 1599,
1601, 1619, 1620, 1646, 1647, 795365, F, USA, Louisiana,
Plaquemines; TU 1581, 1586, 1591, 1592, 1593, 1596, 1597,
1628, 715-326-93-37 (this identifying number is the specimen’s suite of external measurements, total length-tail lengthhind foot length-ear length, because a specimen number was
not listed on specimen tag or skeleton), 755-362-95-37 (this
identifying number is the specimen’s suite of external measurements, total length-tail length-hind foot length-ear length,
because a specimen number was not listed on specimen tag or
skeleton), M, USA, Louisiana, Plaquemines; UGAMNH 4057,
F, USA, Georgia, Brantley; UGAMNH 228, M, USA, Florida,
Columbia; UMMZ 63138, F, Belize, Cayo; UMMZ 61515, F,
Mexico, Tamaulipas; UMMZ 126293, M, Paraguay (Chaco
Region); UMMZ 54239, M, USA, Texas, Kimble; UNSM
20908, M, USA, Nebraska, Otoe; UNSM 19014, M, Nebraska,
Valley; ROM 349, F, USA, Florida, Highlands; USNM 339668,
F, Guyana (Tacatu River); USNM 244905, M, Guatemala,
Petén; USNM 53321, M, Mexico, Hidalgo; USNM 578435,
M, Panama, Bocas del Toro; YPM 5015, M, Guatemala, San
Marcos (Ocós).
Hylobates lar.—MCZ 35943, 35945, 35949, 35950, 41412,
41416, 41418, 41419, 41421, 41424, 41436, 41440, 41449,
41454, 41455, 41458, 41460, 41463, 41469, 41474, 41478,
41487, 41494, 41521, 41523, 41525, 41530, F, Thailand,
Chiang Mai (Mt. Angka); MCZ 35946, 41415, 41427, 41428,
41431, 41434, 41441, 41445, 41446, 41447, 41448, 41451,
41453, 41456, 41459, 41465, 41468, 41472, 41476, 41479,
41481, 41484, 41486, 41489, 41495, 41512, 41522, 41538,
M, Thailand, Chiang Mai (Mt. Angka); USNM 271047,
F, Indonesia (Sumatra, Atjehkungke); USNM 260590,
F, Thailand, Chiang Mai (Mt. Angka).
Saguinus geoffroyi.—MSU 22901, 22902, 22911, 22913,
22938, 22943, 22948, 22951, 22955, 22958, 22963, 22968,
22970, 22976, 22980, 22985, 22990, F, Panama, Panamá; MSU
22919, 22922, 22928, 22933, 22935, 22936, 22965, 22971,
22973, 22981, 23005, F, Panama (Canal Zone); MSU 22899,
22900, 22903, 22905, 22910, 22914, 22940, 22953, 22954,
22959, 22960, 22961, 22964, 22966, 22983, 22984, 22987, M,
Panama, Panamá; MSU 22907, 22918, 22923, 22925, 22926,
22927, 22947, 22974, 22982, 22988, 22989, 22997, 23001, M,
Panama (Canal Zone).
Sciurus carolinensis.—UGAMNH 4381, F, USA, Georgia,
Ware; ISM 689863, F, USA, Illinois, Sangamon; ISM 686321,
F, USA, Illinois, Vermilion; ISM 690267, M, USA, Illinois,
Sangamon; LSUMZ 29102, F, USA, Louisiana, Bienville;
LSUMZ 28786, 28787, 29081, 29371, 29372, 35224, F,
USA, Louisiana, East Baton Rouge; LSUMZ 36171, F,
USA, Louisiana, Iberville; LSUMZ 29090, 29106, F, USA,
Louisiana, Jackson; LSUMZ 239, 28471, F, USA, Louisiana,
Livingston; LSUMZ 26744, 27522, 28663, 28790, F, USA,
Louisiana, St. Tammany; LSUMZ 28587, F, USA, Louisiana,
Tangipahoa; LSUMZ 29112, 29127, F, USA, Louisiana,
516
JOURNAL OF MAMMALOGY
Vernon; LSUMZ 28673, F, USA, Louisiana, West Feliciana;
LSUMZ 23646, F, USA, Mississippi, Jackson; LSUMZ 34305,
F, USA, Mississippi, Lincoln; LSUMZ 28588, 28793, F, USA,
Tennessee, Shelby; LSUMZ 29105, M, USA, Louisiana,
Bienville; LSUMZ 28657, 28789, 34306, M, USA, Louisiana,
East Baton Rouge; LSUMZ 34039, M, USA, Louisiana,
Evangeline; LSUMZ 23644, 35223, 36138, M, USA, Louisiana,
Iberville; LSUMZ 29085, 29087, 29088, M, USA, Louisiana,
Jackson; LSUMZ 34038, M, USA, Louisiana, St. Charles;
LSUMZ 28792, M, USA, Louisiana, St. Tammany; LSUMZ
29124, 29125, 29126, M, USA, Louisiana, Vernon; LSUMZ
28473, M, USA, Louisiana, Washington; LSUMZ 16195,
M, USA, Louisiana, West Feliciana; LSUMZ 28589, 28590,
28675, M, USA, Tennessee, Shelby; MPM 5560, M, USA,
Wisconsin, Milwaukee; TU 4760, F, USA, Florida, Okaloosa;
TU 4830, F, USA, Louisiana, Plaquemines; TU 4358, 4362,
F, USA, Louisiana, West Feliciana; TU 4836, 4837, F, USA,
Mississippi, Claiborne; TU 4469, 4470, 4737, 4742, 4744,
4770, 4973, F, USA, Mississippi, Wilkinson; TU 4759, M,
USA, Florida, Okaloosa; TU 64, 75-800, M, USA, Louisiana,
Orleans; TU 4360, M, USA, Louisiana, West Feliciana; TU
4741, 4769, 4823, 4959, M, Mississippi, Wilkinson; USNM
347938, M, USA, Alabama, Covington; USNM 397216, M,
USA, Florida, Duval; USNM 397211, M, USA, Florida, Leon;
USNM 397180, 397182, M, USA, Georgia, Grady; USNM
347943, 347944, 347948, M, USA, Mississippi, Wilkinson.
Sylvilagus floridanus.—AMNH 131217, F, USA, Florida,
Okaloosa; AMNH 123803, M, USA, Iowa, Johnson; AMNH
137363, M, USA, Kansas, Meade; AMNH 135934, M, USA,
New Jersey, Bergen; AMNH 245055, M, USA, New York,
Suffolk; CM 106428, F, USA, Iowa, Cerro Gordo; CM 7466,
7952, M, USA, Pennsylvania, Crawford; CM 27007, 27010,
M, USA, Pennsylvania, Jefferson; CM 27011, M, USA,
Pennsylvania, Venango; FMNH 156862, F, USA, Illinois,
Cook; FMNH 108331, F, Illinois, DuPage; FMNH 106588,
106591, 106594, 106595, 106597, 106675, 106679, F, USA,
Illinois, Lake; FMNH 57168, F, USA, South Carolina,
Lexington; FMNH 154665, F, USA, Wisconsin, Brown; FMNH
156861, 167018, M, USA, Illinois, Cook; FMNH 106593,
106676, 106677, 106678, 106680, M, USA, Illinois, Lake;
FMNH 41089, M, USA, Illinois, Richland; FMNH 154664,
M, USA, Wisconsin, Brown; ISM 687880, F, USA, Florida,
Hillsborough; ISM 688446, F, USA, Missouri, Jefferson; ISM
690390, M, USA, Illinois, Jackson; ISM 687673, M, USA,
Illinois, Union; ISM 685286, M, USA, Missouri, Benton; ISM
688565, M, USA, Missouri, Jefferson; LSUMZ 29118, F, USA,
Louisiana, Bienville; LSUMZ 28653, M, USA, Tennessee,
Shelby; MWSU 15391, F, USA, Texas, Wichita; MWSU 1211,
10951, M, USA, Texas, Wichita; TU 4315, 4419, 4454, F, USA,
Texas, Wichita; TU 4958, M, USA, Mississippi, Wilkinson; TU
4420, 4437, 4442, M, USA, Texas, Wichita; UGAMNH 1868,
F, USA, Georgia, Jasper; UMMZ 81831, F, USA, Alabama,
Lee; UMMZ 79833, F, USA, Iowa, Winnebago; UMMZ 58238,
F, USA, Kansas, Kingman; UMMZ 165580, F, USA, Nebraska,
Antelope; UMMZ 68481, F, USA, Nebraska, Cherry; UMMZ
123803, F, USA, North Carolina, Dare; UMMZ 79379, F, USA,
Texas, Jeff Davis; UMMZ 83954, M, USA, Alabama, Lee;
UMMZ 81833, M, USA, Alabama, Macon; UMMZ 79832,
M, USA, Iowa, Winnebago; UMMZ 58275, M, USA, Kansas,
Kingman; UMMZ 162680, M, USA, Michigan, Livingston;
UMMZ 81432, M, USA, Oklahoma, Rogers; UNSM 21350, F,
USA, Nebraska, Dawes; ROM 3521, M, Canada, Ontario (York
Regional Municipality); USNM 564063, F, USA, Florida,
Seminole; USNM 505567, F, USA, Maryland, Washington;
USNM 564064, F, USA, Oklahoma, Harper; USNM 265539,
F, USA, Pennsylvania, Lycoming; USNM 349707, F, USA,
Virginia (Alexandria); USNM 348072, M, USA, Georgia,
Floyd; USNM 567615, M, USA, Maryland, Prince George’s;
USNM 567612, M, USA, Maryland (Patuxent Research
Refuge); USNM 564065, M, USA, Virginia, Augusta; USNM
579266, M, USA, Virginia, Northampton; USNM 567962, M,
USA, Virginia (Virginia Beach City); UWZM 21649, F, USA,
Wisconsin, Dane; UWZM 33810, F, USA, Wisconsin, Iowa;
UWZM 18589, F, USA, Wisconsin, Sauk; UWZM 21508,
27698, M, USA, Wisconsin, Dane.
Urocyon cinereoargenteus.—AMNH 4271, F, Mexico,
Chihuahua; AMNH 137028, F, USA, Arizona, Pima; AMNH
214125, F, USA, New Jersey, Monmouth; AMNH 121498, F,
USA, New Jersey, Warren; AMNH 243098, M, USA, Florida,
Highlands; AMNH 248494, M, USA, New Jersey, Bergen;
AMNH 208391, M, USA, Texas, Presidio; CM 10714, F,
USA, Colorado, Garfield; CM 18388, F, USA, Pennsylvania,
Erie; CM 19397, F, USA, Pennsylvania, Fayette; CM 8448,
F, USA, Pennsylvania, Washington; CM 5207, 57359, 61289,
F, USA, Pennsylvania, Westmoreland; CM 28446, M, USA,
Pennsylvania, Fayette; CM 5089, M, USA, Pennsylvania,
Forest; CM 17487, M, USA, Pennsylvania, Westmoreland;
FMNH 121358, 124592, F, USA, Illinois, DuPage; FMNH
129297, F, USA, Illinois, Hardin; FMNH 55708, F, USA,
Illinois, Lake; FMNH 160111, F, USA, Minnesota, Olmsted;
FMNH 113135, F, USA, Wisconsin, Kenosha; FMNH 89859,
M, USA, Arkansas, Stone; FMNH 1152093, 152094, 152095,
M, USA, Illinois, DuPage; FMNH 129296, M, USA, Illinois,
Hardin; FMNH 55734, M, USA, Illinois, Lake; FMNH 126808,
M, USA, Tennessee, Roane; FMNH 129298, M, USA, Texas,
Brewster; FMNH 167187, 167189, M, USA, Wisconsin, Brown;
FMNH 126807, M, USA, Wisconsin, Racine; ISM 614278,
614655, F, USA, Illinois, Sangamon; ISM 684254, F, USA,
Illinois, Union; ISM 685954, F, USA, Missouri, Crawford;
LSUMZ 26856, M, USA, Louisiana, East Baton Rouge; TU
4726, M, USA, Alabama, Clarke; TU 47, M, USA, Texas,
Tyler; UGAMNH 2238, F, USA, Georgia, Decatur; UMMZ
103647, F, USA, Illinois, Ogle; UMMZ 114346, 114347,
F, USA, Texas, Travis; UMMZ 113307, M, USA, Texas,
Travis; UNSM 14579, F, USA, Nebraska, Lancaster; UNSM
14483, M, USA, Illinois, Randolph; UNSM 14592, M, USA,
Illinois, St. Clair; USNM 21215, F, USA, California, Shasta;
USNM 240402, 258563, F, USA, Maryland, Charles; USNM
257652, F, USA, Maryland, Prince George’s; USNM 283642,
F, USA, Maryland, Washington; USNM 564258, F, USA,
Massachusetts, Worcester; USNM 968, F, USA, Washington,
District of Columbia; USNM 21210, 21211, 21212, 21213,
TAGUE—PELVIC DIMORPHISM AND NEWBORN MASS517
21214, M, USA, California, Shasta; USNM 236960, M, USA,
Maryland, Anne Arundel; USNM 505939, M, USA, Maryland,
Frederick; USNM 282140, M, USA, Maryland, Montgomery;
USNM 564257, M, USA, Massachusetts, Worcester; USNM,
256058, M, USA, South Carolina, Allendale; USNM 521046,
M, USA, Virginia, Arlington; USNM 568477, M, USA,
Virginia, Spotsylvania; USNM 23115, M, USA, Virginia
(Alexandria).