ArticlePDF Available

Vahiny depereti, gen. et sp. nov., a New Titanosaur (dinosauria, Sauropoda) from the Upper Cretaceous Maevarano Formation, Madagascar

Authors:

Abstract and Figures

Here we describe Vahiny depereti, gen. et sp. nov., a new titanosaur dinosaur from the Upper Cretaceous Maevarano Formation of northwestern Madagascar. Vahiny is distinguished from other titanosaurs by characteristics of the basal tubera, basipterygoid processes, parasphenoid, and cranial nerve foramina. Diagnostic cranial material of Vahiny formalizes the long-standing hypothesis that two titanosaur genera were present in the Late Cretaceous of Madagascar. The other titanosaur, Rapetosaurus krausei, is the most common dinosaur in the fauna and is known from hundreds of bones, including multiple partial skeletons and skulls. In contrast, non-Rapetosaurus elements are extremely rare. Of these elements, which have been called ‘Malagasy Taxon B,’ we identify a partial braincase as the holotype of Vahiny depereti. Although Vahiny and Rapetosaurus coexisted on the island of Madagascar, differences in their braincases indicate that they are not closely related to one another. Instead, Vahiny is most similar to Jainosaurus from the Late Cretaceous of India, and shares similarities with the South American taxa Muyelensaurus and Pitekunsaurus.
Content may be subject to copyright.
Journal of Vertebrate Paleontology 34(3):606–617, May 2014
©2014 by the Society of Vertebrate Paleontology
ARTICLE
VAHINY DEPERETI, GEN. ET SP. NOV., A NEW TITANOSAUR (DINOSAURIA, SAUROPODA)
FROM THE UPPER CRETACEOUS MAEVARANO FORMATION, MADAGASCAR
KRISTINA CURRY ROGERS*,1 and JEFFREY A. WILSON2
1Biology and Geology Departments, Macalester College, 1600 Grand Avenue, St. Paul, Minnesota 55105, U.S.A.,
rogersk@macalester.edu;
2Museum of Paleontology and Department of Earth & Environmental Sciences, University of Michigan, 1109 Geddes Avenue,
Ann Arbor, Michigan 48109-1079, U.S.A., wilsonja@umich.edu
ABSTRACT—Here we describe Vahiny depereti, gen. et sp. nov., a new titanosaur dinosaur from the Upper Cretaceous Mae-
varano Formation of northwestern Madagascar. Vahiny is distinguished from other titanosaurs by characteristics of the basal
tubera, basipterygoid processes, parasphenoid, and cranial nerve foramina. Diagnostic cranial material of Vahiny formalizes
the long-standing hypothesis that two titanosaur genera were present in the Late Cretaceous of Madagascar. The other ti-
tanosaur, Rapetosaurus krausei, is the most common dinosaur in the fauna and is known from hundreds of bones, including
multiple partial skeletons and skulls. In contrast, non-Rapetosaurus elements are extremely rare. Of these elements, which
have been called ‘Malagasy Taxon B,’ we identify a partial braincase as the holotype of Vahiny depereti. Although Vahiny and
Rapetosaurus coexisted on the island of Madagascar, differences in their braincases indicate that they are not closely related
to one another. Instead, Vahiny is most similar to Jainosaurus from the Late Cretaceous of India, and shares similarities with
the South American taxa Muyelensaurus and Pitekunsaurus.
INTRODUCTION
Titanosaurs were first recognized in Upper Cretaceous strata
from the Mahajanga Basin of northwestern Madagascar by
Dep´
eret (1896a, 1896b), less than 20 years after the first descrip-
tions of titanosaurs in India and Patagonia (Lydekker, 1877, 1879,
1893). Since then, continued exploration of the Maastrichtian
Maevarano Formation of the Mahajanga Basin (1993–present)
has yielded associated and articulated titanosaur skeletons that
resulted in the designation of a new taxon, Rapetosaurus krau-
sei (Curry Rogers and Forster, 2001, 2004; Curry Rogers, 2009).
Rapetosaurus is among the most abundant vertebrates of the
Maevarano Formation. It is known from a wealth of well-
preserved cranial and postcranial remains representing a variety
of ontogenetic stages (e.g., Curry Rogers and Forster 2004; Curry
Rogers, 2009; Curry Rogers et al., 2011) and has proven pivotal
to ongoing revisions of titanosaur anatomy and phylogeny (e.g.,
Upchurch et al., 2004; Curry Rogers, 2005; Wilson, 2005a; Gallina
and Apestegu´
ıa, 2011), as well as understanding the distribution
and function of titanosaur osteoderms (D’Emic et al., 2009; Curry
Rogers et al., 2011).
Throughout the extensive history of paleontological collect-
ing in the Mahajanga Basin, hints of isolated, elongated, and
dorsoventrally compressed caudal centra have persistently sug-
gested that more than one titanosaur species may have populated
the Maevarano Formation ecosystem (Thevenin, 1907; Huene,
1929; Curry Rogers and Forster, 2001, 2004; Curry Rogers, 2002,
2005; Wilson and Upchurch, 2003). Although clearly distinct
from Rapetosaurus (e.g., Curry Rogers and Imker, 2007; McNulty
et al., 2010), the scattered occurrence and lack of association
among these unusual elements prohibited a unified view or defini-
tive diagnosis of ‘Malagasy Taxon B’ (Curry Rogers, 2002, 2005,
2009).
*Corresponding author.
Here we describe Vahiny depereti, gen. et sp. nov., a second
titanosaur species from the Maevarano Formation. A recently
recovered braincase serves as the holotype for this new taxon,
which ensures that the name-bearing material of the second
Malagasy taxon is stable, particularly because both Rapetosaurus
and Vahiny are known from well-preserved and autapomorphic
cranial material. At present, we choose not to assign the fossils
previously ascribed to ‘Malagasy Taxon B’ (e.g., flattened cau-
dal vertebrae) to Vahiny because field work in the Mahajanga
Basin is ongoing, and future discoveries may better resolve as-
sociations. Our conservative taxonomic assignment provides a
basis for establishing these future associations between Vahiny,
Rapetosaurus, and the as yet undesignated materials in the sauro-
pod collection from the Mahajanga Basin.
Institutional AbbreviationsFMNH PR, Field Museum of
Natural History, Chicago, Illinois, U.S.A.; ISI, Indian Statisti-
cal Institute, Kolkata, India; MACN, Museo Argentino Cien-
cias Naturales, Buenos Aires, Argentina; MAU, Museo Munic-
ipal ‘Argentina Urquiza,’ Rinc ´
on de los Sauces, Argentina; PIN,
Russian Academy of Sciences, Moscow, Russia; PVL, Fundacion
Miguel Lillo, Universidad Nacional de Tucuman, San Miguel de
Tucuman, Argentina; SAM, Iziko South African Museum, Cape
Town, South Africa; UA, Universit´
e d’Antananarivo, Antana-
narivo, Madagascar; UCB, Universit´
e Claude Bernard, Lyon,
France; ZPAL, Instytut of Paleobiologii, Polish Academy of Sci-
ences, Warsaw, Poland.
SYSTEMATIC PALEONTOLOGY
SAUROPODA Marsh, 1878
NEOSAUROPODA Bonaparte, 1986
TITANOSAURIA Bonaparte and Coria, 1993
VAHINY DEPERETI, gen. et sp. nov.
(Figs. 1–4, 6)
606
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 607
EtymologyVahiny (Malagasy, pronounced ‘va-heenh’),
meaning traveler or visitor, reflecting the rarity of this taxon
in the Mahajanga Basin. The specific name honors Charles
Dep´
eret, who described the original dinosaur material from
Madagascar and presciently recognized the sauropod nature of
the Malagasy osteoderms.
Holotype—UA 9940, a partial braincase including a par-
tial supraoccipital, basioccipital, basisphenoid, right and left
exoccipital-opisthotics, prootics, laterosphenoids, and or-
bitosphenoids.
Locality and Horizon—MAD 96-07, which is located at coor-
dinates 155456.4S, 463543.1E, Berivotra, Mahajanga Basin,
northwestern Madagascar. The sediments correspond to Facies
2 of the Maastrichtian-aged Anembalemba Member of the Mae-
varano Formation (Rogers et al., 2000; Rogers, 2005). The holo-
typic braincase was found in association with an assortment of
Maevarano Formation vertebrates in a time-averaged assem-
blage, like those preserved at other Maevarano Formation fossil
localities (e.g., Rogers et al., 2000, 2007; Rogers, 2005). Fragmen-
tary titanosaur, crocodile, theropod, and turtle limb bones, as well
as turtle carapace and plastron, crocodile osteoderms, and thero-
pod teeth were also recovered from surface collection at the site.
DiagnosisVahiny depereti can be distinguished as a derived
member of Titanosauria on the basis of a contact between the
quadrate and basal tubera (Wilson, 2002, 2005a). Vahiny shares
more specific characteristics with the Indian titanosaur Jain-
osaurus and with South American titanosaurs Muyelensaurus
(Calvo et al., 2007) and Pitekunsaurus (Filippi and Garrido,
2008). These include the presence of a broad, shallow fossa be-
tween the basal tubera, basipterygoid processes that are oriented
parallel to the occipital plane, and a long prootic spur that ex-
tends onto the basipterygoid process. Vahiny and Muyelensaurus
(Calvo et al., 2007) also share basipterygyoid processes that di-
verge via a broad, ‘U’-shaped embayment. Vahiny and Jain-
osaurus share basal tubera that exhibit a small ventrolateral pro-
cess set off by a notch that is continued by an elongate groove
(Wilson et al., 2009). Vahiny depereti is characterized by the fol-
lowing autapomorphies: basal tubera that are composed mainly
of the basioccipital, with a reduced contribution from the ba-
sisphenoid; a thick web of bone separating the basipterygoid em-
bayment from the basal tubera, visible posteriorly; paired blind
fossae present between the basipterygoid processes and paras-
phenoid rostrum; ventrally keeled and dorsally troughed paras-
phenoid rostrum; and foramina for cranial nerves V, III, and II
are co-linear, and the metotic foramen is displaced ventrally from
them.
Referred Material—FMNH PR 3046, an isolated juvenile ba-
sioccipital (Fig. 5) recovered from Facies 1 of the Anembalemba
Member of the Maevarano Formation at locality MAD 93-18,
Berivotra, Madagascar.
DESCRIPTION
Below we describe the holotypic Vahiny braincase as well as
a referred juvenile basioccipital. The descriptions and compar-
isons that follow are based upon study of collections of Mala-
gasy dinosaurs held at the Universit´
e d’Antananarivo, Stony
Brook University, and the Field Museum of Natural History. Un-
less otherwise stated, descriptions are based primarily upon the
holotypic braincase. Additional insights were gleaned from ti-
tanosaurs housed in collections in India, Argentina, Brazil, and
France, and descriptions available in the published literature. We
employ traditional orientational descriptors (e.g., anterior, pos-
terior) rather than the standardized terms (i.e., cranial, caudal)
applied to mammals and birds (see Wilson, 2006). We refer to
cranial nerve foramina by their number (e.g., foramen for cra-
nial nerve V) rather than their common name (i.e., trigeminal
foramen).
We describe the median and paired paramedian bones that
form the floor, posterior wall, and lateral walls of the braincase,
drawing attention to connections between bones as well as the
foramina passing through or between them. We designate the
surface of the occiput as a vertical reference plane that can be
used to measure angular orientation of projecting structures (e.g.,
basipterygoid processes) for comparative purposes (see Wilson
et al., 2009:25, 34). Principal dimensions of the holotypic brain-
case and referred basioccipital are provided in Table 1.
Preservation
The holotypic braincase lacks the distal ends of the paroc-
cipital processes, basipterygoid processes, and parasphenoid ros-
trum. Most of the supraoccipital and the dorsal margins of the
laterosphenoid and orbitosphenoid have been broken away. The
frontals and parietals were not preserved, which permits a view
into the endocranial cavity and hypophyseal fossa. Openings for
cranial nerves I–XII are preserved on one or both sides of the
skull (Figs. 1–4).
Supraoccipital
The supraoccipital forms the posterodorsal part of the brain-
case, contacting the parietal anterodorsally and the exoccipital-
opisthotic ventrolaterally. The supraoccipital is poorly preserved
in the holotypic braincase; thus, many critical features cannot be
observed (e.g., nuchal crest).
Sutures between the exoccipital-opisthotics and supraoccipi-
tal are completely fused, but low, transversely oriented ridges
at the dorsal margin of the foramen magnum probably mark the
medial-most part of their course. If these indeed represent part of
the suture between the supraoccipital and exoccipital-opisthotic,
the supraoccipital probably had at least a small contribution to
the foramen magnum. The central portion of the nuchal crest
is not preserved in Vahiny, but low, paramedian prominences
mark its ventrolateral margins. These ridges are more laterally
positioned and narrower than the paramedian, more vertically
oriented ridges observed in Rapetosaurus. Due to incomplete
preservation of the supraoccipital, neither its height relative to
the foramen magnum nor the form of its nuchal crest are known.
Basioccipital
The basioccipital is the posterior-most ventral median brain-
case element. It is completely preserved and forms most of the
occipital condyle and basal tubera.
The occipital condyle projects ca. 90from the occipital plane
and is subparallel with the estimated orientation of the skull
roof (Figs. 1–3). As in Jainosaurus (Wilson et al., 2009; Fig.
6A, E), the articular surface of the occipital condyle extends
ventrally, suggesting either the skull was deflected slightly ven-
trally relative to the vertebral axis or that the atlantal articu-
lar surface extended ca. 1.5 cm farther anteriorly than did the
odontoid. The absence of sutures makes it difficult to discern
the relative contribution of the exoccipital-opisthotic to the oc-
cipital condyle in the holotypic braincase, but it clearly con-
tributed at least the dorsolateral shoulders of the condyle, be-
cause the exit for cranial nerve XII is partly preserved on its
neck. In Rapetosaurus, for example, this opening is completely
enclosed by the base of the exoccipital-opisthotic (Curry Rogers
and Forster, 2004:fig. 19). The referred juvenile basioccipital,
which is not fused to the exoccipital-opisthotic, exhibits strongly
beveled sutural facets that nearly meet on the midline, leaving
only a median sliver of basioccipital at the floor of the foramen
magnum (Fig. 5A, D). This indicates a substantial contribution of
the exoccipital-opisthotic to the occipital condyle at least early in
ontogeny. Vahiny lacks the small, ventrally facing depression be-
tween the occipital condyle and basal tubera that is observed in
Downloaded by [141.213.177.149] at 06:07 07 May 2014
608 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 34, NO. 3, 2014
FIGURE 1. Vahiny depereti, holotypic braincase (UA 9940). Stereopairs in A, right lateral view; B, left lateral view. Scale bar equals 3 cm.
Jainosaurus,Rapetosaurus, and some other neosauropods (Wil-
son et al., 2009). The metotic foramen (for cranial nerves IX–XI
and the jugular vein) and foramen ovale (cranial nerve VIII) oc-
cur at the intersection between the basioccipital, basisphenoid,
exoccipital-opisthotic, and prootic.
The basal tubera are anteroposteriorly thin, sheet-like struc-
tures, as they are in Saltasaurus (Wilson, 2002) and certain other
titanosaurs. The basal tubera appear to be primarily composed
of the basioccipital, with a smaller, possibly negligible contribu-
tion from the basisphenoid. This is especially apparent in the
juvenile, unfused basioccipital, in which the basal tubera bear
smooth distal surfaces that lack articular facets for the basisphe-
noid (Fig. 5E). In posterior view, the basal tubera form a trape-
zoid whose upper base is ca. 3 cm and whose lower base is 5 cm.
The symmetrical legs of the trapezoid diverge ventrally to be-
come much broader transversely than is the occipital condyle.
The basal tubera are concave dorsoventrally and transversely,
as in Jainosaurus and the South American titanosaurs Muyelen-
saurus,Pitekunsaurus, and MML 194 (see Wilson et al., 2009;
Figs. 2D, 3B; Table 1). In lateral view, the basal tubera extend
posteriorly as far as the occipital condyle (Figs. 1–3). The basal
tubera are ventrolaterally thickened and roughened, which may
indicate a bony connection between the basal tubera and the me-
dial aspect of the quadrate. A small, lateral, pendant process is
sharply divided from the rest of the basal tubera by a deep notch
(Figs. 2C, 3B, 5B, D). Just ventral to this notch is a conspicu-
ous groove that extends to the preserved end of the basipterygoid
processes. The ventrolateral notch and processes are incipiently
developed in the referred juvenile basioccipital (Fig. 5D), and all
three features are well developed in Jainosaurus (Wilson et al.,
2009), which indicates their close common ancestry.
The basioccipital-basisphenoid suture is fused in the holotypic
specimen, but a well-defined ridge may approximate its posi-
tion (Figs. 1, 2D, 3). The sutural contact between these two
TABLE 1. Principal dimensions (in mm) of the braincase of Vahiny de-
pereti.
Referred
Holotype (FMNH
Region Dimension (UA 9940) PR 3046)
Lateral skull Length 99.4
Occipital condyle Width 29.4 16.6
Height 24.014.3
Length 30.0—
Foramen magnum Width 18.3
Height 24.9
Basal tubera Width 62.411.7
Basipterygoid processes Length, distal to
basal tubera
25.0
Hypophyseal fossa Width 16.9—
Length 7.9—
Asterisks () indicate measurement of an incomplete structure.
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 609
FIGURE 2. Vahiny depereti, holotypic braincase (UA 9940). Stereopairs in A, dorsal view; B, ventral view; C, anterior view; and D, posterior view.
Scale bar equals 3 cm.
Downloaded by [141.213.177.149] at 06:07 07 May 2014
610 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 34, NO. 3, 2014
FIGURE 3. Vahiny depereti, holotypic braincase (UA 9940). Stipple drawings in A, anterior view; B, posterior view; C, left lateral view; and D,
right lateral view. Hatching indicates broken bone; dashed lines indicate probable contour of missing bone. Abbreviations:ahc, adenohypophyseal
canal; bpt, basipterygoid process of basisphenoid; bt, basal tubera; btf, basal tubera fossa; cr pr, crista prootica; ds, dorsum sellae; f.ls, facet for the
laterosphenoid; fm, foramen magnum; ica, foramen for the internal carotid artery; mf, metotic foramen; no, notch; oc, occipital condyle; paf, proatlantal
facet; ps, parasphenoid; IIVII, cranial nerve openings. Scale bar equals 3 cm.
elements is clearly demarcated on the referred juvenile basioc-
cipital (Fig. 5C, E, F). A surface for an abutting articulation is
present on the anterior surface of the proximal half of the re-
ferred juvenile basisphenoid. This surface grades distally into a
striated surface. The sutural surface for the basisphenoid does
not extend to the distal tips of the basal tubera, which are beveled
and smooth.
Basisphenoid
The basisphenoid contacts the basioccipital posteriorly and
forms the majority of the endocranial floor from the metotic
foramen forward. This element is sometimes referred to as
the parabasisphenoid in sauropods (e.g., Witmer et al., 2008;
Balanoff et al., 2010), because it probably incorporates the
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 611
FIGURE 4. Vahiny depereti, holotypic braincase (UA 9940). Three transverse CT sections passing through the braincase at the approximate positions
shown in the surface-rendered braincase at left (ac). The section shown in Ais at level of internal carotid foramen; section in Bis at level of
adenohypophyseal canal; section in Cis at level of parasphenoid rostrum. Abbreviations:ahc, adenohypophyseal canal; bpt, basipterygoid process of
basisphenoid; bt, basal tubera; btf, basal tubera fossa; CN IV, foramen for cranial nerve IV; ds dorsum sellae; ec, endocranial cavity; hpc, hypophyseal
canal; ica, foramen for the internal carotid artery; ls, laterosphenoid; lsp, laterosphenoid pillar; par, parasphenoid rostrum; pop, paroccipital process;
pro, prootic. Scale bar equals 3 cm.
dermal parasphenoid bone as it does in squamates and other
reptiles (e.g., Zaher et al., 2009). Most sauropod braincase
descriptions, however, use the term basisphenoid (e.g., Paulina
Carabajal, 2012), and we retain that term here without implying
a separate parasphenoid bone.
A conspicuous ridge, the dorsum sellae, crosses the floor of
the endocranial cavity just posterior to the hypophyseal fossa,
near the junction of the basisphenoid, prootic, and laterosphe-
noid (Figs. 2A, 3A). The dorsum sellae is continuous with the
laterosphenoid ‘pillars’ (see below), as it is in some titanosaurs
(Rapetosaurus, Curry Rogers and Forster, 2004; unnamed forms,
Paulina Carabajal and Coria, 2007; Garcia et al., 2008). A hint of a
suture on the posteroventral margin of the dorsum sellae may in-
dicate the basisphenoid-laterosphenoid contact. Paired openings
for cranial nerve VI pierce the braincase floor just posterior to the
dorsum sellae and exit ca. 1.5 cm ventral to the line that connects
the openings for cranial nerves II, III, and V (Figs. 1, 3C, D). Just
posterior to these openings on the floor of the braincase is a me-
dian opening that may represent a vascular canal. Just anterior to
the dorsum sellae is the hypophyseal fossa (sella turcica), which
is formed by the basisphenoid, laterosphenoids, and orbitosphe-
noids. The hypophyseal canal expands into a deep fossa that is
visible through the broken anterolateral surface of the basisphe-
noid.
The basipterygoid processes are incomplete distally, but they
would have been fairly elongate. The distance from the base of
the occipital condyle to the ends of the basal tubera is slightly
less than the distance from the ends of the basal tubera to the
ends of the preserved basipterygoid processes (Table 1). The left
and right basipterygoid processes diverge from the mid-sagittal
plane at an angle of approximately 50, and together they are an-
gled 147anteriorly relative to the occipital plane (Figs. 1, 2C, D,
3C–D). A broad, ‘U’-shaped embayment separates the basiptery-
goid processes, as it does in other titanosaurs (e.g., Muyelen-
saurus; Calvo et al., 2007). The embayment extends to within ca.
2 cm of the basal tubera, separated from them by a ribbon of bone
that is visible posteriorly (Figs. 2B, 3B). This latter feature may
be unique to Vahiny.
FIGURE 5. Vahiny depereti, referred juve-
nile basioccipital (FMNH PR 3046) in A, dor-
sal view; B, ventral view; C, right lateral view;
D, posterior view; E, anterior view; F,left
lateral view. Abbreviations:bt, basal tubera;
f.bs, facet for basisphenoid; f.eo-op, facet for
exoccipital-opisthotic; no, notch at distal end
of basal tubera; oc, occipital condyle. Scale
bar equals 1 cm.
Downloaded by [141.213.177.149] at 06:07 07 May 2014
612 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 34, NO. 3, 2014
The long spur of the crista prootica continues onto the ba-
sisphenoid as a conspicuous ridge that is visible on the antero-
lateral edge of the basipterygoid processes. At the level of the
occipital condyle, the crista prootica separates into troughs ex-
tending ventrally from the metotic foramen (posteriorly) and
the foramen for maxillary and mandibular branches of cranial
nerve V (anteriorly). Further ventrally, the continuation of this
ridge separates two openings. The more posteriorly positioned
of these openings is located near the basioccipital-basisphenoid
contact and represents the opening for the internal carotid artery
(Fig. 4A). The more anteriorly positioned of these openings
passes through the basisphenoid and enters the hypophyseal
fossa (Fig. 4B). It may represent a canal into the adenohypoph-
ysis, as suggested by Paulina Carabajal (2012) for that structure
in Bonatitan. Computed tomography (CT) scans show that the
opening for the internal carotid artery probably met the adeno-
hypophyseal canal in the hypophyseal fossa (Fig. 4B). The posi-
tion of the internal carotid arteries on the lateral aspect of the
basipterygoid processes is common in non-titanosaur sauropods
but apparently rare within titanosaurs, in which they open me-
dially, between the basipterygoid processes (Paulina Carabajal,
2012). There are paramedian depressions located between the
basipterygoid processes in Vahiny, but no foramina can be ob-
served, and CT scans reveal that the depressions did extend into
the hypophyseal fossa (Fig. 4). Thus, Vahiny and Jainosaurus
share a primitive feature of laterally positioned openings for the
internal carotid arteries.
The basipterygoid processes do not appear to taper anteropos-
teriorly or transversely. In cross-section, the basipterygoid pro-
cesses are anteroposteriorly elongated and elliptical, with a slight
concavity on the medial surface. Anteriorly, each crista prootica
meets a ridge extending ventrolaterally from the median paras-
phenoid. In anteroventral view, between the paired ventrolat-
eral parasphenoid ridges, are small fossae that contain foramina
that may have transmitted vasculature to the adenohypophysis
(Paulina Carabajal, 2012), as mentioned above. The presence of
these openings in Jainsosaurus cannot be determined in either of
the exemplars, because in both cases that portion of the braincase
was not preserved.
The parasphenoid, or cultriform process, lacks only its dis-
tal end. It is fairly robust, which contrasts with the transversely
thin, sheet-like processes found in some other titanosaurs (e.g.,
Rapetosaurus; Curry Rogers and Forster, 2004). The parasphe-
noid expands anteriorly to form a robust process that terminates
in an anteroventrally directed, distally broadening ‘U’-shaped
trough (Figs. 1, 2). The ventral surface of the parasphenoid bears
a sharp midline ridge with laterally beveled edges. Just postero-
lateral to this ridge, between the basipterygoid processes, are two
blind depressions that are located where the internal carotid ar-
teries open in some titanosaurs (e.g., Saltasaurus).
Exoccipital-opisthotic
The paired exoccipital-opisthotic elements form part of the
posterior wall of the endocranial cavity, as well as the paroc-
cipital processes, the lateral margins of the foramen magnum,
and the dorsolateral portions of the occipital condyle. The
exoccipital-opisthotics are partially preserved, lacking all but
the base of the paroccipital processes and missing the ventro-
lateral border of the foramen magnum (Fig. 2A, B, D). The
exoccipital-opisthotic is fused to both the supraoccipital and
the basioccipital, but part of its suture with the prootic can be
discerned on the left side, in the cross-section through the paroc-
cipital process. Ventrally, between these two bones open the
foramina for cranial nerves VII–XI. The paroccipital processes
do not exhibit the sharp notch in their narrow dorsal border that
characterizes Jainosaurus and Rapetosaurus (Wilson et al., 2009).
Instead, they are anteroposteriorly broad and, where fused
with the prootic, slightly sinuous along the squamosal articular
surface.
The exoccipital-opisthotic articulates with the basioccipital via
a nearly horizontal suture that begins at the metotic foramen and
extends posteriorly onto the occipital condyle. The exoccipital-
opisthotics extend medially nearly to the midline, but they do
not quite exclude the basioccipital and supraoccipital from the
ventral-most and dorsal-most portions of the foramen magnum,
respectively. Just lateral to the dorsal-most point of the foramen
magnum are two transversely oriented low ridges that mark the
part of the contact between the supraoccipital and exoccipital-
opisthotic (Fig. 3B). These facets also serve as the point of ar-
ticulation of the proatlas, which were not preserved with this
specimen.
Prootic
The prootic forms the posterior-most lateral wall of the brain-
case. In the Vahiny holotype, the prootic is completely fused to
the basisphenoid and nearly completely fused to the exoccipital-
opisthotic, with a trace of the suture visible in the paroccipital
process. In a subadult specimen of Rapetosaurus, a gap in the in-
tersection of these elements forms the metotic foramen (Curry
Rogers and Forster, 2004:figs. 23, 24).
The crista prootica is an elongated vertical crest that extends
from a position dorsal to the metotic foramen all the way to the
anterior border of the basipterygoid processes. The crista proot-
ica is posteriorly concave and visible in anterior, posterior, and
lateral views (Figs. 1, 2C, D, 3). At the level of the occipital
condyle, it separates the metotic foramen from the foramen for
cranial nerve V. Further ventrally, it separates the foramen for
the internal carotid artery and the canal into the adenohypoph-
ysis.
No semicircular canals are visible in CT images of the holotype,
but the broken right prootic preserves a depression that may be
the opening into the inner ear.
Laterosphenoid
The laterosphenoid is a transversely oriented element that
abuts the prootic posteriorly and is nearly fused to the basisphe-
noid ventrally and to the orbitosphenoid anteriorly. The pillar-
like laterosphenoids extend medially to contact one another on
the midline, as indicated by a trace of a suture on the poste-
rior aspect of the dorsum sellae. The dorsal portion of the lat-
erosphenoid (e.g., laterosphenoid head) is not preserved in the
holotype. As a consequence, its sutural articulations with the pos-
torbital, frontal, and parietal are unknown. The suture between
the laterosphenoid and orbitosphenoid is not preserved, but it
was likely positioned near the openings for cranial nerves III and
IV, as it is in most other sauropods.
The laterosphenoid forms the crista antotica, which is a ridge
that forms the anterior margin of the foramen for cranial nerve V
and part of the groove for its maxillary and mandibular branches.
The preserved portion of the crista antotica is much lower in re-
lief than the laterosphenoid pillar. This pillar corresponds to the
division between the forebrain and midbrain (e.g., Wilson et al.,
2009).
Orbitosphenoid
The orbitosphenoids enclose the forebrain anteriorly and form
the anterior portion of the hypophyseal fossa. The orbitosphe-
noid, laterosphenoid, and basisphenoid are completely fused in
the holotype (UA 9940). Internally, the optic nerve (cranial nerve
II) foramina open within a fossa in the body of the orbitosphe-
noids. They are directed anteroventrally and divided on the mid-
line by a thin bony isthmus (Figs. 1, 2A, 3C, D). The anterodorsal-
most portion of the orbitosphenoid is most completely preserved
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 613
on the left side, where it is anteroposteriorly and transversely ex-
panded. The slightly dorsally concave portion of the orbitosphe-
noid would have formed the lower margin of the foramen for cra-
nial nerve I.
Cranial Nerves
Openings for the anterior cranial nerves are well preserved on
both sides of the braincase, and the posterior cranial nerve open-
ings (VII–XII) are partially preserved on at least one side of the
holotypic braincase (UA 9940). Openings for cranial nerves II,
III, V, and XII and the metotic foramen form a line that extends
to the foramen magnum (Figs. 1, 3C, D). Openings for cranial
nerves I and IV are positioned dorsal to this line, and the open-
ing for cranial nerve VI is positioned ventral to it, as in other
sauropods (Wilson et al., 2009).
The dorsal margin for cranial nerve I, which is normally formed
by the skull roof, is not preserved. The ventral border is formed
by the conjoined orbitosphenoids. The foramina for cranial nerve
II are completely preserved in the holotypic braincase, where
they occur within an elevated shelf. It appears that this foramen
is mainly composed of the orbitosphenoid, but it may receive a
small midline contribution from the parasphenoid. The opening
for cranial nerve III is small and occurs anterior to the crista an-
totica (Fig. 3), dorsal to the hypophyseal fossa. Its borders are
formed by the orbitosphenoid anteriorly, the laterosphenoid pos-
teriorly, and the basisphenoid ventrally. The opening for cranial
nerve IV also likely occurs along the suture between the lat-
erosphenoid and the orbitosphenoid in a line dorsal to the open-
ing for cranial nerve III (Fig. 3). In Vahiny, the opening for cra-
nial nerve IV is broken dorsally, but it is larger than the foramen
for cranial nerve III. The foramen for cranial nerve V opens be-
tween the laterosphenoid, prootic, and basisphenoid. It is large
and probably accommodated all three branches of cranial nerve
V. The path of the mandibular branch of the trigeminal nerve
is marked by a shallow groove that extends along the anterior
surface of the crista prootica. The opening for cranial nerve VI
is ventral to these other foramina, on the anterolateral surface
of the basisphenoid (Figs. 1–3). It exits slightly below the open-
ing for cranial nerve III. The opening for cranial nerve VII is
a tiny, rounded foramen positioned on the posterior surface of
the crista prootica. The metotic foramen, which transmitted cra-
nial nerves IX–XI and the jugular vein, and the foramen ovale
(cranial nerve VIII) occur in a large fossa posterior to the crista
prootica. Although the ventral margins of the fossa are preserved,
the other borders of these foramina are not visible. Ventrally di-
rected grooves on the posterior side of the crista prootica and
on the basisphenoid indicate the paths of these nerves. A single
opening for cranial nerve XII is located on the dorsal surface of
the neck of the occipital condyle.
COMPARISONS AND PHYLOGENETIC AFFINITES
Titanosaur phylogeny remains poorly resolved, despite nu-
merous recent cladistic analyses that have focused on the group
(e.g., Gonz´
alez Riga, 2003; Curry Rogers, 2005; Calvo et al., 2007;
Zaher et al., 2011). Lack of phylogenetic resolution may be due
to a combination of interrelated factors that include (1) the
nature of the titanosaur body fossil record, which thus far has
produced relatively few skulls or articulated skeletons (e.g., Man-
nion and Upchurch, 2010); (2) undocumented field associations
between elements in a type series, which characterizes many
early descriptions of titanosaur genera, including ‘Titanosaurus
madagascariensis’; (3) the global nature of their fossil record
and resultant wide geographic range of collections, which make
it difficult to make first-hand observations on many different
taxa; (4) genera based on fragmentary remains; and (5) generic
diagnoses based upon features that have obsolesced to describe
a broader distribution of titanosaurs. As such, the precise rela-
tionships of Vahiny and other titanosaurs must await ongoing
taxonomic revision of several titanosaur genera and a more com-
prehensive analysis of titanosaur phylogeny. Here we provide
general comparisons to titanosaur taxa that have basicranial
material from Cretaceous rocks in Madagascar (Rapetosaurus),
India (Jainosaurus,Isisaurus), South America (Antarctosaurus,
Saltasaurus,Muyelensaurus,Pitekunsaurus, Bonatitan), Asia
(Nemegtosaurus), and Africa (Malawisaurus) (Fig. 6).
Rapetosaurus
Two Rapetosaurus skulls, as well as abundant isolated cranial
material pertaining to juvenile and adult individuals, have been
recovered from the Maevarano Formation. The holotypic skull of
Rapetosaurus krausei (UA 8698) included a partial basioccipital
and exoccipital-opisthotic preserving the occipital condyle, right
basal tuber, and paroccipital process. The referred juvenile skull
(FMNH PR 2192), which was found in association with a postcra-
nial skeleton, is better preserved and includes a fused basioc-
cipital and basisphenoid that articulate with a right exoccipital-
opisthotic, supraoccipital, and a left laterosphenoid.
Features of the basicranium clearly distinguish Rapetosaurus
and Vahiny, particularly the morphology of the basal tubera,
basipterygoid processes, and parasphenoid rostrum (Fig. 6A,
B). In Vahiny, the occipital condyle is a little wider than it is
tall, whereas in Rapetosaurus the condyle is slightly taller than
it is wide. The foramen magnum in Vahiny is taller than it is
wide, whereas in Rapetosaurus these dimensions are equal. The
supraoccipital is incompletely preserved in Vahiny, but the pre-
served region appears relatively flat and suggests the presence
of a single nuchal crest, which contrasts with the median groove
present in the nuchal crest of Rapetosaurus.
The basal tubera of Vahiny are composed mainly of the ba-
sioccipital, with a negligible contribution from the basisphenoid.
They are separated by a broad posterior fossa that is not punctu-
ated by vasculature. The basioccipital between the basal tubera
is nearly sheet-like, with a straight ventral border that grades
smoothly into the anteriorly positioned basisphenoid. This mor-
phology contrasts sharply with that of Rapetosaurus,inwhichthe
basal tubera are transversely narrow and have equal contribu-
tions from the basioccipital and basisphenoid. A deep ‘V’-shaped
notch separates the basal tubera in Rapetosaurus,andthereare
vascular pits ventral to the occipital condyle and between the
basal tubera. Rather than grading gently into the basipterygoid
processes, in Rapetosaurus the basal tubera are sharply offset
from them (FMNH PR 2192). In Vahiny, the basipterygoid pro-
cesses are deflected anteriorly relative to the occipital plane and
separated by a broad, ‘U’-shaped embayment that is wider than
the occipital condyle. Although the distal ends of the basiptery-
goid processes are not preserved in Vahiny, the basisphenoid ex-
posure on the posterior surface of the braincase is far smaller
than in Rapetosaurus. Similarly, the basipterygoid processes in
Rapetosaurus are angled anteriorly relative to the occiput (Curry
Rogers and Forster, 2004). The position of the internal carotid
arteries also differs in Rapetosaurus and Vahiny.InVahiny,as
in most non-titanosaur sauropods, the foramina for the inter-
nal carotids are lateral to the basipterygoid processes, but in
Rapetosaurus they are more anteriorly and medially positioned
(FMNH PR 2192).
In Vahiny, the parasphenoid rostrum is a triangular trough with
strongly beveled lateral margins that exhibits ridges and valleys
containing vascular foramina. In Rapetosaurus, the parasphenoid
is sheet-like and lacks foramina or sculpturing.
The foramina for cranial nerves II–V occur at slightly differ-
ent positions in Vahiny and Rapetosaurus (FMNH PR 2192). In
Vahiny, the foramen for cranial nerve IV is positioned slightly
Downloaded by [141.213.177.149] at 06:07 07 May 2014
614 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 34, NO. 3, 2014
FIGURE 6. Titanosaur basicranial comparisons. A,Vahiny depereti (UA 9940); B,Rapetosaurus krausei (FMNH PR 2192); C,Saltasaurus loricatus
(PVL 4017); D,Antarctosaurus wichmannianus (MACN 6904); E,Jainosaurus septentrionalis (ISI R 199); F,Quaesitosaurus orientalis (PIN 3906/2);
G,Pitekunsaurus macayai (MAU-Pv-AG-446). The articulated quadrates and pterygoids were digitally removed from the posterior view of Quaesi-
tosaurus for ease of comparison with other braincases (compare with Wilson, 2005b:fig. 18). Scale bar equals 5 cm.
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 615
dorsal to a line drawn connecting the ventral margins of the
foramina for cranial nerves II and V, whereas the foramen for
cranial nerve III is ventral to this line. In Rapetosaurus,the
foramina for cranial nerves II and V are not in horizontal align-
ment. Instead, the foramen for cranial nerve V is dorsal to cranial
nerve II, and is more aligned with the foramen for cranial nerve
IV. Likewise, cranial nerves II and III are in horizontal alignment
in Rapetosaurus (FMNH PR 2192).
Jainosaurus
The hypodigm of Jainosaurus septentrionalis includes two
braincases (K27/429, ISI R162) and assorted postcranial elements
(Huene and Matley, 1933; Chatterjee and Rudra, 1996; Wilson
et al., 2009, 2011). There are a number of similarities between
the braincases of Vahiny and Jainosaurus, which may indicate
that their sister-taxon relationship, although this hypothesis re-
mains to be tested in a phylogenetic analysis (Fig. 6A, E). Both
Jainosaurus (ISI R162) and Vahiny possess a small, laterally ori-
ented, pendant process and notch on the basal tubera. In both
taxa, this notch extends as a straight groove on the posterolateral
aspect of the basipterygoid process. Jainosaurus,Vahiny,andcer-
tain South American taxa (see below) also share the presence of a
broad, shallow fossa between the basal tubera. The basal tubera
of Vahiny are composed primarily of the basioccipital, whereas
in Jainosaurus they receive a greater contribution from the ba-
sisphenoid (ISI R162). The basipterygoid processes in Vahiny
are angled 147anteriorly relative to the plane formed by the
preserved portion of the braincase dorsal to the foramen mag-
num, in contrast with the basipterygoid processes of Jainosaurus,
which are parallel to the occipital plane (Wilson et al., 2009). The
basipterygoid processes of Vahiny diverge from the midline via a
broad, ‘U’-shaped embayment rather than the acute angle seen
in Jainosaurus (ISI R162; Wilson et al., 2009).
Other Titanosaurs
Isisaurus—The Late Cretaceous Indian titanosaur Isisaurus
(ISI R199; Berman and Jain, 1982) is distinguished from Jain-
osaurus and Vahiny by the strong downward deflection of
the occipital condyle. Other distinguishing features include the
small, circular proatlantal facets and broad supraoccipital wedge
present in Isisaurus.BothVahiny and Jainosaurus bear weaker
proatlantal facets and less prominent supraoccipital wedge.
Muyelensaurus—Calvo et al. (2007:fig. 4) described a brain-
case of Muyelensaurus that, like Vahiny, has basal tubera that
are joined by a thin, concave lamina that terminates laterally in
a sharp ridge. The occipital condyle is wider than the dorsal base
of the basal tubera in Muyelensaurus, but in Vahiny it is slightly
wider than the occipital condyle, as it is in many sauropods. Un-
like Muyelensaurus,Vahiny does not exhibit a wide, ventral de-
pression below the basal tubera. The basipterygoid processes are
not preserved in Muyelensaurus, but Calvo et al. (2007:489) sug-
gest on the basis of the preserved right basipterygoid process that
they are deeply separated and parallel, not laterally divergent as
in Vahiny.
PitekunsaurusVahiny and Pitekunsaurus (Fillipi and Gar-
rido, 2008) share the general features of broadly divergent basal
tubera and the ‘U’-shaped separation between the elongated
basipterygoid processes. However, they differ in several respects.
In Pitekunsaurus, the broad fossa dorsal to the basal tubera is
proportionately shorter than that in Vahiny (Fig.6A,G).The
basipterygoid processes are directed nearly vertically in Pitekun-
saurus,asinJainosaurus, whereas they are more anteriorly di-
rected in Vahiny. The parasphenoid is slightly more elevated rel-
ative to the basal tubera in Pitekunsaurus than in Vahiny.
Bonatitan—Like Isisaurus,Bonatitan (Martinelli and
Forasiepi, 2004) has a ventrally deflected occipital condyle
relative to the occipital plane. This differs from the more
horizontally oriented condyle in Vahiny. The basal tubera also
differ between these taxa. The basal tubera are fairly heavy and
have a stepped margin relative to the basipterygoid processes
in Bonatitan,asinRapetosaurus.Bonatitan and Jainosaurus
share the broadly separated, nearly vertically oriented basiptery-
goid processes. The endocast of Bonatitan has recently been
redescribed by Paulina Carabajal (2012), who noted the medial
positioning of the foramina for the internal carotid arteries. In
addition, Bonatitan has an anteriorly deflected occiput in which
the occipital fossa on the parietals is angled forward and dorsally.
Antarctosaurus—Like Bonatitan,Antarctosaurus (MACN
6904; Huene, 1929:pl. 28; Powell, 2003) has a well-marked occipi-
tal fossa on the parietal that is expressed as a bulge on the dorsal
surface of the skull. The occipital condyle is deflected ventrally, as
is the parasphenoid rostrum. The basal tubera are thin and arched
posteriorly in Antarctosaurus, a feature shared with Vahiny (Fig.
6A, D). The internal carotid arteries are located near the midline
in Antarctosaurus (Paulina Carabajal, 2012) rather than on the
lateral aspect of the basipterygoid process, as they are in Vahiny.
SaltasaurusSaltasaurus and Vahiny share the broad, concave
surface between basal tubera ventral to the occipital condyle, but
they differ in the development of the tubera themselves (Fig. 6A,
C). In Vahiny, the basal tubera are topographically distinctive,
whereas in Saltasaurus they are relatively flat and flush with the
laminar sheet between them (PVL 4017; Powell, 2003). They also
differ in the position of the internal carotid arteries, which open
between the basipterygoid processes in Saltasaurus but lateral to
them in Vahiny,asinJainosaurus. The basipterygoid processes
are autapomorphically thin and platy in Saltasaurus (PVL 4017),
and the occipital condyle is taller and narrower than in Vahiny.
Nemegtosaurus and Quaesitosaurus—Like Rapetosaurus,Ne-
megtosaurus (ZPAL MgD-I19) and Quaesitosaurus (PIN 3906/2)
are easily distinguished from Vahiny on the basis of characters of
the occipital condyle, basal tubera, basisphenoid, and basiptery-
goid processes (Fig. 6A, F). Nemegtosaurus exhibits a downward
facing occipital condyle when the supraoccipital is oriented ver-
tically, in contrast with the subparallel orientations of the skull
roof and occipital condyle in Vahiny. The basal tubera in Nemeg-
tosaurus and Quaesitosaurus are not separated by a bony sheet
as they are in Vahiny, and the basipterygoid processes are sepa-
rated by a deep notch more similar to those of Rapetosaurus and
Malawisaurus (SAM MAL 202). Finally, the openings for the in-
ternal carotid arteries are located between the basipterygoid pro-
cesses in Nemegtosaurus, rather than more laterally as in Vahiny
(Wilson, 2005b).
MalawisaurusMalawisaurus exhibits basipterygoid pro-
cesses that diverge via a deep notch and remain relatively straight
throughout their length (SAM MAL 202; Gomani, 2005), as in
Rapetosaurus and Nemegtosaurus. This contrasts with the broad-
ened divergence of the basipterygoid processes in Vahiny.Vahiny
and Malawisaurus share the presence of a shallow posterior fossa
between their basal tubera, but in Malawisaurus a midline ventral
flange of bone overhangs the proximal end of the basipterygoid
process. This feature is absent in Vahiny.
Synopsis
Although Vahiny and Rapetosaurus coexisted in the same en-
vironment, both share their closest taxonomic affinities with ti-
tanosaurs that lived elsewhere. The cranial material of Vahiny
indicates a close affinity to the Indian titanosaur Jainosaurus
and South American titanosaurs Muyelensaurus and Pitekun-
saurus. On the other hand, Rapetosaurus and Isisaurus share only
general, more broadly distributed anatomical similarities with
their respective contemporaries (Vahiny and Jainosaurus,respec-
tively), or with each other. The hypothesized sister-taxon rela-
tionship between Vahiny and Jainosaurus requires testing in a
phylogenetic context within a broad sample of titanosaurs. At
Downloaded by [141.213.177.149] at 06:07 07 May 2014
616 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 34, NO. 3, 2014
present, the anatomy supports the hypothesis that neither Mada-
gascar nor India present evidence of an endemic fauna produced
by long-term geographic isolation, as first noted by Dep´
eret
(1896a:485).
VAHINY AND MALAGASY ‘TAXON B’
Initial evaluations of sauropod diversity in the Maevarano For-
mation recognized only one genus and species, ‘Titanosaurus
madagascariensis,’ on the basis of poorly preserved and dissoci-
ated postcranial elements (Boule, 1896; Dep´
eret, 1896a, 1896b).
As more remains were recovered in subsequent years, hints of at
least one additional sauropod in the formation began to emerge.
Thevenin (1907:pl. 1, fig. 16) figured a caudal vertebra that dif-
fered from one of those included in the ‘T. madagascariensis
syntype (UCB 92829) in its robustness, elongation, strong pro-
coely, and straighter posterior neural arch margin. The figured
centrum bore more similarities to UCB 92305, the smaller of
the two centra described by Dep´
eret (1896a:pl. 6, fig. 2). Based
on the morphology of UCB 92305 and the vertebra figured by
Thevenin (1907:pl. 1, fig. 16), Huene (1929) questioned the refer-
ral of the Malagasy sauropods to ‘Titanosaurus’ and ultimately
referred the materials in question to cf. Laplatasaurus, a sauro-
pod he described from Upper Cretaceous of South America. This
referral was based largely on the distinctiveness of the Malagasy
titanosaur caudal vertebrae relative to those known from India
and Patagonia referred to ‘Titanosaurus’ (Huene, 1929:91).
Extensive excavations in the Maevarano Formation have
yielded articulated and associated sauropod individuals as well
as hundreds of isolated bones spanning hatchling to adult ontoge-
netic stages (Curry Rogers and Forster, 2001, 2004; Curry Rogers,
2005, 2009; Krause and Rasoamiaramanana, 2006; Krause et al.,
2006, 2010; Curry Rogers and Imker, 2007). Based on some
of this new material, Curry Rogers and Forster (2001) de-
scribed and named Rapetosaurus krausei on the basis of a
well-preserved adult skull and a referred juvenile skull and
its associated postcranial skeleton. Like Huene (1929), Curry
Rogers and Forster (2001, 2004) recognized two caudal vertebral
morphs in the syntypic material of ‘Titanosaurus madagascarien-
sis,’ and they tentatively referred one of the vertebrae (UCB
92829) to Rapetosaurus. Consistent with Huene’s (1929) hypoth-
esis, Curry Rogers and Forster (2001, 2004) speculated that the
other caudal vertebral morph (UCB 92305) indicated a differ-
ent titanosaur taxon that they informally referred to as ‘Mala-
gasy Taxon B’ (Curry Rogers and Forster, 2001; Curry Rogers,
2002; Curry Rogers and Imker, 2007). Shortly after the naming
of Rapetosaurus, Wilson and Upchurch (2003:146) posited that
Laplatasaurus madagascariensis is a nomen dubium because it is
not based upon associated, diagnostic material.
With regard to a second titanosaur taxon in the Maevarano
Formation, several isolated vertebrae, girdle elements, and
limbs that cannot be referred to Rapetosaurus have now been
recovered (e.g., Curry Rogers, 2002; Curry Rogers and Imker,
2007). These enigmatic bones, which constitute less than 10%
of the total sauropod collection from the unit, represent a vari-
ety of ontogenetic stages, and retain their distinctiveness from
Rapetosaurus in both juvenile and adult form (Curry Rogers and
Imker, 2007; McNulty et al., 2010). Despite their distinct mor-
phology, we did not refer these isolated postcranial remains to
Vahiny depereti in this report, and contend that assigning such
disparate and dissociated elements to Vahiny at this time would
likely propagate taxonomic problems. We restricted the holotype
of Vahiny to the braincase, which can be directly compared and
distinguished from the braincases of Rapetosaurus and other ti-
tanosaurs. The name-bearing materials of Vahiny depereti are di-
agnostic and stable, which provides a testable hypothesis for fu-
ture Malagasy titanosaur taxonomy. Field work in the Maevarano
Formation is ongoing, and we anticipate that additional collection
of sauropod fossils will shed light on the relationship between the
Vahiny depereti holotype and non-Rapetosaurus postcranial re-
mains recovered from the Maevarano Formation.
ACKNOWLEDGMENTS
We thank D. Krause, A. Rasoamiaramanana, and the mem-
bers of the Mahajanga Basin Project for their careful recovery
of these materials. In particular, A. Farke deserves recognition
for spotting the fragmentary remains of this skull on a weath-
ered Berivotra hill in 2005. We acknowledge the skillful prepara-
tion efforts of J. Groenke and V. Heisey. B. Simpson provided
collections access at the Field Museum of Natural History. J.
Martinez illustrated the interpretive drawings in Figure 3, and L.
Betti-Nash helped with photography and stereophoto organiza-
tion in Figures 1 and 2. We also thank R. Rogers, M. Carrano,
B. Bagley, and L. Bradtmiller for helpful discussions. A. Randri-
amihaja provided particular help with the correct pronunciation
of Vahiny. Comments by P. Mannion and A. Paulina Caraba-
jal significantly improved the quality of the manuscript. Grants
from the National Science Foundation (DEB-0822957 [to J.A.W.
and K.C.R.] and DEB-9224396, EAR-9418816, EAR-9706302,
DEB-9904045, EAR-0106477, EAR-0116517, and EAR-0446488
[to D. W. Krause]) and the National Geographic Society (to D.
W. Krause) supported this work.
LITERATURE CITED
Balanoff, A. M., G. S. Bever, and T. Ikejiri. 2010. The braincase of Ap-
atosaurus (Dinosauria: Sauropoda) based on computed tomogra-
phy of a new specimen with comment on variation and evolution in
sauropod neuroanatomy. American Museum Novitates 3677:1–29.
Berman, D. S., and S. L. Jain. 1982. Braincase of a small sauropod di-
nosaur (Reptilia: Saurischia) from the Upper Cretaceous Lameta
Group, central India, with review of Lameta Group localities. An-
nals of the Carnegie Museum 51:405–422.
Bonaparte, J. F. 1986. Les dinosaures (Carnosaures, Allosaurid´
es,
Sauropodes, C´
etiosaurid´
es) du Jurassique moyen de Cerro C ´
ondor
(Chubut, Argentina). Annales de Pal´
eontologie 72:325–386.
Bonaparte, J. F., and R. A. Coria. 1993. Un nuevo y gigantesco saur ´
opodo
titanosaurio de la Formaci ´
on R´
ıo Limay (Albiano-Cenomanio) de
la Provincia del Neuqu´
en, Argentina. Ameghiniana 30:271–282.
Boule, M. 1896. Note pr´
eliminaire sur les d´
ebris de dinosauriens envoy´
es
au Mus´
eum par M. Bastard. Bulletin du Mus´
eum d’Histoire Na-
turelle de Paris 2:347–351.
Calvo, J. O., B. J. Gonz´
alez Riga, and J. D. Porfiri. 2007. A new titanosaur
sauropod from the Late Cretaceous of Neuqu´
en, Patagonia, Ar-
gentina. Arquivos do Museu Nacional, Rio de Janeiro 65:484–504.
Chatterjee, S., and D. K. Rudra. 1996. KT events in India: impact, rift-
ing, volcanism and dinosaur extinction. Memoirs of the Queensland
Museum 29:389–532.
Curry Rogers, K. 2002. Titanosaur tails tell two tales: a second Malagasy
titanosaur with saltasaurid affinities. Journal of Vertebrate Paleon-
tology 22(3, Supplement):47A–48A.
Curry Rogers, K. 2005. Titanosauria: a phylogenetic overview; pp. 50–103
in K. Curry Rogers and J. A. Wilson (eds.), The Sauropods: Evo-
lution and Paleobiology. University of California Press, Berkeley,
California.
Curry Rogers, K. 2009. The postcranial osteology of Rapetosaurus krau-
sei (Sauropoda: Titanosauria) from the Late Cretaceous of Mada-
gascar. Journal of Vertebrate Paleontology 29:1046–1086.
Curry Rogers, K., and C. A. Forster. 2001. The last of the dinosaur titans:
a new sauropod from Madagascar. Nature 412:530–534.
Curry Rogers, K., and C. A. Forster. 2004. The skull of Rapetosaurus
krausei (Sauropoda: Titanosauria) from the Late Cretaceous of
Madagascar. Journal of Vertebrate Paleontology 24:121–144.
Curry Rogers, K., and M. J. Imker. 2007. New data on “Malagasy Taxon
B,” a titanosaur from the Late Cretaceous of Madagascar. Journal
of Vertebrate Paleontology 27(3, Supplement):64A.
Curry Rogers, K., M. D. D’Emic, R. Rogers, M. Vickaryous, and
A. Cagan. 2011. Sauropod dinosaur osteoderms from the Late
Downloaded by [141.213.177.149] at 06:07 07 May 2014
CURRY ROGERS AND WILSON—NEW TITANOSAUR FROM MADAGASCAR 617
Cretaceous of Madagascar. Nature Communications 2:564. doi:
10.1038/ncomms1578.
D’Emic, M. D., J. A. Wilson, and S. Chatterjee. 2009. The titanosaur (Di-
nosauria: Sauropoda) osteoderm record: review and first definitive
specimen from the Upper Cretaceous of India. Journal of Verte-
brate Paleontology 29:165–177.
Dep´
eret, C. 1896a. Note sur les dinosauriens sauropodes et th´
eropodes
dans le Cr´
etac´
esup
´
erieur de Madagascar. Bulletin de la Soci´
et´
e
G´
eologique de France 21:176–194.
Dep´
eret, C. 1896b. Sur l’existence de dinosauriens, sauropodes et
th´
eropodes dans le Cr´
etac´
esup
´
erieur de Madagascar. Comptes
Rendus Hebdomadiares des S´
eances de l’Acad´
emie des Sciences,
Paris 122:483–485.
Filippi, L. S., and A. C. Garrido. 2008. Pitekunsaurus macayai gen et
sp. nov., nuevo Titanosauria (Saurischia, Sauropoda) del Cret´
acico
Superior de la Cuenca Neuquina, Argentina. Ameghiniana 45:
575–590.
Gallina, P. A., and S. Apestigu´
ıa. 2011. Cranial anatomy and phylogenetic
position of the titanosaurian sauropod Bonitasaura salgadoi.Acta
Palaeontological Polonica 56:45–60.
Garcia, R. A., A. Paulina-Carabajal, and L. Salgado. 2008. Un nuevo
basicr´
aneo de titanosaurio de la Formaci ´
on Allen (Campaniano-
Maastrichtiano), Provincia de R´
ıo Negro, Patagonia, Argentina.
Geobios 41:625–633.
Gomani, E. 2005. Sauropod dinosaurs from the Early Cretaceous of
Malawi. Palaeontologia Electronica 8(1):27A.
Gonz´
alez Riga, B. J. 2003. A new titanosaur (Dinosauria, Sauropoda)
from the Upper Cretaceous of Mendoza, Argentina. Ameghiniana
40:155–172.
Huene, F. von. 1929. Los saurisquios y ornithisquios del Cret´
aceo Ar-
gentino. Anales del Museo de La Plata 2:1–196.
Huene, F. von, and C. A. Matley. 1933. Cretaceous Saurischia and Or-
nithischia from the Central Provinces of India. Palaeontologica In-
dica 21:1–74.
Krause, D. W., and A. H. Rasoamiaramanana. 2006. Late Cretaceous ver-
tebrates of Madagascar: implications for Gondwanan biogeography;
pp. 59–62 in P. M. Barrett and S. E. Evans (eds.), Ninth Interna-
tional Symposium on Mesozoic Terrestrial Ecosystems and Biota,
Manchester, U.K., 26–30 June 2006. Natural History Museum,
London.
Krause, D. W., P. M. O’Connor, K. Curry Rogers, S. D. Sampson,
G. A. Buckley, and R. R. Rogers. 2006. Late Cretaceous terres-
trial vertebrates from Madagascar: implications for Latin Ameri-
can biogeography. Annals of the Missouri Botanical Garden 93:178–
208.
Krause, D. W., J. J. W. Sertich, R. R. Rogers, S. C. Kast, A. H. Ra-
soamiaramanana, and G. A. Buckley. 2010. Overview of the dis-
covery, distribution, and geological context of Simosuchus clarki
(Crocodyliformes: Notosuchia) from the Late Cretaceous of Mada-
gascar; pp. 4–12 in D. W. Krause and N. J. Kley (eds.), Simosuchus
clarki (Crocodyliformes: Notosuchia) from the Late Cretaceous of
Madagascar. Memoir of the Society of Vertebrate Paleontology 10.
Lydekker, R. 1877. Notices of new and other Vertebrata from Indian Ter-
tiary and Secondary rocks. Records of the Geological Survey of In-
dia 10:30–43.
Lydekker, R. 1879. Indian pre-Tertiary Vertebrata. Part 3. Fossil Reptilia
and Batrachia. Palaeontologica Indica, Series 4 1:20–33.
Lydekker, R. 1893. The dinosaurs of Patagonia. Anales del Museo de La
Plata 2:1–14.
Mannion, P. D., and P. Upchurch. 2010. Completeness metrics and the
quality of the sauropodomorph fossil record through geological and
historical time. Paleobiology 36:283–302.
Marsh, O. C. 1878. Principal characters of American Jurassic dinosaurs.
Part I. American Journal of Science, Series 3 16:411–416.
Martinelli, A. G., and A. M. Forasiepi. 2004. Late Cretaceous vertebrates
from bajo de Santa Rosa (Allen Formation), R´
ıo Negro province,
Argentina, with the description of a new sauropod dinosaur (Ti-
tanosauridae). Revista del Museo Argentino de Ciencias Naturales,
Nuevo Serie 6:257–305.
McNulty, K., K. Curry Rogers, and O. Rudloff. 2010. Morphometric anal-
yses of titanosaur caudal vertebrae: implications for naming “Mala-
gasy Taxon B.” Journal of Vertebrate Paleontology, Program and
Abstracts 2010:133A.
Paulina Carabajal, A. 2012. Neuroanatomy of titanosaurid dinosaurs
from the Upper Cretaceous of Patagonia, with comments on en-
docranial variability within Sauropoda. The Anatomical Record
295:2141–2156.
Paulina Carabajal, A., and R. A. Coria. 2007. Un basicr´
aneo de ti-
tanosaurio (Dinosauria, Sauropoda) del Cret´
acico Superior del
norte de Patagonia: descripci ´
on y aportes al conocimiento del o´
ıdo
interno de los dinosaurios. Ameghiniana 44:109–120.
Powell, J. E. 2003. Revision of South American titanosaurid dinosaurs:
palaeobiological palaeobiogeographical, and phylogenetic aspects.
Records of the Queen Victoria Museum 111:1–94.
Rogers, R. R. 2005. Fine-grained debris flows and extraordinary ver-
tebrate burials in the Late Cretaceous of Madagascar. Geology
33:297–300.
Rogers, R. R., J. H. Hartman, and D. W. Krause. 2000. Stratigraphic anal-
ysis of Upper Cretaceous rocks in the Mahajanga Basin, northwest-
ern Madagascar: implications for ancient and modern faunas. Jour-
nal of Geology 108:275–301.
Rogers, R. R., D. W. Krause, K. Curry Rogers, A. Rasoamiaramanana,
and L. Rahantarisoa. 2007. Paleoenvironment and paleoecology
of Majungasaurus crenatissimus (Theropoda: Abelisauridae) from
the Late Cretaceous of Madagascar; pp. 21–31 in S. D. Sampson
and D. W. Krause (eds.), Majungasaurus crenatissimus (Theropoda:
Abelisauridae) from the Late Cretaceous of Madagascar. Memoir
of the Society of Vertebrate Paleontology 8.
Thevenin, A. 1907. Dinosauriens (Pal´
eontologie de Madagascar IV). An-
nals de Pal´
eontologie 2:121–136.
Upchurch, P., P. M. Barrett, and P. Dodson. 2004. Sauropoda; pp.
259–322 in D. B. Weishampel, P. Dodson, and H. Osm ´
olska (eds.),
The Dinosauria, second edition. University of California Press,
Berkeley, California.
Wilson, J. A. 2002. Sauropod dinosaur phylogeny: critique and cladis-
tic analysis. Zoological Journal of the Linnean Society 136:217–
276.
Wilson, J. A. 2005a. Overview of sauropod phylogeny and evolution; pp.
15–49 in K. Curry Rogers and J. A. Wilson (eds.), The Sauropods:
Evolution and Paleobiology. University of California Press, Berke-
ley, California.
Wilson, J. A. 2005b. A redescription of the skull of Nemegtosaurus
monogliensis (Dinosauria-Sauropoda) and its relevance to Cre-
taceous titanosaur diversity. Journal of Systematic Palaeontology
3:283–318.
Wilson, J. A. 2006. Anatomical nomenclature of fossil vertebrates: stan-
dardized terms or lingua franca? Journal of Vertebrate Paleontology
26:511–518.
Wilson, J. A., and P. Upchurch. 2003. A revision of Titanosaurus Ly-
dekker (Dinosauria-Sauropoda), the first dinosaur genus with a
‘Gondwanan’ distribution. Journal of Systematic Palaeontology
1:125–160.
Wilson, J. A., P. M. Barrett, and M. T. Carrano. 2011. An associ-
ated partial skeleton of Jainosaurus cf. septentrionalis (Dinosauria:
Sauropoda) from the Late Cretaceous of Chhota Simla, central In-
dia. Palaeontology 54:981–998.
Wilson, J. A., M. D. D’Emic, K. Curry Rogers, D. M. Mohabey, and
S. Sen. 2009. Reassessment of the sauropod dinosaur Jainosaurus
(=Antarctosaurus”) septentrionalis from the Upper Cretaceous of
India. Contributions from the University of Michigan Museum of
Paleontology 32:17–40.
Witmer, L. M., R. C. Ridgely, D. L. Dufeau, and M. C. Semones. 2008.
Using CT to peer into the past: 3D visualization of the brain and
ear regions of birds, crocodiles, and nonavian dinosaurs; pp. 67–87
in H. Endo and R. Frey (eds.), Anatomical Imaging: Towards a New
Morphology. Springer-Verlag, Tokyo.
Zaher, H., S. Apestegu´
ıa, and C. A. Scanferla. 2009. The anatomy of
the Upper Cretaceous snake Najash rionegrina Apestegu´
ıa & Za-
her, 2006, and the evolution of limblessness in snakes. Zoological
Journal of the Linnean Society 156:801–826.
Zaher, H., D. Pol, A. B. Carvalho, P. M. Nascimento, C. Riccomini,
P. Larson, R. Juarez-Valieri, R. Pires-Domingues, N. J. Silva Jr.,
and D. A. Campos. 2011. A complete skull of an Early Cretaceous
sauropod and the evolution of advanced titanosaurians. PLoS ONE
6:e16663.
Submitted May 27, 2012; revisions received June 3, 2013; accepted June
30, 2013.
Handling editor: Paul Barrett.
Downloaded by [141.213.177.149] at 06:07 07 May 2014
... The foramen magnum is taller dorsoventrally (42 mm) than it is wide transversely (36 mm), as in many, but not all, macronarians (Martínez et al., 2016). Among titanosaurs, the foramen magnum is taller than wide in Sarmientosaurus (Martínez et al., 2016), Pitekunsaurus (Filippi & Garrido, 2008), Antarctosaurus (Powell, 2003), Bonatitan (Martinelli & Forasiepi, 2004;Salgado et al., 2015a), Quaesitosaurus and Nemegtosaurus (Wilson, 2005), Jainosaurus and Vahiny Curry Rogers & Wilson, 2014; by contrast, the foramen magnum is subcircular in Malawisaurus (Gomani, 2005), Kaijutitan (Filippi et al., 2019), Narambuenatitan (Filippi et al., 2011), Muyelensaurus (Calvo et al., 2007a), Saltasaurus (Powell, 1992(Powell, , 2003, Rapetosaurus (Curry Rogers & Forster, 2004) and MPCM-HUE-8741 (Knoll et al., 2013). Each ventrolateral margin of the foramen magnum bears a small opening that communicates with another opening at the base (medial) of the paroccipital processes; this represents the passage of CN XII (hypoglossal). ...
... As mentioned above, the orbitosphenoid forms the anterior border of CN III, CN IV and CN VI (the rest of which lies on the basisphenoid). The opening for CN VI (abducens) is located posterolateral to CN II and immediately ventral to the opening for CN III, as in Antarctosaurus, Jainosaurus, Malawisaurus and Vahiny (Powell, 2003;Gomani, 2005;Curry Rogers & Wilson, 2014). By contrast, in many other titanosaurs, including Bonatitan, Rapetosaurus, Saltasaurus, Sarmientosaurus and Tapuiasaurus (Powell, 1992(Powell, , 2003Curry Rogers & Forster, 2004;Martinelli & Forasiepi, 2004;Martínez et al., 2016;Wilson et al., 2016), the foramen for CN VI lies anteroventral to that for CN III. ...
... This is the plesiomorphic eusauropod condition Paulina Carabajal et al., 2014;Poropat et al., 2016), which is also retained in the early-diverging titanosaurs Malawisaurus (Andrzejewski et al., 2019) and Sarmientosaurus (Martínez et al., 2016). By contrast, in most titanosaurs and in Tambatitanis (Saegusa & Ikeda, 2014) and, possibly, Mongolosaurus (Mannion, 2011), it opens medial to the basipterygoid process (Paulina Carabajal, 2012); Jainosaurus (Chatterjee & Rudra, 1996), Vahiny (Curry Rogers & Wilson, 2014) and Lirainosaurus (Díez Díaz et al., 2011) show reversals to the plesiomorphic state. ...
Article
The titanosaurian sauropod dinosaur Diamantinasaurus matildae is represented by two individuals from the Cenomanian-lower Turonian 'upper' Winton Formation of central Queensland, northeastern Australia. The type specimen has been described in detail, whereas the referred specimen, which includes several elements not present in the type series (partial skull, atlas, axis and postaxial cervical vertebrae), has only been described briefly. Herein, we provide a comprehensive description of this referred specimen, including a thorough assessment of the external and internal anatomy of the braincase, and identify several new autapomorphies of D. matildae. Via an expanded data matrix consisting of 125 taxa scored for 552 characters, we recover a close, well-supported relationship between Diamantinasaurus and its contemporary, Savannasaurus elliottorum. Unlike previous iterations of this data matrix, under a parsimony framework we consistently recover Diamantinasaurus and Savannasaurus as early-diverging members of Titanosauria using both equal weighting and extended implied weighting, with the overall topology largely consistent between analyses. We erect a new clade, named Diamantinasauria herein, that also includes the contemporaneous Sarmientosaurus musacchioi from southern Argentina, which shares several cranial features with the referred Diamantinasaurus specimen. Thus, Diamantinasauria is represented in the mid-Cretaceous of both South America and Australia, supporting the hypothesis that some titanosaurians, in addition to megaraptoran theropods and possibly some ornithopods, were able to disperse between these two continents via Antarctica. Conversely, there is no evidence for rebbachisaurids in Australia, which might indicate that they were unable to expand into high latitudes before their extinction in the Cenomanian-Turonian. Likewise, there is no evidence for titanosaurs with procoelous caudal vertebrae in the mid-Cretaceous Australian record, despite scarce but compelling evidence for their presence in both Antarctica and New Zealand during the Campanian-Maastrichtian. These later titanosaurs presumably dispersed into these landmasses from South America before the Campanian (~85 Mya), when seafloor spreading between Zealandia and Australia commenced. Although Australian mid-Cretaceous dinosaur faunas appear to be cosmopolitan at higher taxonomic levels, closer affinities with South America at finer scales are becoming better supported for sauropods, theropods and ornithopods.
... The description of titanosaurids from Pakistan is still in an early stage, and a detailed comparison between these two regions is still awaited (Kapur and Khosla 2016). Rogers and Wilson (2014) compared the cranial morphology of two giant titanosaurids, Vahiny depereti (discovered from the Late Cretaceous of Madagascar) and Indian Jainosaurus, and, based on similarity between these two sauropods, proposed a close palaeobiogeographic connection between these two subcontinents. The titanosaurids from the Late Cretaceous of South America also show close morphological similarities with the Indian taxon Isisaurus (Wilson et al. 2011). ...
Article
Full-text available
The Infra- and Intertrappean deposits have yielded diverse vertebrates, especially dinosaurs, mammals, snakes, turtles, crocodiles, and invertebrates. The biotic assemblages demonstrate a remarkable degree of resemblance between these deposits. The palaeobiogeographical affinities of the paleobiota are more intricate, yielding remains of Gondwanan and Laurasian affinities, and some endemic forms. In order to explain the presence of such a complex biota during the northward drift of the Indian plate (Late Cretaceous period), different palaeobiogeographical models have been proposed. Special emphasis has been given in this paper to the palaeobiogeographical implications of Indian Late Cretaceous dinosaurs. The size of the animal played an essential role in determining the nature of the biotic interchange between India and its nearby landmasses. The faunal exchange between India and Asia through the Kohistan Dras volcanic arc system has been considered as the superlative migratory route, which favoured the small fauna during transmaritime dispersal. Conversely, it was difficult for small animals to cross huge marine boundaries, other than for the very large vertebrates (especially dinosaurs). Consequently, a straight terrestrial course, especially in the northern India, is a lesser probability, and the dispersal of these huge vertebrates ought to be seen as part of a ‘Pan Gondwanan’ model.
... Therefore, the trackmakers of Teratopodus were assigned to derived titanosaurs, as shown in the fossil record of the Early Campanian. Titanosaurs were the most diverse and abundant largebodied terrestrial herbivores in the Southern Hemisphere landmasses during the Late Cretaceous (Wilson, 2006;Curry Rogers and Wilson, 2014;Lacovara et al., 2014;Otero and Salgado, 2015). The number of titanosaur taxa has increased drastically in recent years, with roughly 53 species valid from South America . ...
Article
A new titanosaur sauropod ichnotaxon named Teratopodus malarguensis gen. et sp. nov is erected based on a unique association of 10 characters. It is represented by new tracks discovered in the upper section of the Anacleto Formation outcropping in the southern Mendoza Province (middle Campanian, Neuquén Basin). Teratopodus diagnosis includes pes tracks with suboval shaped, with a slightly obtuse ‘V-shaped’ heel and three large claws impressions laterally deflected corresponding to the digits I, II and III; and symmetrical kidney-shaped manus tracks with a slightly concave posterior border. Teratopodus has a medium-gauge trackway and a medium degree of heteropody (∼1/3). In addition, a new characterization of the heels impressions in sauropod tracks is proposed here including the postero-latero-medial angle (PLM°); according to this, Teratopodus pes track can be classified into slightly ‘V-shaped’ heel (PLM° = 95°–100°). The trackmakers are two titanosaurs specimens of relatively small-medium sized of ca. 11 m and 14 m length. These tracks are preserved in a poorly drained floodplain deposits with ephemeral channels controlled by flash flood episodes. In this context, the trackmakers walked from a humid ground to a more flooded one, leaving behind one of the best-preserved pes tracks ever recorded in South America.
... The specimens under preparation at the DAK facility included approximately 30 bones of a then undescribed titanosaur (Malagasy Taxon B of Curry Rogers, 2009), now Vahiny depereti (Curry Rogers and Wilson, 2014), which consisted of partially-articulated and disarticulated postcranial skeletal remains preserved in a loosely cemented, finegrained sediment (Rogers, 2005). These specimens were excavated from Facies 2 of the Anembalemba Member, Maevarano Formation, Quarry MAD93-18. ...
... A palaeobiogeographic connection between Late Cretaceous India, Madagascar and South America was proposed by Wilson et al. (2011) and Rogers and Wilson (2014) based on the comparative cranial morphologies of three titanosaurids (Jainosaurus septentrionalis, Vahiny depereti and Isisaurus). The occurrence of huge titanosaurids in India, South America and Madagascar indicates extensive and long-lasting land associations between these three continents during the Late Cretaceous (Khosla 2019). ...
Chapter
The morphostructural diversity of Indian, French and Argentinean eggshells has been compared and reviewed in great detail and a serious attempt has been made to evaluate their parataxonomy. Indian and French eggshell oospecies show a close resemblance between five Indian and four French oospecies (Megaloolithus khempurensis = M. siruguei; Megaloolithus jabalpurensis = M. mamillare; M. cylindricus = M. siruguei and M. microtuberculata; Fusioolithus padiyalensis = M. microtuberculata; F. baghensis = M. pseudomamillare). Examination of four oospecies from India and Argentina uncovers three groupings, which show similarity between megaloolithids and fusioolithids of the two continents (Fusioolithus baghensis = Patagoolithus salitralensis; Megaloolithus megadermus = Tipo 1e; M. cylindricus = Tipo 1d; M. jabalpurensis = Tipo 1e). The biomineralization aspects of Indian eggshells or any calcified tissue with dental enamel have been compared and studied at four levels (crystallite, unit, morphostructural and megascopic levels). The dinosaur nests, eggs and eggshell fragments are found to be diagenetically altered by calcite recrystallization (herringbone pattern) and chertified silica, which has been commonly noticed in the pores and pore canals.
... With respect to body size evolution in Titanosauria, the record of extremely large-bodied forms has traditionally been mostly fragmentary (Bonaparte and Coria, 1993;Novas et al., 2005). Since the early 2000s, our knowledge of titanosaurian sauropods has notably increased by virtue of the influx of new fossil discoveries (e.g., Novas et al., 2005;Hocknull et al., 2009;Zaher et al., 2011;Curry Rogers and Wilson, 2014;Gorscak et al., 2014;Lacovara et al., 2014;Lü et al., 2014;Diez Díaz et al., 2016;González Riga et al., 2016;Carballido et al., 2017) and detailed descriptions of previously known taxa (e.g., Curry Rogers, 2009;Otero, 2010;Díez Díaz et al., 2011, 2013a, 2013bMannion and Otero, 2012;Gallina and Apesteguía, 2014;Gallina and Otero, 2015;Zurriaguz and Powell, 2015;Poropat et al., 2015Poropat et al., , 2016González Riga et al., 2018). All of this new information contributed to a better understanding of the phylogenetic arrangement of this group thanks to new cladistic analyses that were conducted with the successive addition of taxa to phylogenetic data matrices (e.g., Mannion et al., 2013;Carballido and Sander, 2014;Gorscak and O'Connor, 2016;Mannion et al., 2017;Carballido et al., 2017Carballido et al., , 2020González Riga et al., 2016. ...
Article
Full-text available
Its huge size, excellent preservation, and completeness make Patagotitan mayorum a unique opportunity to explore the anatomy, paleobiological, and phylogenetic aspects linked to gigantism within Sauropoda. In this regard, we describe the appendicular skeleton of this titanosaurian species from the late Albian-aged Cerro Barcino Formation of Chubut Province, Argentina. The diagnosis of Patagotitan mayorum is revised, increasing the number of identified autapomorphies (i.e., lateral surface of the scapular blade with two divergent crests; anterior surface of proximal humerus with paired muscle scars; combined bulges on the deltopectoral area of the humerus; ischium with well-developed and sharp ridge projecting from the ischial tuberosity to the distal blade). Several diagnostic characters of this species correspond to osteological correlates associated to appendicular musculature (e.g., Mm. deltoideus scapularis, deltoideus clavicularis, and teres major; M. coracobrachialis; Mm. supracoracoideus/deltoideus clavicularis and latissimus dorsi; Mm. flexor tibialis 3 and adductor femoris 2), which we discuss in the context of sauropod evolution. In the light of a modification of the scaling equation previously proposed and adjusting the long bone circumference for the humeri of Patagotitan, a new body mass estimate of this species ranges between 42-71 tons, with a mean value of 57 tons. Although considerably less than the value obtained by the original linear equation, the corrected quadratic equation used here provides a mean body mass estimate that is more consistent with those derived from volumetric reconstructions of Patagotitan.
... The description of titanosaurids from Pakistan is still in an early stage, and a detailed comparison between these two regions is still awaited (Kapur and Khosla 2016). Rogers and Wilson (2014) compared the cranial morphology of two giant titanosaurids, Vahiny depereti (discovered from the Late Cretaceous of Madagascar) and Indian Jainosaurus, and, based on similarity between these two sauropods, proposed a close palaeobiogeographic connection between these two subcontinents. The titanosaurids from the Late Cretaceous of South America also show close morphological similarities with the Indian taxon Isisaurus (Wilson et al. 2011). ...
Article
Full-text available
The Infra- and Intertrappean deposits have yielded diverse vertebrates, especially dinosaurs, mammals, snakes, turtles, crocodiles, and invertebrates. The biotic assemblages demonstrate a remarkable degree of resemblance between these deposits. The palaeobiogeographical affinities of the paleobiota are more intricate, yielding remains of Gondwanan and Laurasian affinities, and some endemic forms. In order to explain the presence of such a complex biota during the northward drift of the Indian plate (Late Cretaceous period), different palaeobiogeographical models have been proposed. Special emphasis has been given in this paper to the palaeobiogeographical implications of Indian Late Cretaceous dinosaurs. The size of the animal played an essential role in determining the nature of the biotic interchange between India and its nearby landmasses. The faunal exchange between India and Asia through the Kohistan Dras volcanic arc system has been considered as the superlative migratory route, which favoured the small fauna during trans-maritime dispersal. Conversely, it was difficult for small animals to cross huge marine boundaries, other than for the very large vertebrates (especially dinosaurs). Consequently, a straight terrestrial course, especially in the northern India, is a lesser probability, and the dispersal of these huge vertebrates ought to be seen as part of a ‘Pan Gondwanan’ model.
Article
Despite dominating biodiversity in the Mesozoic, dinosaurs were not speciose. Oviparity constrained even gigantic dinosaurs to less than 15 kg at birth; growth through multiple morphologies led to the consumption of different resources at each stage. Such disparity between neonates and adults could have influenced the structure and diversity of dinosaur communities. Here, we quantified this effect for 43 communities across 136 million years and seven continents. We found that megatheropods (more than 1000 kg) such as tyrannosaurs had specific effects on dinosaur community structure. Although herbivores spanned the body size range, communities with megatheropods lacked carnivores weighing 100 to 1000 kg. We demonstrate that juvenile megatheropods likely filled the mesocarnivore niche, resulting in reduced overall taxonomic diversity. The consistency of this pattern suggests that ontogenetic niche shift was an important factor in generating dinosaur community structure and diversity.
Chapter
The fossil record of dinosaurs from India provides a highly significant contribution to understanding the origin and evolution of dinosaurs and their paleobiogeographic significance. As India rifted from Gondwana and drifted northwards during the age of dinosaurs, the mobile episode in Indian geology provides a unique opportunity to study the diversity of dinosaurs in time and space. The dinosaurs from the Gondwana and post-Gondwana sediments of India have been collected and studied since their discovery in the 1920s, but the full range of their significance and evolutionary history remained fragmentary. After the independence of India, a renaissance arose in the study of dinosaurs at the Indian Statistical Institute (ISI) under the leadership of Pamela Robinson, as more and more dinosaur skeletons were discovered from different localities. This exploration by ISI paleontologists represented a pivotal moment in the history of vertebrate paleontology in India and became a starting point for a remarkable increase in our knowledge of Triassic, Jurassic, and Cretaceous dinosaur faunas. It inspired a new generation of students working under Ashok Sahni’s direction at Panjab University to engage in the Cretaceous research. This paper offers an updated and comprehensive review of the anatomy, systematics, and evolution of Indian dinosaurs within historical, paleobiogeographic, and paleoecologic contexts. The occurrence of Indian dinosaurs is currently restricted to central and southern India, and the record extends across all three Mesozoic periods. It is generally regarded that dinosaurs originated in the Late Triassic Period in Argentina, about 230 million years ago. However, Alwalkeria, a theropod discovered in the Lower Maleri Formation of India, was contemporaneous with the oldest Argentinean dinosaurs. Similarly, Barapasaurus from the Early Jurassic Kota Formation is considered as one of the oldest, gigantic sauropod dinosaurs with a quadrupedal pose. The Late Triassic and Early Jurassic dinosaurs of India are diverse and document their early radiation. With the breakup of Gondwana, India began to disintegrate and drifted northwards, carrying its dinosaur fauna like a passenger ship, until it collided with the Oman-Kohistan-Ladakh Arc in the Late Cretaceous, forming a biotic corridor to Africa and Europe. The Late Cretaceous dinosaurs from the Lameta Formation, consisting of several species of titanosaurs and abelisaurs, provide intimate documentation of the last ‘geologic minutes’ before their extinction. Along with dinosaur bones, the largest titanosaurid hatchery is known from the Lameta Formation, extending for more than 1,000 km. Most egg clutches contain about 10 to 12 spherical eggs ranging in diameter from 15 to 20 cm. Surprisingly, these eggs were empty, showing no signs of embryos, perhaps indicating hatching failure during some environmental crisis. At the Cretaceous-Paleogene boundary, India was ground zero for two catastrophic events—the Shiva impact and Deccan volcanism—both linked to dinosaur extinction. The combination of twin asteroid impacts (Chicxulub and Shiva), with prolonged Deccan volcanism created an unprecedented and ultimately catastrophic environmental crisis across the globe, triggering the end-Cretaceous mass extinction.
Article
Full-text available
Cretaceous dinosaurs were first reported from the Indian subcontinent in the late 1800s, and titanosaur sauropod and abelisauroid theropod remains are now known from central, western, and southern parts of India and from central western Pakistan. Although dinosaur remains are abundant, associated or articulated specimens are extremely rare, and so are complex skeletal elements such as cranial bones and presacral vertebrae. The historical pattern of sampling and collecting has limited the inferences about patterns of diversity, phylogenetic affinity, and paleobiogeographic relationships of Indian dinosaurs. Here we report on three titanosaur vertebrae representing regions of the skeleton that are complex and otherwise poorly represented in the Indian record, including two anterior dorsal vertebrae pertaining to a single individual from Rahioli, in Gujarat State (western India), and an anterior caudal neural arch from Bara Simla, in Madhya Pradesh State (central India). Phylogenetic analysis places the two individuals within Titanosauria, but further resolution of their affinities is precluded by their incompleteness and that of titanosaur vertebral columns in general, lack of coding of character data for titanosaur presacral and anterior caudal vertebrae, and relatively coarse understanding of the evolutionary relationships of titanosaurs. Comparisons with contemporaneous and spatially proximal titanosaurs from Indo-Pakistan, Madagascar, and South America provide insights into their affinities. The dorsal vertebrae share close affinity with Isisaurus from India and Mendozasaurus from Argentina. Few local comparisons are available for the anterior caudal vertebra, which shares characteristics with Tengrisaurus from the Early Cretaceous of Russia.
Article
Full-text available
Rapetosaurus krausei (Sauropoda: Titanosauria) from the Upper Cretaceous Maevarano Formation of Madagascar is the best-preserved and most complete titanosaur yet described. The skull of Rapetosaurus is particularly significant because most titanosaurs are diagnosed solely on the basis of fragmentary postcranial material, and knowledge of the titanosaur skull has remained incomplete. Material referred to Rapetosaurus includes the type skull from an adult that preserves the basicranium, rostrum, mandible, and palate. A second, juvenile skull preserves most of the braincase and cranial vault, as well as some of the palate and lower jaw. Here we provide a detailed description of Ropetosaurus cranial anatomy and highlight comparative relationships among known titanosaur and other neosauropod skulls. The Rapetosaurus skull is similar to those of diplodocoids in its overall shape, with retracted external nares and an elongated snout. However, extensive tooth distribution and bone articulations surrounding the external narial region and orbit are more similar to those of macronarians like Camarasaurus and Brachiosaurus. The maxilla, basicranium, paroccipital process, and pterygoid are among the most diagnostic elements of the Rapetosaurus skull, along with the enlarged antorbital fenestra, anteroventrally oriented braincase, and mandible. Titanosaur crania exhibit a greater diversity than previously recognized and, in light of Rapetosaurus, it is apparent that there is not a narrowly constrained bauplan for the skull of titanosaurs. Broad generalizations about evolution based on previously known, fragmentary fossils require re-evaluation. Ultimately, Rapetosaurus will be key in resolving titanosaur higher-level and ingroup phylogeny.
Article
Whereas the anatomy of birds, domesticated mammals, and humans is described by standardized terminology, the anatomy of most fossil vertebrates is described by nonstandardized terminology. New fossil discoveries increasingly resolve the transitions between these living groups and their fossil outgroups, diminishing morphological differences between them, and vertebrate paleontologists can easily apply more than one system of anatomical terms to such groups. This plurality of systems has led to recent proposals to standardize anatomical terminology for tetrapods, either by applying avian and mammalian anatomical terminology to their respective stem groups (Sauropsida and Synapsida) or by creating an all-encompassing terminology for Tetrapoda from a combination of existing terminologies. The main rationale for implementing standardized anatomical terminology, which requires abandoning competing terminologies, is that it reflects homology and evolutionary descent, eliminates ambiguity, and enhances interdisciplinary communication. The proposed standardized anatomical terminology, however, entails many negative consequences, including reversing character trajectories, misrepresenting complex anatomical transformations and uncertain homologies, and requiring far-reaching terminological conversions. These negative consequences result from increasing the taxonomic scope of standardized anatomical terms that were developed for a specific group, but now: (1) apply to a broader hierarchy of character states; (2) involve additional phylogenetic interpretations or assumptions; and (3) are used for basal, often more generalized conditions. In contrast, traditional non-standardized anatomical terminology, although not strictly phylogenetic, is anatomical ‘lingua franca’ that has been in usage for nearly two centuries and is consistent, ubiquitous, and descriptive.
Article
Mendozasaurus neguyelap gen. et sp. nov. is a new titanosaur from the Upper Cretaceous of Neuquen Basin and the first known dinosaur species from Mendoza Province, Argentina. The remains were found in levels provisionally referred to the Rio Neuquen Formation (late Turonian-late Coniacian) of the Neuquen Group. They consist of 22 mostly articulated caudal vertebrae, a dorsal vertebra, numerous disarticulated appendicular bones and osteoderms. Autapomorphies of Mendozasaurus are: (1) subtriangular infrapostzygapophyseal fossae in anterior dorsal vertebrae; (2) postzygapostspinal laminae parallel to the plane of the postzygapophyseal facets in anterior dorsal vertebrae; (3) interzygapophyseal cavity dorsoventrally extended and limited by the spinopostzygapophyseal and spinoprezygapophyseal laminae in anterior caudal vertebrae; (4) middle caudal vertebrae slightly procoelous with reduced posterior condyles displaced dorsally; (5) laminar mid-posterior caudal neural spines with horizontal and straight dorsal border, and anterodorsal corner forming a right angle; (6) large subconic-spherical osteoderms lacking cingulum. A cladistic analysis permits the inclusion of Mendozasaurus within Titanosauridae, according to the phylogenetic definition of this clade. Although this new species exhibits almost all the titanosaurid synapomorphies proposed by some authors, it lacks prominent posterior condyles in middle caudal centra. This and other plesiomorphic traits suggest that Mendozasaurus is a basal titanosaurid, more derived than Malawisaurus in the caudal procoely.
Article
A new titanosaur is described, Pitekunsaurus macayat gen, et sp. nov., from mudstone levels asigned to Anacleto Formation (Lower - Middle Campaman), corresponding to the uppermost beds of the Neuquén Group (Upper Cretaceous of Neuquén Basin). The specimen is represented by braincase, left frontal, one tooth, four cervical vertebrae, three dorsal vertebrae, four caudal vertebrae, right ulna and scapula, proximal extreme of left femur, rib fragments and uncertain remains. Pitekunsaurus is characterized by the following autapomorphies: (1) basipterygoid processes broadly separated and parallelly projected, (2) anterior cervical vertebrae with small depressions or longitudinal grooves in the spinal sector of spinopostzygapophyseal lamina, (3) centropostzygapophyseal lamina forked proximally in anterior dorsal vertebrae, and (4) posterior centrodiapophyseal lamina with accessory lamina in anterior dorsal vertebrae. The existence of two types of articulations in the posterior caudal vertebrae, one amphicoelous and another biconvex, indicates a close relationship with Rinconsaurus caudamirus Calvo y Gonzalez Riga, suggesting that the caudal morphology of titanosaurs is much more complex and more varied than previously supposed.