with stout roots. Ribs thick and stout. Vertebrae massive.
Pelvic girdle as wide as pectoral girdle; large ischium fused
with pubis, but at proximal end only. Tetradigitate hind-
limb, % length of forefin; femur with two distal facets;
phalanges rounded.
Remarks: The forefin looks very much like that of Yasykovia
Efimov, 1999A, which is probably a subjective junior
synonym for Ophthalmosaurus. This raises the possibility
that this genus might also be synonymous. Maisch &
Matzke (2000B) synonymized Undorosaurus with Ophthalmosaurus,
but the two differ in significant regards: the
pubis and ischium are not completely fused, and the teeth
are large. Therefore, it appears that Undorosaurus is distinct
from Ophthalmosaurus, but the two appear to have close
affinities.
Undorosaurus gorodischensis Efimov, 1999b
Fig. 92
Undorosaurus gorodischensis Efimov, 1999B: 52
Undorosaurus nessovi Efimov, 1999B: 56
Undorosaurus khorlovensis Efimov, 1999B: 57
Holotype: UPM EP-II-20, fragmentary skeleton.
Diagnosis: As for genus.
Occurrence: Ul’yanovsk Province and Moscow region,
Russia.
Stratigraphic range: Upper Jurassic (middle Volgian [Ti-
thonian]).
Remarks: The two other species described, U. nessovi and
U. khorlovensis, are distinguished on features that appear to
be due to differences in maturity. Therefore, they are synonymized
here with Undorosaurus gorodischensis.
Genus Ophthalmosaurus Se e l e y , 1874
Ophthalmosaurus Seeley, 1874: 699
Sauranodon Marsh, 1879: 86 [preoccupied]
Baptanodon Marsh, 1880A: 491
Microdontosaurus Gilmore, 1902: 3
Apatodonosaurus Mehl, 1927: 234
Apatodonosaurus Mehl, 1928: 119
Ancanamunia Rusconi, 1942: 2
Paraophthalmosaurus A rkhangelsky, 1997: 87
Khudiakovia Arkhangelsky, 1999: 88
Yasykovia E fimov, 1999A: 92
Paraophthalmosaurus; Maisch & Matzke, 2000B: 88
Khudiakovia; Maisch & Matzke, 2000B: 90
Yasykovia; Maisch & Matzke, 2000B: 89
Type species: Ophthalmosaurus icenicus Seeley, 1874.
Diagnosis: Humerus with three distal facets, the anterior
one, the smallest, articulating with a preaxial accessory
element; facet for ulna smaller than radial facet and set off
Fig. 92. Undorosaurus gorodischensis, holotype (UPM EP-II-20 (572)),
redrawn from Ef im o v (1999B). Scale equals 50 mm.
obliquely so that its posterior margin is well proximal of its
anterior margin; free margin of anterior distal facet more
acutely pointed than that of posterior distal facet; posterior
edge of shaft pinched, often into a flange; radius probably
wider (anteroposteriorly) than ulna, but is not as deep
(proximodistally); phalanges rounded, not polygonal, and
are not tightly packed, free margins rugose; elements in
longest digit probably <20. Teeth not firmly attached and
usually lost. Orbit large, orbital ratio >0.20; scleral ring
occupies most of orbit; postorbital region narrow. Basioc-
cipital with essentially flat anterior surface, with only an
incipient peg, if at all; extracondylar area less extensive
than condyle, from which it is clearly set off, and is largely
occupied by an extensive pair of stapedial facets. Pelvic
girdle bipartite, ischium and pubis fused to form single
element, foramen marking their union. Preflexural vertebrae
<80.
Remarks: There has been much variation in opinion over
whether Ophthalmosaurus and Baptanodon are distinct. For
example, Lydekker (1888) initially synonymized them, but
later (1889A) changed his mind. Knight (1903), arguing
that they were distinct genera, gave several distinguishing
features. These included that the distal articular facets of
the humerus were elongated in the plane of articulation in
Baptanodon, but at right angles in Ophthalmosaurus, which
is incorrect. Worthy of “special consideration” was that in
Baptanodon the femur had only two distal facets, while Ophthalmosaurus,
by implication, had more than two (Knight,
1903: 79). This again is incorrect. Gilmore (1905:115), who
also considered them distinct genera, repeated the latter
error. Combining the characters given by the various authors
for the two genera, Gilmore (1905:120,124) summarized
their distinguishing features. His comparisons are
quite long, but if they are compared in detail, there are only
two differences between the two genera: in Ophthalmosaurus
“clavicles (typically) separate with the interclavicle
wedged in between them”, whereas it was “firmly united”
in Baptanodon; and the teeth were “perhaps confined to the
anterior portion of the jaws” in Ophthalmosaurus, but “extended
the entire length” in Baptanodon. Neither assertion
is valid. A ndrews (1 9 1 0 :4 9 ) argued that there was only one
genus, and clearly described how the clavicles were firmly
united in Ophthalmosaurus, braced by the interclavicle. Gilmore’s
(1905: 100) claim that the teeth in Baptanodon “extended
along the full length of the jaw” was entirely speculative,
based on the evidence of “faint alveolar partitions
preserved on the inner wall of the right dental groove”
(Gilmore, 1905: 99). The only teeth reported in situ in
Gilmore’s (1902: 9 13) preliminary account, were located
“near the end of the snout”. In her comprehensive account
of Ophthalmosaurus, Kirton (1 9 8 3 :7 0 -7 2 , figs. 14-15) depicted
the teeth as occurring throughout the length of the jaws.
As mentioned elsewhere, there are no complete, articulated
skeletons of Ophthalmosaurus from England. Instead,
the material, which is extensive, is represented by
isolated bones or associations of skeletal elements. Skeletal
reconstructions are, therefore, fraught with difficulties, exacerbated
by the usual problem of distortion. In his study
of Ophthalmosaurus, A ppleby (1956) mentioned how variable
some of the elements were, and how they differed in
their degrees of ossification. Compare, for example, the
differences in shape of two supraoccipitals figured by A ppleby
(1956: figs. 3 ,1 2 ) . Given these problems, it is difficult
to place very much confidence in his stylized reconstruction
of the back o f the skull of Ophthalmosaurus (Appleby,
1956: fig. 21). Gilmore (1905) presented a reconstruction
for the back of the skull of Baptanodon, and, although he
had the advantage of working with two three-dimension-
ally preserved skulls, these were not without distortion.
This can be seen clearly in the posterior view of the skull
that was used for his reconstruction (Gilmore, 1905: pi. 11).
In comparing the cranial reconstructions of Baptanodon
and Ophthalmosaurus, Appleby (1956: 44 2 ) listed six features
in which they differed, using these to support their
separate identities (Table 3). CM considers these minor
differences more apparent than real, attributable to inaccuracies
in both reconstructions. Kirton (1983), whose cranial
reconstruction for Ophthalmosaurus differed from that of
A ppleby (1956), reached a similar conclusion.
Much of the English material referable to Ophthalmosaurus
derives from the Oxford Clay, primarily in the vicinity
of Peterborough where commercial brick works have
provided many collecting opportunities. There is also
material from the Kimmeridge Clay of several counties,
including Dorsetshire and Oxfordshire. Most of the material
is therefore Middle to Late Jurassic in age. Isolated
elements, mostly basioccipitals and humeri, also occur in
the Upper Greensand of Cambridgeshire. Whereas some
of these can be identified as Brachypterygius cantabrigiensis,
others are referable to Ophthalmosaurus. This extends the
geological range of the latter into the Lower Cretaceous
(Albian).
Arkhangelsky (1999) erected a new monotypic genus
Khudiakovia on a humerus with four associated elements
from the Callovian of Russia. The humerus is typically
ophthalmosaurian, and the genus is accordingly synonymized
with Ophthalmosaurus.
The type species of Paraophthalmosaurus, P. saveljevien-
sis Arkhangelsky, (1997: 87, figs. 1, 2d) was founded upon
a partial skull with associated partial forefin, incomplete
pectoral girdle, ribs and vertebrae, from the Upper Jurassic
(Tithonian) of eastern Saratov Province, Russia. A rkhangelsky
(1997:87) compared his new genus with Ophthalmosaurus,
but found several distinct differences. The most
significant of these was that the radius was semilunate,
and the coracoid had an obliquely straight margin, rather
than being rounded. In both of these features, Paraophthalmosaurus
was quite unlike that of any other post-Triassic
ichthyosaur. This raises the question of whether these two
features are natural. The single coracoid lies with its posterior
margin aligned approximately along the edge of the
Table 3. Differences between Baptanodon and Ophthalmosaurus given by A ppleby (1956).
Baptanodon
Most of foramen magnum is formed by supraoccipital
Parietals wrap round supraoccipital and extend
downwards almost to exoccipital
Supratemporal meets parietal close to supraoccipital
Opening between opisthotic, exoccipital, supraoccipital,
parietal and supratemporal is small
Opening between stapes and opisthotic passes sharply inwards
Skull is little wider than high
Ophthalmosaurus
Most of foramen magnum is formed by exoccipitals
Parietals diverge away from supraoccipital only touching
it on dorsal side
Supratemporal meets parietal over distal end of opisthotic
Opening between opisthotic, exoccipital, supraoccipital,
parietal and supratemporal is large
Large opening enclosed by stapes, opishotic, supratemporal
and quadrate
Skull is much wider than high