Content uploaded by Amy Henrici
Author content
All content in this area was uploaded by Amy Henrici on Jan 19, 2016
Content may be subject to copyright.
This article was downloaded by: [Amy Henrici]
On: 09 September 2011, At: 13:47
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of Vertebrate Paleontology
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/ujvp20
An ostodolepid ‘‘microsaur’’ (Lepospondyli) from the
Lower Permian Tambach Formation of central Germany
Amy C. Henrici a , Thomas Martens b , David S Berman a & Stuart S. Sumida c
a Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania, 15213,
U.S.A.
b Abteilung Palääontologie, Museum der Natur, Stiftung Schloss Friedenstein, Gotha,
Germany, D-99867
c California State University San Bernardino, Department of Biology, 5500 University
Parkway, San Bernardino, California, 92407-2307, U.S.A.
Available online: 09 Sep 2011
To cite this article: Amy C. Henrici, Thomas Martens, David S Berman & Stuart S. Sumida (2011): An ostodolepid ‘‘microsaur’’
(Lepospondyli) from the Lower Permian Tambach Formation of central Germany, Journal of Vertebrate Paleontology, 31:5,
997-1004
To link to this article: http://dx.doi.org/10.1080/02724634.2011.596601
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching and private study purposes. Any substantial or systematic
reproduction, re-distribution, re-selling, loan, sub-licensing, systematic supply or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
Journal of Vertebrate Paleontology 31(5):997–1004, September 2011
©2011 by the Society of Vertebrate Paleontology
ARTICLE
AN OSTODOLEPID ‘MICROSAUR’ (LEPOSPONDYLI) FROM THE LOWER PERMIAN
TAMBACH FORMATION OF CENTRAL GERMANY
AMY C. HENRICI,*,1 THOMAS MARTENS,2DAVID S BERMAN,1and STUART S. SUMIDA3
1Carnegie Museum of Natural History, 4400 Forbes Avenue, Pittsburgh, Pennsylvania 15213, U.S.A., henricia@carnegiemnh.org;
2Abteilung Pal¨
aontologie, Museum der Natur, Stiftung Schloss Friedenstein, Gotha, Germany D-99867,
ursauriermartens@t-online.de;
3California State University San Bernardino, Department of Biology, 5500 University Parkway, San Bernardino, California
92407-2307, U.S.A., ssumida@csusb.edu
ABSTRACT—Tambaroter carrolli is a new genus and species of medium-sized ostodolepid ‘microsaur’ from the Lower
Permian Tambach Formation, lowermost formational unit of the Upper Rotliegend, Thuringia, central Germany. Based on
a single skull, it is the first ‘microsaur’ to be described from this formation and the first vertebrate fossil known from the
Tambach Formation outside of the well-known, nearby Bromacker locality. It possesses typical ostodolepid features, including
presence of a recumbent snout, ventral embayment of the cheek, a notch in the posterolateral margin of the tabular, and a
well-developed retroarticular process. It can be distinguished from other ostodolepids by proportional characters and bones
forming the cheek embayment rim. The presence of Tambaroter in Europe, as the first ostodolepid known from outside of the
midcontinent of North America, indicates that the geographic occurrence of this family was more widespread than previously
known.
INTRODUCTION
The Lower Permian Tambach Formation is famous for the
exquisitely preserved tetrapods it has yielded over the past 36
years of excavation at the Bromacker locality in the Thuringian
Forest near Gotha, central Germany. In the spring of 2008, one
of us (T.M.) discovered the remains of fossil vertebrates in loose
blocks of rock on the floor of an excavation site for a new su-
permarket in the Tambach Formation in the village Tambach-
Dietharz, which lies about 1.5 km from the Bromacker locality
(Fig. 1). These fossils, a skull and partial skeleton of a diadec-
tid and a skull of an ostodolepid ‘microsaur,’ represent the first
tetrapods to be found in the Tambach Formation from a site
other than the Bromacker locality. More importantly, the ‘mi-
crosaur,’ which is the subject of this study, is the first of the group
to be discovered in the Tambach Formation.
‘Microsaurs’ are known from the Carboniferous (uppermost
Serpukhovian/lowermost Bashkirian) through the Lower Per-
mian of North America and Europe (Ruta et al., 2003a), being
more diverse in North America and with most of the European
records coming from the Carboniferous. Only three ‘microsaur’
genera are known from the Lower Permian of Europe, the
haspidopareiontid Saxonerpeton Carroll and Gaskill, 1978, and
the brachystelechids Batropetes Carroll and Gaskill, 1971, and
Brachystelechus Carroll and Gaskill, 1978, all from the Lower
Rotliegend of Germany. In their comprehensive review of the
‘Microsauria,’ Carroll and Gaskill (1978) divided the ‘microsaurs’
into two suborders, the Tuditanomorpha and Microbranchomor-
pha. Relatively recent global phylogenetic analyses (Laurin and
Reisz, 1997, 1999; Laurin, 1998; Anderson, 2001, 2007; Ruta
et al., 2003a, 2003b, 2007; Anderson et al., 2008) that have
included ‘microsaurs’ differed in the taxa included, characters,
and results. However, with the exception of one analysis, tree II
in Ruta et al. (2003b:fig. 4), they all agree that ‘Microsauria’ is
*Corresponding author.
paraphyletic. Of the analyses that included a significant sample
of taxa from both suborders, Tuditanomorpha is monophyletic
in three (Anderson, 2001; Ruta et al., 2003a; Anderson et al.,
2008) and Microbrachomorpha is paraphyletic in all (Anderson,
2001, 2007; Ruta et al., 2003a, 2003b; Ruta and Coates, 2007;
Anderson et al., 2008). Most important to this study is that
the analyses including ostodolepids (Anderson, 2001, 2007;
Ruta et al., 2003a, 2003b; Ruta and Coates, 2007; Anderson
et al., 2008) demonstrate Ostodolepidae is monophyletic. In
the phylogenetic analyses conducted by Anderson (2001, 2007;
Anderson et al., 2008), Ostodolepidae is a member of the clade
Recumbirostra Anderson, 2007, which includes gymnarthrids,
pantylids, Rhynchonkos Schultze and Foreman, 1981, and
brachystelechids. In two of these analyses (Anderson, 2001,
2007), ostodolepids are the basal-most clade of Recumbirostra,
whereas in the other (Anderson et al., 2008) it is nested within
Recumbirostra as the sister taxon to the clade of Euryodus +
Cardiocephalus. The phylogenetic analyses of Ruta et al. (2003a,
2003b) and Ruta and Coates (2007), however, do not retrieve
the clade Recumbirostra. In two analyses (Ruta et al., 2003a;
Ruta and Coates, 2007), ostodolepids resolve as the sister taxon
of the clade Rhynchonkos (Cardiocephalus +Euryodus), with
Hapsidopareion,Saxonerpeton,andAsaphestera as successively
more basal sister taxa. Their other analysis (Ruta et al., 2003b)
indicates that ostodolepids are basal to Rhynchonkos,whichin
turn is basal to the representative gymnarthirds.
Until the discovery of the ostodolepid in Germany, the fam-
ily was restricted to four, possibly five, genera from the United
States, from the Lower Permian Wolfcampian and Leonardian of
Texas and Oklahoma. As noted in the following discussion, the
stratigraphy of the Clear Fork Group will follow that of Lucas
(2006), who reinstated the use of Arroyo, Vale, and Choza for-
mations as subdivisions of the Clear Fork Group, which had pre-
viously been dropped by Hentz (1988). Ostodolepis brevispinatus
Williston, 1913, is based on a specimen consisting of seven ar-
ticulated vertebrae and ribs, as well as associated dermal scutes,
997
Downloaded by [Amy Henrici] at 13:47 09 September 2011
998 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 5, 2011
FIGURE 1. Map of Germany with inset showing Thuringian Forest
area, Tambach-Dietharz, and Bromacker Quarry and with outlines of
Tambach Basin and areal extent of Tambach Formation.
from West Coffee Creek, Texas, in the Arroyo Formation, Clear
Fork Group. A nearly complete specimen from most likely the
same horizon and locality was attributed by Case (1929) to O.
brevispinatus because of the close resemblance of their vertebrae
and scutes. At the time Case described this specimen it was unlike
any form previously known, and he speculated that some of the
peculiar features of the skeleton, such as the large orbits, pointed
rostrum, low neural spines, strong limbs and feet, and sharp, con-
ical teeth, were fossorial adaptations that may have enabled this
animal to efficiently capture worms and other soft-bodied forms.
Case (1929) erected a new family, Ostodolepidae, placing it as
an aberrant form of Cotylosauria that exhibited both amphib-
ian and reptilian characters. It was not until 1950 that Romer
(1950) recognized Ostodolepis as a ‘microsaur.’ Over a half of
a century passed before another ostodolepid, Micraroter erythro-
geios Daly, 1973, was described. It was based on a partial skull
with four articulated vertebrae and other partial skull material
from the Hennessey Group of the South Grandfield site, Okla-
homa. A well-preserved, nearly complete skull of an extremely
small ostodolepid genus and species, Nannaroter mckenziei An-
derson et al., 2009, is from the well-known fissure fill deposits
of the Dolese Brothers Quarry at Richards Spur, near Fort Sill,
Oklahoma. These deposits have been long regarded as equiva-
lent to the Arroyo Formation of the Clear Fork Group (Olson,
1991; Sullivan and Reisz, 1999). Recently, however, a radiomet-
ric U-Pb date derived from the speleothem deposits at the Dolese
Brothers Quarry indicates an older Wolfcampian age of 290 Ma
(Woodhead et al., 2010). Woodhead et al. cautioned that cur-
rently accepted biostratigraphic ages for Early Permian fossil-
bearing strata may need to be reconsidered, because of the sig-
nificantly older date for this locality than previous stratigraphic
correlations indicated. Lastly, a nearly complete specimen clearly
assignable to Ostodolepidae was collected by Kitching in 1965
from the Arroyo Formation, Clear Fork Group of West Coffee
Creek, Texas. It may possibly be from the same locality as the
holotype of O. brevispinatus, but its affinities at or below the fam-
ily level are problematic (Carroll and Gaskill, 1978).
In a review of the ‘microsaurs,’ in which Carroll and Gaskill
(1978) discussed taxonomic problems of the Ostodolepidae, they
were not able to determine whether or not the two specimens at-
tributed to O.brevispinatus were conspecific. Furthermore, be-
cause they were unable to determine whether the holotype of
O. brevispinatus could be diagnosed at either the generic or spe-
cific level, they chose to retain the holotypic name, because it is
“... the only long-recognized name by which the group has been
known” (Carroll and Gaskill 1978:76). The nearly complete spec-
imen described by Case (1929) as O.brevispinatus was consid-
ered a new genus and species, Pelodosotis elongatum Carroll and
Gaskill, 1978. They further concluded that the specimen collected
by Kitching was definitely a different genus than the newly named
Pelodosotis, but that it shared some features with the holotype of
Micraroter erythrogeios. Thus, in their description of the Kitching
specimen they assigned it to Micraroter. They noted, however,
that the holotype of Micraroter is considerably smaller and more
juvenile than the more complete Kitching specimen, and sug-
gested that their apparent different ontogenetic stage could ac-
count for the observed differences. Schultze and Foreman (1981),
however, argued that the holotype of Micraroter and the Kitching
specimen represent different ostodolepid genera based on cranial
and vertebral characters. Because none of the differences, other
than size, between the two specimens can be attributed to ontoge-
netic differences, we accept the conclusions of Schultze and Fore-
man (1981), and thus the holotype of Micraroter and the Kitch-
ing specimen will be treated here as separate taxa, with the latter
specimen referred to by its catalog number of BPI 3839.
Institutional Abbreviations—BPI, Bernard Price Institute for
Paleontological Research, University of the Witwatersrand, Jo-
hannesburg; FM, The Field Museum, Chicago; MNG,Museum
der Natur, Gotha.
Anatomical Abbreviations—a, angular; ar, articular; bo,ba-
sioccipital; d,dentary;f, frontal; h, hyoid; j, jugal; l,lacrimal;m,
maxilla; n, nasal; p,parietal;pf, postfrontal; pm, premaxilla; po,
postorbital; pp, postparietal; prf, prefrontal; ps, parasphenoid; pt,
pterygoid; q, quadrate; qj, quadratojugal; sa, surangular; scl,scle-
rotic plates; se, sphenethmoid; sm, septomaxilla; sp, splenial; so,
supraoccipital; spp, postspenial; sq, squamosal; st,stapes;t, tabu-
lar; to, temporal ossicles.
STRATIGRAPHIC OCCURRENCE AND AGE
The Tambach Formation is the lowermost formational unit of
the Upper Rotliegend Group or Series and consists of sediments
deposited in an upland, internally drained basin. It consists of
three informal units: the lower conglomerate or Bielstein con-
glomerate, the Tambach Sandstone, the level the Bromacker fos-
sil locality lies in, and the upper conglomerate or Finsterbergen
conglomerate. The two conglomerate units, hereafter referred to
as lower and upper conglomerates, can be distinguished by their
Downloaded by [Amy Henrici] at 13:47 09 September 2011
HENRICI ET AL.—NEW OSTODOLEPID FROM GERMANY 999
clast type, which were derived from different sources. Rhyolite
clasts from the Oberhof uplift, located south and east of the Bro-
macker locality, occur in the 40 to 125 m thick lower conglomer-
ate. The upper conglomerate, which is 10 to 50 m thick, contains
crystalline clasts derived from the Ruhla crystalline uplift located
northwest of the Bromacker locality in addition to rhyolite clasts
from the Oberhof uplift. A conglomerate near the top of the
exposed section at the discovery site contains crystalline clasts,
indicating that the exposed strata are of the upper conglomer-
ate, most likely from the lowermost portion of this unit. The fos-
sils from the nearby Bromacker locality occur in the uppermost
level of the Tambach Sandstone (Berman et al., 1998; Eberth
et al., 2000), which is considered to be Wolfcampian (Berman
and Martens, 1993; Sumida et al., 1996, 1998, 2004; Berman et
al., 1998, 2000a, 2000b, 2001, 2004; Klembara et al., 2005). Inas-
much as this unit has an interfingering contact with the upper con-
glomerate (Berman and Martens, 1993; Eberth et al., 2000), the
new ‘microsaur’ is most likely contemporaneous with or slightly
younger than the Bromacker assemblage and, therefore, should
be regarded as also Wolfcampian age.
SYSTEMATIC PALEONTOLOGY
LEPOSPONDYLI Zittel, 1888
OSTODOLEPIDAE Romer, 1945
TAMBAROTER, gen. nov.
Type Species—Tambaroter carrolli, sp. nov.
Etymology—Tamb- refers to the Tambach Formation in which
the holotype was found and the Greek aroter, a plowman.
Diagnosis—As for type and only species.
TAMBAROTER CARROLLI,n.sp.
(Figs. 2–6)
Etymology—In honor of Robert L. Carroll’s seminal work on
microsaurs.
Holotype—MNG 14708, a nearly complete skull.
Type Locality and Horizon—An excavation site for a mar-
ket in Tambach-Dietharz, Thuringia, Germany. Lower Permian
Tambach Formation, lowermost portion of upper conglomerate,
Wolfcampian. Exact locality information on file at MNG.
Diagnosis—A medium-sized ostodolepid that differs from
other members of the family in the apomorphies of postorbital
length greater than preorbital length and postorbital forms entire
posterior margin of orbit. Differs from all ostodolepids except
possibly BPI 3839 where these characters are questionable and
can no longer be verified (see Discussion): quadratojugal contacts
or closely approaches postorbital and rim of ventral embayment
of cheek formed by jugal, postorbital, and quadratojugal. With
the exception of Micraroter, it differs in prefrontal contacting nar-
ial margin and parietal having a greater length than width. With
the exception of Pelodosotis, it differs in having a long maxilla.
DESCRIPTION AND COMPARISONS
Skull
The skull (Figs. 2–3) is missing the right anterolateral corner,
most of the occiput, and the posterior end of the right mandible.
Strong dorsoventral compression has caused the right side of the
skull, with the exception of the maxilla, to be splayed outward
at roughly 45◦from the vertical, and the left side to be strongly
compressed and/or folded beneath the skull table. The premax-
illae, maxillae, and mandibles are tightly occluded to the upper
jaw, and lie in a horizontal to near-horizontal plane underneath
the skull roof. Despite imperfection of preservation, an accurate
reconstruction of the skull in lateral view can be given (Fig. 4).
The sutures are easily discerned and, although they are closely
united, they are not strongly interdigitating, suggesting a young
TABLE 1. Tambaroter carrolli, skull measurements in millimeters.
Length Width Height
Skull 30.5
Orbit 7.26.4
Preorbital 9.9
Postorbital 11.5
Interorbital 1.1
Nasal 8.24.9
Frontal 9.24.9
Parietal 9.46.0
adult stage of development. In dorsal view the skull has the out-
line of an isosceles triangle. Sculpturing is strongest on the pre-
maxillae, consisting of pits, grooves, and a few ridges that occur
on the anterior face of the bone. Elsewhere, sculpturing occurs
as small pits and very fine grooves and/or ridges radiating from
centers of ossification. The narial openings are anteroposteriorly
elongated and positioned at the tip of the snout. The subcircular
orbits lie near the midlength of the skull, and in the less distorted
right orbit the length is greater than the height. A narrow cir-
cumorbital ridge of variable development is present, being most
prominent on the lacrimals and postorbitals, barely discernable
on the postfrontals, and indeterminate on the jugal due to poor
preservation. The postorbital length of the skull is greater than
the preorbital length (Table 1). The nasals are shorter than the
frontals, which, in turn, are shorter than the parietals. The length
of the parietals exceeds their width, and a small pineal opening
lies just anterior of their midlength. It should be noted that con-
tra Carroll and Gaskill (1978), an extremely small pineal opening
is present in BPI 3839.
The premaxillae are ventrally exposed, with the exception of
short, narrow dorsal processes that overlap the dorsal surface of
the nasals. The anterior tip of the skull likely overhung the pre-
maxillary dentition, as in other ostodolepids. Anteriorly the pre-
maxillae form broad, subrectangular plates that narrow abruptly
in width along the midline of the skull to form the dorsal pro-
cesses. Five teeth are present in the left premaxilla and four plus
one empty alveolus in the right. The premaxillae form the ante-
rior and most of the ventral border of the external naris. As seen
in the less distorted right side of the skull, the slender maxilla ex-
tends from the posteroventral corner of the naris to the level of
the posteroventral corner of the orbit. Thirteen teeth and possi-
bly three additional tooth positions are present in the left maxilla,
whereas the right maxilla is incomplete posteriorly, preventing
an accurate tooth count. Tips of the premaxillary and maxillary
teeth are missing. For their entire preserved length the teeth are
parallel sided, and their cross-sectional outline describes an an-
teroposteriorly elongated oval.
The lacrimal is a substantial bone that extends from the or-
bit to the external narial opening. The right lacrimal is well pre-
served, although displacement has resulted in it laterally over-
lapping the ventral margin of the prefrontal, whereas the left
lacrimal has been dorsoventrally compressed. The prefrontal also
extends from the orbital margin to the external narial margin,
though its entrance to the narial margin is limited to a small dor-
sal contribution. With the exception of Micraroter, the prefrontal
does not reach the narial margin in other ostodolepids. The nasal
forms the majority of the dorsal margin of the narial opening and
the premaxilla and maxilla form the ventral margin.
The left septomaxilla, which is considerably better preserved
and exposed than the right, is crescent-shaped and occupies the
posterior half of the narial opening. Its vertical component has
the form of a slender, slightly posteriorly bowed pillar of smooth
bone that expands at its dorsal end. A horizontal component ex-
tends anteriorly from the vertical component as a sheet of bone
to the midlength of the narial opening. A septomaxillary canal
Downloaded by [Amy Henrici] at 13:47 09 September 2011
1000 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 5, 2011
FIGURE 2. Tambaroter carrolli, holotype (MNG 14708). Photograph in
(A) dorsal, (B) lateral, and (C) ventral views.
extending anteroposteriorly through the septomaxilla at the level
of the junction of the vertical and horizontal components is ab-
sent. Posterior to the septomaxilla is an opening formed mainly
by the broadly concave anterior margin of the lacrimal, with the
prefrontal forming the dorsal border. A similar opening has been
reported in other ostodolepids. Daly (1973) described its occur-
rence in Micraroter and theorized that it was either filled by a
facial tab of the septomaxilla or housed a marking gland or sen-
sory organ as in caecilians. Carroll and Gaskill (1978) described
this opening in BPI 3839 as a foramen. In their reconstruction
of this skull, however, two foramina are depicted, both of which
are bounded by the lacrimal and septomaxilla. Pelodosotis is too
crushed in this region to determine the presence of a foramen
(Anderson et al., 2009). Nannaroter also possesses two foramina
FIGURE 3. Tambaroter carrolli, holotype (MNG 14708). Outline draw-
ing in (A) dorsal, (B) lateral, and (C) ventral views.
of which the ventral opening is bounded by the lacrimal, sep-
tomaxilla, and maxilla and the dorsal by the lacrimal and sep-
tomaxilla (Anderson et al., 2009). Anderson et al. (2009) spec-
ulated that, although their function is unknown, they could be
analogous to the caecilian tentacle. An opening, the septomaxil-
lary foramen, occurs in a similar position in Dimetrodon limbatus
and other non-mammalian synapsids (Wible et al., 1990:fig. 4A,
B), and it is bounded by the nasal dorsally and the maxilla pos-
teriorly. Although its function is unknown, Wible et al. (1990)
theorized that branches of the superior labial nerve and artery
passed through based on comparisons with living monotremes.
This opening in ostodolepids occupies the same position as the
septomaxillary foramen does in primitive synaspids, indicating
that it most likely accommodated arteries and/or nerves.
Downloaded by [Amy Henrici] at 13:47 09 September 2011
HENRICI ET AL.—NEW OSTODOLEPID FROM GERMANY 1001
FIGURE 4. Reconstruction of Tambaroter carrolli in left lateral view.
The lacrimal forms the anterior orbital rim, and its incomplete
dorsal process likely had also a minor contribution to the dorsal
rim. The prefrontals and postfrontals exclude the frontals from
the orbital margin. Due to the presence of a prominent dorsal
process of the postorbital, the postfrontal has a relatively nar-
row entrance onto the dorsal orbital margin. A well-developed
ventral process of the postorbital restricts the jugal to only the
orbital ventral rim, rather than extending to the posteroventral
corner of the orbit, as in other ostodolepids. The lacrimal and
jugal have subequal contributions to the ventral orbital rim. Al-
though a small gap separates them, the maxilla lacks a small pro-
cess that in Micraroter and BPI 3839 fills this gap. Plates of bone
of the palpebral cup occupy the dorsal-most portion of the orbit,
lying adjacent to the prefrontal, postfrontal, and dorsal processes
of the postorbital. Each palpebral cup consists primarily of one
elongate plate, with one or two smaller plates posterior to it. The
plates fit tightly together and closely conform to the orbital mar-
gin, strongly suggesting that they are preserved in their proper
positions. A few additional, smaller plates are displaced nearby
in each orbit.
The postorbital contacts the postfrontal, jugal, tabular,
squamosal, and possibly the quadratojugal. The ventral margin
of the postorbital cheek region is not well preserved, but a ven-
tral embayment of the cheek was clearly present that was formed
by the jugal, postorbital, and quadratojugal (Figs. 5–6). As can
be seen on the left side of the skull, the jugal extends ventral to
and is slightly overlain by the ventral margin of the anterior half
of the postorbital. At the point where the jugal terminates poste-
riorly, the ventral margin of the postorbital broadens as finished
bone to contribute to the embayment rim. Crushing has caused
the quadratojugal to lie in a near-horizontal plane. It is longer
FIGURE 5. Tambaroter carrolli, holotype (MNG 14708). Photograph of
posterior half of skull in lateral view.
FIGURE 6. Tambaroter carrolli, holotype (MNG 14708). Outline draw-
ing of posterior half of skull in lateral view.
than those of other ostodolepids and either contacts or closely
approaches the postorbital, but this cannot be determined with
certainty due to the presence of small chips of bone in this re-
gion. As preserved, posterior of its midlength the quadratojugal
occupies roughly the same horizontal plane as the ventral rim of
the maxilla, whereas its anterior end is angled anterodorsally at
about 45◦. This configuration is thought to conform to the rim of
the embayment. Small ossicles fill the space of the embayment at
the level of the intersection of the jugal, postorbital, and quadra-
tojugal. Identical ossicles are also present in BPI 3839, and Car-
roll and Gaskill (1978) speculated that they were embedded in
tissue outside of the embayment to protect exposed muscles. An-
derson et al. (2009) also noticed the presence of small ossicles in
the embayment of Nannaroter.
The tabular is a large element of the posterolateral corner of
the skull table that extends onto the cheek and occipital regions.
Crushing, however, has resulted in its lateral-most portion lying
in a subhorizontal plane to the skull table, rather than extend-
ing laterally onto the postorbital cheek region, and to overlap the
dorsal portion of the squamosal, instead of having a marginal, su-
tural contact. The tabular and postparietal are incomplete along
their posterior margins, but enough remains to indicate that they
extended far onto the occipital surface of the skull. As in other
ostodolepids, a notch occurs in the ventrolateral margin of the
tabular that may have, according to Carroll and Gaskill (1978),
allowed for passage of the lateral head vein.
The quadrate, exposed in lateral, posterior, and ventral as-
pects of the skull, is preserved in articulation with the lower jaw.
In lateral view the quadrate forms an anteroposteriorly narrow
condyle, whereas in posterior view it clearly exhibits two condy-
lar facets, a lateral one facing ventrally and a medial one facing
ventrolaterally. The angle between them is slightly greater than
90◦, but distortion of the skull may have altered the true angle.
The palate, which has suffered considerable distortion, was
prepared mainly on the left side to preserve the structural in-
tegrity of the specimen. The tightly occluded lower jaws cover
the ectopterygoid, palatine, and vomer. The sphenethmoid no
longer extends along the midline of the skull, having rotated lat-
erally so as to displace its anterior end to the right of the midline.
The parasphenoid has several breaks and does not completely
lie in a horizontal plane. The pterygoid has undergone the great-
est distortion, with the portion forming the medial border of the
subtemporal fenestra having been folded ventrally over on itself.
The broad palatal ramus of the pterygoid, whose ventral surface
is covered in tiny denticles, contacts its mate anterior to the cul-
triform process of the parasphenoid. The quadrate ramus of the
pterygoid is short, terminating in a broad, flange-like process for
articulation with the quadrate.
Downloaded by [Amy Henrici] at 13:47 09 September 2011
1002 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 5, 2011
The parasphenoid, missing its left posterolateral corner, con-
forms in shape to that of other ostodolepids, consisting of a stout
cultriform process and a broad posterior plate. A field of denti-
cles on a slightly elevated platform on the anterior portion of the
parasphenoid plate continues anteriorly onto a well-developed
ridge of the cultriform process, ending near the end of the pro-
cess. The short basipterygoid processes extend laterally and are
slightly disarticulated from the posteroventrally directed basal
processes of the pterygoid. A process extends laterally from the
parasphenoidal plate and extends underneath the left pterygoid
and mandible. Its proximal portion is covered in smooth bone,
except distally where it grades into roughened bone, making it
difficult to discern from the surrounding matrix. A small oval hole
pierces the parasphenoidal plate at the base of the cultriform pro-
cess. It is interesting to note that several specimens of the gym-
narthrid ‘microsaur’ Euryodus (FM-UR 2296, 1566, 1567) have a
similar opening in the anterior portion of the basal plate. It is not
clear what function the opening may have served.
The sphenethmoid is mostly covered by the parasphenoid, so it
cannot be determined if it is paired, as in Pelodosotis (Case, 1929;
Carroll and Gaskill, 1978). A description of the sphenethmoid in
Nannaroter (Anderson et al., 2009) makes no mention of it being
paired, and it is assumed that it was not. On the basis of what is
visible, the sphenethmoid of Tambaroter forms an elongate, sub-
rectangular plate that underlies the frontals and parietals. The
basioccipital in Tambaroter is incompletely preserved, missing its
right side. The remaining portion indicates a narrow, distally ta-
pering element, in contrast to that of other ostodolepids where it
is broad. Posterior of the basioccipital is an incomplete bone that
most likely is the supraoccipital. A bone that appears to be artic-
ulated with it could be the distal portion of the right postparietal
that has been broken off and folded underneath the skull, expos-
ing the external surface of both the supraoccipital and postpari-
etal. Another interpretation, but far less likely, is that the bone
lying adjacent to the supraoccipital is the exoccipital. The proxi-
mal end of the stapes is visible adjacent to the quadrate ramus of
the left pterygoid. All that can be said about it is that it is broad
and tapers in width distally.
Mandible
The left mandible is complete, whereas the posterior third of
the right mandible is missing. Both mandibles are tightly oc-
cluded and lie in a near-horizontal plane, obscuring their me-
dial surface and dentary teeth. An unusual feature of the lower
jaw is a shift of the symphysial contact between the dentaries
and splenials to the left side of the skull midline. A splenial
and a postsplenial are present along the ventral margin of the
mandible, with the latter being slightly longer. The angular ex-
tends anteriorly to about the midlength of the mandible, with the
surangular forming a wedge between it and the dentary. That a
raised coronoid process formed by the dentary and surangular is
present is indicated by the dorsal extension of these bones and
the steep, nearly vertical angle of the posterior edge of the suran-
gular (Figs. 5–6). Bones of the skull, however, overlie the dorsal
end of the coronoid process, so its full dorsal extent cannot be
determined. The articular and quadrate are preserved in articu-
lation. The near-vertical slope of the raised coronoid region and
a small raised lip on the posterior margin of the articular facet
forms an anteroposteriorly narrow cup-like cotyle for reception
of the quadrate. A distinct, quadrangular retroarticular process
that extends posteriorly well beyond the jaw joint is formed by
the articular with a lateral sheathing of the angular (Figs. 5–6).
All other ostodolepids, excepting Nannaroter,inwhichthelower
jaws are incomplete posteriorly, possess a strong retroarticular
process, but the preservation and/or exposure are not as good
as in Tambaroter. The retroarticular process is best preserved in
the right mandible of Pelodosotis, but the lateral bone is eroded
away, so that the posterior extent of the angular cannot be de-
termined. Carroll and Gaskill (1978) reconstructed the angular
as forming a posteriorly directed wedge that does not reach the
posterior end of the lateral surface of the retroarticular process.
Hyoid
An elongate bone lying adjacent to the anterior end of the cul-
triform process is most likely a hyoid bone. One of similar shape
has also been identified in Pelodosotis (Carroll and Gaskill, 1978).
DISCUSSION
Tambaroter possesses a number of characters that, in combina-
tion, readily identifies it as a member of Ostodolepidae: (1) sculp-
turing on the skull roof is relatively light and consists of scattered
shallow grooves and fine pits; (2) snout overhangs the premaxil-
lary tooth row, giving it a subterminal mouth or recumbent ros-
trum; (3) small pineal opening is present, which is small to minis-
cule in other ostodolepids; (4) postfrontal contributes only to the
dorsal orbital margin; (5) ventral embayment of cheek present
as in other ostodolepids, although the bones contributing to the
ventral rim vary among the taxa; (6) squamosal, tabular, and post-
parietal extend far onto the occipital surface; (7) tabular bears
a notch on its ventrolateral margin; and (8) presence of a well-
developed retroarticular process.
Two, possibly four, unique characters distinguish Tambaroter
from other ostodolepids: (1) postorbital length is greater than
preorbital length, in contrast to Micraroter,Nannaroter,Pelo-
dosotis, and BPI 3839 in which the preorbital length is greater
than the postorbital length; (2) postorbital forms the entire pos-
terior orbital rim. In other ostodolepids, with the exception of
Micraroter in which this character cannot be determined due to
incompleteness of the skull, the jugal forms the posteroventral
corner of the orbital rim; (3) the quadratojugal either contacts or
very closely approaches the postorbital (in Tambaroter asmall
overlying piece of bone prevents determination of their relation-
ship with certainty). In Pelodosotis the squamosal intervenes be-
tween the quadratojugal and postorbital, whereas the quadrato-
jugal is not preserved in Micraroter and has not been identified in
Nannaroter and may be absent (Anderson et al., 2009). The sta-
tus of this character in BPI 3839 is uncertain; Carroll and Gaskill
(1978) state that the quadratojugal appears to reach the jugal on
the right side of the skull but not on the left side. Their speci-
men drawing of the left side of the skull shows the quadratojugal
having a narrow contact with the postorbital but not the jugal.
However, their reconstruction indicates that the quadratojugal
does not contact either the jugal or postorbital but instead con-
tacts the squamosal, as in Pelodosotis. Preparation of the speci-
men apparently done since their study has removed the proximal
end of the right quadratojugal, so this character cannot be veri-
fied. The right quadratojugal extends dorsally and distally under-
lies the squamosal, so it cannot be determined if it would have
contacted the postorbital in an undistorted specimen; (4) a ven-
tral embayment of the cheek is present as in all ostodolepids, but
the bones contributing to the ventral margin vary. In Tambaroter
the quadratojugal, postorbital, and jugal contribute to the ventral
margin of the embayment. The same pattern may occur in BPI
3839 but cannot be verified, as discussed above. In Pelodosotis the
ventral margin of the embayment is formed by the jugal, postor-
bital, squamosal, and quadratojugal. This portion of the skull in
Micraroter is too incomplete to allow for comparison. The condi-
tion in Nannaroter is not definitely known due to incompleteness
of the skull posteriorly. Anderson et al. (2009), however, theo-
rized that the squamosal did not participate in the embayment
and the quadratojugal was absent, resulting in the jugal, postor-
bital, and tabular forming the ventral border of the embayment.
Downloaded by [Amy Henrici] at 13:47 09 September 2011
HENRICI ET AL.—NEW OSTODOLEPID FROM GERMANY 1003
RELATIONSHIPS
Determination of ostodolepid intrarelationships is problem-
atic, because, although each genus appears to be unique, po-
tential synapomorphies are not numerous and some cannot be
verified in the available specimens. Anderson et al. (2009) plan
to revise the family, so a phylogenetic analysis will not be con-
ducted here, but potential synapomorphies between Tambaroter
and other ostodolepids can be discussed. Asaphestera,Saxoner-
peton,andHapsidopareion were chosen as an outgroup to de-
termine polarity of these characters based on their phylogenetic
position relative to ostodolepids in the phylogenies of Anderson
(2001, 2007; Anderson et al., 2008) and Ruta et al. (2003a, 2003b;
Ruta and Coates, 2007). One derived character is possessed by
Tambaroter,Micraroter, and BPI 3839, in which a small gap sepa-
rates the lacrimal and jugal. Tambaroter possibly shares two other
derived characters with BPI 3839, but they unfortunately are no
longer preserved in BPI 3839 and may not be valid (see Discus-
sion): quadratojugal contacts postorbital, and the ventral embay-
ment rim is formed by the quadratojugal, postorbital, and jugal.
The latter character appears to go through successively more-
derived states in Pelodosotis and Nannaroter in which the em-
bayment rim is formed by the jugal, postorbital, squamosal, and
quadratojugal in the former and jugal, postorbital, and tabular in
the latter. In Pelodosotis and Tambaroter the premaxilla forms
most of the ventral rim of the narial opening, a derived charac-
ter, rather than roughly the anterior half as in other ostodolepids.
Specimens of both genera are dorsoventrally compressed in the
snout region, making it difficult to determine whether this char-
acter is real or the result of crushing. Another derived charac-
ter possessed by Tambaroter and Pelodosotis is a long maxilla
that closely approaches the posteroventral corner of the orbit,
whereas in other ostodolepids the maxilla is significantly shorter.
Tambaroter and Micraroter possess the derived character of pre-
frontal reaches the narial opening. If only the unequivocal de-
rived characters are considered, then Tambaroter shares with BPI
3839 and Micraroter the lacrimal and jugal being separated by
a small gap, with Micraroter the prefrontal reaches the narial
opening, and with Pelodosotis a long maxilla. Tambaroter does
not share any derived characters with Nannaroter. Thus, because
three of the six derived characters discussed above are doubt-
ful or cannot be verified, the relationship of Tambaroter to other
ostodolepids cannot be resolved at this time.
CONCLUSIONS
Tambaroter carrolli is a new genus and species of Ostodolep-
idae from the Lower Permian, Wolfcampian, Tambach Forma-
tion, the lowermost formational unit of the Upper Rotliegend,
of central Germany. The only other Wolfcampian occurrence
of an ostodolepid is Nannaroter mckenziei from Richards Spur,
Oklahoma; the other known ostodolepids are from the younger
Leonardian levels in Oklahoma and Texas. The occurrence of the
oldest known ostodolepids in eastern and western Euramerica in-
dicate that ostodolepids had a longer geologic history and were
more diverse than previously recorded. Finally, Tambaroter rep-
resents the youngest ‘microsaur’ known from Europe, as all other
Lower Permian ‘microsaurs’ from that region are from the older
Lower Rotliegend.
ACKNOWLEDGMENTS
This project was supported by the Deutsche Forschungsge-
meinschaft (MA 1472/5–1) to T.M. for field work and by the
Carnegie Museum of Natural History. The senior author is grate-
ful to Dr. Jason Anderson (University of Calgary) who allowed
her to study specimens pertinent to this study that were on loan to
him and for his hospitality during her visit. M. Klinger (Carnegie
Museum of Natural History) is acknowledged for his expert as-
sistance with the figures. We also thank J. Clack, an anonymous
reviewer, and Editor J. Anderson for their helpful comments and
suggestions.
LITERATURE CITED
Anderson, J. S. 2001. The phylogenetic trunk: maximal inclusion of taxa
with missing data in an analysis of the Lepospondyli (Vertebrata,
Tetrapoda). Systematic Biology 50:170–193.
Anderson, J. S. 2007. Incorporating ontogeny into the matrix: a phy-
logenetic evaluation of developmental evidence for the origin of
modern amphibians; pp. 182–227 in J. S. Anderson and H.-D. Sues
(eds.), Major Transitions in Vertebrate Evolution. Indiana Univer-
sity Press, Bloomington, Indiana.
Anderson, J. S., D. Scott, and R. R. Reisz. 2009. Nannaroter mck-
inziei, a new ostodolepid ‘microsaur’ (Tetrapoda, Lepospondyli, Re-
cumbirostra) from the Early Permian of Richards Spur (Ft. Sill), Ok-
lahoma. Journal of Vertebrate Paleontology 29:379–388.
Anderson, J. S., R. R. Reisz, D. Scott, N. B. Frobisch, and S. S. Sumida.
2008. A stem batrachian from the Early Permian of Texas and the
origin of frogs and salamanders. Nature 453:515–518.
Berman, D. S., and T. Martens. 1993. First occurrence of Seymouria
sanjuanensis (Amphibia: Batrachosauria) in the Lower Permian
Rotliegend of central Germany. Annals of the Carnegie Museum
62:63–79.
Berman, D. S., S. S. Sumida, and T. Martens. 1998. Diadectes (Diadecto-
morpha: Diadectidae) from the Early Permian of central Germany,
with a description of a new species. Annals of Carnegie Museum
67:53–93.
Berman, D. S., A. C. Henrici, S. S. Sumida, and T. Martens. 2000a.
Redescription of Seymouria sanjuanensis (Seymouriomorpha) from
the Lower Permian of Germany based on complete, mature spec-
imens with a discussion of paleoecology of the Bromacker lo-
cality assemblage. Journal of Vertebrate Paleontology 20:253–
268.
Berman, D. S., R. R. Reisz, T. Martens, and A. C. Henrici. 2001. A
new species of Dimetrodon (Synapsida: Sphenacodontidae) from
the Lower Permian of Germany records first occurrence of genus
outside of North America. Canadian Journal of Earth Sciences
38:803–812.
Berman, D. S., A. C. Henrici, R. A. Kissel, S. S. Sumida, and T. Martens.
2004. A new diadectid (Diadectomorpha), Orobates pabsti,fromthe
Early Permian of central Germany. Bulletin of the Carnegie Mu-
seum of Natural History 35:1–36.
Berman, D. S., R. R. Reisz, D. Scott, A. C. Henrici, S. S. Sumida, and T.
Martens. 2000b. Early Permian bipedal reptile. Science 290:969–972.
Carroll, R. L., and P. Gaskill. 1971. A captorhinomorph reptile from the
Lower Permian of Europe. Journal of Paleontology 45:450–463.
Carroll, R. L., and P. Gaskill. 1978. The order Microsauria. Memoirs of
the American Philosophical Society 126:1–211.
Case, E. C. 1929. Description of a nearly complete skeleton of
Ostodolepis brevispinatus Williston. Contributions from the Mu-
seum of Paleontology, The University of Michigan 3:81–107.
Daly, E. 1973. A Lower Permian vertebrate fauna from southern Okla-
homa. Journal of Paleontology 47:562–589.
Eberth, D. A., D. S Berman, S. S. Sumida, and H. Hopf. 2000. Lower Per-
mian terrestrial paleoenvironments and vertebrate paleoecology of
the Tambach Basin (Thuringia, central Germany): the upland holy
grail. Palaios 15:293–313.
Hentz, T. F. 1988. Lithostratigraphy and Paleoenvironments of Upper Pa-
leozoic continental red beds, north-central Texas: Bowie (new) and
Wichita (revised) groups. Bureau of Economic Geology, University
of Texas at Austin, Report of Investigation 170:1–55.
Klembara, J., D. S Berman, A. C. Henrici, and A. J. ˇ
Cer ˇ
nansk´y. 2005.
New structures and reconstruction of the skull of the seymouri-
amorph Seymouria sanjuanensis Vaughn. Annals of the Carnegie
Museum 74:217–224.
Laurin, M. 1998. The importance of global parsimony and historical bias
in understanding tetrapod evolution. Part I. Systematics, middle ear
evolution, and jaw suspension. Annals des Sciences Naturelles, Zo-
ologie 19:1–42.
Laurin, M., and R. R. Reisz. 1997. A new perspective on tetrapod phy-
logeny; pp. 9–59 in S. S. Sumida and L. M. Martin (eds.), Amniote
Origins: Completing the Transition to Land. Academic Press, San
Diego, California.
Downloaded by [Amy Henrici] at 13:47 09 September 2011
1004 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 5, 2011
Laurin, M., and R. R. Reisz. 1999. A new study of Solenodonsaurus
janenschi, and a reconsideration of amniote origins and stego-
cephalian evolution. Canadian Journal of Earth Sciences 36:1239–
1255.
Lucas, S. G. 2006. Global Permian tetrapod biostratigraphy and
biochronology; pp. 65–93 in S. G. Lucas, G. Cassinis, and J.
W. Schneider (eds.), Non-Marine Permian Biostratigraphy and
Biochronology. Geological Society, London, Special Publications
265.
Olson, E. C. 1991. An eryopid (Amphibia: Labyrinthodontia) from the
Fort Sill fissures, Lower Permian, Oklahoma. Journal of Vertebrate
Paleontology 11:130–132.
Romer, A. S. 1945. Vertebrate Paleontology. 2nd Edition. University of
Chicago Press, Chicago, Illinois, 687 pp.
Romer, A. S. 1950. The nature and relationships of the Paleozoic mi-
crosaurs. American Journal of Science 248:628–654.
Ruta, M., and M. I. Coates. 2007. Dates, nodes and character conflict: ad-
dressing the lissamphibian origin and problem. Journal of System-
atic Palaeontology 5:1–54.
Ruta, M., M. I. Coates, and D. L. J. Quicke. 2003a. Early tetrapod rela-
tionships revisited. Biological Reviews of the Cambridge Philosoph-
ical Society 78:251–345.
Ruta, M., J. E. Jeffery, and M. I. Coates. 2003b. A supertree of early
tetrapods. Proceedings of the Royal Society of London, Series B
270:2507–2516.
Schultze, H.-P., and B. Foreman. 1981. A new gymnarthrid microsaur
from the Lower Permian of Kansas with a review of the tudi-
tanomorph microsaurs (Amphibia). Occasional Papers of the Mu-
seum of Natural History, University of Kansas 91:1–25.
Sullivan, C., and R. R. Reisz. 1999. First record of Seymouria (Vertebrata:
Seymouriamorpha) from Early Permian fissure fills at Richards
Spur, Oklahoma. Canadian Journal of Earth Sciences 36:1257–
1266.
Sumida, S. S., D. S Berman, and T. Martens. 1996. Biostratigraphic cor-
relations between the Lower Permian of North America and cen-
tral Europe using the first record of an assemblage of terrestrial
tetrapods from Germany. PaleoBios 17:1–12.
Sumida, S. S., D. S Berman, and T. Martens. 1998. A new trematopid
amphibian from the Lower Permian of central Germany. Palaeon-
tology 41:1–25.
Sumida, S. S., D. S Berman, D. A. Eberth, and A. C. Henrici. 2004. A
terrestrial vertebrate assemblage from the Late Palaeozoic of cen-
tral Germany, and its bearing on Lower Permian palaeoenviron-
ments; pp. 113–123 in G. C. Young (ed.), Lower Vertebrates from
the Palaeozoic. Volume 50. Taylor and Francis, Oslo.
Wible, J. R., D. Miao, and J. A. Hopson. 1990. The septomaxilla of fos-
sil and recent synapsids and the problem of the septomaxilla of
monotremes and armadillos. Zoological Journal of the Linnean So-
ciety 98:203–228.
Williston, S. W. 1913. Ostodolepis brevispinatus, a new reptile from the
Permian of Texas. Journal of Geology 21:363–368.
Woodhead, J., R. Reisz, D. Fox, R. Drysdale, J. Hellstrom, R. Maas, H.
Cheng, and R. L. Edwards. 2010. Speleothem climate records from
deep time? Exploring the potential with an example from the Per-
mian. Geological Society of America 38:455–458.
Zittel, K. v. 1888. Handb ¨
uch der Palaeontologie. 1. Abteilung Palaeozo-
ologie. III. Band (Pisces, Amphibia, Reptilia, Aves). Oldenbourg,
M¨
unchen und Leipzig, 971 pp.
Submitted February 17, 2011; revisions received June 7, 2011; accepted
June 8, 2011.
Handling editor: Jason Anderson.
Downloaded by [Amy Henrici] at 13:47 09 September 2011
