Redescription of Elasmosaurus platyurus Cope, 1868 (Plesiosauria: Elasmosauridae) from the Upper Cretaceous (Lower Campanian) of Kansas, U.S.A

Abstract

The type specimen of Elasmosaurus platyurus Cope, 1868 from the Upper Cretaceous (lower Campanian) of Kansas, U.S.A. is redescribed. It consists of part of the skull (e. g., both premaxillae, parts of the maxillae, the occipital condyle and parts of the dentaries), the almost complete vertebral column, including the atlas-axis complex, as well as the pectoral and pelvic girdles (although the latter are now lost). The genus Elasmosaurus can be defined by two unambiguous autapomorphies, the presence of six premaxillary teeth and the high number of 71 cervical vertebrae. It also exhibits a number of advanced features, which are discussed and compared with other elasmosaurs.

Full text

Paludicola 5(3):92-106 November 2005
 by the Rochester Institute of Vertebrate Paleontology
92
REDESCRIPTION OF ELASMOSAURUS PLATYURUS COPE 1868
(PLESIOSAURIA: ELASMOSAURIDAE) FROM THE UPPER CRETACEOUS
(LOWER CAMPANIAN) OF KANSAS, U.S.A.
Sven Sachs
Institut für Paläontologie, Freie Universität Berlin, Haus D, Malteser Strasse 74-100, 12249 Berlin, Germany
Sachs.Pal@gmx.de
ABSTRACT
The type specimen of Elasmosaurus platyurus Cope, 1868 from the Upper Cretaceous (lower Campanian) of Kansas, U.S.A.
is redescribed. It consists of part of the skull (e. g., both premaxillae, parts of the maxillae, the occipital condyle and parts of the
dentaries), the almost complete vertebral column, including the atlas-axis complex, as well as the pectoral and pelvic girdles
(although the latter are now lost). The genus Elasmosaurus can be defined by two unambiguous autapomorphies, the presence
of six premaxillary teeth and the high number of 71 cervical vertebrae. It also exhibits a number of advanced features, which
are discussed and compared with other elasmosaurs.
INTRODUCTION
Elasmosaurus platyurus is the first described and
nominal genus of the Elasmosauridae, a family of long
necked plesiosaurs recorded from the Lower Jurassic to
the Upper Cretaceous. In 1867 a military physician Dr.
Theophilus H. Turner discovered the first remains of
the type specimen, two fragments of vertebrae, in a
ravine, approximately 23 km northeast of Fort Wallace
in western Kansas. In the same year, Dr. John W. Le
Conte brought the specimens to the Academy of
Natural Sciences in Philadelphia, where Edward
Drinker Cope examined them (Almy, 1987). Cope
recognized that they were from a plesiosaur and
noticed the importance as being part of a then barely
represented group of fossil reptile in the New World.
The remaining parts of the skeleton were excavated
and brought to Philadelphia with support by the
Academy in early 1868. Later Cope more fully
described the specimen in a preprint of the final paper.
In his first attempt of a reconstruction Cope
erroneously placed the head at the end of the tail. After
realizing this mistake all copies of the preprint were
recalled and were then reprinted with a new
reconstruction (Storrs, 1984; for further historic
information, also see Almy, 1987, and Davidson, 1997,
2002).
Although Elasmosaurus platyurus is the first
described member of the Elasmosauridae, a detailed
description, including all parts of the type specimen has
only been provided by Cope (1869, 1875). Additional
studies, mainly on the postcranium, have been
undertaken by Welles (1952), Storrs (1999) and
Carpenter (1999). In the present paper the type
specimen of Elasmosaurus platyurus (ANSP 10081) is
redescribed and the diagnostic features are pointed out
in order to state the taxonomic position of the genus
within the Elasmosauridae.
Institutional abbreviation: ANSP: Academy of
Natural Sciences in Philadelphia.
SYSTEMATIC PALAEONTOLOGY
Order PLESIOSAURIA de Blainville, 1835
Superfamily PLESIOSAUROIDEA Welles, 1943
Family ELASMOSAURIDAE Cope, 1869
ELASMOSAURUS PLATYURUS Cope, 1868
Diagnosis—Premaxillae bear six teeth per side;
atlas-axis complex long and low, with distinct sagittal
keel on ventral side; total number of cervical vertebrae
71; mid-cervicals anteroposteriorly elongate; feature-
complex of 4 sacral vertebrae and prominent midline
bar in pectoral and pelvic girdles.
Material—ANSP 10081, incomplete skeleton,
comprising both premaxillae, part of the posterior
section of the right maxilla, 2 maxillary fragments with
teeth, anterior portions of dentaries, three additional
jaw fragments, 2 indeterminable cranial fragments, 72
cervical vertebrae, including the atlas-axis complex, 3
pectoral, 6 dorsal, 4 sacral and 18 caudal vertebrae, as
well as a number of rib fragments. Pectoral and pelvic
girdles originally preserved, but now missing.

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
93
Measurements of cranial elements are given in Table 1;
those of vertebrae in Table 2.
Locality and Horizon—Near McAllaster, Logan
Country, Kansas. Sharon Springs Shale Member, Pierre
Shale (lower Campanian) (for more information see
Storrs, 1999:3-4).
DESCRIPTION
Cranium
Premaxilla—Both premaxillae (Figure 1 A, B)
are almost completely preserved and well co-ossified,
as no medial suture is visible. They fit well together
with the dentary portions, but show clear traces of
dorsoventral crushing. Dorsally, along the midline, a
low keel is present (Figure 2) that might have been
more distinct before the skull was compressed.
Anteriorly the tip of the snout is well-rounded and
almost semicircular in dorsal view. The ventral sides of
the premaxillae are not well preserved. Medially,
behind the alveolar part, the vomeronasal fenestrum is
visible as a broad triangular depression (Figure 2).
Cope (1868:53) stated that eight teeth are present
in each premaxilla. This statement cannot be
confirmed. Indicators for tooth positions are
concavities in the lingual edges of the alveolar parts
(Figure 2). Six of these concavities are present on each
side. Comparing the specimen with the premaxillae of
other elasmosaurs and considering the fact that a suture
to the maxillae is not visible, it can be concluded that
the fragment represents the premaxillae only. The
premaxillary-maxillary contact would most probably
have been immediately behind the last concavity, thus
a total number of six teeth per premaxilla is most
parsimonious. Only fragments of the second and third
tooth are preserved in the right premaxilla, being
formed as large fangs. The first premaxillary teeth, of
which only cross-sections are present, are clearly
smaller than the succeeding ones and are located
between the first two teeth of the dentaries. The apex of
what appears to be a replacement tooth, is visible in the
second alveolus of the left premaxillary. A fragment of
the sixth tooth of the right premaxilla is attached to the
lower jaw as can be seen when reassembling the jaws.
Maxilla—One fragment (Figure 3A) apparently
represents part of the posteriormost section of the right
maxillary. Its medial side is formed as a strong
dorsolaterally curved wall, while the lateral side is less
curved and not as strong. The fragment bears some
alveoli of which only four are clearly visible. They are
very small, compared with the other tooth-bearing
fragments. Among the latter are two (here called
fragment B and C) that are lateromedially flattened,
bearing one large tooth each (Figure 3B, C). Their
general appearance would be unusual for the lower
jaw, thus they are determined as fragments of the
maxillary.
Occipital Condyle—The hemispherical occipital
condyle is present and still articulates with the atlas
centrum (Figure 4A, B), thus no further observations
were possible.
Mandible—The present anterior portions of the
dentaries are well preserved and still bear parts of the
teeth in all alveoli (Fig. 1C, D). Five teeth are present
per side, which are all developed as large fangs and
appear to be about equal in diameter as indicated by
their cross sections. The symphysis is well ossified so
that no suture is visible between the rami, and extends
to the level of dentary tooth four. Dorsally, in about the
posterior half of the symphyseal margin, a prominent
triangular shaped bulge is formed. From here, at the
point where the rami are separated, two medially
situated, elongate, thickened structures are formed that
follow the rami posteriorly.
Additional Jaw Fragments—Among the jaw
fragments are three other tooth bearing elements that
cannot be referred to the maxillary or the dentary with
certainty. One of them (here called fragment D, Figure
3D) bears two large alveoli which served strong fang-
like teeth as they are usually present in the anterior half
of the dentary. The other two (here called fragment E,
Figure 3E and fragment F, Figure 3F) have
comparatively small alveoli, which indicates that were
situated in the posterior half of the jaw.
Additional Cranial Material—Among the
numerous indeterminable fragments are two pieces,
which probably belong into the cranium. One (here
called fragment G, Figure 3G, H) has a robust
appearance and bears a pan-like structure that could
have served for articulation. On one side, this structure
is bordered by a blunt, broad process. Its general
appearance is similar to the articular, in which the pan-
like structure would be part of the glenoid fossa,
however the preservation is to poor to provide an
unambiguous determination. Another possible
interpretation is that the element represents a fragment
of the right squamosal with part of the right articular
fused onto its inner surface. Figure 3G would then
show the inner surface of the eroded or broken off right
quadrate. The blunt process then could be the temporal
arch process of the squamosal. The other fragment is
curved and has one flat side (here called fragment H,
Figure 3I). Again, a clear determination is not possible,
because of the poor preservation. Eventually it could
have been part of the skull roof (e.g. the frontal).
Vertebral Column
Atlas-axis Complex—The atlas-axis complex is
almost completely preserved. Only the cervical ribs
and the most posterodorsal section of the neural spine
are missing. It is horizontally rectangular in outline in

PALUDICOLA, VOL. 5, NO. 3, 2005
94
TABLE 1. Measurements (in cm) of the cranial elements of ANSP 10081.
_______________________________________________________________________________________________________________________
lateral view (Figure 4A-C). Although both centra are
well co-ossified, a suture-like structure is visible at the
right lateral side. However, this structure might be an
artefact, due to lateral pressure onto the centrum. The
co-ossification indicates that the specimen was an adult
individual.
The neural arches are well preserved on the right
side of the complex and are broken off on the left.
Sutures between the centra and neural arches are not
visible. Interestingly, the neural arches are
extraordinarily thin and relatively high. Thus the neural
canal has a high triangular outline in posterior view.
The ventral section of the neural canal is rather narrow
posteriorly, at the axis, where it occupies about half of
the breadth of the centrum. It becomes clearly broader
anteriorly and is expanded almost over the entire
breadth of the centrum of the atlas. Here also the neural
arches are slightly more robust than in the axis and the
neural canal is clearly higher. On the preserved right
side a moderately large circular foramen is present. It is
situated at the base of the neural arches, between the
posterior edge of the neural arch of the atlas and the
anterior side of the neural arch of the axis.
The neural spine is incompletely preserved, as the
posterior half of the axis section is missing. It is very
low and posterodorsally oriented at an angle of
approximately eight degrees to the horizontal plane.
The centra of the atlas and axis vertebra are about
equal in length, each having a quadrate shape in lateral
view. The axis centrum shows prominent depressions
in about the center of the lateral sides. They are
probably enlarged due to the compression onto the
centrum, but appear to have been present in the pre-
compressed state too.
The articular facet of the atlas is still connected to
the occipital condyle and can thus not be observed,

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
95
TABLE 2. Measurements of the vertebrae of ANSP 10081. CV = Cervical vertebrae; PV = pectoral vertebrae; DV = dorsal vertebrae; SV = sacral
vertebrae; CAV = caudal vertebrae; a. = as measurable, i.c. = incomplete.
________________________________________________________________________________________________________________________
whereas the articulation surface of the axis is clearly
visible. It has a broad oval outline, with an excavation
for the neural canal in the middle of its dorsal edge,
and a small notch at about the same level ventrally. The
facet itself is deeply concave and has robust, rounded
edges. Only the basal sections of the cervical ribs are
visible. At the right lateral side, they occur in the
posterior most section of the atlas and are prominently
developed in the axis.
The ventral side of the atlas-axis complex shows
a distinct medial keel that runs between the anterior
edge of the atlas and the posterior edge of the axis.
Although only the basal section of this keel is
preserved, it can be seen that it is transversally broader
at the atlas. Lateral to the keel, the ventral surface is
longitudinally depressed. There is only a slight
depression present in the atlas, whereas the depression
is more prominent in the axis.
Cervical Vertebrae—Including the atlas-axis
complex, a total number of 71 cervical vertebrae are
present (Figure 4D-K). Of these, only two are
incomplete so that the series fits together almost
perfectly and can also be reassembled on basis of the
original matrix that is still attached to many vertebrae.
Most vertebrae are laterally compressed, with this
compression being stronger in the mid-cervical region,
than in the anterior or posterior sections. The
longitudinal lateral crest is well developed. It is slightly
visible in the third cervical, clearly developed from the
6th to the 45th and still visible to the 55th cervical
vertebra, thus into the posterior section of the neck.
The position of the crest is in the middle of the centrum
in the anterior vertebrae and in its dorsal half from
vertebra number 19 onwards.
The shape of the centra differs depending on the
position in the neck. The third cervical is about as long

PALUDICOLA, VOL. 5, NO. 3, 2005
96
FIGURE 1 Elasmosaurus platyurus Cope 1868. A. Dorsal view of premaxillae. B. Ventral view of premaxillae, with the vomeroventral fenestration
visible posteromedially. Anterior portions of dentaries in C dorsal and D. ventral view. Scale = 5 cm.
________________________________________________________________________________________________
as broad. Beginning with the fourth cervical, the centra
become longer than broad. The vertebrae are
extraordinarily elongate in the mid-cervical region. In
the posterior neck, they become again shorter and are
rather broad, so that in number 61 the breadth and
length are about equal and the posteriormost cervicals
are broader than long. The shape of the articulation
facets cannot always be stated accurately, due to the
aforementioned compression. In the most anterior
centra they are broad oval and moderately deepened.
The edges are rounded and thickened and show a clear
excavation in the middle of the dorsal and ventral side.
In the more posterior section of the anterior neck (e.g.
in number 25) the concavity of the ventral edge of the

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
97
FIGURE 2. Elasmosaurus platyurus Cope 1868. Premaxillae in A. dorsal and B. ventral view. Abbreviations: lc = lingual concavitiy, indicating tooth
position, sk = sagittal keel, vf = vomeronasal fenestrum.
________________________________________________________________________________________________________________________
articulation facets becomes more prominent and the
shape of the facets is quadrate with rounded edges.
Cervicals number 38 – 49 are strongly deformed,
whereas numbers 50-60 are still in the matrix so that no
statement about their shape can be given. The centrum
of cervical number 63 is less compressed. Its
articulation facet is also rather quadrate in general
shape, but with rounded edges. The facets of the most
posterior centra are broad oval in outline.
The neural arches are preserved in most cervicals.
Generally they are well fused with the centrum, so that
no suture is visible. The neural canal is rather narrow in
the anterior cervicals and prominently developed in the
most posterior ones. Here it is about as broad as high
and almost circular in the last cervical. Its shape cannot
be observed in the mid-cervicals, due to crushing of the
centrum.
The zygapophyses are present on a number of
vertebrae. Pre- and post-zygapophyses are about equal
in length. The pre-zygapophyses reach over the level
the centrum with their entire length. This can be
observed throughout the cervical vertebral column. The
post-zygapophyses reach over the level of the centrum
with about their posterior half.
The neural spines are completely preserved only
in the 19th and 20th cervical vertebra. Here they are
rather low, so that its shape in the 20th cervical can be
described as almost semicircular in lateral view.
The rib facets are readily visible on all cervicals.
They are situated laterally in the ventral section of the
vertebra almost throughout the cervical vertebral
column and they migrate dorsally only in the last three
vertebrae. In the next to last cervical, they are situated
in the middle of the lateral side of the centrum, are
prominently developed, and high ovals in outline. In
the last cervical vertebra they are situated somewhat
more dorsally and have a more triangular outline. Here
also, a distinct horizontal crest is formed anteriorly at
the lower side of the rib facets.
In a number of vertebrae the cervical ribs are still
attached to the centra, which are completely preserved
only in the anterior cervicals. In the fourth and fifth,
they are almost semicircular in lateral view, while in
the seventh they are more quadrate. In all cases the
cervical ribs are rather straight ventrally directed,
although this position could also be a result of the
lateral compaction.
The ventral side of the centra is mostly not well
preserved. A pair of nutritive foramina is placed in the
middle, separated by a short ridge. This ridge becomes
more prominent towards posterior in the neck and is
thickened in the most posterior cervical vertebrae.
Pectoral Vertebrae—Three pectoral vertebrae
are preserved (Figure 5A). They are still in the matrix,
so that an anterior view is not possible. In the first
pectoral centrum, the rib facets are triangular in outline

PALUDICOLA, VOL. 5, NO. 3, 2005
98
FIGURE 3. Elasmosaurus platyurus Cope 1868. Additional cranial elements. A. Posteriormost part of the right maxilla. B Fragment B, C. Fragment
C, toothbearing fragments probably belonging to the maxilla. D. Fragment D, possibly part of the anterior dentary. E Fragment E, F. Fragment F,
toothbearing fragments from the posterior half of the jaw, G, H. Undeterminable fragment G in two views, I. Undeterminable fragment H. Scale = 5
cm.

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
99
and are situated on small lateromedially-directed
transverse processes. At their anterodorsal margins, a
crest is formed that projects anteromedially and
represents the border of the transverse process. Another
much smaller crest is developed at the bottom of the rib
facets, forming the ventral border of the transverse
process. In the second pectoral vertebra, the rib facets
are still triangular in cross-section, but are not as
elongate ventrally as in the last cervical and first
pectoral vertebra. Thus it has a more robust
appearance. The dorsal margin of the transverse
processes is formed as a plate that is slightly oblique
ventrally. The anterior and posterior edges of this plate
form a clear ridge. The posterior surface of the
transverse process is directed straight vertically,
whereas the anterior one is smoothly curved
ventromedially. At the ventral portion of the transverse
process another, but clearly broader ridge is formed
that is slightly curved, ventrally directed, and there
fused with the centrum in a broad base. The ventral
sides of the centra bear a pair of nutritive foramina in
the middle. They are separated by a broad ridge, which
is formed as a rounded back that becomes more
expanded transversally towards the articulation facets.
In the last pectoral vertebra, there is also a broad, but
only slightly visible, ridge developed at the intersection
of lateral and ventral edges. In the third pectoral
vertebra the rib articulation section differs from that of
the second one in that it is situated more dorsally. The
transverse processes are directed almost straight
vertically. A distinct ridge is present at the posterior
and ventral side of the processes, but both are not as
sharp as in first two pectorals. The rib facets are broad
and triangular in shape, with a long, straight dorsal
margin and shorter, ventromedially directed anterior
and posterior sides. The mentioned ridge at the
intersection of the lateral and ventral edge is broader
and more distinct than in the next to last pectoral
vertebra.
Dorsal Vertebrae—Of the dorsal vertebral
column only six vertebrae are present, thus only about
1/3 of the actual number, compared with e.g.
Hydralmosaurus serpentinus where the number is 19
(Welles, 1952). Cope (1869: 49) thought that 10
dorsals were missing; Welles (1952:54) believed the
number was 18. The first and second preserved dorsals
belong into the anterior section of the column and may
represent positions four and five. The rib facets are
situated at the level of the neural canal. The transverse
processes, which are completely preserved in the ?fifth
vertebra, are more flattened dorsoventrally (although
this is probably also effected by the compaction) and
slightly oblique posterodorsally. The anterior and
posterior sides of the processes bear a distinct ridge
along their margins.
The rib facets are situated at a higher level than
the neck of the transverse processes and are thus
somewhat separated from them. They are broad oval to
rectangular in outline. The centrum of the ?fifth dorsal
is almost circular; its articulation facets are only very
slightly concave. Also the lateral sides of the centrum
show a horizontal depression.
The next preserved centrum might represent the
seventh or eight dorsal. Only the basal parts of the
transverse processes are preserved. As in the ?fifth
dorsal they are dorsoventrally narrow. The articulation
facets of the centrum are circular and the edges are
rather sharp, not well rounded as in the cervicals. Also
a longitudinal concavity is present at the lateral sides of
the centrum, beneath the transverse processes. The next
vertebra is only represented by the centrum. Here the
rib facets were situated clearly over the level of the
centrum, so that this vertebra might belong into the
mid-dorsal section. The last two vertebrae represent
another mid-dorsal, as well as large parts of a posterior
dorsal vertebra. The mid-dorsal is completely
preserved (Figure 5B, C). The shape of its centrum
corresponds with that of the previously described ones.
The rib facets are situated over the level the centrum.
The transverse processes are very elongate. Their
dorsal surfaces are slightly horizontally curved from
the edge of the neural spine to the rib articulation
surface. The later is slightly vertically convex. The
ventral margin is more prominently vertically convex.
The anterior and posterior edges of the transversel
processes bear sharp ridges. The shaft has a broad oval
cross-section. The rib facets are situated prominently
higher than the shaft. They are slightly oblique
posteriorly and are broad oval in outline. The pre-
zygapophyses are shorter than those of the cervical and
pectoral vertebrae and reach over the level of the
centrum only with about the anterior third of their
length. The post-zygapophyses are still very prominent
and reach over the level of the centrum with about the
posterior half of their length. The anterior hiatus is very
long, rather deep, and occupies about 60% of the length
of the neural spines. It thereby thins out dorsally and
thus has an elongated triangular outline. The posterior
hiatus is much shorter and only about half as long as
the anterior one. Dorsal of the hiatuses, the neural spine
becomes extraordinarily thin as can be observed
posteriorly. The last vertebra that belongs in the
posterior section of the column is only partially present
on the left side as the transversel process, the pre-
zygapophyses and most of the centrum are missing.
The neural canal is circular and surrounded by very
thick neural arches. The transverse process is shorter
and more robust than in the previously described
vertebra. Its basal sections at the neural arches are very
elongate anteroposteriorly and are broad ovals in cross
section. Laterally, the process, and especially its

PALUDICOLA, VOL. 5, NO. 3, 2005
100
FIGURE 4. Elasmosaurus platyurus Cope 1868. Cervical verterbrae. A-C Atlas-axis complex with attached occipital condyle in A. lateral, B. ventral
and C. posterior view. D-K Ten anteriormost cervical vertebrae in lateral views. D. cervical vertebra 3, E. cervical vertebra 4, F. cervical vertebra 5,
G. cervical vertebra 6, H. cervical vertebra 7, I. cervical vertebra 8, J. cervical vertebra 9, K. cervical vertebra 10. Scale = 5 cm. L-O. Cervical
vertebra examples from the anterior-, mid- and posterior cervical region in lateral views (size reduced). L. cervical vertebra 20, M. cervical vertebra
30, N. cervical vertebra 40, O. cervical vertebra 49. Scale = 7 cm.

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
101
FIGURE 5. Elasmosaurus platyurus Cope 1868. A. Last cervical and pectoral vertebrae in lateral view. B. Mid-dorsal vertebra in anterior and C.
lateral view. D. Sacral vertebal column associate with the two anterior-most caudal vertebrae in lateral view. E. Posterior caudal vertebrae in lateral
view. Scales = 9 cm.
________________________________________________________________________________________________________________________
anterior side, curve posteriorly. It thereby becomes a
high oval in outline, so that the lateral surface of the
anterior side of the process is formed as a high plate.
The posterior edge is generally thinner and also forms a
distinct ridge in its medial half. The lateral half is
expanded dorsoventrally. The rib facet is oval in
outline. Of the hiatuses, only the posterior one is
visible. It is larger than in the previously described one
and occupies about 50% of the neural spine. The latter
also has a very thin posterior edge above the hiatus.
Sacral Vertebrae—An articulated series of four
sacrals is present (Figure 5 D). In all vertebrae the
transverse processes are very short. In the first sacral,
the rib facets are situated in the posterodorsal section of
the centrum and are not as prominent as in the others.
They are posteromedially directed and have a
somewhat triangular outline, with a concave anterior
edge. On the dorsal side, a distinct, but short and
rather broad, anterior ridge is formed. In the second
sacral vertebra, the rib facets are more prominent than
in the first and are dorsoventrally elongated oval in
shape. They are situated posterior to the midline of the
centrum, but in its dorsal half and are only slightly
posteriorly directed. The anterodorsal ridge is longer
than in the first sacral. In the third sacral, the rib facets
are also dorsoventrally elongated ovals and are situated
at about the same level as in the second one, but are
clearly more prominent and about twice as large as in
the first sacral centrum. They are slightly posteriorly
orientated. The dorsal part of the transverse process is
situated at about the same level as the dorsal edge of
the neural canal. It is well rounded transversely and
forms an anterior edge that is not as distinct as in the
second sacral. The rib facets of the fourth sacral are
slightly larger than those of the third, and occupy
almost the entire lateral margin of the centrum. The
dorsal side of the transverse process is well rounded
and therefore distinctly separated from the surface of
the centrum. The rib facets have an almost straight
lateral orientation. Basal fragments of the sacral ribs
are attached to the second, third and fourth centrum.
The pre-zygapophyses are rather prominent and reach
over the level of the centrum for almost their entire
length. The post-zygapophyses are still connected to

PALUDICOLA, VOL. 5, NO. 3, 2005
102
the pre-zygapophyses so that no clear statement about
their length can be given. The neural spines are present
at the first and second vertebra, but are incomplete and
somewhat compressed laterally so that the shape of the
hiatuses cannot be observed. The ventral sides of the
centra are well rounded and bear a pair of nutritive
foramina that is separated by a broad but low ridge.
Caudal Vertebrae—A series of 18 caudal
vertebrae is present, in which the first two are in the
same block containing the sacral vertebrae (Figure 5D)
and are attached by matrix. The first caudal can clearly
be distinguished from the sacrals by the smaller size of
the rib facets, which are only about half as large as
those of the last sacral, and their position in the ventral
half of the centrum. They are almost circular in shape
and in the first and second caudal vertebrae also bear a
narrow horizontal keel in the middle of their dorsal
side. The rib facets are situated in the ventral section of
the centra throughout the caudal vertebral column.
They become larger and broader oval in shape from the
third caudal vertebra onwards. In the 6th-13th vertebra,
they are very large and occupy almost the entire
lateroventral margin of the centrum. In the 14th to 18th
vertebra, the rib facets are still prominent but
somewhat smaller compared to the size of the centra
(Figure 5E). The edges of the facets are rather sharp in
the anterior caudals and become well rounded in the
posterior ones. As in the other vertebrae, the pre-
zygapophyses reach over the level of the centra for
almost their entire length while the post-zygapophyses
reach over this level by about half of their length. The
neural spines are rather completely preserved in the
first, second, third and seventh caudal, but are
compressed laterally at their basis. Ventrally, the
surfaces for the haemapophyses are first clearly visible
in the fifth caudal, although according structures are
slightly visible in the third and fourth caudal too, but
here no depressions for the articulation are present. The
ventral margin of the centra is well rounded in the first,
second and third vertebra, rather plain and just slightly
concave in the fourth and fifth centrum, and strongly
transversally concave from the sixth to the 18th
vertebra. It should be noted that Cope (1870) stated
that some 30 caudals were missing, however, the usual
number of caudal vertebrae in elasmosaurs is around
30 (Welles, 1952).
Pectoral and Pelvic Girdles—Cope (1869,
1875) described and figured large parts of the pectoral
and pelvic girdles (Figure 6). These elements could not
be located in the collection of the Academy of Natural
Sciences in Philadelphia. Williston (1906: 225) had
already stated: “Unfortunately, the type specimen of
Elasmosaurus no longer has the girdles described by
Cope. What has become of them is not known”.
Carpenter (1999: 152) pointed out that the girdles
possibly were destroyed while on loan to Waterhouse
Hawkins, who, with his assistant, prepared them out of
the hard concretion (Cope, 1869: 51). Cope mentioned
to Turner that Hawkins was going to reconstruct
Elasmosaurus for a planned Paleozoic Park in New
York Central Park (Almy, 1987). However, this project
came to an end on May 3, 1871 when all of Hawkins
work was destroyed by vandals (Colbert and Beneker,
1959). It might well be possible that the missing girdle
elements were in Hawkins workshop at that time and
would have also been destroyed. Conspicuously,
nothing about their loss was mentioned in Cope’s 1875
publication.
The illustrations provided by Cope (1869, 1870)
are not very detailed, but still give an impression about
the shape of the bones. One scapula and the posterior
part of another, as well as the anterior portions of both
coracoids, had been preserved, whereas the pelvic
girdle was complete. Welles (1952: 56-59) redescribed
the missing parts on basis of Cope's figures and
descriptions and concluded that (1) Cope’s claviculae
or procoracoidea are the scapulae, as has been
suggested before by Seeley (1874); (2) the fused
scapulae met in the midline with no trace of a median
bar; (3) the dorsal processes of the scapulae were very
broad and the scapular neck was rather long; (4) the
scapulocoracoid suture was probably running to the
center of the glenoid; (5) the scapular foramen are too
long in Cope's reconstructions and were separated by a
pectoral bar; (6) the anteromedial growth of the
scapulae probably included the clavicular arch and (7)
a pelvic bar was present. Welles (1952: 59)
furthermore pointed out that in Cope's reconstruction
the greatest width of the pubes is at the front of the
acetabulum, whereas in other elasmosaurs the pubes
have anterolaterally directed external borders.
All other features, which can be taken from the
data provided by Cope (1869, 1870) are
synapomorphic for the Elasmosauridae and shall
therefore not be discussed.
DISCUSSION
Although relatively incomplete, the type
specimen of Elasmosaurus platyurus is of importance
as it represents the first described and nominal member
of the Elasmosauridae and the only specimen clearly
referable to this taxon. In the North American
Cretaceous, elasmosaurs are common, but are
represented mostly by isolated skeletal elements. A
number of genera have been described based upon
these remains (see Welles, 1952, 1962). In recent
reviews by Carpenter (1999) and Storrs (1999) many
taxa have been synonymized so that, besides
Elasmosaurus, currently ten elasmosaur genera are
known from the Cretaceous of North America. These
are (in stratigraphic order): Libonectes Carpenter
(1997, Britton Formation, Turonian), Thalassomedon
Welles (1943, Graneros Shale and Niobrara Formation,
Coniacian-Campanian), Hydralmosaurus and
Styxosaurus both Welles (1943, Niobrara Formation

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
103
FIGURE 6. Elasmosaurus platyurus Cope 1868. Original reconstructions of the pectoral- and pelvic girdles of Cope (from Cope, 1869). A. Pectoral
girdle, B. Pelvic girdle, C. Ilium. Not to scale.
________________________________________________________________________________________________________________________
and Pierre Shale formations, Campanian),
Terminonatator Sato (2003, Bearpaw Formation,
Campanian), and Hydrotherosaurus, Welles (1943),
Morenosaurus, Aphrosaurus, and Fresnosaurus all
Welles, (1943, Moreno Formation, Maastrichtian). The
latter genera Morenosaurus, Aphrosaurus and
Fresnosaurus are only based upon postcranial material.
In the following, the features of the type material of
Elasmosaurus platyurus are discussed and compared.
Cranial Remains—The premaxilla bears more
than five teeth (for character discussion see Sachs,
2005: character 2). Primitively plesiosauroids and most
elasmosaurs have five teeth per premaxilla. Exceptions
among the Elasmosauridae are Terminonatator (Sato,
2003) with nine teeth and Aristonectes 10-13 teeth per
premaxilla (Gasparini et al., 2003).
The dentary symphysis extends to a point
between the fourth and fifth tooth in the dentary. This
condition is also present in Tuarangisaurus australis
(Sachs, 2005), Hydrotherosaurus and Libonectes
(Carpenter, 1997: fig. 2). For the other elasmosaurid
taxa, the information is not provided in the literature.
Cervical Vertebrae—Atlas-axis complex is long
and low. In all sauropterygian taxa, except for some
members of the Elasmosauridae, the axis centrum is
either shorter than high or about as long as high. The
only elasmosaur taxa where the condition present in
Elasmosaurus platyurus is known are Styxosaurus
(Welles and Bump 1949; Sachs, 2004) and
Hydralmosaurus (Cope, 1877). It thus can be
considered as an advanced feature (see also Carpenter
1999, character 16; Sachs, 2004).
The number of cervical vertebrae is higher than
60 (for character discussion see Sachs, 2005, character
18). Including the atlas-axis complex, the type
specimen of E. platyurus (ANSP 10081) has 71
vertebrae in the neck. Various researchers have given
different numbers of cervical vertebrae for this
specimen. According to Cope (1869, 1875), 68 1/2
cervicals are present, to which he added another 3 1/2
that he mentioned as being lost. Wegner (1914) gives a
number of 60-78 cervicals; Welles (1949, 1952),
Brown (1981) and Storrs (1999) state 71 cervicals;
Carpenter (1997) reported 74 cervicals and then
(Carpenter, 1999) revised his count to 72 cervicals;
Sachs (2005) mistakenly also stated 72 cervicals;
whereas Welles (1943) also stated that there were 74
cervicals. Williston (1906) gave a total of 76 vertebrae
for the neck of Elasmosaurus. Apparently, some of the

PALUDICOLA, VOL. 5, NO. 3, 2005
104
pectoral vertebrae have been included in the higher
counts.
More than 60 cervical vertebrae is a very
advanced number among the Plesiosauroidea. In the
Jurassic elasmosaurid Occitanosaurus the number is 43
(Bardet et al., 1999). However the early Cretaceous
elasmosaurid Brancasaurus has only 37 cervical
vertebrae (Wegner, 1914). The number is higher than
60 in Libonectes (=62, Carpenter, 1999),
Thalassomedon (=62, Welles, 1943), Styxosaurus (=62,
Carpenter, 1999) and Hydralmosaurus (=63,
Carpenter, 1999). The number of cervicals is unknown
in Aphrosaurus, Fresnosaurus and Morenosaurus
(Welles, 1943). The only sauropterygian where the
number of cervical vertebrae is higher than 70 is
Elasmosaurus platyurus, accordingly this character is a
clear autapomorphy of Elasmosaurus and can be
considered as an advanced feature.
The anterior cervical vertebrae are long and low
(for character discussion see Sachs, 2005: character
17). As with the axis centrum, the shape of the anterior
cervical vertebrae is usually either high and short (e.g.
in the Pliosauridae, Tarlo, 1960) or about as long as
high (e.g. in the Plesiosauridae, Brown, 1981). The
centra become longer than high in most elasmosaur
taxa, like Styxosaurus (Sachs 2004), Hydralmosaurus
(Cope, 1877), Libonectes (Welles, 1949),
Occitanosaurus (Bardet et al., 1999), Callawayasaurus
(Welles, 1962), Terminonatator (Sato, 2003),
Hydrotherosaurus (Welles, 1943) and Tuarangisaurus
(Wiffen and Moisley, 1986; Sachs 2005). The
condition is unknown in Aphrosaurus, Fresnosaurus
and Morenosaurus (Welles, 1943).
A longitudinal lateral crest is present in the
anterior cervical vertebrae. This character has been
discussed before by Brown (1981, 1993), Carpenter
(1999) and Sachs (2004) and is synapomorphic for the
Elasmosauridae.
Pectoral and Dorsal Vertebrae—Three
pectorals are known in Occitanosaurus (Bardet et al.
1999), Brancasaurus (Wegner, 1914), Styxosaurus
(Welles and Bump, 1949), Thalassomedon (Welles,
1943), Aphrosaurus (Welles, 1952), and
Hydralmosaurus (Welles, 1952), thus this number is
common among the Elasmosauridae, although there are
only two pectorals present in Callawayasaurus (Welles
1962), Morenosaurus (Welles, 1952), and
Hydrotherosaurus (Welles, 1943). The number is
unknown for Tuarangisaurus (Wiffen and Moisley,
1986; Sachs, 2005), Fresnosaurus (Welles, 1952) and
Aristonectes (Gasparini et al., 2003) and is not
provided for Libonectes (Welles, 1949).
Dorsal vertebrae are not diagnostic at genus level
in the Upper Cretaceous elasmosaurs and are thus not
useful for comparison.
Sacral and Caudal Vertebrae—The number of
sacral vertebrae is four in the type specimen of
Elasmosaurus platyurus. Among the Elasmosauridae
four sacrals are the usual number as seen in
Occitanosaurus (Bardet et al., 1999), Terminonatator
(Sato, 2003), Styxosaurus (Sachs, 2004) or
Hydralmosaurus (Welles, 1952). Only three are present
in Brancasaurus (Wegner, 1914), Thalassomedon
(Welles, 1943), Morenosaurus (Welles, 1943), and
Hydrotherosaurus (Welles, 1943). The number is
unknown in Libonectes (Welles, 1949),
Tuarangisaurus (Wiffen & Moisley, 1986; Sachs,
2005), Callawayasaurus (Welles, 1962) Aristonectes
(Gasparini et al., 2003), Aphrosaurus (Welles, 1943)
and Fresnosaurus (Welles, 1943).
The tail is relatively complete, consisting of 18
vertebrae, but like the dorsal vertebrae also caudals are
not diagnostic among Cretaceous elasmosaurs and can
thus not be used for comparisons.
Pectoral and Pelvic Girdle—As mentioned
before, the pectoral and pelvic girdles are lost, so that
the comparative observations are totally based upon the
descriptions and figures of Cope (1869, 1870) and the
conclusions and figure of Welles (1952).
In the pectoral girdle the most conspicuous
feature is the long pectoral bar. Carpenter (1999)
discussed this feature and mentioned that according to
Brown (1981) there is an ontogenetic variation of this
character (demonstrated with an ontogenetic series of
Crpytoclidus eurymerus Phillips, 1871), in which the
bar is absent in juvenile and present in adult
individuals. Carpenter (1999) states that in Pistosaurus
and in some plesiosaurids, the bar is not developed in
adult specimen, thus its absence must be considered as
the plesiomorphic condition. Accordingly the presence
of the pectoral bar in Elasmosaurus platyurus is an
advanced feature.
On the scapula, the coracoidal margin appears to
be about as long as the glenoideal one.
Synapomorphically, the coracoidal margin is longer
(usually 1.5 times) than the glenoidal one (Sachs,
2004), although we don’t know how exact Cope’s
(1869) figure is in this respect. Also, the pelvic girdle
has a rather unusual shape in Cope's reconstruction.
Here the pubes are very prominent and are medially
connected over their entire length. Their anterior edge
is anterolaterally directed and almost straight, not
gradually curved as usual in elasmosaurs. Also, the
ischia are connected medially so that along the length
of the pelvic girdle a medial pelvic bar is present. The
latter is the most conspicuous feature because a bar is
usually not present in plesiosaurs. Among the
Elasmosauridae, a pelvic bar is known in Brancasaurus
brancai from the Lower Cretaceous (Berriasian) of
Germany and in a specimen from the Upper Cretaceous
(Campanian-Maastrichtian) of Patagonia, described by
Gasparini & Salgado (2000) as Elasmosauridae gen. et
sp. indet. Of the latter, other elements also are similar
in shape to ANSP 10081. However, on basis of the
preserved material, it cannot be judged whether both
specimens are congeneric.

SACHS--REDESCRIPTION OF ELASMOSAURUS PLATYURUS
105
RESULTS OF THE DISCUSSION
As shown here, Elasmosaurus platyurus
comprises two clear autapomorphies; the presence of
six teeth per premaxilla and 71 cervical vertebrae.
Additionally, it shows one feature-complex that is
unique within the Elasmosauridae: four sacral
vertebrae + presence of a pelvic bar. Elasmosaurus
also shows two advanced feature complexes that are
also only present in Styxosaurus and Hydralmosaurus:
(1) long and low axis centrum + anterior cervical centra
long and low. (2) number of cervicals higher than 60 +
three pectoral vertebrae. Elasmosaurus is more
advanced than most elasmosaur taxa as shown by the
presence of six premaxillary teeth, a long and low axis
centrum, the high number of 71 cervical vertebrae and
the development of both a pectoral and a pelvic bar. A
close relationship of Elasmosaurus platyurus with
Styxosaurus and Hydralmosaurus can be concluded on
basis of the present material.
ACKNOWLEDGMENTS
I am grateful to Dr. Ted Daeschler (ANSP) for his
kind support during my stay at the Academy and for
his help with the photos. I also thank the two reviewers
Kenneth Carpenter and Tamaki Sato, as well as Mike
Everhart for helpful comments. This project was
supported by a Jessup Fellowship of the Academy of
Natural Sciences, Philadelphia.
LITERATURE CITED
Almy, K. 1987. Thof’s dragon and the letters of Capt.
Theophilus H. Turner, M.D. Army. Kansas
History 10: 170-200.
Bardet, N., P. Godefroit, and J. Sciau. 1999. A new
elasmosaurid plesiosaur from the lower Jurassic
of southern France. Palaeontology 42:927-952.
Blainville, H. D. de. 1835. Description de quelques
espèces de reptiles de la Californie, précédée de
l'analyse d'un système général d'Erpetologie et
d'Amphibiologie. Nouvelle Annales de la
Museé de histoire Naturelles de Paris 3:233-296.
Brown, D. S. 1981. The English Upper Jurassic
Plesiosauroidea (Reptilia) and a review of the
phylogeny and classification of the Plesiosauria.
British Museum (Natural History) Bulletin,
Geology Series 35: 253-347.
Brown, D. S. 1993. A taxonomic reappraisal of the
families Elasmosauridae and Cryptoclididae
(Reptilia: Plesiosauroidea). Revue de Paléo-
biologie, Special Volume 7:9-16.
Carpenter, K. 1997. Comparative cranial anatomy of
two North American Cretaceous plesiosaurs.
Pp. 191-216 in J. M. Callaway and E. L.
Nicholls, (eds.), Ancient Marine Reptiles,
Academic Press, San Diego.
Carpenter, K. 1999. Revision of North American
elasmosaurs from the Cretaceous of the western
interior. Paludicola 2:148-173.
Colbert, E. and K. Beneker. 1959. The Paleozoic
Museum in Central Park, or the museum that
never was. Curator 2:137-150.
Cope, E. D. 1868. Remarks on a new enaliosaurian,
Elasmosaurus platyurus. Proceedings of the
Academy of Natural Sciences of Philadelphia
1868:92-93.
Cope, E. D. 1869. Synopsis of the Extinct Batrachia
Reptilia, and Aves of North America. Part I.
Transactions of the American Philosophical
Society, New Series 14:1-104.
Cope, E. D. 1870. Synopsis of the extinct Batrachia,
Reptilia and Aves of North America. Part II.
Transactions of the American Philosophical
Society, New Series 14:105-252.
Cope, E. D. 1875. The Vertebrata of the Cretaceous
formations of the West. Report of the United
States Geological Survey of the Territories, pp.
1-303.
Cope, E. D. 1877. Report on the geology of the region
of the Judith River, Montana, and on vertebrate
fossils obtained on or near the Missouri River.
Bulletin of the United States Geological and
Geographical Survey 3:565-597.
Davidson, J. P. 1997. The bone sharp. The life of
Edward Drinker Cope. Academy of Natural
Sciences of Philadelphia Special Publications
17:1-237.
Davidson, J. P. 2002. Bonehead mistakes: The
background in scientific literature and
illustrations from Edward Drinker Cope's first
restoration of Elasmosaurus platyurus.
Proceedings of the Academy of Natural Sciences
of Philadelphia 152:215-240.
Gasparini Z, and Salgado L. 2000. Elasmosáuridos
(Plesiosauria) del Cretácico Tardío del norte de
Patagonia. Revista Español de Paleontología
15:13-21.
Gasparini Z., Bardet N., Martin J., and Fernández M.
2003. The elasmosaurid plesiosaur Aristonectes
cabrera from the latest Cretaceous of South
America and Antarctic. Journal of Vertebrate
Paleontology 23:104-115.
Gray, J.E. 1825. A synopsis of the genera of reptiles
and Amphibia, with a description of some new
species. Annales of the Philosophical Society of
London 26:193-217.
Phillips, J. 1871. Geology of Oxford and the valley of
the Thames. 523 pp. Oxford.
Sachs, S. 2004. Redescription of Woolungasaurus
glendowerensis (Plesiosauria: Elasmosauridae)
from the Lower Cretaceous of Northeast

PALUDICOLA, VOL. 5, NO. 3, 2005
106
Queensland. Memoirs of the Queensland
Museum 49:215-233.
Sachs, S. 2005. Tuarangisaurus australis sp. nov.
(Plesiosauria: Elasmosauridae) from the Lower
Cretaceous of northeastern Queensland, with
additional notes on the phylogeny of the
Elasmosauridae. Memoirs of the Queensland
Museum 50:429-444.
Sato, T. 2003. Terminonatator ponteixensis, a new
elasmosaur (Reptilia: Sauropterygia) from the
Upper Cretaceous of Saskatchewan. Journal of
Vertebrate Paleontology 23:89-103.
Seeley, H. G. 1874. Note on some of the generic
modifications of the plesiosaurian pectoral arch.
Quarterly Journal of the Geological Society of
London 30:436-449.
Storrs, G. W. 1984. Elasmosaurus platyurus and a page
from the Cope-Marsh war. Discovery 17:25-27.
Storrs, G. W. 1999. An examination of Plesiosauria
Diapsida: Sauropterygia) from the Niobrara
Chalk (Upper Cretaceous) of central North
America. University of Kansas Paleontological
Contributions 11:1-15.
Tarlo, L. B. 1960. A review of the Upper Jurassic
pliosaurs. British Museum (Natural History)
Bulletin, Geology Series 4:145-189.
Wegner, T. 1914. Brancasaurus brancai n. g., n. sp.,
ein Elasmosauride aus dem Wealden
Westfalens. Pp. 234-305 in F. Schoendorf (ed.)
Branca Festschrift, Gebrüder Borntraeger,
Stuttgart.
Welles, S. P. 1943. Elasmosaurid plesiosaurs with a
description of the new material from California
and Colorado. University of California Memoirs
13:125-254.
Welles, S. P. 1949. A new elasmosaur from the Eagle
Ford Shale of Texas. Fondren Science Series
1:1-28.
Welles, S. P. 1952. A review of the North American
Cretaceous elasmosaurs. University of
California Publications in Geological Sciences
29:47-143.
Welles, S. P. 1962. A new species of elasmosaur from
the Aptian of Columbia and a review of the
Cretaceous plesiosaurs. University of California
Publications in Geological Sciences 44:1-96.
Welles, S. P., and J. Bump. 1949. Alzadasaurus
pembertoni, a new elasmosaur from the Upper
Cretaceous of South Dakota. Journal of
Paleontology 23:521-535.
Wiffen, J. and W. L. Moisley. 1986. Late Cretaceous
reptiles (Families Elasmosauridae and
Pliosauridae) from the Mangahouanga Stream,
North Island, New Zealand. New Zealand
Journal of Geology and Geophysics 29:205-252.
Williston, S. W. 1906. North American plesiosaurs:
Elasmosaurus, Cimoliasaurus, and Polycotylus.
American Journal of Science Series 4, 21:221-
234.

Corrections
p. 92: Systematic Palaeontology – Elasmosaurus platyurus Cope, 1869
p. 92: Material – 71 cervical vertebrae, including the atlas-axis complex
p. 104: Pectoral and Pelvic Girlde – Cryptoclidus eurymerus (Phillips, 1871)