Post on 25-Mar-2023
transcript
Cranial sexual dimorphism in New World marsupials and a test ofRensch’s rule in Didelphidae
DIEGO ASTUA*
Laboratorio de Mastozoologia, Departamento de Zoologia, Centro de Ciencias Biologicas, Universidade Federal de
Pernambuco, Avenida Professor Moraes Rego, s/n. Cidade Universitaria, 50670-420 Recife, Pernambuco, Brasil
* Correspondent: diegoastua@ufpe.br
This study assessed the occurrence of sexual size dimorphism (SSD) and sexual shape dimorphism (SShD) in
the skull and mandible of representatives of most species within the 3 orders of living New World opossums,
Didelphimorphia, Paucituberculata, and Microbiotheria, using geometric morphometrics. Centroid sizes and
partial warps were extracted from landmarks set on images of the dorsal, ventral, and lateral views of the skull
and lateral view of the mandible, and were compared between sexes to estimate SSD and SShD. Specimens
totaling 2,932 from 71 species of Didelphidae, 5 species of Caenolestidae, and 1 species of Microbiotheriidae
were analyzed. SSD was variable in the Didelphimorphia and the Paucituberculata and absent in
Microbiotheria. Similar results were found for SShD, but SSD and SShD are not clearly coupled. I also
evaluated the validity of Rensch’s rule—the widely observed phenomenon of correlated increases in SSD and
body size for male-biased sexual dimorphism, or correlated decreases in SSD in body size for female-biased
sexual dimorphism—in the Didelphidae. Didelphids span 2 orders of magnitude in body size, and, when
present, sexual dimorphism is male-biased. Regressions of SSD and SShD estimators onto size, using
phylogenetic independent contrasts, showed either no significant relationship between SSD or SShD with size
in any of the structures and views analyzed, or a trend contrary to Rensch’s rule (smaller species more
dimorphic, but with male-biased dimorphism). Lack of adherence to Rensch’s rule in Didelphimorphia may
relate to a lack of social interactions and male territoriality, usually associated with such a trend via sexual
selection. If the trend contrary to Rensch’s rule is real, an explanation may lie in the increasing amount of small-
bodied species that recently have been found to be semelparous and thus subject to stronger selection for larger
males. DOI: 10.1644/09-MAMM-A-018.1.
Key words: Caenolestidae, Didelphidae, Microbiotheriidae, Rensch’s rule, sexual shape dimorphism, sexual size
dimorphism
E 2010 American Society of Mammalogists
Living New World opossums are currently classified in 3
orders. Didelphimorphia contains a single family (Didelphi-
dae) and includes 19 genera and .90 species that range in
body size from 10 g (Monodelphis and Hyladelphys) to
.2,000 g (Didelphis). The 2 remaining orders, considerably
less diverse, are probably relict lineages. Living representa-
tives of Paucituberculata include only 3 genera and 6 species
from a single family (Caenolestidae) and are all small-bodied,
weighing from 25 to 50 g. Microbiotheria contains a single
monotypic family (Microbiotheriidae), with a sole, small-
bodied (627 g) species, Dromiciops gliroides.
Sexual dimorphism in qualitative and quantitative charac-
ters in New World opossums has been reported for several
species of Didelphidae (Bergallo and Cerqueira 1994; Maunz
and German 1996; Oliveira et al. 1992; Pine et al. 1985),
Caenolestidae (Bublitz 1987), and Microbiotheriidae (Hersh-
kovitz 1999). However, most morphometric studies that
included quantitative appraisals of sexual dimorphism were
restricted to a single genus (Cerqueira and Lemos 2000;
Lemos and Cerqueira 2002; Ventura et al. 1998), addressed
variation within a single species (Lopez-Fuster et al. 2000,
2002; Pine et al. 1985; Steiner and Catzeflis 2003), or included
only a single representative of each genus analyzed (Astua de
Moraes et al. 2000). Therefore, a comprehensive analysis of
sexual dimorphism in size and shape over a broad taxonomic
range is still lacking for didelphids.
Rensch’s rule states that when males are larger than
females, greater amounts of sexual dimorphism in body size
will be found in larger species. Inversely, greater amounts of
sexual dimorphism are expected in smaller species when
w w w . m a m m a l o g y . o r g
Journal of Mammalogy, 91(4):1011–1024, 2010
1011
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
females are larger than males. Rensch’s rule has been
evaluated in taxa ranging from arthropods to vertebrates
(Abouheif and Fairbairn 1997). The validity of the rule has
been assessed in a wide variety of mammals, but most
analyses lacked statistical tests or did not use phylogenetic
information when performing regressions. Groups of mam-
mals adequately tested for the occurrence of Rensch’s rule
include carnivores, primates, and ungulates (Abouheif and
Fairbairn 1997; Smith and Cheverud 2002), but an appraisal of
Rensch’s rule in marsupials other than kangaroos and
wallabies, which did not include phylogenetic information,
never has been made.
Recently, geometric morphometrics have been used for the
evaluation of sexual dimorphism in a variety of mammal
groups (Cardini and Tongiorgi 2003; Monteiro-Filho et al.
2002; O’Higgins and Collard 2004) because they allow for a
formal separation of size and shape (Zelditch et al. 2004) and
thus independent estimates of size and shape dimorphism
(Franklin et al. 2006; Gidaszewski et al. 2009; Hood 2000).
The purpose of this study was to use geometric morphometric
descriptors to evaluate and quantify the occurrence of sexual
dimorphism in size and shape of the skull and mandible within
the 3 living orders of New World opossums. The data set
obtained here, combined with recent advances in the study of
the phylogenetic relationships of species within the Didelphi-
dae, allowed an evaluation of the application of Rensch’s rule
to this group, assessing the relation of both sexual size and
shape dimorphism to body size.
MATERIALS AND METHODS
Samples.—I used museum specimens and included only
those unambiguously identifiable based on either published
diagnostic characters or geographic distributions. Whenever
possible I included specimens from limited geographic regions
and from a single subspecies. To maximize taxonomic
representation I limited samples of well-represented species
to approximately 30 males and females. For rarer taxa I
examined all available specimens. Specific criteria used for
each species’ sample are detailed in Astua de Moraes (2004).
To avoid ontogenetic variation I used only adult specimens.
Specimens were considered adults when presenting fully
erupted P3, p3, M4, and m4 (Astua and Leiner 2008; Luckett
and Hong 2000; Tribe 1990; Tyndale-Biscoe and Mackenzie
1976). Taxonomy follows Gardner (2007). Rare and more
recently described didelphid genera such as Cryptonanus and
Chacodelphys were not included.
I obtained data for a total of 2,932 specimens from 71
species of Didelphidae, 5 species of Caenolestidae, and 1
species of Microbiotheriidae, for a grand total of 21 genera
and 77 species (Table 1; a complete list of the specimens
examined is available upon request). This represented all
TABLE 1.—Species examined and sample sizes used in this analysis. Sample sizes are presented as ranges of sample sizes for each sex,
because some specimens could not be used in all views due to missing or broken structures that prevented the setting of 1 or more landmarks.
Genus Species and sample sizes (males/females)
Didelphimorphia
Caluromys C. derbianus: 30–36/32–39; C. lanatus: 29–30/27–31; C. philander: 44–52/40–50
Caluromysiops C. irrupta: 3/1–2
Chironectes C. minimus: 30–38/21–25
Didelphis D. albiventris: 29–32/28–31; D. aurita: 28–42/27–35; D. imperfecta: 9/7; D. marsupialis: 26–27/33–35; D. pernigra: 26–27/34–36;
D. virginiana: 3–30/16–31
Gracilinanus G. aceramarcae: 4/3; G. agilis: 30–35/31–33; G. dryas: 5/2–3; G. marica: 6–7/1–2; G. microtarsus: 22–25/7–8
Hyladelphys H. kalinowskii: 3/2
Lestodelphys L. halli: 2/1
Lutreolina L. crassicaudata: 29–32/23–26
Marmosa M. lepida: 2–3/3–4; M. mexicana: 27–29/17–20; M. murina: 27–34/28–30; M. robinsoni: 25–31/30–33; M. rubra: 9–11/5–7; M.
tyleriana: 4–6/2; M. xerophila: 24–28/32–33
Marmosops M. fuscatus: 20–22/9–10; M. impavidus: 32–34/17–23; M. incanus: 31–34/30–31; M. invictus: 3/5; M. noctivagus: 27–31/34–38; M.
ocellatus: 9–12/4–5; M. parvidens: 8–10/6–7; M. paulensis: 15–17/15–17; M. pinheiroi: 10–11/6–9
Metachirus M. nudicaudatus: 28–33/24–27
Micoureus M. alstoni: 3–4/4–4; M. constantiae: 10–11/6–8; M. demerarae: 28–33/24–26; M. paraguayanus: 23–28/23–26; M. phaeus: 6–7/6–8;
M. regina: 26–33/34–37
Monodelphis M. adusta: 8–10/3–4; M. americana: 7–12/12–15; M. brevicaudata: 23–28/25–27; M. dimidiata: 2–5/1–4; M. domestica: 28–33/27–
32; M. glirina: 27–31/29–33; M. palliolata: 1/7; M. sorex: 2/2
Philander P. andersoni: 16–21/15–18; P. frenatus: 29–33/30–32; P. mcilhennyi: 8–9/4–5; P. opossum: 26–31/23–27
Thylamys T. cinderella: 6–7/1; T. elegans: 25–26/20–22; T. karimii: 8/8; T. pallidior: 34–40/22–25; T. pusillus: 7–8/3–6; T. sponsorius: 10–12/
5–8; T. tatei: 4/4–6; T. venustus: 2/1
Tlacuatzin T. canescens: 25–29/18–20
Paucituberculata
Caenolestes C. caniventer: 3/10–12; C. convelatus: 5/8–10; C. fuliginosus: 49–57/36–38
Lestoros L. inca: 38–43/32–34
Rhyncholestes R. raphanurus: 8–10/14–15
Microbiotheria
Dromiciops D. gliroides: 19–23/12–18
1012 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
recognized species for most genera. Adequate samples for
sexual dimorphism analyses were unavailable for several
species, and only specimens with �3 males and females were
included in statistical comparisons. Data for species with
smaller sample sizes are presented only to provide qualitative
comparisons; these include rare taxa such as Hyladelphys,
Lestodelphys, and Caluromysiops.
Images and landmarks.—I took digital images of the skull
(in dorsal, ventral, and lateral view) and the mandible with a
Nikon Coolpix 995 camera (Nikon, Inc., Melville, New York)
at a resolution of 1,280 3 960 pixels. Skulls and mandibles
always were oriented similarly; that is, frontal plane passing
through the root of incisors and base of occipital condyle (for
dorsal views), palate plane (for ventral views), midsagittal
plane (for lateral views), and the plane including coronoid
process and horizontal ramus (for the mandible), always
parallel to the lens plane. All images included a ruler for scale.
For some species I also used images taken for a previous study
(Astua de Moraes et al. 2000).
I determined 12 landmarks for the dorsal view of the skull, 14
for the ventral view, 19 for the lateral view, and 19 for the
lateral view of the mandible (Fig. 1). Landmarks were digitized
using TPSDig (Rohlf 2006). For the dorsal and ventral view of
the skull I set landmarks on both the left and right and averaged
coordinates from both sides using the midline as reflection axis.
This controlled for variation due to skull asymmetry and
resulted in using coordinates representing 1 side of the skull in
the analyses. When only a specific landmark was missing on 1
side of the skull, I used the coordinates from the opposite side
instead of the average coordinates.
I tested all landmarks for repeatability. For each view 30
specimens from 1 species were selected randomly, and all
landmarks were digitized twice, randomly reordering the
specimens between the 2 sampling events. I estimated repeat-
ability as the intraclass correlation coefficient, which was
derived from an analysis of variance on the x and y coordinates
of each landmark using individuals as the factor. This took into
account intraclass variability (error in locating landmark
position) and interclass variation (real differences between
individuals—Falconer 1989). Repeated measurements of the x
and y coordinates of all landmarks exhibited ,10% error, and
the error rate was ,5% in 95% of the cases. All landmarks were
considered satisfactory and included in subsequent analyses.
Statistical analyses.—Landmark configurations were sub-
mitted to a generalized Procrustes alignment (Rohlf and Slice
1990) to remove effects of position, orientation, and isometric
size in landmark configurations, thus formally separating size
and shape. The size variable used in generalized Procrustes
alignment was centroid size, the square root of the sum of
squared distances between each landmark and the centroid of
the whole landmark configuration. This procedure reduces
size to a single univariate measure that incorporates the
multivariate nature of size and was therefore used for
evaluation of sexual size dimorphism (SSD). The remaining
landmark configurations retained only shape information and
were used for evaluation of sexual shape dimorphism (SShD).
I evaluated SSD using centroid sizes for each view. To test
for existence of SSD in each species I compared centroid sizes
between males and females using t-tests (Zar 1996) whenever
allowed by sample sizes. Because 4 tests for each species were
performed, a Bonferroni correction within each species was
applied, thus yielding a P-value of 0.0125. I evaluated SShD
using the complete set of partial warps, including uniform
components, as shape variables. To assess the existence of
SShD in each species I compared male and female shapes
using Goodall F-tests whenever allowed by sample sizes. To
confirm these results, especially for species with reduced
sample sizes, species that presented significant SShD were
FIG. 1.—Landmarks used in the analyses presented on skulls of A,
B) Marmosa robinsoni (A: dorsal view, B: ventral view) and C)
Didelphis albiventris (lateral view) and D) mandible of Lutreolina
crassicaudata. Bars 5 1 cm.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1013
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
submitted to a resampling Goodall F-test with 2,500
repetitions. Goodall F-tests and resampling F-tests were
performed with the Two-Groups module of the IMP suite
(Sheets 2004). Mean Procrustes distances between sexes for
each species also were calculated to visualize the amount of
SShD.
Size and shape sexual dimorphism and Rensch’s rule
in Didelphidae.—To assess the existence of Rensch’s rule in
the Didelphidae I regressed the natural logarithm of the ratio
between centroid sizes of males and females (SSD estimators)
onto the natural logarithm of female centroid size (Smith
1999; Smith and Cheverud 2002). An absence of correlation
between SSD and size would result in a slope that did not
differ significantly from 0. Regressions used a least-squares
fit, because this reflects the distribution of error when the
dependent variable is a ratio but the independent variable is a
direct measurement (Smith 1999).
Traditional analyses of Rensch’s rule usually assess the
existence of increasing SSD with increasing body size, but
with a few exceptions (Drovetski et al. 2006), most analytical
approaches do not consider differences in shape between
sexes. I assessed the relation between SShD and body size to
determine whether shape differences increased with increasing
body size (hereafter termed ‘‘Rensch’s rule for shape’’ to
avoid confusion). SShD was estimated based on the mean
tangent distance between sexes. Tangent distances are
obtained by projecting Procrustes distance between sexes
(which represent the distance between any 2 shapes in
Kendall’s shape space, thus, in this case, the difference in
shape between sexes) onto a plane tangent to Kendall’s
multidimensional curved shape space. Because Procrustes
distances actually lie in the curved shape space, they are not
suitable for use in standard statistical procedures that assume
Euclidean linear distances. The correspondence between
Procrustes distances and tangent distances can be estimated
by correlating these 2 distances for each pair of specimens in
the analyses. All analyses yielded a correlation coefficient of
1, demonstrating that tangent distances were good proxies for
Procrustes distances. I then regressed tangent distances onto
the natural logarithm of female centroid size (the same
indicator of size used for SSD) to test for the relation between
SShD and body size. As for SSD, the absence of a relationship
would yield a slope that does not differ significantly from 0
(i.e., shape difference does not change with body size).
Phylogenies.—Because these data are not statistically
independent due to phylogenetic relationships among species,
it was necessary to conduct statistical analyses within a
phylogenetic framework (Harvey and Pagel 1991; Martins and
Hansen 1997). I used phylogenetic independent contrasts to
account for relationships between the studied taxa (Felsenstein
1985).
Although recent studies have increased the number of taxa
used in phylogenetic analyses of Didelphidae (Jansa and Voss
2005; Voss and Jansa 2003, 2009; Voss et al. 2004, 2005),
they do not encompass all of the taxa available for this study
(.60 species). Therefore, I manually assembled a larger tree
to include all taxa for which I had SSD or SShD data. I
rearranged the original supertree of Cardillo et al. (2004) to
match the topology of Voss and Jansa (2009) for relationships
at the generic level. This assumes monophyletic genera which,
based on current knowledge, is probably reasonable for most
taxa except Marmosa (Voss and Jansa 2009). However, the
clade Marmosa + Micoureus (after exclusion of Tlacuatzin)
was retained because it has been recovered as monophyletic in
most analyses, including the supertree (Cardillo et al. 2004)
and all molecular and morphological analyses (Voss and Jansa
2009).
Phylogenetic independent contrasts were calculated with the
PDAP:PDTREE module (Midford et al. 2008) of Mesquite
(Maddison and Maddison 2008). All analyses were repeated
for the 3 views of the skull and the mandible. Because this
phylogeny lacked branch lengths, I evaluated 2 methods of
assigning arbitrary branch lengths: all branch lengths equal to
1, and the branch length estimation method proposed by Pagel
(1992), as implemented in Mesquite. Of the 2, only the method
proposed by Pagel (1992) did not violate the assumptions of
independence of absolute values of standardized contrasts and
their standard deviations. Even with this method, some views
were excluded from the analyses for violating such assump-
tion.
The tree included polytomies (Fig. 2), reflecting phyloge-
netic uncertainty in some nodes. Therefore, I used the bounded
degrees of freedom approach suggested by Purvis and Garland
(1993), because previous tests of this approach have shown
that independent contrasts can be used even in the presence of
such phylogenetic uncertainty (Garland and Diaz-Uriarte
1999). The method involves calculating upper and lower
limits for the degrees of freedom used to determine the
significance of the regression. All polytomies were converted
to a series of bifurcations of length 0, and thus m 2 1 contrasts
are computed in each polytomy with m branches. However,
because we cannot know whether multifurcating nodes
represent soft or hard polytomies (representing an unknown
phylogeny or a true simultaneous speciation event, respec-
tively), the calculated degrees of freedom are bounded. The
upper bound is calculated assuming that all multifurcations are
hard polytomies, and the corresponding degrees of freedom
are n 2 2 (where n is the number of terminal taxa and n 2 1 is
the number of contrasts). The lower bound is calculated
assuming that all polytomies are soft polytomies, and is
calculated as p 2 1 (where p is the number of real bifurcating
nodes). Regressions then are repeated using upper and lower
bounds to determine how this affects results.
RESULTS
Sexual size and shape dimorphism.—Size comparisons
between sexes for all species with adequate sample sizes are
presented in Table 2. With the exception of Tlacuatzin and the
rare genera with small samples (Hyladelphys and Lestodel-
phys), all remaining genera of Didelphidae and Caenolestidae
contained some species with SSD, at least in 1 of the views. D.
1014 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
gliroides did not exhibit significant SSD in the skull or the
mandible. Shape comparisons between sexes for all species
with adequate sample sizes are presented in Table 3. All
genera presented some species with SShD, at least in 1 of the
views.
Allometric relation between SSD or SShD and size in
Didelphidae.—The relationship between SSD or SShD and
size was evaluated for the skull in dorsal and ventral views
and the mandible. Regressions based on the dorsal view of the
skull and the mandible were not significant, but a significant
FIG. 2.—Tree used to produce the phylogenetic independent contrasts used in regressions of sexual size dimorphism (SSD) and sexual shape
dimorphism (SShD) on size. Refer to text for additional details.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1015
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
TABLE 2.—Sexual size dimorphism in New World opossums. For each view of the skull and the mandible, mean centroid size and standard
deviation (CS 6 SD), and significance of t-tests (boldface type: P , 0.0125, after Bonferroni correction) between male and female centroid sizes
are indicated.
Skull (dorsal) Skull (ventral) Skull (lateral) Mandible
== RR == RR == RR == RR
Didelphimorphia
Caluromys derbianus 68.4 6 3.4 68.1 6 2.9 72.8 6 3.8 72.2 6 3.5 88.7 6 4.5 87.5 6 4.2 70.5 6 4.3 69.0 6 3.3
C. lanatus 72.2 6 2.4 71.0 6 2.8 76.2 6 2.6 75.3 6 3.4 93.1 6 3.1 91.8 6 4.3 73.2 6 2.9 71.6 6 2.9
C. philander 63.2 6 3.7 61.9 6 4.0 66.5 6 4.5 64.0 6 4.6 81.0 ± 5.2 78.0 ± 5.8 63.9 6 4 62.2 6 4.5
Caluromysiops irrupta 76.5 6 0.9 74.0 6 0.3 80.6 6 1.4 77.9 6 0.4 97.0 6 1.1 94.0 6 0.2 75.5 6 2.3a 73.2 6
Chironectes minimus 85.8 6 3.9 83.9 6 4.1 93.4 6 4.5 92.0 6 4.4 113.9 6 5.8 110.7 6 6.2 94.9 6 4.3 92.9 6 4.6
Didelphis albiventris 105.5 ± 8.0 97.8 ± 7.5 116.0 ± 8.7 107.0 ± 8.1 137.4 ± 10.7 126.3 ± 10.1 115.0 ± 6.9 107.7 ± 8.4
D. aurita 122.6 ± 12.0 108.6 ± 7.5 132.9 ± 12.9 119.2 ± 7.9 160.5 ± 15.7 143.0 ± 10.5 133.3 ± 11.6 123.1 ± 7.3
D. imperfecta 101.5 6 9.4 99.5 6 10.3 111.7 6 11.5 109.6 6 11.1 132.6 6 12.2 129.5 6 14.2 109.6 6 10.5 108.6 6 11.4
D. marsupialis 123.3 6 6.0 120.1 6 6.8 136.5 6 6.4 133.4 6 7.4 163.3 6 8.7 158.9 6 10.0 135.0 6 5.4 133.2 6 7.1
D. pernigra 117.1 6 9.1 112.4 6 7.3 129.6 6 9.8 126.2 6 8.6 153.7 6 11.9 149.1 6 10.6 128.3 6 9.3 124.8 6 8.5
D. virginiana 138.3 ± 8.7 127.7 ± 9.2 156.7 ± 10.7 144.0 ± 12.2 188.0 ± 13.3 171.8 ± 15.6 155.0 6 3.5 146.0 6 14.0
Gracilinanus aceramarcae 33.5 6 0.7 31.8 6 0.7 34.0 6 0.6 32.7 6 0.8 42.1 6 1.3 40.1 6 1.4 32.2 6 0.5 30.5 6 0.8
G. agilis 33.0 ± 1.3 31.4 ± 1.4 34.8 ± 1.8 32.9 ± 1.8 42.0 ± 2.1 40.2 ± 2.1 33.1 ± 1.5 31.5 ± 1.7
G. dryas 32.0 6 1.2 31.8 6 0.2 3 6 1.5 32.6 6 0.6 40.3 6 1.8 40.2 6 0.5 30.0 6 1.2 30.6 6 0.5
G. marica 32.2 6 1.1a 29.1 31.5 6 1.4 29.8 6 1.2
G. microtarsus 34.4 ± 1.7 32.3 ± 1.6 36.2 ± 2.4 33.6 ± 2.0 44.0 ± 2.8 40.9 ± 2.7 34.7 ± 2.0 32.1 ± 2.1
Hyladelphys kalinowskii 25.7a 26.0 6 0.4 26.2 6 0.3 26.7 6 0.6
Lestodelphys halli 41.9 6 0.8a 36.4 45.1 6 0.9a 39.3 54.7 6 0.9a 47.0 44.0 6 1.2a 37.2
Lutreolina crassicaudata 79.5 ± 5.8 70.7 ± 5.8 89.4 ± 7.0 79.1 ± 6.8 104.0 ± 8.2 91.7 ± 7.6 85.9 ± 6.5 76.9 ± 6.3
Marmosa lepida 33.1 6 0.9 33.0 6 1.3 33.2 6 0.6 33.1 6 1.9 41.5 6 0.4 41.3 6 1.8 31.4 6 0.3 31.8 6 1.0
M. mexicana 39.7 ± 2.5 37.2 ± 1.5 42.0 ± 3.1 38.7 ± 1.9 51.0 ± 3.7 47.0 ± 2.0 39.0 ± 3.0 36.6 ± 1.5
M. murina 42.1 ± 2.2 39.9 ± 1.8 44.9 ± 2.7 41.0 ± 2.3 54.5 ± 3.5 51.0 ± 2.8 41.9 ± 2.2 40.0 ± 2.2
M. robinsoni 47.5 ± 2.5 44.1 ± 1.7 50.0 ± 2.8 46.2 ± 2.1 61.7 ± 3.2 56.7 ± 2.5 47.6 ± 2.7 44.0 ± 2.1
M. rubra 43.7 6 1.3 42.3 6 1.1 46.3 6 1.0 45.0 6 1.9 55.8 6 1.5 55.3 6 2.5 44.6 6 1.3 43.7 6 1.3
M. tyleriana 37.8 6 0.7 41.1 6 6.6 40.0 6 1.2 44.1 6 9.0 48.9 6 1.2 53.2 6 10.1 38.1 6 0.8 41.7 6 6.5
M. xerophila 38.9 ± 2.0 36.5 ± 1.3 41.7 ± 2.2 38.8 ± 1.5 50.1 ± 2.6 46.8 ± 1.9 40.3 ± 2.2 37.5 ± 1.5
Marmosops fuscatus 42.8 ± 2.7 38.9 ± 1.9 45.6 ± 3.0 41.2 ± 2.1 55.0 ± 3.9 49.3 ± 2.5 43.6 ± 2.9 40.0 ± 2.1
M. impavidus 40.1 ± 2.4 38.1 ± 1.8 42.6 ± 2.8 39.8 ± 2.2 51.2 ± 3.4 48.0 ± 2.9 39.6 ± 2.5 37.6 ± 2.3
M. incanus 45.6 ± 2.8 41.8 ± 2.3 48.9 ± 3.7 44.6 ± 2.7 59.0 ± 4.5 53.5 ± 3.2 46.7 ± 3.4 42.3 ± 2.8
M. invictus 35.2 6 0.2 34.4 6 0.7 36.8 6 0.6 36.0 6 0.9 44.2 6 0.3 43.4 6 0.8 35.0 6 0.5 34.0 6 0.7
M. noctivagus 44.5 ± 2.3 42.3 ± 2.1 47.6 ± 3.1 45.1 ± 2.3 57.5 ± 3.7 54.4 ± 2.7 44.5 ± 2.7 42.6 ± 2.5
M. ocellatus 39.6 6 1.0 37.4 6 2.0 42.5 ± 1.8 38.9 ± 2.3 51.1 6 2.5 47.0 6 3.1 39.6 ± 1.9 36.9 ± 1.8
M. parvidens 33.7 6 0.9 33.2 6 1.9 34.6 6 1.1 34.5 6 2.1 41.8 6 1.5 41.0 6 2.5 33.0 6 0.9 33.1 6 1.6
M. paulensis 42.1 ± 2.1 39.8 ± 2.4 45.1 ± 2.6 42.5 ± 2.6 54.3 ± 2.9 51.1 ± 3.3 42.8 ± 1.8 40.5 ± 2.5
M. pinheiroi 34.4 ± 0.7 33.2 ± 1.0 35.9 ± 0.8 34.1 ± 0.5 43.3 ± 1.2 41.7 ± 1.2 33.9 6 1.0 33.0 6 0.8
Metachirus nudicaudatus 68.5 6 2.9 63.8 6 2.5 71.6 6 3.8 70.2 6 3.5 85.2 6 4.1 83.8 6 3.4 74.1 ± 3.5 71.6 ± 2.6
Micoureus alstoni 52.5 6 1.3 50.5 6 1.9 55.7 6 1.5 58.6 6 8.7 68.7 6 1.9 65.6 6 2.6 53.0 6 1.7 51.9 6 1.8
M. constantiae 48.3 6 3.6 46.0 6 1.8 51.5 6 3.8 49.5 6 1.9 62.7 6 5.0 59.0 6 2.4 50.2 6 3.9 47.1 6 1.9
M. demerarae 51.9 ± 2.5 49.2 ± 2.5 55.2 ± 2.6 52.4 ± 3.0 67.4 ± 3.0 63.6 ± 3.7 53.6 6 2.3 52.0 6 2.8
M. paraguayanus 51.3 6 2.4 50.4 6 2.8 54.9 6 3.2 54.0 6 3.5 67.0 6 4.0 65.1 6 4.3 53.7 6 2.7 52.8 6 3.7
M. phaeus 45.3 6 1.4 44.7 6 2.1 47.9 6 1.7 46.9 6 2.3 58.9 6 2.3 57.5 6 3.5 45.5 6 1.7 44.6 6 1.1
M. regina 52.4 ± 2.4 50.1 ± 2.6 56.1 ± 3.4 53.4 ± 3.1 68.9 ± 3.7 65.4 ± 3.8 53.7 ± 3.6 51.6 ± 3.0
Monodelphis adusta 31.2 6 2.2 29.4 6 1.4 35.0 6 2.6 33.1 6 1.2 41.2 6 2.9 38.5 6 2.0 32.2 6 2.3 30.6 6 1.3
M. americana 32.9 6 3.7 32.0 6 1.7 36.0 6 4.6 35.1 6 2.2 43.0 6 5.4 41.1 6 2.4 35.4 6 3.3 33.5 6 2.3
M. brevicaudata 44.1 ± 3.1 40.2 ± 2.1 48.9 ± 3.6 44.6 ± 2.3 57.8 ± 4.2 52.0 ± 2.7 46.0 ± 3.3 42.1 ± 2.1
M. dimidiata 32.4 6 4.3a 26.7 39.3 6 8.7a 29.8 35.8 6 9.1 31.1 6 5.3
M. domestica 46.0 6 3.9 44.3 6 2.3 51.8 ± 4.1 49.0 ± 2.8 60.3 6 4.8 57.5 6 3.3 48.8 ± 3.5 46.8 ± 2.6
M. glirina 46.5 ± 2.7 41.8 ± 1.7 52.2 ± 3.2 47.0 ± 2 61.5 ± 3.5 54.9 ± 2.5 48.7 ± 2.9 44.1 ± 1.9
M. palliolata 44.0 6 2.4a 42.5
M. sorex 31.6 6 0.7 29.2 6 4.3 35.5 6 1.1 32.6 6 6.0 41.9 6 1.3 38.7 6 6.4 32.1 6 0.3 30.1 6 5.1
Philander andersoni 84.6 ± 3.7 79.4 ± 3.3 94.1 ± 4.6 88.1 ± 3.7 111.9 ± 5.5 104.8 ± 4.2 92.3 ± 3.6 86.7 ± 4.0
P. frenatus 77.0 ± 6.4 69.7 ± 4.2 84.3 ± 7.0 75.8 ± 4.2 100.8 ± 8.3 90.6 ± 5 85.5 ± 7.0 78.2 ± 4.0
P. mcilhennyi 86.7 6 4.2 82.8 6 4.8 97.2 6 4 92.3 6 5.9 115.0 6 5.2 109.0 6 7.2 94.6 6 4.9 91.0 6 5.3
P. opossum 84.2 6 5.1 82.9 6 4.6 93.9 6 6.2 92.6 6 5.9 111.1 6 7.6 108.4 6 7.0 92.6 6 4.7 90.9 6 5.1
Thylamys cinderella 34.6 6 2.2a 35.7 36.6 6 2.5a 37.9 44.0 6 3.5 46.0 35.1 6 2.5 35.4
T. elegans 34.3 6 1.5 33.4 6 1.6 36.6 6 1.9 35.6 6 2.2 44.1 6 2.1 42.7 6 2.5 34.1 6 1.8 33.0 6 1.7
T. karimii 35.3 6 1.4 33.5 6 1.5 38.0 6 1.6 35.7 6 2.1 46.2 ± 1.8 43.1 ± 2.4 35.7 6 1.3 33.6 6 1.9
T. pallidior 31.8 6 1.6 30.9 6 1.3 33.8 6 1.8 32.8 6 1.5 40.5 6 2.2 39.0 6 1.6 31.0 6 1.7 30.6 6 1.4
T. pusillus 30.8 6 1.0 29.7 6 1.6 32.4 6 1.1 31.1 6 2.0 39.2 6 1.6 38.6 6 1.6 32.7 6 2.9 30.1 6 1.5
1016 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
negative slope was found for the analysis of the ventral view
of the skull (Fig. 3A). Results for SShD are similar to those for
SSD, with significant negative slopes found for the skull in
both views (Figs. 3B and 3C) and no relationship between
SShD and size in the mandible.
DISCUSSION
Significant sexual dimorphism in size, shape, or both was
found in Didelphimorphia and Paucituberculata, with signif-
icantly dimorphic species being found in almost all genera.
The only living representative of Microbiotheria is not
sexually dimorphic. In the Didelphidae 50–60% of the species
present significant sexual dimorphism in cranial size or shape,
and in all cases males are larger than females. If these species
follow Rensch’s rule, the larger species should be more
dimorphic. However, didelphids do not seem to follow this
pattern, because slopes either did not differ significantly from
0 (indicating that in these cases the amount of SSD or SShD is
uncorrelated to size) or were negative (indicating that smaller
species were more dimorphic, even though males were larger
than females). Possible explanations for the lack of adherence
to Rensch’s rule may include the lack of rigid social systems
or a possible prevalence of reproductive strategies leading to
greater SSD or SShD in smaller species.
Most previous analyses of sexual dimorphism as it relates to
Rensch’s rule refer to body size dimorphism. I used cranial
size as a proxy for body size when comparing these results
with previous estimates, because cranial size can be expected
to vary accordingly to body size. A correlation between mean
body weight (Smith et al. 2003) and mean cranial size
(centroid size, from this study) for 46 didelphid species
showed a correlation of r 5 0.91, thus allowing for a
comparison between these and previous results.
Sexual dimorphism in Didelphidae.—The only woolly
opossum with data on sexual dimorphism was Caluromys
philander. Presence of SSD in C. philander was detected in
a large sample from French Guiana and in specimens from
a broader geographic area (Astua de Moraes et al. 2000;
Richard-Hansen et al. 1999), but SSD varied across localities
(Caramaschi 2005). Significant SSD was confirmed here for
C. philander only (skull and mandible), yet significant SShD
was found in Caluromys derbianus and C. philander in at least
1 view each (Table 3).
Opossums of the genus Didelphis are some of the best-
studied didelphid species, although existing assessments of
their sexual dimorphism are contradictory. I found SShD for
all 6 currently recognized species (Table 3), but only D.
virginiana, D. aurita, and D. albiventris presented significant
SSD (Table 2). The significant SSD found here in D. aurita,
D. albiventris, and D. virginiana and SShD in D. virginiana
were reported in other studies (Cerqueira and Lemos 2000;
Gardner 1973; Lemos and Cerqueira 2002). This study also
corroborates previous evidence for a lack of SSD in D.
imperfecta, D. pernigra, and D. marsupialis (Ventura et al.
2002). The latter species, however, also has been reported to
exhibit SSD (Cerqueira and Lemos 2000; Richard-Hansen et
al. 1999; Tyndale-Biscoe and Mackenzie 1976).
The remaining genera of large-bodied opossums showed
different levels of SSD and SShD. Although Lutreolina
crassicaudata proved to be highly dimorphic in size, SSD
was absent in Chironectes minimus and found only in the
mandible in Metachirus nudicaudatus (Table 3). Two species
of Philander (P. andersoni and P. frenatus) were highly
dimorphic in size, and 2 (P. mcilhennyi and P. opossum) were
not. Results for SShD were similar, with half the species being
dimorphic. L. crassicaudata has been reported as a highly
dimorphic species (Graipel et al. 1996; Lemos et al. 2001), but
contrary to previous studies (Lemos et al. 2001; Richard-
Hansen et al. 1999) I found no SSD in C. minimus and M.
nudicaudatus. As in Didelphis, contradictory evidence exists
for SSD in M. nudicaudatus (Patton et al. 2000; Richard-
Hansen et al. 1999). These contrasting results are indicative of
significant morphological variation in a widely geographically
distributed taxon that possibly is represented by several
species, as suggested by ongoing studies (Silva 2005; Vieira
2006). Dimorphism in P. frenatus had already been reported in
a previous analysis (Astua de Moraes et al. 2000), but the lack
TABLE 2.—Continued.
Skull (dorsal) Skull (ventral) Skull (lateral) Mandible
== RR == RR == RR == RR
T. sponsorius 35.0 6 1.7 36.2 6 1.4 37.2 6 2.1 38.8 6 1.9 44.9 6 2.5 46.5 6 2.6 34.6 6 2.0 35.6 6 2.7
T. tatei 35.8 6 1.0 34.4 6 1.4 38.2 6 1.1 37.5 6 1.9 45.8 6 1.1 44.8 6 2.5 35.4 6 1.1 34.0 6 2.1
T. venustus 33.5 6 0.8a 33.4 35.3 6 1.1a 36.9 42.6 6 1.3 44.0 32.9 6 1.4a 34.3
Tlacuatzin canescens 35.7 6 2.0 35.4 6 2.7 37.2 6 3.0 37.0 6 3.1 45.6 6 3.6 45.0 6 3.6 35.9 6 2.0 35.7 6 3.2
Paucituberculata
Caenolestes caniventer 37.8 6 2.9 35.6 6 1.2 40.1 6 2.9 37.9 6 1.6 47.1 6 3.1 44.6 6 1.6 33.1 6 2.1 30.8 6 1.0
C. convelatus 39.1 6 2.1 37.1 6 0.9 42.0 ± 2.6 39.4 ± 1.0 49.3 6 3.2 46.5 6 1.1 33.0 6 2.1 32.0 6 0.6
C. fuliginosus 35.9 ± 1.9 33.9 ± 1.3 37.8 ± 2.3 35.4 ± 1.5 44.3 ± 2.4 41.7 ± 1.7 30.2 ± 1.6 28.7 ± 1.3
Lestoros inca 34.0 ± 1.0 32.7 ± 0.7 35.5 ± 1.1 34.2 ± 1.0 41.9 ± 1.4 40.4 ± 1.1 28.5 ± 0.9 27.5 ± 0.9
Rhyncholestes raphanurus 36.5 6 1.1 35.6 6 0.7 39.0 6 2.0 38.5 6 1.2 44.9 6 1.8 44.0 6 1.3 30.7 6 1.4 29.9 6 1.0
Microbiotheria
Dromiciops gliroides 31.7 6 0.9 32.2 6 0.8 33.5 6 1.1 3 6 1.0 39.6 6 1.1 40.3 6 1.0 29.7 6 1.0 30.0 6 1.0
a Species (or views) were not tested and are presented for qualitative comparison purposes only.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1017
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
of SSD found in P. opossum contradicts previous reports
(Lemos et al. 2001; Richard-Hansen et al. 1999).
Most Micoureus species were not dimorphic in size, except
for M. demerarae and M. regina. Micoureus paraguayanus
and M. demerarae recently have been diagnosed as separate
species (Patton and Costa 2003; Patton et al. 2000). Further
differences between these 2 species were observed here,
because only the former presented SSD and SShD in the skull
(Patton et al. 2000). With the exception of Patton et al. (2000),
estimates of sexual dimorphism in Micoureus were unknown
prior to this study.
In Marmosa the presence of SSD in M. murina, M.
xerophila, M. robinsoni, and M. mexicana confirm several
previous reports, although some of these consisted only of
measurement ranges with no statistical tests (Alonso-Mejıa
and Medellın 1992; Lopez-Fuster et al. 2000, 2002; O’Connell
1983; Rossi 2005). The 3 other species of the genus (M.
lepida, M. rubra, and M. tyleriana) are rare in mammal
collections and thus had much smaller sample sizes, which
could have contributed to the absence of SSD. Based on
similarly limited samples, Rossi (2005) also observed low
levels of SSD in these 3 species.
Marmosops was highly dimorphic in size, with most species
showing significant SSD (Table 2). Dimorphism was reported
previously in qualitative characters for the genus (Lunde and
Schutt 1999; Voss and Jansa 2003; Voss et al. 2001). In this
study only M. invictus and M. parvidens did not present
significant SSD, as previously reported for the latter (Pine
1981). This is a problematic species of Marmosops, because
morphological and molecular analyses have suggested that it
contains multiple species (Patton and Costa 2003; Voss and
Jansa 2003; Voss et al. 2001). Estimates of dimorphism may
need to be reevaluated after elucidation of its proper
taxonomic status. Semelparity or partial semelparity have
been proposed for Marmosops incanus and M. paulensis,
based on age classes of museum specimens and field studies
(Leiner et al. 2008; Lorini et al. 1994). Either reproductive
strategy would be consistent with significant dimorphism.
TABLE 3.—Sexual shape dimorphism in New World opossums. For
each view of the skull and mandible mean Procrustes distances
between sexes and results for Goodall F-tests are indicated (boldface
type: significant differences found in Goodall F-tests after
resamplings, using P , 0.0125 for Bonferroni correction).
Mean Procrustes distances:
== 2 RR (31022)
Skull
MandibleDorsal Ventral Lateral
Didelphimorphia
Caluromys derbianus 1.26 0.51 1.46 1.13
C. lanatus 0.85 0.78 1.32 1.18
C. philander 0.70 0.79 1.07 0.91
Caluromysiops irrupta 3.44 2.59 3.90 —
Chironectes minimus 1.43 1.00 1.57 1.79
Didelphis albiventris 2.64 2.77 2.58 2.67
D. aurita 2.56 3.72 3.41 3.75
D. imperfecta 2.29 2.93 3.44 3.42
D. marsupialis 1.43 1.30 1.67 1.13
D. pernigra 1.48 1.41 1.37 1.74
D. virginiana 2.12 1.91 2.31 3.53
Gracilinanus aceramarcae 2.84 1.62 2.62 2.38
G. agilis 1.43 1.49 1.44 1.52
G. dryas 1.27 1.32 1.61 1.96
G. marica — — — 1.99
G. microtarsus 1.42 1.91 2.27 1.83
Lutreolina crassicaudata 2.03 1.79 1.84 2.08
Marmosa lepida 2.91 3.06 2.70 1.94
M. mexicana 1.79 0.96 1.76 1.69
M. murina 1.07 1.56 1.47 1.40
M. robinsoni 1.67 1.71 2.02 2.35
M. rubra 1.69 1.17 1.31 1.10
M. tyleriana 3.15 2.39 3.54 2.79
M. xerophila 1.55 1.16 1.59 1.63
Marmosops fuscatus 2.70 2.21 2.81 2.68
M. impavidus 1.38 1.76 1.93 1.57
M. incanus 1.91 1.42 2.25 1.84
M. invictus 1.65 1.30 1.80 1.33
M. noctivagus 0.99 1.22 1.35 1.38
M. ocellatus 1.81 1.60 1.21 1.90
M. parvidens 2.88 1.47 1.79 1.71
M. paulensis 1.03 1.07 1.30 1.25
M. pinheiroi 1.63 2.03 1.81 2.09
Metachirus nudicaudatus 1.30 1.12 1.35 1.90
Micoureus alstoni 2.09 1.64 2.18 1.79
M. constantiae 1.99 0.96 1.91 2.06
M. demerarae 1.17 1.02 1.29 1.35
M. paraguayanus 0.67 0.72 0.75 1.51
M. phaeus 2.31 0.92 2.50 1.23
M. regina 1.24 1.59 1.55 1.82
Monodelphis adusta 4.02 2.38 2.22 1.99
M. americana 1.26 1.75 2.27 2.42
M. brevicaudata 1.44 1.99 1.95 2.55
M. dimidiata — — — 3.87
M. domestica 0.99 1.09 1.14 1.31
M. glirina 2.34 2.32 2.43 3.05
M. sorex 2.59 3.47 3.31 2.98
Philander andersoni 1.28 1.12 1.41 1.77
P. frenatus 1.64 2.06 2.54 2.68
P. mcilhennyi 2.27 1.15 2.17 1.87
P. opossum 1.72 0.80 1.19 1.34
Thylamys elegans 1.70 0.59 0.93 0.94
T. karimii 1.59 1.12 1.80 2.17
T. pallidior 1.39 0.73 0.90 0.84
TABLE 3.—Continued.
Mean Procrustes distances:
== 2 RR (31022)
Skull
MandibleDorsal Ventral Lateral
T. pusillus 2.82 1.22 2.40 3.08
T. sponsorius 1.40 1.87 2.33 1.41
T. tatei 2.45 1.02 2.03 1.81
Tlacuatzin canescens 1.31 1.09 1.84 1.73
Paucituberculata
Caenolestes caniventer 1.92 2.02 2.55 1.95
C. convelatus 1.79 2.45 2.87 3.26
C. fuliginosus 1.62 1.22 1.99 1.41
Lestoros inca 1.01 0.74 1.30 0.71
Rhyncholestes raphanurus 1.62 0.83 1.36 1.43
Microbiotheria
Dromiciops gliroides 1.39 0.71 1.14 1.13
1018 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
Two species of Gracilinanus, G. microtarsus and G. agilis,
exhibited SSD, and SShD was found for the latter (Tables 2
and 3). Costa et al. (2003) also found high levels of sexual
dimorphism in G. agilis, but only a few measurements were
dimorphic for G. microtarsus. In contrast, my results indicate
that both species present similar levels of size and shape
dimorphism. Martins et al. (2006b) reported significant
differences in the diets of males of females of G. microtarsus
and related these differences to reproductive data to suggest
partial semelparity (Martins et al. 2006a). This could explain
the high levels of SSD and SShD found here, because the
specimens referred to as G. microtarsus by Martins et al.
(2006a) are probably G. agilis (S. Loss, Universidade Federal
do Espirito Santo, pers. comm.).
The genus Thylamys has been studied and rearranged in the
last decade and currently includes 10 species (Gardner 2007). I
obtained representatives of 8 of these, with very variable
sample sizes (Table 1). Of those tested, only Thylamys karimii
presented significant SSD, and only in 1 view. This is in
agreement with previous studies of T. karimii, T. pallidior, and
T. elegans (Carmignotto and Monfort 2006; Solari 2003).
The genus Monodelphis, which includes all short-tailed
opossums, contains .20 species (Gardner 2007). Although its
taxonomy likely will be altered in the future, the choice of the
samples used here increases the chances of dealing with a
single taxonomic unit (particularly for such widely distributed
species as M. domestica). Marked sexual dimorphism is
frequently cited for Monodelphis spp. (Bergallo and Cerqueira
1994; Pine et al. 1985; Ventura et al. 1998). Among the 8
species I sampled, significant SSD was present in M.
brevicaudata, M. domestica, and M. glirina (Table 1),
confirming previous results for the first 2 (Bergallo and
Cerqueira 1994; Ventura et al. 1998), and M. domestica also
presented significant SShD. I could not confirm dimorphism
in M. dimidiata because I had only 1 female specimen.
However, males were larger than the single female in all
views. Based on the observation of sexual dimorphism in M.
dimidiata, Pine et al. (1985) stated that marsupials should be
regarded as extremely dimorphic. However, this study
illustrates that sexual dimorphism is not homogeneous in the
Didelphidae.
Sexual dimorphism in shrew opossums and in the monito
del monte.—This study presents the 1st comprehensive and
comparable assessment of SSD and SShD for most species in
Caenolestidae. Three species presented SSD and SShD in at
least 1 view, particularly Caenolestes fuliginosus and Lestoros
inca (Tables 2 and 3). Previous appraisals of SSD or SShD in
caenolestids are scarce. Bublitz (1987) evaluated sexual
dimorphism in Caenolestidae but reported mainly qualitative
differences between sexes. Albuja and Patterson (1996) report
measurements that suggest sexual dimorphism from species of
Caenolestes but did not provide a statistical analysis of the
data. The lack of sexual dimorphism I found in Dromiciops
gliroides (Microbiotheriidae) confirms previous comments by
Hershkovitz (1999), who stated that living microbiotheriids
were not dimorphic.
Possible sources of sexual dimorphism in New World
opossums.—Sexual dimorphism is usually explained by sexual
selection or as a strategy to avoid niche overlap between sexes
(Shine 1989). Assessing the influence of these factors on
sexual dimorphism in New World marsupials requires
information on social structure, interactions, and precise
ecological niche measurements, all of which are scarce for
the vast majority of species.
In Didelphidae sexual dimorphism increases after sexual
maturity, and it is possibly related to females interrupting growth
by transferring growth-allocated energy to pregnancy and
lactation (Bergallo and Cerqueira 1994; Gardner 1973). Such
an increase in sexual dimorphism also would allow the use of a
wider niche (Cerqueira 1984) and support resource partitioning
(Leite et al. 1994; Pine et al. 1985), but the link between
dimorphism and resource partitioning remains to be tested
properly. I used only adult specimens (i.e., specimens with full
dentition) to avoid influence of ontogenetic differences between
sexes, but recent analyses have shown that sexual maturity occurs
much earlier than the full eruption of all molars and premolars in
several didelphids (Astua and Geise 2006; Dıaz and Flores 2008).
Not only do individuals grow after sexual maturity, but they also
may continue to grow after full eruption of the cheek teeth, thus
possibly further confounding estimates of SSD.
In several Australian marsupials strong SSD also is related
to semelparity (Tyndale-Biscoe and Renfree 1986), where
FIG. 3.—Regressions of sexual size dimorphism (SSD) and sexual shape dimorphism (SShD) on centroid size (CS), using phylogenetic
independent contrasts (PICs), for those views where a significant negative relation was found: A) skull, ventral view, SSD; B) skull, dorsal view,
SShD; C) skull, ventral view, SShD. Both upper bounds (UB) and lower bounds (LB) on degrees of freedom (d.f.) and respective P-values are
presented, according to Purvis and Garland (1993). TD 5 tangent distances. See text for additional details.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1019
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
success in agonistic encounters with other males is particularly
important. This is also the proposed reason for the strong
dimorphism and aggressive behavior found in Monodelphis
dimidiata (Gonzalez and Claramunt 2000; Pine et al. 1985).
Gracilinanus agilis and Marmosops paulensis recently have
been reported to be at least partially semelparous (Leiner et al.
2008; Martins et al. 2006a), and both are dimorphic in size and
shape (Tables 2 and 3). So far, semelparity has been reported
only in small-bodied species. Because other taxa with similar
levels of SSD still lack reproductive data, the relation between
body size, semelparity, and dimorphism awaits investigation.
Examination of the scarce available data on social interactions
in New World marsupials suggests that these are apparently
rudimentary, limited mostly to mating and parental care or to
simple agonistic encounters (Charles-Dominique 1983). This
complicates the establishment of a relationship between sexual
dimorphism and social interactions that has been recognized
for other mammals (Lindenfors and Tullberg 1998).
The results presented here do not agree with all previous
estimates of sexual dimorphism. A precise comparison of
these results with previous analyses is hindered by the use of
traditional morphometrics to derive all previous estimates,
thus confounding effects of size and shape. Another possible
source for these differences would be the existence of
geographical variation in sexual dimorphism, as proposed by
Ventura et al. (2002) and empirically confirmed in Caluromys
philander and Metachirus nudicaudatus (Caramaschi 2005;
Silva 2005; Vieira 2006). Contrasting estimates of dimor-
phism in some species of Didelphis (see above) also could be
due to geographic variation. Considering that this is one of the
most abundant genera in collections, this pattern should be
investigated more thoroughly.
Absence of Rensch’s rule within Didelphidae.—Based on a
phylogenetic analysis of 21 taxa, Abouheif and Fairbairn
(1997) concluded that Rensch’s rule is widespread and
general. Explanations for the general occurrence of Rensch’s
rule include a variety of hypotheses that can be summarized as
evolutionary constraints, natural selection, or sexual selection
(Abouheif and Fairbairn 1997; Dale et al. 2007; Serrano-
Meneses et al. 2008). Sexual selection provides the best
explanation of Rensch’s rule in groups such as shorebirds
(Szekely et al. 2004), hummingbirds and flower mites
(Colwell 2000), Aves as a whole (Dale et al. 2007), primates
(Lindenfors and Tullberg 1998), and odonates (Serrano-
Meneses et al. 2008). For all of these taxa natural history,
ecology, and mating systems are sufficiently known from
large, species-rich data sets. Unfortunately, the same is not
true for the vast majority of didelphids. Knowledge of
population ecology and basic behavior are limited, and
possible explanations for the presence or absence of Rensch’s
rule for the majority of these taxa can only be speculative.
The lack of a significant trend in SSD and SShD in relation
to size among didelphids reflects something that can be
observed qualitatively: small-, medium-, and large-bodied
species exist, and in all of these size classes we find some
species that are dimorphic and others with no significant SSD
or SShD. Sexual selection is presumed to have a major
influence in taxa with polygynous mating systems, where it
may lead to selection for larger males. Social structure is
poorly known for most marsupial species, and existing data
refer to a particular community only (Charles-Dominique
1983). Based on available home range data, didelphids appear
to have a promiscuous social system in which males have
overlapping territories and females tend to be more territorial.
Consequently, didelphid males probably do not compete
directly for territories, releasing them from the selective
pressure for increasing body size. This appears to occur in
small-bodied (e.g., Marmosa murina), medium-bodied (e.g.,
Micoureus paraguayanus), and large-bodied (Didelphis and
Philander) species (Caceres 2006; Charles-Dominique 1983).
Apart from mating periods and preweaning periods (when the
young are attached to the mother), didelphids are usually
solitary and do not engage in the formation of structured social
bonds (Charles-Dominique 1983). Thus, examination of the
available data on social interactions and space use in
didelphids does not suggest situations that would lead to
increasing SSD in larger species due to sexual selection.
Contrary to the predictions of Rensch’s rule, I found a
significant negative relationship between sexual dimorphism
and body size in 3 of 5 cases tested. In these cases smaller
species are more dimorphic than larger species, with males
larger than females. A possible explanation could reside in the
reproductive strategy of several smaller species. Over 2
decades ago the highly dimorphic species Monodelphis
dimidiata was suggested to be semelparous (Pine et al.
1985), which would explain its highly aggressive behavior in
male–male encounters (Gonzalez and Claramunt 2000).
Subsequently, based on age classes of museum specimens,
Lorini et al. (1994) proposed that Marmosops incanus also is
semelparous. More recently, detailed population studies have
indicated that Marmosops paulensis and Gracilinanus agilis
exhibit semelparity or partial semelparity (Leiner et al. 2008;
Martins et al. 2006a). In contrast, no records of semelparity are
found for well-studied, large-bodied species, because these
survive for .1 year and reproduce more than once (Tyndale-
Biscoe and Renfree 1986).
Semelparity is associated with strong, male-biased sexual
dimorphism in the well-studied Australian dasyurid marsupi-
als of the genus Antechinus (Fisher and Cockburn 2006;
Kraaijeveld-Smit et al. 2003). Likewise, each of the 4
didelphid species in which semelparity or partial semelparity
has been hypothesized present strong SSD and SShD. In
Antechinus male-biased sexual dimorphism in semelparous
species appears to be driven by sexual selection. Examination
of behavioral and paternity data has shown that larger males
are dominant, more successful in competing for females,
preferred by females, and survive longer than smaller males
(Fisher and Cockburn 2006). Larger males also obtain longer
copulations, produce more sperm, and fertilize a larger
number of females and produce more offspring than smaller
males (Holleley et al. 2006; Kraaijeveld-Smit et al. 2003).
Although no similar data are available for neotropical species,
1020 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
a similar process may be acting on small, semelparous
didelphid marsupials.
Reproductive patterns are unknown for most short-tailed
and mouse opossums, but many of these small-bodied genera
have significant SSD or SShD, or both. This presents the
possibility that they share social or reproductive patterns that
favor male-biased SSD. Larger didelphid species tend to
survive longer thus potentially reducing the strength of sexual
selection and leading to less-marked SSD. If this is true, it
could help to explain the occurrence of the inverse of Rensch’s
rule observed in the Didelphidae.
RESUMO
Este estudo avaliou a ocorrencia de dimorfismo sexual de
tamanho (DST) e dimorfismo sexual de forma (DSF) no cranio
e na mandıbula de representantes da maioria das especies das
tres ordens de marsupiais do Novo Mundo, Didelphimorphia,
Paucituberculata e Microbiotheria, atraves de morfometria
geometrica. Tamanhos de centroide e deformacoes parciais
foram extraıdos de marcos anatomicos colocados em imagens
das vistas dorsal, lateral e ventral do cranio e lateral da
mandıbula, e foram comparados entre os sexos para estimar o
DST e DSF. Foram analisados 2932 especimes de 71 especies
de Didelphidae, 5 especies de Caenolestidae e 1 de Micro-
biotheriidae. O DST foi variavel em Didelphimorphia e em
Paucituberculata, e ausente em Microbiotheria. O DSF seguiu
padroes similares, mas o DST e DSF nao estao claramente
acoplados. Tambem foi avaliada a validade da regra de
Rensch—o fenomeno amplamente observado de aumento do
dimorfismo sexual correlacionado com um aumento do
tamanho corporal, quando machos sao maiores que femeas,
ou com uma diminuicao do tamanho corporal, quando femeas
sao maiores que machos—em Didelphidae. Os didelfıdeos
variam em ate duas ordens de grandeza nos seus tamanhos
corporais, e quando existe dimorfismo, machos sao maiores
que femeas. Regressoes dos indicadores de DST e DSF sobre
o tamanho, usando contrastes filogeneticos independente,
indicaram ausencia de relacao significativa entre DST ou DSF
com o aumento do tamanho corporal em nenhuma das
estruturas e vistas analisadas, ou um padrao contrario a regra
de Rensch (especies menores com mais dimorfismo, apesar de
machos serem maiores). Explicacoes para a falta de aderencia
a regra de Rensch em Didelphimorphia podem estar
relacionadas a falta de interacoes sociais ou de territorialidade
em machos, geralmente associadas a este padrao atraves de
selecao sexual. Caso o padrao inverso a regra de Rensch seja
real, uma explicacao pode estar na quantidade crescente de
especies de pequeno tamanho corporal que recentemente
foram descritas como semelparas e portanto sujeitas a uma
selecao mais forte por machos maiores.
ACKNOWLEDGMENTS
I am grateful to the following institutions and professionals
(curators and collection managers) for access to collections under
their care, help during my visits, and sending additional information:
R. Voss (American Museum of Natural History); C. Conroy
(Museum of Vertebrate Zoology, University of California, Berkeley);
J. Salazar-Bravo and W. Gannon (Museum of Southwestern Biology,
University of New Mexico); J. Braun and M. Revelez (Sam Noble
Oklahoma Museum of Natural History); R. Timm (Museum of
Natural History, University of Kansas); B. Patterson and M.
Schulenberg (Field Museum of Natural History); J. Kirsch and P.
Holahan (University of Wisconsin Zoological Museum); A. Gardner,
L. Gordon and C. Ludwig (National Museum of Natural History); L.
Costa, Y. Leite and B. Andrade (Universidade Federal de Minas
Gerais); J. A. Oliveira, L. F. Oliveira, L. Salles and S. Franco (Museu
Nacional, Unversidade Federal do Rio de Janeiro); M. de Vivo and J.
Barros (Museu de Zoologia da Universidade de Sao Paulo); L. Geise
(Universidade do Estado do Rio de Janiero); R. Cerqueira
(Universidade Federal do Rio de Janeiro); and V. Pacheco and E.
Vivar Pinares (Museo de Historia Natural de la Universidad Nacional
de San Marcos). R. Voss also granted me access to material he
personally was studying, including rare specimens. For loans of
several important specimens I am indebted to A. Brunet (James Ford
Bell Museum of Natural History), M. Hafner (Louisiana State
University, Museum of Natural Science), and P. C. A. Simoes-Lopes
and M. Graipel (Universidade Federal de Santa Catarina). I am
grateful to R. Voss and B. Patterson for information on dentition and
morphology of Caenolestidae, D. Flores and S. Solari for help in
determining Thylamys species, and M. Guenther, E. Dumont, and an
anonymous reviewer for several revisions and suggestions that
improved the clarity and quality of this manuscript. The author was
supported throughout this project by a doctoral fellowship from
Fundacao de Amparo a Pesquisa do Estado de Sao Paulo, (00/11444-
7), and a Grant-in-Aid of Research from the American Society of
Mammalogists. This work is presently supported by a grant from
Fundacao de Amparo a Ciencia e Tecnologia do Estado de
Pernambuco (APQ-0351-2.04/06).
LITERATURE CITED
ABOUHEIF, E., AND D. J. FAIRBAIRN. 1997. A comparative analysis of
allometry for sexual size dimorphism: assessing Rensch’s rule.
American Naturalist 149:540–562.
ALBUJA, L., AND B. D. PATTERSON. 1996. A new species of northern
shrew-opossum (Paucituberculata: Caenolestidae) from the Cor-
dillera del Condor, Ecuador. Journal of Mammalogy 77:41.
ALONSO-MEJIA, A., AND R. A. MEDELLIN. 1992. Marmosa mexicana.
Mammalian Species 421:1–4.
ASTUA, D., AND L. GEISE. 2006. Early reproductive onset in the white-
eared opossum, Didelphis albiventris Lund, 1840 (Didelphimor-
phia, Didelphidae). Mammalian Biology 71:299–303.
ASTUA, D., AND N. O. LEINER. 2008. Tooth eruption sequence and
replacement pattern in woolly opossums, genus Caluromys
(Didelphimorphia: Didelphidae). Journal of Mammalogy 89:244–
251.
ASTUA DE MORAES, D. 2004. Evolucao morfologica do cranio e
elementos pos-cranianos dos marsupiais americanos (Didelphi-
morphia, Paucituberculata, Microbiotheria). Ph.D. dissertation,
Universidade de Sao Paulo, Sao Paulo, Brasil.
ASTUA DE MORAES, D., E. HINGST-ZAHER, L. F. MARCUS, AND R.
CERQUEIRA. 2000. A geometric morphometric analysis of cranial
and mandibular shape variation of didelphid marsupials. Hystrix,
Italian Journal of Mammalogy, New Series 11:115–130.
BERGALLO, H. G., AND R. CERQUEIRA. 1994. Reproduction and growth
of the opossum Monodelphis domestica (Mammalia, Didelphidae)
in northeastern Brazil. Journal of Zoology (London) 232:551–563.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1021
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
BUBLITZ, J. 1987. Untersuchungen zur Systematik der rezenten
Caenolestidae Trouessart, 1898: unter Verwendung craniomea-
trischer Methoden. Bonner Zoologische Monographien 23:1–96.
CACERES, N. C. 2006. Uso do espaco por marsupiais: fatores
influentes, comportamento e heterogeneidade espacial. Pp. 203–
215 in Os marsupiais do Brasil: biologia, ecologia e evolucao (N.
C. Caceres and E. L. A. Monteiro-Filho, eds.). Editora da
Universidade Federal do Mato Grosso do Sul, Campo Grande,
Brasil.
CARAMASCHI, F. P. 2005. Variacao geografica em Caluromys
philander (Linnaeus, 1758) no Brasil (Didelphimorphia: Didelphi-
dae). M.S. thesis, Museu Nacional/Universidade Federal do Rio de
Janeiro, Rio de Janeiro, Brasil.
CARDILLO, M., O. R. P. BININDA-EMONDS, E. BOAKES, AND A. PURVIS.
2004. A species-level phylogenetic supertree of marsupials.
Journal of Zoology (London) 264:11–31.
CARDINI, A., AND P. TONGIORGI. 2003. Yellow-bellied marmots
(Marmota flaviventris) ‘in the shape space’ (Rodentia, Sciuridae):
sexual dimorphism, growth and allometry of the mandible.
Zoomorphology 122:11–23.
CARMIGNOTTO, A. P., AND T. MONFORT. 2006. Taxonomy and
distribution of the Brazilian species of Thylamys (Didelphimor-
phia: Didelphidae). Mammalia 70:126–144.
CERQUEIRA, R. 1984. Reproduction de Didelphis albiventris dans le
nord-est du Bresil (Polyprotodontia, Didelphidae). Mammalia
48:95–104.
CERQUEIRA, R., AND B. LEMOS. 2000. Morphometric differentiation
between neotropical black-eared opossums, Didelphis marsupialis
and D. aurita (Didelphimorphia, Didelphidae). Mammalia 64:319–
327.
CHARLES-DOMINIQUE, P. 1983. Ecology and social adaptations in
didelphid marsupials: comparison with eutherians of similar
ecology. Pp. 341–435 in Advances in the study of mammalian
behavior (J. F. Eisenberg and D. G. Kleiman, eds.). Special
Publication 7, The American Society of Mammalogists.
COLWELL, R. K. 2000. Rensch’s rule crosses the line: convergent
allometry of sexual size dimorphism in hummingbirds and flower
mites. American Naturalist 156:495–510.
COSTA, L. P., Y. L. R. LEITE, AND J. L. PATTON. 2003. Phylogeography
and systematic notes on two species of gracile mouse opossums,
genus Gracilinanus (Marsupialia: Didelphidae) from Brazil.
Proceedings of the Biological Society of Washington 116:275–
292.
DALE, J., P. O. DUNN, J. FIGUEROLA, T. LISLEVAND, T. SZEKELY, AND L.
A. WHITTINGHAM. 2007. Sexual selection explains Rensch’s rule of
allometry for sexual size dimorphism. Proceedings of the Royal
Society of London, B. Biological Sciences 274:2971–2979.
DIAZ, M. M., AND D. A. FLORES. 2008. Early reproduction onset in
four species of Didelphimorphia in the Peruvian Amazonia.
Mammalia 72:126–130.
DROVETSKI, S. V., S. ROHWER, AND N. A. MODE. 2006. Role of sexual
and natural selection in evolution of body size and shape: a
phylogenetic study of morphological radiation in grouse. Journal of
Evolutionary Biology 19:1083–1091.
FALCONER, D. S. 1989. Introduction to quantitative genetics. Long-
man, Essex, United Kingdom.
FELSENSTEIN, J. 1985. Phylogenies and the comparative method.
American Naturalist 125:1–15.
FISHER, D. O., AND A. COCKBURN. 2006. The large-male advantage in
brown antechinuses: female choice, male dominance, and delayed
male death. Behavioral Ecology 17:164–171.
FRANKLIN, D., L. FREEDMAN, N. MILNE, AND C. E. OXNARD. 2006. A
geometric morphometric study of sexual dimorphism in the crania
of indigenous southern Africans. South African Journal of Science
102:229–238.
GARDNER, A. L. 1973. The systematics of the genus Didelphis
(Marsupialia: Didelphidae) in North and Middle America. Special
Publications, The Museum, Texas Tech University 4:4–81.
GARDNER, A. L. 2007. Mammals of South America. University of
Chicago Press, Chicago, Illinois.
GARLAND, T., AND R. DIAZ-URIARTE. 1999. Polytomies and
phylogenetically independent contrasts: examination of the
bounded degrees of freedom approach. Systematic Biology
48:547–558.
GIDASZEWSKI, N. A., M. BAYLAC, AND C. P. KLINGENBERG. 2009.
Evolution of sexual dimorphism of wing shape in the Drosophila
melanogaster subgroup. BMC Evolutionary Biology 9:110.
GONZALEZ, E. M., AND S. CLARAMUNT. 2000. Behaviors of captive
short-tailed opossums, Monodelphis dimidiata (Wagner, 1847)
(Didelphimorphia, Didelphidae). Mammalia 64:271–285.
GRAIPEL, M., P. R. M. MILLER, AND A. XIMENEZ. 1996. Contribuicao a
identificacao e distribuicao das subespecies de Lutreolina crassi-
caudata (Desmarest) (Marsupialia, Didelphidae). Revista Brasi-
leira de Zoologia 13:781–790.
HARVEY, P. H., AND M. D. PAGEL. 1991. The comparative method in
evolutionary biology. Oxford University Press, Oxford, United
Kingdom.
HERSHKOVITZ, P. 1999. Dromiciops gliroides Thomas, 1894, last of the
Microbiotheria (Marsupialia), with a review of the family
Microbiotheriidae. Fieldiana: Zoology (New Series) 93:1–60.
HOLLELEY, C. E., C. R. DICKMAN, M. S. CROWTHER, AND B. P.
OLDROYD. 2006. Size breeds success: multiple paternity, multivar-
iate selection and male semelparity in a small marsupial,
Antechinus stuartii. Molecular Ecology 15:3439–3448.
HOOD, C. S. 2000. Geometric morphometric approaches to the study
of sexual size dimorphism in mammals. Hystrix, Italian Journal of
Mammalogy, New Series 11:77–89.
JANSA, S. A., AND R. VOSS. 2005. Phylogenetic relationships of the
marsupial genus Hyladelphys based on nuclear gene sequences and
morphology. Journal of Mammalogy 86:853–865.
KRAAIJEVELD-SMIT, F. J. L., S. J. WARD, AND P. D. TEMPLE-SMITH.
2003. Paternity success and the direction of sexual selection in a
field population of a semelparous marsupial, Antechinus agilis.
Molecular Ecology 12:475–484.
LEINER, N. O., E. Z. F. SETZ, AND W. R. SILVA. 2008. Semelparity and
factors affecting the reproductive activity of the Brazilian slender
opossum (Marmosops paulensis) in southeastern Brazil. Journal of
Mammalogy 89:153–158.
LEITE, Y. L. R., J. R. STALLINGS, AND L. P. COSTA. 1994. Particao de
recursos entre especies simpatricas de marsupiais na Reserva
Biologica de Poco das Antas, Rio de Janeiro. Revista Brasileira de
Biologia 54:525–536.
LEMOS, B., AND R. CERQUEIRA. 2002. Morphological differentiation in
the white-eared opossum group (Didelphidae: Didelphis). Journal
of Mammalogy 83:354–369.
LEMOS, B., G. MARROIG, AND R. CERQUEIRA. 2001. Evolutionary rates
and stabilizing selection in large-bodied opossum skulls (Didel-
phimorphia: Didelphidae). Journal of Zoology (London) 255:181–
189.
LINDENFORS, P., AND B. S. TULLBERG. 1998. Phylogenetic analyses of
primate size evolution: the consequences of sexual selection.
Biological Journal of the Linnean Society 64:413–447.
1022 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
LOPEZ-FUSTER, M. J., R. PEREZ-HERNANDEZ, J. VENTURA, AND M.
SALAZAR. 2000. Effect of environment on skull-size variation in
Marmosa robinsoni in Venezuela. Journal of Mammalogy 81:829.
LOPEZ-FUSTER, M. J., M. SALAZAR, R. PEREZ-HERNANDEZ, AND J.
VENTURA. 2002. Craniometrics of the orange mouse opossum
Marmosa xerophila (Didelphimorphia: Didelphidae) in Venezuela.
Acta Theriologica 47:201–209.
LORINI, M. L., J. A. OLIVEIRA, AND V. G. PERSSON. 1994. Annual age
structure and reproductive patterns in Marmosa incana (Lund,
1841) (Didelphidae, Marsupialia). Zeitschrift fur Saugetierkunde
59:65–73.
LUCKETT, W. P., AND N. HONG. 2000. Ontogenetic evidence for dental
homologies and premolar replacement in fossil and extant
Caenolestids (Marsupialia). Journal of Mammalian Evolution
7:109–127.
LUNDE, D. P., AND W. A. SCHUTT. 1999. The peculiar carpal tubercles
of male Marmosops parvidens and Marmosa robinsoni (Didelphi-
dae: Didelphinae). Mammalia 63:495–504.
MADDISON, W. P., AND D. R. MADDISON. 2008. Mesquite: a modular
system for evolutionary analysis. Version 2.5. http://mesquiteproject.
org. Accessed 9 June 2008.
MARTINS, E. G., V. BONATO, C. Q. DA-SILVA, AND S. R. F. DOS REIS.
2006a. Partial semelparity in the neotropical didelphid marsupial
Gracilinanus microtarsus. Journal of Mammalogy 87:915–920.
MARTINS, E. G., V. BONATO, H. P. PINHEIRO, AND S. F. DOS REIS. 2006b.
Diet of the gracile mouse opossum (Gracilinanus microtarsus)
(Didelphimorphia: Didelphidae) in a Brazilian cerrado: patterns of
food consumption and intrapopulation variation. Journal of
Zoology (London) 269:21–28.
MARTINS, E. P., AND T. F. HANSEN. 1997. Phylogenies and the
comparative method: a general approach to incorporating phylo-
genetic information into the analysis of interspecific data.
American Naturalist 149:646–667.
MAUNZ, M., AND R. Z. GERMAN. 1996. Craniofacial heterochrony and
sexual dimorphism in the short-tailed opossum (Monodelphis
domestica). Journal of Mammalogy 77:992.
MIDFORD, P. E., T. GARLAND, JR., AND W. P. MADDISON. 2008.
PDAP:PDTREE package for Mesquite, version 1.13. http://
mesquiteproject.org. Accessed 30 July 2008.
MONTEIRO-FILHO, E. L. A., L. R. MONTEIRO, AND S. F. REIS. 2002. Skull
shape and size divergence in dolphins of the genus Sotalia: a tridimen-
sional morphometric analysis. Journal of Mammalogy 83:125–134.
O’HIGGINS, P., AND M. COLLARD. 2004. Sexual dimorphism and facial
growth in papionin monkeys. Journal of Zoology (London)
257:255–272.
O’CONNELL, M. A. 1983. Marmosa robinsoni. Mammalian Species
203:1–6.
OLIVEIRA, J. A., M. L. LORINI, AND V. G. PERSSON. 1992. Pelage
variation in Marmosa incana (Didelphidae, Marsupialia) with
notes on taxonomy. Zeitschrift fur Saugetierkunde 57:129–136.
PAGEL, M. D. 1992. A method for the analysis of comparative data.
Journal of Theoretical Biology 156:431–442.
PATTON, J. L., AND L. P. COSTA. 2003. Molecular phylogeography and
species limits in rainforest didelphid marsupials of South America.
Pp. 63–81 in Predators with pouches: the biology of carnivorous
marsupials (M. E. Jones, C. R. Dickman, and M. Archer, eds.).
CSIRO Publishing, Collingwood, Australia.
PATTON, J. L., M. N. F. DA SILVA, AND J. R. MALCOLM. 2000. Mammals
of the Rio Jurua and the evolutionary and ecological diversification
of Amazonia. Bulletin of the American Museum of Natural History
244:1–306.
PINE, R. H. 1981. Reviews of the mouse opossums Marmosa parvidens
Tate and Marmosa invicta Goldman (Mammalia: Marsupialia:
Didelphidae) with description of a new species. Mammalia 45:55–70.
PINE, R. H., P. L. DALBY, AND J. O. MATSON. 1985. Ecology, postnatal
development, morphometrics, and taxonomic status of the short-
tailed opossum, Monodelphis dimidiata, an apparently semelparous
annual marsupial. Annals of Carnegie Museum 54:195–231.
PURVIS, A., AND T. GARLAND, JR. 1993. Polytomies in comparative
analyses of continuous data. Systematic Biology 42:569–575.
RICHARD-HANSEN, C., J. C. VIE, N. VIDAL, AND J. KERAVEC. 1999. Body
measurements on 40 species of mammals from French Guiana.
Journal of Zoology (London) 247:419.
ROHLF, F. J. 2006. TPSDig. Available from the author at Department
of Ecology and Evolution, State University of New York at Stony
Brook, Stony Brook.
ROHLF, F. J., AND D. SLICE. 1990. Extensions of the Procrustes method
for the optimal superimposition of landmarks. Systematic Zoology
39:40–59.
ROSSI, R. V. 2005. Revisao taxonomica de Marmosa Gray, 1821
(Didelphimorphia: Didelphidae). Ph.D. dissertation, Universidade
de Sao Paulo, Sao Paulo, Brasil.
SERRANO-MENESES, M. A., A. CORDOBA-AGUILAR, M. AZPILICUETA-
AMORIN, E. GONZALEZ-SORIANO, AND T. SZEKELY. 2008. Sexual
selection, sexual size dimorphism and Rensch’s rule in Odonata.
Journal of Evolutionary Biology 21:1259–1273.
SHEETS, H. D. 2004. IMP suite—integrated morphometrics package.
http://www.canisius.edu/,sheets/morphsoft.html. Accessed 5 Au-
gust 2004.
SHINE, R. 1989. Ecological causes for the evolution of sexual
dimorphism: a review of the evidence. Quarterly Review of
Biology 64:419–461.
SILVA, H. S. 2005. Variacao geografica em Metachirus nudicaudatus
(Didelphimorphia, Didelphidae) na Mata Atlantica. M.S. thesis,
Museu Nacional/Universidade Federal do Rio de Janeiro, Rio de
Janeiro, Brasil.
SMITH, F. A., ET AL. 2003. Body mass of late Quaternary mammals.
Ecology 84:3403.
SMITH, R. J. 1999. Statistics of sexual size dimorphism. Journal of
Human Evolution 36:423–459.
SMITH, R. J., AND J. M. CHEVERUD. 2002. Scaling of sexual
dimorphism in body mass: a phylogenetic analysis of Rensch’s
rule in primates. International Journal of Primatology 23:1098–
1135.
SOLARI, S. 2003. Diversity and distribution of Thylamys (Didelphi-
dae) in South America, with emphasis on species from the
western side of the Andes. Pp. 82–101 in Predators with pouches:
the biology of carnivorous marsupials (M. E. Jones, C. R.
Dickman, and M. Archer, eds.). CSIRO Publishing, Collingwood,
Australia.
STEINER, C., AND F. M. CATZEFLIS. 2003. Mitochondrial diversity and
morphological variation of Marmosa murina (Didelphidae) in
French Guiana. Journal of Mammalogy 84:822.
SZEKELY, T., R. P. FRECKLETON, AND J. D. REYNOLDS. 2004. Sexual
selection explains Rensch’s rule of size dimorphism in shorebirds.
Proceedings of the National Academy of Sciences 101:12224–
12227.
TRIBE, C. J. 1990. Dental age classes in Marmosa incana and other
didelphoids. Journal of Mammalogy 71:566–569.
TYNDALE-BISCOE, C. H., AND R. B. MACKENZIE. 1976. Reproduction in
Didelphis marsupialis and D. albiventris in Colombia. Journal of
Mammalogy 57:249–265.
August 2010 ASTUA—SEXUAL DIMORPHISM IN AMERICAN MARSUPIALS 1023
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022
TYNDALE-BISCOE, C. H., AND M. B. RENFREE. 1986. Reproductive
physiology of marsupials. Cambridge University Press, Cam-
bridge, United Kingdom.
VENTURA, J., R. PEREZ-HERNANDEZ, AND M. J. LOPEZ-FUSTER. 1998.
Morphometric assessment of the Monodelphis brevicaudata group
(Didelphimorphia: Didelphidae) in Venezuela. Journal of Mam-
malogy 79:104.
VENTURA, J., M. SALAZAR, R. PEREZ-HERNANDEZ, AND M. J. LOPEZ-
FUSTER. 2002. Morphometrics of the genus Didelphis (Didelphi-
morphia: Didelphidae) in Venezuela. Journal of Mammalogy
83:1087.
VIEIRA, C.L.G.C. 2006. Sistematica do jupati Metachirus Burmeister,
1854 (Mammalia: Didelphimorphia). M.S. thesis, Universidade
Federal do Espirito Santo, Vitoria, Brasil.
VOSS, R. S., AND S. A. JANSA. 2003. Phylogenetic studies on didelphid
marsupials II. Nonmolecular data and new IRBP sequences:
separate and combined analyses of didelphine relationships with
denser taxon sampling. Bulletin of the American Museum of
Natural History 276:1–82.
VOSS, R. S., AND S. A. JANSA. 2009. Phylogenetic relationships and
classification of didelphid marsupials, an extant radiation of New
World metatherian mammals. Bulletin of the American Museum of
Natural History 322:1–177.
VOSS, R. S., D. P. LUNDE, AND S. JANSA. 2005. On the contents
of Gracilinanus Gardner and Creighton, 1989, with the
description of a previously unrecognized clade of small
didelphid marsupials. American Museum Novitates 3482:
1–34.
VOSS, R. S., D. P. LUNDE, AND N. B. SIMMONS. 2001. The mammals of
Paracou, French Guiana: a neotropical lowland rainforest fauna—
part 2. Nonvolant species. Bulletin of the American Museum of
Natural History 263:3–236.
VOSS, R. S., T. TARIFA, AND E. YENSEN. 2004. An introduction to
Marmosops (Marsupialia: Didelphidae), with the description of a
new species from Bolivia and notes on the taxonomy and distribution
of other Bolivian forms. American Museum Novitates 3466:1–40.
ZAR, J. H. 1996. Biostatistical analysis. 3rd ed. Prentice Hall, Inc.,
Upper Saddle River, New Jersey.
ZELDITCH, M. L., D. L. SWIDERSKI, H. D. SHEETS, AND W. L. FINK. 2004.
Geometric morphometrics for biologists: a primer. Elsevier
Academic Press, Boston, Massachusetts.
Submitted 14 January 2009. Accepted 25 January 2010.
Associate Editor was Elizabeth R. Dumont.
1024 JOURNAL OF MAMMALOGY Vol. 91, No. 4
Dow
nloaded from https://academ
ic.oup.com/jm
amm
al/article/91/4/1011/965366 by guest on 15 July 2022