■■I
NOTICE: Return or renew all Library Materials! The Minimum Fee tor each Lost Book is $50.00.
The person charging this material is responsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below.
Thett, mutilation, and underlining ot books are reasons for discipli- nary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400
UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN
novo saw
LI6I— O-I096
ILLINOIS LI Y ST JRBAN/\- PAIGN
AUG 1 3 tPPR
550-O
FlELDlANA
Geology
NEW SERIES, NO. 28
Osteology of Simosaurus gaillardoti and the Relationships of Stem-Group Sauropterygia
Olivier Rieppel
December 30, 1994 Publication 1462
PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY
Information for Contributors to Fieldiana
General: Fieldiana is primarily a journal for Field Museum staff members and research associates, although manuscripts from nonaffiliated authors may be considered as space permits.
The Journal carries a page charge of $65.00 per printed page or fraction thereof. Payment of at least 50% of page charges qualifies a paper for expedited processing, which reduces the publication time. Contributions from staff, research associates, and invited authors will be considered for publication regardless of ability to pay page charges, however, the full charge is mandatory for nonaffiliated authors of unsolicited manuscripts. Three complete copies of the text (including title page and abstract) and of the illustrations should be submitted (one original copy plus two review copies which may be machine-copies). No manuscripts will be considered for publication or submitted to reviewers before all materials are complete and in the hands of the Scientific Editor.
Manuscripts should be submitted to Scientific Editor, Fieldiana, Field Museum of Natural History, Chicago, Illinois 60605-2496, USA.
Text: Manuscripts must be typewritten double-spaced on standard-weight, 8Vfe- by 11-inch paper with wide margins on all four sides. If typed on an IBM-compatible computer using MS-DOS, also submit text on 5V4-inch diskette (WordPerfect 4.1, 4.2, or 5.0, MultiMate, Displaywrite 2, 3 & 4, Wang PC, Samna, Microsoft Word, Volkswriter, or WordStar programs or ASCII).
For papers over 100 manuscript pages, authors are requested to submit a Table of Contents," a "List of Illustrations," and a "List of Tables" immediately following title page. In most cases, the text should be preceded by an "Abstract" and should conclude with "Acknowledgments" (if any) and "Literature Cited."
All measurements should be in the metric system (periods are not used after abbreviated measurements). The format and style of headings should follow that of recent issues of Fieldiana.
For more detailed style information, see The Chicago Manual of Style (13th ed.), published by The University of Chicago Press, and also recent issues of Fieldiana.
References: In "Literature Cited," book and journal titles should be given in full. Where abbreviations are desirable (e.g., in citation of synonymies), authors consistently should follow Botanico-Periodicum-Huntianum and TL-2 Taxonomic Literature by F. A. Stafleu & R. S. Cowan (1976 et seq.) (botanical papers) or Serial Sources for the Biosis Data Base (1983) published by the BioSciences Information Service. Names of botanical authors should follow the "Draft Index of Author Abbreviations, Royal Botanic Gardens, Kew," 1984 edition, or TL-2.
References should be typed in the following form:
Croat, T. B. 1978. Flora of Barro Colorado Island. Stanford University Press, Stanford, Calif., 943 pp.
Grubb, P. J., J. R. Lloyd, and T D. Pennington. 1963. A comparison of montane and lowland rain forest in Ecuador.
I. The forest structure, physiognomy, and fioristics. Journal of Ecology, 51: 567-601. Langdon, E. J. M. 1979. Yage among the Siona: Cultural patterns in visions, pp. 63-80. In Browman, D. L.,and R. A.
Schwarz, eds., Spirits, Shamans, and Stars. Mouton Publishers, The Hague, Netherlands. Murra, J. 1946. The historic tribes of Ecuador, pp. 785-821. In Steward, J. H., ed., Handbook of South American
Indians. Vol. 2, The Andean Civilizations. Bulletin 143, Bureau of American Ethnology, Smithsonian
Institution, Washington, D.C. Stolze, R. G. 1981. Ferns and fern allies of Guatemala. Part II. Polypodiaceae. Fieldiana: Botany, n.s., 6: 1-522.
Illustrations: Illustrations are referred to as "figures" in the text (not as "plates"). Figures must be accompanied by some indication of scale, normally a reference bar. Statements in figure captions alone, such as "x 0.8," are not acceptable. Captions should be typed double-spaced and consecutively. See recent issues of Fieldiana for details of style.
All illustrations should be marked on the reverse with author's name, figure number(s), and "top."
Figures as submitted should, whenever practicable, be SY2 x 11 inches (22 x 28 cm), and may not exceed 1VA x 16V* inches (30 x 42 cm). Illustrations should be mounted on boards in the arrangement to be obtained in the printed work. This original set should be suitable for transmission to the printer as follows: Pen and ink drawings may be originals (preferred) or photostats; shaded drawings must be originals, but within the size limitation; and photostats must be high-quality, glossy, black-and-white prints. Original illustrations will be returned to the corresponding author upon publication unless otherwise specified.
Authors who wish to publish figures that require costly special paper or color reproduction must make prior arrangements with the Scientific Editor.
Page Proofs: Fieldiana employs a two-step correction system. The corresponding author will normally receive a copy of the edited manuscript on which deletions, additions, and changes can be made and queries answered. Only one set of page proofs will be sent. All desired corrections of type must be made on the single set of page proofs. Changes in page proofs (as opposed to corrections) are very expensive. Author-generated changes in page proofs can only be made if the author agrees in advance to pay for them.
THIS PUBLICATION IS PRINTED ON ACID-FREE PAPER.
fiKfc06YUBRABY
FIELDIANA
Geology
NEW SERIES, NO. 28
Osteology of Simosaurus gaillardoti and the Relationships of Stem-Group Sauropterygia
Olivier Rieppel
Department of Geology
Field Museum of Natural History
Roosevelt Road at Lake Shore Drive
Chicago, Illinois 60605-2496
U.S.A.
Accepted July 19, 1994 Published December 30, 1994 Publication 1462
PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY
© 1 994 Field Museum of Natural History Library of Congress Catalog Card Number: 94-61792
ISSN 0096-2651 PRINTED IN THE UNITED STATES OF AMERICA
Table of Contents
Abstract 1
zusammenfassung 1
Introduction 1
Systematic Paleontology 4
slmosaurus gaillardoti h.v. meyer, 1 842 . . 4
Morphological Description 10
Skull 10
Lower Jaw 14
Postcranial Skeleton 16
Vertebral Column 16
Ribs 20
Pectoral Girdle 22
Pelvic Girdle 23
Forelimb 25
Hindlimb 27
Functional Morphological Correlates in the Skeleton of Simosaurus gail-
LARDOTl 30
Phylogenetic Analysis 36
Definition of Characters 37
Summary and Conclusions 73
Acknowledgments 76
Literature Cited 76
Appendix I 81
Appendix II 82
List of Illustrations
11.
12.
13. 14. 15. 16.
17.
18. 19.
20. 21. 22. 23. 24.
25.
26.
27. 28. 29. 30. 31.
32.
33. 34.
1. Competing hypotheses of sauropterygian 35. interrelationships 3
2. Distribution of outcrops yielding Simo- 36. saurus gaillardoti in the upper Muschel-
kalk basin 6 37.
3. Mandibular symphysis of Simosaurus gaillardoti, Nothosaurus mougeoti, and 38. Nothosaurus mirabilis 7 39.
4. Holotype of Simosaurus guilielmi 8 40.
5. Sutures identified on holotype of Simo- 41. saurus guilielmi 9 42.
6. Skull of holotype of Simosaurus guiliel- mi var. angusticeps 10 43.
7. Jaw fragment of cf. Lamprosauroides 44. goepperti 10
8. Skull of Simosaurus gaillardoti 11 45.
9. Skull of Simosaurus gaillardoti in dorsal 46. view 12
10. Skull of Simosaurus gaillardoti in ven- 47.
tral view 13
Lower jaw of Simosaurus gaillardoti ... 15 Posterior part of the lower jaw of Notho- saurus mirabilis 16
Cervical vertebra of Simosaurus gaillar- doti 17
Dorsal vertebrae of Simosaurus gaillar- doti 18
Sacral vertebrae of Simosaurus gaillar- doti 18
Dorsal centrum of Simosaurus gaillar- doti and Nothosaurus 19
Sacral vertebrae of Simosaurus gaillar- doti 19
Cervical ribs of Simosaurus gaillardoti ... 20
Cervical ribs of Nothosaurus 20
Ribs of Simosaurus gaillardoti 21
Sacral ribs of Simosaurus gaillardoti ... 22
Caudal ribs of Simosaurus gaillardoti . . 22
Gastral rib of Nothosaurus 23
Pectoral girdle of Simosaurus gaillar- doti 23
Dermal pectoral girdle of Simosaurus
gaillardoti 24
Endochondral pectoral girdle of Simo- saurus gaillardoti 25
Left ilium of Simosaurus gaillardoti .... 26
Pelvic girdle of Simosaurus gaillardoti ... 27
Pubis of Nothosaurus sp 28
Forelimb of Simosaurus gaillardoti .... 29 Intermedium and astragalus of Simo- saurus gaillardoti 31
Metacarpal series of Simosaurus gaillar- doti 31
Hindlimb of Simosaurus gaillardoti .... 32 Astragalus of Simosaurus gaillardoti ... 33 Metatarsal series of Simosaurus gaillar- doti 33
Reconstruction of the skeleton of Simo- saurus gaillardoti 34
Dual jaw adductor system of Simosau- rus 35
Skull of Pistosaurus longaevus 38
Incomplete skull of Cymatosaurus 40
Parietal skull table of cf. Cymatosaurus . . 40
Skull fragment of Placodus gigas 41
Skull of Nothosaurus mirabilis and Pla- codus gigas 43
Skull of Placodus gigas 45
Mandibular symphysis of Lariosaurus.
Cymatosaurus, and Nothosaurus 46
Lower jaw of Placodus gigas 47
Mandibular symphysis of Nothosaurus
mirabilis 47
Dorsal vertebra of a placodont (?Cy-
amodus) 49
in
48. Dorsal centrum of Simosaurus, Notho- saurus, and Placodus 50
49. Dorsal centrum of a pachypleurosaur
and of Pistosaurus 50
50. Ilium of Nothosaurus and Placodus gi- gas 50
51. Pelvis of Nothosaurus "raabr 50
52. Pectoral girdle of Placodus gigas 51
53. Clavicle of Nothosaurus 52
54. Scapula of Nothosaurus 53
55. Interclavicle of Simosaurus, dermal pec- toral girdle of Nothosaurus 54
56. Interclavicle of an unidentified saurop- terygian 54
57. Humerus of Cymatosaurus 55
58. Humeri of Sauropterygia 56
59. Humerus of Placodus gigas and Notho- saurus 57
60. Distal end of humerus of Nothosaurus, humerus of Placodus gigas 57
6 1 . Isolated ulnae of unidentified sauropter- ygians 58
62. Femur of Nothosaurus and Placodus gi- gas 59
63. Proximal head of the femur of Simo- saurus gaillardoti, Nothosaurus, and Cya- modus 59
64. Tarsus of Nothosaurus "raabi" 60
65. A. Strict consensus tree (unrooted) for ingroup taxa. B. Strict consensus tree (unrooted) for ingroup taxa plus four potential outgroups 60
66. Strict consensus tree for the Sauropteryg- ia (ingroup) rooted on an all-zero ances- tor 71
67. Strict consensus tree with the Sauropte- rygia (ingroup) rooted on a paraphyletic outgroup 71
68. Strict consensus tree for 22 reptile taxa (including Sauropterygia) rooted on an all-zero ancestor 72
69. Sauropterygian interrelationships among the Reptilia with the Testudines omitted from the analysis 72
70. Testing sauropterygian versus turtle in- terrelationships by the inclusion of ad- ditional taxa 82
7 1 . Testing sauropterygian versus turtle in- terrelationships by the inclusion of pa- reiasaurs and procolophonids, but ex- cluding very incompletely known taxa . . 82
List of Tables
1 . Measurements of coracoids of Simosau- rus gaillardoti 25
2. Measurements of pubis and ischium of Simosaurus gaillardoti 30
3. Measurements of the humerus of Simo- saurus gaillardoti 30
4. Measurements of the radius of Simosau- rus gaillardoti 30
5. Measurements of the ulna of Simosaurus gaillardoti 31
6. Measurements of the femur of Simosau- rus gaillardoti 31
7. Measurements of the tibia of Simosaurus gaillardoti 33
8. Data matrix for 29 taxa included in cla- distic analysis 61
IV
Osteology of Simosaurus gaillardoti and the Relationships of Stem-Group Sauropterygia
Olivier Rieppel
Abstract
Simosaurus gaillardoti H.v. Meyer, 1842, is recognized as the only species of its genus. An amended diagnosis of the genus is given here. Simosaurus mougeoti H.v. Meyer, 1842, is a lower jaw of Not hosaurus; Simosaurus guilielmi H.v. Meyer, 1847-1855, is a junior synonym of Simosaurus gaillardoti. The skeletal morphology of Simosaurus is redescribed in detail and compared to that of other stem-group (non plesio- and pliosaur) Sauropterygia (including Placodus). Ninety-four skeletal characters are defined and used in a phylogenetic analysis. Placodus is shown to be the sister-taxon of the Eosauropterygia, placodonts and eosauroptery- gians together constituting the monophyletic Sauropterygia. Within the Eosauropterygia, Coro- saurus is the sister-taxon of an unnamed clade comprising Pachypleurosauroidea plus Eusaurop- terygia. Simosaurus is the sister-taxon to all other eusauropterygians included in the analysis (the Cymatosaurus-Nothosaurus-Lariosaurus-Pistosaurus clade). Functional morphological correlates of the skeleton suggest that Simosaurus was capable of sustained swimming in the shallow epicontinental sea in pursuit of prey such as "holostean"-grade fishes and, perhaps, ammonites.
Zusammenfassung
Simosaurus gaillardoti H.v. Meyer, 1842 wird als alleiniger Vertreter seiner Gattung aner- kannt. Simosaurus mougeoti H.v. Meyer, 1 842, ist ein Unterkieferrest von Nothosaurus; Si- mosaurus guilielmi H.v. Meyer, 1847-1855, ist ein jungeres Synonym von Simosaurus gail- lardoti. Die Gattung Simosaurus wird durch eine erweiterte Diagnose besser begriindet. Die Morphologie des Skelettes von Simosaurus wird neu beschrieben und mit dem Skelett anderer Stammgruppen-Vertreter der Sauropterygia, und mit Placodus, verglichen. Die phylogenetische Analyse weist die Placodontier als Schwestergruppe der Eosauropterygia aus; beide Gruppen zusammen bilden die monophyletische Gruppe der Sauropterygia. Innerhalb der Sauropterygier erweist sich Corosaurus als Schwestergruppe eines unbenannten Taxons, das die Pachypleu- rosauroidea und die Eusauropterygia umfasst. Simosaurus ist die Schwestergruppe aller andern in der Analyse beriicksichtigten Eusauropterygier (Cymatosaurus Nothosaurus Lariosaurus Pis- tosaurus). Funktionell-morphologische Betrachtungen lassen Simosaurus als ausdauernden Schwimmer erscheinen, der im flachen epicontinentalen Meer Ganoidfische und, vielleicht, Ammoniten gejagt haben mag.
Introduction tral Europe. The Sauropterygia include a diversity
of taxa such as the enigmatic genera Cymatosaurus
The Sauropterygia are a group of marine Me- and Pistosaurus (Sues, 1987a) from the Middle
sozoic reptiles with a fossil record dating back to Triassic, the relatively small pachypleurosaurs from
the uppermost Lower Triassic and a first radiation the Middle Triassic (Rieppel, 1 987), the Triassic
in the lower Muschelkalk (lower Anisian) of cen- nothosaurs (Ceresiosaurus, Lariosaurus, Notho-
FIELDIANA: GEOLOGY, N.S., NO. 28, DECEMBER 30, 1994, PP. 1-85 1
saurus, Paranothosaurus; Rieppel, 1 994a) extend- ing from the lower Anisian (possibly uppermost Scythian) to the uppermost Carnian, the Middle Triassic genus Simosaurus, and the crown-group plesio- and pliosaurs with a fossil record extending from the Triassic-Jurassic boundary (Storrs & Taylor, 1993) to the Upper Cretaceous. (For a complete and annotated list of generic names for Triassic Sauropterygia, see Storrs, 1991.) Whereas the crown-group plesio- and pliosaurs show a cos- mopolitan distribution, the stem-group taxa listed above are restricted to the Old World with the exception of Corosaurus from the upper Scythian of Wyoming (Storrs, 1991). Recent fieldwork, however, indicates the presence of other stem- group sauropterygians in the Middle Triassic (Ani- sian) of northwestern Nevada (Sander et al., 1 994).
The monophyly of the Sauropterygia has been recognized for a long time, but the relationships of the Sauropterygia to placodonts, a group of du- rophagous marine reptiles extending from the up- permost Scythian to the Rhaetian (Pinna, 1990), have remained controversial to the present date. The discussion of phylogenetic interrelationships of the Sauropterygia (Fig. 1) has had a long history (summarized, in part, by Rieppel, 1989a). Owen (1860) recognized the relationships of "notho- saurs" and plesiosaurs with placodonts and in- cluded these taxa with some other enigmatic fossils in his Sauropterygia. A revised concept of the Sau- ropterygia was proposed by Williston (1925), who included the nothosaurs, plesiosaurs, and placo- donts in the Synaptosauria (a term taken from Baur, 1 887) on the basis of the presence of a single upper temporal fossa. Williston's (1925) Synap- tosauria were renamed as Euryapsida by Colbert (1955), and the Lower Permian genus Araeoscelis (Williston, 1914; Vaughn, 1955) was included within the group, as it shares the presence of a single upper temporal fossa (Romer, 1968).
The concept of the Euryapsida (including Arae- oscelis, Sauropterygia, and Placodontia) contrasts with the hypothesis of a diapsid relation of the Sauropterygia (not the placodonts), first proposed by Jaekel (1910) following his investigation of the cranial structure of the pachypleurosaur Anaro- saurus and of Simosaurus. A diapsid derivation of sauropterygians was further supported by Kuhn- Schnyder (1967, 1980), who again used the skull of Simosaurus (Kuhn-Schnyder, 1961) to support Jaekel's (1910) hypothesis of the loss of a lower temporal arch in the Sauropterygia. Following the description of the marine diapsid genus Claudio-
saurus from the Permo-Triassic of Madagascar, Carroll (1981) further supported the diapsid status of the Sauropterygia.
Subsequent cladistic analysis of reptile interre- lationships (Gauthier, 1984; Reisz et al., 1984; Benton, 1985; Evans, 1988) showed Araeoscelis to be a stem-group diapsid and modified the di- agnosis of the Diapsida to refer to the presence of an upper temporal fossa (among other characters) since the lower temporal fossa is a plesiomorphic trait at the level of the Diapsida. The relations of the Sauropterygia within the Diapsida were not addressed in any of these cladistic analyses, and although the Placodontia were recognized as diap- sids (Sues, 1987b), a close relationship between them and sauropterygians continued to be rejected (Kuhn-Schnyder, 1967, 1980; Sues, 1987b; Car- roll & Currie, 1991).
The most recent published contributions to sau- ropterygian interrelationships, presented in a cladis- tic framework, are those of Sues (1987a), S. Schmidt (1987), Rieppel (1987, 1989a), Tschanz (1989), and Storrs (1991, 1993a). An important insight was the recognition (by Sues, 1987a; see also S. Schmidt, 1 987) of the monophyly of the "Nothosauridae" (in- cluding Ceresiosaurus, Lariosaurus, Nothosaurus, and Paranothosaurus; see also Rieppel, 1994a, for ad- ditional synapomorphies of Nothosauridae), and the corroboration of a basal dichotomy, the Pachypleu- rosauroidea representing the sister-group of all other sauropterygians, the Eusauropterygia (Tschanz, 1 989). Within the Eusauropterygia, Simosaurus was hypothesized to represent the sister-group of all other taxa (included in the Nothosauria sensu Tschanz, 1989). Rieppel (1989a) defended a sister-group re- lationship of the Placodontia and Sauropterygia, res- urrecting the Euryapsida as a subgroup of the Diap- sida (see also Zanon, 1 989), but Storrs' ( 1 99 1 , 1 993a) analysis showed the placodonts to be the sister-group of the Eusauropterygia within the Sauropterygia (Storrs, 1991, grouped the Placodontia and Eu- sauropterygia in a new taxon, the Nothosauri- formes). In view of these competing hypotheses (Fig. 1), I reanalyzed Storrs' data matrix using the same software package as he did (Swofford, 1 990) and obtained the same results, which per- sisted even after manipulation of his data matrix in a critical reassessment of character definitions and/or polarities (Rieppel, 1993a). Characters supporting the position of placodonts as sister- group of the Eusauropterygia are few and in some cases problematical: the enlarged upper temporal fossa, the elongated mandibular symphysis, the
FIELDIANA: GEOLOGY
CO
o ■o
o
CO
c o
■D O U
to
Q.
Eusauropterygia Nothosauriformes Sauropterygia
|
CO |
|
|
0) |
|
|
■o |
|
|
o |
|
|
1_ |
|
|
3 |
|
|
(D |
|
|
00 |
|
|
O |
">l |
|
3 |
3 |
|
CD |
3 |
|
CL |
(0 |
|
>» |
(0 |
|
.C |
o |
|
O (D |
E |
|
Q- |
col |
CTJ
!■
CO
o
o
© CO
3 Q.
Storrs, 1991
Eusauropterygia Sauropterygia Euryapsida
Rieppel, 1989a
Fig. 1 . Competing hypotheses of sauropterygian interrelationships. For further discussion, see text.
robust and distinctly curved humerus, and the ap- proximately equal length of radius and ulna. Among these characters, temporal elongation and elongation of the mandibular symphysis may be functionally correlated characters, as an increase in the physiological cross-section of the jaw ad- ductor musculature may require increased stabil- ity of the lower jaw in these durophagous reptiles. Humerus morphology and limb proportions may again reflect functional demands of aquatic loco- motion; furthermore, these characters have been shown to be sexually dimorphic in at least some sauropterygian taxa (Rieppel, 1989a, 1993b; Sander, 1989), which adds to the problem of char- acter definitions.
In view of possible functional constraints in sec- ondarily aquatic tetrapods, the problem of con- vergence must be addressed by inspection of the original material, by increased scrutiny of char- acter definitions, and by the attempt to broaden the data base for testing homology versus homo- plasy (convergence or reversals) by congruence. This study is designed to test the interrelationships of Triassic Sauropterygia by a critical reevaluation of the characters taken from the literature by Storrs (1991, 1993a) and Rieppel (1993a). The osteology of Simosaurus will be reviewed in detail, in view of the hypothesized crucial position of the genus
as sister-taxon to all other eusauropterygians (Sues, 1987a; S. Schmidt, 1987; Tschanz, 1989) and the fact that Simosaurus is known from complete and partially articulated specimens. A close compari- son of the osteology of Simosaurus to that ofNoth- osaurus will provide the basis for a critical char- acterization of eusauropterygian morphology. Placodus, again known from at least one complete skeleton (Drevermann, 1933), is here chosen as a paradigm for placodont morphology and is com- pared to isolated placodont material in the effort to characterize the basal placodont morphology (a complete skeleton of Paraplacodus [Kuhn-Schny- der, 1942] remains inaccessible for study at this time; see also Zanon, 1989). The Pachypleurosau- roidea will be used as a terminal taxon even though some of the genera referred to this taxon are poorly known and future revision of the pachypleurosaurs (Rieppel, work in progress) may alter our current understanding of the group. Accordingly, charac- ters for the Pachypleurosauroidea will be taken from Dactylosaurus (Sues & Carroll, 1985, and pers. obs.) and the well-known Serpianosaurus- Neusticosaurus clade (Rieppel, 1989a; Sander, 1989).
The material described and illustrated in this study is listed in Appendix II, which also provides the relevant institutional abbreviations.
RIEPPEL: SIMOSAURUS GAILLARDOTI
Systematic Paleontology Sauropterygia Owen, 1860
Definition— A monophyletic group of Meso- zoic marine reptiles including the Placodontia and Eosauropterygia.
Diagnosis— Premaxillae large, forming most of the snout; anterior teeth procumbent; lacrimal lost; clavicles positioned anteroventral to interclavicle; clavicles applied to medial surface of scapula; pos- terior stem of interclavicle reduced or absent; cor- acoid foramen enclosed between coracoid and scapula; pectoral girdle fenestrate; three or more sacral ribs.
Eosauropterygia, new taxon
Definition— A monophyletic group including Corosaurus, Pachypleurosauroidea, and Eusau- ropterygia.
Diagnosis— Lateral basioccipital tubera in complex relation to pterygoid; zygosphene-zygan- trum articulation present; expanded articular fac- ets of centrum support neural arch; clavicles with expanded anterolateral corners; scapular blade re- duced.
Eusauropterygia Tschanz, 1989
Definition— A monophyletic group including Ceresiosaurus, Cymatosaurus, Lariosaurus, Noth- osaurus, Pistosaurus, Simosaurus, and all plesio- and pliosaurs.
Diagnosis— Frontal in contact with premaxilla; frontals fused (in adult); parietal fused (in adult); vertebrae platycoelous.
Simosaurus H.v. Meyer, 1842
Type Species— Simosaurus gaillardoti H.v. Meyer, 1 842, from the upper Muschelkalk (Ladin- ian) of Luneville, France (the holotype is now lost; see discussion below).
Diagnosis— Large (3-4 m total length) eusau- ropterygians with a brevirostrine skull; tooth crowns broad and blunt, distinctly set off from tooth base, enamel distinctly striated; snout not constricted; upper temporal fossae much larger than orbits; pineal foramen displaced posteriorly; man- dibular articulation displaced to a level well be-
hind the occipital condyle, occiput deeply exca- vated; squamosal with distinct lateral process; vertebrae platycoelous and non-notochordal with infraprezygapophysis and infrapostzygapophysis in addition to zygosphene and zygantrum; clavicle with short anterolateral process; humerus without entepicondylar foramen; ulna with broad proxi- mal head.
Distribution— Middle Triassic of central Eu- rope.
Simosaurus gaillardoti H.v. Meyer, 1842
1 844 Simosaurus gaillardoti, H.v. Meyer & T.
Plieninger, pp. 45-47, PI. 11, Fig. 1. 1847 Simosaurus gaillardoti, Giebel, p. 162. 1847- Simosaurus gaillardoti, H.v. Meyer, p.
1855 86, PI. 65, Figs. 1-2. 1 847- Simosaurus guilielmi, H.v. Meyer, p. 93,
1855 PI. 20, Fig. 1. 1859 Simosaurus gaillardoti, Gervais, 1859,
pp. 475^76, PI. 55, Fig. 2, PI. 56, Figs.
1-4. 1 864 Simosaurus gaillardoti, v. Alberti, p. 225. 1864 Simosaurus guilielmi, v. Alberti, p. 225. 1896 Simosaurus gaillardoti, Fraas, p. 11,
PI. 3. 1896 Simosaurus guilielmi, Fraas, p. 13. 1899 Simosaurus gaillardoti, Schrammen, PI.
24, Figs. 2a-c. 1 905 Simosaurus gaillardoti, Jaekel, pp. 60, 72,
Figs. 4-7. 1910 Simosaurus gaillardoti, Jaekel, pp. 327,
72, Fig. 3. 1914 Simosaurus gaillardoti, Schroder, p. 96,
text Figs. 14, 28. 1921 Simosaurus gaillardoti, Huene, pp. 201-
239, Figs. 1-13, Pis. I— III. 1921 Simosaurus guilelmi, Huene, pp. 227-
228, Fig. 14. 1 924 Simosaurus gaillardoti, Arthaber, p. 470,
Figs. 8-9. 1924 Simosaurus guilelmi, Arthaber, p. 470. 1928 Simosaurus gaillardoti, Corroy, p. 122,
Fig. 11. 1928 Simosaurus guilielmi, Corroy, pp. 122—
123. 1928 Simosaurus gaillardoti, M. Schmidt, p.
404, Fig. 1134. 1928 Simosaurus guilielmi, M. Schmidt, pp.
404-405, Fig. 1135. 1934 Simosaurus gaillardotii, Kuhn, p. 38.
FIELDIANA: GEOLOGY
1934 Simosaurus guillelmi, Kuhn, p. 39.
1935 Simosaurus gaillardoti, Edinger, p. 330,
Figs. 6, 9a. 1948 Simosaurus gaillardoti, Huene, p. 41,
Fig. 1. 1952 Simosaurus gaillardoti, Huene, p. 163ff.,
Figs. 1-66. 1956 Simosaurus gaillardoti, Huene, p. 390,
Fig. 429. 1956 Simosaurus guile! mi, Huene, p. 390. 1959 Simosaurus guilielmi var. angusticeps,
Huene, pp. 180-184, Figs. 1-3, PI. 19.
1961 Simosaurus gaillardoti, Kuhn-Schnyder,
p. 95ff., Figs. 1-3, 4a, 5, 7b, 8, Pis. 9- 10.
1 962 Simosaurus gaillardoti, Kuhn-Schnyder,
p. 135, Figs. 1-2.
1963 Simosaurus gaillardoti, Kuhn-Schnyder,
Figs, la, 2a, 3a. 1965 Simosaurus gaillardoti, Kuhn-Schnyder,
p. 153, Fig. 7. 1967 Simosaurus gaillardoti, Kuhn-Schnyder,
p. 342, Figs. 7a, 8b. 1 970 Simosaurus gaillardoti, Schultze, pp. 230-
231, Fig. 15. 1987 Simosaurus guilielmi, Wild, pp. 19-20,
Fig. 8. 1989a Simosaurus gaillardoti, Rieppel, p. 61,
Fig. 15a. 1994a Simosaurus gaillardoti, Rieppel, p. 9ff.,
Figs. 2B, 3B, 4, 5A, 6.
Holotype— The original skull from the Mu- schelkalk of Luneville (France), described by Mey- er ( 1 842) as Simosaurus gaillardoti, can no longer be located today, but the description and illustra- tions given by Meyer (1847-1855) validate the name. The holotype of Simosaurus guilielmi Mey- er, 1847-1855, a subjective junior synonym of Simosaurus gaillardoti, is kept at the Staatliches Museum fur Naturkunde Stuttgart (smns 16700) and is considered to be the neotype of Simosaurus gaillardoti.
Locus Typicus— Upper Muschelkalk (Ladin- ian) of Luneville, France.
Distribution— Upper Muschelkalk (Ladinian, Middle Triassic) of eastern France, Wurttemberg and Franconia, SW Germany; Lettenkeuper and Gipskeuper (restricted to upper Ladinian: H. Hag- dorn, in lit. 7 Apr. 1994) of Wurttemberg, Ger- many (Fig. 2).
The occurrence of Simosaurus in the upper Muschelkalk of Wurttemberg (Germany) is cor- related with particular ammonite faunas within
the "Discoceratitenschichten" (H. Hagdorn, in lit. 18 Dec. 1993). The genus is first recorded from the nodosus biozone (Ceratites [Ceratites] nodo- sus) of the upper Ceratitenschichten (Wenger, 1957; Hagdorn, 1991) of the upper Muschelkalk and is most frequently encountered in the dorsoplanus biozone, characterized by Ceratites (Discocera- tites) dorsoplanus. In the semipartitus biozone (Ceratites [Discoceratites] semipartitus), Simosau- rus becomes rare again and remains so until the end of the Muschelkalk. Simosaurus is exceedingly rare in the Keuper.
Diagnosis— Same as for genus, of which this is the only known species.
Comments— The genus Simosaurus was first described by Meyer (1842) on the basis of speci- mens from the Muschelkalk of Luneville, France. Meyer (1842) provided no formal diagnosis and no illustration of the holotype of the genotypical species, but he recognized Simosaurus gaillardoti as very distinct from Nothosaurus by its broad, flat, and brevirostrine skull with a deeply exca- vated occiput (type species by priority on p. 192; named after Claude Antoine Gaillardot [1774- 1833]). The holotype is now lost, but the descrip- tion and illustrations provided by Meyer (1847- 1855) validate the species.
Announcing his work on the fossil reptiles from Luneville to F.v. Andriani, Meyer wrote in a letter dated 27 December 1841 that he was surprised not to find Nothosaurus as part of the fauna; he indicated, however, that he recognized two species of Simosaurus, S. gaillardoti and S. mougeoti (Freyberg, 1972, p. 25). Simosaurus mougeoti was based on a mandibular symphysis and was for- mally described by Meyer in 1 842. The same spec- imen was later figured by Meyer (1847-1855, PI. 1 5, Fig. 3) but designated as Nothosaurus mougeoti in the figure caption (p. 165). In the text, Meyer (1847-1855, p. 19) referred to the specimen as "the first fragment of the lower jaw of Nothosaurus ever found." Indeed, the characteristic elongation of the symphyseal area of the lower jaw with, in its posterior part, large alveoli for fangs opposing the paired maxillary fangs is diagnostic for the genus Nothosaurus (Fig. 3B). Again, the holotype of Simosaurus mougeoti can no longer be located today.
A skull of Simosaurus (Fig. 4) from the Letten- keuper of Hoheneck near Ludwigsburg, Baden- Wiirttemberg, was sent to H.v. Meyer by the Duke Wilhelm of Wurttemberg (Meyer, in lit. to F.v. Andriani, 25 Jan. 1 842; Freyberg, 1 972, p. 26) and first described by Meyer and Plieninger (1844).
RIEPPEL: SIMOSAURUS GAILLARDOTI
Fig. 2. Distribution of outcrops yielding Simosaurus in the upper Muschelkalk basin. Dotted: upper Muschelkalk; black: Keuper. (Courtesy of H. Hagdorn, Ingelfingen.)
Coming from the Lettenkeuper, this specimen was geologically younger than other Simosaurus spec- imens from the upper Muschelkalk. Meyer and Plieninger (1844, p. 46) mentioned problems of preparation and described proportional relations that differed from the skull of Simosaurus gail- lardoti. Their final conclusion was, however, that "all of those differences, including the different contours of the upper temporal fossae, are not sufficient to refer the skull from Ludwigsburg to a species separate from Simosaurus gaillardoti." Later, Meyer (1847-1855, p. 93, PI. 20, Fig. 1) changed his views and named a new species, Si- mosaurus guilielmi, to refer to the skull from Lud- wigsburg. The diagnosis of Simosaurus guilielmi is to be deduced from its comparison with 57- mosaurus gaillardoti (Meyer, 1847-1855, p. 72): the skull of Simosaurus gaillardoti is generally larger and is said to have a broader and less point- ed snout, a less deeply excavated occiput, and rel-
atively larger upper temporal fenestrae (whose lon- gitudinal diameter exceeds twice the longitudinal diameter of the orbit); the distance from the pos- terior margin of the orbit to the anterior margin of the upper temporal fossa, compared to the dis- tance from the posterior margin of the external naris to the anterior margin of the orbit, is larger in Simosaurus gaillardoti; and the dorsal bridge (frontal bone) between the orbits is relatively broader in Simosaurus gaillardoti. Huene (1921, pp. 227-228) confirmed the validity of the new species (which he misspelled as Simosaurus gui- lelmi) on the basis of differences in the cranial suture pattern that he claimed to be able to identify on the holotype. M. Schmidt (1928) again listed Simosaurus guilielmi (which he misspelled as Si- mosaurus guilielmi) as a separate species based on the proportional differences listed by Meyer (1 847- 1855). Inspection of the holotype of Simosaurus gui-
FIELDIANA: GEOLOGY
Fig. 3. Mandibular symphysis in dorsal view. A, Simosaurus (smns 786 1 , upper Muschelkalk, Crailsheim; original of Fraas, 1896, PI. 3); B, Nothosaurns mougeoti (from Meyer, 1847-1855, PI. 15, Fig. 3); C, Nothosaurus mirabilis (smns 59817, upper Muschelkalk [nodosus biozone], Hegnabrunn). Scale bar = 20 mm.
lielmi (smns 16700), and its comparison with all other Simosaurus skulls from the upper Muschel- kalk deposited in public repositories, shows that this species must be considered a junior synonym of Simosaurus gaillardoti. Absolute size differs considerably among the skulls from the upper Mu- schelkalk (usually referred to Simosaurus gaillar- doti), some of them being as small as the skull of Simosaurus guilielmi, indicating ontogenetic vari- ation. The holotype of Simosaurus guilielmi does not show an appreciably more pointed snout than is observed in the skulls from the upper Muschel- kalk. The holotype of Simosaurus guilielmi is very poorly prepared (see Meyer & Plieninger, 1 844, p. 46), with most of the bone surface severely dam- aged. The natural bone surface is damaged all around the contours of the snout, rendering the assessment of its natural shape problematical. Preparation may also account for the apparent slight constriction of the snout referred to by M. Schmidt (1928). What little can be identified of the sutural pattern does not support Huene's (1921) observation (Fig. 5). The posterior projection of the jugal is very distinct, as is the suture between premaxilla and maxilla at the anterolateral edge of the external naris. In front of the right orbit there is a vague indication of a contact of the pre- frontal with the nasal. This character does not dif-
ferentiate Simosaurus guilielmi from Simosaurus gaillardoti, since the latter shows bilateral vari- ability in at least one of the skulls (smns 10360, see Kuhn-Schnyder, 1961, PI. 9, contra Kuhn- Schnyder, 1961, Fig. 2; see also Figs. 8-9 of this paper). The second specimen, Simosaurus gui- lielmi var. angusticeps (Fig. 6), described by Huene (1959) from the lower Gipskeuper of Oberson- theim, is very poorly preserved and allows no identification of sutural details in the skull.
To use cranial proportional relations in the di- agnosis of different species of Simosaurus is highly problematical, because of 1) incomplete preser- vation of most skulls, 2) extensive crushing and deformation of the skulls during fossilization, and 3) poor preparation. In the holotype of Simosaurus guilielmi, the margins of the external nares, orbits, and upper temporal fossae are, partially at least, severely damaged, and the skull dorsoventrally compressed, rendering proportional relations un- clear. With reference to the proportional relations considered to be diagnostic at the species level by Meyer (1847-1855), the following are considered to be least affected by crushing and flattening of the skull: 1) longitudinal diameter of the upper temporal fossa in relation to the longitudinal di- ameter of the orbit; 2) width of the postorbital arch in relation to the distance between the exter-
RIEPPEL: SIMOSAURUS GAILLARDOTI
Fig. 4. Holotype of Simosaurus guilielmi (Meyer, 1847-1855), from the Lettenkeuper of Hoheneck near Lud- wigsburg (smns 16700). Scale bar = 50 mm.
nal naris and the orbit; and 3) width of the frontal (see Appendix II for the material included in this
bone between the orbits, related to the width of study). For measurements that differ on the left
the bony bridge separating the external nares. and right side of the skull, the arithmetic mean of
Measurements were taken from a total of 1 2 Si- the two values was used.
mosaurus skulls deposited in public repositories The ratio (longitudinal diameter of upper tem-
FIELDIANA: GEOLOGY
poral fossa/longitudinal diameter of orbit) ranges from 1.48 to 2.44. The value for the holotype of Simosaurus guilielmi (smns 16700) is 2.01, and the Tubingen specimen of Simosaurus guilielmi shows a ratio of 1.48, whereas the ratios repre- sented by skulls from the upper Muschelkalk of Crailsheim range from 1.53 (smns 18274) to 2.44 (original of Meyer, 1847-1855, p. 86ff., PI. 65, Figs. 1-2; smns uncatalogued). The ratio (width of postorbital arch/distance between external naris and orbit) ranges from 0.44 to 1.15. The value for the holotype of Simosaurus guilielmi (smns 1 6700) is 0.52, and the same ratio is shown by the Tu- bingen specimen of Simosaurus guilielmi. The ratios represented by skulls from the upper Mu- schelkalk of Crailsheim range from 0.44 (smns 16639) to 1.15 (bsp 1932.1.13). The ratio (width of frontal bridge between orbits/width of bony bridge between external nares) ranges from 1.38 to 2.44. The value for the holotype of Simosaurus guilielmi (smns 16700) is 2.0, and the Tubingen specimen of Simosaurus guilielmi shows a ratio of 1.70, whereas the ratios represented by skulls from the upper Muschelkalk of Crailsheim range from 1.38 (bsp 1932.1.13) to 2.44 (smns 18550). As far as they can be established with any degree of confidence, proportional relations in the skull lend no support to the validity of two separate species of Simosaurus.
Opeosaurus suevicus is a genus and species de- scribed by Meyer (1847-1855, p. 82, PI. 14, Figs. 7-9) on the basis of lower jaw fragments, including parts of both left and right dentary, from the upper Muschelkalk of Stuttgart-Zuffenhausen (smns 4141). An alleged diagnostic feature is a "man- dibular fenestra" in the posterior part of the left dentary. Preservation and/or preparation of the specimen is very poor, and the mandibular fenes- tra is an artifact. The genus was considered a pos- sible junior synonym of Simosaurus by Kuhn (1934, p. 40; see also Arthaber, 1924, p. 470; Huene, 1956, p. 390; Storrs, 1991, p. 137), but tooth fragments preserved on both dentaries in- dicate that Opeosaurus must be a junior synonym of Nothosaurus (pers. obs.). Fraas ( 1 896, p. 9) con- sidered Opeosaurus suevicus a junior synonym of Nothosaurus aduncidens (see also Schrammen, 1899, p. 408; Romer, 1956, p. 662). The lower jaw is not diagnostic, and Opeosaurus and suevicus are nomina dubia (Storrs, 1991, p. 137).
Another taxon questionably considered a junior synonym of Simosaurus by Kuhn ( 1 934, p. 38; see also Arthaber, 1924, pp. 470-471) is Lamprosau- roides goepperti {Lamprosauroides K. P. Schmidt, 1927, replaces Lamprosaurus Meyer, 1860), rep-
Fig. 5. Sutures identified on the holotype of Simo- saurus guilielmi (smns 16700). Abbreviations: ju, jugal; m, maxilla; n, nasal; pm, premaxilla; po, postorbital; pof, postfrontal; prf, prefrontal. Scale bar = 40 mm.
resented by a right maxilla from the lower Mu- schelkalk of Upper Silesia (now Poland) and first described by Meyer (1860) (the holotype is at the Institute of Geological Sciences of the University of Wroclaw, MGU Wr. 387 ls). If that and similar specimens (Fig. 7) were, indeed, Simosaurus, it would significantly extend the geological range of that genus from the upper down into the lower Muschelkalk. The teeth, however, are different from those of Simosaurus (slender and pointed), and Lamprosauroides goepperti Meyer ( 1 860) may be a senior synonym of Cymatosaurus Fritsch (1894; see also Schrammen, 1899, p. 408) if it is not, in fact, a nomen dubium. The consequences of such synonymy for the validity of the genus name Cymatosaurus must await the proper as- sessment of the type material of Conchiosaurus clavatus Meyer (1834), which may include Cy- matosaurus material (R. Wild, in lit. 7 Oct. 1993).
RIEPPEL: SIMOSAURUS GAILLARDOTI
Fig. 6. Skull of the holotype of Simosaurus guilielmi var. angusticeps (Huene, 1959) from the lower Gipskeuper of Obersontheim (Geologisch-Palaontologisches Institut und Museum, University of Tubingen, uncatalogued). Scale bar = 50 mm.
Morphological Description
Skull
The skull of Simosaurus has been the object of a large number of studies. The most important specimen is smns 10360 (Figs. 8-10). It was do- nated to the Staatliches Museum fur Naturkunde in Stuttgart by R. Blezinger in 1901 and formed the original of Jaekel's ( 1 905; see also Jaekel, 1 907, Fig. 19) description. Huene (1921, p. 222) de-
P-V1V ^— -Jrw
Fig. 7. Jaw fragment of cf. Lamprosauroides goep- perti (Meyer, 1860) (lower Muschelkalk [mul], Sacrau near Gogolin [Poland], bgr uncatalogued).
scribed a second skull (smns 1 1 364), erroneously identifying it as Jaekel's (1905) specimen. Jaekel's (1905) original specimen was reprepared and re- described by Huene (1952). The same specimen was later used by Romer (1956) and, after acid preparation, by Kuhn-Schnyder (1961; see also Kuhn-Schnyder, 1962, 1963, 1965, 1967) and Rieppel (1989a, 1994a) in their discussion of cra- nial anatomy.
Dorsal View (Fig. 9)— The skull of Simosaurus is broad and flat, with large and oval temporal fossae and a short and broad rostrum. A rostral constriction of the skull at the level of the anterior margin of the external nares is absent or only very weakly developed. The premaxilla meets the max- illa at the anterolateral edge of the oval external naris. Posterior (nasal) processes of the premax- illae (identified as nasals by Jaekel, 1 905; see Kuhn- Schnyder, 1961) extend backward between the ex- ternal nares to meet the frontal at a level shortly behind the posterior margin of the external nares. The nasals (identified as postnasals by Jaekel, 1 905; see Kuhn-Schnyder, 1 96 1) are relatively small tri- angular elements defining the posteromedial mar- gin of the external naris; the posteriorly pointing apex of the nasal is embraced between antero- medial and anterolateral processes of the frontal.
10
FIELDIANA: GEOLOGY
'#<#
~3* ■
Fig. 8. Skull of Simosaurus gaillardoti (smns 10360) in dorsal (left) and ventral (right) views. Scale bar = 50 mm. (Dorsal view: smns neg. #9160; ventral view smns neg. #9159.)
The prefrontal (identified as lacrimal by Jaekel, 1905; see Kuhn-Schnyder, 1961) defines the an- teromedial margin of the orbit. It remains sepa- rated from the postfrontal along the dorsal margin of the orbit, smns 10360 is interesting because it shows an asymmetry of the anterior relations of the prefrontal on the right and left side of the skull. Jaekel (1905, Fig. 4) showed the prefrontal in con- tact with the nasal on both sides of the skull, whereas Kuhn-Schnyder (1961, Fig. 2) recon- structed the skull with a contact of the anterolat- eral process of the frontal with the maxilla, sep- arating the prefrontal from the nasal on both sides. In fact, the skull shows the prefrontal in contact with the nasal on the right side, whereas the an- terolateral process of the frontal reaches the max- illa on the left side (Fig. 9; see also Kuhn-Schnyder, 1961, PI. 9).
The prefrontal meets the palatine by means of a distinct yet slender pillar-like structure (identi- fied as lacrimal by Huene, 1921), which in its dor-
sal part carries a depression, or one or two small foramina (Huene, 1921). This prefrontal-palatine pillar divides the anteroventral part of the orbit into two large openings, a lateral one situated be- tween the ascending process of the maxilla (not lacrimal, as claimed by Kuhn-Schnyder, 1 96 1 ) and the palatine-prefrontal pillar, the medial one lo- cated medial to the palatine-prefrontal pillar. The lateral opening represents the infraorbital foramen (sensu Oelrich, 1 956), which in Simosaurus is con- fluent with the lacrimal foramen and, hence, served the passage of the lacrimal duct in its dorsal part and the passage of the superior alveolar nerve (maxillary division of the trigeminal nerve) along with its artery in its ventral part. The medial open- ing may have served the passage of the palatine nerve and of the ophthalmic division of the tri- geminal nerve.
The frontals are fused and define the concave dorsal margin of the orbit (between prefrontal and postfrontal). The bone lacks clearly defined pos-
RIEPPEL: SIMOSAURUS GAILLARDOTI
11
Fig. 9. Skull of Si mosaurus gaillardoti (smns 10360) in dorsal view. Abbreviations: bo, basioccipital; f, frontal; ju, jugal; m, maxilla; n, nasal; p, parietal; pm, premaxilla; po, postorbital; pof, postfrontal; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; so, supraoccipital; sq, squamosal. Scale bar = 40 mm.
terolateral processes and meets the parietal in a deeply interdigitating suture somewhat behind the level of the anterior margin of the upper temporal fossa. The postfrontal defines the posteromedial margin of the orbit as well as the anteromedial margin of the upper temporal fossa, forming the dorsal part of the narrow postorbital arch. The ventral process of the postfrontal embraces a ta- pering dorsal process of the postorbital. The post- orbital is a curved (rather than triradiate) element defining the anterolateral and most of the lateral margin of the upper temporal fossa through its participation in the formation of the upper tem- poral arch. Anteriorly, the postorbital has a rela- tively narrow exposure at the posterior margin of the orbit, between the postfrontal and jugal bone. The jugal is a fairly prominent element in 57- mosaurus. A slender anterior process extends along the entire lateral (morphologically ventral) margin
of the orbit but does not reach the prefrontal an- teriorly. The width of the jugal is increased at the posterolateral margin of the orbit, but the element becomes narrow again as it extends farther pos- teriorly, situated between the postorbital and the maxilla. The maxilla forms the greater part of the lateral (morphologically ventral) margin of the ex- ternal naris and expands to form a low ascending process (topologically a medial process) between the external naris and the orbit. Behind the orbit, the bone shows a narrow lateral exposure and ex- tends backward , below the jugal, to a level slightly in front of the squamosal-postorbital suture with- in the upper temporal arch. The anterior half of the upper temporal fossa is thus closed laterally by a bony arcade composed of postorbital, jugal, and maxilla, the maxillary tooth row thus extend- ing backward to a level well behind the orbit. The jugal, together with the maxilla, forms a short,
12
FIELDIANA: GEOLOGY
Fig. 10. Skull of Simosaurus gaillardoti (smns 10360) in ventral view. Abbreviations: bo, basioccipital; ec, ectopterygoid; in, internal naris; m, maxilla; pi, palatine; pm, premaxilla; pt, pterygoid; q, quadrate; qj, quadratojugal; v, vomer. Scale bar = 40 mm.
free-ending posterior process (broken in many specimens), which prompted Jaekel ( 1 9 1 0) to pos- tulate a diapsid derivation for the Sauropterygia. The parietal is unpaired (fused) and forms a narrow bridge between the upper temporal fossae. The pineal foramen is displaced to a position somewhat behind the midpoint of the parietal skull table. Posteriorly, the parietal is broadened, with a concave posterior margin. A poorly defined oc- cipital crest (for the insertion of the spinalis capitis muscle) separates the parietal skull table from the occipital exposure of the parietal (identified as postparietal by Kuhn-Schnyder, 1961; but see Schultze, 1970). The parietal forms a laterally de- scending flange along the medial and posterome- dial margin of the upper temporal fossa for the origin of the deep jaw adductor musculature (m. adductor mandibulae extemus medialis and, pos- sibly, m. pseudotemporalis superficialis; Rieppel, 1989a).
The squamosal caps the quadrate suspension and defines the posterolateral margin of the upper temporal fossa. Posterolateral to the upper tem- poral fossa, the squamosal is drawn out into a distinct but small lateral projection. The poorly defined and shallow occipital crest separates the dorsal exposure of the squamosal from its occipital exposure (identified as tabular by Kuhn-Schnyder, 1961; but see Schultze, 1970). The quadrate bone is covered in lateral view by a descending process of the squamosal. The latter meets a small qua- dratojugal bone, positioned lateral to the ventral part of the quadrate bone, immediately above the mandibular condyles of the quadrate. The qua- dratojugal of Simosaurus was first identified by Schultze (1970, Fig. 15; see also Sues, 1987a).
Ventral View (Fig. 10)— The palate of Simo- saurus is characterized by relatively small internal nares of almost circular contours. The internal na- res are separated by the paired vomers, which are
RIEPPEL: SIMOSAURUS GAILLARDOTI
13
long and slender elements. Anteriorly, the vomers reach a gap separating the premaxillaries along the midline (see below). Posteriorly, the vomers em- brace an anteromedial projection of the ptery- goids.
The premaxillae show five tooth positions on an "alveolar ridge." Behind the alveolar ridge, a gap in the ossification results in the formation of a "foramen incisivum" enclosed between the two premaxillaries and closed posteriorly by the vo- mers. Between the vomer and maxilla, the pre- maxilla extends backward, entering (and defining) the anterior margin of the internal naris.
The palatal shelf of the maxilla is broadest in its anterior part, where it forms the lateral margin of the internal naris. From there it tapers toward its posterior tip along a gently curved medial mar- gin, which establishes sutural contact with the pal- atine (anteriorly) and ectopterygoid (posteriorly). The palatine is an almost triangular or trapezoidal element, wedged in between the vomer and pter- ygoid medially, the maxilla laterally, and the ec- topterygoid posteriorly. It defines the posterior margin of the internal naris. The ectopterygoid lies behind the palatine and between the pterygoid (medially) and the maxilla (laterally). It defines the anterolateral margin of the subtemporal fossa.
The pterygoids are the dominant elements in the dermal palate. The bones meet in a ventro- medial suture, obliterating the interpterygoid fossa and covering the basicranium in ventral view. Only the basioccipital is narrowly exposed behind their posterior border. The palate is akinetic due to the fusion of the pterygoids with the basicranium.
The anterior (palatine) processes of the two pterygoids form a combined anteromedial process that enters between the vomers, extending to a level between the internal nares. Transverse pro- cesses of the pterygoid are weakly expressed and meet the ectopterygoid laterally. There is no ven- tral flange of the pterygoids for the origin of the pterygoideus externus muscle. In front of the an- terior margin of the subtemporal fossa, the palate (pterygoid and ectopterygoid) is characteristically vaulted, which results from a depression of the dorsal surface of the palate in the corresponding area. This indicates the presence of a well-devel- oped dorsal (anterior) portion of m. pterygoideus externus, which took its origin from the dorsal surface of the pterygoid and ectopterygoid behind the orbit. The quadrate ramus of the pterygoid defines the medial margin of the subtemporal fossa and carries a well-developed ventral flange along
the medial margin of its anterior part from which originated m. pterygoideus internus.
Occipital View and Braincase— The occiput of Simosaurus was described by Schultze (1970) and Rieppel ( 1 994a), who also described braincase structure in detail. The present account will em- phasize the discussion of characters not dealt with in detail in these publications but essential with respect to the definition of characters for phylo- genetic analysis (see below). The occiput of 57- mosaurus is deeply concave due to the displace- ment of the mandibular articulation to a position well behind the occipital condyle. The posttem- poral fossa is reduced to a small opening posi- tioned between the occipital exposure of the op- isthotic and squamosal. The occiput is essentially closed and platelike, and no well-defined paroc- cipital process is differentiated. A deep incisure separates the occipital exposure of the squamosal from the laterally descending process of the same bone covering the quadrate in lateral view. This incisure was identified by Kuhn-Schnyder (1961) as an "otic notch" and taken as evidence for the presence of a small (reduced?) and dorsally posi- tioned tympanic membrane in Simosaurus. How- ever, while showing a slight concavity of its pos- terior aspect, the quadrate of Simosaurus slants backward (in correlation with the posterior dis- placement of the mandibular joint), and it lacks the posterior concavity indicative of the presence of a tympanic membrane in other diapsids. A tym- panic membrane most probably was absent in Si- mosaurus, as it presence would also interfere with the depressor mandibulae muscle (Rieppel, 1 989a).
The occipital condyle is formed by the basioc- cipital only, which on either side of the condyle forms large lateral basioccipital tubera defining the medial margin of the eustachian foramen (Riep- pel, 1994a). The exoccipitals do not meet dorsal to the occipital condyle. An ossified epipterygoid is absent in Simosaurus (Rieppel, 1 994a).
Lower Jaw
The lower jaw of Simosaurus (Fig. 1 1) is slender and lacks a coronoid process. The greater part of the lateral surface of the lower jaw is covered by the dentary (anteriorly) and surangular (posteri- orly). The angular forms the ventral margin of the lower jaw ramus below the surangular. The lateral exposure of the articular ossification below the mandibular joint and on the short retroarticular
14
FIELDIANA: GEOLOGY
ang
sang
sang pra
ang
ch.t
Fig. 1 1 . Lower jaw of Simosaurus gaillardoti (smns 1 6638; original of Huene, 1 952, Fig. 58). Abbreviations: ang, angular; ar, articular; c, coronoid; ch.t., chorda tympani foramen; d, dentary; pra, prearticular; sang, surangular; sp, splenial. Scale bar = 40 mm.
process cannot be delineated with any certainty (in Nothosaurus, the articular ossification is not exposed in lateral view, due to a posterior exten- sion of the surangular; pers. obs.). The supposed anterior margin of the articular bone shown by Huene (1952, Fig. 58a, drawn from smns 16638) is a crack. A distinct ridge on the lateral surface of the surangular marks the ventral edge of the insertion area of the m. adductor mandibulae ex- ternus superficialis on the lateral surface of the lower jaw. The anterior and dorsal edge of the insertion area of the m. pterygoideus externus is indicated by a distinct, curved ridge on the pos- teroventrolateral aspect of the lower jaw, ventral to and in front of the mandibular articulation.
In medial view, the lower jaw shows the elon- gated splenial bone. Fairly high posteriorly, the bone defines the anterior margin of the adductor fossa. The bone closes the Meckelian canal me- dially, tapering anteriorly to a blunt tip. The sple- nial does not reach the mandibular symphysis (contra Huene, 1952). smns 16638 is the original of Huene (1952, Fig. 58). The mandibular ramus measures 366 mm in length, and the anterior tip of the splenial lies about 30 mm behind the sym- physis. The splenial gains no ventral exposure along the ventral margin of the lower jaw.
In front of the splenials, Meckel's canal opens on the medial side of the lower jaw. Meckel's car- tilage must have emerged from the anterior open- ing of the Meckelian canal to meet with its coun-
terpart from the other side in the mandibular symphysis. The dentary provides a dermal cov- ering of the lateral and ventral surface of Meckel's cartilage; each dentary also forms a medial sym- physeal shelf covering the anterior tips of Meckel's cartilages dorsally. As the dentaries wrap around the anterior tips of Meckel's cartilages, the dermal mandibular symphysis appears recessed in pos- terior (medial) view. An anteroposterior elonga- tion of the mandibular symphysis is not prominent in Simosaurus.
The coronoid is a small element lying at the anterodorsal margin of the adductor fossa. It gains no exposure on the lateral surface of the lower jaw. The prearticular is exposed within the adductor fossa. Along the ventral margin of the adductor fossa, the angular bone forms a distinct ledge that must have served for the insertion of the ptery- goideus internus muscle.
The articular surface is saddle-shaped. Its an- terior margin is relatively low compared to the distinctly elevated posterior margin. This mor- phology of the articular facet locks the lower jaw against the quadrate, blocking translational move- ment caused by the anterior pull of the pterygoide- us muscle (which had a strongly developed dorsal [anterior] portion; see above).
The retroarticular process is present but short. Its lateral surface bears a deep facet, which must have served for the insertion of the depressor man- dibulae muscle. Deep to this facet lies a distinct
RIEPPEL: SIMOSAURUS GAILLARDOTI
15
f.m.add
sang
ch.t?
ch.t?
sang
Fig. 1 2. Posterior part of the lower jaw of Nothosau- rus mirabilis (smns 598 1 8, upper Muschelkalk [nodosus- spinosus biozone], Hegnau). A, Left lateral view; B, me- dial view; C, dorsal view. Abbreviations: ang, angular; ar, articular; ch.t., chorda tympani foramen; d, dentary; f.m.add, facet for the insertion of superficial jaw adduc- tor muscle fibers; pra, prearticular; sang, surangular; sp, splenial. Scale bar = 20 mm.
foramen. The medial surface of the retroarticular process, together with the medially overhanging mandibular articular facet, forms a deep recess that faces medioventrally and probably served as the insertion site of the m. pterygoideus internus (posterior fibers). Immediately posteroventral to the mandibular articular facet, the retroarticular process shows another distinct foramen (smns 16638) that lies in the classical position of the chorda tympani foramen. The foramen on the lat- eral surface of the retroarticular process cannot also serve the entry of the chorda tympani into the lower jaw and must therefore be interpreted as the posterior supra-angular foramen (sensu Oelrich, 1956). Although sutural relations are not distinct, the position of the posterior supra-angular fora- men indicates a posterior extension of the suran- gular in a pattern similar to Nothosaurus (Fig. 1 2).
Postcranial Skeleton
The best preserved skeleton of Simosaurus is smns 14733 (Huene, 1952) from the upper Mu-
schelkalk of Tiefenbach near Crailsheim. The skeleton is fairly complete but was fully disartic- ulated. The skull and the pectoral girdle were miss- ing, but its association with teeth of Simosaurus allowed its unequivocal identification. The skel- etal elements were completely removed from the matrix, but a cast was made of the specimen before its preparation in order to document the original position of its parts. This cast was lost during World War II. Only one photograph of the specimen in its original position survived, but in a very poor, almost unrecognizable condition. The original specimen was packed and sheltered during the war. In 1 945, the specimen was returned to provisional quarters in Ludwigsburg. To make it available for description by Huene (1952), the specimen was unpacked by Dr. Staesche, Curator at the smns, and parts went on loan to Huene in Tubingen (R. Wild, pers. comm.). Redescription of this skeleton of Simosaurus below will be augmented by ref- erence to additional material deposited in public repositories and referred to by the appropriate col- lection numbers.
Vertebral Column
smns 14733 provided six centra and three neural arches of the cervical region, 26 centra and 28 neural arches of the dorsal region, five vertebrae counted as sacrals by Huene (1952), and two prox- imal as well as one distal caudal vertebra. The articulated skeleton of Simosaurus from the Gips- keuper of Obersontheim (gtpi uncatalogued) shows a total of 32 dorsal vertebrae in situ (counting three rather than five sacral vertebrae; cf. Huene, 1959), with the possibility that there were one or two additional dorsal vertebrae in front of the pre- served series. Huene (1959) estimated a similar number of elements (30-3 1 ) in the cervical ver- tebral column, but with no factual basis. The dis- tance between the skull and the preserved torso as reconstructed for the specimen was estimated and not based on the position of the parts in the field (see discussion below).
The cervical vertebrae (Fig. 1 3) show a weakly amphicoelous and non-notochordal centrum with a slightly constricted middle portion. The ventral surface of the centrum is distinctly keeled. The length of the centra increases from front to back. Centrum # 1 24a (smns 1 4733), an anterior cervical element, measures 16 mm in length, whereas cen- trum #1 30 (smns 14733), located more posteriorly within the cervical series, is 20.5 mm long. The neural spines slant slightly backward. Their an-
16
FIELDIANA: GEOLOGY
tenor edges are thin, and the posterior margins are thickened. The height of the neural spines increas- es from front to back: 47.5 mm is the total height of a neural arch from the middle of the cervical region (the dorsal tip of a neural spine from a more anterior position is broken); total arch height in- creases to 55 mm in the posterior cervical region. In addition to the zygapophyseal articulations, the neural arches of the cervical region show the ac- cessory zygosphene-zygantrum articulation, but not the differentiation of infrapre- and infrapost- zygapophyses (see below). Width across the post- zygapophyses is slightly larger than across the pre- zygapophyses, but pre- as well as postzygapophyses are generally wider (45-47 mm) than the centra (26-28 mm). The articular surface of the prezyg- apophyses in the cervical region is slightly inclined facing dorsomedially (approx. 30° from the hori- zontal in the midcervical region); the postzyga- pophyses correspondingly face ventrolaterally. The neurocentral suture is unfused in the cervical re- gion (allowing the neural arches to separate from the centra). The dorsal surface of the centrum is broadened and flattened, providing a platform for articulation with the neural arch. This character is further accentuated in the dorsal region, as will be discussed below.
The centra of the dorsal vertebral column (Fig. 14) average 26-27 mm in length, 33-35 mm in height, and 21-29 mm in width. The central part of the centrum is constricted in ventral view, the articular surfaces are weakly amphicoelous or platycoelous, and the centra are non-notochordal. The available neural arches show differences in height of the neural spine, indicating that the neu- ral spines were somewhat higher in the pectoral region than elsewhere in the dorsal region. The total height of the dorsal neural arches is 72-77 mm, the total width across the transverse pro- cesses is 55-59 mm, and the average total width across the pre- and postzygapophyses is 35-40 mm. The zygapophyses are hence generally broader than the centrum, but this character depends on the inclination of the articular surfaces, which increas- es from front to back, and may be reversed in the sacral and caudal regions (see below). In an an- terior dorsal neural arch, the articular surface of the prezygapophysis is inclined by approximately 30° from the horizontal (facing dorsomedially); in a posterior neural arch, the inclination increases to approximately 40° from the horizontal. Neural arches of the dorsal region differ from those of the cervical region by the development of infrapre- and infrapostzygapophyses (Huene, 1952, p. 167) as accessory articulations in addition to the zygo-
Fig. 13. Cervical vertebra of Si mosaurus gaillardoli (smns 14733). Left: centrum in dorsal view (#130 of Huene, 1952); right: neural arch in anterior view (#144 of Huene, 1952). Scale bar = 20 mm.
sphene-zygantrum articulation (Fig. 1 5). The neu- rocentral suture is open throughout the dorsal re- gion, as is again indicated by the separation of the neural arches from the centra. The dorsal surface of the centrum is conspicuously widened, provid- ing a very characteristic "cruciform" or "butterfly- shaped" platform for the articulation with the neu- ral arch. This platform, increasing in width along an anteroposterior gradient, is traversed by the longitudinal neural canal (forming a longitudinal groove on the centrum, which is somewhat broad- er and deeper than in Nothosaurus), and its surface bears a pattern of radiating grooves and ridges participating in sutural interdigitation (Fig. 16). Posterolaterally, the articular platform carries tri- angular areas on both sides with a conspicuous pattern of pitting; similar pitted areas, but much smaller and less well defined, are located in the anterolateral corner of the articular surface.
The interpretation of the sacral region of Si- mosaurus remains problematical to some degree. Huene (1952) counted five sacral vertebrae, but a count of three sacral vertebrae is probably correct. Two vertebrae of smns 14733 (with no individual numbers) are unquestionably sacral elements, and they were interpreted as such by Huene (1952, Fig. 19). The vertebrae (Fig. 17) are in articulation, and the neural arches have not separated from the centrum, although the neurocentral suture re- mains visible. It passes through the lower part of the rib articulation, the centrum thus contributing to the formation of a short yet high transverse process. The infrapre- and infrapostzygapophyses are reduced as compared to those of the dorsal neural arches. The dorsomedial inclination of the articular surface on the prezygapophysis (latero- ventral inclination of the articular surface on the
RIEPPEL: SIMOSAURUS GAILLARDOTI
17
Fig. 14. Dorsal vertebrae of Simosaurus gaillardoti (smns 14733). A, Neural arches in anterior view, centra in dorsal view; B, neural arches in posterior view, centra in posterior view. Scale bar = 20 mm.
postzygapophysis) is very pronounced, and the width across the zygapophyses no longer exceeds the width across the platycoelous articular surface of the centrum. The width across the pre- or post- zygapophyses of these two vertebrae ranges from 21.5 to 23 mm, and the width across the articular surface of the centrum measures 28-29 mm. The middle portion of the centrum is constricted in ventral view.
The isolated vertebra #147 (smns 14733) (Fig. 17) was not figured by Huene (1952). It shows
prz
poz
iprz
ipoz
Fig. 15. Sacral vertebrae of Simosaurus gaillardoti (smns 14733, #147 of Huene, 1952). A, Anterior view; B, posterior view. Abbreviations: ipoz, infrapostzyg- apophysis; iprz, infraprezygapophysis; poz, postzyg- apophysis; prz, prezygapophysis; zyg, zygantrum; zyp, zygosphene. Scale bar = 20 mm.
reduced infrapre- and infrapostzygapophyses, the articular surface of the prezygapophysis is strongly inclined dorsomedially (approximately 45° from the horizontal), and the neural arch has not sep- arated from the centrum. The transverse process is high dorsoventrally but short proximodistally, and the neurocentral suture passes through its middle portion, indicating that the centrum con- tributes to the rib support. This indicates that the element derives from the far posterior dorsal series and/or may represent the first sacral vertebra.
Vertebra #113 (smns 14733) shows even more reduced infrapre- and infrapostzygapophyses, and the articular surface of the prezygapophysis is even more inclined than in vertebra #147 (approxi- mately 55° from the horizontal). Again, the neural arch has not separated from the centrum, and again, the neurocentral suture passes through the middle of the short yet high transverse process, the cen- trum thus contributing to the rib support. Huene (1952, Fig. 18) interpreted the vertebra as a "cau- dosacral element," which is here interpreted to mean an anterior caudal. As the centrum shows no clear-cut posteroventral articular facets for the chevron bone, the element could also represent the posteriormost of four sacral vertebrae.
Huene (1952, p. 168) considered vertebra #153 (smns 14733) as the first sacral element, a claim that cannot be supported on morphological grounds. The inclination of the articular surface of the prezygapophysis indicates a position in the posterior dorsal region, the neural arch has sepa-
FIELDIANA: GEOLOGY
Fig. 16. A, Dorsal centrum of Simosaurus gaillardoti (smns 54763; upper Muschelkalk, Hildesheim) in dorsal view. B, Dorsal centrum of Not hosaurus sp. in dorsal view (smns 598 1 9, upper Muschelkalk, Bindlach near Bayreuth). Anterior is at the top. Scale bar = 20 mm.
rated from the centrum, and the centrum did not contribute significantly to the formation of the ar- ticular facet for the rib at the distal end of the transverse process. On morphological grounds, no more than four vertebrae can possibly be inter- preted as sacral vertebrae of Simosaurus. On func- tional grounds, Simosaurus may have had no more than three sacral vertebrae (see the discussion of sacral ribs and of the ilium, below), with the con- sequence that the first caudal element would not bear a chevron. The problem with the identifica- tion of the number of sacral elements in saurop- terygians (and other reptiles in general) is that pos- teriormost dorsal and/or anteriormost caudal ribs
may converge on the ilium in addition to the gen- uine sacral ribs.
The tail of smns 14733 is very incompletely preserved. Vertebra #98 represents an isolated dis- tal tail vertebra. It is relatively long (26.2 mm) in relation to its total height (36 mm; neural arch somewhat incomplete) and total width (approxi- mately 1 1 mm across the prezygapophyses). The neural arch has not separated from the centrum, and the prezygapophyses are almost vertically ori- ented and hence do not exceed the width of the centrum. The middle part of the centrum is con- stricted in ventral view, and the ventral surface is keeled. The keel bifurcates and diverges toward
Fig. 17. Sacral vertebrae of Simosaurus gaillardoti (smns 14733). A, Sacral vertebrae in right lateral view (left, original of Huene, 1952, Fig. 19), vertebra #147 (right, original of Huene, 1952) in anterior view. B, Sacral vertebrae in left lateral view (left, original of Huene, 1952, Fig. 19), vertebra #147 (right, original of Huene, 1952) in posterior view. Scale bar = 20 mm.
RIEPPEL: SIMOSAURUS GAILLARDOTI
19
Fig. 18. Cervical ribs of Si mosaurus gaillardoti (smns 14733). A, Anterior cervical rib (#140?); B, posterior cervical rib (#150); C, #19 of Huene, 1952; D, #140? of Huene, 1952; E, #150 of Huene, 1952. Scale bar = 10 mm.
the posterior margin of the centrum, where its limbs expand into articular facets for the chevron.
Ribs
A total of four cervical ribs are available (smns 14733, #19, #140, #145, and #150; Fig. 18), of
Fig. 1 9. Cervical ribs of Nothosaurus sp. (mb R. 1 50, lower Muschelkalk [Schaumkalk], Oberdorla, Thiirin- gen; original of Peyer, 1939, Fig. 23). Scale bar = 10
which the anteriormost one is the smallest and shortest. It is dichocephalous, whereas in the more posterior cervical ribs the tuberculum and capit- ulum tend to become confluent. All cervical ribs show an expanded (broadened) "shoulder," but none shows a distinct free anterior process. Whether this is the natural condition or due to breakage (incurred during casting or an incident in which the cabinet holding the specimen was overturned) can no longer be determined. How- ever, Huene (1952, Fig. 21) figured such a process in at least one anterior cervical rib, and a free anterior process on cervical ribs is a general feature of sauropterygians (Fig. 1 9).
The dorsal ribs (Fig. 20G) are holocephalous, evenly curved, and show a sharp and slightly over- hanging posterior margin. The anterior dorsal ribs show a moderate distal expansion. Posterior dor- sal ("lumbar") ribs are distinctly reduced in size (see also Huene, 1959) and lack the distal expan- sion.
Problems surface again with identification of sa- cral ribs (Figs. 20-21). Among the available skel- etal elements of smns 14733, a total of seven sacral ribs (both left and right) have been identified, dis- tributed over an assumed total of five sacral ribs on each side by Huene (1952). Among these, rib #1 1 (Huene, 1952, Fig. 30) is the principal sacral rib. It is a very robust and stout element with a length of 70 mm and a distinct proximal and distal expansion (proximal width: 29.9 mm; minimal width at middiaphysis: 12.7 mm; distal width: 23.5
20
FIELDIANA: GEOLOGY
<*
Fig. 20. Ribs of Simosaurus gaillardoti (smns 14733). A, Sacral rib (#73 of Huene, 1952) in lateral view; B, sacral rib (#65 of Huene, 1952) in lateral view; C, sacral rib (#1 1 of Huene, 1952) in lateral view; D, anterior caudal rib (#102 of Huene, 1952) in lateral view; E, anterior caudal rib (#102 of Huene, 1952) in dorsal view; F, sacral rib (#1 1 of Huene, 1952) in dorsal view; G, dorsal rib. Scale bar = 20 mm.
mm). A second sacral rib, not as robust as # 1 1 but still more massive than all others, is rib #65 (Huene, 1952, Fig. 31), with a length of 75.0 mm, a prox- imal width of 2 1 . 1 mm, a distal width of 1 9.3 mm. and a width at middiaphysis of 9.3 mm. A third sacral rib, identifiable by the presence of both proximal and distal expansions, is rib #73 (Huene, 1952, Fig. 33), with a length of 73.2 mm, a prox- imal width of 23.5 mm, a distal width of 13 mm.
and a width at middiaphysis of 8.3 mm. Assuming three sacral ribs to be the plesiomorphic condition at the level of the Eusauropterygia (with reference to pachypleurosaurs and Placodus), these three elements (#11, #65, and #73) would be the "an- cestral" sacral ribs, and the question arises wheth- er Simosaurus has added additional sacral ribs to the plesiomorphic number by "sacralization" of a posterior dorsal or an anterior caudal pair of ribs.
RIEPPEL: SIMOSAURUS GAILLARDOTI
21
Fig. 2 1 . Sacral ribs of Simosaurus gaillardoti (smns 14733). A, #1 1 of Huene, 1952; B, #65 of Huene, 1952; C, #73 of Huene, 1952. Scale bar = 20 mm.
Because a posterior dorsal or anterior caudal rib would have to converge on the ilium in a pos- terolateral or anterolateral direction, respectively (rather than being more or less transversely ori- ented, as are the three principal sacral ribs), one would expect the fourth pair of sacral ribs to be longer than the elements described above. Besides the three ribs just described (#1 1, #65, and #73), a total of four additional ribs (without separate numbers) have been described, two of which are shorter than (65.5 mm and 68 mm, respectively) and two the same length as (both 73.5 mm) the sacral ribs identified above. All four show a prox- imal and distal expansion, but the two longer el- ements (of these four) are distinctly more robust, their proximal and distal expansion is more pro- nounced, and their curvature is less, although one side is strongly concave. These two pairs of ribs may represent posteriormost dorsal ("lumbar") or anteriormost caudal ribs converging toward the ilium without providing much further support (see description of the ilium, below). Five pairs of ribs are converging toward the ilium in the skeleton from the Gipskeuper described by Huene (1959), but, judged by the size of the ilium, a maximum of three sacral ribs could possibly have supported
Fig. 22. Caudal ribs of Simosaurus gaillardoti (smns 14733). A, #102 of Huene, 1952; B, #168 of Huene, 1952. Scale bar = 20 mm.
the bone. Ilium morphology also indicates that the most massive of the three sacral ribs (#11) would be the posteriormost one of the three, articulating in the large posterior articular facet on the medial side of the ilium (see below).
Typical caudal ribs are rather straight elements with an expanded proximal articular head that does not fuse with the transverse process of the respec- tive vertebral element (Fig. 22). The caudal ribs retain a broad but not expanded distal tip in dorsal or ventral view. The posterior margin of the rib continues to provide a sharp and slightly over- hanging rim, as is the case in the dorsal ribs.
Gastral ribs were present in Simosaurus, but none is well enough preserved to allow a detailed description of their morphology. Huene (1952, p. 170) mentioned that medial gastral rib elements may co-ossify, a phenomenon also reported for Nothosaurus (Fig. 23; see also Koken, 1893, PI. 1 1, Figs. 7-9) and Corosaurus (Storrs, 1991).
Pectoral Girdle
The dermal pectoral girdle is composed of the interclavicle and the clavicle (Fig. 24). The inter- clavicle is essentially a T-shaped element, with a broad anterior portion extending into tapering lat- eral processes and a relatively short yet distinct posterior stem. A complete interclavicle (smns 15995) measures 158 mm across the tips of the tapering anterior lateral processes, while the total length of the element is 88.3 mm (with the pos- terior stem approximately 68 mm long). The entire element is thus distinctly wider than long. Huene (1952, p. 174, Fig. 62) figured an isolated inter-
22
FIELDIANA: GEOLOGY
Fig. 23. Medial element of a gastral rib of Nothosau- rus sp. (mb R.7 14, upper Muschelkalk, Bayreuth; original of Koken, 1893, PI. 11, Fig. 9). Scale bar = 20 mm.
clavicle of large size from the Grenzbonebed of the upper Muschelkalk separating the Muschel- kalk from the Keuper, which shows a relatively much longer posterior stem than is observed in smns 15995 and which may or may not be attrib- utable to Simosaurus. Variation is also observed in the anterior part of the interclavicle as smns 1 5995 is compared to smns 7862 (Fig. 25). In the latter specimen, the anterolateral processes are rel- atively shorter and do not taper to a pointed tip, while the posterior stem (judged by its relative width at the posterior break) may well have been relatively longer than would appear in smns 1 5995.
The clavicles (Figs. 25A,C) are rather distinct, curved elements that are received in an antero- ventral facet on the interclavicle and that estab- lished a broad mutual contact along the ventral midline of the body in front of the interclavicle. The anterolateral edge of the interclavicle is ex- panded into a distinct blade forming the "clavic- ular corner" described by Storrs (1991, 1993a). A character diagnostic (autapomorphic) for Simo- saurus is a small anterior process projecting from the anterior edge of the clavicle. The posterior ramus of the clavicle tapers to a blunt tip as it trends in a posterodorsal direction. It is received in a distinct facet on the anterior and medial sur- face of the dorsal scapular wing.
The endochondral pectoral girdle is composed of two elements ossifying separately, the scapula and coracoid (Fig. 26). The dorsal wing (scapular blade) of the scapula is much reduced, forming essentially a posterodorsally ascending process. At its anterior base, this process expands into the ventral portion of the scapula, contributing to the formation of the glenoid laterally and to the clo- sure of the coracoid foramen posteromedially. The anterior aspect of the dorsal wing supports the posterior process of the clavicle. The clavicular
Fig. 24 . Pectoral gi rdle of Simosaurus gaillardoti (smns 7862, upper Muschelkalk, Crailsheim; original of Fraas, 1896, p. 12; Huene, 1952, Fig. 59). Scale bar = 50 mm.
facet expands onto the medial surface of the an- terior part of the dorsal wing up to the deep furrow separating the dorsal wing from the expanded ven- tral portion of the scapula.
The coracoid is a characteristically waisted bone with distinctly concave anterior and posterior margins and convex lateral and medial margins. The lateral margin is thickened anteriorly where it participates in the formation of the glenoid facet. Immediately behind the latter, the lateral margin of the coracoid is notched, forming the medial margin of the coracoid foramen. Measurements for two well-preserved coracoids are given in Ta- ble 1.
Pelvic Girdle
The ilium of Simosaurus (Figs. 27-28) is a small element with a much reduced iliac blade. The lat- ter is short and separated from the expanded ven- tral portion by a distinct constriction ("neck"). Both ends of the dorsal portion taper to a pointed tip, the anterior one corresponding to the spina praeacetabuli (Siebenrock, 1894), and the poste- rior one pointing in a posterodorsal direction but not quite reaching the level of the posterior margin of the expanded ventral portion. The ventral por- tion of the ilium participates in the formation of the acetabulum, but no supra-acetabular buttress is developed. The medial surface of the ventral portion of the ilium shows a complex morphology indicating a maximum of three articular facets for sacral ribs. The first and second sacral ribs artic-
RIEPPEL: SIMOSAURUS GAILLARDOTI
23
« 1
Fig. 25. The dermal pectoral girdle of Simosaurus gaillardoti. A, Interclavicle, clavicle, and scapula (smns 7862, upper Muschelkalk, Crailsheim; original of Fraas, 1896, p. 12; Huene, 1952, Fig. 59); B, interclavicle (smns 15995, upper Muschelkalk, Tiefenbach near Crailsheim); C, clavicle (smns 17097, upper Muschelkalk, Tiefenbach near Crailsheim). Abbreviations: cl, clavicle; cof, coracoid foramen; icl, interclavicle; sc, scapula. Scale bar = 20 mm.
ulated in two facets arranged one above the other on the anterior part of the ventral portion of the ilium. A third articular facet lies behind the two anterior ones, separated from the latter by a deep groove (or furrow). The morphology of the ilium thus indicates the presence of three functional sa- cral ribs in Simosaurus, and the same number is indicated for Nothosaurus by the articulated skele- ton of Nothosaurus "raabi" (mb 1.007.18) de- scribed by Schroder (1914; see discussion below
and Fig. 51). Isolated ilia of Nothosaurus show a morphology almost identical to that of Simosau- rus, although the deep furrow separating the an- terior two from the posterior articular facet is ab- sent (see discussion below and Fig. 50). There is one single or poorly subdivided articular facet on the medial surface of the ilium of Nothosaurus, offering room for no more than three functional sacral ribs. The pubis of Simosaurus (Fig. 28) shows deeply
24
FIELDIANA: GEOLOGY
Table 1 . Measurements (in millimeters) for two well- preserved coracoids of Simosaurus gaillardoti.
cof
Fig. 26. The endochondral pectoral girdle of Si- mosaurus gaillardoti. A, Right scapula, medial view (composite, mainly based on smns 18373 from upper Muschelkalk, Heldenmuhle); B, right coracoid, lateral (ventral) view (smns 10046, upper Muschelkalk, Crail- sheim; original of Huene, 1952, Fig. 64). Abbreviations: cl.f, clavicular facet on scapula; cof, coracoid foramen. Scale bar = 20 mm.
concave anterior and posterior margins and ex- panded lateral (morphologically dorsal) and me- dial (morphologically ventral) heads. The lateral (dorsal) head is expanded as it participates in the formation of the acetabulum. The obturator fo- ramen is located immediately anteroventral to the acetabulum and is completely enclosed by bone. The medial (ventral) margin shows a generally
|
Coracoid measurements |
|
|
Specimen number |
Lateral Minimal Medial Length width width width |
|
smns 10046 smns 16736 |
139 83 36.5 81.8 218 125 44.5 106.7 |
convex margin, which is distinctly notched, how- ever. A similar notch (concavity) is observed in the medial (ventral) margin of the pubis ofNotho- saurus, which shows an open (slitlike) obturator foramen in specimens of variable size (Fig. 29).
The ischium is again similar to that of Notho- saurus (Fig. 28) with strongly concave anterior and posterior margins, a thickened lateral (morpho- logically dorsal) head participating in the forma- tion of the acetabulum, and a widely expanded medial (morphologically ventral) portion with an evenly convex margin. The concavity of the pos- terior margin of the pubis and anterior margin of the ischium indicates the presence of a large thy- roid fenestra in Simosaurus. Measurements for well-preserved pelvic elements are given in Ta- ble 2.
Forelimb
The humerus is represented by a number of specimens (see Appendix II and Table 3). The element is generally wider distally than proxi- mally. The proximal head is thicker, however, and the distal end is rather flat. The humerus shows a distinct curvature, with the posterior (postaxial) margin being evenly concave, while the anterior (preaxial) margin shows a distinct angulation be- tween the first and second thirds of its length. The deltopectoral crest is fairly strongly developed and expanded into an anterior ledge accentuating the angulation of the humerus (Fig. 30A). Huene ( 1 952) described an entepicondylar foramen in the hu- merus of Simosaurus that is, in fact, absent. The specimen described by Huene ( 1 952; smns 1 4733) was reprepared, and the entepicondylar foramen described by Huene ( 1952) turned out to be a shal- low groove at best. A shallow groove may also be observed in smns 52095 and smns 1 7590, whereas all other humeri investigated (see Appendix II) show no sign of an entepicondylar foramen or groove. Both proximal and distal articular surfaces are covered by unfinished bone. The distal artic-
RIEPPEL: SIMOSAURUS GAILLARDOTI
25
Fig. 27. Left ilium of Simosaurus gaillardoti (smns 14733). A, Lateral view; B, medial view. Scale bar = 20 mm.
ular surface is evenly rounded with no apparent separation of radial or ulnar condyles.
Among pachypleurosaurs, a sexual dimorphism in skeletal morphology was observed that is best expressed in proportional relations of the humerus (Rieppel, 1989a, 1993b; Sander, 1989), one sex (sex y) showing a greater degree of distal expansion of the humerus than the other (sex x). The sample of humeri of Simosaurus is rather restricted and hence not well suited for the analysis of sexual dimorphism in this genus. Additional problems were created by crushing of the bones during fos- silization. Nevertheless, there is a distinct dis- crepancy in the values obtained for the ratio of humerus length/distal width. The ratio is 5.91 for smns 7956 and 5.02 for an uncatalogued smns specimen, whereas for five other intact humeri the ratio ranges from 4.17 to 4.5. This may be taken as indication that sexual dimorphism cannot be ruled out for Simosaurus.
The radius (Table 4) is a rather featureless bone (Fig. 30B) with the proximal head slightly more expanded than the distal end. The outer (preaxial) margin is somewhat angulated; the inner (post- axial) margin is evenly concave.
The most striking feature of the ulna (Fig. 30C and Table 5) is the distinct expansion of the prox- imal head and a much lesser expansion of the distal end. The anterior (preaxial) margin is strongly con- cave, which, together with the concavity of the postaxial margin of the radius, contributes to the
formation of a distinct spatium interosseum. The posterior (postaxial) margin of the ulna is more or less straight.
The manus was completely disarticulated, which renders its reconstruction problematical. Huene (1952) assumed the presence of three ossified car- pal elements, a reasonable assumption in view of the similar condition known from articulated Nothosaurus specimens {Nothosaurus "raabi," MB 1.007.18; see Schroder, 1914). One of the carpal elements (#210) can safely be interpreted as the intermedium on the basis of its shape, which in- cludes a concavity on the proximal margin (Fig. 31). The maximal diameter of the element is 30.5 mm and its thickness is 7.5 mm. It was situated distal to the spatium interosseum, and the con- cavity of its proximal margin indicates that the perforating artery passed proximal to the inter- medium between the distal heads of radius and ulna (Rieppel, 1993c). The other two carpals most probably represent the ulnare (#212) and distal carpal 4 (#154). Additional carpal elements, if present, remain unknown.
Five metacarpals are known and have been ar- ranged by Huene (1952) from digits I through V as follows (Fig. 32): #16 (50.4 mm), #1 12 (53 mm), #37 (58.5 mm), #151 (53.5 mm), and #42 (53 mm). Following that reconstruction, metacarpal I would be the shortest, metacarpal III the longest in the series. The phalangeal formula of the manus is unknown.
26
FIELDIANA: GEOLOGY
spa
scr.f
B
Fig. 28. The pelvic girdle of Simosaurus gaillardoti (smns 14733). A, Left ilium, lateral view; B, left ilium, medial view; C, right pubis, lateral (ventral) view; D, right ischium, lateral (ventral) view. Abbreviations: ilb, iliac blade; obf, obturator foramen; scr.f, sacral rib facets; spa, spina praeacetabuli. Scale bar = 20 mm.
Hindlimb
The femur (Table 6) of Simosaurus is a slender and only very weakly curved bone (Fig. 3 3 A). In the associated material of smns 14733, the femur (#81) is distinctly shorter than the humerus and less expanded both proximally and distally (the same is true for the specimen described by Huene, 1959). Relating the distal expansion to the length of the bone, however, the ratio for the femur (4. 1- 5.3) is not significantly different from the ratio
obtained for the humerus (4.2-5.9). The same re- sults if distal width is related to proximal width (femur: 1.1-1.2; humerus: 0.8-1.2). The slender appearance of the femur results from a greater degree of constriction in the middiaphyseal part. If minimal width of the femur is related to the length of the bone, the ratio ranges from 8.6 through 1 1 .9; comparable values (length/minimal width) for the humerus are 6.8-9.5.
The ventral side of the proximal head shows a fairly well developed trochanter, but the latter's
RIEPPEL: SIMOSAURUS GAILLARDOTI
27
Fig. 29. The pubis of Not hosaur us sp. A, smns 18516, upper Muschelkalk, Heldenmiihle near Crailsheim; B, smns 55853, Grenzbonebed, Zwingelhausen; C, Oberfrankisches Erdgeschichtliches Museum, Bayreuth, bt uncata- logued (original of Meyer, 1847-1855, PI. 41, Fig. 3). Abbreviation: obf, obturator foramen. Scale bar = 20 mm.
head is not distinctly set off from the shaft of the femur by a deep intertrochanteric fossa. The de- velopment of the trochanter results in a triangular cross-section of the proximal head of the femur. The articular condyles for tibia and fibula are not separated from one another and in level; there is, however, a shallow intercondylar fossa separating the tibial and fibular articulations.
The tibia (Table 7) is a relatively robust element, generally thicker than the fibula, with rather straight lateral edges and only weakly expanded proximal and distal articular heads (Fig. 33B). The fibula (#182; length: 90 mm; proximal width: 26 mm; minimal width: 12.5 mm; distal width: 26.5 mm) is of similar length to the tibia, but generally more slender and distinguished by a strongly concave
anterior (preaxial margin) (Fig. 33C). The latter contributes to the formation of a spatium inter- osseum but also emphasizes the appearance of markedly expanded proximal and distal articular heads.
The pes, again, is fully disarticulated, but the tarsus certainly had three ossifications (astragalus, calcaneum, and distal tarsal 4); Huene (1952) en- tertained the possibility of the presence of a fourth ossification in the tarsus (distal tarsal 3). As a gen- eral rule, however, the carpus retains a greater number of ossifications in those sauropterygians with a different number of ossified elements in the carpus and tarsus. With a disarticulated hand and foot, the morphology of the hand and foot may never be completely known, but the logical alter-
Fig. 30. The forelimb of Simosaurus gaillardoti (smns 14733). A, Left humerus (#1 of Huene, 1952); B, left radius (#89 of Huene, 1952); C, left ulna (#93 of Huene, 1952). Abbreviations: dpcr, deltopectoral crest; ecg, ectepi- condylar groove. Scale bar = 20 mm.
28
FIELDIANA: GEOLOGY
RIEPPEL: SIMOSAURUS GAILLARDOTI
Table 2. Measurements (in millimeters) of the pubis and ischium of Simosaurus gaillardoti (smns 14733).
Table 4. Measurements (in millimeters) of the ra- dius of Simosaurus gaillardoti (smns 14733).
|
Measurements |
Specimen part #112, right #89, left |
Measurements |
|||||
|
Specimen part Left pubis Right pubis |
Breadth 116 118 121 |
Lateral length 72.5 70.9 51.7 52.5 |
Minimal length 37.8 39 31.5 31 |
Medial length 98 95.5 119.5 118.5 |
Length 128 126 |
Lateral Minimal width width 31.5 15.5 30.5 15 |
Medial width 20.5 22 |
|
Left ischium Right ischium |
natives available are 1) Simosaurus presents an exception to the rule, with three carpal and four tarsal ossifications, 2) the carpus of Simosaurus is incompletely preserved, or 3) the fourth tarsal os- sification counted by Huene (1952) has a different identity.
One of the tarsal elements (#123) can unequiv- ocally be identified as the astragalus (Figs. 3 IB, 34). It is larger than the intermedium, with a max- imal diameter of 33.8 mm and a thickness of 15 mm, but it shares with the latter bone a distinct concavity on its proximal margin. This indicates, again, that the perforating artery did not pass be- tween the astragalus and calcaneum, but proximal to the astragalus through the spatium interosseum between the distal heads of the tibia and fibula.
A total of six metatarsals are preserved, out of which four elements (Fig. 35) can be attributed to the left metatarsus (R. Wild, pers. coram.). Pro- ceeding from digits I through V, their respective absolute length is as follows: #101,38.5 mm; #192, 44.5 mm; #66, 46 mm; metatarsal 4, missing; and #55, 31 mm. Following this reconstruction, meta- tarsal 3 is the longest and metatarsal 5 the shortest of the series. The phalangeal formula of the pes is unknown.
Functional Morphological Correlates in the Skeleton of Simosaurus gaillardoti
Simosaurus gaillardoti was a sauropterygian of fairly large size (Fig. 36). The trunk of the subadult specimen from the Gipskeuper of Obersontheim measures 120 cm. Huene (1959) reconstructed the specimen with a cervical vertebral column of sim- ilar length as the trunk. This claim may be sup- ported with reference to the articulated skeleton of Nothosaurus raabi (mb 1.1007.18; Schroder, 1914), whose glenoid-acetabulum length mea- sures approximately 255 mm and the neck region approximately 225 mm. In that specimen, the neck is therefore approximately 88% of trunk length. In the largest pachypleurosaur (Neusticosaurus ed- wardsii; Carroll & Gaskill, 1985), the neck is rel- atively shorter, on the average about 55% of trunk length. In view of other close skeletal resemblances with Nothosaurus (see the cladistic analysis be- low), similar body proportions may be proposed for Simosaurus, which results, for the subadult specimen of Obersontheim (Huene, 1952, mea- sured 1 9 cm for basicranial length), in an approx- imate snout-vent length of 248 cm. Paranotho- saurus amsleri (Peyer, 1939; probably congeneric
Table 3. Measurements (in millimeters) of the humerus of Simosaurus gaillardoti.
|
Measurements |
||||
|
Proximal |
Minimal |
Distal |
||
|
Specimen number |
Length |
width |
width |
width |
|
smns 14733 |
262 |
56 |
27.5 |
59.5 |
|
smns 14733 |
261 |
52 |
28 |
60.5 |
|
smns 7956 |
270 |
53.5 |
31.3 |
45.7 |
|
smns 17590 |
342 |
65 |
45 |
82 |
|
smns 18287 |
288 |
57.9 |
37.5 |
64 |
|
smns 18658 |
210 |
39.5 |
29 |
ca. 48.5 |
|
smns 18686 |
ca. 307 |
ca. 67 |
45.2 |
— |
|
smns 52095 |
ca. 315 |
59.9 |
43.9 |
— |
|
smns uncatalogued |
276 |
51 |
38 |
53 |
30
FIELDIANA: GEOLOGY
Table 5. Measurements (in millimeters) of the ulna of Simosaurus gaillardoti (smns 14733).
Table 6. Measurements (in millimeters) of the femur of Simosaurus gaillardoti.
|
Measurements |
|||
|
Specimen part |
Length |
Lateral Minimal width width |
Medial width |
|
#74, right #93, left |
127.5 127.5 |
43.5 17.7 44.6 16.9 |
27.4 28.7 |
with Nothosaurus: Rieppel & Wild, 1994) is a nothosaur from the Alpine Triassic of Europe known from complete and articulated specimens. In the adult specimen described by Peyer (1939), tail length averages about 65% of trunk length. A fully grown Simosaurus may thus have reached a total length of up to 4 m. The skeletal reconstruc- tion on display at the Stuttgart Museum measures approximately 3 m.
The shape of the distally expanded dorsal ribs results in a broad, somewhat dorsoventrally flat- tened body in the skeletal reconstruction. The skull is brevirostrine, broad, and flat in general outline, with a relatively long temporal region, large upper temporal fossae, and posteriorly displaced man- dibular articulation. As is characteristic for some eusauropterygians, the tooth row is extended back- ward well beyond the posterior margin of the orbit, indeed well beyond the anterior margin of the up- per temporal fossa. The anterior teeth of Simo- saurus are somewhat enlarged and procumbent,
Fig. 31. A, Intermedium of Simosaurus gaillardoti (smns 14733, #210 of Huene, 1952), in dorsal (left) and ventral (right) views; B, astragalus of Simosaurus gail- lardoti (smns 14733, #173 of Huene, 1952), in dorsal (left) and ventral (right) views. Scale bar = 20 mm.
|
Measurements |
||||
|
Mini- |
||||
|
Specimen |
Proximal |
mal |
Distal |
|
|
number |
Length |
width |
width |
width |
|
smns 14733 |
203 |
46 |
17 |
39 |
|
smns 17223 |
ca. 215 |
ca. 49.5 |
21 |
41 |
|
smns 18038 |
142 |
34 |
14 |
ca. 28.8 |
|
smns 18676 |
129 |
29.5 |
12.7 |
24.3 |
|
smns 18689a |
219 |
51.9 |
25.5 |
49 |
|
smns 19052 |
256 |
- |
25.5 |
62 |
suited for prey capture. The lateral teeth are rather short and blunt, characterized by an expanded and strongly striated crown. The morphology and ver- tical position of the lateral teeth of Simosaurus, correlated with a vertical mode of tooth replace- ment (Jaekel, 1905, 1907), are a striking contrast to the laterally directed and needle-shaped teeth of Nothosaurus and potentially indicative of du- rophagous habits in Simosaurus.
Mandibular mechanics appear less well suited to crush hard-shelled prey, however. Durophagy in reptiles is generally correlated with a deep lower jaw, a high coronoid process, and a load arm that is relatively short compared to the force arm (Rieppel & Labhardt, 1979, and references there- in). None of these features is observed in Simo- saurus. The lower jaw is a slender, almost delicate structure without a coronoid process, and the pos- terior extension of the tooth row results in a short force arm relative to a potentially very long load arm. A short force arm relative to a potentially long load arm is usually correlated with a quick snapping bite (Robinson, 1973). The mechanical
Fig. 32. Metacarpal series of Simosaurus gaillardoti (smns 14733). Scale bar = 20 mm.
RIEPPEL: SIMOSAURUS GAILLARDOTI
31
Fig. 33. Hindlimb of Simosaurus gaillardoti (smns 14733). A, Right femur (#81 of Huene, 1952); B, tibia (#51 of Huene, 1952); C, fibula (#182 of Huene, 1952). Abbreviation: tr, trochanter. Scale bar = 20 mm.
advantage of the jaw adductor musculature is de- pendent on two principal parameters: 1) the in- sertional angle of the jaw adductor muscle fibers and/or of their insertional tendon, relative to the long axis of the lower jaw, and 2) the distance of the muscular and/or tendinous insertion from the fulcrum point, that is, from the mandibular joint (Gans & Bock, 1965; Alexander, 1983; Rieppel, 1978). A third parameter of importance is the height of the coronoid process, which becomes
more critical the more acute the angle of insertion (DeMar & Barghusen, 1972).
The greatest torque is transmitted to the lower jaw by vertically inserting adductor musculature. With a skull as flat as that of Simosaurus, vertically positioned muscle fibers would have a short ab- solute length and hence a very restricted excursion range within a tolerable range of the length/tension ratio (see Rieppel, 1984, 1985, for further discus- sion). The relative degree of passive stretching in-
32
FIELDIANA: GEOLOGY
Table 7. Measurements (in millimeters) of the tibia of Simosaurus gaillardoti.
|
Measurements |
||||
|
Specimen number |
Length |
Lateral width |
Minimal width |
Medial width |
|
smns 14733 smns 15978 smns uncatalogued |
ca. 90 90.7 90 |
33.5 29.5 29 |
ca. 22 17.5 18 |
ca. 30.5 29.5 28 |
curred by a jaw adductor muscle fiber upon jaw opening depends on the insertional angle and is maximal with a vertical orientation of the fiber relative to the long axis of the lower jaw (Rieppel, 1978). Vertically oriented jaw adductor muscle fi- bers would not only be the shortest in absolute terms (due to the flat skull) but would also incur the greatest degree of relative stretching upon jaw opening. The force a muscle may generate upon contraction depends on the degree of passive stretching prior to contraction (Gans & Bock, 1 965; Alexander, 1983). Muscle force decreases with an increasing degree of passive stretching, according to the length/tension ratio, which depends on the degree of overlap of actin and myosin filaments within each sarcomere. The absolute excursion range of a muscle fiber therefore increases with increasing absolute length, since passive stretching will be spread across a larger number of individual sarcomeres, each remaining within a tolerable range of the length/tension ratio. However, all simple relations of muscle fiber action and jaw closure will be complicated by the complex jaw adductor muscle architecture generally encountered in rep- tiles (Gans & De Vree, 1987; Gans et al., 1985). The large upper temporal fenestrae of Simosau- rus indicate a large physiological cross-section of the jaw adductor musculature, but most fibers ap- pear to have been positioned at an oblique angle
relative to the long axis of the lower jaw (Rieppel, 1989a). The lateral ledge on the surangular indi- cates a well-developed m. adductor mandibulae externus superficialis, which would have originat- ed from the lower surface of the posterior part of the upper temporal fossa and from the anterior surface of the posteroventrally slanting quadrate, and which would have inserted into the lateral surface of the surangular. Accordingly, these fibers would have slanted from an anteroventrally po- sitioned insertion to a posterodorsally positioned point of origin. The same orientation of muscle fibers has to be assumed for deeper layers of the jaw adductor musculature originating from the posterior temporal region and inserting into the central insertional tendon and adductor fossa of the lower jaw. The mechanical advantage of these fibers increases with increasing degree of jaw clo- sure. Aside from the adductive component, these fibers would exert a retractive force on the lower jaw ramus, which increases with increasing degree of jaw opening.
Muscle fibers originating from the broad ante- rior corner of the upper temporal fossa would have to slant in a posteroventral direction to insert into the central insertional tendon fastened to the lower jaw ramus behind the posteriorly extended tooth row. Aside from the adductive component, these
Fig. 34. Astragalus of Simosaurus gaillardoti (smns Fig. 35. Metatarsal series of Simosaurus gaillardoti
14733, #123 of Huene, 1952). Scale bar = 10 mm. (smns 14733). Scale bar = 20 mm.
RIEPPEL: SIMOSAURUS GAILLARDOTI
33
fibers would exert a strong anterior pull, as would the pterygoideus muscle, which was strongly de- veloped with a large anterior (dorsal) portion orig- inating on the dorsal surface of the palate behind the orbit. The mechanical advantage of these mus- cle portions with respect to jaw adduction increas- es with increasing degree of jaw opening. The an- terior pull of the anterior adductor muscle fibers and of the pterygoideus muscle would increase with increasing degree of jaw closure. The mor- phology of the lower jaw joint (see the description above) appears to reflect these mechanical corre- lates, in that the lower jaw is locked against the quadrate to counteract the anterior pull from the pterygoideus muscle and anterior portions of the external jaw adductor muscle, which may have been only partially compensated for by the re- tractive component generated by the posterior jaw adductor muscle fibers.
It appears, therefore, that the pterygoideus mus- cle as well as anterior portions of the adductor musculature were able to effect a strong and rapid jaw closure in Simosaurus. A strong pterygoideus musculature, as indicated by the development of an anterior (dorsal) portion in Simosaurus, would be particularly critical to achieving rapid jaw clo- sure against hydrodynamic drag (Taylor, 1987, p. 176). Any crushing effect with the jaws near to closure would have to be effected by the posterior portions of the jaw adductor musculature with fi- bers pulling in a posterodorsal direction, and whose retractive force component may be compensated by isometric contraction of the antagonistic an- terior adductor fibers and pterygoideus muscle portions. The only means to increase the crushing force on the tooth row generated by the postero- dorsally slanting posterior adductor fibers is by increasing the height of the coronoid process (DeMar & Barghusen, 1972; Rieppel, 1985), but this process is entirely absent in Simosaurus.
It may be noted at this juncture that Olson (1961) presented a biomechanical model of the evolution of tetrapod jaw mechanics, which are postulated to have changed from a kinetic inertial system favoring rapid initial jaw closure in anthracosau- rian amphibians to the static pressure system fa- voring strong adductive forces with the jaws near to closure in reptiles. Olson (1961) also suggested that secondarily aquatic reptiles tend to revert to a kinetic inertial system. As argued on an earlier
Fig. 36. Reconstruction of the skeleton of Simosau- rus gaillardoti (approx. 7.5% of natural size).
34
FIELDIANA: GEOLOGY
Fig. 37. The dual jaw adductor system of Simosaurus. For further discussion, see text.
occasion (Rieppel, 1989a), this model of the evo- lution of jaw mechanics in secondarily aquatic rep- tiles seems too simple, since Simosaurus shows a dual system (Fig. 37), with an anterior group of muscles favoring rapid jaw closure ("kinetic in- ertial fibers") and a posterior group of fibers fa- voring adduction with the jaws near to closure ("static pressure fibers"). A similar dual system was reconstructed by Taylor (1992) for the large plesiosaur Rhomaleosaurus from the Lower Ju- rassic of England and may, indeed, be character- istic of eusauropterygians in general.
Simosaurus lacks the longirostrine skull with jaws bearing long, slender, and laterally directed teeth as are observed in the piscivorous Notho- saurus species. Simosaurus also lacks the high cor- onoid process and the deep posterior temporal re- gion observed in the lower jaw and skull of the obviously durophagous placodonts. The latter also show strongly procumbent and chisel-shaped an- terior teeth in the upper and lower jaw, which were presumably used to pick up hard-shelled sessile prey such as bivalves from the substrate and which, again, are absent in Simosaurus. On the other hand, Simosaurus appears to have been able to perform a strong and quick snapping bite, followed by a moderate crushing action with the jaws near to closure. Such a pattern of jaw mechanics might be expected in a predator of moderately hard-shelled prey such as hard-scaled fishes (of "holostean" grade) or, perhaps, ammonoids (Ceratites).
The orientation of the intervertebral articula- tions indicates that lateral undulation may have played a reduced role in the locomotion of Si- mosaurus. The standard zygapophyseal articula- tions are strengthened not only by an additional zygosphene-zygantrum articulation, but also by the differentiation of additional infrapre- and in-
frapostzygapophyses in the trunk, in the sacrum, and at least in the proximal tail region. The pre- and postzygapophyses as well as the infrapre- and infrapostzygapophyses show a distinct tendency toward dorsolateral inclination of the articular sur- face, increasing along an anteroposterior gradient in the vertebral column. A tendency toward dor- solateral inclination of the prezygapophyses (ven- trolateral inclination of postzygapophyses) is also observed in the vertebral column of Placodus, again increasing along an anteroposterior gradient, but the same structural trend is not observed in the vertebral column of Nothosaurus. Accordingly, paraxial propulsion is also assumed to have been the predominant mode of locomotion in placo- donts, especially in those taxa characterized by the development of dermal armor such as Cyamodus (Pinna & Nosotti, 1989; see summary in Storrs, 1993b). In plesiosaurs, zygapophyseal articula- tions between vertebrae are strongly inclined (Brown, 1981), such that the width across the zyg- apophyses is less than centrum width throughout the vertebral column (Storrs, 1 993a, character 48). The morphology of intervertebral articulations in plesiosaurs correlates with a shift from axial pro- pulsion to paraxial propulsion, also reflected by the shortening of trunk and tail and modification of the limbs to form paddles (Storrs, 1993b). In Simosaurus, an initial shift to paraxial locomotion may be reflected by the changing orientation of vertebral zygapophyses and the dorsoventrally flattened trunk. Interlocking of pre- and postzyg- apophyses and infrapre- and infrapostzygapophy- ses with almost vertically oriented articular sur- faces effectively reduces the potential for lateral undulation in the posterior part of the axial skel- eton, whereas the neck and anterior trunk regions with more horizontally oriented pre- and postzyg-
RIEPPEL: SIMOSAURUS GAILLARDOTI
35
apophyses would preserve their potential for ex- tensive lateral bending during prey capture. The flat skull would tend to minimize drag as the head was swept laterally in pursuit of prey (Taylor, 1987, p. 190).
Principal thrust in forward locomotion seems to have been generated by the forelimbs of Si- mosaurus rather than through lateral undulation. The stylo- and zeugopodial elements of the fore- limbs are much more strongly developed than those of the hindlimbs in the two associated skeletons available (ontogenetic variation of limb propor- tions remains unknown for Simosaurus). The strong reduction of the dorsal blade of the scapula would seem to prevent the upstroke of the front limb from becoming a major component of the propulsion-generating limb movement. Simosau- rus, therefore, had not developed full capabilities of underwater flight. The strongly developed ven- tral parts of the pectoral girdle correlate with a well-developed deltopectoral crest on the humerus and indicate that the downstroke of the front limb may have been the principal thrust-produc- ing movement. This would result in a type of lo- comotion intermediate between true underwater flight and rowing, corresponding to the otariid model developed by Godfrey (1984) for plesio- saurs (see also Storrs, 1991, 1993b); Carroll and Gaskill (1985) postulated a similar type of loco- motion for the pachypleurosaur Neusticosaurus edwardsii, and Storrs (1991, 1993b) for Corosau- rus. In Simosaurus smns 14733, the ratio of hu- merus length/femur length is 1 .28; the correspond- ing ratio ranges from 1 . 1 3 to 1 .84 in Neusticosaurus edwardsii.
Functional morphological correlates in the skel- eton of Simosaurus result in the reconstruction of its mode of life as an animal of the shallow open water, roaming the sea at moderate speed in the search for free-swimming fish and, perhaps, mol- lusk prey such as ammonoids (Ceratites). Such conclusions contrast with S. Schmidt's (1988) re- construction of Simosaurus as a primarily terres- trial animal.
Phylogenetic Analysis
Phylogeny reconstruction will be based on a number of "paradigm" terminal taxa, selected ac- cording to sufficient current knowledge of their skeletal morphology to be useful for phylogenetic analysis. This procedure is justified with reference to the fact that missing data may influence cladistic
analysis in rather unpredictable ways (Platnick et al., 1991; see also Appendix I). However, as more terminal taxa become better known in the future and are added to the analysis of cladistic interre- lationships of the Sauropterygia, the results pre- sented in this study may change.
The Pachypleurosauroidea is generally treated as a monophyletic taxon, sister-group to the Eu- sauropterygia (plus Placodontia sensu Storrs, 1991, 1993a; see also Rieppel, 1987, 1993a). The mono- phyly of the Pachypleurosauroidea is, however, rather weakly supported (by the probable absence of the ectopterygoid and the complex morphology of the retroarticular process; Rieppel, 1989a) and may be called into question by current revisionary work focusing on a redescription of the pachy- pleurosaurs from the lower Muschelkalk (Rieppel, 1993d) and the redescription of the Chinese ma- terial of Keichousaurus by Lin Kebang (Ph.D. the- sis in progress). It is for these reasons that the well- known genera Dactylosaurus (Sues & Carroll, 1985; and pers. obs. on the holotype of Dactylosaurus schroederi Nopcsa, 1928), Serpianosaurus (Riep- pel, 1 989a), and Neusticosaurus (Carroll & Gaskill, 1985; Sander, 1989) are chosen to represent the pachypleurosaurs. Serpianosaurus and Neustico- saurus are here coded as a single clade (terminal taxon) based on earlier evidence of monophyly (e.g., extreme reduction of the upper temporal fos- sa, gastral rib morphology; Rieppel, 1987, 1993b).
The Nothosauridae (sensu Rieppel, 1984) com- prise the genera Lariosaurus, Ceresiosaurus, Noth- osaurus, and Paranothosaurus. Of these, the gen- era Nothosaurus and Lariosaurus will be used as terminal taxa. Nothosaurus is known from abun- dant cranial and postcranial material deposited in public repositories (see Appendix II). The genus Lariosaurus is currently under revision (Rieppel & Tschanz, work in progress) and remains con- troversial until this work is completed (Tschanz, 1989; Kuhn-Schnyder, 1990; Storrs, 1993a). The holotype of Lariosaurus balsami was destroyed during World War II, and the "Munich specimen" (bsp AS 1 802) was designated the neotype by Kuhn- Schnyder (1987). Another well-preserved and complete specimen of Lariosaurus is the "Frank- furt specimen" (smf R- 1 3), purchased by E. Riip- pell in 1850 and described by Boulenger (1898). The specimen comes from Perledo, the type lo- cality for the genus, and together with the "Munich specimen" it forms the basis for inclusion of Lario- saurus in the present analysis. Paranothosaurus is probably a junior synonym of Nothosaurus (Riep- pel & Wild, 1994). The genus Ceresiosaurus is known from Peyer's (1931a) published work; this
36
FIELDIANA: GEOLOGY
material has since been reprepared, and new ma- terial has been collected. The material of Ceresio- saurus currently available requires redescription before it can be entered into phylogenetic analysis, but it is not available for study at the present time.
Additional terminal taxa used in this study are the genera Cymatosaurus and Pistosaurus. Data for Pistosaurus result from personal inspection of the only surviving skull of Pistosaurus longaevus (see Edinger, 1935, for a history of the specimens of Pistosaurus); the postcranial skeleton of Pisto- saurus was described by Sues (1987a). Cymato- saurus is one of the most problematical taxa used in this analysis, because of the very incomplete knowledge of its postcranial skeleton. However, two isolated humeri are included in this study (see Appendix II) that were identified as those of Cy- matosaurus. The reason is that both elements are very distinctive, and both come from lower Mu- schelkalk deposits that have yielded Dactylo- saurus and/or Anarosaurus, Nothosaurus, and Cy- matosaurus but no other taxon. Data for Coro- saurus are taken from Storrs (1991) and fmnh specimens.
Placodus will be used as a representative genus for the Placodontia, on the basis of an ongoing revision of its osteology and taxonomy. Parapla- codus, purportedly the most plesiomorphic pla- codont with respect to some characters (i.e., den- tition; Peyer, 1931b, 1935), requires redescription of a fairly complete specimen (Kuhn-Schnyder, 1942) before it can reliably be used in phytogeny reconstruction. Some important characters in the skull of Paraplacodus have been recorded by Zan- on (1989, and pers. obs.).
The phylogenetic analysis of sauropterygian in- terrelationships presented here relies heavily on the work of Gauthier, Kluge, and Rowe (1988) on amniote relationships. Their data matrix for am- niotes forms the core of the present analysis, after deletion of all those characters informative with respect to or within the Synapsida only. The re- maining list of characters was further augmented by the addition of characters taken from Evans (1988) and Storrs ( 1 99 1 , 1 993a) and by characters that emerged from personal inspection of the sau- ropterygian material referred to in Appendix II. Due to the addition of new characters and rede- finition of old ones, the numbering of the char- acters given below no longer corresponds to char- acter numbers in Gauthier, Kluge, and Rowe (1988), Evans (1988), Storrs (1991, 1993a), or Rieppel (1993a). Character codings of archosau- romorph taxa are mostly the same as in Gauthier, Kluge, and Rowe (1988) except for the Protoro-
sauria (for which terminal taxon they were unable to recover diagnostic characters; Gauthier, Kluge, & Rowe, 1988, p. 205). Their Protorosauria is here replaced by the Prolacerti formes (see discussion in Evans, 1988, p. 227), and the prolacerti form characters are taken from Wild ( 1 973), Gow ( 1 975), Chatterjee (1980), Evans (1988), and Rieppel ( 1 989b). Claudiosaurus (Carroll, 1 98 1 ) was added to their list of taxa, and Lepidosauromorpha were broken down into potential subgroups, viz., Youn- giniformes, Kuehneosauridae, Rhynchocephalia (Gephyrosaurus and Sphenodontida; Gauthier, Estes, & de Queiroz, 1988), and Squamata. If dif- ferent from Gauthier, Kluge, and Rowe ( 1 988) and Evans (1988), character codings are based on the following published descriptions:
Captorhinidae: Heaton (1979), Heaton and Reisz (1980, 1986)
Testudines: Gaffney (1979b, 1990)
Araeoscelidia: Vaughn (1955), Reisz (1981), Reisz etal. (1984)
Younginiformes: Gow (1975), Carroll (1981), Currie(1981, 1982)
Kuehneosauridae: Robinson (1962), Colbert (1970)
Rhynchocephalia: Robinson (1973), Evans (1980, 1981), Fraser(1982, 1988a,b), Fraser and Walkden (1984), Whiteside (1986)
Squamata: Gauthier, Estes, and de Queiroz (1988), and pers. obs.
Rhynchosauria: Chatterjee (1974), Benton (1983)
Choristodera: Sigogneau-Russell (1979, 1981), Sigogneau-Russell and Russell (1978)
Trilophosaurus: Gregory ( 1 944)
Definition of Characters
1. Premaxillae small (0) or large (1), forming most of snout in front of external nares.
The character translates into posteriorly posi- tioned ("retracted") external nares.
2. Premaxilla without (0) or with ( 1 ) postnarial process, excluding maxilla from posterior margin of external naris.
3. Snout unconstricted (0) or constricted (1). Constriction of the rostrum at the level of the
anterolateral margin of the external naris where the premaxilla meets the maxilla is characteristic of longirostrine sauropterygians such as the Noth- osauridae (sensu Rieppel, 1 994a, including Notho- saurus and Lariosaurus) and Cymatosaurus. The character also occurs in Placodus, but it is absent
RIEPPEL: SIMOSAURUS GAILLARDOTI
37
Fig. 38. Skull of Pistosaurus longaevus (the "second" specimen of Meyer, 1847-1855, PI. 22, Fig. 1; Oberfrank- isches Erdgeschichtliches Museum, Bayreuth, bt uncatalogued). A, Oblique dorsolateral view; B, lateral view, partially restored. Abbreviations: ec, ectopterygoid; ep, epipterygoid; f, frontal; ju, jugal; m, maxilla; n, nasal; p, parietal; pm, premaxilla; po, postorbital; pof, postfrontal; prf, prefrontal; pt, pterygoid, q, quadrate; sq, squamosal. Scale bar = 20 mm.
in the pachypleurosaur genera Dactylosaurus, Ser- pianosaurus, and Neusticosaurus. Simosaurus shows no conspicuous constriction of the broad and relatively short snout at the level of the pre- maxillary-maxillary suture, although a weak em- bayment can be seen in relatively small skulls (Si- mosaurus "guilielmi"). This may reflect initial development of snout constriction or secondary loss of that character. Snout constriction is coded as absent (0) for Simosaurus in order to avoid influence on pattern reconstruction by a priori hy- potheses of character transformation. Pistosaurus has a longirostrine skull with a tapering and point- ed snout but without distinct rostral constriction.
4. Nasals shorter (0) or longer (1) than fron- tal(s).
5. Nasals entering posterior margin of external naris (0) or reduced (absent) with no contact with the margin of the external naris (1).
Reduction of the nasals is characteristic of Cy- matosaurus. Much reduced nasals have been de- scribed for Pistosaurus (Koken, 1893; Edinger, 1935; Sues, 1987a), but unequivocal identification of the nasal bones on the only available specimen of Pistosaurus longaevus is difficult (Fig. 38). The reduced nasals may have been better exposed on a second skull {Pistosaurus "grandaevus" of Mey- er, 1847-1855), which is now lost.
38
FIELDIANA: GEOLOGY
6. Nasals meet in dorsomedial suture (0) or are separated from one another by nasal processes of the premaxillae extending back to the frontal bone(s) (1).
This character is difficult to assess because of extreme variability. Both conditions occur in the pachypleurosaur genus Neusticosaurus, which shows the nasal processes of the premaxillae al- ways extending back to meet the frontal in ventral view. The nasals meet in a dorsomedial suture when they are broad enough to cover the premax- illary processes in dorsal view. In Dactylosaurus the nasals meet in the dorsal midline of the skull, but character variation cannot be assessed in this genus due to the lack of specimens. Within the Nothosauridae, the best known skulls are those of Nothosaurus. The nasals meet in a dorsomedial suture in all well-preserved skulls, but one speci- men of Nothosaurus (cf. N. mirabilis, smns 10977) shows the nasal process of the premaxilla meeting the frontal. The character is coded as polymorphic for that genus. Reduction of the nasals in Cyma- tosaurus and Pistosaurus results in an invariable contact of the premaxilla with the frontal: char- acters 5 and 6 are interdependent in these two genera. In Placodus the nasals are fused and broad- ly separate the premaxillae from the frontals (Sues, 1987b).
7. The lacrimal is present and enters the exter- nal naris (0) or remains excluded from the external naris by a contact of maxilla and nasal ( 1 ), or the lacrimal is absent (2). The character is coded as unordered.
8. The prefrontal and postfrontal are separated by the frontal along the dorsal margin of the orbit (0), or a contact of prefrontal and postfrontal ex- cludes the frontal from the dorsal margin of the orbit (1).
9. Preorbital and postorbital region of skull of subequal length (0), preorbital region distinctly longer than postorbital region (1), or postorbital region distinctly longer (2). The character is coded as unordered.
The elongated postorbital region of the skull is one of the few characters supporting Storrs's ( 1 99 1 , 1993a) conclusion that the Placodontia is the sis- ter-group of the Eusauropterygia (Rieppel, 1993a). In pachypleurosaurs, the postorbital skull is dis- tinctly shorter than the preorbital skull. In Placo- dus, the postorbital region of the skull is not as markedly elongated as it is in the Eusauropterygia included in this study, yet it still ranges from 1 1 5% (bt 13, original of Sues, 1987b) to 120% (in the holotype of Placodus hypsiceps"; Meyer, 1863) of the preorbital skull. In the Cyamodontoidea
(Cyamodus), the postorbital region is distinctly elongated, and Placodus is coded accordingly in this analysis. In the brevirostrine eusauropteryg- ian Simosaurus (smns 1 3060), the postorbital re- gion is about 1 50% (if measured to the mandibular condyles of the quadrate; approximately 135% if measured to the posterior margin of the squa- mosal) of the preorbital skull. In Nothosaurus mi- rabilis, the nothosaur with the relatively longest rostrum, the postorbital region of the skull still amounts to 145% of the preorbital skull length (measurements based on the complete skull of bm[nh] 42829; 142% in the somewhat less well preserved specimen of Meyer, 1847-1855, PI. 2, Figs. 1-2; postorbital region measured from pos- terior margin of orbit to posterior end of squa- mosal).
10. Upper temporal fossa absent (0), present and subequal or larger than the orbit ( 1 ), or present but distinctly smaller than orbit (2). The character is coded as unordered.
Absence of an upper temporal fossa is the plesio- morphic condition; its presence is diagnostic of the Diapsida (Evans, 1988; Gauthier, Kluge, & Rowe, 1988). If present, the plesiomorphic size of the upper temporal fossa is roughly comparable to that of the orbit or somewhat smaller (1). In the pachypleurosaurs, the upper temporal fossa is very distinctly smaller than the orbit, although its relative size still varies: the upper temporal fossa is relatively larger in Dactylosaurus (Sues & Car- roll, 1985) than in the Serpianosaurus-Neustico- saurus clade (Carroll & Gaskill, 1985; Rieppel, 1989a; Sander, 1989). The significant increase in the size of the upper temporal fossa in relation to the size of the orbit in Placodus and in the Eusau- ropterygia is one of the characters supporting Storrs's ( 1 99 1 , 1 993a) conclusion that Placodus is the sister-group of the Eusauropterygia (Rieppel, 1993a). However, enlargement of the upper tem- poral fossa in these taxa is logically correlated with elongation of the postorbital region of the skull (character 9; see discussion above) and therefore is not coded separately in this study.
1 1 . Frontal(s) paired (0) or fused ( 1 ) in the adult.
12. Frontal(s) without (0) or with (1) distinct posterolateral processes.
1 3. The frontal does not enter the margin of the upper temporal fossa (0) or narrowly enters the anteromedial margin of the upper temporal fossa
(1).
The frontal is excluded from the anteromedial margin of the upper temporal fossa in Cymato- saurus cf. C. silesiacus (smns 10977; see also Schrammen, 1 899), but the frontal bone enters the
RIEPPEL: SIMOSAURUS GAILLARDOTI
39
Fig. 39. Incomplete skull of Cymatosaurus sp. (bgr uncatalogued, lower Muschelkalk, Sacrau, Gogolin [Po- land]. A, Dorsal view; B, occipital view. Abbreviations: bo, basioccipital; f, frontal; p, parietal; po, postorbital; pof, postfrontal; prf, prefrontal; pt, pterygoid, q, quad- rate; sq, squamosal. Scale bar = 20 mm.
anteromedial margin of the upper temporal fossa in Cymatosaurus friedericianus (Fritsch, 1 894, and pers. obs.; see also Schrammen, 1899) and in an unidentified partial skull of Cymatosaurus kept at the Bundesanstalt fur Geowissenschaften und Rohstoffe, Berlin (Fig. 39). The character is coded as polymorphic for Cymatosaurus. The frontal also narrowly enters the anteromedial margin of the upper temporal fossa in Pistosaurus.
14. Parietal(s) paired (0) or fused (1) in adult.
15. Pineal foramen at the middle of the skull table (0), displaced posteriorly (1), displaced an-
UP III ■lll.^....,.
Fig. 40. Parietal skull table of cf. Cymatosaurus (brg uncatalogued, lower Muschelkalk [Pecten- and Dado- crinus-beds], Gogolin [Poland]). Abbreviations: f, fron- tal; p, parietal. Scale bar = 20 mm.
teriorly (2), or absent (3). The character is coded as unordered.
The plesiomorphic condition of this character is the pineal foramen centered in the middle of the parietal skull table. Among sauropterygians, this condition is observed in pachypleurosaurs and in Cymatosaurus. In some taxa, including Placodus (bsp 1968.1.75, where the frontal enters the ante- rior margin of the foramen) and Pistosaurus, the pineal foramen is displaced anteriorly. In other Eusauropterygia, the pineal foramen is displaced to a posterior position in the parietal skull table.
16. Parietal skull table broad (0), strongly con- stricted (1), or forming a sagittal crest (2). The character is coded as unordered.
Constriction of the parietal skull table results in larger upper temporal fossae and provides a great- er area of origin for the deep jaw adductor mus- culature. The pachypleurosaurs retain a broad pa- rietal skull table, as does Placodus. Cymatosaurus usually shows a constricted parietal, although one (isolated) parietal retains a relatively broad skull table (Fig. 40). At this time, Cymatosaurus is cod- ed polymorphic for this character, although future revisionary work may prove the broad parietal to be characteristic of Eurysaurus (currently treated as a junior synonym of Cymatosaurus, but see Eurysaurus schafferi Arthaber, 1924). Simosaurus and Nothosaurus show a markedly constricted pa- rietal skull table, as does Pistosaurus.
In sauropterygians with a strongly constricted parietal skull table, the area for origin of deep jaw adductor muscles may be further increased by the development of a sagittal crest on the posterior part of the parietal. The character is polymorphic within the genus Nothosaurus {Nothosaurus edin- gerae; Schultze, 1970; Rieppel & Wild, 1994) and a sagittal crest is well developed in Pistosaurus.
17. Postparietals present (0) or absent (1).
18. Tabulars present (0) or absent (1).
19. Supratemporals present (0) or absent (1).
40
FIELDIANA: GEOLOGY
20. The jugal extends anteriorly along the ven- tral margin of the orbit (0), is restricted to a po- sition behind the orbit but enters the latter's pos- terior margin (1), or is restricted to a position behind the orbit without reaching the latter's pos- terior margin (2). The character is coded as unor- dered.
In the Serpianosaurus-Neusticosaurus clade, the jugal is a slender and curved element defining the posteroventral and ventral margin of the orbit, meeting the prefrontal anteriorly in most if not all specimens. A similar configuration of the jugal may be assumed for Dactylosaurus, where the character remains unknown. In Placodus, the jugal forms the ventral margin of the orbit and extends anteriorly beyond the level of the anterior margin of the orbit, an autapomorphic feature of the genus (Fig. 41). In Simosaurus, the main body of the jugal lies behind the orbit, but a slender process continues anteriorly to define the lower margin of the orbit. In Nothosaurus, the jugal remains re- stricted to a position behind the orbit, without entering the latter's posterior margin in most spec- imens. There is, however, a partial skull of Notho- saurus (smns 56838) that shows a narrow entry of the jugal into the posterior margin of the orbit. The genus is, therefore, coded polymorphic ( 1 and 2) for this character at this time, although a slight possibility exists that the generic identity of this skull may have to be revised in the future (the jugal has been said to enter the back of the orbit in a small (juvenile?) specimen of Ceresiosaurus [K. Tschanz, pers. comm.], and the presence of Ceresiosaurus in the German Triassic [Lettenkeu- per] may be indicated by an ulna [see below]). The relation of the jugal to the orbit in Lariosaurus remains controversial, due to the fact that none of the material is well enough preserved to allow the unequivocal identification of the character. In Cymatosaurus (smns 10977), the jugal bone de- fines the posteroventral margin of the orbit and extends anteriorly up to a level shortly behind the midpoint of the orbit (the genus is coded 1). In Pistosaurus, the anterior process of the jugal is hard to distinguish from the posterior part of the maxilla but seems to define the ventral margin of the orbit (coded 0).
2 1 . The jugal extends backward no farther than to the middle of the cheek region (0) or nearly to the posterior end of the skull (1).
In pachypleurosaurs, the jugal remains broadly separated from the squamosal, the plesiomorphic condition also observed in other diapsids (with the exception of some squamates). The jugal also remains separate from the squamosal in Placodus
Fig. 4 1 . Skull fragment of Placodus gigas (smf 4162, upper Muschelkalk, Bayreuth). Abbreviations: ju, jugal; m, maxilla; n, nasal; prf. prefrontal. Scale bar = 20 mm.
(Sues, 1987b), Cymatosaurus, and Simosaurus, al- though the latter genus shows the posterior exten- sion of the maxilla and jugal discussed as character 46. In Pistosaurus, the jugal meets the squamosal along the ventral margin of the cheek (upper tem- poral arch). In Nothosaurus, the posterior exten- sion of the maxilla and jugal results in a close approximation of the posterior tip of the jugal and the anterior tip of the squamosal, but whether a contact is established by the two bones remains unknown due to the problematical preservation of the material. The presence of a contact between squamosal and jugal is therefore not entered as a separate character in the phylogenetic analysis.
22. Lower temporal fossa absent (0), present and closed ventrally ( 1 ), or present but open ven- trally (2). Since the presence of a ventrally open lower temporal fenestra logically implies loss of the lower temporal bar, and hence the previous presence of a ventrally closed temporal fenestra, this character is coded as ordered.
Some controversy surrounded the interpreta- tion of the lower cheek region in sauropterygians, that is, whether the cheek had become emargin- ated ventrally or a lower temporal arch had been lost (Kuhn-Schnyder, 1962, 1967). Cladistic anal- ysis initially did not resolve the question (Rieppel, 1989a; Zanon, 1989), but a more recent investi- gation (Rieppel, 1993b) showed the assumption of loss of a lower temporal arch to be more par- simonious than the assumption of ventral emar- gination of the cheek. Since presence of the lower temporal arch (character 1) must logically precede loss of the lower temporal arch (character 2), this character is coded as ordered, but reversible (for a discussion of a secondary development of the
RIEPPEL: SIMOSAURUS GAILLARDOTI
41
lower temporal arch, see Whiteside, 1986, and Fraser, 1988).
Placodus is exceptional among the taxa here dis- cussed because it shows an upper but no lower temporal fossa; the latter is thought to have been closed secondarily (Sues, 1987). The jugal meets the quadratojugal (if present) along the lower mar- gin of the cheek (as is also the case in the Cyamo- dontoidea). Paraplacodus, on the other hand, shows a deeply excavated cheek region with a curved jugal defining the posterior margin of the orbit and without a posterior process, suggesting the loss of a lower temporal arch (Zanon, 1989; Pinna, 1989, and pers. obs.), and a deep excavation of the lower cheek region is also characteristic of at least some Cyamodontoidea (e.g., Placochelys; Jaekel, 1907, and pers. obs.). The hypothesis of secondary clo- sure of the cheek region in Placodus is thus ac- cepted.
23. Squamosal descends to (0) or remains broadly separated from (1) ventral margin of skull. See discussion of character 34.
24. Quadratojugal present (0) or absent (1). The quadratojugal is present in pachypleuro-
saurs, where it is situated in front of the ventrally descending process of the squamosal, which covers the quadrate laterally. The presence of a small quadratojugal in Simosaurus was first reported by Schultze (1970), but the squamosal still extends far down on the lateral surface of the quadrate. Rieppel ( 1 994a) described a quadratojugal in acid- prepared braincase material of Nothosaurus, but this report may be erroneous, the suture suppos- edly delineating the quadratojugal resulting from erosion of the bone surface. No quadratojugal is present in the well-preserved and well-prepared specimens of Nothosaurus described by Schroder (1914; see Appendix II), where the squamosal ex- tends far down toward the ventral margin of the skull. The same is true of Cymatosaurus (Fig. 39B) and Pistosaurus (Fig. 38), two genera in which the quadratojugal is absent. Sues (1987b) described a large quadratojugal in Placodus, restricting the squamosal to a dorsal position in the posterior temporal region of the skull, but its delineation from the squamosal is controversial (cf. Broili, 1912; Pinna, 1989; Sues found a faint suture de- lineating the squamosal from the quadratojugal in bt 13 but not in the smf specimens; pers. comm.). Personal observation of bt 13 and smns 18644 and of the holotype of Placodus hypsiceps Meyer, 1863 (a junior synonym of Placodus gigas), in my view supports Broili's (1912) conclusion that a quadratojugal may be absent, that is, fused to the squamosal in Placodus. However, since a large
quadratojugal is present in the Cyamodontoidea (Cyamodus kuhnschnyderi, Nosotti & Pinna, 1 993a, and pers. obs.; Placochelys placodonta, Jae- kel, 1907, and pers. obs.), the reconstruction of the skull of Placodus given by Sues (1987b) is here provisionally accepted, and Placodus coded 1 for character 23 and 0 for character 24 (quadratojugal present but fused to squamosal). Restriction of the squamosal of Placodus to a dorsal position on the posterior temporal region of the skull (due to the presence of a large quadratojugal) follows from both Sues's (1987b) and Pinna's (1989) recon- structions.
25. Quadratojugal with (0) or without (1) an- terior process.
The character is partially correlated with char- acters 21 and 22 defined above. Absence of the quadratojugal logically correlates with the absence of an anterior process of that bone. However, a complete lower temporal arch may be formed by the jugal alone, the posterior process of the jugal extending all along the lower margin of the lower temporal fenestra to meet the quadratojugal in a posterior position as the latter lost the anterior ramus.
26. Occiput with paroccipital process forming the lower margin of the posttemporal fossa and extending laterally (0), paroccipital processes trending posteriorly (1), or occiput platelike with no distinct paroccipital process and with strongly reduced posttemporal fossae (2). The character is coded as unordered.
In pachypleurosaurs, Simosaurus, Nothosaurus (Fig. 42A), and Lariosaurus, the occiput is plate- like and the posttemporal fossae much reduced or absent. Closure of the occiput is effected by the occipital exposure of the parietal and squamosal bones as well as by an expansion of the occipital exposure of the opisthotic and exoccipital bones. The occiput of Cymatosaurus and Pistosaurus is still insufficiently known at this time. In Placodus (as well as in the Cyamodontoidea), well-defined paroccipital processes extend posterolaterally (Fig. 42B).
27. Mandibular articulations approximately at level with occipital condyle (0) or displaced to a level distinctly behind occipital condyle (1), or po- sitioned anterior to the occipital condyle (2). The character is coded as unordered.
This character is coded as polymorphic for the genus Nothosaurus, taking into account the deeply excavated occiput in Nothosaurus juvenilis (Rieppel, 1994b). A posterior displacement of the mandibular joints generally results in a deeply ex- cavated occiput with posterolaterally trending par-
42
FIELDIANA: GEOLOGY
occipital processes; hence, this character would seem to be partially correlated with character 32. However, the mandibular joints are level with the occipital condyle in Placodus, where the paroccip- ital processes are trending posterolaterally.
28. Exoccipitals do (0) or do not (1) meet dorsal to the basioccipital condyle.
The exoccipital and basioccipital are completely fused in Captorhinus, such that the contribution of the exoccipital to the occipital condyle cannot be determined. In turtles, the exoccipitals usually participate in the formation of the occipital con- dyle (Gaffney, 1979b, p. 1 14). In diapsid reptiles, the occipital condyle is generally formed by the basioccipital only, and the exoccipitals remain separated above the occipital condyle. This is also the condition characteristic for the Eosauropteryg- ia and placodonts (except for an autapomorphous contact of the exoccipitals above the occipital con- dyle in Cyamodus).
29. Quadrate with straight posterior margin (0) or quadrate shaft deeply excavated (concave) pos- teriorly (1).
The posterior excavation of the shaft of the quadrate is indicative of the presence of a large tympanum and hence of an impedance-matching middle ear, as was present in pachypleurosaurs (Rieppel, 1989a). In Placodus, the quadrate is deeply excavated posteriorly, and there is no rea- son to believe that a tympanum was absent, al- though the relative size of the tympanum was re- duced in adaptation to aquatic habits. If a tympanum was present in Simosaurus, it would have been much reduced and housed in a dorsal recess formed by the squamosal bone (see discus- sion above, and Kuhn-Schnyder, 1961; Rieppel, 1989a). There is no posterior excavation of the quadrate in Simosaurus, nor in Nothosaurus, Cy- matosaurus, and Pistosaurus, which all seem to have lacked a tympanum.
30. Quadrate covered by squamosal and qua- dratojugal in lateral view (0), or quadrate exposed in lateral view (1).
This character is easily understood in a compar- ison of the plesiomorphic condition {Captorhinus, Petrolacosaurus, Youngina) and the apomorphic condition (lizard). Assignment of character states is less easy in other taxa such as Placodus and eosau- ropterygians. The quadrate is covered in lateral view in Placodus (by the squamosal and, possibly, the quadratojugal) and in Pistosaurus (by the squamosal only). In pachypleurosaurs, Simosaurus, Cymato- saurus, and nothosaurs, the quadrate is exposed in lateral view to some degree, at least, but this seems to result from a narrowing of the lateral edge of the
Fig. 42. A, Skull of Nothosaurus mirabilis (smns 56286, upper Muschelkalk. Berlichingen a.d. Jagst) in occipital view; B, skull of Placodus gigas (smf R-360, upper Muschelkalk, Bayreuth; original of Broili, 1912, p. 147) in occipital view. The arrow points to the par- occipital process of Placodus. Scale bar = 20 mm.
descending process of the squamosal and, if present, of the quadratojugal. Morphologically, the squa- mosal and, if present, the quadratojugal descend to a level narrowly above the mandibular condyle of the quadrate, such that the latter is coded as being laterally covered in these groups.
31. Lateral conch on quadrate absent (0) or present ( 1 ).
32. Palate kinetic (0) or akinetic (1).
This is a complex character that involves a suite of structures that relate the dermal palate to the base of the braincase in a palatobasal articulation. The plesiomorphic condition is characterized by some separation of the pterygoids in front of the palatobasal articulation. The pterygoids articulate with basicranial basipterygoid processes in a sy- novial joint. The basipterygoid processes form by ossifications of the basi trabecular processes which develop as lateral projections from the posterior end of the trabeculae cranii just in front of the point of fusion of the trabeculae with the acro- chordal cartilage. The acrochordal cartilage is
RIEPPEL: SIMOSAURUS GAILLARDOTI
43
sometimes considered to be part of the basicranial plate (Bellairs & Kamal, 1981), and it is positioned behind the hypophysis. As a consequence, the pal- atobasal articulation lies lateral to the sella turcica, a depression in the floor of the braincase that re- ceived the hypophysis in the fully developed skull.
An akinetic palate is characterized by the fusion of the pterygoids to the basicranium, which oblit- erates the basitrabecular process during ontogeny (Rieppel, 1 977) as well as the interpterygoid fossa. The development of an akinetic palate would thus seem to logically correlate with the absence (loss) of the basipterygoid processes. However, Sues (1987b) described large "basipterygoid processes" in Placodus in combination with an akinetic pal- ate. These processes lie behind the level of the sella turcica, and even if one takes into account the shortening of the posterior basicranium that characterizes the skull of Placodus, it is hard to see how basitrabecular processes may have con- tributed to the formation of the "basipterygoid processes" described by Sues (1987b). These pro- cesses emerge from the ventral surface of the bas- icranium immediately in front of the occipital con- dyle, and whereas there is some contribution of the basisphenoid to the anterior aspect of these processes, they seem to derive mainly from the basioccipital (Zanon, 1989; Nosotti & Pinna, 1993b). This interpretation is supported by in- spection of the cyamodontoid Placochelys placo- donta (mb R. 1 765), where basioccipital tubera are distinct and the basioccipital remains separate from the basisphenoid.
33. Basioccipital tubera free (0) or in complex relation to the pterygoid, as they extend ventrally (1) or laterally (2). The character is coded as unor- dered. This character was first recognized by Za- non (1989) and will be redefined here.
The plesiomorphic condition is the presence of posterolateral projections on the basioccipital (the spheno-occipital tubercle of Oelrich, 1956) serving the insertion of the longus colli muscle. In eosau- ropterygians and Placodus, the ventral surface of the basisphenoid and basioccipital are largely cov- ered by the posteriorly expanded pterygoids that meet all along the midline in the formation of an akinetic palate (a partial exposure of the posterior braincase floor constitutes a character reversal in Pistosaurus; Sues, 1 987a). On topological grounds, the "basipterygoid processes" of Placodus de- scribed by Sues represent enlarged basioccipital tubera supporting the posterior (quadrate) ramus of the pterygoid. Some of the flexor musculature of the head may still have inserted into their pos-
terior surface. In Simosaurus and Nothosaurus, basioccipital tubera are again present but no longer located ventral to the occipital condyle. Instead, they extend lateral to the occipital condyle, en- closing between them and the pterygoid the eusta- chian foramen (Rieppel, 1994a), which may be closed in some specimens of Nothosaurus. Al- though the topological relations of the basioccip- ital tubera are different as compared to Placodus, they still establish a complex relation to the quad- rate ramus of the pterygoid. The occiput of pachy- pleurosaurs, Cymatosaurus, and Pistosaurus is not well enough known to identify the character state for these taxa.
34. Suborbital fenestra (infraorbital foramen of Oelrich, 1956) absent (0) or present (1).
Sues (1987b) described a suborbital depression in the palate of Placodus, which he considered homologous to the suborbital fenestra of other diapsids. A distinct cleft is seen between ectoptery- goid, palatine, and maxilla in the skull of Placodus (Fig. 43), and a similar cleft is observed in Pla- cochelys placodonta (mb R. 1765). The infraorbital cleft is well developed in a fragmentary skull of Placodus (smns 18641), which shows, however, that the foramen has no connection to the floor of the orbit. A suborbital fenestra was reported for Paraplacodus by Zanon (1989), but personal in- spection of the material showed nothing but a tooth pushed upward and located in the posterolateral corner of the orbit. Placodus was accordingly cod- ed both for the presence and absence of the sub- orbital fenestra in the cladistic analysis, with no effect on tree topology. Since Placodus (and other sauropterygians) are diapsids, a reduction of the suborbital fenestra must be assumed. The subor- bital fenestra is absent in pachypleurosaurs as well as in all other Eusauropterygia included in this analysis.
Turtles present a particular problem with re- spect to this character. Gaffhey (1972) described a foramen palatinum posterius in the palate of turtles which has not generally been hypothesized to be homologous with the suborbital fenestra in diapsids. During early stages of ossification of the palate, however, the foramen palatinum posterius appears in the same topological position as does the suborbital fenestra (Rieppel, 1993e), and a ho- mology of these foramina has recently been pro- posed in a review of amniote phylogeny by Laurin and Reisz (1994). Again, turtles were coded for the presence and absence of this character.
Of all combinations of alternative coding of this character in turtles and Placodus, only one had
44
FIELDIANA: GEOLOGY
iof?
Fig. 43. A, Skull of Placodus gigas (smf R-359, upper Muschelkalk, Bayreuth) in ventral view; B, magnified view of the infraorbital fenestra of Sues (1987b). Abbreviations: ec, ectopterygoid; iof, infraorbital foramen; m, maxilla; pi, palatine. Scale bar = 20 mm.
any impact on the tree, and that was coding the infraorbital fenestra absent in turtles but present in Placodus. The tree preserved its topology (see discussion below), but its length increased by one step (using DELTRAN optimization), as the loss of the infraorbital fenestra was treated as conver- gence in turtles and Eosauropterygia.
35. Pterygoid flanges well developed (0) or strongly reduced (1).
The plesiomorphic condition is a dermal palate with well-developed transverse flanges formed by the pterygoid and ectopterygoid. These define the anterior margin of the suborbital fenestra and serve as the site of origin of a strong pterygoideus muscle (Carroll, 1969). Placodus retains well-developed pterygoid flanges, as does Corosaurus (Storrs, 1 99 1 , p. 1 8), Cymatosaurus, and Pistosaurus among the Eusauropterygia under consideration. Pterygoid flanges are very reduced or virtually absent in pachypleurosaurs as well as in Simosaurus, Notho- saurus, and Lariosaurus.
36. Premaxillae enter internal naris (0) or are excluded (1).
This character may, to some degree, be inter-
dependent with the elongation of the snout in lon- girostrine sauropterygians but is less ambiguously defined than snout elongation. Although charac- terized by a distinct rostrum set off from the pre- orbital skull by a rostral constriction, Placodus can hardly be viewed as a longirostrine skull in com- parison, for example, to (some) Nothosaurus, Cy- matosaurus, and Pistosaurus. Yet the premaxillae are excluded from the internal naris in Placodus (the confluent internal nares are autapomorphic for Placodus at the level of the present analysis), but not in the brevirostrine Simosaurus, nor in the Serpianosaurus-Neusticosaurus clade (Carroll & Gaskill, 1985; Sander, 1989).
37. Ectopterygoid present (0) or absent (1).
This character is entered here as a synapomor- phy (loss of ectopterygoid) of Dactylosaurus and the Serpianosaurus-Neusticosaurus clade, al- though it is admitted that empirical evidence for the absence (loss) of the ectopterygoid in pachy- pleurosaurs is weak because of the difficult nature of the material. However, the ectopterygoid is a rather distinct element in the palate of Simosaurus and Nothosaurus, so that the expectation would
RIEPPEL: SIMOSAURUS GAILLARDOTI
45
Fig. 44. A, Mandibular symphysis of Lariosaurus (smf R-13, upper Ladinian, Perledo), ventral view; B, man- dibular symphysis of Cymatosaurus (bgr uncatalogued, lower Muschelkalk [Pecten- and Dadocrinus-beds], Gogolin [Poland]); C, mandibular symphysis of Not hosaurus (mb R.6, lower Muschelkalk, Rudersdorf; original of Schuster & Bloch, 1925).
be that at least one or two among the hundreds of specimens of the Serpianosaurus-Neusticosaurus clade would have documented its presence.
38. Retroarticular process of lower jaw absent (0) or present (1).
39. Distinct coronoid process of lower jaw ab- sent (0) or present (1).
The coronoid process is distinct in Placodus but absent in pachypleurosaurs, Simosaurus, Notho- saurus, and Lariosaurus. The lower jaws of Cy- matosaurus and Pistosaurus remain incompletely known and unknown, respectively. Characters re- lated to the presence or absence of a coronoid process are the presence of a posterior coronoid process of the dentary bone and the presence or absence of the lateral exposure of the coronoid bone. Placodus shows a very limited lateral ex- posure of the coronoid bone, most of the coronoid process being formed by the dentary (Drever- mann, 1933; Huene, 1936). In eosauropterygians with no coronoid process, the coronoid bone is not exposed in the lateral view of the lower jaw.
40. Mandibular symphysis short (0) or elon- gated (1).
An elongated (deep) mandibular symphysis is a derived character supporting Storrs's (1991, 1 993a) conclusion that the Placondontia is the sister-group of the Eusauropterygia. In the plesiomorphic con- dition, the mandibular symphysis is short, as is true for the pachypleurosaur genera here consid- ered (there is one alleged pachypleurosaur, Ana- rosaurus multidentatus Huene, 1958, with an ex-
tended symphysis, but its affinity is, in fact, with Cymatosaurus [Rieppel, 1995]). The mandibular symphysis is not expanded in Simosaurus, but it is to a variable degree in the genera Nothosaurus, Lariosaurus, and Cymatosaurus (Fig. 44). The lower jaw of Pistosaurus remains unknown.
Elongation of the mandibular symphysis may be correlated with a distinct pattern of tooth re- placement. Whereas the maxillary, dentary, and palatine teeth of Placodus show a distinct vertical replacement, the replacement teeth on the pre- maxilla and on the mandibular symphysis show significant lateral (anterior) migration through the jaw bones during the replacement cycle (Fig. 45). A lateral (anterior) migration of premaxillary and symphyseal replacement teeth is also distinct in Nothosaurus (Edmund, 1960, 1969) and Cyma- tosaurus (Fig. 44B), and at least for Nothosaurus there is also evidence for a lateral migration of the lateral maxillary and dentary replacement teeth (Fig. 46). A lateral migration of the replacement teeth has also been described for Simosaurus (Ed- mund, 1960, using the figures in Huene, 1921), a claim that conflicts with Jaekel's (1905, 1907) de- scription of vertical tooth replacement in this ge- nus. Personal observations indicate lateral (ante- rior) migration of anterior replacement teeth and vertical tooth replacement in lateral teeth of Si- mosaurus. However, patterns of tooth replace- ment throughout the Sauropterygia require further study before the significance of this character can be assessed.
46
FIELDIANA: GEOLOGY
Fig. 45. Lower jaw of Placodus gigas (bsp AS VII 1 209, upper Muschelkalk, Bayreuth) in dorsal view. Note the replacement pit exposed on the left symphyseal sur- face. Scale bar = 20 mm.
There is an important difference in the mandib- ular symphysis of Placodus versus the eusauropte- rygians considered here. A substantial segment of the mandibular symphysis of Placodus is formed by the splenial (Drevermann, 1933), which re- mains excluded from the symphysis in Nothosau- rus (Fig. 46), Lariosaurus, and Cymatosaurus.
41. Splenial bone enters the mandibular sym- physis (0) or remains excluded therefrom (1).
42. Teeth set in shallow or deep sockets (0) or superficially attached to bone (1).
43. Anterior (premaxillary and dentary) teeth upright (0) or strongly procumbent (1).
Chisel-shaped anterior teeth of Placodus are strongly procumbent, as are the anterior premax- illary and dentary fangs of the Eusauropterygia (including Lariosaurus; Mazin, 1985) included in this study.
44. Premaxillary and anterior dentary fangs ab- sent (0) or present (1).
Fig. 46. Mandibular symphysis of Nothosaurus mi- rabilis (smns 59817, upper Muschelkalk, Hegnabrunn) in dorsal view. Note the replacement pits medial to the functional tooth positions. Scale bar = 20 mm.
In Lariosaurus and Nothosaurus, the (usually five) premaxillary and (usually five) anterior den- tary teeth are enlarged and fang-like. The large alveoli of the fourth and fifth tooth on the dentary are reflected in a broad posterior part of the sym- physis, set off* from the remaining lower jaw ramus by a distinct constriction (Fig. 44C). Such a con- striction at the posterior part of the symphysis is not observed in Cymatosaurus (Fig. 44B), but the premaxillary and anterior dentary teeth are dis- tinctly enlarged. The character remains unknown for Pistosaurus. In Corosaurus, only the anterior dentary teeth are enlarged, but not the anterior premaxillary teeth (Storrs, 1991).
RIEPPEL: SIMOSAURUS GAILLARDOTI
47
45. One or two caniniform teeth present (0) or absent (1) on maxilla.
46. The maxillary tooth row is restricted to a level in front of the posterior margin of the orbit (0), or it extends backward to a level below the anterior corner of the upper temporal fossa (1), or it extends backward to a level below the middle part of the upper temporal fossa (2). The character is coded as unordered.
The plesiomorphic condition is a maxillary tooth row extending posteriorly to a level somewhere in front of the posterior margin of the orbit, as seen in pachypleurosaurs and Placodus. In Simosaurus, Nothosaurus, Lariosaurus, and Corosaurus (Storrs, 199 1), the maxilla, and with it the maxillary tooth row, is extended backward. In Lariosaurus and Corosaurus (Storrs, 1991), this continues to a level below the anterior corner of the upper temporal fossa (see also Mazin, 1985; Tintori & Renesto, 1990), and in Simosaurus and Nothosaurus to a level below the middle part of the upper temporal fossa. The posterior extension of the tooth row beyond the orbit is correlated with the develop- ment of the dual jaw adductor system discussed above.
47. Teeth on pterygoid flange present (0) or ab- sent (1).
48. Vertebrae notochordal (0) or non-noto- chordal (1).
49. Vertebrae amphicoelous (0), platycoelous (1), or other (2). The character is coded as unor- dered.
Haas (1966) reported a derived ossification pat- tern in the centra of "nothosaurs" and "placo- donts," characterized by what he called "terminal plugs." These observations recall the ossification pattern in limb bones of sauropterygians, in which "terminal plugs" had been identified as epiphyses by Lydekker (1889, p. 149), an interpretation re- jected by Moodie (1908). Unfortunately, the na- ture of this ossification pattern and its taxonomic distribution among amniotes is not well enough known at this time to use the character in phy- logenetic analysis.
50. Dorsal intercentra present (0) or absent (1).
51. Cervical intercentra present (0) or absent
(1).
52. Cervical centra rounded (0) or keeled (1) ventrally.
53. Zygosphene-zygantrum articulation absent (0) or present (1).
A zygosphene-zygantrum articulation, as ob- served in Nothosaurus, Simosaurus, and Pistosaurus (Sanz, 1983), for example, consists of a bipartite
anterior process located at the base of the neural arch above the prezygapophyses (the zygosphene), which fits into a notch at the base of the neural arch of the preceding vertebra, again located above the postzyg- apophyses. Placodus shows a hyposphene-hypan- trum articulation: the hyposphene is a projection from the base of the neural spine, located below the postzygapophyses, that fits into a socket at the base of the neural arch of the succeeding vertebra, again located below the prezygapophyses (Fig. 47). On to- pological grounds, the zygosphene-zygantrum artic- ulation cannot be compared to the hyposphene-hy- pantrum articulation. Placodus is accordingly coded with the zygosphene-zygantrum articulation absent (the hyposphene-hypantrum articulation is an aut- apomorphy of Placodus among the genera included in this analysis). In pachypleurosaurs, the projection is from between the prezygapophyses, and the socket is between the postzygapophyses of the preceding vertebra. Accordingly, pachypleurosaurs are coded as sharing the presence of a zygosphene-zygantrum articulation (see also Kuhn-Schnyder, 1959; Carroll & Gaskill, 1985).
54. Sutural facets receiving the pedicels of the neural arch on the dorsal surface of the centrum in the dorsal region are narrow (0) or expanded into a cruciform or "butterfly-shaped" platform
(1).
In the presacral region, the neurocentral suture runs between the centrum and the neural arch be- low the rib articulation (diapophysis), but through the rib articulation in the sacrum and caudal re- gion. The neurocentral suture commonly fuses during ontogeny, particularly in terrestrial ani- mals, but where it remains open the pedicel of the neural arch is received on the dorsal surface of the centrum in a contact that does not exceed the transverse diameter of the centrum to a significant degree.
The neurocentral suture remains open in sau- ropterygians (as in many marine reptiles). As a consequence, the neural arch frequently dissoci- ates from the centrum in pachypleurosaurs, Si- mosaurus, and Nothosaurus in the cervical and dorsal region of the vertebral column. This dis- sociation reveals, in these taxa, a broadened facet formed by the dorsal surface of the centrum to receive the pedicels of the neural arch (Figs. 16, 48-49). This facet has lateroventrally inclined lat- eral lappets in the cervical region but becomes fully horizontal within the dorsal region, where it ex- ceeds the width of the body of the centrum as seen in ventral view. Pistosaurus shows the same apo- morphic contact of the centrum and neural arch
48
FIELDIANA: GEOLOGY
Fig. 47. Isolated dorsal vertebra of a placodont (ICyamodus, smns 59825, upper Muschelkalk, Hegnabrunn), showing the hyposphene-hypantrum articulation. A, Anterior view; B, posterior view. Abbreviations: hyp, hypantrum; hys, hyposphene; ns, neurocentral suture; poz, postzygapophysis; prz, prezygapophysis. Scale bar = 20 mm.
in dorsal vertebrae, except that in this taxon the contact area on the centrum bears distinct sockets on either side of the neural canal for the reception of peglike projections from the neural arch (Fig. 49B). The vertebral column of Cymatosaurus re- mains unknown. Outside the Eosauropterygia, the same lateral flaring of the neural arch facets on the centrum is observed, as far as is known, in choris- toderes (Sigogneau-Russell, 1981), but it is absent in Hovasaurus (an aquatic younginiform; Currie, 1981) and Palaeopleurosaurus (an aquatic rhyn- chocephalian; Carroll, 1985).
In Placodus, the neural arch separates less easily from the centrum in the cervical and dorsal ver- tebral column, although the neurocentral suture can still be identified. In cases where a dissociation of the elements has occurred during fossilization, the centrum can be seen to form narrow ridges on either side of the neural canal for the reception of the narrow pedicels of the neural arches. That the pedicels of the neural arch are received on ridges rather than in grooves on the centrum may be correlated with the derived shape of the neural canal in this taxon, which is unusually high and narrow. Whereas Placodus may not exactly rep- resent the plesiomorphic condition with respect to this character (but could be autapomorphic), it certainly lacks the specialized contact seen in pachypleurosaurs and in the eusauropterygians in- cluded in this study.
55. Transverse processes of neural arches of the
dorsal region relatively short (0) or distinctly elon- gated (1).
56. Cervical ribs without (0) or with (1) a dis- tinct free anterior process.
In pachypleurosaurs, Nothosaurus, and Pisto- saurus (Sues, 1987a), the anterior tip of the cer- vical ribs is drawn out into a free process. Such is not the case in Simosaurus, whose cervical ribs show a broad "shoulder" but no distinct process, but for the reasons discussed above, the absence of the anterior process might well be due to break- age. The character is coded as unknown for Si- mosaurus. A free anterior process is present on the cervical ribs of Placodus.
57. Dorsal ribs without (0) or with ( 1 ) pachyos- tosis.
Rib pachyostosis is a character sometimes dif- ficult to establish, because it may show gradual variation and may also be subject to ontogenetic variation. Rib pachyostosis is present in Neusti- cosaurus (coded as polymorphic for the Serpiano- saurus-Neusticosaurus clade), and it is also present in Lariosaurus.
58. The number of sacral ribs is two (0), three ( 1 ), or four or more (2). The character is coded as unordered.
The plesiomorphic character is two sacral ribs. Such has been reported to be the case in Keichou- saurus, a pachypleurosaur from China (Lin Ke- bang, confirmed by Sues, 1 987a), but in the pachy- pleurosaurs entered in this analysis (Dactylosaurus,
RIEPPEL: SIMOSAURUS GAILLARDOTI
49
Fig. 48. A, Isolated dorsal centrum of Simosaurus (smns 54763, upper Muschelkalk, Crailsheim) in dorsal view; B, isolated dorsal centrum of Nothosaurus (smns 59820, upper Muschelkalk, Bindlach near Bayreuth) in dorsal view; C, isolated dorsal centrum of Placodus (smns 59826, upper Muschelkalk, Hegnabrunn) in dorsal view. Scale bar = 20 mm.
Fig. 50. A, B, Right ilium of Nothosaurus (smns 59821, upper Muschelkalk [Trochitenkalk], ?Bindlach near Bayreuth) in lateral (A) and medial (B) views; C, D, left ilium of Placodus gigas (smf 1035, cast of Drever- mann's [1933] specimen) in lateral (C) and medial (D) views. Scale bar = 20 mm.
Serpianosaurus-Neusticosaurus clade), three sa- cral ribs is the typical number. As argued above, three functional sacral ribs were present in Simo- saurus and Nothosaurus (Schroder, 1914), and the same is true of Placodus (Drevermann, 1933), which shows the same basic morphology of the ilium (Fig. 50) as do Simosaurus and Nothosaurus. Lariosaurus is reported to have five sacral ribs (Peyer, 1933-1934; Tintori & Renesto, 1990). The number of sacral ribs cannot be established in smf R-13, and it is four with the possible "sacraliza- tion" of a lumbar rib in bsp AS I 802. Revisionary work in progress will establish the number of func-
tional sacral ribs in Lariosaurus (probably four in the adult; see Tintori & Renesto, 1990, PI. 1, and bsp AS I 802) as well as any variation, but a num- ber higher than three would be autapomorphic for Lariosaurus in the present analysis and hence is uninformative. The number of sacral ribs remains unknown for Cymatosaurus and Pistosaurus. Out- side the Eosauropterygia and the placodonts, three sacral ribs are observed in choristoderes (Sigo- gneau-Russell, 1981).
59. Sacral ribs with (0) or without (1) distinct expansion of distal head.
As described above, the sacral ribs of Simosau-
Fig. 49. A, Isolated dorsal centrum from an uniden- tified pachypleurosaur (mhi uncatalogued, lower Mu- schelkalk [lower Dolomites], Neidenfels) in dorsal view (scale bar = 5 mm); B, isolated dorsal centrum of Pis- tosaurus (mhi 1278, upper Muschelkalk, Neidenfels) in dorsal view (scale bar = 20 mm).
Fig. 5 1 . Pelvic region in the holotype of Nothosaurus "raabi" Schroder, 1914 (mb 1.007-18, lower Muschel- kalk, Riidersdorf). Scale bar = 20 mm.
50
FIELDIANA: GEOLOGY
Fig. 52. Pectoral girdle of Placodus gigas (smf 1035, cast of Drevermann's [1933] specimen). A, Left coracoid in lateral (ventral) view; B, left scapula in lateral view; C, left scapula in medial view; D, left clavicle in dorsal view; E, left clavicle in medial view. Abbreviations: cl.f, clavicular facet on medial surface of scapula; cof, coracoid foramen; gl, glenoid; icl.f, interclavicular facet on left clavicle. Scale bar = 50 mm.
rus are characterized by a distinct expansion of their distal head, and the same is true of the sacral ribs of Placodus, where the third functional sacral rib is again the most massive one. In pachypleuro- saurs, the distal expansion of sacral ribs is not pronounced, the articular head scarcely or not at all set off from the shaft of the rib. The same is true of Nothosaurus (Fig. 5 1 ) and of Lariosaurus (Peyer, 1934-1935; Tintori & Renesto, 1990). Sa- cral rib morphology remains unknown for Cy- tnatosaurus and Pistosaurus.
60. Cleithrum present (0) or absent (1).
6 1 . Clavicles broad (0) or narrow ( 1 ) medially.
62. Clavicles positioned dorsally (0) or antero- ventrally ( 1 ) to the interclavicle.
The anteroventral position of the clavicles with respect to the interclavicle was first recognized as a eosauropterygian synapomorphy by Carroll and Gaskill (1985), and it is shared by Placodus (Dre- vermann, 1933), as well as by choristoderes (Sigo- gneau-Russell, 1981).
63. Clavicles do not meet in front of the inter- clavicle (0) or meet in an interdigitating anterome- dial suture (1).
In pachypleurosaurs, Simosaurus. Nothosaurus, and Lariosaurus, the clavicles meet in an inter- digitating suture in front of the interclavicle. In Placodus (Fig. 52), the clavicles reach closely to the ventral midline with their tapering anterome- dial tip but fail to meet each other. Drevermann's ( 1 933) reconstruction of the dermal pectoral girdle shows the lack of an anteromedial suture between the clavicles, a conclusion supported by the round- ed anteromedial tips of the interclavicles. An an- teromedial suture between the two clavicles is also absent in Corosaurus (Storrs, 1991).
64. Clavicles without (0) or with ( 1 ) anterolat- eral^ expanded corners.
The clavicle of Dactylosaurus shows an antero- lateral expansion of the clavicle into a horizontal shelf of bone defining a pronounced anterolateral corner. The same character is seen in the clavicles of Simosaurus and Nothosaurus (Fig. 53). An an- terolateral expansion of the clavicle is absent in the Serpianosaurus-Neusticosaurus clade as well as in Placodus (Fig. 52). The character appears to be present in Lariosaurus but obscured by pachy- ostosis of the clavicle. The character is here coded
RIEPPEL: SIMOSAURUS GAILLARDOTI
51
Fig. 53. The clavicle in Nothosaurus sp. A, Right clavicle and scapula (bm[nh] 38669, Lettenkeuper?, Hoheneck) in dorsal view; B, right clavicle (mb R.728, lower Muschelkalk, Upper Silesia) in dorsal view; C, left clavicle (smns 59822, upper Muschelkalk, Hegnabrunn) in dorsal view. Abbreviations: cl, clavicle; cof, coracoid foramen; sc, scapula. Scale bar = 20 mm.
as unknown for Lariosaurus, pending further re- vision of the genus.
65. Clavicle applied to the anterior (lateral) (0) or to the medial (1) surface of scapula.
The reverse relation of the clavicle and scapula was first recognized as a eosauropterygian syna- pomorphy by Carroll and Gaskill (1985), and it is shared by Placodus (Drevermann, 1933). How- ever, the detailed relationship of the clavicle to the scapula differs in pachypleurosaurs, Simosau- rus, Nothosaurus, Lariosaurus, and Pistosaurus (Sues, 1987a) as compared to Placodus, in corre- lation with a different morphology of the scapula. In the eosauropterygian genera, the scapula has a broad ventral (glenoid) portion that extends into a narrow and short posterodorsal process. The clavicle is applied to the anterior aspect of the dorsal process, with an extension onto the medial surface of the scapula (Fig. 54). In Placodus, the articulation of the clavicle is restricted to the me- dial side of the anterior part of the scapula, which forms a high and narrow blade (Fig. 52C).
66. Interclavicle rhomboidal (0) or T-shaped (1).
67. Posterior process on (T-shaped) interclav- icle elongate (0), short (1), or rudimentary or ab- sent (2). The character is partially correlated with character 66 (the absence of a posterior stem pre- supposes the presence of a T-shaped interclavicle) and is coded as unordered.
An elongated posterior stem is the plesiomor- phic condition for a T-shaped interclavicle. The interclavicle of Dactylosaurus is not known in de- tail, but a much abbreviated posterior stem may be present in the Serpianosaurus-Neusticosaurus clade (Rieppel, 1989a, Fig. 8h). It is, however, completely absent in the vast majority of the spec- imens and, if present, is very rudimentary. A sim- ilarly much reduced posterior stem of the inter- clavicle may occur in the pachypleurosaur Keichousaurus (Lin Kebang, pers. comm., and cm 91.92.1). The character is coded as unknown for Dactylosaurus and polymorphic (2 and 3) for the Serpianosaurus-Neusticosaurus clade. Simosau- rus retains a distinct yet relatively short posterior stem of the interclavicle. Huene (1952, Fig. 62) figured a large interclavicle with a long posterior stem from the Grenzbonebed of the upper Mu- schelkalk separating the Muschelkalk from the Keuper, but whether or not this element is to be referred to Simosaurus remains questionable. Mention may be made at this junction of an iso- lated interclavicle (Figs. 55 A-B) of distinctly larger size than seems otherwise typical for Simosaurus, retaining the base of a distinct posterior process. Because the taxonomic status of these elements remains unclear at the present time, Simosaurus is coded for a short posterior stem of the inter- clavicle (1) as documented by the articulated skel- eton (smns 14733; Huene, 1952). Nothosaurus was described as having no posterior stem on the interclavicle by Koken (1893), and the same is true for an as yet undescribed specimen of Nothosaurus from the upper Muschelkalk (Fig. 55C). There are, however, two isolated interclavicles with pointed posterior tips, one of which (Fig. 55D) is associated with parts of the clavicles that establish a broad anteromedial contact. The other (Fig. 56) retains a facet on its anterior margin indicating an antero- medial contact of the clavicles. On this basis, these elements are referred to Nothosaurus, and the ge- nus accordingly coded for a rudimentary or absent posterior stem (2 and 3).
Placodus (Drevermann, 1933) retains a short but distinct posterior stem on the interclavicle. The interclavicle of Cymatosaurus and Pistosau- rus remains unknown.
68. Scapula broad (0) or narrow (1) above glenoid.
52
FIELDIANA: GEOLOGY
m.cl.f
a.cl.f
Fig. 54. Scapula of Nothosaurus. A, B, Right scapula (mhi 1 1 75/1, upper Muschelkalk [Discoceratitenschichten], Wittighausen) in lateral (A) and medial (B) views; C, D, left scapula (mhi 1277, upper Muschelkalk [Dorsoplanus biozone], Gottwollshausen) in lateral (C) and medial (D) views. Abbreviations: a.cl.f, anterior clavicular facet on left scapula; cof, coracoid foramen; m.cl.f., medial clavicular facet on left scapula. Scale bar = 20 mm.
69. Supraglenoid buttress present (0) or absent (1).
70. One (0) or two (1) coracoid ossifications.
7 1 . Coracoid foramen enclosed by coracoid os- sification (0) or between coracoid and scapula (1).
In the plesiomorphic condition, the coracoid fo- ramen is enclosed within the coracoid ossification (Romer, 1956; between coracoid and scapula in Proganochelys, Gaffney, 1 990). In the Eosauropte- rygia, as well as in Placodus, the coracoid foramen is represented by a notch in the coracoid, closed by the adjacent notch in the scapula. The coracoid foramen is present in Dactylosaurus (and Kei- chousaurus; pers. obs.).
72. Pectoral fenestration absent (0) or present (1).
Storrs (1991) defined the pectoral fenestration as a synapomorphy of the Sauropterygia. The pectoral fenestra is the open space located between the dermal pectoral girdle anteriorly and the coracoids posteri- orly. In all sauropterygians, the coracoids meet in the ventral midline, except for Placodus (Drever- mann, 1933). Storrs (199 1) concluded that this must be due to an autapomorphic reduction of the cora- coid plates in Placodus, and his practice to code
presence of the pectoral fenestration in the latter genus is followed in this study.
73. Limbs short and stout (0) or long and slen- der (1).
74. Humerus rather straight (0) or angulated ("curved") (1).
The "curved humerus" has long been cited as a "nothosaurian" character. The sauropterygian humerus appears "curved" in adult individuals due to a distinct angulation along its anterior mar- gin (related to the deltopectoral crest on its ventral surface) and an evenly concave posterior margin. Curvature of the humerus is one of the characters used to support Storrs's (1991, 1993a) hypothesis that the Placodontia is the sister-group to the Eusauropterygia.
An angulated humerus is characteristic of adult Dactylosaurus (sex y; Rieppel, 1993b, Fig. 8) and adult representatives of the Serpianosaurus-Neus- ticosaurus clade (sex y; Carroll & Gaskill, 1985; Rieppel, 1989a; Sander, 1989), as well as of Si- mosaurus, Nothosaurus, Lariosaurus, and Placo- dus. The humerus of Pistosaurus is weakly angu- lated (Sues, 1 987a), whereas that of Cymatosaurus (smns 58463) is not angulated (Fig. 57). A com-
RIEPPEL: SIMOSAURUS GAILLARDOTI
53
BNHMBH
Fig. 55. A, B, Isolated interclavicle, probably of Simosaurus (smns 59823, upper Muschelkalk [spinosus biozone], Hegnabrunn); C, anterior dermal pectoral girdle of Not hosaurus (smns 56618, upper Muschelkalk [upper nodosus biozone], Berlichingen) showing the absence of a posterior stem on the interclavicle; D, fragment of the dermal pectoral girdle of an unidentified sauropterygian (mb R.328, upper Muschelkalk, Bayreuth) in dorsal view. Abbreviations: cl, clavicle; icl, interclavicle. Scale bar = 20 mm.
Fig. 56. Isolated interclavicle of an unidentified sau- ropterygian, probably Placodus (smns 59824, upper Muschelkalk [robustus biozone], Bindlach near Bay- reuth). Anterior is at the bottom. Scale bar = 20 mm.
parison of humeri of Dactylosaurus, Neusticosau- rus, and Not hosaurus (Fig. 58) shows very close similarities in the general shape of the bone, which indicates that humerus angulation ("curvature") is not a character restricted to the placodonts and Eusauropterygia.
The "curved" humerus is interdependent with the general tendency, within the Sauropterygia, to have a forelimb that is of a more robust construc- tion than the hindlimb. Storrs ( 1 99 1 , 1 993a) coded this character separately, although "forelimb more robust than hindlimb" is a trait difficult to define, because of both gradual variation between taxa and ontogenetic variation within taxa. The char- acter cannot be expressed as a simple length re- lation of humerus/femur. Whereas the humerus shows, in general, strongly positive allometric growth within the Sauropterygia (with a relatively small humerus in juveniles), the genus Anarosau- rus (not included in this study) shows an excep- tionally long femur (see Rieppel, 1993b, for com- parative data), which still is much more slender than the humerus; in Lariosaurus smf R-13, the
54
FIELDIANA: GEOLOGY
enc
cap
Fig. 57. Humerus of Cymatosaurus sp. A, smns 58463, lower Muschelkalk, Winterswijk; B, Martin-Luther University Halle, uncatalogued, lower Muschelkalk (Schaumkalk), Freyburg/Unstrut. Abbreviations: cap, capitellum; dpcr, deltopectoral crest; ecc, ectepicondyle; ecg, ectepicondylar groove; enc, entepicondyle; enf, entepicondylar foramen; tr, trochlea. Scale bar = 10 mm.
femur again is distinctly longer but less robust than the humerus. The other extreme is provided by genera such as Ceresiosaurus (Peyer, 1931a) with a forelimb that, in the adult at least, is much more robust than the hindlimb. The zeugopodial ele- ments of the forelimb are longer than the same elements of the hindlimb in Simosaurus (this re- lation is variable in the Serpianosaurus-Neusti- cosaurus clade; tibia and fibula are unknown in Dactylosaurus), but such is not also the case in Nothosaurns (mb 1.007. 1 8, N. "raabi" of Schroder, 1914). If any zeugopodial elemet.t is distinctly en- larged, it is the ulna (accounted for by character 87), but in Lariosaurus (smf R-13) the tibia is longer yet not as broad as the ulna (the fibula is incompletely exposed in this specimen). If the numbers of carpal and tarsal ossifications differ in sauropterygians, the larger number is always found in the carpus, but this relation is also true of non- sauropterygians with a hindlimb much more strongly developed than the forelimb. If hyper- phalangy occurs in sauropterygians (Keichousau- rus, Lariosaurus, Ceresiosaurus), it occurs in the manus, but, among the taxa considered in this analysis, hyperphalangy of the manus is an aut- apomorphy of Lariosaurus and hence uninforma- tive. For the reasons detailed above, neither the increased robustness of the forelimbs nor the rel- ative length of the hindlimb will be coded as a
separate character in this analysis (Storrs, 1991. 1993a).
75. Humerus with prominent (0) or reduced (1) epicondyles.
The entepicondyle and ectepicondyle of the hu- merus are generally reduced in the Eosauropteryg- ia and in Placodus. Within that group, the epicon- dyles are, however, comparatively well developed in Dactylosaurus (adult sex y; Rieppel. 1 993b) and in Cymatosaurus.
76. The ectepicondylar groove is open and notched anteriorly (0), open without anterior notch ( 1 ), or closed (2) (i.e., ectepicondylar foramen pres- ent). The character is coded as unordered.
The plesiomorphic character is an ectepicondyl- ar groove that sets ofT the ectepicondyle from the humeral shaft, thereby creating a more or less dis- tinct notch at the distal end of the humerus. The plesiomorphic condition is well exemplified by an isolated humerus from the lowermost Muschel- kalk (lower Gogolin beds) of Gogolin (Poland) de- scribed by Huene (1944) (Fig. 58A). The ectepi- condylar notch is distinct in eosauropterygians with a distinct ectepicondyle (in Dactylosaurus, occa- sionally in the Serpianosaurus-Neusticosaurus clade [adult sex y], and in Cymatosaurus), but it is absent in the humeri of Simosaurus and Pis- tosaurus. An ectepicondylar notch is occasionally present in small humeri of Nothosaurus (bm[nh]
RIEPPEL: SIMOSAURUS GAILLARDOTI
55
ecg
dpcr
ecg
Fig. 58. Humeri of Sauropterygia. A, Unidentified sauropterygian (smns 16253, lower Muschelkalk, Gogolin [Poland]; original of Huene, 1944); B, Dactylosaurus schroederi (smf R-4097a, lower Muschelkalk, Kamien Gorny Slaski [Poland], drawing of cast), right humerus in ventral view; C, Dactylosaurus schroederi (smf R-4097a, lower Muschelkalk, Kamien Gorny Slaski [Poland], drawing of cast), left humerus in dorsal view; D, Nothosaurus "raabi" (holotype, MB 1.007- 1 8, lower Muschelkalk, Riidersdorf); E, Nothosaurus sp. (Martin-Luther University Halle, uncata- logued, lower Muschelkalk, Halle); F, Nothosaurus sp. (bm[nh] 40052, Muschelkalk, Niirnberg; original of Lydekker, 1889, Pt. II, Fig. 84). Abbreviations: dpcr, deltopectoral crest; ecg, ectepicondylar groove; enc, entepicondyle; enf, entepicondylar foramen; spr, supinator ridge. Scale bar = 5 mm in B and C, 20 mm in A, D, E, and F.
R-40052; Lydekker, 1889, p. 296, Fig. 84) but absent in larger specimens (Fig. 59B). It is un- known whether this reflects ontogenetic or taxo- nomic variation, or both.
Placodus presents a particular problem because the distal end of the humerus described by Drever- mann (1933) as showing an ectepicondylar fora- men is, in fact, an incomplete humerus of Notho- saurus. The humerus of Placodus lacks an ectepicondylar foramen, but it shows a distinct ectepicondylar notch (Figs. 5 9 A, 60).
77. Entepicondylar foramen present (0) or ab- sent (1).
An entepicondylar foramen is generally present in the pachypleurosaurs, the plesiomorphic con-
dition. As described above, the entepicondylar fo- ramen is absent in Simosaurus, as well as in Pis- tosaurus (Sues, 1987a), but the foramen is present in Nothosaurus and Cymatosaurus. Disregarding the nothosaur humerus attributed to Placodus by Drevermann (1933), the entepicondylar foramen is also absent in Placodus.
78. Radius shorter than ulna (0), or longer than ulna (1), or approximately of the same length (2). The character is coded as unordered.
With an ossified olecranon, the ulna is usually longer than the radius, the plesiomorphic condi- tion. In sauropterygians, the olecranon does not ossify, which reduces the relative length of the (ossified) ulna. Equal length of radius and ulna is
56
FIELDIANA: GEOLOGY
Fig. 59. A, Humerus of Placodus gigas (smns 59827, upper Muschelkalk [spinosus-nodosus biozone], Heg- nabrunn), B, humerus of Nothosaurus sp. (smns 16250b, upper Muschelkalk, Bindlach near Bayreuth).
one of the characters supporting Storrs's (1991, 1993a) conclusion that the Placodontia is the sis- ter-group of the Eusauropterygia (Rieppel, 1 993a).
The radius is longer than the ulna in Dactylo- saurus (the radius/ulna ratio is 1.14 in Nopcsa's [1928] specimen), a condition that is also found in Neusticosaurus (Carroll & Gaskill, 1985, Table 3; Sander, 1989, Table 4). In Serpianosaurus, the radius is only slightly longer than the ulna, or the two bones are equal in length. In Simosaurus, the radius and ulna are equal in length. In Nothosaurus (mb 1.007.18, N. "raabr of Schroder, 1914), the radius/ulna ratio is 1.02, that is, practically equal in length, as they also are in Pistosaurus (Sues, 1987a). Radius and ulna are incompletely pre- served in the skeleton of Placodus described by Drevermann (1933), but the two bones are very similar in length in Paraplacodus (Peyer, 1935).
In Sauropterygia with an unossified olecranon, the proximal head of the ulna is generally ex- panded, although to a variable degree. Within pachypleurosaurs, the proximal head of the ulna is not distinctly broadened except for Keichousau- rus, in which the entire ulna is greatly broadened (Young, 1958, and pers. obs.). The ulna of Kei-
enf
ecg/- f
ecg
Fig. 60. A, Distal end of the humerus of Nothosaurus sp., attributed to Placodus by Drevermann (1933; smf 1035); B, humerus of Placodus gigas (smns 59827, upper Muschelkalk [spinosus-nodosus biozone], Hegnabrunn); C, humerus of Placodus (mhi 776, upper Muschelkalk [Discoceratitenschichten] Grombach). Abbreviations: ecg, ectepi- condylar groove; enf, entepicondylar foramen. Scale bar = 20 mm.
RIEPPEL: SIMOSAURUS GAILLARDOTI
57
Fig. 6 1 . Isolated ulnae of unidentified sauropteryg- ians (referred to Nothosaurus sp. in the smns collections). A, smns 7175, upper Lettenkeuper, Hoheneck; B, smns 1892, Lettenkeuper, Hoheneck; C, smns 56686, upper Muschelkalk, Sommerhausen near Wiirzburg. Scale bar = 20 mm.
chousaurus is, in fact, very reminiscent of the ulna in Lariosaurus; however, the shape of the ulna varies quite extensively among the specimens cur- rently referred to Lariosaurus, which is the reason why Peyer (1933-1934) chose to diagnose Lario- saurus by a much expanded proximal head of the ulna (among other characters). His character def- inition is accepted here. For the specimens cur- rently referred to Lariosaurus, the ratio of the ulna length/proximal width ranges from 1 .4 to 2. 1 , with smf R-13 showing a ratio of 1.65. In Simosaurus smns 14733, the ratio is 2.89; for Nothosaurus (mb 1.007.18, N. "raabi" of Schroder, 1914), the ratio is 2.23. Based on Sues (1987a, Fig. 4A), the ratio is 2.4 in Pistosaurus. For Placodus, the ratio is unknown, but the preserved proximal part of the ulna does not appear strongly expanded either. A number of isolated ulnae have been referred to Nothosaurus (smns 7175, smns 1892, smns 5668) that document a certain variability of shape of that element (Fig. 61), but one specimen (smns 7175)
seems much broader than would be expected in Nothosaurus and shows a ratio of ulna length/ proximal width of 1.82. However, this particular element may well be referable to Ceresiosaurus (rather than Nothosaurus), a genus currently viewed as sister-taxon of Lariosaurus (Tschanz, 1989; Storrs, 1993a). With the terminal taxa used in this study, the widening of the ulna in Keichousaurus and in Lariosaurus has no resolving power, and the character is not entered into the data matrix.
79. Iliac blade well developed (0), reduced (1), or absent (2).
The plesiomorphic condition is a well-devel- oped iliac blade ascending in a posterodorsal di- rection and carrying an anterior spina praeacetab- uli (Siebenrock, 1894). The posterior extent of the iliac blade relative to the posterior margin of the ischium is variable and depends not only on the relative size of the iliac blade but also on its ori- entation (horizontal or slanting in a posterodorsal direction). Placodus and the Eusauropterygia in- cluded in this study (the ilium is incomplete, but the iliac blade appears elongate and not reduced in Pistosaurus [Sues, 1987a] and unknown in Cy- matosaurus) share a similar morphology of the ilium characterized by an iliac blade of reduced size, which retains a spina praeacetabuli. Pachy- pleurosaurs, however, have effectively lost the iliac blade; the ilium retains only a narrow process ex- tending upward from the broadened acetabular portion. There is no indication of an iliac blade in Lariosaurus, which again retains a narrow dor- sal process of the ilium only.
80. Pubis without (0) or with (1) distinctly notched (concave) ventral (medial) margin.
Simosaurus and Nothosaurus show a notch or concavity in the ventral (medial) margin of the pubis. This contrasts with the evenly convex ven- tral (medial) margin of the pubis in pachypleuro- saurs. A concavity is also absent on the ventral (medial) margin of the pubis in Placodus (due to reduction of the pubis?). The character is here treated as presently unknown for Lariosaurus. The same is true for the pubis of Cymatosaurus and Pistosaurus.
81. Thyroid fenestra absent (0) or present (1).
In the plesiomorphic condition, pubis and is- chium form a closed plate. In the derived condi- tion, a large fenestra thyroidea develops between pubis and ischium, which is present in pachypleu- rosaurs, Simosaurus, and Nothosaurus. Due to the incomplete knowledge of the ventral elements of the pelvic girdle in Pistosaurus (Sues, 1987a), the character is coded as unknown for that genus. The
58
FIELDIANA: GEOLOGY
Fig. 62. A, Femur of Nothosaurus sp. (bm[nh] 40053, Muschelkalk, Numberg); B, femur of Placodus gigas (smns uncatalogued, coll. M. Wild #1798, upper Muschelkalk, Bindlach near Bayreuth). Scale bar = 20 mm.
same is true for Cymatosaurus. A small fenestra, closed ventrally (medially) by a contact of the pu- bis with the ischium, is present in Placodus, and a convincing argument has been made that this represents a reduced thyroid fenestra (Sues, 1 987b; Storrs, 1991).
82. Acetabulum oval (0) or circular (1).
83. Femoral shaft stout and straight (0) or slen- der and sigmoidally curved (1).
In the Sauropterygia, the femur is generally slen- der and may either be straight or show some slight sigmoidal curvature. A slender and more or less straight femur is coded 1 .
84. Intertrochanteric fossa well denned (0) or reduced (1), or rudimentary or absent (2).
A well-developed trochanter separated from the shaft of the femur by a well-developed intertro- chanteric fossa is the plesiomorphic condition. A reduced intertrochanteric fossa is retained in Plac- odus (Fig. 62), where the trochanter is well set off from the shaft of the femur. In pachypleurosaurs.
Fio. 63. Proximal head of the femur. A, Simosaurus gaillardoti (smns 14733, #81 of Huene, 1952); B, Notho- saurus sp. (smns 59829, upper Muschelkalk, Bindlach near Bayreuth); C, Cyamodus (smns 59828, upper Mu- schelkalk, Hegnabrunn). Scale bar = 20 mm.
Simosaurus, Nothosaurus, and Pistosaurus (Sues, 1987a), the trochanter is represented by a more or less distinct thickening on the ventral surface of the proximal end of the femur, which results in a triangular cross-section of the bone at that level. The intertrochanteric fossa is rudimentary or ab- sent (Fig. 63). The femur of Cymatosaurus is un- known. A well-developed trochanter is present in Corosaurus, separated from the shaft by a distinct intertrochanteric fossa (Storrs, 1991).
85. Distal femoral condyles prominent (0) or not projecting markedly beyond shaft (1).
86. Anterior femoral condyle relative to pos- terior condyle larger and extending further distally (0) or smaller and of subequal extent distally (1).
87. The perforating artery passes between as- tragalus and calcaneum (0) or between the distal heads of tibia and fibula proximal to the astragalus
(1).
This character is discussed in detail by Rieppel ( 1 993c). The plesiomorphic condition (for this lev- el of analysis) is represented by a tarsus in which the astragalus and calcaneum meet in a suture that is pierced by a foramen though which passes the perforating artery. In turtles and lepidosauriforms, the perforating artery passes proximal to the as- tragalus and calcaneum (which may fuse during ontogeny) through the spatium interosseum be- tween the distal ends of the tibia and fibula. In the Sauropterygia, secondarily aquatic habits have re- sulted in skeletal reductions, the astragalus and calcaneum commonly appearing as separate rounded ossifications in the proximal tarsus. The position of the astragalus at the distal end of the spatium interosseum, together with the marked
RIEPPEL: SIMOSAURUS GAILLARDOTI
59
Fig. 64. Tarsus of Not hosaurus "raabi" Schroder, 1914. A, Holotype (mb 1.007-18, lower Muschelkalk, Riiders- dorf); B, smf R-4546, lower Muschelkalk, Oberdorla, Thuringen.
|
Eu |
"N |
|||||||
|
Da |
Se-Ne |
f*, |
pi |
|||||
|
N^P||IF' |
||||||||
|
V |
J |
L |
l |
J |
||||
Ar Yo CI
Ca
B
Fig. 65. A, Strict consensus tree of nine equally parsimonious unrooted networks (TL = 85, CI = 0.694, RC = 0.421) for ingroup taxa only; B, strict consensus tree of six equally parsimonious unrooted networks (TL = 167, CI = 0.707, RC = 0.535) for ingroup taxa plus the four taxa used as outgroups by Storrs (1991, 1993a). Abbreviations: Ar, Araeoscelidia; Ca, Captorhinidae; CI, Claudiosaurus; Co, Corosaurus; Da, Dactylosaurus; Eu, Eusauropterygia; PI, Placodus; Se-Ne, Serpianosaurns-Neusticosaurus clade. The Nothosauriformes are circled with a dashed line. For further discussion, see text.
60
FIELDIANA: GEOLOGY
Table 8. Data matrix for the taxa included in the cladistic analysis. For discussions on changing character codings, see text.
|
1 |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
1 0 |
|
|
1 |
Ancestor |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
2 |
Captorhinidae |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
3 |
Testudines |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0&1 |
0&2 |
0 |
|
4 |
Araeoscelidia |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
|
|
5 |
Younginiformes |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
1 |
|
|
6 |
Kuehneosauridae |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
|
|
7 |
Rhynchocephalia |
0 |
0 |
0 |
0 |
0 |
0 |
1&2 |
0 |
0 |
|
|
8 |
Squamata |
0 |
0 |
0 |
0 |
0 |
0 |
1&2 |
0&1 |
0&1&2 |
|
|
9 |
Rhynchosauria |
1 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
||
|
1 0 |
Prolacertiformes |
1 |
0 |
1 |
0 |
0 |
1 |
0 |
1 |
||
|
1 1 |
Trilophosaurus |
2 |
0 |
0 |
0 |
0 |
9 |
1 |
1 |
||
|
1 2 |
Choristodera |
1 |
0 |
1 |
0 |
0 |
1 |
0 |
2 |
||
|
1 3 |
Archosauriformes |
1 |
0 |
1 |
0 |
0 |
1 |
0 |
0&1 |
||
|
1 4 |
Claudiosaurus |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
1 |
|
|
1 5 |
Dactylosaurus |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
1 |
2 |
|
|
1 6 |
Serpiano-Neustico |
0 |
0 |
0 |
0 |
0&1 |
2 |
0 |
1 |
2 |
|
|
1 7 |
Simosaurus |
0 |
0 |
0 |
0 |
1 |
2 |
0 |
2 |
||
|
1 8 |
Ncihosaurus |
0 |
1 |
0 |
0 |
0&1 |
2 |
0 |
2 |
||
|
1 9 |
Lariosaurus |
0 |
1 |
0 |
0 |
9 |
2 |
0 |
2 |
||
|
2 0 |
Corosaurus |
0 |
0 |
0 |
0 |
0 |
2 |
0 |
2 |
||
|
2 1 |
Cymatosaurus |
0 |
1 |
0 |
1 |
1 |
2 |
1 |
2 |
||
|
2 2 |
Pistosaurus |
0 |
0 |
0 |
1 |
1 |
2 |
0 |
1 |
||
|
2 3 |
Placodus |
0 |
1 |
1 |
0 |
0 |
2 |
1 |
2 |
||
|
2 4 |
Pareiasauria |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
|
2 5 |
Procolophonidae |
0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
0 |
0 |
|
2 6 |
Palaeagama |
0 |
9 |
0 |
0 |
0 |
0 |
9 |
0 |
1 |
|
|
2 7 |
Paliguana |
0 |
9 |
0 |
0 |
0 |
0 |
9 |
0 |
1 |
|
|
2 8 |
Saurosternon |
9 |
9 |
9 |
9 |
? |
9 |
9 |
9 |
9 |
9 |
|
29 |
Coelurosauravus |
1 |
0 |
0 |
1 |
0 |
0 |
9 |
9 |
0 |
1 |
concavity on its proximal margin (Fig. 64; see also Romer, 1956, Fig. 189D, for the tarsus of Lario- saurus), is here taken as evidence that the perfo- rating artery passed proximal to the astragalus be- tween the distal heads of tibia and fibula (on the same basis, a proximal course of the perforating artery is assumed for rhynchosaurs; Chatterjee, 1974).
88. Calcaneal tuber absent (0) or present (1).
89. Foot short and broad (0) or long and slender
(1).
90. Distal tarsal 1 present (0) or absent (1).
91. Distal tarsal 5 present (0) or absent (1).
92. Total number of tarsal ossifications four or more (0), three (1), or two (2). The character is coded as unordered.
As far as is known at this time, the maximum number of tarsal ossifications recorded in pachy- pleurosaurs and the Eusauropterygia is four (La- riosaurus smf R-13, with distal tarsals 4 and 3 ossified in the left pes; whether Simosaurus also retained four tarsal ossifications is questionable). Three tarsal ossifications (astragalus, calcaneum, and distal tarsal 4) are present in Nothosaurus (mb 1.997.18; N. "raabi" of Schroder, 1914), and Simosaurus may have shared the same number. The tarsal ossifications of Dactylosaurus are un- known. In the Serpianosaurus-Neusticosaurus clade, the number of tarsal ossifications drops to two (astragalus and calcaneum), and two tarsal ossifications are thought to be present in Placodus (Drevermann, 1933; the same number of tarsal
RIEPPEL: SIMOSAURUS GAILLARDOTI
61
Table 8. Data matrix for the taxa included in the cladistic analysis (continued).
|
2 |
1 1 |
1 2 |
1 3 |
1 4 |
1 5 |
1 6 |
1 7 |
1 8 |
1 9 |
20 |
|
|
1 |
Ancestor |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
2 |
Captorhinidae |
0 |
0 |
? |
0 |
0&2 |
0 |
0 |
1 |
0 |
0 |
|
3 |
Testudines |
0 |
0 |
? |
0 |
3 |
0 |
1 |
1 |
0&1 |
0 |
|
4 |
Araeoscelidia |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
5 |
Younginiformes |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
0&1 |
0 |
0 |
|
6 |
Kuehneosauridae |
0 |
1 |
0 |
0 |
2 |
0 |
1 |
1 |
0 |
|
|
7 |
Rhynchocephalia |
0&1 |
0&1 |
0 |
0&1 |
0&2 |
0&1 |
1 |
0&1 |
0&1 |
|
|
8 |
Squamata |
0&1 |
0 |
0 |
0&1 |
0&2&3 |
0&1 |
1 |
0&1 |
0&1 |
|
|
9 |
Rhynchosauria |
0 |
0 |
0 |
1 |
3 |
2 |
1 |
1 |
0 |
|
|
1 0 |
Prolacertiformes |
0 |
0&1 |
0 |
0&1 |
2&3 |
0 |
1 |
0&1 |
0 |
|
|
1 1 |
Trilophosaurus |
0 |
0 |
1 |
0 |
3 |
2 |
1 |
? |
0 |
|
|
1 2 |
Choristodera |
0 |
0 |
0 |
0 |
3 |
1 |
1 |
1 |
0 |
|
|
1 3 |
Archosauriformes |
0&1 |
0&1 |
0&1 |
0&1 |
0&3 |
0&1 |
1 |
0&1 |
0 |
|
|
1 4 |
Claudiosaurus |
0 |
0 |
0 |
0 |
0 |
1 |
1 |
0 |
||
|
1 5 |
Dactylosaurus |
0 |
0 |
0 |
0 |
0 |
1 |
1 |
? |
||
|
1 6 |
Serpiano-Neustico |
0 |
0 |
0 |
0 |
0 |
1 |
1 |
0 |
||
|
1 7 |
Simosaurus |
1 |
0 |
0 |
1 |
1 |
1 |
1 |
1 |
0 |
|
|
1 8 |
Nothosaurus |
1 |
0 |
1 |
1 |
1&2 |
1 |
1 |
1&2 |
||
|
1 9 |
Lariosaurus |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
? |
||
|
2 0 |
Corosaurus |
0 |
0 |
0 |
0 |
1 |
1 |
1 |
2 |
||
|
2 1 |
Cymatosaurus |
0 |
0&1 |
1 |
0 |
0&1 |
1 |
1 |
1 |
||
|
2 2 |
Pistosaurus |
1 |
1 |
1 |
2 |
2 |
1 |
1 |
0 |
||
|
2 3 |
Placodus |
0 |
0 |
0 |
2 |
0 |
1 |
1 |
0 |
||
|
2 4 |
Pareiasauria |
0 |
0 |
? |
1 |
2 |
0 |
0 |
1 |
0 |
0 |
|
2 5 |
Procolophonidae |
0 |
1 |
? |
0 |
2 |
0 |
1 |
0 |
0 |
|
|
2 6 |
Palaeagama |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
|
|
2 7 |
Paliguana |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
? |
0 |
|
2 8 |
Saurosternon |
? |
? |
? |
? |
? |
? |
? |
9 |
? |
? |
|
2 9 |
Coelurosauravus |
0 |
0 |
0 |
0 |
? |
0 |
1 |
1 |
0 |
1 |
ossifications has been reported for Paraplacodus, Kuhn-Schnyder, 1942).
93. Metatarsal 5 long and slender (0) or dis- tinctly shorter than the other metatarsals and with a broad base (1).
94. Metatarsal 5 straight (0) or "hooked" (1). The data discussed above (characters 1 through
94) as coded in the data matrix (Table 8) were an- alyzed using the software package PAUP version 3.1.1, developed by David L. Swofford (Swofford, 1990; Swofford & Begle, 1993). The principal goal of the present analysis is the test of two alternative hypotheses of sauropterygian interrelationships, that is, the monophyly of the Nothosauriformes (Storrs, 1991, 1993a; nesting placodonts within the Saurop- terygia) versus the monophyly of the "Euryapsida"
(Rieppel, 1989a; Zanon, 1989; grouping placodonts as sister-taxon of the Eosauropterygia). The test will be based on local (ingroup without or with an all- zero ancestor) and global (ingroup plus multiple out- groups) parsimony.
In a first step, an unrooted network was con- structed for the ingroup only, which is here con- sidered to comprise the taxa Corosaurus, Cyma- tosaurus, Dactylosaurus, Lariosaurus, Nothosaurus, Pistosaurus, Serpianosaurus-Neusticosaurus, Si- mosaurus, and Placodus. Deleting all other taxa from the analysis causes a number of characters to become constant and/or uninformative; these were ignored in the analysis (characters 1-2, 4, 7, 12, 17-19, 21-23, 25, 28, 30-34, 38, 41^*2, 47- 48, 50-56, 58, 60, 62, 65, 68-74, 81-83, 85-91,
62
FIELDIANA: GEOLOGY
Table 8. Data matrix for the taxa included in the cladistic analysis (continued).
|
3 |
2 1 |
2 2 |
23 |
24 |
25 |
2 6 |
27 |
28 |
29 |
30 |
|
|
1 |
Ancestor |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
2 |
Captorhinidae |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
? |
0 |
0 |
|
3 |
Testudines |
0 |
0 |
1 |
0 |
0 |
0 |
0&2 |
0 |
1 |
1 |
|
4 |
Araeoscelidia |
0 |
0&1 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
|
5 |
Younginiformes |
0 |
1 |
0 |
0 |
0 |
0 |
0 |
1 |
||
|
6 |
Kuehneosauridae |
0 |
2 |
1 |
1 |
1 |
0 |
0 |
1 |
||
|
7 |
Rhynchocephalia |
1 |
1&2 |
0 |
0 |
1 |
0&1 |
0 |
1 |
||
|
8 |
Squamata |
0 |
2 |
1 |
1 |
? |
0 |
0&1&2 |
0&1 |
||
|
9 |
Rhynchosauria |
1 |
1 |
1 |
0 |
0 |
1 |
0 |
0 |
||
|
1 0 |
Prolacertiformes |
0&1 |
2 |
1 |
0&1 |
1 |
1 |
0 |
1 |
||
|
1 1 |
Trilophosaurus |
? |
0 |
? |
0 |
? |
1 |
2 |
? |
||
|
1 2 |
Choristodera |
0 |
1 |
1 |
0&1 |
0 |
1 |
1 |
1 |
||
|
1 3 |
Archosauriformes |
1 |
1 |
1 |
0 |
0 |
0&1 |
0&1&2 |
1 |
||
|
1 4 |
Claudiosaurus |
0 |
2 |
0 |
0 |
1 |
0 |
0 |
? |
0 |
0 |
|
1 5 |
Dactylosaurus |
0 |
2 |
0 |
0 |
1 |
2 |
0 |
? |
||
|
1 6 |
Serpiano-Neustico |
0 |
2 |
0 |
0 |
1 |
2 |
0 |
? |
||
|
1 7 |
Simosaurus |
0 |
2 |
0 |
0 |
1 |
2 |
1 |
1 |
0 |
|
|
1 8 |
Nothosaurus |
0 |
2 |
0 |
1 |
? |
2 |
0&1 |
1 |
0 |
|
|
1 9 |
Lariosaurus |
0 |
2 |
0 |
? |
? |
2 |
0 |
? |
0 |
|
|
20 |
Corosaurus |
0 |
2 |
0 |
? |
? |
1 |
1 |
1 |
0 |
|
|
2 1 |
Cymatosaurus |
0 |
2 |
0 |
1 |
? |
? |
1 |
? |
0 |
|
|
2 2 |
Pistosaurus |
0 |
2 |
0 |
1 |
? |
? |
1 |
? |
0 |
|
|
2 3 |
Placodus |
0 |
2 |
0 |
0 |
0 |
1 |
0 |
1 |
1 |
|
|
24 |
Pareiasauria |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0 |
0 |
0 |
|
25 |
Procolophonidae |
0 |
0 |
1 |
0 |
0 |
0 |
2 |
0&1 |
1 |
0 |
|
2 6 |
Palaeagama |
0 |
2 |
1 |
1 |
? |
0 |
? |
? |
1 |
|
|
2 7 |
Paliguana |
0 |
2 |
1 |
? |
? |
0 |
? |
? |
1 |
|
|
28 |
Saurosternon |
? |
? |
? |
? |
? |
? |
? |
? |
? |
? |
|
29 |
Coelurosauravus |
0 |
2 |
0 |
0 |
1 |
? |
9 |
? |
1 |
1 |
and 93-94). With the branch-and-bound algo- rithm implemented (addition sequence: stepwise), nine equally most parsimonious reconstructions (MPRs) were obtained, a network with a tree length (TL) of 85 steps, a consistency index (CI) of 0.694, and a rescaled consistency index (RC) of 0.42 1 . Lack of resolution is restricted to the Eusauropte- rygian taxa. The strict consensus tree of those nine MPRs (Fig. 6 5 A) is equivocal with respect to the potential grouping of Placodus versus pachypleu- rosaur taxa within a monophyletic group com- prising the Eusauropterygia.
Rooting the ingroup network by implementa- tion of an all-zero ancestor polarizes the characters as discussed in the character definitions given above and specifies sister-group relationships contra-
dicting the concept of the Nothosauriformes (Storrs, 1991, 1993a, placing the Placondontia as sister-group of the Eusauropterygia). Instead, Placodus falls out as sister-taxon of all other Sau- ropterygia (Fig. 66), here referred to as Eosaurop- terygia. Using the ingroup taxa only as before, but rooting the tree by implementation of the as- sumption of an all-zero ancestor, again renders numerous characters constant and/or uninforma- tive; these characters were ignored in the analysis: 1-2, 4, 7, 17-19, 21-23, 28, 30-34, 38, 42, 47, 50-52, 55-56, 58, 60, 62, 65, 69-73, 81-83, 85- 91, and 93-94. With the branch-and-bound al- gorithm implemented (addition sequence: step- wise), six equally MPRs were obtained, with a TL of 1 10 steps, a CI of 0.673, and an RC of 0.391.
RIEPPEL: SIMOSAURUS GAILLARDOTI
63
Table 8. Data matrix for the taxa included in the cladistic analysis (continued).
|
4 |
3 1 |
32 |
3 3 |
34 |
35 |
3 6 |
3 7 |
3 8 |
3 9 |
4 0 |
|
|
1 |
Ancestor |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
2 |
Captorhinidae |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
0 |
0 |
|
3 |
Testudines |
0 |
0&1 |
0 |
0 |
0&1 |
1 |
0&1 |
0 |
0 |
|
|
4 |
Araeoscelidia |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
|
|
5 |
Younginiformes |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
||
|
6 |
Kuehneosauridae |
1 |
0 |
0 |
0 |
? |
? |
0 |
0 |
||
|
7 |
Rhynchocephalia |
0&1 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
||
|
8 |
Squamata |
1 |
0 |
0 |
0 |
0&1 |
0 |
1 |
0 |
||
|
9 |
Rhynchosauria |
0 |
0 |
0 |
0 |
1 |
0 |
1 |
0 |
||
|
1 0 |
Prolacertiformes |
0 |
0 |
0 |
0 |
0 |
0 |
0&1 |
0 |
||
|
1 1 |
Trilophosaurus |
0 |
0 |
0 |
0 |
0 |
0 |
1 |
0 |
||
|
1 2 |
Choristodera |
0 |
? |
0 |
0 |
1 |
0 |
0 |
0 |
||
|
1 3 |
Archosauriformes |
0 |
0&1 |
0 |
0 |
0&1 |
0&1 |
0&1 |
0 |
||
|
1 4 |
Claudiosaurus |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
0 |
||
|
1 5 |
Dactylosaurus |
0 |
? |
0 |
0 |
1 |
0 |
0 |
|||
|
1 6 |
Serpiano-Neustico |
0 |
? |
0 |
0 |
1 |
0 |
0 |
|||
|
1 7 |
Simosaurus |
0 |
2 |
0 |
0 |
0 |
0 |
0 |
|||
|
1 8 |
Nothosaurus |
0 |