Zoology 106 (2003): 73–84

© by Urban & Fischer Verlag

http://www.urbanfischer.de/journals/zoology

Carpal ontogeny in Monodelphis domestica and

Caluromys philander (Marsupialia)
Jan Prochel and Marcelo R. Sánchez-Villagra*
Zoological Institute, University of Tübingen, Tübingen, Germany

Received August 29, 2002 · Revised version received December 10, 2002 · Accepted December 11, 2002

Summary
Carpal bones have experienced numerous changes during marsupial evolution, even though their diversity and development remain
poorly studied. The aim of this work was to document adult form and the pattern of mesenchymal tissue condensation and formation of
chondrification and ossification centers in the hand of two marsupials. Two fundamental questions were asked: whether the loss of em-
bryonic precursors was associated with the loss of adult elements, or whether there were developmental signs of ancestral mammalian
elements that have been fused or lost in marsupial taxa. We were also interested to find out whether there is sexual dimorphismus in the
carpals, as has been reported for some didelphids. Histological sections, cleared and stained specimens and macerated skeletons repre-
senting an ontogenetic series of Monodelphis domestica were used to document carpal development. Comparisons were made with
perinatal stages of Caluromys philander and with adult specimens of other marsupials. A prenatal M. domestica in the 13th day after
conception has a cell condensation that because of its position is homologized with a centrale, which is at birth already lost or fused.
Neonatal M. domestica and C. philander have the number and arrangement of their adult carpal anatomy. Trapezium and trapezoid start
ossification later than most other carpals, while pisiform and prepollex are the last to do so. Adult males of M. domestica have relatively
larger and more robust pisiforms, compared to other carpals, than females. This sexual dimorphism develops relatively late as it was not
recorded in male specimens around 160 days old. An extra sesamoid bone located just distal to the radius and proximo-palmar to the
scaphoid was recorded in specimens of C. philander, C. derbianus and Didelphis virginiana.

Key words: hand, phylogeny, Marsupialia, Didelphidae, Monodelphis

Introduction 1989; Szalay, 1994). However, studies on their on-

The anatomy of carpal bones has experienced numer-
ous changes during mammalian evolution (Weber,
1928; Yalden, 1972; Lewis, 1989; Stafford and Thor-
ington, 1998; Ji et al., 2002). The development of these
elements in most groups of mammals remains largely
unstudied, although ontogenetic studies can often de-
liver information about homology and adaptations and
constraints during development (Hinchliffe and Grif-
fiths, 1983; Shubin and Alberch, 1986).
Preliminary studies have shown that a large diversity in
carpal anatomy exists within the Marsupialia (Lewis,
togeny are few and limited. Emery (1897) has been the
only researcher to present a comprehensive study of
this kind, having summarized the little information
available at the time and having studied 12 marsupial
species, including Didelphis aurita, as sole representa-
tive of the Didelphidae. Schmidt-Ehrenberg (1942)
studied the development of carpals in Didelphis marsu-
pialis. Nesslinger (1956) described the sequence of os-
sification of the carpals in her investigation of ossifica-
tion centers in Didelphis marsupialis, as did Oliveira et
al. (1998) for Didelphis albiventris. Szalay (1994) and
colleagues (Szalay and Trofimov, 1996; Szalay and

*Corresponding author: Marcelo R. Sánchez-Villagra, Zoological Institute, University of Tübingen, Auf der Morgenstelle 28,
D-72076 Tübingen, Germany; phone: ++49-7071-297 4631; fax: ++49-7071-295 150; e-mail: marcelo.sanchez@uni-tuebingen.de

0944-2006/03/106/01-073 $ 15.00/0
J. Prochel and M. R. Sánchez-Villagra

Sargis, 2001) presented hypotheses about the adult PND-5 represents approximately the middle point of the
groundplan in the carpals of Marsupialia and of the period of fixation (Luckett, 1993), the end of which is
major clades within this group, including didelphids. represented by PND-12. Two histologically-sectioned
In this paper we document the development of carpals specimens of Caluromys philander, a neonate and a very
in an ontogenetic series of Monodelphis domestica and early pouch-young in the period of fixation were also
in perinatal stages of Caluromys philander. We exam- studied. Most sections had a thickness of 10 µ, except
ine some aspects of carpal anatomy among adult Didel- the oldest M. domestica with a thickness of 15 µ, and all
phidae, and discuss the results in the context of marsu- were embedded in paraffin and stained with Azan/Dog-
pial evolution. This is part of a project led by the junior mak. The 3-D-computer reconstruction method used in
author that aims at documenting adult form and the pat- this work was described in Sánchez-Villagra et al.
tern of mesenchymal tissue condensation and formation (2002), who also presented photographs of several of the
of chondrification and ossification centers in the marsu- specimens studied in this work before sectioning.
pial hand. General questions asked in this investigation In addition to the histological sections of M. domestica,
were: Is the loss of embryonic precursors associated cleared and stained specimens of this species were
with the loss of adult elements of the marsupial hand, as available, including a neonate and specimens of ap-
has often been the case during limb evolution (Hall, proximate ages PND 6, 30 and 50, all from the personal
1999)? Or does recent marsupial ontogeny recapitulate collection of MRSV. We also examined the carpal
(Lovejoy, 2000) the numerous reductions and fusions bones in macerated skeletons of Caluromys philander,
that have ocurred during the evolution of tetrapods and C. derbianus, Didelphis virginiana, and an ontogenetic
mammals? In this context, it is of interest to examine series of M. domestica from the collection of A. van
whether there is evidence of the centralia in early devel-
Nievelt (AVN, Duke University) consisting of eight
opment, given that at least one centrale probably char-
specimens ranging in age from PND-110 to adulthood
acterized the last common ancestor of Theria. There are
(Table 2). Species names used follow Gardner (1993).
reports of centralia for several eutherians early in devel-
opment, that are later lost or fused in development (e.g.,
Leboucq, 1884; Schmidt-Ehrenberg, 1942). However,
some of these reports might have been the result of re-
capitulatory preconceptions superimposed on unclear Table 1. Specimens studied histologically of Monodelphis do-
histological evidence (see Hinchliffe and Griffiths, mestica and Caluromys philander. CRL, crown-rump length; HL,
1983; Shubin and Alberch, 1986). head length; PND, postnatal day; PY, pouch young.
In this study we also wanted to establish whether Mon- Species Age CRL (mm) HL (mm)
odelphis domestica was sexually dimorphic in the
carpal region, as in some close didelphid relatives M. domestica Stage 33 – –
(Jansa and Voss, 2000) reported by Lunde and Schutt M. domestica PND 0 ca. 10 ca. 4.5
(1999). Aspects of growth and the development of sex- M. domestica PND 5 ca. 11.5 ca. 6.5
ually dimorphic features have been studied in marsu- M. domestica PND 12 ca. 20.5 ca. 8.5
C. philander ca. PND 0 ca. 11.5 ca. 6
pial cranial anatomy and body size (e.g., Bergallo and C. philander PY ca. 15 ca. 7.5
Cerqueira, 1994; Maunz and German, 1996) but not
postcranial traits.

Table 2. Macerated skeletons examined of Monodelphis domes-

Materials and methods tica, Caluromys philander, C. derbianus and Didelphis virgini-
ana. See Table 1 and text for abbreviations.
Histological sections of the hand of specimens in the col- Species Age Sex Number
lection of the Zoologisches Institut of the University of
Tübingen (ZS) were examined (Table 1). In most cases M. domestica PND 120 Female AVN 98002
the exact age of the specimens is known, as well as their M. domestica Adult Female AVN 97001
size measured as crown-rump length (CRL) and head M. domestica ca. 2.5 years Female AVN 98005
length (HL). All specimens of Monodelphis domestica M. domestica Adult Female AVN 98011
M. domestica PND 110 Male AVN 98004
were kindly donated to us by K.K. Smith (Duke Univer- M. domestica PND 120 Male AVN 98008
sity). Stage 33 was the earliest (stages 31 and M. domestica PND 132 Male AVN 98014
32 were also examined, see Sánchez-Villagra et al., 2002 M. domestica PND 164 Male AVN 98010
for details) in which carpal elements could be clearly C. philander Adult Female AMNH 234989
identified in the precartilaginous stage. Stage 33 repre- C. philander Adult Male AMNH 234979
sents approximately the second half of the 13th day after C. derbianus Adult Female LACM 25525
D. virginiana Adult – LACM 31042
conception. PND (postnatal day) 0 is the day of birth,

74 Zoology 106 (2003) 1


Institutional abbreviations used are: American Museum
of Natural History (AMNH), Natural History Museum
of the Los Angeles County (LACM), and Zoologische
Sammlung of the Zoologisches Institut at the Univer-
sity of Tübingen (SZ).
Aspects of the biology and growth of Monodelphis do-
mestica and Caluromys philander are discussed by
Saunders et al. (1989) and Atramentowicz (1995) re-
spectively.
Results
Figure 1A shows a schematic drawing of the dorsal
view of the carpal bones and proximal metacarpals in
an adult specimen of Monodelphis domestica. Speci-
mens of Caluromys spp. and Didelphis virginiana show
a similar arrangement, with no deviations in terms of
the number and general arrangement of the carpal
bones of the hypothesized didelphid groundplan as pre-
sented by Szalay (1994, see also Szalay and Sargis,
2001). One important difference is an extra sesamoid
bone located just distal to the radius and proximo-pal-
mar to the scaphoid (Fig. 1B) recorded in adult C. phi-
lander (AMNH 234979), C. derbianus (LACM 25525)
and Didelphis virginiana (LACM 31042), not present
in M. domestica.
Development of carpals in embryonic and
pouch-young Monodelphis domestica
Stage 33 (Figs. 2, 3): The youngest specimen has nine
cell-condensations, eight of which are homologized
with later carpal elements. In this stage we saw no signs
of a prepollex anlage. The 3-D reconstructions do not
Fig. 1. Camera lucida drawings of the right
distal forearm, carpals and proximal
metacarpals in adult specimens of (A) Mon-
odelphis domestica (AvN 98011), dorsal
view, and (B) Caluromys philander
(AMNH 234979), palmar view. Abbrevia-
tions: ha, hamate; lu, lunate; m1–m5,
metacarpals 1–5; ca, capitate; pi, pisiform;
r, radius; sc, scaphoid; td, trapezoid; tm,
trapezium; tr, triquetrum; u, ulna. The aster-
isk (*) points to the extra sesamoid element
in C. philander, described for Didelphis au-
rita by Emery (1897) as ‘Randknochen’.
Zoology 106 (2003) 1
Carpal ontogeny in didelphids
show any major articular surfaces. In the following de-
scription we will use the names of the carpals even
though at this stage these elements are just cell conden-
sations. The sections illustrated in Figure 3 might be
confusing because of the section plane and the fact that
the hand was clenched when fixated, apparently distort-
ing the relationship between the structures.
Both capitate (= magnum) and lunate have a cylindrical
shape. Hamate (= unciform) and scaphoid are the rela-
tively largest elements. The pisiform is in a similar his-
tological condition as the other elements and is egg-
shaped. The scaphoid has a triangular shape, with its
shortest vertice pointing towards the lunate and the cell
condensation identified here as a centrale. The latter is
ca. 50 µ thick and is found only ca. 10 µ away from the
lunate. The triquetrum (= cuneiform) is more or less
cylindrical in the dorso-palmar plane and shows con-
cavities both towards the pisiform and the hamate. Both
trapezium and trapezoid are relatively small and undif-
ferentiated in terms of shape.
Stage PND-0 (Figs. 4, 5): There are nine Anlagen of the
carpals, already in a precartilaginous stage. Some struc-
tures represent the first signs of articular surfaces that
appear later in development.
The capitate has a round vertice pointing to the trape-
zoid. One of the ends of the capitate is found in be-
tween metacarpals 3 and 4. The hamate is the largest el-
ement. The lunate is relatively long (‘stretched’ in the
dorso-palmar plane) and its dorsal end is found in be-
tween a relatively large concavity of the scaphoid. The
cross section of the middle portion of the lunate is cir-
cular, but oval towards its more palmar end. The pre-
pollex is the smallest element and shows a relatively
large distance to the trapezium and scaphoid. The
scaphoid has an almost U-shape from the dorsal view,
75
J. Prochel and M. R. Sánchez-Villagra

Fig. 3. Histological sections of the carpal region in the right hand

of Monodelphis domestica, stage 33 (ca. 1 day before birth). Ab-
breviations: ce, centrale; ha, hamate; lu, lunate; m1–m5,
metacarpals 1–5; ca, capitate; pi, pisiform; r, radius; sc, scaphoid;
Fig. 2. Dorsal (top) and palmar (bottom) views of three-dimen- td, trapezoid; tm, trapezium; tr, triquetrum; u, ulna. Scale, 0.1 mm.
sional reconstructions of portions of the right hand and forearm of
Monodelphis domestica, stage 33 (ca. 1 day before birth). Abbre-
viations: ce, centrale; ha, hamate; lu, lunate; m1–m5, metacarpals
1–5; ca, capitate; pi, pisiform; r, radius; sc, scaphoid; td, trape-
zoid; tm, trapezium; tr, triquetrum; u, ulna. Not to scale.

Fig. 4. Dorsal (top), palmar (2nd row), proximal (3rd row) and distal (bottom) views of three-dimensional reconstructions of the sec-
tioned right carpals of neonate (left) and PND-5 (right) Monodelphis domestica. Abbreviations: ha, hamate; lu, lunate; ca, capitate; pi,
pisiform; pp, prepollex; sc, scaphoid; td, trapezoid; tm, trapezium; tr, triquetrum. Not to scale.

76 Zoology 106 (2003) 1

 Carpal ontogeny in didelphids

Zoology 106 (2003) 1 77

J. Prochel and M. R. Sánchez-Villagra

Fig. 5. Histological sections of the carpal region in the right hand of Monodelphis domestica, neonate (left) and PND-5 (right). Abbrevi-
ations: ha, hamate, lu, lunate, m1–m5, metacarpals 1–5, ca, capitate; pp, prepollex; r, radius; sc, scaphoid; td, trapezoid; tm, trapezium;
tr, triquetrum; u, ulna. Scale for PND-0, 0.1 mm; for PND-5, 0.5 mm.

with its concave end partially enclosing the lunate. The
larger of the two arms of the scaphoid is situated to-
wards the capitate, the short one is found in between the
ulna and the proximal end of the lunate. The triquetrum
shows a concavity that will later become the surface of
articulation with the pisiform.
Stage PND-5 (Figs. 4, 5): There are no signs of ossifi-
cation in any of the carpals yet, but future articular sur-
faces are clearly recognizable. The anlage of all the
processes that characterize the adult carpals are already
present. The dorsal portion of the capitate in between
metacarpal 3 and the distal end of the lunate is rela-

78 Zoology 106 (2003) 1


tively flat. The palmar portion is rounder and is in
close association with the hamate and triquetrum. The
dorsal end of the hamate is relatively large and the later
surfaces of articulation with the capitate and the tri-
quetrum are clearly recognizable. The anlage of the
surfaces of articulation with metacarpals 3 and 4 are
seen for the first time. On the palmar side there is a
small process. The lunate is round on its palmar por-
tion and gracile and elongated towards the dorsal end
of the triquetrum. The pisiform is elongated and has a
more or less circular cross-section, expanded in its dor-
sal-most portion. The prepollex is the smallest element
and has an oval shape when viewed dorsally. Concern-
ing the scaphoid, differences with the neonate are
minor. The palmar process of the scaphoid has become
longer relative to the rest of this element in comparison
to that of the neonate. The concavity of the scaphoid
enclosing the lunate has further developed. The dorsal
portion of the trapezium is round while the distal one is
triangular. Its palmar portion is more robust than its
dorsal one. The trapezoid is dorsally flat, distally it has
a triangular shape. The palmar end of the triquetrum
has a concavity facing the pisiform, separate from that
facing the ulna. A distal curvature fits in an identation
of the hamate.
Stage PND-12: There are no major differences in form
to that recorded in the PND-5 specimen, even though
all elements have significantly increased in size and the
articular surfaces are more sharply defined. Radius and
ulna show clear signs of ossification.
A description with illustrations of histological sections
and 3-D reconstructions of perinatal stages of Caluromys
philander examined are available upon request from the
corresponding author.
Sequence of ossification in the carpals
of Monodelphis domestica
Examination of the cleared and stained specimens al-
lowed us to study the sequence of ossification of the
carpals. The small number of specimens available,
however, limited the resolution though some differ-
ences were recorded. At ca. PND-30 the distal epiph-
ysis of the radius is the largest center of ossification
recorded in this area. As already revealed by histologi-
cal sections and confirmed with a cleared and stained
specimen, by PND-12 no ossification in the carpals had
started. By ca. PND-30 (Fig. 6), one of the specimens
showed ossification of all adult carpals except for the
pisiform, prepollex, trapezoid and trapezium. The next
stage (represented by another specimen of similar age),
showed that ossification of the trapezium has started
(Fig. 6). A later stage in this series of age also ca. PND-
30 showed ossification in the trapezoid. Pisiform and
prepollex ossified later.
Zoology 106 (2003) 1
Carpal ontogeny in didelphids
Sexual dimorphism in Monodelphis domestica
In M. domestica the pisiform is relatively larger and
more robust in adult males than in adult females, with a
disto-palmar thickening in males. This was recorded in
fully grown specimens and not juveniles (Fig. 7). In a
series of males of ages between PND-110 and 164
(Table 2) the pisiform was still relatively small and
slender, as in females of the same age.
Discussion
In very early postnatal life of both species of didelphids
examined, the number and arrangement of the adult
carpals can already be identified, a phenomenon that
characterizes all other marsupials examined to date
(Emery, 1897; Sánchez-Villagra and Döttling, in re-
view). Consistent with the idea that the adult carpal pat-
tern is set early in ontogeny is the fact that at PND 5,
most of the anlage of the processes of the adult carpals
of M. domestica are already present and the articular
surfaces are recognizable.
There is not a single report of the postnatal fusion of
carpal elements in marsupials known to us (Emery,
1897; Schmidt-Ehrenberg, 1942. In placentals, postna-
tal fusion of carpal elements is rare, as expected given
their much more altricial condition than marsupials at
birth (Sánchez-Villagra and Sultan, 2002). Stafford and
Thorington (1998) investigated several ‘Archonta‘ and
found a postnatal fusion of carpal elements only in
some Chiroptera and Dermoptera.
The isolated centrale characteristic of the therian
groundplan is missing in all adult marsupials examined
to date and was hypothesized to have been incorporated
in the scaphoid by Szalay (1994), a good null hypothe-
sis considering the topographical position of these ele-
ments in the therian groundplan. However, in investi-
gating the fate of the therian centrale in Marsupialia, it
may be appropiate to remember the cautionary note of
Romer (1956: 381): ‘Loss of elements from the primi-
tive reptilian carpal (or tarsal) pattern is commonly as-
sumed to be due to fusion. Sometimes this is the case,
but often assumptions of fusion are unwarranted, and
more frequently reductions appear to be due to actual
loss of elements. Embryological evidence is of major
importance, but in some cases evidence claimed on the
basis of precartilaginous or mesenchyme condensations
is inadequate or dubious’. Here we reported a centrale
as a separate mesenchymal condensation in a late pre-
natal specimen of Monodelphis domestica, which was
no longer present in the approximately one day older
neonate specimen. Because of its topographical rela-
tions, it is parsimonious to homologize this element
with the centrale characteristic of the therian ground-
plan. The fate of this mesenchymal element observed in
79
80 Zoology 106 (2003) 1
this early stage of M. domestica is impossible to estab-
lish with the data available. No signs of a separated
centrale or of its fusion with the scaphoid or with any
other carpal has been found in the early postnatal devel-
opment of any of the marsupial species studied to date
(D. viverrinus: Sánchez-Villagra and Döttling, unpub-
lished material; Didelphis aurita, Isoodon obesulus,
Dasyurus hallucatus, Petaurus norfolcensis, Trichosu-
rus vulpecula, Aepyprymnus rufescens, Bettongia
gaimardi, Macropus giganteus, and Phascolarctos
cinereus: Emery, 1897).
For adults, the only deviation from the general didel-
phid carpal pattern (Szalay, 1994) recorded here was
the additional sesamoid element located just distal to
the styloid process of the radius, found in adult speci-
mens of Caluromys philander, C. derbianus and Didel-
phis virginiana. This element has the same position and
topographical relations to that described for Didelphis
aurita by Emery (1897). According to Emery, this ele-
ment appears relatively late in the ontogeny of D. au-
rita; it was not present in the perinatal stages we stud-
ied of C. philander.

Sequence of ossification of the carpals

The relatively late onset of ossification of the pisiform
and prepollex reported here for Monodelphis domestica
characterizes other therians studied to date. In most pla-
cental species for which information is available
(Curgy, 1965; Vogel, 1972), the pisiform is among the
latest carpals to ossify. This is also the case for the mar-
supial Didelphis virginiana (Nesslinger, 1956). In those
placentals in which the prepollex is present and for
which this kind of information is available, this element
is the last one to ossify. This includes Canis familiaris,
Rattus norvegicus and Mus musculus (Curgy, 1965). In-

Fig. 7. Development of sexual dimorphism in the pisiform of c

Monodelphis domestica. (Top) Adult female (AvN 97001) and
(bottom) adult male (AVN 98011) showing the robusticity of the
pisiform in the latter, absent in the (middle) PND-164 (AvN
98010) male. Abbreviations: pi, pisiform; tr, triquetrum; u, ulna.
Scale, 1 mm.

b Fig. 6. Cleared and stained specimens of Monodelphis do-

mestica in left, dorsal and right, palmar view. The three speci-
mens are ca. PND-30/40 old, from top to bottom increasing in
age. Abbreviations: dre, distal epiphysis of the radius; due, dis-
tal epiphysis of the ulna; ha, hamate; lu, lunate; m1–m5,
metacarpals 1–5; ca, capitate; r, radius; sc, scaphoid; td, trape-
zoid; tm, trapezium; tr, triquetrum; u, ulna. Scale, 1 mm.

Zoology 106 (2003) 1 81

J. Prochel and M. R. Sánchez-Villagra
formation about the timing of ossification of the pre-
pollex in comparison to other carpals is not available
for marsupials, but apparently this element is also the
last one among carpals to ossify in Didelphis virginiana
(Nesslinger, 1956).
The late ossification of the pisiform in Monodelphis
domestica is in agreement with the interpretation that
this element is a sesamoid and not a ‘true carpal‘
(Rieppel, 1992), a matter that has been discussed for a
long time in the literature about mammals (see Maier,
1971 for a review). However, relative timing of ossifi-
cation of the pisiform among carpals in mammals is
not uniform, as summarized by Curgy (1965). For ex-
ample, in the rat it is among the first carpals to begin
ossification, in Macaca it is intermediate, and in hu-
mans it is the last. Establishing a hypothetical primitive
pattern for mammals will require consideration of a
relevant sample in a phylogenetic context. No conclu-
sion concerning the homology or evolutionary origin
of the pisiform can be reached based on the relative
timing of its development.
Whereas we found a relatively late ossification of pre-
pollex and pisiform as discussed above, the examina-
tion of histological sections did not reveal any differ-
ences in early ontogeny between these two elements
and all the other carpals. The onset of cell condensation
and chondrification appears to be similar in all carpals
that characterize the adults, and the onset of chondrifi-
cation is also roughly similar. The onset of ossification
though is clearly retarded in prepollex and pisiform as
compared to other carpals, as revealed by the examina-
tion of cleared and stained specimens.
The mechanisms that determine the ossification se-
quence of the carpals or of the skeleton in general are
poorly understood. What determines the onset of ossifi-
cation of a bone in the first place? Skeletal tissues pro-
vide rigidity, support and muscle attachment. The func-
tion of a bone may influence the onset of its ossifica-
tion, as it is known that muscle contractions and move-
ments (intra- or extrauterine) influence bone formation
(Weisel, 1967; Verraes, 1975; Mook, 1977; Morris and
Gaudin, 1982; Amprino, 1985; Hall, 1986; Müller and
Streicher, 1989; Mikic and Carter, 1995; Carter et al.,
1998). If species vary in their hand use for example,
then they may have different sequences of ossification
in the carpals. On the other hand, if the specific order of
ossification of bones is relatively unimportant to devel-
oping functionality of the postcranium, variability may
reflect this lack of constraint. Detailed information re-
garding the temporal sequence of first muscle activity
in the arm of developing mammals is lacking. It is
therefore impossible at present, even when dealing with
well known species, to test whether the sequence of os-
sification and its changes in evolution correspond di-
rectly to muscular activity.
82
Sexual dimorphism in Monodelphis domestica
We reported here for the first time for Monodelphis do-
mestica a sexually dimorphic feature in the postcra-
nium. We found that the pisiform is relatively larger
and more robust in adult males than in adult females, as
was reported by Lunde and Schutt (1999) for several
other species of didelphids. According to Lewis (1989),
a robust pisiform increases the firmness or stability of
the palm of the hand. This could be the case in males of
M. domestica. It remains to be studied whether the
outer form of the hand of M. domestica is sexually di-
morphic, as it is in Marmosops parvidens (Lunde and
Schutt, 1999; fig. 2). If this is the case, then the sexually
dimorphic pisiform of M. domestica is likely to be cor-
related with the copulatory behavior of this species.
The male holds the female during copulation using its
hands, and copulation takes place in a lateral position
(Barnes and Barthold, 1969). Copulatory behavior was
also cited by Lunde and Schutt (1999) to explain simi-
lar sexual dimorphism in other didelphid species.
The sexual dimorphism in the pisiform of M. domestica
develops relatively late in ontogeny, perhaps correlated
with that fact that males of this species have a longer
period of growth than females. Bergallo and Cerqueira
(1994) found that head and body size in males grow in
age classes (based on dentition) in which females have
ceased to grow. Maunz and German (1996) established
that while in females the skull grows until around PND-
228, in males this occurs until PND-380. During the pe-
riod of fixation (ca. the first 12 days of postnatal life)
the pisiform grows mostly in length, remaining rela-
tively slender. Only later in ontogeny (later at least than
ca. PND 164) does the thickness of this element start
developing at a relatively higher rate until achieving its
adult form. It appears that this period of growth is pro-
longed in males in comparison with females.
Lunde and Schutt (1999) reported that Marmosops and
Gracilinanus have an enlarged and modified pisiform,
while Marmosa and Micoureus have an enlarged and
modified prepollex. In this context Monodelphis do-
mestica is similar to the first pair of taxa, even though
the genus Monodelphis is phylogenetically closer to the
second pair (Jansa and Voss, 2000). Consideration of
these traits in the context of didelphid phylogeny (Jansa
and Voss, 2000) implies that there has been either con-
vergence in the evolution of sexual dimorphism in
didelphids or that our current notions of didelphid phy-
logeny need to be revised.
Conclusion
The data presented in this and other papers on the same
topic (Sánchez-Villagra and Döttling, in review; Ham-
rick, 2001) show that there is neither Haeckelian nor von
Zoology 106 (2003) 1

Baerian recapitulation (Lovejoy, 2000) in the develop-
ment of the marsupial hand. Comparison of our results
for two didelphids with those of other mammals
(Schmidt-Ehrenberg, 1942) and tetrapods (Hinchcliffe
and Griffiths, 1983) confirms that there is no common
pattern of condensations or carpal anlage for any of these
groups, not even for Marsupialia (Emery, 1897). As
stated by Hinchcliffe and Griffiths (1983), ‘the identifi-
cation of a condensation archetype has arisen because in-
vestigators have superimposed their preconceptions on
histological evidence which was unclear’. Ontogenetic
data are useful for studying complex characters in time
and adaptations and constraints in development, not in
order to directly establish polarity or discover ancestral
states. In order to understand the phylogenetic transfor-
mations of the synapsid hand, a lot can be learned from
the careful study of adult forms, including fossils, in a
comparative and phylogenetic framework (e.g., Hopson,
1995; Stafford and Thorington, 1998; Ji et al., 2002).
Acknowledgements
For permitting access to materials on which this study
is based, we thank I. Horovitz, R. D. E. MacPhee, W.
Maier, K. K. Smith and A. van Nievelt. We thank T. T.
Fussneger and M. Meinert for histological work, M.
Hoffmann for help with the clearing and staining tech-
nique, A. van Nievelt for help with preparation of mac-
erated skeletons, W. Maier for suggestions, and P.
Lemelin and an anonymous reviewer for constructive
comments that improved the manuscript. This work
was financially supported by the Department of Zool-
ogy of the University of Tübingen, a Research Study
Grant from the Natural History Museum of the Los An-
geles County to MRSV and the German National Sci-
ence Foundation (DFG Grant SA-883/4 to MRSV).
References
Amprino, R. 1985. The influence of stress and strain in the early
development of shaft bones. Anat. Embryol. 172: 49–60.
Atramentowicz, M. 1995. Growth of pouch young in the bare-
tailed woolly opossum, Caluromys philander. J. Mammal. 76:
1213–1219.
Barnes, R. D. and S. W. Barthold. 1969. Reproduction and breed-
ing behaviour in an experimental colony of Marmosa mitis
Bangs (Didelphidae). J. Reprod. Fertil., Supplement 6:
477–482.
Bergallo, H. G. and R. Cerqueira. 1994. Reproduction and growth
of the opossum Monodelphis domestica (Mammalia: Didelphi-
dae) in northeastern Brazil. J. Zool., Lond. 232: 551–563.
Burke, A. C. and P. Alberch. 1985. The development and homol-
ogy of the chelonian carpus and tarsus. J. Morphol. 186:
119–131.
Carter, D. R., B. Mikic and K. Padian. 1998. Epigenetic mechani-
cal factors in the evolution of long bone epiphyses. Zool. J.
Linn. Soc. 123: 163–178.
Zoology 106 (2003) 1
Carpal ontogeny in didelphids
Curgy, J. J. 1965. Apparition et soudure des points d’ossification
des membres chez les mammifères. Mém. Mus. Nat. d’Hist.
Natur. Série A. Zoologie 32: 173–307.
Emery, C. 1897. Zoologische Forschungsreisen in Australien und
dem Malayischen Archipel. In: Monotremen und Marsupialier.
(R. Semon, ed.). Gustav Fischer, Jena: vol. 2, pp. 369–400.
Gardner A. L. 1993. Order Didelphimorphia. In: Mammal
Species of the World, 2nd edition. (D. E. Wilson and D. M.
Reeder, eds.). Smithsonian Institution Press, Washington. pp.
15–23.
Hall, B. K. 1986. The role of movement and tissue interactions in
the development and growth of bone and secondary cartilage
in the clavicle of the embryonic chick. J. Embryol. Exp.
Morph. 93: 133–152.
Hall, B. K. 1999. Evolutionary Developmental Biology. Chap-
man and Hall, New York.
Hamrick, M. W. 2001. Primate origins: evolutionary change in
digital ray patterning and segmentation. J. Human Evol. 40:
339–351.
Hinchliffe, J. R. and P. J. Griffiths. 1983. The prechondrogenic
patterns in tetrapod limb development and their phylogenetic
significance. In: Development and Evolution. Symposium of
the British Society of Developmental Biology, Volume 6 (B.
Goodwin, N. Holden and C. Wylie, eds.). Cambridge Univer-
sity Press, Cambridge, pp. 99–121.
Hopson, J. A. 1995. Patterns of evolution in the manus and pes of
non-mammalian therapsids. J. Vert. Paleontol. 15: 615–639.
Jansa, S .A. and R. S. Voss. 2000. Phylogenetic studies on didel-
phid marsupials I. Introduction and preliminary results from
nuclear IRBP. J. Mammal. Evol. 7: 43–77.
Ji, Q., Z. X. Luo, C.X. Yuan, J. R. Wible, J. P. Zhang and J. A.
Georgi. 2002. The earliest known eutherian mammal. Nature
416: 816–822.
Leboucq, H. 1884. Recherches sur la morphologie du carpe chez
les mammifères. Arch. Biol. 5: 35–102.
Lewis, O. J. 1989. Functional Morphology of the Evolving Hand
and Foot. Clarendon Press, London.
Lovejoy, N. R. 2000. Reinterpreting recapitulation: systematics
of the needlefishes and their allies (Teleostei: Beloniformes).
Evolution 54: 1349–1362.
Luckett, W. P. 1993. An ontogenetic assessment of dental ho-
mologies in the therian mammals. In: Mammal Phylogeny.
Mesozoic Differentiation. Multituberculates, Monotremes,
Early Therians and Marsupials. (F. S. Szalay, M. J. Novacek
and M. C. McKenna, eds.). Springer, New York, pp. 182–204.
Lunde, D. and W. Schutt. 1999. The peculiar carpal tubercles of
male Marmosops parvidens and Marmosa robinsoni (Didel-
phidae: Didelphinae). Mammalia 63: 495–504.
Maier, W. 1971. Vergleichend- und funktionellanatomische Un-
tersuchungen an der Vorderextremität von Theropithecus
gelada (Rüppell 1835). Abh. Senckenb. Naturforsch. Gesell.
527: 1–284.
Maunz, M. and R. Z. German. 1996. Craniofacial heterochrony
and sexual dimorphism in the short-tailed opossum (Monodel-
phis domestica). J. Mammal. 77: 992–1005.
Mikic, B. and D. R. Carter. 1995. Bone strain gauge data and the-
oretical models of functional adaptation. J. Biomechanics 28:
465–469.
Mook, D. 1977. Larval and osteological development of the
sheepshead, Archosargus probatocephalus (Pisces: Sparidae).
Copeia 1977: 126–133.
Morris, S. L. and A. J. Gaudin. 1982. Osteocranial development
in the viviparous surfperch Amphistichus argenteus (Pisces:
Embiotocidae). J. Morphol. 174: 95–120.
83
J. Prochel and M. R. Sánchez-Villagra
Müller, G. B. and J. Streicher. 1989. Ontogeny of the syndesmo-
sis tibiofibularis and the evolution of the bird hindlimb: a
caenogenetic feature triggers phenotypic novelty. Anat. Em-
bryol. 179: 327–339.
Nesslinger, C. L. 1956. Ossification centers and skeletal develop-
ment in the postnatal Virginia Opossum. J. Mammal. 37:
382–394.
Nowak, R. M. 1999. Walker’s Mammals of the World. 6th Ed.
John Hopkins University Press, Baltimore.
Oliveira de, C. A., J. C. Nogueira and B. Mahecha. 1998. Sequen-
tial order of appearance of ossification centers in the opossum
Didelphis albiventris (Didelphidae) skeleton during develop-
ment in the Marsupium. Ann. Anat.180: 113–121.
Rieppel, O. 1992. Studies on skeleton formation in reptiles. I. The
postembryonic development of the skeleton of Cyrtodactylus
pubisulcus (Reptilia: Gekkonidae). J. Zool., Lond. 227:
87–100.
Romer, A. S. 1956. Osteology of Reptiles. University of Chicago,
Chicago.
Sánchez-Villagra, M. R. and F. Sultan. 2002. The cerebellum at
birth in therian mammals, with special reference to rodents.
Brain, Behav. Evol. 59: 101–113.
Sánchez-Villagra, M. R., S. Gemballa, S. Nummela, K. K. Smith
and W. Maier. 2002. Ontogenetic and phylogenetic transfor-
mations of the ear ossicles in marsupial mammals. J. Morphol.
251: 219–238.
Saunders, N. R., E. Adam, M. Reader and K. Møllgård. 1989.
Monodelphis domestica (grey short-tailed opossum): an acces-
sible model for studies of early neocortical development. Anat.
Embryol. 180: 227–236.
Schmidt-Ehrenberg, E. C. 1942. Die Embryogenese des Extrem-
itätenskeletts der Säugetiere. Rev. Suiss. Zool. 49: 33–133.
84
Shubin, N. H. and P. Alberch. 1986. A morphogenetic approach
to the origin and basic organization of the tetrapod limb. Evol.
Biol. 5: 319–381.
Stafford, B. J. and R. W. J. Thorington. 1998. Carpal develop-
ment and morphology in archontan mammals. J. Morphol.
235: 135–155.
Szalay, F. S. 1994. Evolutionary History of the Marsupials and an
Analysis of Osteological Characters. Cambridge University
Press, New York.
Szalay, F. S. and E. J. Sargis. 2001. Model-based analysis of
postcranial osteology of marsupials from the Palaeocene of
Itaborai (Brazil) and the phylogenetics and biogeography of
Metatheria. Geodiversitas 23: 139–302.
Szalay, F. S. and B. A. Trofimov. 1996. The Mongolian late Cre-
taceous Asiatherium, and the early phylogeny and paleobio-
geography of Metatheria. J. Vert. Paleontol. 16: 474–509.
Verraes, W. 1975. Some functional aspects of ossifications in the
cartilaginous ceratohyale during postembryonic development
in Salmo gairdneri Richardson, 1836 (Teleostei: Salmonidae).
Forma et functio 8: 27–32.
Vogel, P. 1972. Vergleichende Untersuchung zum Ontogenese-
modus einheimischer Soriciden (Crocidura russsula, Sorex ara-
neus und Neomys fodiens). Rev. Suiss. Zool. 79: 1201–1332.
Weber, M. 1928. Die Säugetiere, Einführung in die Anatomie und
Systematik der Recenten und Fossilen Mammalia. Gustav Fis-
cher, Jena.
Weisel, G. F. 1967. Early ossification in the skeleton of the
sucker (Catostomus macrocheilus) and the guppy (Poecilia
reticulata). J. Morphol. 121: 1–18.
Yalden, D. W. 1972. The form and function of the carpal bones in
some arboreally adapted mammals. Acta Anatomica 82:
383–406.
Zoology 106 (2003) 1