C allaway, 1994). This feature, however, later turned out to
be plesiomorphic for ichthyopterygians (Motani 1999B).
Therefore, at least some of the fragmentary specimens
assigned to Cymbospondylus may not actually belong to this
genus.
Cymbospondylus piscosus Leidy, 1868
Cymbospondylus piscosus L eidy, 1868:178
Cymbospondylus petrinus L eidy, 1868: 178
Cymbospondylus grandis Leidy, 1868: 178
CymbospondylusO) grandis L eidy, 1868; Merriam, 1902:106
Holotype: A neotype should be designated (see Remarks).
Diagnosis: Cymbospondylus with at least some mesiodis-
tally compressed tooth crowns.
Occurrence: West Humboldt Range and New Pass Range,
northern Nevada, USA.
Stratigraphic range: Prida Formation; Middle Triassic
(Anisian).
Remarks: L eidy (1868) erected three species within Cymbospondylus,
C. piscosus, C. petrinus, and C. grandis, without
specifying a type species. Merriam (1902) designated C.
piscosus the type species because it was the first in L eidy’s
(1868) list, and this designation rules according to the
principle of the first revisor (contra Motani, T999B). He
subsequently synonymized C. grandis with C. petrinus
(Merriam, 1908), thus recognizing two species among L ei-
dy’s (1868) original material. The two species differ in the
shape of amphicoelous concavities of vertebral centra as
well as in size (Merriam, 1902, 1908). In many vertebral
centra of C. petrinus, the amphicoelous concavity is restricted
to the middle part, leaving somewhat flat areas peripherally
(Merriam, 1908: fig. 135). This is unlike the vertebrae
known for C. piscosus, in which the concavity occupies the
entire articular surface without leaving circumferential flat
areas (Merriam, 1908: fig. 136). This difference, however,
may not be as diagnostic as it seems. In basal ichthyopterygians,
the shape of amphicoelous concavities changes
within a single individual depending on the position of a
vertebra in the vertebral column. The more posterior the
position, the more restricted the amphicoelous concavity
to the middle, and the more extensive the peripheral flat
area. This transition may occur more anteriorly in some
species than in others. In C. petrinus, the peripheral flat
area becomes conspicuous as far anteriorly as the mid or
anterior dorsal area (Merriam, 1908: fig. 135), but at least
the cervical and anteriormost dorsal vertebrae are without
such flat areas (UCMP 9950). Therefore, it is likely that the
holotype of C. piscosus represents the anteriormost dorsal
vertebral series.
The holotype of C. piscosus comprises five vertebrae,
only one of which is complete. Therefore, it is difficult to
establish the diagnostic features of the species. This poses
a nomenclatural problem: If C. piscosus were to be found a
nomen dubium, the generic name Cymbospondylus would
become one as well. However, another species of Cymbospondylus,
C. petrinus, is known from well-preserved
specimens described by Merriam (1905, 1908). The most
complete skeleton of C. petrinus (UCMP 9950) has been
treated as the best example of Cymbospondylus for over 90
years. In the interest of nomenclatural stability, it seems
most appropriate to designate this specimen as the neotype
of C. piscosus in accordance the Fourth Edition of the
ICZN. An appropriate formal appeal to the Commission is
in preparation.
Cymbospondylus buchseri Sander, 1989
Cymbospondylus buchseri Sander, 1989: 164
Holotype: PIVIL'Z T 4351, anterior half of the skeleton.
Diagnosis: Cymbospondylus with short humerus (see Remarks).
Occurrence: Monte San Giorgio, Tessin, Switzerland.
Stratigraphic range: Grenzbitumen horizon; Middle Triassic
(Amsian-Ladinian boundary).
Remarks: The best-known skeleton referred to Cymbospondylus
piscosus (UCMP 9950) has a relatively small humerus
with poorly ossified head, and thus the animal was
not fully mature despite its estimated length of 9.1 m (Merriam,
1908). The holotype of Cymbospondylus buchseri is
smaller, with an estimated length of about 5.5 m (Sander,
1989). It has a humerus that is relatively wide compared to
that of Cymbospondylus piscosus, and its-head is very poorly
ossified, indicating immaturity. Motani & You (1998B)
found that the L/W ratio of the radius of the basal ichthyosaur
Chaohusaurus geishanensis increased during growth.
Unpublished data show that the humerus also became
more elongated with, growth in this species.-Considering
the close similarity of humeral shape between the two
genera (Motani, 1999A), it seems reasonable to assume
that a similar growth-related elongation of the humerus
occurred in Cymbospondylus. If so, the humeral shape of C.
buchseri possibly approached that of Cymbospondylus piscosus
with growth.
(Unranked) Hueneosauria
M a isc h et M a tzke, 2000B
Definition: The last common ancestor of Mixosaurus cor-
nalianus and Ophthalmosaurus icenicus, and all its descendants.
Suborder Mixosauria M otan i, 1999B
Definition: All hueneosaurians more closely related to
Mixosaurus comalianus than to Ophthalmosaurus icenicus.
Family Mixosauridae B a u r , 1887A
Definition: The last common ancestor of Mixosaurus cor-
nalianus and M. nordenskioeldii, and all its descendants.
Diagnosis: Premaxilla posteriorly pointed, scarcely entering
external naris; long sagittal crest reaching nasal; large
anterior terrace of supratemporal fenestra, reaching nasal;
pubis more than twice as large as ischium; high, narrow
neural spines; mid-caudal vertebral centra with increased
size; posterior teeth more robust than anterior ones; slightly
concave terrace present proximally in humerus, anterior
to humeral head.
Genus Mixosaurus Baur, 1887A
Mixosaurus Baur, 1887A: 2
Phalarodon Merriam, 1910: 3 82
Sangiorgiosaurus Brinkmann, 1998: 132
Contectopalatus Maisch & Matzke, 1998B: 113
Type species: Mixosaurus comalianus (Bassani, 1886).
Diagnosis: As for the family.
Mixosaurus comalianus (Bassani, 1886)
Ichthyosaurus comalianus Bassani, 1886: 20
Mixosaurus comalianus; Baur, 1887A: 2
Syntypes: Bassani (1886) stated that there were five skeletons
belonging to this species, four complete and one
partial, stored at Museo Civico di Storia Naturale di Milano.
He did not specify the holotype, so these five should
be considered syntypes. Repossi (1902) figured the smallest
individual of the five and parts of the others, but did not
select a lectotype. Pinna (1967) reported that 12 (not five)
specimens of M. comalianus, resulting from the excavation
in 1863-1878, were lost during a fire in 1943, including the
one he considered the holotype (MCSNM 14624).
Neotype: Brinkmann (1999) officially designated a neotype.
The designation of the neotype (MCSNM type no. 45)
by Pinna (1967) did not satisfy the requirements of ICZN
(Article 75d). It was neither associated with a revision in
which he found it necessary to designate a neotype for
nomenclatural stability nor did he specify the characters
upon which the identity of the proposed neotype was
based.
Diagnosis: Small Mixosaurus, probably not very much
larger than 100 cm; tooth size small relative to skull width
(less than 0.05); anterior teeth well spaced.
Occurrence: Besano region, Italy, and Monte San Giorgio,
Tessin, Switzerland.
Stratigraphic range: Grenzbitumen horizon; Middle Triassic
(Anisian-Ladinian boundary).
Mixosaurus atavus (Quenstedt, 1852)
Ichthyosaurus atavus Quenstedt, 1852
Mixosaurus atavus atavus; F raas, 1891: 38 (as M. atavus var.
minor)
Mixosaurus atavus; Huene, 1916: 3
Contectopalatus atavus; Maisch & Matzke, 1998B: 115
Lectotype: Maisch & Matzke (1998B: 107 and fig. 1) designated
a lectotype.
Diagnosis: Small Mixosaurus, similar in size to M. cornali-
anus; tooth size relative to skull width large (more than
0.1); teeth laterally compressed, at least posteriorly.
Occurrence: Widely distributed throughout the Germanic
Basin: Althengstett, Riidersdorf, and many other localities
in Germany; Gogolin, Gomy Slask, Poland.
Stratigraphic range: Lower Muschelkalk; Middle Triassic
(Anisian-?Ladinian).
Remarks: Maisch & Matzke (1998B, 2000B) used the publication
date of 1851-1852 for this species, but it should be
1852 following ICZN Article 21. The designation of a neotype
by Maisch & Matzke (1998B) is invalid (Motani,
1999C). Maisch & Matzke (1998B) diagnosed the species
as a large mixosaurid with the skull length of up to 30 cm.
However, there is insufficient evidence to infer such a large
size for the species. Fragments of ichthyosaurs larger than
M. atavus do occur in Muschelkalk, and have been referred
to Mixosaurus^) major H uene, 1916. However, they are
non-diagnostic.
Fraas (1891) found two varieties of Mixosaurus atavus
(Quenstedt, 1852), naming them M. atavus var. minor and
M. atavus var. major, the former being typical of the species.
Huene (1916) raised the latter variety to the species rank,
and therefore the two names are deemed to be subspecific
since 1891 (ICZN Article 45g). The name of a nominotyp-
ical subspecies has to be in accordance with ICZN Article
47, so the two subspecific names are M. atavus atavus
(Quenstedt, 1852) and M. atavus major Fraas, 1891. The
latter was later found non-diagnostic (Mazin, 1983B), leaving
only the former as valid.
Mixosaurus nordenskioeldii (Hulke, 1873)
Ichthyosaurus nordenskioeldii Hulke, 1873: 4
Mixosaurus nordenskioeldii; Dames, 1895: 1047
Cymbospondylus(?) nordenskioeldii; Merriam, 1908: 148