Anatomical Details of Aetosauria (Archosauria: Pseudosuchia) as Revealed by an Articulated Posterior Skeleton from the Upper Triassic Ischigualasto Formation, San Juan Province, Argentina

Abstract

Aetosaurs are an early-diverging clade of “crocodile-line” archosaurs whose oldest records come from Argentina and Brazil. Articulated skeletons of aetosaurs are rare, but offer insight into their paleobiology. We describe here an incomplete, articulated posterior skeleton of an aetosaur from the Upper Triassic Ischigualasto Formation of San Juan Province, Argentina. It resembles Aetosauroides, the only aetosaur genus currently recognized from the Ischigualasto Formation, but lacks apomorphies that would allow us to assign it to that taxon, appears to be more robust than the holotype of A. scagliai, and preserves a different tail morphology. We identify the specimen as a basal aetosaur in part because the few exposed dorsal and lateral osteoderms have a typical ornamentation of radially distributed pits, grooves, and ridges emanating from a “center of ossification.” Although the specimen is incomplete and exposed primarily in ventral view from the sacrum posteriorly, it preserves many anatomical features not often preserved in aetosaurs, including extensive appendicular armor and a well- preserved caudal ventral carapace. The latter apparently consists of only two columns of ventral osteoderms, and preserves a large cloacal vent proximally. Posteriorly, the ventral paramedian osteoderms fuse to form a single element, something that has not previously been demonstrated in aetosaurs. The arrangement of osteoderms around the vent is distinct from that seen in A. scagliai. The ventral caudal osteoderms differ from many other aetosaurs in that they do not transition from wider than long to longer than wide, indicating that the specimen had a relatively abbreviated tail.

Full text

Submitted
:
J
anuary
1
3
th
, 202
1
–
Accepted
:
S
eptember
5
th
, 202
1
–
Posted online
:
September
1
4
th
, 202
1
To link and cite this article:
doi: 10.5710/AMGH.05.09.2021.3426

1
ANATOMICAL DETAILS OF AETOSAURIA (ARCHOSAURIA:PSEUDOSUCHIA)
AS REVEALED BY AN ARTICULATED POSTERIOR SKELETON FROM THE
UPPER TRIASSIC ISCHIGUALASTO FORMATION, SAN JUAN PROVINCE,
ARGENTINA
DETALLES ANATÓMICOS DE AETOSAURIA (ARCHOSAURIA:PSEUDOSUCHIA)
REVELADOS POR UN ESQUELETO POSTCRANEAL ARTICULADO DE LA
FORMACIÓN ISCHIGUALASTO (TRIASICO SUPERIOR), PROVINCIA DE SAN JUAN,
ARGENTINA.
ANDREW B. HECKERT1 RICARDO N. MARTÍNEZ2 AND MATTHEW D. CELESKEY3
1Department of Geology, Appalachian State University, ASU Box 32067, Boone, NC 28608-
2067, USA heckertab@appstate.edu
2Instituto y Museo de Ciencias Naturales, Universidad Nacional de San Juan, España 400
(norte), CP 5400, San Juan, Argentina; martinez@unsj.edu.ar
3New Mexico Museum of Natural History and Science, 1801 Mountain Road NW,
Albuquerque, NM 87104, USA; deadanimaldesign@hmnh.org
36 pag. (text+references); 6 fig. 2 tables
Running Header: HECKERT ET AL: ARTICULATED ISCHIGUALASTO FORMATION
AETOSAUR POSTERIOR SKELETON
Short Description: Description of an incomplete skeleton of an aetosaur including novel
anatomical information regarding the vent and fusion of caudal osteoderms.
Corresponding author: Andrew B. Heckert (heckertab@appstate.edu)

2
Abstract. Aetosaurs are an early-diverging clade of “crocodile-line” archosaurs whose oldest
records come from Argentina and Brazil. Articulated skeletons of aetosaurs are rare, but offer
insight into their paleobiology. We describe here an incomplete, articulated posterior skeleton
of an aetosaur from the Upper Triassic Ischigualasto Formation of San Juan Province,
Argentina. It resembles Aetosauroides, the only aetosaur genus currently recognized from the
Ischigualasto Formation, but lacks apomorphies that would allow us to assign it to that taxon,
appears to be more robust than the holotype of A. scagliai, and preserves a different tail
morphology. We identify the specimen as a basal aetosaur in part because the few exposed
dorsal and lateral osteoderms have a typical ornamentation of radially distributed pits,
grooves, and ridges emanating from a “center of ossification.” Although the specimen is
incomplete and exposed primarily in ventral view from the sacrum posteriorly, it preserves
many anatomical features not often preserved in aetosaurs, including extensive appendicular
armor and a well-preserved caudal ventral carapace. The latter apparently consists of only
two columns of ventral osteoderms, and preserves a large cloacal vent proximally.
Posteriorly, the ventral paramedian osteoderms fuse to form a single element, something that
has not previously been demonstrated in aetosaurs. The arrangement of osteoderms around
the vent is distinct from that seen in A. scagliai. The ventral caudal osteoderms differ from
many other aetosaurs in that they do not transition from wider than long to longer than wide,
indicating that the specimen had a relatively abbreviated tail.
Key words. Triassic. Ischigualasto Formation. Aetosaur. South America. Argentina.
Osteoderm. Armor. Archosaur.

3
Resumen. DETALLES ANATÓMICOS DE AETOSAURIA
(ARCHOSAURIA:PSEUDOSUCHIA) REVELADOS POR UN ESQUELETO
POSTCRANEAL ARTICULADO DE LA FORMACIÓN ISCHIGUALASTO (TRIASICO
SUPERIOR), PROVINCIA DE SAN JUAN, ARGENTINA. Los aetosaurios son un clado de
arcosaurios que divergieron temprano del linaje de los cocodrilos, cuyos registros más
antiguos provienen de Argentina y Brasil. Los esqueletos articulados de aetosaurios son
raros, pero ofrecen información sobre su paleobiología. Aquí describimos un esqueleto
posterior incompleto y articulado de un aetosaurio de la Formación Ischigualasto del Triásico
Superior de la Provincia de San Juan, Argentina. Se parece a Aetosauroides, el único género
de aetosaurios actualmente reconocido de la Formación Ischigualasto, carece de apomorfias
que nos permitan asignarlo a ese taxón y parece ser más robusto que el holotipo de A.
scagliai. Identificamos al espécimen como un aetosaurio basal en parte porque los pocos
osteodermos laterales y dorsales expuestos tienen una ornamentación típica de hoyos, surcos
y crestas distribuidos radialmente que emanan de un "centro de osificación". Aunque el
espécimen está incompleto y expuesto principalmente en vista ventral desde el sacro
posteriormente, es importante porque conserva muchas características anatómicas que a
menudo no se conservan en los aetosaurios, incluida una extensa armadura apendicular y un
caparazón ventral caudal bien conservado. Aparentemente este último consta de solo dos
columnas de osteodermos ventrales y conserva una gran ventilación cloacal proximalmente.
Posteriormente, los osteodermos paramediales ventrales se fusionan para formar un solo
elemento, algo que no se había demostrado previamente en los aetosaurios. Los osteodermos
caudales difieren de muchos otros aetosaurios en que disminuyen rápidamente en longitud, lo
que indica que el espécimen tenía una cola relativamente corta. Algunas de estas
características pueden representar la condición plesiomórfica de Aetosauria.

4
Palabras clave. Triassic. Formación de Ischigualasto. Aetosaurio. América del Sur.
Argentina. Osteodermos. Armadura. arcosaurio

5
AETOSAURS are a clade of pseudosuchian archosaurs known from Upper Triassic deposits
across much of Pangea, most recently reviewed by Desojo et al. (2013). In South America,
aetosaurs are known from Argentina, Brazil, and Chile, with a total of five genera
(Aetosauroides, Neoaetosauroides, Aetobarbakinoides, Polesinesuchus, and Chilenosuchus)
recognized (Desojo et al., 2013; da-Silva et al., 2014). All but Chilenosuchus, which is too
poorly known to ascribe to a precise taxonomic position, pertain to a grade of relatively
plesiomorphic aetosaurs characterized, in part, by narrow bodies and relatively simple armor
lacking prominent spikes or other ornamentation (Heckert & Lucas, 2000; Desojo et al.,
2013; Parker, 2016a,b, 2018a,b).
In Argentina, the two stratigraphic units that yield aetosaurs are the Ischigualasto and
Los Colorados formations in the Ischigualasto-Villa Unión Basin in San Juan and La Rioja
provinces (Fig. 1). Casamiquela (1960) named and described (Casamiquela, 1961, 1967)
specimens of Aetosauroides scagliai and Argentinosuchus bonapartei from the Ischigualasto
Formation and Bonaparte (1967) named Neoaetosauroides engaeus from the Los Colorados
Formation. Although the Ischigualasto aetosaurs were assigned to Stagonolepis by Heckert
and Lucas (2000, 2002), Desojo and Ezcurra (2011; Ezcurra, 2016) rejected this assignment
and Argentinosuchus is considered a nomen dubium best assigned to an indeterminate
aetosaurine (e.g., Desojo & Ezcurra, 2011), so, currently, the only known aetosaur from the
Ischigualasto Formation is Aetosauroides (Desojo & Ezcurra, 2011; Ezcurra, 2016; Desojo et
al., 2013; Taborda et al., 2013). As a consequence, most records of aetosaurs from the
Ischigualasto have been assigned to Aetosauroides (e.g., Bonaparte, 1978; Cerda & Desojo,
2011; Martínez et al., 2013; Cerda et al., 2018; Desojo et al., 2020). Specimens of
Aetosauroides have also been reported from Brazil (Zacarias, 1981) and were subsequently
referred to Stagonolepis (Heckert & Lucas, 2000; Lucas & Heckert, 2001), an assignment
rejected by Desojo and Ezcurra (2011; Taborda et al., 2015) who referred Zacarias’ (1981)

6
material to Aetosauroides. Recent discoveries of new taxa in Brazil (Aetobarbakinoides
brasiliensis Desojo et al., 2012; Polesinesuchus aurelioi da-Silva et al., 2014), as well as
additional specimens of Aetosauroides from that unit (e.g., Biacchi Brust et al., 2018; Paes
Neto et al., 2021) have demonstrated that there is greater diversity of aetosaurs in the Santa
Maria Formation there than is presently known from the Ischigualasto Formation. Most
recent phylogenetic analyses find the South American aetosaurs, especially Aetosauroides, to
be among the most basal taxa in the clade (Parker, 2016a; Biacchi Brust et al., 2018;
Hoffman et al., 2018). To date, essentially all published descriptions of aetosaurs from the
Ischigualasto Formation have relied on the collections originally described by Casamiquela
(1960, 1961, 1967), so the specimen we describe here is the first novel articulated aetosaur
fossils from Argentina to be described and illustrated in decades. Our description here
supersedes a previous presentation (Martínez & Heckert, 2015).
Institutional abbreviations. CPE2, Coleção Municipal, São Pedro do Sul, Brazil; NCSM,
North Carolina Museum of Natural Sciences, Raleigh, North Carolina, USA; NMMNH, New
Mexico Museum of Natural History and Science, Albuquerque, New Mexico, USA; PVL,
Instituto de Miguel Lillo, Tucumán, Argentina; PVSJ, Instituto y Mueso de Universidad
Nacional de San Juan, San Juan, Argentina; SMNS, Staatliches Museum für Naturkunde,
Stuttgart, Germany; ULABRA PVT, Universidade Luterana do Brasil, Coleção de
Paleovertebrados, Canoas, Rio Grande do Sul, Brazil.
MATERIAL AND METHODS
The specimen described here, PVSJ 691, is housed at the Instituto y Museo de
Universidad Nacional de San Juan, San Juan, Argentina. It was collected in 1996 and
prepared using mechanical methods. All consolidants used in the field and for final
preparation were cyanoacrylates. It has not been described scientifically except as one of the
specimens assigned to Aetosauroides scagliai in Desojo’s (2005) Ph.D. dissertation and in

7
our earlier abstract (Martínez & Heckert, 2015). All specimens of Aetosauroides used for
comparison were examined first-hand by at least one of the authors (typically ABH),
although we make numerous references to the literature to facilitate future work. Thus,
throughout this paper comparisons to previously named taxa include the holotypes and, in
some cases, referred material as follows: Aetosauroides scagliai—holotype (PVL 2073) and
referred specimens (PVL 2052, PVL 2059, PVL 2091); Stagonolepis robertsoni—holotype
and referred specimens as described by Walker (1961; see also Parker 2018a);
Aetobarbakinoides brasiliensis—holotype (CPE2 168, Desojo et al., 2012); Polesinesuchus
aurelioli—holotype (ULABRA PVT003; Roberto-da-Silva et al., 2014). Where
measurements are listed they were made with electronic digital calipers.
PROVENANCE
The specimen described here was collected in 1996 by an expedition of the Instituto y
Museo de Ciencias Naturales de San Juan to the Parque Provincial de Ischigualasto (Fig.
1.2). It was discovered by Eriko Nagao, a Japanese scientific writer who participated in
several expeditions of the PVSJ conducted by one of us (RNM) and Oscar Alcober and
sponsored by Earthwatch Institute. The specimen was found in the area called “Cementerio
de Mate Jim” in outcrops located north of the Agua de la Peña River, in strata composed of
grey-green sandy siltstones corresponding to a well-drained floodplain and located
approximately 180 m above the base of the Ischigualasto Formation (Fig. 1.3). This horizon
corresponds to the top layers of the Cancha de Bochas Member (sensu Currie et al., 2009)
and middle levels of the Hyperodapedon-Exaeretodon-Herrerasaurus biozone (sensu
Martínez et al., 2011).
Martínez et al. (2013) reported that aetosaurs are most common in the lower portion
of the Ischigualasto Formation, specifically in the Cancha de Bochas and lower Valle de la
Luna members sensu Currie et al. (2009). The lower interval corresponds to the

8
Hyperodapedon-Exaeretodon-Herrerasaurus biozone, and the two higher specimens are
from the Exaeretodon biozone (Martínez et al., 2013; Fig. 1.3). The specimens indexed by
Martínez et al. (2013) and reproduced in Figure 1.3 are solely aetosaurs collected by the
PVSJ; as others (e.g., Taborda et al., 2015) have noted, it is not readily possible to integrate
historic specimens, including the holotype of Aetosauroides scagliai, into either the column
presented by Martínez et al. (2011, 2013) or the lithostratigraphic framework of Currie et al.
(2009; see also Desojo et al., 2020). Thus it appears that PVSJ 691, the focus of this study,
was collected from the middle part of the Aetosauroides-bearing stratigraphic interval (Fig.
1.3), but it is unclear how this may relate to the PVL specimens used for comparison.
DESCRIPTION
The specimen we describe here, PVSJ 691, consists of an incomplete, largely
articulated postcranial skeleton preserving most of the pelvis, hind limbs, and tail. It was
found in life position, with much of the dorsal side exposed and weathered. As currently
prepared, the specimen is exposed largely in ventral view, with the hind limbs directed
posteriorly in a single block that is approximately 84 cm long, as much as 33 cm wide (Figs.
2.1–2, 3). Hematitic concretions and several large fractures somewhat obscure some
anatomical details. Anterior to the pelvis the specimen includes some of the dorsal armor and
vertebrae, but these are, relative to the more posterior elements, more jumbled and
disarticulated (possibly scavenged during the Triassic prior to burial) and some have been
removed to facilitate further preparation. An ulna found with the specimen that bears the
same number is too large (13 cm long) and robust to pertain to this individual, although it
could represent another, larger-bodied aetosaur (Desojo, 2005). The exposed vertebrae are
visible in ventral to left lateral view. Other preserved elements visible in the primary block
include part or all of both pubes, the ischia, parts of the left ilium, much of both hind limbs,
and most of the armor, including numerous lateral, ventral, and appendicular osteoderms in

9
external view and the ventral surface of some paramedian osteoderms. The left femur, tibia,
and fibula, tarsals, proximal metatarsals, and some phalanges are all partially visible,
although parts are obscured by appendicular osteoderms and/or hematitic concretions. The
right hind limb is largely covered by appendicular osteoderms, but the distal femur, much of
the tibia and fibula, astragalus, calcaneum, and part of the pes is visible. In the following
paragraphs we describe these elements in greater detail.
Axial Skeleton
Parts of three dorsal vertebrae are visible anterior to the articulated pelvis (Fig. 2.1–
2.3). We use D1, D2, and D3 to identify these elements, as we are not certain of their position
(although we suspect that D1 corresponds to the 12th or more posterior dorsal vertebra). D1 is
exposed in partial lateral view, D2 in ventral and lateral views, and D3 in anterior view. D1
was at least 24–25 mm long originally, with a posterior articular facet 22 mm high. The body
of the centrum is 19 mm tall from the ventral margin to the apparent position of the
neurocentral suture, and the vertebra as a whole was more than 32 mm tall. D2 was at least
25 mm long, 21 mm wide across the posterior facet and at least 13–15 mm tall to the suture
between the neural arch and the centrum. D3 is barely exposed, but 18 mm across the anterior
articular facet. The centra are u-shaped in cross-section, with no fossae developed on their
lateral faces. The neurocentral sutures are closed, and in fact difficult to discern, indicating
that the specimen is relatively mature (Irmis, 2007).
Haemal arches. Posterior to the sacrum, no vertebrae are clearly discernable. The cloacal
vent does reveal two pairs of haemal arches (chevrons), each in articulation, and there is
indication of the caudal vertebrae under (dorsal to) these (Figs 2.1–2.2, 4.3, 5.3). The haemal
arches are each at least 45 mm long and as much as 20 mm across proximally. They are
typical of such structures generally and extremely similar in size and shape to the one such
element preserved with PVL 2073, the holotype of Aetosauroides scagliai.

10
Pelvis
The pelvis is somewhat disarticulated due to post-mortem crushing. The right ilium is
missing or else buried under matrix and other elements and almost none of the left ilium is
distinct, but the left pubis is visible in two pieces and the right pubis is similarly visible, both
in essentially lateral view (Fig. 2.1–2.3). Much of the ischia are visible, albeit in ventral view
and somewhat obscured by osteoderms (Figs. 2.1–2.2, 3, 4.5, 5.5).
Ilium. Portions of the left ilium are preserved on both sides of the large crack that runs
oblique to the long axis of the skeleton, the right ilium is not visible (Figs. 2.1–2.2, 3, 4.2,
5.2). Anterior to the fracture some of the lateral (ventral) face of the ilium may be visible
below the femur, but most of the bone visible here probably pertains to osteoderms exposed
in ventral view. As preserved, the acetabulum surrounding the femur appears somewhat
ventrally directed (or at least not vertical), although this could be a taphonomic artifact.
Posterior to the fracture we interpret most of the exposed bone to be part of the ilium based
on its relative position to the ischia and femur. If this interpretation is correct, then the
postacetabular blade appears to be relatively long and tabular, not short as is typical of
aetosaurs (e.g., Walker, 1961). It therefore appears somewhat similar to that of
Revueltosaurus (Parker et al., 2005, fig. 3c), although it may simply be fractured and the
visible portion is the postacetabular blade. It is possible that some of this bone is part of a
sacral rib visible in ventral view (Figs. 4.2, 5.2).
Pubis. The proximal portion of the left pubis abuts the femoral head and preserves much of
the obturator plate—the distal portion is displaced by a large crack but is otherwise preserved
in lateral view (Fig. 2.1–2.3). The iliac and ischiatic facets are subequal in size (each
approximately 16 mm across). The dorsal margin of a relatively large obturator foramen is
preserved, although the ventral margin is lacking, so the presence of a second fenestra as
described in Stagonolepis and Aetosaurus by Walker (1961) cannot be evaluated (Figs. 2.3, 4

11
.5, 5.5). Both the left and the right pubis appear to preserve a medial “tongue” similar to that
of Polesinesuchus (da-Silva et al., 2014), visible at the edge of the specimen near the contact
with the ischium on the right side and medial to the obturator plate on the left (Figs. 4.5, 5.5).
The obturator plate is broken away on the right side. As is typical in aetosaurs, the element
narrows to a thin shaft before twisting and flaring medially and distally, this can be seen in
postero-ventral view on the right side. The right side indicates a minimum length from the
acetabulum to this boot as ~80 mm, and, based on the left side, it could be as much as 95
mm. The pubic apron is offset by the large oblique crack, but may have been fused as there
are bone fragments that may be part of the left pubis on the right side.
Ischium. Much of both ischia are exposed in ventral view, but obscured by some matrix and
two columns of small ventral osteoderms (Figs. 2.1–2.2; 3; 4.5, 5.5). The ischiatic plate is
broad (at least 85 mm across at the acetabula) and flares laterally to as much as 110 mm wide
before tapering posteriorly to a rounded terminus ~15 mm across. The element is
approximately 72 mm long, with 65 mm of this apparently fused along the midline in a
fashion similar to that described by Walker (1961) for Stagonolepis robertsoni.
Hind limbs
Large portions of both hind limbs are preserved, including all or parts of both pairs of
femora, tibiae, fibulae, and pedes (Figs. 2.1–2, 3, 4.1–2, 4.4, 5.1–5.2, 5.4). Appendicular
armor, hematitic concretions, and fractures complicate description, as does the articulation of
the specimen. Measurements follow the protocols of Martz (2002, fig. 4.13) for consistency
with his and subsequent analyses (e.g., Heckert et al., 2010) and are presented in Table 1.
Femur. The left femur is exposed, slightly displaced from the hip socket, in more-or-less
antero-medial view proximally and antero-dorsal view distally (Figs. 4.2, 5.2). The right
femur is almost completely covered by osteoderms and probably lacks a head (Figs. 4.1, 5.1).
Given the extensive covering of osteoderms, it is probably twisted to show a more lateral

12
view, although the distal end is visible in antero-dorsal view. The left femur is relatively
gracile, more than 150 mm long, and strongly twisted, with the twist occurring relatively far
down the shaft, as indicated by a change in the orientation of the crest running down the
anteromedial surface of the shaft approximately 90 mm from the femoral head (Table 1). The
fourth trochanter is not visible for taphonomic reasons and appears to have been damaged in
the fracture distal to the femoral head.
Tibia. The left tibia is preserved but split by the “fault-like” fissure so that the medial portion
of the shaft and distal end are not in contact with the proximal end (Figs. 4.2, 5.2). Length
estimates are approximately 120 mm (Table 1). On the right side the situation is better, but
with most of the tibia visible in anteromedial view (Figs. 4.1, 5.1). The proximal end is
robust, but covered by a thin screen of appendicular osteoderms. The midshaft is nearly
triangular in cross-section. The distal ends of both tibiae are articulated with their respective
astragalus, but details are obscured by hematitic crust and appendicular osteoderms. Desojo
(2005) described the articular surface for the tarsus as flat anteriorly and convex posteriorly.
A break on the right side allowed us to measure the cross-section of the tibia as
approximately 50 mm in circumference, with mid-shaft diameters of approximately 12
(minimum) and 16 (maximum) mm at the narrowest point (Table 1).
Fibula. Parts of the left fibula are visible, again on opposite sides of the fissure, the shaft and
iliofibularis tubercle covered by appendicular osteoderms (Figs. 4.2, 5.2). The right side is
better, with most of the distal shaft visible (Figs. 4.1, 5.1). The iliofibularis tubercle is
covered by a mass of concreted matrix and the proximal end is obscured by appendicular
osteoderms. The cross-section of the fibula is approximately 33 mm in circumference with a
minimum diameter of 8.3 mm and maximum of 10.5 at this point.
Pes. Parts of both feet are preserved. The left is visible in dorsal (plantar) view and the right
in ventral (palmar) view (Figs 2.1–2.2, 3, 4.1–4.2, 4.4, 5.1–5.2, 5.4). Both are heavily

13
encrusted in hematite. Table 1 includes the measurements we were able to determine given
the state of preservation. On the left side parts of the astragalus, four metatarsals (I–IV), and
at least one phalanx are preserved (Figs. 4.2, 5.2). The medial metatarsal (I) lacks a distal
end. A single relatively short (~10 mm) but incompletely preserved phalanx is articulated
with metatarsal II. There is a distinct gap between the distal tibia and the astragalus on the
right side (Figs. 4.2, 5.2).
The right pes is more easily seen from multiple angles (Figs. 4.1, 4.4, 5.1, 5.4). Parts
of the astragalus and calcaneum are preserved in posterodorsal view, including the calcaneal
tuber in dorsal view. The medioventral surface of the astragalus is also visible in ventral
view. None of the other tarsals are distinct. The astragalus is distinct from the distal tibia
(Figs. 4.4, 5.4). Metatarsal I, the proximal phalanx, and an ungual are exposed in both ventral
and dorsal views. Metatarsals II and III are visible in ventral view, as is a proximal phalanx
articulated with II. Fragments of the proximal and distal ends of metatarsal IV are evident on
the lateral surface of III. A relatively well preserved ungual probably belongs to digit II.
There is evidence of at least two phalanges and, possibly, an ungual for III as well. The
phalangeal formula for the first three digits (2-3-3+) therefore matches other aetosaurs where
the foot is known (most considered 2-3-4-5-x, where x ranges from 2–4) (Walker, 1961;
Desojo & Báez, 2005; Schoch, 2007; Heckert et al., 2010; Lucas & Heckert, 2011). The
more distal phalanges are always considerably shorter than the next most proximal element,
usually by a factor of 50%, except for the unguals, both of which are approximately 25 mm
long in side view (with articular facet oriented vertically) but 30–35 mm long along the curve
of the claw. The unguals are somewhat compressed, but clearly triangular in cross-section,
with the base of the triangle situated ventrally and the apex dorsally.
Osteoderms
Preserved osteoderms associated with the primary block include parts of the dorsal,

14
lateral, ventral, and appendicular carapace (Fig. 2.1–2.2). Portions of several dorsal
paramedians are visible in separate pieces (e.g., Fig. 2.4). Two precaudal and many caudal
lateral osteoderms are visible in external view. The ventral surfaces of five pairs of proximal
caudal dorsal paramedians are visible in ventral view. Ventral osteoderms consisting of both
less organized pairs (over the ischia) and two well-developed columns from the base of the
tail posteriorly, are also visible. Articulated appendicular osteoderms cover much of both
hind limbs, but are especially well-preserved and articulated on the right side.
Dorsal Paramedian Osteoderms. Anterior to the well-preserved and articulated material,
one block of matrix yields portions of at least one vertebra and several paramedians (Fig.
2.4). As many as five paramedian osteoderms are visible in dorsal view or cross section (Fig.
2.1–2.2, 2.4). These appear to preserve part of the lateral area and are characterized by a
prominent anterior bar and strongly radial ornamentation of elongate ridges and grooves.
Although clearly radial, the pattern is relatively subdued, and not deeply incised
All other paramedian osteoderms presently visible are exposed solely in ventral
(internal) view. On the right side of the tail immediately posterior to the pelvis there are five
essentially intact paramedians, although probably less than half of the ventral surface is
exposed (Figs. 2.1–2.2, 3.1). They all bear arcuate lateral margins, and none appear to be
keeled. They do not quite align with the ventrals (described later), as they outnumber them
5:4 before being covered by preserved laterals.
Dorsal Lateral Osteoderms. Two of the osteoderms exposed on the anatomical left side
anterior to the pelvis may pertain to laterals (Fig. 2.1–2.3). The more anterior preserves the
lateral flange and may be associated with the anteriormost preserved dorsal vertebra. The
more posterior also appears to be the lateral flange, but with a prominent anterior process. It
may be a sacral lateral. The pattern on the first is strongly radial, emanating from the ridge
along the straight flexure between the missing medial and preserved lateral flanges.

15
Unfortunately, no homologous osteoderms are known from the holotype of Aetosauroides
scagliai (PVL 2073), so comparisons are difficult.
Caudal Lateral(?) Osteoderms. Osteoderms we interpret as lateral osteoderms are present
on the left side in the vicinity of the cloacal vent (Figs. 4.3, 5.3) and on the right side further
posteriorly (Fig. 4.6). These could conceivably be ventral lateral osteoderms, but we interpret
them as laterals because they are not necessarily out of position and could easily be
articulated with paramedians. The gap between them from the ventral paramedian osteoderms
is also larger than the midline gap that separates the left and right column of ventral
osteoderms (Fig. 4.6). They are also strongly keeled and deeply pitted, more so than the
nearby ventrals (described later). The longitudinal keel bifurcates the osteoderms into
roughly subequal lateral and medial flanges, and is typical of lateral osteoderms but not
ventral osteoderms in aetosaurs with ventral osteoderms (Walker, 1961; Heckert et al., 2010).
Walker (1961) and Schoch (2007) have indicated that armor in the tail of Stagonolepis
robertsoni and Aetosaurus ferratus forms tight bands wherein the lateral and ventral
osteoderms are practically articulated, and we feel that this is the case here, which is why the
“lateral” margin of some of the laterals appear to match the lateral margin of a corresponding
ventral. The osteoderms we describe as caudal laterals thus correspond more to the
morphology of known plesiomorphic aetosaur lateral osteoderms than they do ventral
osteoderms.
The first five preserved laterals on the anatomical right side are fragmentary, and
coincide with caudal dorsal paramedian and ventral osteoderm rows 6–10, although they
match the dorsals better than the ventrals (Fig. 2.1–2.3). Laterals from row 10 posteriorly
well match the ventrals (essentially articulate with them) as well (e.g., Fig. 4.6). The
ornamentation on these laterals consists primarily of pits of subequal size emanating from the
ridge or keel. The pits are not distributed particularly radially, however, and correspond more

16
closely with the “anastomosado” (anastamosing) pattern identified by Taborda (2011;
Taborda et al., 2015). The ridge is straight and divides the osteoderms more or less equally,
although the median portion is marginally larger and more rectangular. The lateral margin is
more trapezoidal, with a narrow anterior side and a wider posterior margin.
Ischiatic Ventral Osteoderms. There are 10 osteoderms preserved on the ischia that
represent at least five and perhaps as many as eight pairs, but are considered distinct from the
ventral carapace here, in part because they are much smaller and do not correspond to
vertebrae, as is typical of other osteoderms (Fig. 4.5). They are of similar size, or smaller
than, most appendicular osteoderms (10–20 mm maximum dimension), but more ovate and
less polygonal, so we think it unlikely that they are translocated laterals or appendicular
osteoderms. All are pitted and have a slightly raised eminence, or boss, in the center. The pits
are small, subcircular, and not obviously distributed in any sort of pattern. Smaller
osteoderms have fewer than 10 pits, the larger ones often have 15 or more. The anteriormost
may have an articular surface, but most do not appear to possess this feature. Similar
osteoderms are were described, but not illustrated, on the holotype of Coahomasuchus
kahleorum (Heckert & Lucas, 1999, p. 57).
Caudal Ventral Paramedian Osteoderms. The only unambiguously preserved articulated
ventral osteoderms are those of the caudal paramedian series (Figs. 2.1–2.3, 3, 4.3, 4.6, 5.3).
These include six anterior pairs (12 preserved osteoderms) that stretch from the posterior
margin of the ischia to the posterior margin of the cloacal opening (Figs. 4.3, 5.3). All or
parts of 14 more pairs are preserved posterior to this (Fig. 4.6). We rely primarily on the right
side, as these osteoderms are better exposed there—the left side tends to be less well
preserved and/or covered by lateral osteoderms. All are ornamented by pits emanating from a
center of ossification that ranges from a pronounced longitudinal keel (typical of the more
anterior osteoderms) to a barely discernable bump posteriorly. Where present, keels run

17
diagonally across the ornamented surface, such that they begin anterolaterally and end at or
near the middle of the posterior margin of the osteoderm. The surface texture is also variable,
sometimes consisting of irregularly shaped pits delineated by thin, anastomosing ridges
(especially on the first two pairs) and other times consisting of more regular, ovoid pits
separated by thicker ridges. The most posterior osteoderms have elongated grooves and
ridges. All are marked by an anterior articular surface (“anterior bar”) that is more or less
flush with the pitted surface. Anteriorly this articular surface is thin—3 mm or less, but in the
4–5th osteoderms it is as much as 10 mm long and appears to have subsumed some of the
pitting.
Most of osteoderms in the first six rows of caudal ventral paramedians are in some
way shaped to accommodate the cloacal vent (V1-V6 on Figs. 4.3, 5.3). The anteriormost
pair probably coincides with the first caudal vertebra (which is not visible) and is poorly
preserved, with the right directed laterally toward the hind limb and the left covered in part
by a lateral osteoderm. Both are strongly pitted. The second pair is better preserved,
especially on the right. There is a thin anterior bar and a pronounced ornamentation of
irregularly shaped, densely packed, pits. The lateral margin is sinuous—concave anteriorly
and convex posteriorly. The third pair is trapezoidal in outline, with the broad end anterior
and tapering posteriorly. The fourth pair is proportionately much longer, with a much larger
articular surface (~20% instead of ~10% of the total length) and bears a more pronounced
keel than do any of the preceding osteoderms. The fifth pair is the longest of the ventral
osteoderms, but is displaced and/or missing on the left side. The preserved right osteoderm
appears superficially heptagonal (seven-sided) as the medial margin is angulated to
accommodate the cloacal vent. The sixth pair is the last pair modified for this purpose and is
the broadest. It also has a faintly radial aspect to the pitting.
Posterior to the cloacal vent, the ventral paramedian osteoderms more strongly

18
resemble dorsal paramedian osteoderms. The first pair of post-cloacal osteoderms (V7 on
Figs. 4.3, 5.3) is badly damaged, but each is wider than long and appears to bear at least some
radial grooves in addition to the pits. The eighth pair is also very broad and extensively pitted
(Figs. 4.3, 5.3). There is a low, rounded boss situated near the middle of the osteoderm.
Posteriorly the boss diminishes in size and is barely discernable four rows more posteriorly,
although the pits do appear to emanate from it, so it is like the “center of ossification” (e.g.,
Martz, 2002; Cerda and Desojo, 2011; Desojo et al., 2013; Taborda et al., 2013) on dorsal
paramedian osteoderms. For the next seven pairs, the lateral margins taper posteriorly, so that
the tail narrows rapidly even as the length of the osteoderms remains relatively constant
(Table 2).
The left and right ventral paramedian osteoderms of the last five rows (beginning with
the 16th row) are actually fused together, a feature we have not previously seen in a ventral
carapace (Figs. 2.1_2.2, 3, 4 .6), although similar fusion has been reported in dorsal
paramedian osteoderms of Calyptosuchus by Parker (2018b). The bosses are slightly further
apart anteriorly and very close together posteriorly. The first of these conjoined osteoderms
tapers rapidly from ~50.9 m wide immediately behind the anterior bar to 44.3 mm along the
posterior margin, but subsequent osteoderms do not shrink as dramatically, although each
pair narrows more rapidly than they get shorter (Table 2). We discuss this further in the
section on comparisons.
Appendicular Osteoderms. The right hind limb of PVSJ 691 preserves an extensive coating
of appendicular osteoderms over the femur, tibia, and fibula. Approximately 35 largely
imbricated osteoderms so thoroughly cover the proximal hindlimb that most of the femur is
completely obscured (Figs. 2.1–2, 3, 4.1, 5.1). Similarly at least 20 more osteoderms cover
much of the proximal tibia and fibula. The osteoderms covering the femur tend to be slightly
larger than the more distal osteoderms. The shapes of appendicular osteoderms vary from

19
rhomboid to irregular, but most, especially over the femur, are somewhat quadrangular. All
are ornamented to some extent, mostly with subcircular pits and some more elongate ridges
which appear to emanate from the center of the osteoderm. The proximal osteoderms
(covering the femur) are generally better-preserved and more distinct than the more distal
osteoderms covering the tibia and fibula. The latter are not preserved over the distal tibia and
fibula, with only a few evident distal to the iliofibularis process on the fibula on the right
side. However, many small appendicular osteoderms are visible covering the distal tibia and
fibula on the right hindlimb, covering the ~anterior surface (Figs. 2.1–2.2, 3, 4.2, 5.2).
Many of the osteoderms over the femur preserve an elongate keel that runs proximo-
distally, and bisects the osteoderm along its long axis. This ridge or keel, where present,
tends to run parallel to the long axis of the underlying bone. In at least some cases this keel is
more pronounced distally. Each side typically bears many (> 10) small pits, some of which
are more elongate and thus appear to be radiating out from the ridge. The most medial series
of osteoderms appears to comprise a proximal-distal series of seven osteoderms, but the
others are not as readily traced, although there are clumps of three to four osteoderms,
suggesting that 6–8 osteoderms constituted a single proximo-distal series over the proximal
hind limb.
COMPARISONS
Although details of the morphology of paramedian and lateral osteoderms are often
used to identify different aetosaur genera (e.g, Heckert & Lucas, 2000), these are generally
only useful in the more derived typothoracisine and desmatosuchine aetosaurs (Parker, 2007,
2016a,b, 2018a,b; Desojo et al., 2013), as basal aetosaurians are typically characterized by
relatively plesiomorphic armor. Most South American taxa have osteoderms broadly similar
to that of PVSJ 691, in that most dorsal paramedians are wider than long, bear a dorsal
eminence consisting of a boss or longitudinal keel, and a pattern of radially oriented pits,

20
ridges, and grooves radiating from that eminence (e.g., Heckert & Lucas, 2002; Taborda et
al., 2013, 2015). PVSJ 691 is distinct from Coahomasuchus in that the ridges and grooves in
the paramedian armor are more clearly radial, and the caudal ventral carapace apparently
consists of only two columns of caudal ventral osteoderms, unlike the four in Coahomasuhus
(Heckert & Lucas, 1999). Details of the cloacal armor (which could, of course, be dimorphic
within taxa) and the possible presence of only two columns of caudal ventral osteoderms
distinguish PVSJ 691 from Aetosaurus (Walker, 1961; Schoch, 2007). The apparent presence
of only two columns of ventral osteoderms also distinguishes PVSJ 691 from the Elgin
aetosaur tentatively assigned to Stagonolepis by Keeble and Benton (2020). Because the
lateral osteoderms are not much broader than they are long they are also distinct from
Neoaetosauroides (Bonaparte, 1967).
Among the other basal aetosaurs, the preserved armor of PVSJ 691 is either not
distinct, or else there is no overlap of homologous osteoderms, so taxonomic determinations
must necessarily rely on other elements. Desojo and Ezcurra (2011) demonstrated that the
dorsal vertebrae of Aetosauroides are distinct from those of Stagonolepis, and
contemporaneous taxa named since then (Aetobarbakinoides and Polesinesuchus; Desojo et
al., 2012; da-Silva et al., 2014) also have distinct suites of vertebral characters. Key to
Desojo and Ezcurra’s (2011) description of Aetosauroides are the presence of a lateral fossa
on the dorsal vertebrae as well as an infradiapophyseal lamina with corresponding fossae on
either side. These are not present in Aetobarbakinoides (Desojo et al., 2012) or
Polesinesuchus (da-Silva et al., 2014).
Although only two posterior dorsals are preserved in lateral view in PVSJ 691, both
lack a conspicuous lateral fossa but appear to possess an infradiapophyseal lamina. Well-
rimmed lateral fossae are clearly evident on the holotype of Aetosauroides scagliai (e.g.,
Heckert and Lucas, 2002, fig. 2.11). No such fossae are present on PVSJ 691. The absence of

21
lateral fossae distinguishes PVSJ 691 from Aetosauroides (Desojo & Ezcurra, 2011), and the
centra of PVSJ 691 are also proportionately taller and less laterally constricted than those of
the holotype of Aetosauroides (PVL 2073). Irmis (2007) considered the presence of fossae in
phytosaur centra as somewhat ontogenetically variable, but PVL 2073 and PVSJ 691 have
centra of approximately the same size, so it seems unlikely tht this difference would be the
result of ontogenetic variation. Desojo and Ezcurra (2011) also considered the presence of an
infradiapophyseal lamina an autapomorphy of Aetosauroides scagliai, although its possible
presence is much harder to evaluate in PVSJ 691, as the neural arches are only partially
exposed and preserved in D1 and D2 (Fig. 2.3).
Although osteoderms are largely unknown in Polesinesuchus, the known vertebral
series overlaps with that preserved in PVSJ 691. Generally, the centra of Polesinesuchus,
which are similar in that they also lack a well-developed fossa, differ from PVSJ 691 in that
they lack an infradiapophyseal lamina and consist of dorso-ventrally shorter centra (e.g., da-
Silva et al., 2014, p. 251). Aetobarbakinoides also lacks lateral fossae and has only incipient
to absent infradiapophyseal laminae (e.g., Desojo et al., 2012).
Both the holotype of Aetosauroides scagliai (PVL 2073) and PVSJ 691 preserve
much of the hind limbs, and both specimens are of roughly similar size. Table 1 shows a
comparison of measurements of the femora, tibiae, and fibulae of these specimens using
protocols established by Martz (2002, fig. 4.13) in his thesis and frequently used by other
aetosaur workers (e.g., Heckert et al., 2010). These demonstrate that, although PVSJ 691
possesses elements that are approximately the same length as their counterparts in PVL 2073,
the PVSJ 691 elements are almost always more robust, with greater widths proximally and
distally and thicker shafts. The other articulated Ischigualasto Formation specimen of
Aetosauroides, PVL 2059, appears to be wider, and thus likely more robust, than PVL 2073,
possibly as a result of sexual dimorphism (e.g., Taborda et al., 2015), but lacks hind limb

22
elements that could be used for comparisons. All are much smaller than PVL 2052, a large
specimen of Aetosauroides described by Casamiquela (1967). Although difficult to evaluate
due to articulation and taphonomic issues, the twist of the femora on PVSJ 691 occurs more
distally (~60% of the way down the shaft) than occurs in the similarly sized femora of PVL
2073, the holotype of Aetosauroides scagliai (Casamiquela, 1967, fig. 20; Heckert & Lucas,
2002, fig. 3.9–10), where the same twist is evident on the proximal half of the shaft.
Similarly, PVSJ 691 lacks the pronounced sulcus visible on the anterodorsal face of both
femora in PVL 2073 (Figs. 4.2, 5.2; compare to Casamiquela, 1961, fig. 20a). Given that
PVL 2073 is generally better preserved, especially in terms of details of bone texture, we
think it unlikely that the sulci on that specimen are a taphonomic artifact.
The ankle joints appear typical of more basal aetosaurs, in that most of the motion
between the lower hind limb and the pes is accommodated between the tibia-fibula and
astragalus-calcaneum. This is unlike the condition in Typothorax coccinarum, which has a
more mesotarsal articulation (Heckert et al., 2010). This is not to say that there was (or was
not) significant motion accommodated between the astragalus and calcaneum, as PVSJ does
not preserve the astragalus and calcaneum in sufficient detail to ascertain details of their
articulation, we note merely that there are distinct gaps between the astragalus and tibia on
both sides that suggest that the foot was capable of at least some hinge-like movement
between the distal leg (tibia and fibula) and proximal tarsals (astragalus and calcaneum).
The caudal armor of PVSJ 691 is distinct from that of PVL 2073, the holotype of
Aetosauroides scagliai. While the latter has been disarticulated, Casamiquela (1967,
lamina I) appears to show the posterior portion of PVL 2073 in ventral view after initial
preparation but prior to the separation of the osteoderms from the rest of the skeleton. In this
photo, the caudal ventral osteoderms are strongly deflected to the animal’s left side, which
matches a general arc in the preserved articulated dorsal and lateral osteoderms, now

23
mounted as a separate block. Four columns of ventral paramedian osteoderms are apparent,
with three columns represented by multiple osteoderms—the right lateral ventral osteoderms
appear to be only impressions. Each osteoderm is wider than long. The two medial columns
have broader osteoderms than the more lateral ones. At least six rows are preserved in the
medial columns. Each osteoderm preserves an elongate keel, this is especially pronounced
anteriorly (in the first three rows). A somewhat radial (intermediate of Taborda et al., 2015)
pattern of grooves and ridges emanates from that ridge. On at least some of the osteoderms it
appears that the ridge marking the center of ossification is situated lateral to the middle of the
osteoderm. Importantly, there is no obvious accommodation for the cloacal vent. We
hypothesize that in PVL 2073 the armor “unzips,” with the cloaca lying between the three
most anterior pairs. We cannot rule out that additional, more anterior, osteoderms were
removed prior to this photograph, but our interpretation (Fig. 6; Supp. Fig. 1) is based on
what is evident in Casamiquela (1967, lamina X). Even if some osteoderms were removed,
the remaining osteoderms differ from those of PVSJ 691 in terms of shape, position of the
center of ossification, apparent accommodation of the vent, and other details. The first two or
three caudal vertebrae are visible in ventral view, but it is not clear which, if any, of the
caudal ventral osteoderms may be associated with these osteoderms. At least a few of the
osteoderms illustrated in the proximal tail by Casamiquela (1967, lamina I) are still present,
but disarticulated, in the PVL collections (ABH, pers. obs.). Other, irregularly shaped
osteoderms associated with PVL 2073 presently do not appear to match those and so
probably are not cloacal ventral osteoderms and instead may represent more distal caudal
ventral osteoderms.
DISCUSSION
Although PVSJ 691 is incomplete, it is arguably the third-best preserved aetosaur

24
specimen from the Ischigualasto Formation behind the holotype of Aetosauroides scagliai
(PVL 2073), which consists of much of a postcranial skeleton, and PVL 2059, which
preserves the best aetosaur skull from the formation and an incomplete carapace (Taborda et
al., 2015). As such, PVSJ 691 provides insight into the pelvis, caudal morphology, and
cloacal vent, all of which have paleobiological implications. This is especially important as
intraspecific variation in aetosaurs is poorly understood (aside from the spectacular block of
Aetosaurus in Germany—Schoch, 2007), and only recently explored in South American taxa
(e.g., Desojo & Ezcurra, 2011; Taborda et al., 2015) and other basal stagonolepids (Antczak
2016). Furthermore, while Aetosauroides and other South American aetosaurs are often
recovered as among the earliest-diverging aetosaurs (e.g., Desojo et al., 2013), none are as
well known from the posterior skeleton. Thus, if PVSJ 691 represents a specimen of
Aetosauroides, it provides insight into the ancestral condition of numerous pelvic, hind limb,
and caudal features. This is especially important as other taxa provide insight into the
immediate outgroups to Aetosauria, including Revueltosaurus and Acaenasuchus (e.g.,
Nesbitt, 2011; Marsh et al., 2020).
Pelvis. As in several other aetosaurian taxa, the pelvis of PVSJ 691 appears to be oriented
subvertically. This is true in Aetosauroides (in both the holotype PVL 2073 and the larger
PVL 2052) and Typothorax (both T. antiquum Lucas et al., 2002 and T. coccinarum, see
Heckert et al, 2010). We cannot completely rule out some crushing, so it is possible that the
actual life position of the acetabula was more laterally directed, but the relatively intact sheet-
like ischia suggest that there has been little displacement of the preserved pelvic elements.
The acetabula, as indicated by the margins of the ischia, are at least 85 mm apart in PVSJ
691, and only 100 m apart in the much larger specimen of T. coccinarum described by
Heckert et al. (2010). The holotype of Aetosauroides scagliai, PVL 2073, has ilia that are as
little as 27 mm apart, although measurements of the distorted ilia using the same landmarks

25
preserved on PVSJ 691 yields an intracetabular width of at least 40 mm.
Appendicular osteoderms. Appendicular osteoderms are known from many aetosaurs,
including specimens assigned to Aetosauroides from the Ischigualasto Formation (e.g.,
Casamiquela, 1960, 1961, 1967), but most reports are of a few osteoderms, so they lack the
formal nomenclature that characterizes other osteoderms. The right hind limb of PVSJ 691
preserves the most extensive and/or complete series of appendicular osteoderms known. This
suggests that at least some aetosaurs had similar, dense, imbricated and closely articulated
sets of appendicular osteoderms covering the hind limb, as Drodz (2018) demonstrated for
the forelimb of Stagonolepis olenkae. Similar osteoderms were reported (PVL 2073,
Casamiquela, 1961, fig. 25; PVL 2052; Casamiquela, 1967, lamina XV), including a few on
the holotype right humerus of “Argentinosuchus bonapartei” (PVL 2091; Casamiquela, 1961,
fig. 27d), and others are preserved loose in the PVL collections associated with the holotype
of Aetosauroides scagliai (ABH pers. obs.). We suspect that PVL 691 preserves armor that is
more typical of early aetosaurs, and that appendicular armor on other aetosaur specimens is
either not preserved due to scavenging and other taphonomic processes, or else has been
removed to facilitate preparation and examination of the underlying elements.
Caudal morphology. Relatively few articulated tails of aetosaurs have been described and
illustrated in multiple views. Thus, most descriptions rely on Walker’s (1961) treatment of
Aetosaurus (followed by Schoch, 2007, who did not discuss the tail in great detail). Although
incomplete, the tail of PVSJ 691 is distinct from that of most other aetosaurs in that its width
tapers rapidly, yet the individual osteoderms do not appear to be getting longer. Most
restorations of aetosaurs show distal caudal osteoderms that get more elongate than wide
(e.g., Casamiquela, 1967, Lámina XV; Desojo et al., fig. 3).
Of aetosaurs for which ventral caudal armor is known, PVSJ 691 is nearly unique in
that it appears to possess only two columns of ventral osteoderms. Most other taxa, including

26
Aetosaurus, Stagonolepis robertsoni, Coahomasuchus, and Typothorax all exhibit at least
four columns (Walker, 1961; Heckert & Lucas, 1999; Heckert et al., 2010) (Fig. 6). Recently
Keeble and Benton (2020, figs. 2-3) published CT imagery of an articulated, albeit distorted,
tail from the Lossiemouth Sandstone near Elgin, Scotland that likely pertains to Stagonolepis.
Although distorted, and incomplete, this tail preserves at least eight total columns of
osteoderms. Lucas et al. (1998, text-fig. 4) illustrated a block preserving part of the tail of an
aetosaur (NCSM 11756) that they referred to Aetosaurus from the Upper Triassic of North
Carolina, USA but whose taxonomic status is presently uncertain (Desojo et al., 2013). Lucas
et al. (1998) did not describe NCSM 11756 in detail, and there is a taphonomic artifact of this
specimen that makes determination of the number of preserved osteoderms difficult, but the
dorsal surface preserves two columns of paramedians, and the ventral surface four columns
of osteoderms, so this specimen appears to possess only six columns of osteoderms as well.
We tentatively hypothesize that PVSJ 691 exhibits the ancestral condition of only six total
rows of caudal osteoderms (two paramedian, two lateral, two ventral).
The fusion of the two medial columns of ventral osteoderms beginning at the 16th row
of ventral paramedian osteoderms is also a feature of PVSJ 691 that has not been
conclusively documented previously. This fusion apparently began relatively early in
ontogeny, as there are two relatively distinct bosses (centers of ossification), one on either
side of the midline suture, but these are more medially situated than the bosses on the more
anterior osteoderms. Consequently, the pattern of pitting on the lateral margin is more
elongate, consisting of elongate grooves and ridges radiating from the bosses. We see no
indication that it is pathological (either from injury or developmental irregularity), as there is
no loss of symmetry between left and right elements, and the fused surface is regular and
clearly delineated, suggesting that the fusion is the result of a normal process. This is
therefore distinct from the appearance of fusion in ventral paramedian osteoderms assigned to

27
Stagonolepis by Keeble and Benton (2020), who considered the appearance of paired ventral
osteoderms in CT-scans as “likely more due to taphonomic processes…rather than their
being fused in life” (Keeble & Benton, 2020, p. 58) or the dorsal paremedian osteoderms of
Calyptosuchus Parker (2018b) described as fused.
The fusion appears to have taken place earliest in the caudal-most preserved
osteoderms and is somewhat less well marked anteriorly. This follows the pattern of
neurocentral fusion (of vertebral centra to their respective neural arches) in crocodylians
(Brochu, 1996) and parasuchians (Irmis, 2007), and thus may further indicate that PVSJ 691
was beginning to attain skeletal maturity, although onteogenetic stages based on the fusion of
postcranial elements are particularly difficult to interpret in saurians (Griffin et al., 2020).
Heckert and Lucas (1999) speculated that the distal tip of the tail they illustrated for
Coahomasuchus was composed of paired columns of paramedian and lateral osteoderms and
only a single ventral column. This would appear to be validated by PVSJ 691, which is the
first aetosaur we have seen to definitively have fused two columns of osteoderms into a
single column. This also suggests that distally the tail of Coahomasuchus is reduced from 8
columns (two dorsal paramedian, two lateral, four ventral) to six (presumably by the loss of
ventral osteoderm columns, probably ventral lateral osteoderms) and eventually five by the
fusion of the midline ventral columns.
Parker (2018b, fig. 17a–b) described and illustrated fused caudal osteoderms of the
holotype of Calyptosuchus wellesi. Based on his descriptions and the figure captions, these
are dorsal paramedian osteoderms, unlike the ventral paramedian osteoderms of PVSJ 691.
Furthermore, the two “fused” osteoderms he illustrates are still clearly distinct elements
bound by a midline suture (MLS in Parker, 2018b, fig. 17a–b), and essentially identical,
minus the suture, to a similar individual osteoderm. In PVSJ 691, the ventral paramedians are
so thoroughly fused together that their symmetry is the only indication of where the midline

28
suture would be. Parker (2018b) indicates that the fusion in Calyptosuchus is similar to that
of a paratype specimen of Scutaryx deltatylus (e.g,. Parker 2016b, fig. 25), although we
cannot discern details of the fusion in Scutaryx from either the text or figures of Parker
(2016b).
Cloacal vent. Because of their extensive armor, aetosaurs are sometimes difficult to interpret
as the armor obscures morphological details of the underlying bones. However, the armor
also has to correspond to details of the soft tissues, and in this way can yield paleobiological
insights not obtainable from other bones. One such example of this is the cloacal vent, which
has now been described in Aetosaurus ferratus (Walker, 1961), Coahomasuchus kahleorum
(Heckert & Lucas, 1999) and Typothorax coccinarum (Heckert et al., 2010). In PVSJ 691 the
vent has experienced some post-mortem disruption but, minimally, it was 130 mm long. It
may have been as much as 155–180 mm long based on the interpretation of other osteoderms,
but we prefer the more conservative, lower number. The maximum width as preserved is
approximately 50 mm. Relative to other taxa, PVSJ 691 involves the most osteoderm rows
(6) and, proportionately, has a larger apparent opening. Although we cannot be certain that
Casamiquela (1967) illustrated the entirety of the armor in the cloacal region for PVL 2073,
the morphology of that specimen is distinct from PVL 691, and appears to show a simple
“unzipping” of the ventral armor to accommodate the vent proximally. Whether these
differences are taxonomically significant is difficult to ascertain, as sample size is so low that
we cannot rule out dimorphism (although both individuals of Typothorax coccinarum for
which the vent is intact exhibit identical morphology of the surrounding osteoderms—
Heckert et al., 2010). Still, it is tempting to hypothesize that the condition exhibited by PVL
2073, with the vent accommodated by fewer rows of osteoderms, including none proximally,
would be closer to the ancestral condition, and that the expansion of the vent to a structure
accommodated by more specialized osteoderms in other taxa as in most other aetosaurs (Fig.

29
6), is an indication that PVSJ 691 does not exhibit the plesiomorphic condition.
Relatively little of the dorsal paramedian osteoderms of PVSJ 691 are exposed, but
what is visible appears extremely similar to the morphology seen in PVL 2073, the holotype
of Aetosauroides scagliai. However, the vertebrae, many features of the limb bones, and
details of the caudal armor provide preliminary evidence that PVSJ 691 is distinct from the
similarly-sized holotype of Aetosauroides scagliai and thus may represent a different taxon.
Each of these features could be due to individual variation, and Taborda et al. (2015) posited
that PVL 2059 and PVL 2073 might be sexual dimorphs, but the consistent differences
between PVSJ 691 and PVL 2073 cause us to caution workers that there may in fact be more
than one taxon of aetosaur in the Ischigualasto Formation. Without further preparation, and
knowing that there is little anterior skeleton preserved, we consider the morphological
differences documented here insufficient to warrant erection of a new taxon name for PVSJ
691, but it supports the possibility that there may be additional taxonomic diversity in the
Ischigualasto Formation. This would not be unexpected, as most basins that yield aetosaur
fossils, including the broadly contemporaneous Santa Maria sequence in Brazil, preserve
representatives of multiple taxa (Desojo et al., 2012, 2013; da Silva et al., 2014).
CONCLUSIONS
Because of the rarity of articulated aetosaurs, even an incomplete specimen such as
PVSJ 691 has the potential to yield new insights into aetosaur morphology. The right hind
limb preserves the most complete set of appendicular armor known, and we are the first to
illustrate the small osteoderms over the ischia and fused mid- to distal caudal osteoderms.
The apparent presence of only two columns of ventral caudal osteoderms may be a
plesiomorphic condition for aetosaurs. The large cloacal opening surrounded entirely by
ventral paramedian osteoderms differs from the holotype of Aetosauroides scagliai and is
more similar to later-appearing aetosaurs. The fusion of midcaudal ventral osteoderms at the

30
midline to form one double-keeled column has not been documented previously, although it
was put forth as a possibility to explain the distal tail configuration of Coahomasuchus
(Heckert & Lucas, 1999).
ACKNOWLEDGMENTS
We thank the Earthwatch Institute and its volunteers who sponsored the 1996 PVSJ
expedition to Ischigualasto. We also thank Eriko Nagao who found the specimen. J.B. Desojo
made one of us (ABH) aware of the specimen and provided constructive comments on an
earlier draft. The specimen was prepared primarily by Raul Gordillo and later by Diego
Abelin. Jaime Powell and Rodrigo Gonzalez (MVL) were extremely helpful in our studies of
PVL material. The map in Figure 1.2 was modified from FreeVectorMaps.com. Portions of
Figure 6 were generated by MDC as part of a project for S.G. Lucas and are used here with
Lucas’ permission. C. Kammerer (NCSM) checked on specimen NCSM 11756 at our
request. We thank two anonymous reviewers for their comments and editor D.G Lazo for
handling the manuscript. Editor Diego Pol handled an earlier version of this manuscript and
provided additional valuable suggestions.
REFERENCES
Antczak, M. (2016). Late Triassic aetosaur (Archosauria) from Krasiejów (SW Poland): new
species or an example of individual variation? Geological Journal, 51(5), 779-788.
Biacchi Brust, A. C., Desojo, J. B., Schultz, C. L., Paes-Neto, V. D., & Da-Rosa, Á. A. S.
(2018). Osteology of the first skull of Aetosauroides scagliai Casamiquela 1960
(Archosauria: Aetosauria) from the Upper Triassic of southern Brazil
(Hyperodapedon Assemblage Zone) and its phylogenetic importance. PLOS ONE,
13(8), e0201450. https://doi.org/10.1371/journal.pone.0201450
Bonaparte, J. F. (1967). Dos nuevas "faunas" de reptiles Triasicos de Argentina. Gondwana
Symposium Proceedings and Papers, 1, 283-306.

31
Bonaparte, J. F. (1978). El Mesozoico de America del Sur y sus tetrapodos. Opera Lilloana,
26, 1-596.
Brochu, C. A. (1996). Closure of neurocentral sutures during crocodilian ontogeny: implications for
maturity assessment in fossil archosaurs. Journal of Vertebrate Paleontology, 16(1), 49-62.
Casamiquela, R. M. (1960). Notica preliminar sobre dos nuevos estagonlepoideos
Argentinos. Ameghiniana, 2, 3-9.
Casamiquela, R. M. (1961). Dos nuevos estagonolepoideos Argentinos de Ischigualasto, San
Juan. Revista Asocíation Geológia Argentina, 16, 143-203.
Casamiquela, R. M. (1967). Materiales adicionales y reinterpretación de Aetosauroides
scagliai (de Ischigualasto, San Juan). Revista del Museo de La Plata, Paleontologia
Serie 5, 33, 173-196.
Cerda, I. A., & Desojo, J. B. (2011). Dermal armour histology of aetosaurs
(Archosauria:Pseudosuchia), from the Upper Triassic of Argentina and Brazil.
Lethaia, 44, 417-428.
Cerda I. A., Desojo J. B., & Scheyer T. M. (2018). Novel data on aetosaur (Archosauria,
Pseudosuchia) osteoderm microanatomy and histology: palaeobiological implications.
Palaeontology, 61(5), 721-745.
Currie, B.S., Colombi, C., Tabor, N.J., Shipman, T.C. and Montañez, I.P. 2009. Stratigraphy
and architecture of the Upper Triassic Ischigualasto Formation, Ischigualasto
Provincial Park, San Juan, Argentina. Journal of South American Earth Sciences 27:
74–87.
Desojo, J. B. (2005). Los aetosaurios (Amniota, Diapsida) de América Del Sur: sus
relaciones y aportes a la biogeografíıa y bioestratigrafía del Triásico continental
[PhD Dissertation, Universidad de Buenos Aires]. Buenos Aires.

32
Desojo, J. B., & Báez, A. M. (2005). El esquelto postcraneano de Neoaetosauroides
(Archosauria: Aetosauria) del Triásico Superior del centro-oeste de Argentina.
Ameghiniana, 42, 115-126.
Desojo, J. B., Fiorelli, L. E., Ezcurra, M. D., Martinelli, A. G., Ramezani, J., Da Rosa, Á. A. S., . . .
Langer, M. C. (2020). The Late Triassic Ischigualasto Formation at Cerro Las Lajas (La
Rioja, Argentina): fossil tetrapods, high-resolution chronostratigraphy, and faunal
correlations. Scientific Reports, 10(1), 12782. doi:10.1038/s41598-020-67854-1
Desojo, J. B., & Ezcurra, M. D. (2011). A reappraisal of the taxonomic status of
Aetosauroides (Archosauria, Aetosauria) specimens from the Late Triassic of South
America and their proposed synonymy with Stagonolepis. Journal of Vertebrate
Paleontology, 31(3), 596 - 609.
Desojo, J. B., Ezcurra, M. D., & Kischlat, E.-E. (2012). A new aetosaur genus (Archosauria:
Pseudosuchia) from the early Late Triassic of southern Brazil. Zootaxa, 3166, 1-33.
Desojo, J. B., Heckert, A. B., Martz, J. W., Parker, W. G., Schoch, R. R., Small, B. J., &
Sulej, T. (2013). Aetosauria: a clade of armoured pseudosuchians from the Late
Triassic continental beds. Geological Society, London, Special Publications, 379,
203-239.
Dróżdż, D. (2018). Osteology of a forelimb of an aetosaur Stagonolepis olenkae (Archosauria:
Pseudosuchia: Aetosauria) from the Krasiejów locality in Poland and its probable
adaptations for a scratch-digging behavior. PeerJ, 6, e5595. doi:10.7717/peerj.5595
Ezcurra, M. D. (2016). The phylogenetic relationships of basal archosauromorphs, with an emphasis
on the systematics of proterosuchian archosauriforms. PeerJ, 4, e1778.
https://doi.org/10.7717/peerj.1778

33
Griffin, C. T., Stocker, M. R., Colleary, C., Stefanic, C. M., Lessner, E. J., Riegler, M.,
Formoso, K., Koeller, K., & Nesbitt, S. J. (2020). Assessing ontogenetic maturity in
extinct saurian reptiles. Biological Reviews, https://doi.org/10.1111/brv.12666
Heckert, A. B., & Lucas, S. G. (1999). A new aetosaur (Reptilia: Archosauria) from the
Upper Triassic of Texas and the phylogeny of aetosaurs. Journal of Vertebrate
Paleontology, 19(1), 50-68
Heckert, A. B., & Lucas, S. G. (2000). Taxonomy, phylogeny, biostratigraphy,
biochronology, paleobiogeography, and evolution of the Late Triassic Aetosauria
(Archosauria:Crurotarsi). Zentralblatt für Geologie und Paläontologie Teil I 1998
Heft 11-12, 1539-1587.
Heckert, A. B., & Lucas, S. G. (2002). South American occurrences of the Adamanian (Late
Triassic: latest Carnian) index taxon Stagonolepis (Archosauria:Aetosauria) and their
biochronological significance. Journal of Paleontology, 76(5), 854-863.
Heckert, A. B., Lucas, S. G., Rinehart, L. F., Celeskey, M. D., Spielmann, J. A., & Hunt, A. P.
(2010). Articulated skeletons of the aetosaur Typothorax coccinarum Cope (Archosauria:
Stagonolepididae) from the Upper Triassic Bull Canyon Formation (Revueltian: early-
mid Norian), eastern New Mexico, USA. Journal of Vertebrate Paleontology, 30(3), 619-
642.
Hoffman, D. K., Heckert, A. B., & Zanno, L. E. (2018). Under the armor: X-ray computed
tomographic reconstruction of the internal skeleton of Coahomasuchus chathamensis
(Archosauria: Aetosauria) from the Upper Triassic of North Carolina, USA, and a
phylogenetic analysis of Aetosauria. PeerJ, 6, e4368.
https://doi.org/10.7717/peerj.4368

34
Irmis, R. B. (2007). Axial skeleton ontogeny in the Parasuchia (Archosauria:Pseudosuchia)
and its implications for ontogenetic determination in archosaurs. Journal of
Vertebrate Paleontology, 27(2), 350-361.
Keeble, E., & Benton, M. J. (2020). Three-dimensional tomographic study of dermal armour
from the tail of the Triassic aetosaur Stagonolepis robertsoni. Scottish Journal of
Geology, 56, 55-62.
Lucas, S. G., & Heckert, A. B. (2001). The aetosaur Stagonolepis from the Upper Triassic of
Brazil and its biochronological significance. Neues Jahrbuch für Geologie und
Paläontologie Monatsheft, 2001, 719-732.
Lucas, S. G., & Heckert, A. B. (2011). Late Triassic aetosaurs as the trackmaker of the
tetrapod footprint ichnotaxon Brachychirotherium. Ichnos, 18(4), 197-208.
Lucas, S. G., Heckert, A. B., & Huber, P. (1998). Aetosaurus (Archosauromorpha) from the
Upper Triassic of the Newark Supergroup and its biochronological significance.
Palaeontology, 41, 1215-1230.
Lucas, S. G., Heckert, A. B., & Hunt, A. P. (2002). A new species of the aetosaur Typothorax
(Archosauria: Stagonolepididae) from the Upper Triassic of east-central New Mexico.
New Mexico Museum of Natural History and Science Bulletin, 21, 221-233.
Martínez, R. N., & Heckert, A. B. (2015). Novel morphological insights from an incomplete,
articulated skeleton of a primitive aetosaur (Archosauira, Pseudosuchia) from the Upper
Triassic Ischigualasto Formation, San Juan province, Argentina. Journal of Vertebrate
Paleontology, SVP Program and Abstracts Book, 2015, 176.
Martínez, R. N., Sereno, P. C., Alcober, O. A., Colombi, C. E., Renne, P. R., Montañez, I. P.,
& Currie, B. S. (2011). A basal dinosaur from the dawn of the dinosaur era in
southwestern Pangaea. Science, 331, 206-210.

35
Martínez, R. N., Apaldetti, C., Alcober, O. A., Colombi, C. E., Sereno, P. C., Fernandez, E.,
Malnis, P. S., Correa, G. A., & Abelin, D. (2013). Vertebrate succession in the
Ischigualasto Formation. Journal of Vertebrate Paleontology, 32(sup1), 10-30.
Marsh, A. D., Smith, M. E., Parker, W. G., Irmis, R. B., & Kligman, B. T. (2020). Skeletal anatomy
of Acaenasuchus geoffreyi Long and Murry, 1995 (Archosauria: Pseudosuchia) and its
implications for the origin of the aetosaurian carapace. Journal of Vertebrate Paleontology,
e1794885. doi:10.1080/02724634.2020.1794885
Martz, J. W. (2002). The morphology and ontogeny of Typothorax coccinarum (Archosauria,
Stagonolepididae) from the Upper Triassic of the American Southwest [M.S., Texas
Tech University]. Lubbock.
Nesbitt, S. J. (2011). The early evolution of archosaurs: relationships and the origin of major clades.
Bulletin of the American Museum of Natural History, 352, 1-292.
Paes Neto, V. D., Desojo, J. B., Brust, A. C. B., Ribeiro, A. M., Schultz, C. L., & Soares, M. B.
(2021). The first braincase of the basal aetosaur Aetosauroides scagliai (Archosauria:
Pseudosuchia) from the Upper Triassic of Brazil. Journal of Vertebrate Paleontology,
e1928681. doi:10.1080/02724634.2021.1928681
Parker, W. G. (2007). Reassessment of the aetosaur 'Desmatosuchus' chamaensis with a
reanalysis of the phylogeny of the Aetosauria (Archosauria: Pseudosuchia). Journal
of Systematic Palaeontology, 5, 41-68. https://doi.org/10.1017/S1477201906001994
Parker, W. G. (2016a). Revised phylogenetic analysis of the Aetosauria (Archosauria:
Pseudosuchia); assessing the effects of incongruent morphological character sets.
PeerJ, 4, e1583. https://doi.org/10.7717/peerj.1583
Parker, W. G. (2016b). Osteology of the Late Triassic aetosaur Scutarx deltatylus
(Archosauria: Pseudosuchia). PeerJ, 4, e2411. https://doi.org/10.7717/peerj.2411

36
Parker, W. G. (2018a). Anatomical notes and discussion of the first described aetosaur
Stagonolepis robertsoni (Archosauria: Suchia) from the Upper Triassic of Europe,
and the use of plesiomorphies in aetosaur biochronology. PeerJ, 6, e5455.
https://doi.org/10.7717/peerj.5455
Parker, W. G. (2018b). Redescription of Calyptosuchus (Stagonolepis) wellesi (Archosauria:
Pseudosuchia: Aetosauria) from the Late Triassic of the Southwestern United States with a
discussion of genera in vertebrate paleontology. PeerJ, 6, e4291.
https://doi.org/10.7717/peerj.4291
Parker, W. G., Irmis, R. B., Nesbitt, S. J., Martz, J. W., & Browne, L. S. (2005). The Late
Triassic pseudosuchian Revueltosaurus callenderi and its implications of the diversity
of early ornithischian dinosaurs. Proceedings of the Royal Society of London, B, 272,
963-969.
Robert-da-Silva, L. Desojo, J. B., Cabreira, S. F., Aires, A. S. S., Müller, R. T., Pacheco, C.
P., & Dias-da-Silva, S. (2014). A new aetosaur from the Upper Triassic of the Santa
Maria Formation, southern Brazil. Zootaxa, 3764(3), 240-278.
Schoch, R. R. (2007). Osteology of the small archosaur Aetosaurus from the Upper Triassic
of Germany. Neues Jahrbuch für Geologie und Paläontologie Abhandlungen, 246(1),
1-35.
Taborda, J. R. A. (2011). Dinámica del crecimiento de Aetosauroides scagliai Casamiquela,
1960 (Archosauria: Aetosauria) del Triásico Tardío de América del Sur [MS,
Universidad de Buenos Aires]. Buenos Aires.
Taborda, J. R. A., Cerda, I. A., & Desojo, J. B. (2013). Growth curve of Aetosauroides
scagliai Casamiquela 1960 (Pseudosuchia: Aetosauria) inferred from osteoderm
histology. Geological Society, London, Special Publications, 379, 413-423.

37
Taborda, J. R. A., Heckert, A. B., & Desojo, J. B. (2015). Intraspecific variation in
Aetosauroides scagliai Casamiquela (Archosauria: Aetosauria) from the Upper
Triassic of Argentina and Brazil: Evidence of sexual dimorphism? Ameghiniana,
52(2), 173-187.
Walker, A. D. (1961). Triassic reptiles from the Elgin area: Stagonolepis, Dasygnathus, and
their allies. Philosophical Transactions of the Royal Society of London, B, 244, 103-
204.
Zacarias, J. D. (1981). Una nova especie de tecodonte aetosaurio (Aetosauroides
subsulcatus, sp. nov.) de Foramacáo Santa Maria, Triassico de Rio Grande du Sul,
Brasil [M.S., University Federal of the state of Rio Grande]. Porta Alegre.
Figure captions
Figure 1. Index map showing the general location and stratigraphic position of the aetosaur
specimen described here. 1, General reconstruction of Pangaea in the Late Triassic, diamond
marks approximate position of the Ischigualasto Basin; 2, Position of the locality (4-pointed
star) in the Ischigualasto basin of San Juan Province (SJ), Argentina (map of Argentina
modified from FreeVectorMaps.com); 3, Stratigraphy of the Ischigualasto Formation
showing age estimates, distribution of aetosaur fossils, and approximate stratigraphic position
of PVSJ 691, 180 m above the base of the formation (modified from Martínez et al., 2011,
2013). PVSJ, Instituto y Mueso de Universidad Nacional de San Juan, San Juan, Argentina.
Figure 2. Articulated skeleton (1–3) and isolated osteoderms (4) of PVSJ 691, an aetosaur
from the Upper Triassic Ischigualasto Formation. 1, Skeleton in ventral aspect, left image of
stereo pair; 2, Skeleton in ventral aspect, right image of stereo pair with anatomical labels ; 3,
View of disarticulated elements in anterior portion of preserved skeleton, scale blocks are on
matrix; 4, Dorsal paramedian osteoderms in matrix from an associated block. Abbreviations:

38
ab, anterior bar; app, appendicular osteoderms; c, chevron; co, center of ossification; cv,
cloacal vent; d, dorsal vertebra; dp, distal pubis; dpm, dorsal paramedian osteoderm; f,
femur; fi, fibula; fv, fused ventral osteoderms, i, ischium; io, ischiatic osteoderms l, lateral
osteoderm; mt, metatarsals (I–IV refer to number), of, obturator foramen; p, pubis; pes, pes;
ph, phalanx; pp, proximal pubis; t, tibia; v, ventral osteoderms. Scale bars equal 10 cm (1–
2),s2 cm (3–4). Scale blocks in (3) are in cm.
Figure 3. Interpretive line drawing of the skeleton of PVSJ 691, an aetosaur from the Upper
Triassic Ischigualasto Formation illustrated in Figure 2.1. Abbreviations: app, appendicular
osteoderms; as, astragalus; c, chevron; cd, caudal centrum; d, dorsal vertebra; f, femur; fi,
fibula; i, ischium; io, ischiatic osteoderms l, lateral osteoderm; mt, metatarsals (I–IV refer to
number), of, obturator foramen; p, pubis; pes, pes; pm, paramedian osteoderm; pp, proximal
pubis; sr, sacral rib; t, tibia; v, ventral osteoderms. Scale bar equals 10 cm.
Figure 4. Details of PVSJ 691, an aetosaur from the Upper Triassic Ischigualasto Formation
(specimen is exposed in primarily ventral aspect). 1, Right hind limb, primarily in anterior
view; 2, Left hind limb, primarily in anterior view; 3, Close-up of the cloacal region in
ventral view; Arabic numerals refer to specific ventral osteoderms; 4, Close-up of right distal
tibia, fibula, and pes in ~plantar view, Roman numerals refer to specific metatarsals; 5,
Close-up of the ischiatic region in ventral view showing the small ischiatic osteoderms
(anterior to left), Arabic numerals refer to specific osteoderms; 6, Close-up of mid-caudal
region in ventral view. Abbreviations: app, appendicular osteoderms; as, astragalus; c,
chevron; co, center of ossification; cv, cloacal vent; dpm, dorsal paramedian osteoderm; f,
femur; fi, fibula; i, ischium; ib?, possible iliac blade; io, ischiatic osteoderms l, lateral
osteoderm; ml, midline; mt, metatarsals (I–IV refer to number), of, obturator foramen; p,

39
pubis; pes, pes; ph, phalanges; t, tibia; u, ungual phalanx; v, ventral osteoderms. Scale bars
equal 5 cm (1–2), 2 cm (3–6). Scale blocks are in cm.
Figure 5. Interpretive line drawings of the skeletal features of PVSJ 691, an aetosaur from
the Upper Triassic Ischigualasto Formation illustrated in Figure 3.1–3.5. 1, Right hind limb,
primarily in anterior view; 2, Left hind limb, primarily in anterior view; 3, Close-up of the
cloacal region in ventral view; Arabic numerals refer to specific ventral osteoderms; 4, Close-
up of right distal tibia, fibula, and pes in ~plantar view, Roman numerals refer to specific
metatarsals; 5, Close-up of the ischiatic region in ventral view showing the small ischiatic
osteoderms (anterior to left), Arabic numerals refer to specific osteoderms. Abbreviations:
app, appendicular osteoderms; as, astragalus; c, chevron; co, center of ossification; cv,
cloacal vent; dpm, dorsal paramedian osteoderm; f, femur; fi, fibula; i, ischium; ib?, possible
iliac blade; io, ischiatic osteoderms l, lateral osteoderm; ml, midline; mt, metatarsals (I–IV
refer to number), of, obturator foramen; p, pubis; pes, pes; ph, phalanges; t, tibia; u, ungual
phalanx; v, ventral osteoderms. Scale bars equal 5 cm (1–2) and 2 cm (3–5).
Figure 6. Comparisons of the caudal armor series of diverse aetosaurs. 1, Aetosaurus ferratus
(SMNS 5770-S21); 2, Coahomasuchus kahleorum holotype specimen (NMMNH P-18496);
3, The specimen tentatively assigned to Aetosauroides and described here (PVSJ 691); 4,
The holotype of Aetosauroides scagliai as illustrated by Casamiquela (1967, lamina I; PVL
2073); 5, Typothorax coccinarum (NMMNH P-56299), after Heckert et al. (2010). All scale
bars equal 10 cm.

TABLE 1 — Measurements of limb bones of PVSJ 691 and PVL 2073 (in mm). Numbers in
parentheses are the equivalent measurement in protocols described by Martz (2002, fig.
4.13).
PVSJ 691 PVL 2073
Femur Left Right Left Right
Length (1): 154.7 — 143.7 152.2
Distance from head to 4
th
trochanter (2) — — 49.8 54.3
Width of femoral head (3): ~30 — 37.0 30.9AP
Length of femoral head (4): 18.9+ — 17.9 16.9
Thickness across 4
th
trochanter (5) — — 22.9 23.7
Mid-shaft long axis (6): ~24 — 20.1 20.5
Mid-shaft short axis (7): 16+ — 13.7 13.6
Distal width (8): ~42 ~44 40.3 42.7
Tibia measurements Left Right Left Right
Length (1): ~117.6 120.2 — 115.9
Proximal medio-lateral width (2): 38 AP ~50 — 35.6
Proximal antero-posterior thickness 3): — — — 20.8
Mid-shaft long axis (4): 15.7 — — 14.3
Mid-shaft short axis (5): 10.5 — — 11.9
Distal width (6): 22.5 — — ~20
Distal length (7): ~13.4 — — 8.0
Fibula Left Right Left Right
Length (1): 125? 125? — 118.9
Proximal width (2): — — — 19.8
Proximal thickness (3): — — — 7.4
Length from proximal head to
peak of iliofibularis trochanter (4): ~44 ~44 — 44.3
Thickness at iliofibularis trochanter (5): — ~24.74 — 15.6
Distal width (6) — — — 20.5
Distal length (7): ~11.9 ~11.9 — 8.0
Right Pes (all measurements are lengths)
Metatarsal I: 36
Phalanx I-1: 20.3
Ungual II: ~25
Metatarsal II 50
Phalanx II-1 ~19
Metatarsal III ~51
Explanations
~ = approximate measurement
AP = As preserved, measurement is incomplete

+ = minimum measurement based on nearly complete preservation but specimen may be
greater

TABLE 2 — Measurements of caudal ventral osteoderms of PVSJ 691 (in mm).
Osteoderm row Midline length Anterior Width Posterior Width
Left Right Left Right
7 ~24 49+ 50.6 — ~48
8 24 — 58 54 49+
9 ~24.5 ~49+ 48.5 46 42
10 ~24.5 47 41.4AP 43.8 44
11 ~24.5 44 43 39 39
12 23 39 39 37 37
13 23 38 38 36 36
14 23 30 31 — 31
15 21 — — — —
Fused osteoderms
Osteoderm row Width: Anterior Posterior
16 50.9 44.3
17 41.3+ ~39
18 36.8 35.5
19 ~33 28.6 (estimated)
Explanations
~ = approximate measurement
+ = minimum measurement based on nearly complete preservation but specimen may be
greater