Content uploaded by Andreas T. Matzke
Author content
All content in this area was uploaded by Andreas T. Matzke on Sep 17, 2018
Content may be subject to copyright.
116
Journal of Vertebrate Paleontology 23(1):116–127, March 2003
q
2003 by the Society of Vertebrate Paleontology
THE CRANIAL OSTEOLOGY OF THE ICHTHYOSAUR LEPTONECTES CF. TENUIROSTRIS FROM
THE LOWER JURASSIC OF ENGLAND
MICHAEL W. MAISCH and ANDREAS T. MATZKE
Institut und Museum fu¨r Geologie und Pala¨ontologie der Eberhard-Karls-Universita¨t
Sigwartstr. 10, D-72076 Tu¨bingen, Germany
ABSTRACT—For the first time, a three-dimensionally preserved skull of the Lower Jurassic leptonectid ichthyosaur
Leptonectes cf. tenuirostris is described. The specimen reveals a wealth of new data on the cranial anatomy of basal
leptonectid neoichthyosaurs. The osteology of the skull roof differs fundamentally from other well-known post-
Triassic ichthyosaurs and is highly autapomorphic. The structure of the sclerotic ring, which is well preserved in
situ, is described and it appears plausible that L. cf. tenuirostris was capable of stereoscopic vision. Inadequate
description makes comparison to other leptonectids difficult. It can be demonstrated, however, that L. cf. tenuirostris
is most similar to L. tenuirostris among known leptonectids. It is fundamentally different from the Upper Liassic
Eurhinosaurus longirostris in the osteology of the skull roof and the size and orientation of the major skull openings,
making generic distinction of these two species a necessity. At the moment, no autapomorphies are known of the
genus Leptonectes, which therefore must be considered as a metataxon that comprises basal leptonectids from the
Lower to Middle Liassic.
INTRODUCTION
Ichthyosaurs, a major clade of marine Mesozoic amniotes,
are known from the Lower Triassic (Spathian) to the Upper
Cretaceous (Cenomanian). Particularly in the Triassic and Ju-
rassic, they achieved a considerable diversity. Leptonectes ten-
uirostris (Conybeare, 1822) was one of the first ichthyosaur
species to be formally named and described. Many specimens
are known from the Rhaetian to Sinemurian of southern Eng-
land (Conybeare, 1822; Hawkins, 1834, 1840; Owen, 1881;
McGowan, 1974, 1989a, 1993, 1996), the Lower Sinemurian
of southern Germany (von Huene, 1922; Maisch, 1999) and
probably the Lower Pliensbachian of Belgium (Godefroit,
1992), including several complete skeletons and many skulls.
Even though the postcranial anatomy of L. tenuirostris is rela-
tively well known, particularly through the works of Hawkins
(1840), Owen (1881), and von Huene (1922), the cranial anat-
omy of the species was never investigated in detail. Although
Maisch (1999) provided a detailed account of a skull from
southern Germany, the specimen was much crushed and dam-
aged, so that many points of the cranial anatomy remained ob-
scure.
Here we report a well-preserved, three-dimensional skull,
most probably referable to this species and identified here as
Leptonectes cf. tenuirostris, from the Hettangian or Sinemurian
beds (unfortunately detailed stratigraphical data are lacking) of
Lyme Regis, Dorset, England. Although a drawing of its dorsal
aspect and a short discussion of its skull roof (with particular
respect to the foramen parietale) were published by Jaekel
(1903:34, fig. 4), it has so far escaped the notice of subsequent
ichthyosaur workers.
The specimen is kept at the Museum fu¨r Naturkunde der
Alexander-von-Humboldt-Universita¨t zu Berlin, where it is cat-
alogued under the number MBI 007.11. It was collected (and
apparently later also prepared) by the famous German paleon-
tologist Otto Jaekel himself during a visit to England in 1902.
The skull is embedded in a dark grey calcareous nodule, as are
numerous vertebrate specimens, including many ichthyosaurs,
from the Hettangian and Sinemurian of Lyme Regis.
The purpose of this paper is twofold. First, a detailed de-
scription and illustration of the cranial anatomy of the specimen
of L. cf. tenuirostris is provided. Second, a comparison of L.
cf. tenuirostris to other known leptonectid ichthyosaurs is
made. This allows a clearer definition of several taxa currently
included in the Leptonectidae, but it also reveals the consider-
able lack of knowledge on the details, particularly of the cranial
anatomy, of most members of the family; this lack hampers
phylogenetic analysis considerably.
Institutional Abbreviations BMNH, Natural History Mu-
seum, London; BRSMG, Bristol City Museum and Arts Gal-
lery; MBI, Museum fu¨r Naturkunde, Berlin; MHN, Museon,
The Hague.
SYSTEMATIC PALEONTOLOGY
Order ICHTHYOSAURIA de Blainville, 1835
HUENEOSAURIA Maisch and Matzke, 2000
LONGIPINNATI von Huene, 1948
MERRIAMOSAURIA Motani, 1999
PARVIPELVIA Motani, 1999
NEOICHTHYOSAURIA Sander, 2000
Family LEPTONECTIDAE Maisch, 1998b
Genus LEPTONECTES McGowan, 1996
Type Species Ichthyosaurus tenuirostris Conybeare, 1822.
LEPTONECTES CF. TENUIROSTRIS (Conybeare, 1822)
McGowan, 1996
(Figs. 1–5)
Emended Diagnosis Medium-sized ichthyosaur, total
length around 4,000 mm at maximum; skull length not exceed-
ing 1,000 mm; snout very long and slender, orbital ratio of skull
less than 0.25; snout ratio exceeding 0.70; prenarial ratio larger
than 0.56; sclerotic ratio much larger than 0.34; snout without
or with only rather inconspicuous overbite; fenestra supratem-
poralis of rounded shape, about one third the size of orbita;
jugal with narrow, dorsoventrally depressed ramus suborbitalis
and considerably wider ramus postorbitalis; forefin relatively
wide with four well-developed primary digits; humerus with
strongly constricted shaft and well-developed leading edge fac-
et; foramen interosseum remaining usually open between radius
and ulna; radius and tibia notched; other fin elements usually
117MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
unnotched; phalanges discoidal, large, rather widely spaced dis-
tally; femur with slender shaft, expanded distally.
Remark As there remains some doubt about the specific
identification of the Berlin specimen (see discussion), only fea-
tures that are displayed by specimens that can be confidently
identified on the specific level have been included in the above
diagnosis.
Distribution Rhaetian to Lower Sinemurian, southern Eng-
land, Lower Sinemurian, southern Germany, ?Lower Pliensba-
chian, Belgium.
Additional Species Included in the Genus Leptonectes
solei (McGowan, 1993) McGowan, 1996; Leptonectes moorei
McGowan and Milner, 1999.
DESCRIPTION
Preservation
The skull is three-dimensionally preserved from a level short-
ly behind the anterior narial margin up to its posterior end (Figs.
1, 2). Both lower jaw rami remain in natural articulation, but
their posterior margins are slightly damaged. Both sclerotic
rings are complete, undistorted and situated in natural position
within the orbitae (Figs. 1, 2A).
The occipital bones and palate, with the exception of part of
the right pterygoid in posterior and dorsal view (Fig. 4A, B),
are not freed from matrix. The left cheek region lacks the squa-
mosal, quadratojugal and quadrate (Fig. 1B), whereas its coun-
terpart is complete and articulated (Figs. 1A, 2A).
General Skull Shape
The skull is dominated by the very large orbitae as best seen
in lateral view, where the dorsal skull roof is hardly visible
(Figs. 1, 2A). The orbito–narial bar is short. The ramus subor-
bitalis of the jugal is extremely thin. The postorbital skull seg-
ment is very short (Figs. 1, 2A).
The skull roof is much narrower at the level of the anterior
than at the level of the posterior orbital margin (Figs. 2B, 3A,
B). This leads to a lateral and anterior orientation of the orbitae
(Fig. 3C) not found in other ichthyosaurs, except Eurhinosaurus
longirostris (Maisch and Matzke, 2000) where the anterior ori-
entation of the orbitae is more considerable.
From the level of the external naris backwards, the skull roof
ascends gently and continuously. It reaches its greatest height
at the level of the parietals. From the foramen parietale to the
anterior preserved end of the snout, a sagittal depression is pre-
sent. It is expressed as a deep, well-marked groove between the
frontals and widens to a shallow trough, the excavatio inter-
nasalis (Maisch, 1997), between the nasals. This is widest at
the level of the orbito-narial bar (Figs. 2B, 3A, B).
Lateral to the sagittal depression, the skull roof is convex,
particularly the nasals, which form a ridge-like convexity on
either side of the dorsal snout surface. This becomes less
marked from the level of the anterior orbital margin backwards
and tapers out at orbital mid-length. There, the skull roof is
essentially flat, except for the sagittal depression.
The lateral snout surface dorsal to the external naris meets
the dorsal nasal ridge in an angle of approximately 50
8
. The
orbito–narial bar is almost perpendicularly oriented to the dor-
sal surface. No crest or ridge delimits it from the skull roof.
The lateral surface of the anteroventral portion of the prefrontal
is markedly concave up to the anterior end of the sclerotic ring
(Figs. 1, 2A). Farther backwards, the dorsal prefrontal surface
parallels the dorsal convexity of the orbit. At the level of the
prefrontal–postfrontal suture, the supraorbital skull roof be-
comes essentially flat.
The parietals form a low sagittal crest (Figs. 2B, 3B). The
lateral parietal shelf is oriented at a small angle of about 25
8
to
the horizontal, before it turns abruptly into the vertical descend-
ing parietal flange.
At about the level of the posterior third of the orbit, the
orbito-temporal bar becomes convex again and is posteroven-
trally inclined. It turns in gentle curvature around the fenestra
supratemporalis and merges uninterruptedly with the occipital
surface of the skull. Farther ventrally, the lateral surface of the
postorbital skull segment is essentially a flat, vertical, very nar-
row bar of bone delimited against both the supraorbital and
occipital skull surfaces by the crista postorbitalis lateralis
(Maisch, 1998a).
The occipital skull surface ventral to the fenestra supratem-
poralis is generally convex transversely and more or less flat
dorsoventrally, to form the posterior limits of the adductor
chamber (Fig. 4A, B).
Skull Openings
External Naris The anterior margin of the external naris
(Figs. 1, 2A) is preserved on neither side. Nevertheless, the
remaining portion (probably at least two thirds) shows that the
naris was a big, low, anteroposteriorly-elongated aperture. Its
outline is somewhat complicated by a laterally and slightly ven-
trally protruding nasal wing along its posterior preserved third,
which protrudes slightly inside the external naris from above
but stops short before the aperture’s posterior end, so that a
slight posterodorsal lobe of the naris is formed.
The ventral narial margin runs parallel to the alveolar margin
of the upper jaw for most of its length, until it bows dorsally,
gently merging into the posterior margin, which is oriented per-
pendicular to the ventral margin. The dorsal margin of the naris
parallels, except for the aforementioned posterodorsal lobe, the
ventral margin of the nasal, so that the naris expands very slow-
ly and continuously in height from front to back. The dorsal
and posterior margins meet each other in a pronounced corner.
Orbita The orbit is an unusually large opening which dom-
inates the entire architecture of the skull (Figs. 1, 2A, B, 3A–
C, 5). As the cranium is undistorted, there is the rare possibility
to determine the natural outline of the orbit (see Fig. 5 for
measurements). It is not perfectly round but slightly compressed
dorsoventrally, with an almost perfectly oval shape in lateral
view. All margins are gently rounded, the dorsal and ventral
margins much less so than the anterior and posterior ones.
There is no area circumorbitalis anteriorly, anteroventrally, or
dorsally, but posteriorly and posteroventrally the postorbital
skull segment’s lateral surface, formed largely by the postor-
bital, and the posterior half of the ventral margin, formed by
the ramus suborbitalis of the jugal, are completely flat and
smooth and form a partial area circumorbitalis (which sur-
rounds the entire orbit in some Upper Jurassic forms such as
Aegirosaurus leptospondylus, pers. obs.). Otherwise the mar-
gins of the orbit are pronounced into narrow ridges.
When the skull is seen in dorsal view, the ventral orbital
margin protrudes somewhat more laterally than the dorsal mar-
gin, as long as the latter is formed by the prefrontal. Further
posteriorly, the dorsal margin, now formed by postfrontal and
postorbital, extends somewhat farther laterally.
In anterior view the posterior orbital margin protrudes some-
what more laterally than the anterior margin for almost the en-
tire orbital height, most so dorsally, least so (almost not) ven-
trally. It is also seen that the anterior margin is not completely
regular, but is distinctlyemarginated medially in its middlethird
for a distance that approximately equals the height of the naris.
Fenestra Supratemporalis The fenestra supratemporalis
is, as compared to the orbit, very small (Figs. 2B, 3A, B, 4A).
It is of almost circular, rounded shape, but anterolaterally and
posteriorly it is slightly drawn out. All its margins are gently
convex and merge uninterruptedly into one another. The ante-
118 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 23, NO. 1, 2003
FIGURE 1. Skull of Leptonectes cf. tenuirostris (Conybeare, 1822) from the Lower Lias of Lyme Regis, MBI 007.11 in A, right lateral and B,
left lateral view. Scale bar equals 10 mm.
rior margin (seen in posterior view) is somewhat inclined me-
dioventrally. The posterior margin is quite considerably emar-
ginated ventrally in its lateral half.
Foramen Parietale The foramen parietale is of elongated
teardrop shape, widest anteriorly and tapering posteriorly (Figs.
2B, 3A, B). Its anterior margin is almost straight transversely
with a slight sagittal notch.
Osteology of the Cranium and Mandible
Maxilla The anterior ends of both maxillaries are not pre-
served (Figs. 1, 2A, B, 3A, C, 4C). The maxilla forms almost
the entire ventral narial margin, except for a slight posterior
overlap of the small processus ventralis anterior of the lacrimal.
The alveolar margin and subnarial margin are completely
straight and parallel. Posterior to the external naris, the maxilla
curves gently downward and tapers rapidly. It ends about 90
mm posterior to the external naris, thus forming a long, slender
processus suborbitalis which extends to mid-orbital length and
is not completely visible in lateral but only in ventral view.
Dorsally, the maxilla is sutured to the lacrimal posterior to the
naris for a length of approximately 30 mm. There is a very
slight overlap of the lacrimal on the maxilla, but the suture
between the two elements remains unserrate. Farther posteri-
orly, the maxilla establishes a straight, unserrate sutural contact
with the jugal by which it is also slightly overlapped dorsally.
Remains of seven (left side) and five (right side) teeth are
found in the maxillaries. The posteriormost tooth, seen on the
left side, is found at the level of the posterior narial margin.
Nasal The anterior ends of the nasals are also absent (Figs.
1, 2, 3, 4C). The nasals form the entire dorsal roof of the snout
and extend posteriorly onto the skull roof for about 150 mm
(measured on the left side, where the nasal is more completely
119MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
FIGURE 2. Skull of Leptonectes cf. tenuirostris (Conybeare, 1822) from the Lower Lias of Lyme Regis, MBI 007.11 in A, lateral and B, dorsal
view. Abbreviations:a, angular; d, dentary; f, frontal; j, jugal; l, lacrimal; mx, maxilla; n, nasal; p, parietal; po, postorbital; pof, postfrontal;
prf, prefrontal; q, quadrate; qj, quadratojugal; sa, surangular; sc, sclerotic ring; sq, squamosal; st, supratemporal. Half natural size.
preserved). The contribution of the nasal to the skull roof is
remarkably narrow, as compared to other ichthyosaurs. The ex-
cavatio internasalis is, as mentioned above, widest at the level
of the orbito-narial bar, where it measures about 40 mm trans-
versely. The laterally-protruding wing along the posterior third
of the preserved dorsal narial margin is better preserved on the
right side, where it is 25 mm long and has an almost semicir-
cular outline. Farther anteriorly, the lateroventral margin of the
nasal is slightly convex.
Posterior to the external naris, the nasal sends down a flange
(which is not well preserved on either side) for a length of about
10 mm; the flange is posteriorly largely overlapped by the pre-
frontal. This flange forms an external point-contact with the
lacrimal at about half narial height. Farther posteriorly, a long,
almost unserrate suture with the prefrontal is established.
There is no contact between nasal and postfrontal (because
of the unusual condition that the prefrontal meets the parietal,
as described below). The nasal-prefrontal suture curves mark-
edly medially, from the posterior narial margin onwards, up to
two thirds orbital length. Then it runs approximately parallel to
the long axis of the skull. Posteriorly it becomes lanceolately
serrated. The nasal overlaps the frontal medially, as in most
120 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 23, NO. 1, 2003
FIGURE 3. Skull of Leptonectes cf. tenuirostris (Conybeare, 1822) from the Lower Lias of Lyme Regis, MBI 007.11. A, anterior region in
dorsal view; B, posterior region in dorsal view; C, anterior view. Scale bar equals 10 mm.
ichthyosaurs. The medial naso-frontal suture is squamous and
runs gently anteromedially. On the left side only, a point-con-
tact between the nasal and the parietal is established. The nasals
meet about 65 mm anterior to the foramen parietale and from
there on form the sagittal suture of the skull roof for their entire
preserved length. The posterior end of the nasal is situated at
about mid-orbital length.
Lacrimal The lacrimal forms most of the orbito–narial bar,
the lower half of the posterior narial margin and the posteri-
ormost 15 mm of the ventral narial margin (Figs. 1, 2, 3A, C).
Along the anterior and anteroventral orbital margin, to which
the lacrimal equally contributes, a marked crista lacrimalis la-
teralis is developed, anterior to which the lateral surface of the
lacrimal is essentially flat and inclined anteromedially. Dorsally,
the lacrimal forms a short point-contact with the nasal and a
squamous suture with the anteroventral portion of the prefron-
tal, which overlaps the lacrimal.
The posteroventral suture of the processus lacrimalis poste-
rior with the ramus suborbitalis of the jugal is straight. The
lacrimal forms approximately the first 35 mm of the ventral
orbital margin.
Jugal The jugal has a very low and slender ramus subor-
bitalis which is dorsoventrally flattened for most of its length
(Figs. 1, 2A, 3C). Its lateral margin is produced into a narrow
ridge. Anteriorly, a low, slender spur is pinched in between the
lacrimal dorsally and the processus suborbitalis of the maxilla
121MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
FIGURE 4. A–E, details of the skull of Leptonectes cf. tenuirostris (Conybeare, 1822), MBI 007.11. A, posterior view of right occipital region;
B, the same but in posteroventral and slightly medial view; C, cross-section at the anterior break of the snout; D, ventral aspect of posterior end
of right angular, showing a pathology; E, posterior dentary tooth. Abbreviations:f.op, facet for the opisthotic; m.d.m, area of origin of musculus
depressor mandibulae; pmx, premaxilla; pt, pterygoid; v, vomer; otherwise as in Figure 1.
ventrally. The jugal reaches only slightly anterior to the orbit.
Along the posterior third of the orbit, the ramus suborbitalis
expands dorsally and forms part of the area circumorbitalis. The
lateral ridge merges with the ventral margin of the jugal pos-
teriorly. There is no posteroventral flange at the point where
ramus suborbitalis and ramus postorbitalis meet. Instead they
turn into each other in gently convex curvature. The jugal nev-
ertheless is widest dorsoventrally at this point.
The ramus postorbitalis reaches dorsally to about two fifths
orbital height. Its lateral surface is quite flat. The posteroventral
margin is slightly damaged on both sides. The postorbital over-
laps the jugal, whereas the jugal in turn very slightly overlaps
the quadratojugal. The jugal-quadratojugal contact is not very
well preserved and was apparently very short (but could have
been considerably longer originally). The postorbital-jugal con-
tact more or less parallels the posteroventral and posterior or-
bital margin.
Postorbital The postorbital forms almost the entire poste-
rior margin of the orbit (Figs. 1, 2, 4A). The pars postorbitalis
is laterally flattened and contributes to the partial area circu-
morbitalis. It is delimitated by a pronounced ridge, the crista
postorbitalis lateralis, posteriorly. The lamina posterior (Maisch,
1998a) of the postorbital is almost completely covered by the
posteriorly adjacent temporal elements, supratemporal, squa-
mosal and quadratojugal. Between the supratemporal and squa-
mosal and along the anterior squamosal margin, a very slight
part of it is, however, exposed. Probably this is due to slight
damage of the anterior margin of the squamosal. The suture
between postorbital and supratemporal is remarkably long.
Dorsally, the postorbital is completely covered by the post-
frontal, but it remains exposed on the underside of the dorsal
orbital margin, where it reaches anteriorly for 20 mm to con-
tribute to the supraorbital skull roof. The supraorbital portion
is twisted at an angle of 90
8
to the postorbital portion. The
lamina posterior to the postorbital also forms a strong angle of
almost 90
8
with the pars postorbitalis of the bone, so that the
ventral temporal elements, squamosal and quadratojugal, are
almost entirely in an occipital position, as it is typical for the
members of the Leptonectidae (Maisch and Matzke, 2000).
Postfrontal The postfrontal is a large bony plate which
forms the entire anterior half of the orbito–temporal bar, as well
as much of the supraorbital skull roof (Figs. 1, 2, 3A, B, 4A).
There is no overlap on the prefrontal anteriorly along the dorsal
orbital margin, but farther medially the postfrontal strongly
overlaps the prefrontal for about 30 mm; then, however, the
prefrontal overlaps the postfrontal. The pre–postfrontal suture
is therefore a classic example of a turnover suture, as found in
many ichthyosaurs where the skull bones show a large amount
of telescoping.
Medially, the postfrontal overlaps the parietal and forms most
of the anterior margin of the fenestra supratemporalis, as well
as the anterior third of the lateral margin. Posteriorly, the su-
122 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 23, NO. 1, 2003
pratemporal overlaps the postfrontal, whereas the postfrontal
covers the postorbital posteroventrally. Along the anterolateral
margin of the fenestra supratemporalis, the postfrontal is slight-
ly elevated above the general surface of the surrounding skull
roof.
Prefrontal The prefrontal is of most unusual size and
shape (Figs. 1, 2, 3A, B). In fact it is the largest skull roof
element. It forms the entire anterior half of the dorsal orbital
margin and the dorsal third of the anterior orbital margin, along
both of which it is exceedingly thin. The lamina anteroventralis
of the prefrontal strongly overlaps the lacrimal, whereas it is
itself slightly covered by the nasal, and almost reaches the pos-
terior margin of the external naris. The medial and lateral por-
tions of the lamina anteroventralis form a strong angle, so that
a trough-like depression extends medial to the orbital margin
for approximately 40 mm. Farther posteriorly, the entire dorsal
surface of the prefrontal is gently convex or flat. Posteriorly,
the above described turnover suture with the postfrontal is
formed. Where the prefrontal covers the postfrontal, it extends
back up to a point only 20 mm anterior to the fenestra tem-
poralis and forms an extensive posterior suture with the parietal,
which it even overlaps.
On the right side there is also a long medial suture with the
frontal, which is overlapped by the prefrontal, whereas on the
left side the two elements remain separated from each other as
the nasal and parietal form a short point-contact. Where the
prefrontal-frontal suture is exposed, it is completely straight and
parallel to the long axis of the skull. This example of left-right-
variation (which is most probably due to a very slight damage
of the right nasal) should be borne in mind if one uses detailed
sutural relationships of skull roof elements for taxonomic pur-
poses in Jurassic ichthyosaurs.
Frontal The frontal has a restricted exposure on the dorsal
skull roof (Figs. 2B, 3A, B). This is mainly due to the fact that
all the immediately surrounding elements, nasal, prefrontal and
parietal, overlap the frontal to some extent. Nevertheless, the
externally visible portion of the frontal is of considerable size,
particularly if compared to Eurhinosaurus longirostris (Maisch
and Matzke, 2000) where the frontal is completely covered by
the surrounding skull roof bones. The frontal extends 70 mm
anterior to the foramen parietale. The maximum width of the
two frontals is 37 mm (measured slightly anterior to the fora-
men parietale). The lateral margins of the foramen are almost
entirely formed by the frontals, which posteriorly underlap the
parietals considerably.
Each frontal is slightly convex transversely. The two ele-
ments together enclose a deep groove along their sagittal suture
for their entire length. This groove becomes both wider and
deeper anteriorly. At the anterior end of the frontals it is 7 mm
wide and the frontals are almost vertically oriented plates along
the lateral margins of the groove.
Parietal The parietals form the wide intertemporal bar
(Figs. 2A, B, 3A, B, 4A, B). Anteriorly they are in broad con-
tact with the prefrontals along an 18 mm long transverse suture.
Anterolaterally the parietal is overlapped and also partially un-
derlapped by the postfrontal, so that the anterolateralmost ex-
tension of the parietal is sandwiched between dorsal and ventral
medial lamellae of the postfrontal. Externally, the parietal con-
tributes less than half of the anterior margin of the fenestra
supratemporalis, but internally it extends farther laterally with
its pinched-in portion for about two thirds of the anterior mar-
gin’s length.
The lateral margin of the dorsal parietal surface is gently
concave and forms the entire medial border of the fenestra su-
pratemporalis. Medially, the dorsal parietal surface is pro-
nounced into a sagittal elevation, which is narrower and more
ridge-like lateral to the foramen parietale, but becomes flatter
and wider posteriorly. The lamina descendens of the parietal,
which is vertically oriented, is exposed on the right side for the
entire length of the fenestra supratemporalis. Posteriorly it is
18 mm, centrally about 20 mm high. The lamina descendens is
set off by a distinct ridge from the dorsal surface and merges
into the dorsal ridge of the occipital surface.
The posteriormost portion of the dorsal parietal surface is set
off from the rest of the bone by a distinct anteromedially-run-
ning ridge which also extends from the dorsal ridge of the oc-
cipital ramus. This surface slopes gently posteroventrally. Lat-
erally it merges uninterruptedly into the more steeply inclined
posterior surface of the occipital ramus. Laterally, the occipital
ramus is sutured to the ramus occipitalis of the supratemporal.
The suture is situated about 40 mm lateral to the midline on
the dorsal edge, then runs laterally for 10 mm and finally turns
directly ventrally. The occipital ramus of the parietal accom-
panies the ramus occipitalis of the supratemporal for its entire
height and terminates immediately above the opisthotic facet
on the occipital surface of the supratemporal. On the anterior
surface of the posterior temporal bar, the supratemporal-parietal
suture runs essentially dorsoventrally, whereas it is strongly ser-
rate on the ventral surface.
The ventral surface of the posterior temporal bar is more or
less flat and inclined anteroventrally. The anteroventral margin
of the ventral surface is reinforced by a sharp crest, which ex-
tends onto the supratemporal. Another ridge on the ventral sur-
face extends also onto the supratemporal where it forms the
anteromedial border of the opisthotic facet. From this ventral
ridge, a thin triangular extension protrudes ventrally at the level
of the posterior third of the medial border of the fenestra tem-
poralis. This extension must have established contact with the
epipterygoid, of which nothing is exposed.
Supratemporal The supratemporal is the largest element
in the postorbital skull segment and of very complex shape
(Figs. 1, 2, 3B, 4A, B). It forms the lateral half of the posterior
border of the fenestra supratemporalis and posterior temporal
bar by means of a stout, massive ramus occipitalis which is 20
mm long along the dorsal margin of the posterior temporal bar
and 29 mm deep at the level of the anteromedial border of the
opisthotic facet. It also forms two thirds of the lateral border
of the fenestra supratemporalis, being replaced farther anteri-
orly by the postfrontal, which it overlaps extensively in contrast
to Maisch’s (1999) hypothesis, which assumed (with due cau-
tion) the opposite condition in a badly crushed skull of L. ten-
uirostris from Germany.
The ramus anterior of the supratemporal is 44 mm deep at
the posteroventral corner of the skull table and is only separated
by 10 mm from the orbital margin on the right side. In dorsal
view, the supratemporal appears to enter the orbital margin, but
in fact it remains clearly separated by the postfrontal and post-
orbital, as seen in lateral aspect.
The occipital surface of the supratemporal is ornamented by
radial striations and rugosities where the occipital and anterior
rami meet. The centre of these is situated slightly laterodorsal
to the opisthotic facet and probably marks the point of origin
of the musculus depressor mandibulae. The opisthotic facet is
situated as a rounded depression in the corner between the oc-
cipital, anterior, and ventral rami. It is ca. 22 mm high and 13
mm wide on the right side. Anteromedially, it is bordered by a
ridge, which continues onto the ventral surface of the posterior
temporal bar. The anteroventral ridge of the posterior temporal
bar also extends onto the ventral occipital ramus of the supra-
temporal medial and anterior to the opisthotic facet in such a
way that an anteriorly-running and medially-facing concavity is
formed medial to the facet, which is overhung dorsally by a
medially protruding flange. The suture with the parietal is very
complexly serrated in this area. The medially-protruding flange
is definitely formed by the parietal, as is the dorsal part of the
anteromedial border of the opisthotic facet.
123MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
The ventral ramus of the supratemporal is an extensive, thin
plate of bone with a convex medial margin. It covers much of
the lamina ascendens of the pterygoid and the ascending lamina
of the quadrate in posterior view. The supratemporal-pterygoid
suture is particularly complex and deeply serrated, whereas
there appears to be no firm sutural contact with the quadrate.
Laterally, the ventral ramus does not reach as far ventrally as
it does medially, where contact with the pterygoid is estab-
lished.
Laterally, the ventral ramus is in contact with the squamosal,
which it slightly overlaps, in a rather straight suture. The an-
terior ramus also establishes a rather straight anteroposteriorly-
running suture with the squamosal ventrally, with no major
overlap between the two, as well seen on the left side, where
the squamosal is lost.
Squamosal The squamosal is a small, thin element, which
forms part of the posterolateral surface of the cheek (Figs. 1,
2, 4A). Dorsally, it contacts the anterior ramus of the supratem-
poral. Posteriorly, it is in contact with the lateral margin of the
ventral ramus of the supratemporal for ca. 10 mm. Then, for a
length of about 20 mm, it contacts the lateral margin of the
lamina ascendens of the quadrate.
The squamosal is somewhat depressed with respect to the
surrounding bones and, as seen on the left side, where it is lost,
was not supported internally by an extensive lamina posterior
of the postorbital.
Ventral to the contact with the quadrate, the squamosal is
drawn out into a narrow triangular tip medial to the quadrato-
jugal, which takes part in the lateral border of the foramen
quadrati. Anterolaterally, it forms a rather straight suture with
the postorbital, which it overlaps.
Quadratojugal The right quadratojugal is incompletely ex-
posed (Figs. 1, 2A, 4A). The processus quadratus is still matrix-
covered and the remaining part of the bone is slightly damaged,
particularly along its anteroventral border.
The quadratojugal is small. Anteroventrally it forms an ap-
parently very short contact with the jugal, which it overlaps.
Anterodorsally, a short suture is made with the postorbital,
which is also covered by the quadratojugal to some extent.
There is a moderately developed jugal-quadratojugal notch. The
exposed portion of the quadratojugal lateral to the foramen qua-
drati is convex mediolaterally and about 13 mm wide. Along
the dorsal margin, which is apparently somewhat squashed over
the squamosal, it is 17 mm wide. The quadratojugal forms the
ventral half of the exposed part of the lateral margin of the
foramen quadrati, which is 14 mm wide at maximum and seen
for a height of 18 mm. The posteroventral portion is partially
covered by a heavily damaged, unidentifiable element.
Quadrate Only part of the lamina ascendens of the quad-
rate is exposed (Figs. 2A, 4A, B). It is apparently a rather robust
bony plate, convex transversely and slightly convex dorsoven-
trally, more so in the medial half. Laterally, it contactsthe squa-
mosal and farther ventrally forms the medial border of the fo-
ramen quadrati. Dorsally and dorsomedially it is covered by the
lateral half of the ramus ventralis of the supratemporal, and
ventromedially by the lamina ascendens of the pterygoid with
which a straight, medially concave suture is formed. The con-
dylar area is completely covered by matrix and the abovemen-
tioned unidentifiable element.
Vomer Along the anterior cross break is seen a cross sec-
tion through both vomers, which are ventrally supported by the
posterior palatal processes of the premaxillaries (Fig. 4C). They
show no noteworthy features.
Pterygoid Only part of the posterior and mediodorsal sur-
faces of the right pterygoid is well exposed (Fig. 4A, B). The
lamina ascendens is partially covered anteromedially by the me-
dial part of the ramus ventralis of the supratemporal, with which
it establishes a serrate/squamous sutural contact. Farther later-
ally it in turn covers the ventromedial portion of the lamina
ascendens of the quadrate.
From the point where the supratemporals ventral ramus
reaches deepest ventrally over the pterygoid, a pronounced
ridge extends straight in posteromedial direction on the dorsal
surface of the pterygoid. Lateral to this ridge, about 5 mm pos-
teromedial to the ventralmost extension of the supratemporal, a
small foramen, only 2 mm in diameter, is situated. The dorsal
pterygoid surface medial to the ridge, which is part of the dorsal
surface of the medial pterygoid wing, is deeply concave trans-
versely.
Dentary Only a small portion of each dentary is preserved
as a thin lamella of bone situated dorsolaterally on the crista
supraangularis (Figs. 1, 2A, 3C). Four teeth remain in the right,
three in the left dentary. As preserved, the bone reaches back
only for about one quarter orbital length, but it is apparently
incomplete on both sides. Its ventral extension reaches to the
ventral half of the lower jaw on the anterior preserved end of
the left mandibular ramus, but otherwise is also slightly dam-
aged and incomplete.
Splenial Both splenials are exclusively visible in ventral
view as thin strips of bone applied extensively to the medial
surface of the lower jaw. The splenial terminates somewhat
anterior to the posterior orbital margin on both sides. On the
left mandibular ramus, the splenial ends about 100 mm anterior
to the preserved posterior end of the mandible. Posteriorly, the
splenial tapers out to an extremely narrow splint of bone, but
gets gradually wider anteriorly. At the anterior cross-break, the
ventral width of the splenial is about 8 mm.
The ventral surface is markedly convex transversely. The
bone itself is essentially straight anteroposteriorly. At the an-
terior cross-break, the two splenials are still 20 mm apart. At
their posterior ends they are separated by a distance of 92 mm
as the two lower jaw rami diverge at an angle of 35–40
8
.
Angular The angular extends for the entire preserved
length of the lower jaw (Figs. 1, 2A, 4D). Anteriorly it is very
low, being exposed only for a height of 4 mm as it is dorsally
overlapped by the surangular. Up to two thirds orbital length,
the angular is restricted to the ventral surface of the lower jaw,
then it becomes gradually higher and, at the level of the pos-
terior end of the ramus suborbitalis of the jugal, the angular-
surangular suture turns quite abruptly dorsally. From there on
backwards, the angular forms about half of the external man-
dibular surface, being 21 mm high at maximum.
In ventral view, the angular is strongly convex transversely.
Anteriorly it is only 5 mm wide, but at the posterior end of the
jaw it reaches its maximum width of 16 mm, which is the entire
width of the jaw ramus as it forms the ventral surface com-
pletely. In lateral view, the ventral margin is almost straight up
to three quarters orbital length. Then it bows gently but mark-
edly dorsally, as does the entire lower jaw. The dorsal margin
of the lower jaw at the level of the processus glenoideus is thus
about 23 mm, the ventral margin 17 mm, higher than the an-
terior preserved end.
The right angular shows a pathology (Fig. 4D), which ex-
tends along the ventral and lateroventral margin for a length of
50 mm from the preserved posterior end forwards. It looks like
a healed wound with scarry bone tissue, probably the result of
a bite. Nothing similar is found on the left side.
Surangular Consistent with the structure of the entire low-
er jaw, the surangular is a very low and slender element (Figs.
1, 2A). It overlaps the angular ventrally and is itself dorsally
overlapped by the dentary, as far as the latter is preserved. The
posterior margins of both surangular and angular are damaged
on both sides.
The lateral surface of the surangular is practically flat up to
mid-orbital level. Further backwards it becomes slightly convex
dorsoventrally. Anterior to the processus glenoideus, which is
124 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 23, NO. 1, 2003
FIGURE 5. Measurements of the orbitae and sclerotic rings of MBI
007.11. A, measurements of the internal aperture of the left sclerotic
ring. The circle symbolises the aperture, the lines indicates the direction
of the measurement, and the numbers give the measurements in mm.
B, the same for the right side. C,D, measurements of the plates of the
left sclerotic ring. The small circle symbolises the internal aperture, the
larger circle the outer margin of the sclerotic ring, otherwise as in A,
B.E,F, the same for the right side. G, measurements of the left orbita.
The circle symbolises the orbital aperture; otherwise as above. H, the
same for the right orbita. I, areas of the left and right sclerotic rings
where the sclerotic plates show an externodorsal sulcus (1, indicated in
black), cross sections through sclerotic plates with (1) and without (2)
the sulcus are seen on the right side.
largely exposed on the left side, the lateral surface is concave.
The processus glenoideus is inconspicuous and developed as a
small, triangular dorsal projection. Posterior to it, the dorsal
surangular margin is straight; anterior to it, it is markedly con-
cave.
About 45 mm anterior to the processus glenoideus, a deep
fossa supraangularis is found, above which the surangular is
produced into a pronounced crest, the crista supraangularis,
which is anteriorly covered, as described, by the dentary. At its
posterior end, the crista supraangularis is only 5 mm high; at
the anterior end of the lower jaw (as preserved) it has a height
of 9 mm, but has become low and inconspicuous and probably
tapered out only some cm in advance of the preserved portion
of the mandible. In ventral view it is clearly seen that the sur-
angular is for most of its length, at least ventrally, a very thin
bony plate, 3 mm wide at maximum.
Dentition Only very few teeth of the maxilla and dentary
are available for study (Figs. 1, 2A, 4E). As seen in the anterior
cross-break, the teeth are aulacodontously implanted, as in all
post-Triassic ichthyosaurs. The tooth crowns are very slender,
conical, and about straight with a very slight distolingual cur-
vature. The enamel is macroscopically smooth, as in all lepto-
nectids (Maisch, 1998b) and when examined with a hand lens
(magnification
3
10) shows still no distinct ornamentation. The
largest maxillary tooth crown is ca. 18 mm long. The roots are
not well exposed in any tooth.
Sclerotic Ring Both sclerotic rings are exquisitely pre-
served in their natural position (Figs. 1, 2, 3, 5). Their orien-
tation with respect to the margins of the orbit and the long axis
of the skull is identical on both sides, and it appears that MBI
007.11 is one of the very rare examples of an ichthyosaur skull
with the sclerotic rings in situ.
The detailed measurements of both sclerotic rings and orbits
are given in Figure 3. Each sclerotic ring consists of 17 plates.
As for the orbit, the sclerotic ring has a somewhat ovoid shape
with the long axis of the ellipse oriented at an angle of ap-
proximately 45
8
(left side) and 50
8
(right side) respectively to
the horizontal axis of the skull. The sutures between the indi-
vidual plates are strongly serrate and show strong overlap. The
overlap between individual plates does not accord to any intel-
ligible system. The growth centre of each sclerotic plate is sit-
uated at the point where the external, flat portion of the plate
turns in strong curvature into the internal portion. In the anter-
odorsal quarter of each sclerotic ring, the sclerotic plates show
a marked, trough-like depression on the dorsolateral surfaces of
their internal portions.
Both sclerotic rings appear to be in the position they would
have assumed in vivo with all the eye muscles relaxed. The
posterior margin of the sclerotic ring is situated within the orbit,
whereas the anterior margin protrudes laterally for about 22
mm. In this way the anterior and posterior margins of the in-
ternal aperture of each sclerotic ring are in line and parallel to
the long axis of the skull, and the internal aperture is not visible
when the cranium is seen in frontal view. In this way the animal
had its eyes directed completely laterally and had the largest
visual sector possible within view. It is obvious that if the scle-
rotic rings could be rotated through an angle of approximately
30
8
, in such a way that the anterior portion of the sclerotic ring
is brought closer to the midline and the posterior portion swung
laterally (which was easily possibly by action of the ophthalmic
musculature), the eyes would have been directed so much an-
teriorly that the visual fields of both eyes overlapped, and ste-
reoscopic vision was at least theoretically possible.
TAXONOMIC DISCUSSION AND COMPARISON
Identification of the well-preserved skull MBI 007.11 is dif-
ficult, because the quality of previous studies on leptonectid
cranial osteology is largely poor. The family Leptonectidae en-
compasses the two genera Leptonectes McGowan, 1996, with
the three species L. tenuirostris (Conybeare, 1822), L. solei
(McGowan, 1993), and L. moorei McGowan and Milner, 1999,
and Eurhinosaurus Abel, 1909, with the two species E. longi-
rostris (Owen and Jaeger in Jaeger, 1856) and E. costini
(McGowan, 1986) (the latter is the type species of the genus
Excalibosaurus McGowan, 1986, identified as a junior subjec-
tive synonym of Eurhinosaurus by Maisch and Matzke, 2000).
125MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
In addition there is the nominal species Ichthyosaurus latif-
rons Ko¨nig, 1825, which is based on a three-dimensional skull
(BMNH R1122) from the Lower Lias of Barrow-on-Soar (Ko¨n-
ig, 1825; Lydekker, 1889 contra Owen, 1881; see also Mc-
Gowan, 1989a) and some associated vertebral material. It is
identifiable as a leptonectid, as it combines very large orbits
with small supratemporal fenestrae. A modern description of
this second named leptonectid species, considered as a nomen
dubium by McGowan (1989a), is highly desirable. As it is of
doubtful validity and modern descriptions are lacking, no fur-
ther comparison was made.
Generally, it is evident that MBI 007.11 represents a lepto-
nectid. Except for the extremely long and delicate snout, which
is missing in the specimen, it shows all features considered
diagnostic of the family by Maisch (1998b, 1999), Maisch and
Matzke (2000), and McGowan and Milner (1999), namely, a
very large orbit, small fenestra supratemporalis, extremely short
postorbital skull segment with the quadratojugal rotated almost
entirely to the occipital surface of the skull, and slender, elon-
gated, conical teeth which lack distinct surface ornamentation
of the enamel.
The skull of the genus and species Eurhinosaurus longiros-
tris (Fig. 2H) has been the focus of a recent re-study, some of
the results of which are found in Maisch and Matzke (2000).
This re-study was based on a well preserved three-dimensional
skull which can be well compared to MBI 007.11. E. longiros-
tris shows numerous considerable differences. In the skull roof,
the nasal is much wider posteriorly and totally overlaps the
frontal, which remains unexposed in the articulated state. The
foramen parietale is thus uniquely bordered by the nasal and
parietal. The prefrontal is very small and exposed only as a
narrow strip along the dorsal orbital margin, without any con-
tact to the parietal or frontal. The postfrontal is very short and
stout, even more so than in MBI 007.11. The supratemporal
reaches the dorsal orbital margin. The fenestra supratemporalis
is only about half as large as in MBI 007.11. The orbits are
much more anteriorly directed. There is no wing-like process
of the nasal along the dorsal narial margin etc.
It is therefore evident that MBI 007.11 can not be identified
with E. longirostris. The latter is probably the most highly de-
rived and specialised member of the Leptonectidae. This is in
accordance with it being the youngest species of the family
known (Lower Toarcian). Unfortunately, E. longirostris is the
only other leptonectid that can be compared to MBI 007.11. in
such detail, as it is the only one adequately described.
The second species of Eurhinosaurus,E. costini (McGowan,
1986), is only known from few specimens from the Lower Lias
of England, and only the holotype (BRSMG Cc 881) has hith-
erto been published. It preserves a strongly compressed but
complete and well-prepared skull. Unfortunately, McGowan’s
papers (1986, 1989b) reveal little detail about the cranial oste-
ology of this species, except that it has an overbite that is less
considerable than that in E. longirostris.
Personal observation of the holotype of E. costini revealed,
however, some significant anatomical differences to MBI
007.11. In E. costini the ramus anterior of the supratemporal is
lower and generally less extensive, the quadratojugal is larger
and the jugal is much more evenly rounded and more slender
where the ramus suborbitalis and ramus postorbitalis meet each
other. The fenestra supratemporalis is apparently much smaller
than in MBI 007.11 and comes close to the condition in E.
longirostris. The naris has a strongly bilobed appearance, not
due to a lateral wing of the nasal but to a small processus
lacrimalis narialis, as is, e.g., also found in Temnodontosaurus
nuertingensis (von Huene, 1931; Maisch and Hungerbu¨hler,
1997). From these qualitative anatomical comparisons, it is
clear that MBI 007.11 is not conspecific with the holotype of
E. costini.
L. solei (McGowan, 1993) is a rare species known only from
two specimens (BRSMG Ce 9856, the holotype, and MHN
96270; McGowan, 1993). The skull of the holotype is in a very
bad state of preservation. The second specimen (MHN 96270)
preserves, however, an apparently complete and well preserved
three-dimensional skull. Unfortunately, McGowan (1993:1203,
fig. 5B) only provides a very small photographic figure of rather
low quality in dorsal view of this important specimen, which
is not described in detail.
What can be said from the little information available on the
cranial osteology of L. solei is that it is first a much larger
animal than the one to which MBI 007.11 must have belonged
(skull length in L. solei is over 1,000 mm, whereas the skull of
MBI 007.11 was at maximum 700 mm long, in the upper part
but not even close to the limits of the known size range of L.
tenuirostris, which can reach a skull length of 950 mm; see
McGowan, 1974, 1989a, 1993). It must be noted in this context
that there is no indication that MBI 007.11 represents a juve-
nile. More important than the size difference is that the orbits
are apparently even larger and essentially laterally and dorsally
oriented in L. solei, the interorbital skull roof thus being much
narrower than in MBI 007.11. The fenestra supratemporalis is
much elongated anteroposteriorly whereas in MBI 007.11 it is
almost completely rounded. These differences in the shape, size
and orientation of the major skull openings are, together with
the considerable size difference, convincing arguments against
an identification of MBI 007.11 with L. solei.
L. moorei McGowan and Milner, 1999 is even rarer than L.
solei. It is exclusively known from a single incomplete skeleton
(BMNH R 14370) from the Lower Pliensbachian of Seatown,
Dorset. It is a much smaller animal than MBI 007.11 with a
skull length of only 328 mm. It differs from all other leptonec-
tids in its short snout, which makes it evidently a valid species.
Unfortunately this character can not be directly assessed in MBI
007.11. Nevertheless, the shortness of the snout segment of the
skull of L. moorei is also reflected in some other proportions,
such as the shortness of the external naris, which is otherwise
universally an elongated, low opening in leptonectids. It is clear
that MBI 007.11 had a typically leptonectid, low and elongate
external narial aperture, and therefore almost certainly a very
elongated, typically leptonectid snout which makes it complete-
ly different from L. moorei in this respect.
The details of the cranial osteology that can be learned from
McGowan and Milner’s (1999) figures otherwise indicate an
animal similar to MBI 007.11, and to L. tenuirostris. This is
particularly evident in the structure of the lower jaw, the shape
of the jugal, and the structure of the postorbital skull segment.
In this context some misidentifications of McGowan and Milner
(1999) should be noted. The element that they label as maxilla
(McGowan and Milner, 1999:765, fig. 3) belongs to the pre-
maxilla, showing well the processus subnarialis and supranari-
alis along its posterior border. The element that they label as
squamosal in the same figure is the quadratojugal. The squa-
mosal and supratemporal are situated farther dorsally and not
labelled at all, although the sutures are well seen on the pho-
tograph (McGowan and Milner, 1999:762, fig. 1) and even cor-
rectly indicated in the line drawing (McGowan and Milner,
1999:fig. 3). The difference in size and age, shape of the naris,
and—in all probability—length of the snout makes it impossi-
ble to identify specimen MBI 007.11 with Leptonectes moorei.
Finally, MBI 007.11 has to be compared to the most common
representative of the group in the Rhaetic-Sinemurian beds of
southern England. Already from the size and general shape of
the skull it is highly tempting to identify the new specimen as
a representative of this, the first ever described leptonectid spe-
cies, L. tenuirostris.
Good figures and descriptions of L. tenuirostris are, however,
rare. The few data which can be gained from the classical ac-
126 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 23, NO. 1, 2003
counts of Conybeare (1822), Hawkins (1834, 1840), Owen
(1881), and von Huene (1922) and the modern ones by Mc-
Gowan (1974, 1989a) and Maisch (1999) do, however, not
speak against an identification. General skull proportions, shape
and size of the major skull openings, and shape of the naris,
jugal, postorbital skull segment, lower jaw, and all the individ-
ual skull bones for which sutures have been identified in other
known specimens agree quite well.
The only significant difference is that in the L. tenuirostris
skull described by Maisch (1999:494, fig. 1) the maxilla is ap-
parently excluded from the ventral margin of the external naris.
However, the region in question is not particularly well pre-
served in this specimen. The narial aperture is partially covered
by the jugal and its outlines are not totally clear. There is, fur-
thermore, no direct evidence that premaxilla and lacrimal met
below the naris. On the other hand, there is at least the possi-
bility that the contribution of the maxilla to the external naris
is a variable feature in L. tenuirostris, as it clearly is, e.g., in
Stenopterygius quadriscissus (Maisch and Matzke, 2000). Oth-
erwise the skull described by Maisch (1999) agrees well with
the present specimen. Some slight discrepancies (such as the
tentative different interpretation of the overlap relationships be-
tween postfrontal and supratemporal in Maisch’s paper) can be
explained by the bad preservation, compression, and partial dis-
articulation of the German specimen.
In sum there is little reason to identify MBI 007.11 other
than as Leptonectes tenuirostris (Conybeare, 1822). The differ-
ence in the configuration of the circumnarial bones and the lack
of the snout segment of the skull are, however, an obstacle to
a confident identification. We therefore prefer, for the time be-
ing, to identify the specimen in open nomenclature as Lepto-
nectes cf. tenuirostris, pending a detailed re-description of the
cranial osteology of other leptonectids, currently under study
by C. McGowan and R. Motani (Toronto).
The genus Leptonectes itself cannot be defined, at present,
by a valid synapomorphy. It remains a metataxon, or, in stron-
ger words, a waste-basket for non-eurhinosaurian, basal, lep-
tonectid ichthyosaurs. Whether Leptonectes is monophyletic or
a paraphyletic assemblage of Eurhinosaurus’ successive sister
species is at present impossible to tell. The main reason for this
is the lack of adequate data on and modern descriptions of the
cranial osteology of all species, except E. longirostris (Maisch
and Matzke, 2000) and L. cf. tenuirostris (this study). Before a
complete revision of the crania of all known leptonectids, which
obviously, as the above discussion has shown, display a con-
siderable number of specializations characteristic of individual
taxa, has been undertaken, the phylogeny of the Leptonectidae
cannot be adequately analyzed.
CONCLUSIONS
A well-preserved three-dimensional skull of the Lower Ju-
rassic ichthyosaur Leptonectes cf. tenuirostris (Conybeare,
1822) is described in detail and clarifies the osteology and
three-dimensional structure of the skull of this taxon. It shows
several unique autapomorphies of the skull roof, and also a
trend towards the development of stereoscopic vision, which is
carried further in the genus Eurhinosaurus.
It is shown that L. cf. tenuirostris differs fundamentally in
its cranial osteology from the Upper Liassic Eurhinosaurus lon-
girostris, the only other leptonectid of which the cranial oste-
ology is, at present, completely and well known from three-
dimensional material. It is therefore without question that the
two species, which so far could almost only be distinguished
by the presence or absence of an overbite and by cranial mea-
surements, must be kept apart generically.
Discussion of the taxonomy of the Leptonectidae leads to the
conclusion that only monophyly of the genus Eurhinosaurus is
at present supported by a single convincing synapomorphy,
whereas the genus Leptonectes remains a metataxon, as already
pointed out by Maisch (1999). The phylogeny of the Leptonec-
tidae must, at present, remain largely unresolved due to the lack
of detailed knowledge of the osteology, particularly of the cra-
nia, of most species. Nevertheless this study presents a wealth
of new data which hopefully will form an important step to
arriving at a data set which will make an analysis of leptonectid
relationships possible.
ACKNOWLEDGMENTS
Our sincerest thanks go to Wolf-Dieter Heinrich and Dave
Unwin (Institut fu¨r Pala¨ontologie der Alexander-von-Humboldt-
Universita¨t, Berlin), for the possibility to study the material de-
scribed there and their hospitality and assistance during our stay
at Berlin. Further thanks for access to material go to Rupert
Wild and Rainer R. Schoch (Staatliches Museum fu¨r Naturkun-
de, Stuttgart) and Jon Radley, Roger Clark and Roger Vaughan
(Bristol City Museum and Art Gallery).
LITERATURE CITED
Abel, O. 1909. Cetaceenstudien. 1. Mitteilung: Das Skelett von Eurhi-
nodelphis cocheteuxi aus dem Obermioza¨n von Antwerpen. Sit-
zungsberichte der kaiserlich-ko¨niglichen Akademie der Wissen-
schaften zu Wien, mathematisch-naturwissenschaftliche Klasse
118:241–253.
Blainville, H. M. D., de. 1835. Description de quelques espe`ces de la
Californie, pre´cede´e de l’analyse d’un syste`me ge´ne´ral
d’e´rpetologie et d’amphibiologie. Nouvelles Archives de Muse´um
d’ Histoire Naturelle 4:236–296.
Conybeare, W. D. 1822. Additional notes on the fossil genera Ichthy-
osaurus and Plesiosaurus. Transactions of the Geological Society
of London (2), 1:103–123.
Godefroit, P. 1992. Pre´sence de Leptopterygius tenuirostris (Reptilia,
Ichthyosauria) dans le Lias moye´n de Lorraine belge. Bulletin de
l’Institute Royale des Sciences Naturelles Belgique 62:163–170.
Hawkins, T. 1834. Memoirs of Ichthyosauri and Plesiosauri, Extinct
Monsters of the Ancient Earth. Royal Folio, London, 58 pp.
——— 1840. The Book of the Great Sea Dragons, Gedolim Taninim,
of Moses: Extinct Monsters of the Ancient Earth. W. Pickering,
London, 27 pp.
Huene, F., von. 1922. Die Ichthyosaurier des Lias und ihre Zusammen-
ha¨nge. Gebru¨der Borntraeger, Berlin, 114 pp.
——— 1931. Neue Ichthyosaurier aus Wu¨rttemberg. Neues Jahrbuch
fu¨r Mineralogie, Geologie und Pala¨ontologie, Abteilung B, 65:305–
320.
——— 1948. Short review of the lower tetrapods; pp. 65–106 in A. du
Toit (ed.), Robert Broom Commemorative Volume, Royal Society
of South Africa Special Publications, Pretoria.
Jaeger, G. F., von. 1856. U
¨ber eine neue Spezies von Ichthyosauren
(Ichthyosaurus longirostris Owen et Ja¨ger) nebst Bemerkungen
u¨ber die u¨brigen in der Liasformation Wu¨rttembergs aufgefundenen
Reptilien. Nova acta Academiae caesareae Leopoldino-Carolinae
germanicae naturae curiosorum 25:937–967.
Jaekel, O. 1903. U
¨ber die Epiphyse und Hypophyse. Sitzungsberichte
der Gesellschaft naturforschender Freunde Berlin 2:2–58.
Ko¨nig, C. D. E. 1825. Icones: Fossilium Sectiles. Centuria prima, Lon-
don, 4 pp.
Lydekker, R. 1889. Catalogue of the Fossil Reptiles and Amphibians in
the British Museum (Natural History): Part II. Containing the Or-
ders Ichthyopterygia and Sauropterygia. Trustees of the British Mu-
seum, London, 307 pp.
Maisch, M. W. 1997. The cranial osteology of Ichthyosaurus interme-
dius Conybeare, 1822 from the Lias of Great Britain. Stuttgarter
Beitra¨ge zur Naturkunde, Serie B, 258:1–27.
——— 1998a. The temporal region of the Middle Jurassic ichthyosaur
Ophthalmosaurus—further evidence for the non-diapsid cranial ar-
chitecture of the Ichthyosauria. Neues Jahrbuch fu¨r Geologie und
Pala¨ontologie Monatshefte 1998:401–414.
——— 1998b. A new ichthyosaur genus from the Posidonia Shale
(Lower Toarcian, Jurassic) of Holzmaden, SW-Germany with com-
127MAISCH AND MATZKE—ICHTHYOSAUR CRANIAL OSTEOLOGY
ments on the phylogeny of post-Triassic ichthyosaurs. Neues Jahr-
buch fu¨r Geologie und Pala¨ontologie Abhandlungen 209:47–78.
——— 1999. Leptonectiden und Temnodontosauriden (Ichthyosauria)
aus dem Alpha-O
¨lschiefer (Sinemurium) von Baden-Wu¨rttemberg
(SW-Deutschland). Neues Jahrbuch fu¨r Geologie und Pala¨ontologie
Monatshefte, 1999:490–512.
———, and A. Hungerbu¨hler. 1997. Revision of Temnodontosaurus
nuertingensis (v. Huene, 1931), a large ichthyosaur from the Lower
Pliensbachian (Lower Jurassic) of Nu¨rtingen, South Western Ger-
many. Stuttgarter Beitra¨ge zur Naturkunde, Serie B, 248:1–11.
———, and A. T. Matzke. 2000. The Ichthyosauria. Stuttgarter Beitra¨ge
zur Naturkunde, Serie B, 298:1–159.
McGowan, C. 1974. A revision of the longipinnate ichthyosaurs of the
Lower Jurassic of England, with descriptions of two new species
(Reptilia: Ichthyosauria). Life Science Contributions of the Royal
Ontario Museum 97:1–37.
——— 1986. A putative ancestor for the swordfish-like ichthyosaur
Eurhinosaurus. Nature 322:454–455.
——— 1989a. Leptopterygius tenuirostris and other long-snouted ich-
thyosaurs from the English Lower Lias. Palaeontology 32:409–
427.
——— 1989b. Computed tomography reveals further details of Excal-
ibosaurus, a putative ancestor for the swordfish-like ichthyosaur
Eurhinosaurus. Journal of Vertebrate Paleontology 9:269–281.
——— 1993. A new species of large, long-snouted ichthyosaur from
the English lower Lias. Canadian Journal of Earth Sciences 30:
1197–1204.
——— 1996. The taxonomic status of Leptopterygius Huene, 1922
(Reptilia, Ichthyosauria). Canadian Journal of Earth Sciences 33:
439–443.
———, and A. C. Milner. 1999. A new Pliensbachian ichthyosaur from
Dorset, England. Palaeontology 42:761–768.
Motani, R. 1999. Phylogeny of the Ichthyopterygia. Journal of Verte-
brate Paleontology 19:473–496.
Owen, R. 1881. Monograph on the fossil Reptilia of the Liassic For-
mations. Part 3. Ichthyopterygia. Palaeontographical Society
Monographs 35:83–134.
Sander, P. M. 2000. Ichthyosauria: their diversity, distribution and phy-
logeny. Pala¨ontologische Zeitschrift 74:1–35.
Received 28 February 2001; accepted 19 February 2002.
