![Lateral and palatal views of the skulls of: A and D, the theropod Tyrannosaurus (redrawn from Carr [1999]); B and E, the "sphenosuchian" Dibothrosuchus (redrawn from Wu & Chatterjee [1993]); and C and F, a generalized peirosaurid based on both Uberabasuchus (Carvalho et al., 2004) and Lomasuchus (Gasparini et al., 1991). Scale bar equals 25 cm in A and D, and 1 cm in B, C, E, and F. The coloured rectangle approximately indicates the portion of the skull corresponding to the specimen MSNM V5770. For abbreviations see text.](https://www.researchgate.net/profile/Simone-Maganuco/publication/40661608/figure/fig6/AS:668899871825931@1536489565454/Lateral-and-palatal-views-of-the-skulls-of-A-and-D-the-theropod-Tyrannosaurus-redrawn_Q320.jpg)
Content uploaded by Simone Maganuco
Author content
All content in this area was uploaded by Simone Maganuco
Content may be subject to copyright.
Abstract – Here we report on the finding of new vertebrate remains from the Middle Jurassic
(Bathonian) of the Mahajanga basin, NW Madagascar, that represent one of the largest predatory
archosaurs from Gondwana. The remains consist of a fragmentary right maxilla bearing three teeth and
seven isolated teeth that clearly belong to a single taxon. Although the scarcity of the remains renders
establishment of affinities difficult, the presence of a unique combination of features including a well
developed medial shelf of the maxilla, rostral and lateral teeth respectively U-shaped and sub-oval in
cross-section, and very large tooth denticles (1 per mm), allows us to erect a new taxon, Razanandrongobe
sakalavae. Additionally, the structure of the teeth, the peculiar wear of their enamel, and the morphol-
ogy of the maxilla strongly suggest that Razanandrongobe sakalavae often crushed the bones of its
prey.
Key words: Archosauria, teeth, maxilla, Middle Jurassic, Madagascar.
Riassunto – Un nuovo grande arcosauro predatore nel Giurassico medio (Bathoniano) del
Madagascar.
E’ qui segnalato il ritrovamento di nuovi resti di vertebrati, provenienti dal Giurassico medio
(Bathoniano) del bacino di Mahajanga, Madagascar nord-occidentale, che appartengono a uno dei più
grandi arcosauri predatori che abitarono le terre di Gondwana. I resti consistono in un frammento di
mascella con ancora impiantati tre denti, più sette denti isolati, che appartengono chiaramente ad un
solo taxon. Sebbene la frammentarietà dei reperti renda difficile stabilirne le relazioni, la peculiare
combinazione di caratteri, tra cui una mensola palatale del mascellare ben sviluppata, denti rostrali e
laterali rispettivamente a forma di U e sub-ovali in sezione trasversa, e carene con denticoli decisa-
mente grandi (1 per mm) ci consente di erigere un nuovo taxon, Razanandrongobe sakalavae. La strut-
tura dei denti, il particolare grado di usura del loro smalto e la morfologia del mascellare suggerisco-
no inoltre che Razanandrongobe sakalavae fosse solito triturare le ossa delle prede.
Parole chiave: Archosauria, denti, mascella, Giurassico medio, Madagascar.
Atti Soc. it. Sci. nat. Museo civ. Stor. nat. Milano, 147 (I): 19-51, Gennaio 2006
Simone Maganuco*, Cristiano Dal Sasso**
&
Giovanni Pasini***
A new large predatory archosaur
from the Middle Jurassic (Bathonian) of Madagascar
*Museo Civico di Storia Naturale, Corso Venezia 55, 20121 Milano, Italy,
e-mail: simonemaganuco@iol.it
**Museo Civico di Storia Naturale, Corso Venezia 55, 20121 Milano, Italy,
e-mail: cdalsasso@yahoo.com (corresponding author).
***Museo Civico dei Fossili di Besano, Via Prestini 5, 21050 Besano (Varese), Italy,
e-mail: museodibesano@tiscali.it
20 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Introduction
The Mahajanga Basin (Fig. 1) comprises a large region of north-western
Madagascar, where deposition of continental, deltaic and partly marine sediments
occurred since Permian times (Fig. 2A). In recent years, the Mahajanga Basin has
become famous in the palaeontological literature for its Late Cretaceous
(Maastricthian) invertebrates and vertebrate-rich outcrops (Maevarano and
Berivotra Formations). The equally fossiliferous Late Triassic and Jurassic deposits
(Isalo ‘Group’), exposed over a several-hundred-kilometre-long arc along the east-
ern and southern margins of the basin, have been object of few research subsequent
to work by Lydekker (1894) (e.g., Thevenin, 1907; Besairie, 1936; Lavocat, 1955;
Bonaparte, 1986), only in the last decade have they received more attention (Flynn
et al., 1997, 1999; Dal Sasso & Pasini, 2003; Maganuco et al., 2005; Buffetaut,
2005). The Middle Jurassic deposits of the Mahajanga Basin contain a peculiar and
poorly known fossil vertebrate fauna. Teeth of thalattosuchian crocodyliforms, ple-
siosaurs and possibly ichthyosaurs, as well as dinosaurian remains (mainly
sauropods), were reported in the last century (Thevenin, 1907; Besairie, 1936, 1972;
Lavocat, 1955; Ogier, 1975; Bonaparte, 1986). Remains of an early tribosphenic
mammal (Flynn et al., 1999a), pterosaur teeth (Dal Sasso & Pasini, 2003), sauropod
teeth and one dentary (Buffetaut, 2005), and diverse theropod teeth and vertebrae
have been also reported (Flynn et al., 1997; Maganuco et al., 2005).
Bevetzaka
Ambondromamy
Tsinjorano
Andranomamy
Antananarivo
RN4
Mahajanga
Mahajanga
Andobokely lake
Bernakamba lake
Kamoro river
Antananarivo
5Km
*
Fig. 1 - Map of the Triassic and Jurassic outcrops of the Mahajanga Basin (black areas on the left) and
close-up of the localities recently prospected by the authors. Part of the material here described (the
isolated teeth) comes from the area marked by the asterisk.
Fig. 1 - Mappa dei giacimenti triassici e giurassici del Bacino di Mahajanga (aree annerite nella carti-
na a sinistra) e particolare delle località recentemente prospettate dagli autori. Parte del materiale qui
descritto (i denti isolati) proviene dall’area indicata dall’asterisco.
In October 2001, within an agreement between the Museo di Storia Naturale di
Milano and the Ministère de l’Energie et des Mines together with the Direction des
Mines et de la Géologie de Madagascar, some inspection by one of the authors (G.
Pasini) verified the presence and the extension of fossiliferous outcrops of the Isalo
IIIb Subunit (Fig. 2B), and their potential for further research as well. Under that
aim, four temporary concessions to prospection (Bureau du Cadastre Minier
n.3402-3405/Carte 41) were acquired in the area between Ambondromamy and
Ambalanjanakomby. In April 2003, a joint expedition of the Museo di Storia
Naturale di Milano and the Museo Civico dei Fossili di Besano, promoted by an
Italian private sponsor (BIOE s.r.l.), gave the authors the opportunity to re-prospect
the same area and to collect the first Malagasy remains of pterosaurs (Dal Sasso &
Pasini, 2003), as well as the isolated teeth here examined. During the 2001 field
21
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 2 - Schematic geological sections of the whole Mahajanga basin (A) and near the village of
Ambondromamy (B). The arrow indicates the Isalo IIIb Subunit “Faciés mixte - dinosauriens”
Bathonian sensu Besairie (1972), which our specimens come from. Modified from Besairie (1972).
Fig. 2 - Sezione geologica schematica generale del bacino di Mahajanga (A) e in particolare nei pres-
si di Ambondromamy (B). La freccia indica la Subunità Isalo IIIb “Faciés mixte - dinosauriens”
Bathonian sensu Besairie (1972), dalla quale provengono gli esemplari. Tratta da Besairie (1972), sem-
plificata.
Fig. 3 - One of the outcrops of the Middle Jurassic (Bathonian) Isalo IIIb subunit near the village of
Ambondromamy.
Fig. 3 - Uno degli affioramenti della subunità stratigrafica Isalo IIIb, riferibile al Giurassico medio
(Bathoniano), nei pressi del villaggio di Ambondromamy.
Geological setting
Between the villages of Ambondromamy and Ambalanjanakomby, about 170 km
E-SE of Mahajanga, the national road to Tananarive (RN4) crosses a succession of
badlands, locally known as tanety (hills), created by weathering into the Jurassic sed-
SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
22
survey, local collectors reported the fortuitous finding on the surface, in the neigh-
bourhood of the village of Ambondromamy (Fig. 3), of two cranial fragments with
well preserved teeth still in place, belonging to two different kinds of fossil rep-
tiles. After several events, the two cranial fragments were acquired in Europe by
Gilles Emringer and Francois Escuillié of the Eldonia society of Gannat, France,
with the purpose to make them available for scientific description. The first spec-
imen, a sauropod mandibular fragment now housed in the Toulouse Natural
History Museum, was studied by E. Buffetaut (Buffetaut, 2005). G. Pasini was able
to see the second specimen in June 2003 and to recognize that the teeth still
implanted in the bone were identical to the teeth he had collected in the field in the
meantime, during the prospecting activity in April 2003. Therefore the Museo
Civico di Storia Naturale di Milano acquired the specimen, in order to study it
together with the isolated teeth collected around the same area. Here we describe
all those remains as diagnostic of a new taxon, Razanandrongobe sakalavae, which
potentially represents the largest predator of the Malagasy Middle Jurassic terres-
trial ecosystem.
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR 23
iments that constitute the eastern margin of the Mahajanga Basin (Fig. 1). These out-
crops were correlated to the upper levels of the Isalo ‘Group’ (W Madagascar) and
named ‘Isalo III-Facies Continental’ by Besairie (1936). Subsequently, the same
author (Besairie 1972) distinguished the top of the series as ‘Isalo IIIb-Bathonien
Facies Mixte Dinosauriens’ because of its heterolithic nature, which was verified by
Boast & Nairn (1982). According to Besairie (1972) this subunit is about 190 m thick
and consists of alternating, variegated sandstones and siltstones arranged in cross-
bedding layers, calcareous paves and multicoloured claystone banks. Silicified wood
is sometimes present. Besairie (1972) estimated a Bathonian age for Isalo IIIb (ICS
stage Bathonian 167.7 – 164.7 Ma in Gradstein et al. [2004]) on the basis of its rich
invertebrate fauna including the index echinoids Nucleolites amplus and Acrosalenia
colcanapi. The isolated teeth were collected in loose sediments on the surface in the
Isalo IIIb subunit composing the tanety located about 8 km W-NW of the village of
Andranomamy, at an altitude of ~60 m (Fig. 3). During the preparation of the maxil-
la we collected the matrix to analyze the sediment. The matrix is composed of a fine-
grained, well sorted sandstone, with very fine-grained carbonatic cement. Quartz is
the main clastic component; among the accessories, garnet and zircon are scarce and
are present with about the same ratio, whereas ilmenite is more abundant, as usually
seen in the alluvial deposits. Granulometry ranges from 0.2 to 0.3 mm. Silty, black-
ish fragments, rich in organic matter, are also present (Pezzotta, pers. com. 2005). The
latter might be related to long exposure of the specimens on the ground surface. In
fact, some teeth underwent the action of the so-called “feu de brousse” (among the
locals, to set fire to the grass is a common way to renovate their cultivation land). In
summary, the sedimentologic parameters indicate a fluvial-lacustrine, non-marine
depositional environment.
Our analysis is consistent with those reported by Besairie (1972) on the sections
he made along the rim of the Mahajanga basin; in particular, there are affinities
with the section called “Coupe-Route de Majunga”, which is referred as continen-
tal Bathonian with mixed facies, and named “Isalo IIIb-Bathonien-Facies mixte
dinosauriens”. Some discrepancy of data occurs in the granulometry of the sedi-
ment (0.15 mm in Besairie, 1972) and in the garnet-zircon ratio, with the latter
scarcer than the former, according to the same author. However, the high variabil-
ity of that facies must be taken into account because of its genesis, which is testi-
fied also by local heteropy. In the end, it is worth mention that the sample here ana-
lyzed comes from the sediment that was in direct contact with the bone. As already
noted by Lavocat (1955), there is evident difference in composition between the
homogeneous, more sandy and softer beds, and the silt encasing the fossils. The
encrusting nature of the silt raises the hypothesis of a pseudo-nodulisation of the
fossils, linked to a peculiar diagenesis of the carcasses, in a peculiar environment
of decomposition of the organic remains.
Material and methods
The material here described is housed in the Collection of Fossil Vertebrates of
the Museo Civico di Storia Naturale di Milano (MSNM V). The teeth, found isolat-
ed in loose sediments, are pictured in digital photographs (Nikon Coolpix 995, 3.34
Megapixel) which were taken on them as they were collected in the field. As men-
tioned above, some preparation was done on the maxilla, in order to bring to light
the teeth and their denticles, which in the end revealed the presence of a third erupt-
ing tooth. Measurements were taken with a digital calliper; cross-sections of the
teeth were obtained by casting them with silicon RTV rubber and cutting of their
epoxy-resin replicas. The cross-section drawings were made under camera lucida.
In describing the teeth we utilized the following terms: TCH = tooth crown height;
FABL = fore-aft basal length; BW = basal width; BCR (FABL/BW) = basal com-
pression ratio; ER (FABL/TCH) = elongation ratio (all these terms are from Currie
et al. [1990], and Farlow et al. [1991]); DSDI = denticle size difference index
(Rauhut & Werner, 1995). Denticle count, where possible, was taken at mid-crown
height because, as explained in the description, the size of the denticles decreases at
the apical and basal ends of both carinae. The systematic terms are taken mainly
from Brochu (2001). Additional systematic terms concerning the relationships
among theropods, basal Crurotarsi and mesoeucrocodylians are taken respectively
from Weishampel et al. (2004), Parrish (1993) and Sereno et al. (2001).
Institutional abbreviations
MNHN, Muséum national d’Histoire naturelle, Paris (France); MOR, Museum
of the Rockies, Bozeman (Montana, USA); MSNM, Museo Civico di Storia
Naturale, Milano (Italy); TMP, Royal Tyrrell Museum of Palaeontology,
Drumheller (Alberta, Canada).
Anatomical abbreviations
antfe/al/aj, antorbital fenestra/articular surface for lacrimal/articular surface
for jugal; apal/aect, articular surface for palatine/articular surface for ectoptery-
goid; ic/sof, internal choana/suborbital fenestra; a2,2
nd alveolus; a4, 4th alveolus;
antfe, antorbital fenestra; df, dental foramina; ect, ectopterygoid; ic, internal
choana; ip, interdental plates; j, jugal; l, lacrimal; m, maxilla; mfe, maxillary fen-
estra; mms, maxillary medial shelf; n, nasal; pal, palatine; pt, pterygoid; rt,
replacement teeth; s, sulcus; sof, suborbital fenestra; v, vomer.
Systematic Paleontology
Archosauromorpha Huene, 1946
Archosauriformes Gauthier, Kluge & Rowe, 1988
Archosauria Cope, 1869 sensu Gauthier, 1986
Razanandrongobe sakalavae, gen. et sp. nov.
Holotype: MSNM V5770 Fragmentary right maxilla bearing three unerupted
teeth.
Referred material: MSNM V5771-5777, isolated teeth.
Age and stratigraphic horizon: Middle Jurassic (Bathonian), 167.7-164.7
MA, Mahajanga Basin, Isalo IIIb subunit, ‘Facies Mixte Dinosauriens’.
24 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Locality: maxilla: undetermined Hills W of Ambondromamy – teeth: hills N-
NW of Andranomamy (Mahajanga Province), NW Madagascar.
Etymology: Razanandrongobe, Malagasy composite name for ‘ancestor’
(Razana-) of the ‘large’ (-be) ‘lizard’ (-androngo-); sakalavae, latin for ‘of
Sakalava’, the ethnic group that inhabits the Mahajanga region.
Diagnosis: large predatory archosaur with deep, robust maxilla bearing a
prominent maxillary medial shelf; thecodont tooth implantation; alveolar channels
nearly straight in the sagittal plane; alveoli (as preserved) sub-rectangular in ven-
tral view; fused interdental plates with a surface texture consisting of marked
ridges and furrows extending for a short distance also above the dorsal margin of
the interdental plates; heterodont dentition; incisiform rostral teeth, U-shaped in
cross-section; stout lateral teeth, sub-oval in cross-section; smallest lateral teeth,
globe-shaped; denticles present on both carinae in all the teeth, and very large (0.8-
1.4 per mm) (unambiguous autapomorphy of the taxon).
Description
Maxilla (specimen MSNM V5770)
The specimen consists of a maxillary fragment (Figs. 4-6), 192 mm long, 124 mm
high, and 96 mm wide (as preserved) that preserves many features such as the inter-
dental plates and a robust maxillary medial shelf. It also exposes on its lateral side
five alveoli and three in situ unerupted teeth. On the basis of the curvature of the
crowns and the general morphology of the bone, it should pertain to the right side of
the skull. Its position in the skull and the interpretation of some features (such as the
attachment areas for the palatine/ectopterygoid and for the jugal/lacrimal or the pres-
ence/absence of the antorbital fenestra) remain unclear, as these features vary accord-
ing to which group the specimen is compared with. The external bony surface under-
went little weathering, and is better preserved ventro-medially.
Lateral view (Fig. 4A). The preserved dorsal margin of the bone is nearly horizon-
tal to the level of the third alveolus and then gradually slopes ventrally and forms a
concave area (see dorsal view). The original ventral margin of the maxilla is pre-
served from the alveolus 2 to the caudal end of the bone, and seems to show that the
maxilla had the same convexities in the lateral outline visible in the maxillary mid-
caudal portion of several taxa. The lateral wall of the bone is totally eroded, leaving
exposed the medial surface of five alveoli and three substitution teeth. According to
the definitions given by Motani (1997) and Zaher & Rieppel (1999) the tooth implan-
tation here is clearly thecodont. The dorsal end of the alveolar channels is preserved,
so that we can infer that the missing lateral wall of the maxilla did not terminate far
dorsally to that margin, where it would have formed the ventral margin of the antor-
bital fenestra or, alternatively, met the nasal. Judging from both the ER of the pre-
served tooth crowns (Tab. I) and the rostro-caudal diameter of the alveoli, we can also
infer that the length of each alveolar channel was major than its corresponding TCH.
Substitution teeth are housed in alveoli 3-5, illuminating the process of tooth replace-
ment: replacement tooth 5 represents an early stage and was emerging in a latero-ven-
tral direction (at about 45°) from the lateral wall of its alveolar channel; replacement
25
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
26 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Fig. 4 - Specimen MSNM V5770 in lateral (A) and dorsal (B) views. Scale bar = 5 cm. For abbrevia-
tions see text.
Fig. 4 - Esemplare MSNM V5770 nelle viste laterale (A) e dorsale (D). Scala metrica = 5 cm. Per le
abbreviazioni consultare il testo.
27
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 5 - Specimen MSNM V5770 in medial (A) and ventral (B) views. Scale bar = 5 cm. For abbrevi-
ations see text.
Fig. 5 - Esemplare MSNM V5770 nelle viste mediale (A) e ventrale (B). Scala metrica = 5 cm. Per le
abbreviazioni consultare il testo.
28 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
tooth 3, after having rotated to a subvertical position, was growing medial to the root
of the functional tooth that occupied the alveolar channel; finally, replacement tooth
4 was growing in its alveolar channel where the resorption of the root of the function-
al tooth was almost complete. The tooth denticles are exceptionally large (1-1.25 per
mm) and wrap around the tip of the tooth. Further characteristics of the teeth will be
discussed below, in the dentition section.
Medial view (Fig. 5A).The interdental plates are fused to each other and to the
ventromedial surface of the maxilla, and bear fine vertical wrinkles and marked
ridges and furrows, slightly caudally inclined, that give the medial maxillary wall a
rugose texture. The interdental plates taper caudally, following the curvature of the
alveolar ventral margin of the maxilla. A rostro-caudally oriented row of dental
foramina, that presumably served for the passage of the dental arteries, delimits the
interdental plates dorsally. Within the rugose area, smaller foramina and pits are
present, mostly above that row. Just dorsal to the interdental plates, the maxilla
gradually becomes more inclined medially to form a stout, prominent maxillary
medial shelf, which lies in a sub-horizontal plane at the level of its broken medial
margin. The surface texture in this area is very much like that of the interdental
plates for a height of about 15-20 mm above the row of the dental foramina, after
which the maxilla is marked by an abrupt change of texture and becomes smooth.
A single broad depression possibly coinciding with the positions of dentary teeth at
closed jaws can be observed at the level of the second preserved alveolus. The bro-
ken medial margin of the maxillary medial shelf appears slightly sigmoid in outline,
and is dorso-ventrally thicker in the mid portion of the preserved maxilla, as if it
would have formed a strong median bar in that point. Adjacent to this thickened
margin the ventral side of the maxillary medial shelf is stouter than in the adjacent
areas, and pierced by at least 3 foramina. Rostrally, the broken medial margin of the
shelf tapers to halving its dorsoventral depth. Caudally, the shelf abruptly terminates
rostrally to the rostral portion of the attachment area for the palatine or ectoptery-
goid. This attachment area is preserved as a well-delimited, sub-triangular depres-
sion, more marked dorsally, that lie in a vertical plane at the caudal end of the max-
illa. Dorsal to the maxillary medial shelf, the maxilla is slightly inclined laterally
and faintly concave, up to its lateral-most preserved dorsal margin.
Dorsal view (Fig. 4B). The most striking feature in dorsal aspect is a broad area
in the caudo-lateral corner that could represent the rostro-ventral margin of the
antorbital fenestra, or more simply, a concave depression on the medial side of the
missing lateral wall of the bone. Another possibility is that this area represents the
rostro-medial portion of the attachment area for the jugal/lacrimal. This uncertain-
ty is due to the bad preservation of the bony surface in this point. This renders it
difficult to determine if its texture indicates the attachment area of a bone or an
external surface. A sulcus, presumably for the passage of a large artery (see
Colbert, 1946), runs rostro-caudally 10 to 20 mm from the lateral broken edge of
the maxilla for its entire length, dividing the alveolar portion of the bone from the
maxillary medial shelf.
Ventral view (Fig. 5B). Judging from the preserved portions of the alveoli, they
were probably subrectangular in transverse section, with their lateral alveolar mar-
gin extended farther ventrally than the medial one. Toward the caudal end of the
bone the tooth row gently arches in a lateral direction, indicating an increase in
skull width. The palatal shelf, as preserved, is rather wide but its total medial exten-
sion cannot be reconstructed with accuracy. The maxilla tapers caudally forming a
concave medial margin where the bone would have bordered rostrally the internal
choana or the suborbital fenestra. Along this concave margin the undeformed bony
wall is unbroken, continuous and naturally curved.
29
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 6 - Specimen MSNM V5770 in rostral (A) and caudal (B) views. Scale bar = 5 cm. For abbrevi-
ations see text.
Fig. 6 - Esemplare MSNM V5770 nelle viste rostrale (A) e caudale (B). Scala metrica = 5 cm. Per le
abbreviazioni consultare il testo.
Rostral view (Fig. 6A). In rostral view the lateral portion of MSNM V5770
appears gently convex. The alveoli match this convexity, being slightly curved
medially, but their orientation can be considered definitely sub-vertical. The bro-
ken section of the maxillary medial shelf appears almost triangular, with the con-
cave dorsal and ventral margins converging medially. The transition from the ver-
tical portion of the maxilla bearing the interdental plates to the sub-horizontal
medial-most portion of the maxillary medial shelf, which gives the medial wall a
concave profile, can be seen in this view.
Caudal view (Fig. 6B). In caudal view, the wide area which is variably inter-
pretable (see dorsal view) is visible on the dorso-lateral corner of the specimen.
This area faces dorso-caudally, at an angle of about 25° from the horizontal plane.
On the medial side the stouter portion of the maxillary medial shelf can be seen,
and in this view its ventral wall is inclined at 45° degree and nearly straight. On the
contrary, in rostral view it is regularly concave, because the shelf is less robust and
the palate vault deeper. The attachment area for the palatine/ectopterygoid, present
on the medio-ventral side of the bone, preserves a rim thicker dorsally. At the same
level on the lateral side the ventro-laterally directed tip of the replacement tooth in
the caudalmost alveolus is visible.
Dentition (specimens MSNM V5770 – V5777)
All the tooth crowns here described (Tab. I; Fig. 7) are stout, pointed, recurved,
bent lingually, and bear very large denticles on both the carinae. Since no basal-
most portion of the crown is intact, it is difficult to understand if there was a basal
constriction between the crown and the root. All the teeth can be considered true
ziphodont sensu Prasad & de Lapparent de Broin (2002), in having clearly individ-
ualized denticles on the carinae, which are not the result of prolongation of the
enamel ridges of the crown. On the basis of the shape of the crown, we recognized
three morphotypes that in all likelihood come from one taxon. The variation of the
teeth is consistent with the range of variation documented for other archosaurs in
which teeth vary in size, shape, and carinal orientation depending on where they
occur in the jaw. It is not known whether the teeth come from the upper or the lower
jaw but we can infer their position in the jaws. The first morphotype, including
MSNM V5775, represents a rostral tooth. Specimens MSNM V5771-5774, 5776,
and all the substitution teeth implanted into the maxilla are lateral teeth and pertain
to the second morphotype. The third morphotype is represented by MSNM V5777,
a small lateral tooth with a very short crown that does not appear as a scaled ver-
sion of the larger lateral teeth, being rather globular. It was probably one of the cau-
dal-most teeth. The lacking of the rostral portions of the jaws and the paucity of
our sample do not allow us to understand if the smaller size of V5775 respect to
most of the lateral teeth signifies that the rostral teeth were smaller than the later-
al ones or if V5775 belongs to a small individual. In V5773 and V5777 only the
very apical portion of the crown shows faint traces of wearing. In all other speci-
mens the tip is markedly consumed and the wear surfaces extend on both the lin-
gual and labial side, where the dentine is broadly exposed and preserve microscop-
ic wear striations. Some of these striations are oriented vertically or slightly
inclined caudally, but the majority of them is oriented heterogeneously. The later-
30 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
31
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 7 - Specimens MSNM V5771-5777 in lingual (A) and distal (B) views. Scale bar = 10 cm.
Fig. 7 - Esemplari MSNM V5771-5777 nelle viste linguale (A) e distale (B). Scala metrica = 10 cm.
al teeth of R. sakalavae show their greatest wear on the lingual side (Fig. 8H), on
which a U-shaped spalled surface occupies most of the apical third of the crown,
whereas a long and narrower spalled surface, contacting the former apically, flat-
tened the apical half of the twisted mesial carina. The lingual spalled surfaces cross
over the tip of the crown and extend onto the labial side. The surficial enamel in
32 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Tab. I - Razanandrongobe sakalavae n. gen. n. sp.: basic numbers and measure-
ments of the teeth. With the exception of the specimen MSNM V5773 the tips of
the crown are consumed; as a consequence, the TCHs and ERs here reported are
respectively slightly lower and higher than they should be.
Tab. I - Razanandrongobe sakalavae n. gen. n. sp.: misure principali dei denti.
Poiché, ad eccezione dell’esemplare MSNM V5773, le punte delle corone sono
consumate, i valori relativi a TCH ed ER riportati sono rispettivamente leggermen-
te più bassi e più alti che in origine.
------4-
46.2 -----5-
-----441.00
45.7 26.7 23.1 1.16 0.58 4 4 1.00
33.5 23.1 18.6 1.24 0.69 6 5.5 1.09
35.5 20.6 15.1 1.36 0.58 5 4.5 1.11
30.3 18.2 14.2 1.28 0.60 5 5 1.00
23.4 12.8 15.4 0.83 0.55 5 5 1.00
20.7 14.5 12.2 1.19 0.70 - 6.5 -
11.8 12.8 11.1 1.15 1.09 7 6.5 1.08
Specimen
Tooth crown height (TCH) in mm
Fore-aft basal length (FABL) in mm
Tooth basal width (BW) in mm
Basal compression ratio (FABL/BW)
Elongation ratio (FABL/TCH)
Number of serrations per 5 mm of
mesial carina
Number of serrations per 5 mm of
distal carina
Denticle size difference index (DSDI)
MSNM
V5770*
MSNM
V5770**
MSNM
V5770***
MSNM
V5771
MSNM
V5772
MSNM
V5773
MSNM
V5774
MSNM
V5775
MSNM
V5776
MSNM
V5777
*replacement tooth in alveolus 3.
** replacement tooth in alveolus 4.
*** replacement tooth in alveolus 5.
the rostral tooth is mostly worn away on the labial side (Fig. 8I), with a U-shaped
spalled surface, whereas it is still preserved lingually. In all the teeth the carinae
grade into the tooth body. In the lateral teeth, both carinae are close to the midline
of the crown near the tip, but not far from it they twist, with the distal carina mov-
ing slightly labially and the mesial carina moving decidedly lingually (Fig. 8G).
The marked twist of the mesial carina is visible also in the erupting tooth occupy-
ing the fifth alveolus of MSNM V5770. Due to the twisting of the carinae, in cross-
section at mid-height (Fig. 8A, B) the crowns of these lateral teeth appear strong-
ly asymmetrical, with a gently arched lingual side and a strongly curved labial side
that looks like the curved portion of a D. The rostral tooth V5775 differs from the
others in having both carinae on the lingual side, and the labiolingual width of the
tooth greater than the distance between the two carinae. As a consequence, its
strongly curved labial side appears U-shaped in cross-section at mid-height the
crown (Fig. 8C). A similar pattern of carinal variation occurs in the teeth of sever-
al not strictly related theropod dinosaurs (e.g., tyrannosauroids and abelisaurids) in
which the carinae are strictly in a parasagittal plane in the rearmost teeth, but with
a more mediolateral orientation towards the premaxilla or front of the dentary. The
cross-sections at the base of all the teeth are oval to sub-circular (Fig. 8D-F).
Again, the specimen V5775 differs from the others in having the axes inverted, thus
resulting more expanded labio-lingually than meso-distally (Fig. 8F). The distal
carina reaches the base of the crown in all the isolated teeth, whereas the mesial
carina terminates at two thirds of the crown height. In all the specimens, including
the rostral tooth V5775, both carinae bear denticles for all their length. The denti-
cles are remarkably large relative to the TCH and compared with other archosauri-
form denticles (Farlow et al., 1991; Benton, 1986; Sues et al., 1996; Clark et al.,
2000; Gow, 2000; Campos et al., 2001; Prasad & de Lapparent de Broin, 2002;
Senter, 2003; Carvahlo et al., 2004). Measured at mid-crown, there are 0.8-1.4 den-
ticles per mm (Fig. 8K-N). However they vary in size along both carinae: the api-
calmost denticles are slightly smaller than those immediately below, they maintain
the same size up to the mid of the crown, and then gradually decrease in size
toward the basal end of the carina, where they are as large as half of the denticles
at mid height of the crown. In all the specimens the mesial denticles are quite erod-
ed. However they seem as tall as the distal denticles or, in some cases, only slight-
ly shorter. The denticles are relatively simple in form, almost as high as long, with
slightly rounded tips. As one can see in mesial and distal views (Fig. 8K-N), in the
denticles of R. sakalavae the basal width is greater than the length (between 1.6
and 1.9 times). Where preserved, the denticles of the mesial carina wrap around the
tip of the tooth onto the distal carina. However this feature has limited taxonomic
utility, being common in both theropod dinosaurs (Currie & Carpenter, 2000) and
ziphodont crocodyliforms (Prasad & de Lapparent de Broin, 2002). The enamel
forms narrow junctions between the denticles at their top, giving more continuity
to the cutting edges. These junctions, together with the denticle shape, render the
slits between the denticles flask-shaped in lateral view (Fig. 8K), and hourglass-
shaped looking at the crown in mesial and distal views (Fig. 8N). These rounded
spaces are analogous to the high pressure slots visible on the exterior surface,
between the denticles in several taxa such as the basal synapsid Dimetrodon, the
33
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
34 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
parasuchians, and the tyrannosaurids (Abler, 2001). A fracture along the medial
sagittal plane of a portion of the distal carina in the specimen MSNM V5774
exposes the interior surface of some denticles (Fig. 8J). Although the interior sur-
face appears partly damaged, the rounded spaces are preserved on it, resembling
the condition of tyrannosaurids and differing from both Dimetrodon and parasuchi-
ans (Abler, 2001), which lack them. Looking at the edge of both carinae the denti-
cles appear slightly inclined toward the tip of the crown. In all the teeth of R.
sakalavae, blood grooves extend from the base of the spaces (i.e. between the bases
of adjacent denticles) onto the surface of the crown. They are macroscopically vis-
ible and, following the denticle orientation, are slightly inclined toward the base of
the crown. The enamel, where preserved, is smooth and has the same texture sur-
face in all the specimens, except for V5775 which bears vertical striations on the
basal third of the lingual surface. A dense pitting, due to surface micro-damaging
by sediment granulation, notched the enamel surface.
Taxonomic affinities
Due to the fragmentary nature of our material, Razanandrongobe sakalavae
lacks almost all the characters used in archosauromorph phylogenetic analyses
(Sereno, 1991; Parrish, 1993; Juul, 1994; Benton & Walker, 2002). In spite of that,
the general morphology of both maxilla and dentition allowed us to make compar-
ison with several archosauromorph taxa and identify some features that could help
to establish its affinities. In order to better understand the taxonomic significance
and the evolutionary history of these features we compared our material not only
to the derived archosaur groups with which it shows potential affinities, but also to
non-archosaurian archosauromorphs (Fig. 10).
Among archosauromorphs, the presence of tooth denticles has been reported in
several groups, including parasuchians and rauisuchians (Abler, 1997),
Ornithosuchus (Farlow et al., 1991), euparkerids (Wu & Russell, 2001; Senter,
2003), dinosaurs (Weishampel et al., 2004), “sphenosuchians” (Wu & Chatterjee,
1993; Clark et al., 2000), “protosuchians” (Farlow et al., 1991; Sues et al., 1996;
Gow, 2000) and ziphodont crocodyliforms (Colbert, 1946; Langston, 1975; Willis
& Mackness, 1996; Prasad et al., 1999; Campos et al., 2001; Carvalho et al., 2004).
On the contrary, they are absent in the teeth of basal archosauromorphs, such as
Protorosauria (Nosotti, pers. comm. 2005) and Rhynchosauria (Carroll, 1988), and
basal diapsids (Carroll, 1988). Therefore this character could represent a synapo-
morphy of Archosauriformes, including R. sakalavae. Unfortunately, with few
exceptions (e.g. Euparkeria [Senter, 2003]) only the presence/absence of denticles
has been reported in close outgroups to Archosauria. This lack of published
descriptions of other denticle-related characters makes it difficult to test if denticles
contain informative taxonomic characters at high systematic levels among
archosauromorphs (for example, the blood grooves at the bases of the denticles are
visible in both Euparkeria [Senter, 2003] and some theropods [Currie et al., 1990;
Maganuco et al., 2005] but in literature we were not able to find information about
their presence/absence in other taxa including ziphodont crocodylomorphs).
According to Juul (1994) it is not yet clear even at which hierarchic level the occur-
35
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 8 – Some features of the teeth of Razanandrongobe sakalavae n. gen. n. sp.: specimens MSNM
V5772 (A, D), V5773 (B, E), and V5775 (C, F) in cross-sections at mid-height (A-C) and at the base
of the crown (D-F); specimen MSNM V5773 in apical view (G), showing the twisting of the carinae;
tooth wearing visible as spalled surfaces on the lingual side of the specimen MSNM V5772 (H) and
labial side of the specimen MSNM V5775 (I); interior of the denticles (J) visible thanks to a fracture
along the medial sagittal plane of the distal carina in the specimen MSNM V5774; close-ups of the
denticles of the specimen MSNM V5771 seen in lingual (K) and distal (L) views; close-ups of the den-
ticles of the specimen MSNM V5773 seen in lingual (M) and distal (N) views. In A-G the directions
of the axes are abbreviated as follows: di, distal; la, labial; li, lingual; me, mesial. Scale bars are in mm.
Fig. 8 – Alcune caratteristiche dei denti di Razanandrongobe sakalavae n. gen. n. sp.: esemplari
MSNM V5772 (A, D), V5773 (B, E) e V5775 (C, F) in sezione trasversa a metà altezza (A-C) e alla
base della corona (D-F); vista apicale dell’esemplare MSNM V5773 (G), che mostra la torsione delle
carene; grado di usura dei denti, evidenziato da piani di levigatura sul lato linguale dell’esemplare
MSNM V5772 (H) e su quello labiale dell’esemplare MSNM V5775 (I); superficie interna dei denti-
coli (J) visibile grazie ad una frattura lungo il piano sagittale mediale della carena distale nell’esem-
plare MSNM V5774; particolare dei denticoli dell’esemplare MSNM V5771 in vista linguale (K) e
distale (L); particolare dei denticoli dell’esemplare MSNM V5773 in vista linguale (M) e distale (N).
In A-G le direzioni degli assi sono abbreviate nel seguente modo: di, distale; la, labiale; li, linguale;
me, mesiale. Le scale metriche sono in mm.
AB
C
me
di
la li
DEF
rence of thecodont dentition may be considered a diagnostic feature, relative length
of root to crown has not yet been quantified for many archosauromorphs, and the
tooth implantation of some basal forms has been described in a confusing manner.
In any case such a kind of tooth implantation clearly seems to be diagnostic for
more inclusive taxa than Archosauria (Sereno, 1991), being shared without doubt
by most of the Archosauriformes. Again, as pointed out by Juul (1994), on the basis
of the published material available to us it was not possible to establish the exact
distribution and condition of the interdental plates within Archosauromorpha.
Nevertheless, they are absent in rhynchosaurs and Prolacerta (Juul, 1994); present
but not fused in: Euparkeria (Senter, 2003), the rauisuchids (Juul, 1994), the
ornithosuchid Ornithosuchus (Juul, 1994), the sauropodomorph Massospondylus
(Juul, 1994), many theropods (Holtz, 2000), and the “protosuchians” (Sues et al.,
1996); and present and fused to each other in some theropod taxa (Holtz, 2000;
Senter, 2003), MSNM V5770, the poposaurid Teratosaurus (Galton, 1985),
mesoeucrocodylians (pers. obs. 2004-2005), and, at least judging from photos, also
in other Crocodyliformes. So the condition “interdentental plates present but
unfused” could represent at the same time a synapomorphy of Archosauriformes,
and the plesiomorphic condition for Theropoda (in which they became fused inde-
pendently in different groups), poposaurids, and Crocodyliformes (among whom
fused interdental plates developed once). So, the dentition and the tooth implanta-
tion suggest that R. sakalavae belongs to Archosauriformes, and its condition of the
interdental plates places it within Archosauria. Looking at the general morphology
of both maxilla and dentition, R. sakalavae does not show any significant resem-
blance either with basal Archosauromorpha such as the Protorosauria or the highly
specialized Rhynchosauria (Carroll, 1988), or basal Archosauriformes such as the
Proterosuchidae (Carroll, 1988), Erythrosuchidae (Gower & Sennikov, 2000),
Proterochampsidae (Nash, 1975), and Euparkeridae (Wu & Russell, 2001; Senter,
2003), basal ornithodirans such as Scleromochlus (Benton, 1999), and the crurotar-
sian ornithosuchids (Carroll, 1988; Sereno, 1991) and Erpetosuchus (Benton &
Walker, 2002). All these forms can be easily dismissed, because they have homod-
ont dentition, TCH major than maxillary depth, maxilla relatively low in profile,
and do not have maxillary medial shelves and fused interdental plates (but for the
distribution of the interdental plates see above). The heavily built skull and denti-
tion of R. sakalavae do not show any resemblance with Pterosauria, so the latter can
be dismissed too without taking care of its controversial affinities (Brochu, 2001).
Also the herbivorous aetosaurs and the long snouted parasuchians (Carroll, 1988;
Sereno, 1991), belonging to the Crurotarsi, can be excluded, because of their high-
ly specialized skulls. The basalmost crurotarsian showing some affinities with R.
sakalavae are the Prestosuchidae (Parrish, 1993). In lateral view, the Porto Alegre
specimen of Prestosuchus figured by Parrish (1993: 4C) superficially resembles R.
sakalavae in maxillary depth, ratio between the TCH and depth of the maxilla,
shape of the caudal portion of the maxilla and sutural surface between maxilla and
jugal. In addition, also the prestosuchids attain a medium to large size, and present
a certain degree of heterodonty (Parrish, 1993), with caniniform rostral teeth, and
compressed and recurved caudal teeth. Moreover, in the prestosuchid Saurosuchus
galilei (Alcober, 2000) the rostralmost medial portions of the maxillae form two
36 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
37
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
medial shelves meeting medially and forming a short secondary palate. Also the
poposaurid Teratosaurus superficially resembles R. sakalavae in lateral view, but
following Parrish (1993) we can conclude that the resemblance in skull profile
among the large headed carnivorous taxa such as Prestosuchidae, Rauisuchidae,
Poposauridae, some theropods and Erythrosuchidae is probably the result of their
adaptation to the same ecological role as medium-sized to large terrestrial carni-
vores. Summing up, thanks to its features Razanandrongobe sakalavae can be rea-
sonably placed within the Archosauria but not among the basal members of this
taxon. Therefore it must be compared with more derived predatory archosaurs, i.e.
theropods and crocodylomorphs. This is consistent with the fossil record of
Archosauria - Crocodylomorpha and Theropoda are the only carnivorous
archosaurian lineages known to have survived the Triassic (Sereno, 1991; Parrish,
1993; Brochu, 2001). The possible affinities of R. sakalavae with theropods and
crocodylomorphs are discussed in the following sections.
Theropod affinities
In comparing Razanandrongobe sakalavae with Theropoda (Fig. 9A, D), the
attachment area located on the medial side of the maxilla must be considered the
attachment for the palatine; the recess just caudal to the maxillary medial shelf must
be considered the rostro-lateral corner of the internal choana; and the concave area
on the lateral side of the specimen could alternatively represent the attachment for
the jugal, the rostral portion of the antorbital fenestra, or a depression on the medial
side of the missing lateral wall of the bone. Turning the identification around, among
theropods (Rauhut, 2003; Weishampel et al., 2004) we can easily dismiss the follow-
ing groups (or at least their currently known members) because they do not show any
of the hallmarks of R. sakalavae: Coelophysoidea, Ceratosauridae, Noasauridae,
Allosauroidea (including Allosauridae, Sinraptoridae, and Carcharodontosauridae),
Compsognathidae, Ornithomimosauria, Oviraptorosauria, Therizinosauridae,
Alvarezsauridae, Troodontidae, Dromaeosauridae, and Avialae. It is worth mention-
ing, however, that among predatory dinosaurs only the small-bodied troodontids have
teeth with denticles as large as - or even larger than - R. sakalavae relative to TCH
(Currie et al., 1990; Farlow et al., 1991). Eventually, we did not find any resemblance
in size and shape between the teeth of R. sakalavae and the isolated teeth from the
same Middle Jurassic strata recently referred to Theropoda incertae sedis (Maganuco
et al., 2005). Razanandrongobe sakalavae shows some hallmarks of the Cretaceous
Abelisauridae (Sereno et al., 2004), such as subrectangular alveoli, fused interdental
plates heavily textured and rugose, relatively low-crowned teeth (except for
Carnotaurus, which has elongate and slender crowns), and, possibly, broad maxil-
lary-jugal contact. However, the corresponding portion of the maxilla in the
abelisaurids (Sampson, 1998), roughly comprised between the alveoli 7-11 in a
whole series of about 15-16, has not a well-developed maxillary medial shelf, and
bears a row of dental foramina that are larger, and more regular in shape than those
of R. sakalavae, and are placed in a tidy row just above the interdental plates. The
teeth of R. sakalavae do not show the typical features of the abelisaurid teeth, such
as: lateral crowns drop-shaped in cross-section (Candeiro et al., 2004) and decided-
ly more flattened labio-lingually than those of R. sakalavae; rostral teeth D-shaped,
38 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Fig. 9 – Lateral and palatal views of the skulls of: A and D, the theropod Tyrannosaurus (redrawn from
Carr [1999]); B and E, the “sphenosuchian” Dibothrosuchus (redrawn from Wu & Chatterjee [1993]);
and C and F, a generalized peirosaurid based on both Uberabasuchus (Carvalho et al., 2004) and
Lomasuchus (Gasparini et al., 1991). Scale bar equals 25 cm in A and D, and 1 cm in B, C, E, and F.
The coloured rectangle approximately indicates the portion of the skull corresponding to the specimen
MSNM V5770. For abbreviations see text.
A
B
C
39
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Fig. 9 – Viste laterale e palatale dei seguenti crani: A e D, il teropode Tyrannosaurus (ridisegnato a par-
tire da Carr [1999]); B ed E, lo “sfenosuco” Dibothrosuchus (ridisegnato a partire da Wu & Chatterjee
[1993]); C ed F, un peirosauride generalizzato basato sui due generi Uberabasuchus (Carvalho et al.,
2004) e Lomasuchus (Gasparini et al., 1991). La scala metrica è di 25 cm in A e D, e di 1 cm in B, C,
E ed F. Il rettangolo colorato indica approssimativamente la porzione del cranio corrispondente all’e-
semplare MSNM V5770. Per le abbreviazioni consultare il testo.
D
E
F
i.e. with a labio-lingual axis considerably shorter than that of R. sakalavae; blood
grooves markedly inclined toward the base of the crown (Kellner & Campos, 2002;
pers. obs. 2004); in the abelisaurids the areas of the crown adjacent to the carinae are
flat or concave whereas the denticles of R. sakalavae grade into the tooth body; the
abelisaurid denticles are pointed and decidedly smaller than those of R. sakalavae
(pers. obs. on abelisaurid teeth housed in MSNM; Kellner & Campos, 2002). The
rostral tooth of Razanandrongobe sakalavae resembles the premaxillary teeth of
Tyrannosauroidea in being incisiform and U-shaped in cross-section. The sides of the
rostral tooth of R. sakalavae are not flattened as in Tyrannosauridae, looking more
like the premaxillary teeth of Stokesosaurus (Holtz, pers. comm. 2004), Eotyrannus
(Hutt et al., 2001), or the Jurassic teeth described by Zinke (1998) and referred to the
basal tyrannosauroid Aviatyrannis by Rauhut (2003), but not the tooth referred to
Aublysodon (Lehman & Carpenter, 1990: fig. 4) that lacks denticles on the carinae.
On the other hand, contrary to the basal Tyrannosauroidea (Holtz, 2001b; Hutt et al.,
2001) R. sakalavae possesses lateral teeth incrassated as in – or even more than - the
Cretaceous Tyrannosauridae (cross-section greater than 60% wide mediolaterally as
long rostrocaudally is a tyrannosaurid synapomorphy [Holtz, 2001a]). Despite of the
relatively great size of the denticles in Tyrannosauroidea (Farlow et al., 1991; pers.
obs. 2004 on MOR and TMP material), the denticles of R. sakalavae (0.8-1.4 per
mm) are larger than those borne in tyrannosaurid teeth with comparable TCH. In
most of our specimens, the denticles are even larger than those preserved in the
largest teeth of Tyrannosaurus (1.2 per mm [Farlow et al., 1991]). The presence of
round, voided strengthening structures between the denticles have not yet been stud-
ied in a sufficient number of taxa to test their systematic significance. In R.
sakalavae the denticles are simpler in form and do not possess the thin groove that
in tyrannosaurid teeth runs along the apices of the denticles (Abler, 1997). Lastly, the
blood grooves in R. sakalavae resemble those of the tyrannosaurids in having a sim-
ilar degree of orientation toward the base of the crown (Currie et al., 1990), but dif-
fer from those of basal tyrannosauroids, in which they are perpendicular to the edge
of the crown (Zinke, 1998). If related to the Tyrannosauroidea, MSNM V5770 would
represent the caudal end of a tyrannosaurid right maxilla, being in contact with the
jugal dorso-caudally, with the palatine ventrally, being located just below the antor-
bital fenestra, and bordering rostro-laterally the internal choana. The alveoli of R.
sakalavae, as preserved, appear slightly less rounded than in the tyrannosauroids
(pers. obs. 2004 on MNHN, MOR and TMP tyrannosaurid material; Hutt et al.,
2001). Although in tyrannosaurids the maxillary medial shelf is well-developed and
placed well dorsal to the level of the tooth row, the maxillary medial shelf of MSNM
V5770, at scale, is too massive and too extended medially (Brochu, 2003; Carr, pers.
comm. 2004; pers. obs. 2004). Moreover, the angle formed by the maxillary medial
shelf with the main body of the maxilla in Tyrannosauridae ranges from 125° to 140°
(pers. obs. 2004 on MNHN, MOR, MSNM and TMP tyrannosaurid material), where-
as in R. sakalavae it is 110° to 120° (Fig. 6A, B). Finally, the interdental plates are
separated in Tyrannosauridae (Currie, 2003; Hurum & Sabath, 2003), although they
are not clearly differentiated from the medial surface of the maxilla in some basal
tyrannosauroids like Eotyrannus (Hutt et al., 2001). The only other large-sized
theropods with a developed maxillary medial shelf and labiolingually expanded teeth
40 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
are the spinosaurids. Unfortunately, basal spinosaurids from the Jurassic are not yet
known, and the highly derived snouts of the known spinosaurids (Sereno et al., 1998;
Dal Sasso et al., 2005), which all come from Cretaceous sediments, differ from R.
sakalavae in having: palatal shelves that meet forming an acute angle (about 45°);
highly modified palatal shelves in which the not well-defined interdental plates are
placed lateral to the paradental lamina that covers the alveoli secondarily medially;
subcircular alveoli; homodonty in tooth shape; high crowns; and carinae unserrated
or with very fine serration. Very large denticles (even 1.2 per mm) are present in
some teeth (Farlow et al., 1991; pers. obs. 2005) referred to the Torvosauridae, a
taxon known from the Middle Jurassic and usually considered to be the sister group
of the Spinosauridae (Sereno et al., 1998). However, the teeth of the torvosaurids are
labiolingually compressed, and their maxilla (Allain, 2002; pers. obs. 2004 on
MNHN material) is low and lacks a well-developed maxillary medial shelf.
Crocodylomorph affinities
Among the crocodylomorphs, “protosuchians” (Nash, 1975; Sues et al., 1996;
Gow, 2000), thalattosuchians (pers. obs. 2005), dyrosaurids (pers. obs. 2005), noto-
suchids (Buckley et al., 2000; Gomani, 1997), Libycosuchus (Stromer, 1914),
Sphagesaurus (Pol, 2003), Araripesuchus (Ortega & Gasparini, 2000), and
Sarcosuchus (Sereno et al., 2001) do not show any significant resemblance with
our specimens. To throw light on the hypothetic crocodylomorph affinities of our
specimen we looked at the different evolutionary stages of development of the
maxillary medial shelf in the remnant crocodylomorph groups (Langston, 1973),
interpreting consequently the other features visible in our specimens. An initial
stage of development of the maxillary medial shelf can be seen in some basal croc-
odylomorphs, the “sphenosuchians” (Walker, 1972; Walker, 1990; Wu &
Chatterjee, 1993; Sues et al., 1996; Clark, 2000; Clark & Sues, 2002). In the
“sphenosuchians” Sphenosuchus (Walker, 1990) and Dibothrosuchus (Wu &
Chatterjee, 1993) (Fig. 9B, E) there is a relatively short secondary palate formed
mainly by the rostral half of the maxillae that is rostral to the internal choanae. The
maxillae form a consistent part of the lateral margin of the internal choanae and do
not contribute to border the suborbital fenestrae, the palatines contact medially the
caudal halves of the maxillae and form the caudolateral margins of the internal
choanae, and the ectopterygoids do not contact the maxillae. The rostral margin of
the internal choana is at the same level of the rostralmost portion of the antorbital
fossa, the maxillary border gently arches laterally and is deep, deeper than the
TCH. Razanandrongobe sakalavae is compatible with a similar disposition of the
bones bordering the internal choanae. No traces of the rostromedial wall of the
antorbital fossa are visible in R. sakalavae, but this could depend by the fact that
our maxilla lacks all its lateral portion. Other than for its large size, R. sakalavae
differs from the “sphenosuchians” in having the attachment area for the palatine
reaching the level of the rostral margin of the internal choana, and the ventral mar-
gin of the maxilla tapering caudally. Taking into account these features, in accept-
ing the sphenosuchian topology of the bones we can hypothesize that R. sakalavae
had a more developed maxillary medial shelf, the internal choanae more retracted
caudally, and a probably smaller rostral margin of the antorbital fenestra, displaced
41
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
42 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
more caudally too. In short, it should have been like an intermediate form between
the basal “sphenosuchians” and the more derived mesoeucrocodylians (see below),
a possibility consistent with its Bathonian age (the “sphenosuchians” lived up to
the Early Jurassic, while the terrestrial mesoeucrocodylians [i.e. metasuchians] are
known from the Late Jurassic [Gomani, 1997]). The “sphenosuchians” differ from
R. sakalavae also in the dentition (Wu & Chatterjee, 1993; Sues et al., 1996; Clark,
2000), which is heterodont, but the rostral teeth do not bear denticles, and the lat-
eral teeth are labiolingually compressed, base-constricted, and bear small denticles
(about 8 per mm). In mesoeucrocodylians the maxillary contribution to the second-
ary palate is significantly increased with respect to the “sphenosuchians”. The
maxillae participate in bordering the suborbital fenestrae and often contact the
ectopterygoids (this would be the case in R. sakalavae, which could have had small
suborbital fenestrae or a great participation of the ectopterygoids to the lateral mar-
gin of the suborbital fenestrae), but are nearly excluded from bordering the inter-
nal choane, which are located well behind the caudal end of the tooth row (not pre-
served in our specimen), back to a caudal extension of the palatines that are close
to the midline of the skull. Additionally, in the mesoeucrocodylians the antorbital
fenestra is greatly reduced or absent, and the caudal portion of the maxilla broad-
ly contacts the lacrimal (this attachment area would correspond to the concavity on
the dorso-caudal margin of the specimen MSNM V5770). The specimen MSNM
V5770 matches the shape of the mid part of the maxilla of the Late Cretaceous
Peirosauridae (Fig. 9C, F) in both lateral (Carvahlo et al., 2004) and palatal
(Gasparini et al., 1991) views. As in Razanandrongobe sakalavae, in Lomasuchus
(Gasparini et al., 1991) the concavity of the maxillary medial shelf is more marked
rostrally, and broad depressions at the level of the middle of the tooth row can be
seen on the palate where occlusal contact occurs with the large mandibular teeth.
In our specimen, the cracking along the stouter portion of the maxillary medial
shelf would be close to the suture between the maxillae and rostral portion of the
palatine. However, in R. sakalavae the maxilla and the alveolar channels are more
vertical, the palate is deeper, and the “wave” in the tooth row (sensu Gasparini et
al., 1991) is more marked and closer to the attachment area for the lacrimal, sug-
gesting a snout higher and shorter than that of the peirosaurids. Also the dentition
is not particularly similar to that of the peirosaurids, which have rostral teeth sub-
circular in cross-section, and caudalmost lateral teeth globe-shaped but with a pro-
nounced neck. Moreover, in all the known members of the Peirosauridae the den-
ticles are fine (Carvalho et al., 2004). Among the mesoeucrocodylians, the bau-
rusuchids (Campos et al., 2001) and sebecosuchids (Colbert, 1946) have deep and
nearly vertical maxillae, but their similarities with R. sakalavae are less numerous
than those the peirosaurids have. Indeed in the baurusuchids the shape of the max-
illa and its relations with the surrounding bones appear hardly compatible with the
shape of MSNM V5770 [Riff, pers. comm. 2005], whereas in the maxilla of the
sebecosuchids, although more compatible with the topology of the bones in
MSNM V5770, the alveolar border does not form “waves” and the attachment for
the ectopterygoid is retracted further caudally. Some details of the maxilla of
Sebecus icaerorhinus, figured in medial view by Colbert (1946: fig. 2), resulted
however important for comparison purpose. In Sebecus, as well as in extant
43
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Crocodylia (pers. obs. 2005 on skulls of Melanosuchus niger and Crocodylus
niloticus in MSNM) and some other mesoeucrocodylians (pers. obs. 2004 on
MNHN material), the maxillary medial shelf lies just above the level of the tooth
row, whereas in R. sakalavae the palate is deeper. Unfortunately, we were not able
to check this feature in the “sphenosuchians”. R. sakalavae resembles extant rep-
resentatives of the Crocodylia (Brochu, 1999; pers. obs. 2005) in having the attach-
ment of the ectopterygoid that reaches the level of the tooth row, and a small sub-
orbital fenestra. In extant crocodiles, however, the snout is definitely flattened
dorso-ventrally, and the alveoli curve medially not far above the palatal shelf
(Brochu, pers. comm. 2004) and do not form vertical channels as they do in
MSNM V5770. Moreover, the teeth of the ziphodont Crocodylia (such as the
Pristichampsinae) bear fine denticles (Prasad & de Lapparent de Broin, 2002).
Summary
Razanandrongobe sakalavae shows the following combination of features:
alveolar channels nearly vertical as in theropods and some terrestrial crocodylo-
morphs such as “sphenosuchians”, baurusuchians, peirosaurids, and sebeco-
suchids; fused and highly sculpted interdental plates very similar to those of the
abelisaurids in shape and texture, and forming an alveolar medial surface flatter
than usual in crocodylomorphs; dental foramina resembling those of the
Crocodylomorpha in shape, density, and distribution; maxillary medial shelf croc-
odylomorph in shape (with the exception of the “protosuchians”), similar to the
shelf in “sphenosuchians” or alternatively in mesoeucrocodylians, different from
that in spinosaurid theropods, and more extended medially than in any other
theropod dinosaur; maxillary medial shelf higher to the tooth row than in croco-
dylomorphs we examined; shape and position of the broad depression on the
palate where occlusal contact occurs with the large mandibular teeth recalling the
peirosaurid crocodylomorphs (usually more laterally placed in theropods); rostral
teeth U-shaped in cross-section as in tyrannosauroid theropods; lateral teeth even
more labiolingually inflated than usually found in ziphodont crocodylomorphs
and in tyrannosaurid theropods (rounded cross-sections are commonly present in
the teeth of the non-ziphodont crocodylomorphs and spinosaurid dinosaurs);
tooth crowns stout and relatively short, as usual in crocodylomorphs; smallest lat-
eral teeth globe-shaped (the smallest teeth in Theropoda are usually scaled version
of the largest ones) as in several crocodylomorphs (in which, however, they usu-
ally have a distinct neck); theropod-like carinal variation; very large autapomor-
phic tooth denticles, even larger respect to TCH than the largest ones in the large-
bodied theropods and considerably larger than the fine denticles of any other
ziphodont crocodylomorph. Taking into account this combination of features,
Razanandrongobe sakalavae clearly appears as a new taxon, but its definitive
attribution to the Crocodylomorpha or Theropoda cannot be supported (Fig. 10).
For this reason, we refer our material to Archosauria incertae sedis, pending more
complete material.
Paleobiology
Both theropods and ziphodont crocodylomorphs with nearly vertical maxillae
44 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Fig. 10 – Phylogenetic relationships among the archosauromorph taxa cited in the discussion, based
mainly on the trees published by Brochu (2001). Additional information about the relationships among
basal Crurotarsi and mesoeucrocodylians are taken respectively from Parrish (1993) and Sereno et al.
(2001). The two hypothetic positions of Razanandrongobe sakalavae, marked by a question mark, are
represented as unresolved polytomies at the base of Theropoda and Crocodylomorpha.
Fig. 10 – Relazioni filogenetiche tra gli arcosauromorfi citati nella discussione. Schema elaborato
principalmente sulla base degli alberi pubblicati da Brochu (2001). Le informazioni addizionali ine-
renti le relazioni tra i crurotarsi basali e i mesoeucoccodrilli sono tratte rispettivamente da Parrish
(1993) e Sereno et al. (2001). Le due ipotetiche posizioni di Razanandrongobe sakalavae, contrasse-
gnate da un punto interrogativo, sono riportate come politomie irrisolte alla base di Theropoda e
Crocodylomorpha.
45
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
come from deposits that clearly indicate a terrestrial environment. In this regard,
Razanandrongobe sakalavae does not represent an exception. A size-comparison
between the maxilla and teeth of R. sakalavae and other large archosauriform
predators clearly demonstrates that it was indeed a very large animal, surpassing
by far the size of the other known Middle Jurassic Malagasy predators (Maganuco
et al., 2005), and being potentially large enough to feed on the sympatric
sauropods (Buffetaut, 2005) such as Lapparentosaurus and Archaeodontosaurus.
The thickness of the bone and the size of the alveoli of MSNM V5770 are com-
parable to those of the neosuchian Sarcosuchus and the tyrannosaurid theropods
Daspletosaurus and Tarbosaurus (pers. obs. 2004). According to Shubert &
Ungar (2005), the wear surfaces on the teeth of R. sakalavae (see description and
Fig. 8H, I) appear closer to ante-mortem enamel spalling caused by tooth-food
contact rather than to attritional facets related to tooth-tooth contact. According to
Currie et al. (1990), it appears verisimilar that flakes of enamel were splayed off
the underlying dentine surfaces during traumatic events that occurred as the ani-
mal bit into the bones or other hard objects. D. Tanke and P. Currie (pers. comm.
2004) showed us that a similar type of wear has been found in an average of one
tooth out of ten among the isolated theropod teeth in Dinosaur Provincial Park.
Although our sample is small and, consequently, its statistical significance limit-
ed, it is worth to underline that in R. sakalavae five teeth of seven present such a
kind of spalled surface. The incisiform rostral tooth referred to Razanandrongobe
sakalavae closely resembles the tyrannosaurid premaxillary teeth, which were
more likely used in scraping meat from bone than in capturing prey (Erickson,
1999; Holtz, 2003). Currie (2000) pointed out that the shape of the tyrannosaurid
lateral teeth, which are labiolingually inflated and sub-circular in cross-section,
increase their strength. We can infer that also R. sakalavae had strengthened lat-
eral teeth, because it had teeth even more inflated labiolingually and rounded in
cross-section than tyrannosaurids. Also the denticles of R. sakalavae appear very
similar to those of tyrannosaurids (Currie et al., 1990) in having many features
(e.g. size, shape and basal width) typical of strengthened denticles well adapted to
bite into bone.
Additionally, the rounded structure at the base of the junction between neigh-
bouring denticles described in this paper is comparable to the round void present
in the tyrannosaurids which, as pointed out by Abler (2001), would have protect-
ed the posterior edge of the tooth against the propagation of cracks formed by the
junction between neighbouring denticles, distributing forces over an increased
area. Following Abler (1997) it cannot be excluded either that the narrow junc-
tions between the denticles griped and held meat fibres, possibly facilitating an
infectious bite as in extant monitor lizards. As reported by Langston (1973), the
expansion and union of the palatal medial shelves represent a strengthening of the
cranial skeleton by acquiring a stronger tubular construction. Busbey (1995)
demonstrated that the thickening of the bones and the addition of a secondary
palate (even an incomplete one) significantly increase the ability of the snout to
withstand both vertical bending and axial torsion. Both massive maxillary medi-
al shelf and thickness of the maxillary bony walls are visible in MSNM V5770.
According to Senter (2003) the fused interdental plates create a medial wall
around the teeth, which suggests better resistance to transverse force. In conclu-
sion, on the basis of the data collected it is equally parsimonious to postulate that
Razanandrongobe sakalavae usually killed its prey or not. What can be inferred
for certain is that: R. sakalavae had a robust snout; the above mentioned features
of the maxilla and dentition strongly suggest a diet including also hard tissues
such as bones and tendons.
Conclusions
The present study demonstrates that a new species of very large, terrestrial,
predatory archosaur with a unique combination of maxillary and tooth features
lived in Madagascar in the Middle Jurassic. Many features of Razanandrongobe
sakalavae, gen. et sp. nov., strongly suggest that it fed also on hard tissues such as
bones and tendons. Besides the autapomorphic denticles, R. sakalavae would pos-
sess different autapomorphic, synapomorphic and homoplasic features according
to which taxon it is compared with. Although R. sakalavae differs in some aspects
from any currently known member of the Crocodylomorpha and Theropoda, it
belongs certainly to one of those two taxa. The nature of our material renders
impossible to perform a real phylogenetic analysis, thus at present no one of the
two phylogenetic assignments hypothesized above can be definitively confirmed
or refuted. Therefore we refer our material to Archosauria incertae sedis.
Acknowledgements
The authors are deeply grateful to Gilles Emringer, who made possible that the
maxilla became available to science; Luciano Allievi (BIOE s.r.l., Milano), who
made possible the 2003 expedition; the firm Fossilia s.n.c., who financed the 2001
expedition; the Ministère de l’Energie et des Mines and the Direction des Mines
et de la Géologie de Madagascar (in particular Georges Rakotonirina) for their
indispensable collaboration. Many thanks also to the people of Andranomamy for
the kind help in the field, to Chris Brochu for the careful review of the manuscript,
and to Eric Buffetaut and one anonymous reviewer for their comments and criti-
cism. Chris Brochu also provided helpful suggestions on the presentation of the
manuscript. We thank also: Michael Holland and Jack Horner for access to the
MOR collection; Philippe Taquet for access to the MNHN collection; Phil Currie
and Jim Gardner for access to the TMP collection; Scott Sampson for permission
to our collaborator (Federico Fanti) to take photos of the Malagasy dinosaur spec-
imens in his care; Marco Auditore, Bernard Battail, Chris Brochu, Thomas Carr,
Andrea Cau, Phil Currie, Douglas Riff Gonçalves, Didier Dutheil, Federico Fanti,
Thomas Holtz Jr., Octávio Mateus, Stefania Nosotti, Karin Peyer, Federico
Pezzotta, Sebastièn Steyer, Darren Tanke, and Philippe Taquet for the helpful dis-
cussions; Eric Buffetaut and Francois Escuillié for useful information; and
Debora Affer and Lorenzo Magnoni for lab preparation. Photos are by Cristiano
Dal Sasso and Giovanni Pasini, drawings are by Simone Maganuco. Simone
Maganuco acknowledges the MSNM for his grant. The opinions expressed in the
paper remain our own.
46 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
References
Abler W. L., 1997 - Tooth Serrations in Carnivorous Dinosaurs. In: The Encyclopedia
of Dinosaurs. Currie P. J. & Padian K. (eds). Academic Press, New York: 740-743.
Abler W. L., 2001 - A kerf-and-drill model of tyrannosaur tooth serrations. In:
Mesozoic Vertebrate Life. Tanke D. H. & Carpenter K. (eds.). Indiana
University Press: 84-89.
Alcober O., 2000 - Redescription of the skull of Saurosuchus galilei (Archosauria:
Rauisuchidae). Journal of Vertebrate Paleontology, 20 (2): 302–316.
Allain R., 2002 - The discovery of megalosaur (Dinosauria: Theropoda) in the
Middle Bathonian of Normandy (France) and its implications for the phyloge-
ny of basal Tetanurae. Journal of Vertebrate Paleontology, 22: 548-563.
Benton M. J., 1986 - The Late Triassic reptile Teratosaurus- a rauisuchian, not a
dinosaur. Palaeontology, 29: 293-301.
Benton M. J. & Walker A. D., 2002 - Erpetosuchus, a crocodile-like basal
archosaur from the Late Triassic of Elgin, Scotland. Zoological Journal of the
Linnean Society, 136: 25-47.
Besairie H., 1936 - Recherches géologiques à Madagascar, première suite. La
Géologie du Nord-Ouest. Mémoires Academie Malgache Tananarive, 21.
Besairie H., 1972 - Géologie de Madagascar. I. Les Terraines Sédimentaires.
Annales Géologiques de Madagascar, 35: 1-465.
Boast J. & Nairn A. E. M., 1982 - An outline of the geology of Madagascar. In: The
ocean basins and margins. Vol. 6. The Indian Ocean. Nairn A. E. M. & Stehli F.
G. (eds.). Plenum Press, New York: 649-696.
Bonaparte J. F., 1986 - Faunal change across the Triassic - Jurassic boundary. In:
The beginning of the age of dinosaurs. Padian K. (ed.). Cambridge University
Press, New York: 247-258.
Brochu C. A., 1999 - Phylogeny, systematics, and historical biogeography of
Alligatoroidea. Society of Vertebrate Paleontology, Memoir 6.
Brochu C. A., 2001 - Progress and future directions in archosaur phylogenetics.
Journal of Paleontology, 75: 1185-1201.
Brochu C. A., 2003 - Osteology of Tyrannosaurus rex: Insights from a nearly com-
plete skeleton and high-resolution computed tomographic analysis of the skull.
Society of Vertebrate Paleontology, Memoir 7: 1-138.
Buckley G. A., Brochu C. A., Krause D. W. & Pol D. 2000 - A pug-nosed croco-
dyliform from the Late Cretaceous of Madagascar. Nature, 405: 941-944.
Buffetaut E., 2005 - A new sauropod dinosaur with prosauropod-like teeth from the
Middle Jurassic of Madagascar. Bulletin de la Société Géologique de France, t.
176, n° 5: 467-473.
Busbey A. B., 1995 - The structural consequences of skull flattening in crocodil-
ians. In: Functional Morphology in Vertebrate Paleontology. Thomason J. J.
(ed.). Cambridge University Press: 173-192.
Campos D. A., Suarez J. M., Riff D. & Kellner A. W. A., 2001 - Short note on a
new Baurusuchidae (Crocodyliformes, Metasuchia) from the Upper Cretaceous
of Brazil. Boletim do Museu Nacional, Geologia, 57: 1-7.
47
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
Candeiro C. R. A., Abranches C. T., Abrantes E. A., Avilla L. S., Martins V. C.,
Moreira A. L., Torres S. R. & Bergqvist L. P., 2004 - Dinosaurs remains from
western São Paulo state, Brazil (Bauru Basin, Adamantina Formation, Upper
Cretaceous). Journal of South American Earth Sciences, 18: 1–10.
Carr T. D., 1999 - Craniofacial ontogeny in Tyrannosauridae (Dinosauria,
Coelurosauria). Journal of Vertebrate Paleontology, 19 (3): 497-520.
Carroll R. L., 1988 - Vertebrate Paleontology and Evolution, WH Freeman & Co..
Carvalho I. S., Ribeiro L. C. B. & Avilla L. S., 2004 - Uberabasuchus terrificus sp.
nov., a New Crocodylomorpha from the Bauru Basin (Upper Cretaceous),
Brazil. Gondwana Research, 7 (4): 975-1002.
Clark J. M., 1994 - Patterns of evolution in Mesozoic Crocodyliformes. In: In the
Shadow of the Dinosaurs. Fraser N. C. & Sues H.-D. (eds.). Cambridge
University Press: 84-97.
Clark J. M., Sues H.-D. & Berman D. S., 2000 - A new specimen of Hesperosuchus
agilis from the Upper Triassic of New Mexico and the interrelationships of
basal crocodylomorph archosaurs. Journal of Vertebrate Paleontolology, 20:
683-704.
Clark J. M. & Sues H.-D., 2002 - Two new basal crocodylomorph archosaurs from
the Lower Jurassic and the monophyly of the Sphenosuchia. Zoological Journal
of the Linnean Society, 136: 77-95.
Colbert E. H., 1946 - Sebecus, representative of a peculiar suborder of fossil
Crocodilia from Patagonia. Bulletin of the American Museum of Natural
History, 87: 217-270.
Currie P. J., 2000 - Theropods from the Cretaceous of Mongolia. In: The Age of
Dinosaurs in Russia and Mongolia. Benton M. J., Shishkin M. A., Unwin D. M.,
& Kurochkin E. N. (eds.). Cambridge University Press: 434-455.
Currie P. J., 2003 - Cranial anatomy of tyrannosaurid dinosaurs from the Late
Cretaceous of Alberta, Canada. Acta Palaeontologica Polonica, 48 (2): 191-226.
Currie P. J., Rigby J. K. Jr. & Sloan R. E., 1990 - Theropod teeth from the Judith
River Formation of southern Alberta, Canada. In: Dinosaur Systematics:
Perspectives and Approaches. K. Carpenter K. & Currie P. J. (eds.). Cambridge
University Press, New York: 107-125.
Dal Sasso C. & Pasini G., 2003 - First record of pterosaurs (Pterosauria,
Archosauromorpha, Diapsida) in the Middle Jurassic of Madagascar. Atti della
Società Italiana di Scienze Naturali e del Museo Civico di Storia Naturale in
Milano, 144 (II): 281-296.
Dal Sasso C., Maganuco S., Buffetaut E. & Mendez A. M., 2005 - New informa-
tion on the skull of the enigmatic theropod Spinosaurus, with remarks on its
size and affinities. Journal of Vertebrate Paleontology, 25 (4): 888-896.
Erickson G. M., 1999 - Breathing life into Tyrannosaurus rex. Scientific American,
281: 42-49.
Flynn J. J., Parrish J. M., Simpson W. F., Razafimanantsoa L., Andriatompohavana
R. & Totovolahy A., 1997 - New Triassic and Jurassic Vertebrates from
Madagascar. Journal of Vertebrate Paleontology, 17 (3, suppl.): 46A.
48 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Flynn J. J., Parrish J. M., Rakotosaminanana B., Simpson W. F. & Wyss A. R., 1999
- A Middle Jurassic mammal from Madagascar. Nature, 401: 57-60.
Galton P. M., 1985 - The poposaurid thecodontian Teratosaurus suevicus v. Meyer,
plus referred specimens mostly based on prosauropod dinosaurs, from the
middle Stubensandstein (Upper Triassic) of Nordwürttemberg. Stuttgarter
Beiträge zur Naturkunde, Serie B, 116: 1-29.
Gasparini Z., Chiappe L. M. & Fernandez M., 1991 - A new Senonian peirosaurid
(Crocodylomorpha) from Argentina and a synopsis of the South American
Cretaceous crocodilians. Journal of Vertebrate Paleontology, 11 (3): 316-333.
Gomani E. M., 1997 - A crocodyliform from the Early Cretaceous Dinosaur Beds,
Northern Malawi. Journal of Vertebrate Paleontology, 17 (2): 280-294.
Gow C. E., 2000 - The skull of Protosuchus haughtoni, an Early Jurassic crocodyli-
form from southern Africa. Journal of Vertebrate Paleontology, 20 (1): 49-56.
Gower D. J. & Sennikov A. G., 2000 - Early archosaurs from Russia. In: The Age
of Dinosaurs in Russia and Mongolia, Benton M. J., Shishkin M. A., Unwin
D., and Kurochkin E. N. (eds.). Cambridge University Press, Cambridge:
120-139.
Gradstein F. M., Ogg J. G. & Smith, A. G., (eds.), 2004 - A Geologic Time Scale
2004. Cambridge University Press.
Holtz T. R. Jr., 2000 - A new phylogeny of the carnivorous dinosaurs. Gaia, 15: 5-
61. (The papers of the December 1998 issue of Gaia are technically dated 2000).
Holtz T. R. Jr., 2001a - The phylogeny and taxonomy of the Tyrannosauridae. In:
Mesozoic Vertebrate Life. Tanke D. H. & Carpenter K. (eds.). Indiana
University Press, USA: 64-83.
Holtz T. R. Jr., 2001b - Pedigree of the tyrant kings: New information on the ori-
gin and evolution of the Tyrannosauridae. Journal of Vertebrate Paleontology,
21 (3): 62A-63A.
Holtz T. R. Jr., 2003 - Dinosaur Predation. Evidence and ecomorphology. In:
Predator-Prey Interactions in the fossil record. Kelley P. H., Kowalewsky M., &
Hansen T. A. (eds.). Kluwer Academic/Plenum Publishers, New York: 325-340.
Hurum J. H. & Sabath K., 2003 - Giant theropod dinosaurs from Asia and North
America: Skulls of Tarbosaurus bataar and Tyrannosaurus rex compared. Acta
Palaeontologica Polonica, 48 (2): 161-190.
Hutt S., Naish D., Martill D. M., Barker M. J. & Newbery P., 2001 - A prelimina-
ry account of a new tyrannosauroid theropod from the Wessex Formation (Early
Cretaceous) of southern England. Cretaceous Research, 22: 227-242.
Janensch W., 1925 - Die Coelurosaurier und Theropoden der Tendaguru-Schichten
Deutsch-Ostafrikas. Palaeontographica 1(Suppl. 7): 1-99.
Juul L., 1994 - The phylogeny of basal archosaurs. Palaeontologia Africana: 1-38.
Kellner A. W. A. & Campos D. A., 2002 - On a theropod dinosaur (Abelisauria)
from the continental Cretaceous of Brazil. Arquivos do Museu Nacional, 60 (3):
163-170.
Langston W., 1973 - The crocodilian skull in historical perspective. In: Biology of the
Reptilia, 4. Gans C. & Parson T. S. (eds.). Academic Press, New York: 263-284.
Langston W., 1975 - Ziphodont crocodiles: Pristichampsus vorax (Troxell), new com-
bination, from the Eocene of North America. Fieldiana Geology, 33: 291-314.
49
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
50 SIMONE MAGANUCO, CRISTIANO DAL SASSO & GIOVANNI PASINI
Lavocat R., 1955 - Etude des gisements de dinosauriens de la region de Majunga
(Madagascar). Travaux du Bureau Géologique, 69: 1-19.
Lehman T. M. & Carpenter K., 1990 - A partial skeleton of the tyrannosaurid
dinosaur Aublysodon from the Upper Cretaceous of New Mexico. Journal of
Paleontology, 64: 1026-1032.
Lydekker R., 1895 - On bones of a sauropodous dinosaur from Madagascar.
Quarterly Journal of the Geological Society of London, 51: 329-336.
Maganuco S., Cau A. & Pasini G., 2005 - First description of theropod remains from
the Middle Jurassic (Bathonian) of Madagascar. Atti della Società Italiana di
Scienze Naturali e del Museo Civico di Storia Naturale in Milano, 146 (II): 165-202.
Motani R., 1997 - Temporal and spatial distribution of tooth implantation in
ichthyosaurs. In: Ancient Marine Reptiles. Callaway J. M. & Nicholls E. L.
(eds.). Academic Press: 81-103.
Nash D. S., 1975 - The morphology and relationships of a crocodilian,
Orthosuchus stormbergi, from the Upper Triassic of Lesotho. Annals of South
African Museum, 67 (7): 227-329.
Ogier A., 1975 - Etude de nouveaux ossements de Bothriospondylus (Sauropode)
d’un gisement du Bathonien de Madagascar. Doctoral thesis (3eCycle),
Universitè Paris VI.
Ortega F. & Gasparini Z., 2000 - A new species of Araripesuchus (Crocodylomorpha,
Mesoeucrocodylia) from the Lower Cretaceous of Patagonia (Argentina). Journal
of Vertebrate Paleontology 20 (1): 57-76.
Parrish J. M., 1993 - Phylogeny of the Crocodylotarsi, with reference to archosauri-
an and crurotarsan monophyly. Journal of Vertebrate Paleontology, 13: 287-308.
Pol D., 2003 - New remains of Sphagesaurus huenei (Crocodylomorpha:
Mesoeucrocodylia) from the Late Cretaceous of Brazil. Journal of vertebrate
Paleontology, 23 (4): 817-831.
Prasad G. V. R. & de Lapparent de Broin F., 2002 - Late Cretaceous crocodile
remains from Naskal (India): comparisons and biogeographic affinities.
Annales de Paléontologie, 88: 19-71.
Rauhut O. W. M., 2003 - The interrelationships and evolution of basal theropod
dinosaurs. Special Papers in Palaeontology, 69: 1-213.
Sampson S. D., Witmer L. M., Forster C. A., Krause D. W., O’Connor P. M.,
Dodson P. & Ravoavy F., 1998 - Predatory dinosaur remains from Madagascar:
Implications for the Cretaceous biogeography of Gondwana. Science, 280:
1048-1051.
Schubert B. W. & Ungar P. S., 2005 - Wear facets and enamel spalling in tyranno-
saurid dinosaurs. Acta Palaeontologica Polonica 50 (1): 93-99.
Senter P., 2003 - New information on cranial and dentary features of the Triassic
Archosauriform reptile Euparkeria capensis. Palaeontology, 46 (3): 613-621.
Sereno P. C., 1991 - Basal archosaurs: phylogenetic relationships and functional
implications. Society of Vertebrate Paleontology, Memoir 2: 1-53.
Sereno P. C., Beck A. L., Dutheil D. B., Gado B., Larsson H. C. E., Lyon G. H.,
Marcot J. D., Rauhut O. W. M., Sadleir R. W., Sidor C. A., Varricchio D. J.,
Wilson G. P. & Wilson J. A., 1998 - A long-snouted predatory dinosaur from
Africa and the evolution of Spinosaurids. Science, 282 (5392): 1298-1302.
Sereno P. C., Larsson H. C., Sidor C. A. & Gado B., 2001 - The Giant
Crocodyliform Sarcosuchus from the Cretaceous of Africa. Science, 294: 1516-
1519.
Sereno P. C., Wilson J. A. & Conrad J. L., 2004 - New dinosaurs link southern land-
masses in the Mid-Cretaceous. Proceedings of the Royal Society of London, B:
1471-2954.
Stromer [von Reichenbach], Ernst. 1914. Ergebnisse der Forschungsreisen Prof. E.
Stromers in den Wüsten Ägyptens. II. Wirbeltier-Reste der Baharije-Stufe (unter-
stes Cenoman). 1. Einleitung und 2. Libycosuchus. Abhandlungen der Königlich
Bayerischen Akademie der Wissenschaften, Mathematisch-Physikalische Klasse,
München, 27 (3): 1-16.
Sues H.-D., Shubin N. H., Olsen P. E. & Amaral W. W., 1996 - On the cranial struc-
ture of a new protosuchid (Archosauria: Crocodyliformes) from the McCoy Brook
Formation (Lower Jurassic) of Nova Scotia, Canada. Journal of Vertebrate
Paleontology, 16 (1): 34-41.
Thevenin A., 1907 - Paléontologie de Madagascar. IV Dinosauriens. Annales de
Paléontologie, I (II): 121-136.
Walker A. D., 1972 - New light on the origin of birds and crocodiles. Nature, 137:
257-263.
Walker A. D., 1990 - A revision of Sphenosuchus acutus Haughton, a crocodylo-
morph reptile from the Elliot Formation (late Triassic or early Jurassic) of
South Africa. Philosophical Transactions of the Royal Society of London,
Biological Sciences, 330: 1-20.
Weishampel D. B., Dodson P. & Osmólska H., (eds.), 2004 - The Dinosauria.
Second Edition. University of California Press.
Willis P. M. A. & Mackness B., 1996 - Quinkana babarra, a new species of
ziphodont mekosuchine crocodile from the Early Pliocene Bluff Downs Local
fauna, Northern Australia, with revision of the genus. Proceedings and Journal
of the Linnean Society of New South Wales, 116: 143-151.
Wu X.-C. & Chatterjee S., 1993 - Dibothrosuchus elaphros, a crocodylomorph
from the Lower Jurassic of China and the phylogeny of the Sphenosuchia.
Journal of Vertebrate Paleontology, 13 (1): 58-89.
Wu X.-C. & Russell A. P., 2001 - Redescription of Turfanosuchus dabanensis
(Archosauriformes) and new information on its phylogenetic relationships.
Journal of Vertebrate Paleontology, 21 (1): 40-50.
Zaher H. & Rieppel O., 1999 - Tooth Implantation and Replacement in Squamates,
with Special Reference to Mosasaur Lizards and Snakes. American Museum
Novitates, 3271: 1-19.
Zinke J., 1998 - Small theropod teeth from the Upper Jurassic coal mine of
Guimarota (Portugal). Palaontologische Zeitschrift, 72 (1/2): 179-189.
Ricevuto: 26 agosto 2005
Approvato: 5 gennaio 2006
51
A NEW LARGE PREDATORYARCHOSAUR FROM THE MIDDLE JURASSIC (BATHONIAN) OF MADAGASCAR
