Marshall University Marshall Digital Scholar
Biological Sciences Faculty Research Biological Sciences
7-2003 Preliminary Report on the Osteology and Relationships of a New Aberrant Cryptocleidoid Plesiosaur from the Sundance Formation, Wyoming F. Robin O’Keefe Marshall University, [email protected]
William Wahl Jr.
Follow this and additional works at: http://mds.marshall.edu/bio_sciences_faculty Part of the Animal Sciences Commons, and the Ecology and Evolutionary Biology Commons
Recommended Citation O’Keefe, F. R. & Wahl, W. (2003). Preliminary report on the osteology and relationships of a new aberrant cryptocleidoid plesiosaur from the Sundance Formation, Wyoming. Paludicola 4, 48–68.
This Article is brought to you for free and open access by the Biological Sciences at Marshall Digital Scholar. It has been accepted for inclusion in Biological Sciences Faculty Research by an authorized administrator of Marshall Digital Scholar. For more information, please contact [email protected], [email protected]. Paludicola4(2):4&.68 July 2003 C bythe Roche$terInstitute of VertebratePaleontology
PRELIMINARY REPORT ON THE OSTEOLOGY AND RELATIONSHIPS OF A NEW ABERRANT CRYPTOCLEIDOID PLESIOSAUR FROM THE SUNDANCE FORMATION, WYOMING
F. Robin O'Keefe1 and William Wahl, JR.l
1-Department of Anatomy, New York College of Osteopathic Medicine, Old Westbury, New York 11568
ABSTRACT The cryptocleidoid plesiosaur Taktwcfltslaram.eruiS, new genus, is described from the RedwaterShale:Memberof the Sundancefonnation. County, Narrona Wyommg.. Thebolotypcof thisspecies was a thatpllrtial skeleton since bas been lost. A neotypc:is designatedthat preserves $lime the elementspresent in the bolotype. A seooodspecimen is referred tothe taxon, IOd tlus specimen includes amialmaterial.TheprtSen�edcnarial elementsan: the left squm10581, a partial right ftonal. several isolated teeth, tbe parasphenoid,IDdlargeJ)OftlOnSof the left lind right ptcrygoids. Tbe skull sbaramany traits Wlththator Kllfurtero!/ltlUI'V.J, a cryptocleidoid plesiosaurfrom the KimmcridJC Clay of England. However, the palale is derived, and resembles those ofthe pooriy-undustood cimolia.uurid pie!>i()SIIlm of the Cretacc:Qus of the southern bcmisphen:. This similerity is established via cornperiJOnthe with skull of an undescribed taxoo fromlate Jurassic of Cuba. Thecryptocleidoid plesiosaurs underwent an extensive radiation in the Late Jurassic, and ll"IOfe research attention is needed, beginning with additionalprcpantion and collection of Ta�twctu
INTRODUCTION 'Piesiosaurus shirleyensis' (now lost) did preserve teeth and a fragment of mandible possibly from this taxon Upper Jurassic plesiosaur materiaJ has been (Knight 1900). known from the (Oxfordian) Redwater Shale member This paper offers a preliminary description of of the Sundance Formation of Wyoming since the end a second cryptocleidoid taxon from the Redwater of the 19th century, being first mentioned by Marsh in Shale. Thecase will be madethat this taxonis the same 1891, and later elaborated by Marsh (1893, 1895) and as the taxon 'Tricleidus?' laramiensis erected by Mehl by Knight (1898, 1900). Mehl (1912) advanced the (1912), itself a revision of the taxon 'Cimoliosaurus' hypotheses that two small plesiosaur taxa occurred in /aramiensis Knight 1900. However, the taxonomy of the Redwater SbaJe, and that both showed affinities to this species is quite complex, and the difficulties are the cryptocleidoids (sensu O'Keefe 200I) of the compounded by the fact that Knight's holotype cannot (Callovian) Oxford Clay of England. The Sundance be located today. The tack taken here is to erect a Formation plesiosaurs received no further attention neotype of this species based on Knight's original until the 1990s,when field crews of the Tate Museum (valid) description, assign a new genus name as the in Casper, Wyoming began collecting new plesiosaur species is currently without a valid one, and then refer material from Natrona county. The taxonomic history other material to the taxon. The taxonomic issues of Sundance plesiosaurs, and the status of Pantosaurus involvedare discussedat length below. striatus, are reviewed in the preceeding paper by O'Keefe and Wahl (2003). Pantosaurus seems to have Cryptocleidoid Phylogeny and Relevance-A� had the longer neck of the two taxa present in the will be shown below, the cranial anatomy of this Redwater Shale. This taxon possesses at least 35 second RedwaterShale taxon is quite derived and very cervical vertebrae that are long antero-posteriorly, and important, because it sheds light on the anatomy and that are very simHar to those of the long-necked taxon relationships of the group of animals defined as the Muraenosaurus. No cranial material is currently known Cimoliasauridae by O'Keefe (200I). This group of from Pantosaurus, although Knight's specimen of bizarre animals is best known from the Cretaceous taxa
48 O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 49
Kiliwhekea and Aristonectes, but also includes the Etymoloc-Tate, in honor of Marion and Inez Jurassic taxon KimmerosauniS (O'Keefe 200I). A Tate, founders of the Tate Museum in Casper, cladistic analysis perfonned below shows that the Wyoming in 1980. -Nectes,Greek, meaning diver. second Redwater Sbale taxon is yet another member of this group. Sorting out the relationships of the Tatenectes laramiensisKnight 1900, new combination Cimoliasawidae has assumed new importance with the (Figures 1,2,3,4,5). recent publication of a paper contending that Aristonectes (and material referred to it) is an Holotype-W. C. Knight, uncatalogued. elasmosaur (Gasparini et al. 2003). It is our view that Disarticulated axial skeleton and nearly complete this assignment is incorrect; this view is supported by forelimb. This specimen is lost, but was figured and much character evidence in the cladistic analysis, and described by Knight ( 1900) in adequate detail to will be discussed below. Some necessary background validate the name. on plesiosaurphylogeny is therefore introduced here. , - Neotype-UW 15943 & UW 24801, a partial The phylogeny of the Plesiosauria. and of clade skeleton comprising axial skeleton, ribs, pectoral Cryptocleidoidea in particular, has undergone girdle, and forelimb elements. extensive taxonomic revision of late. Carpenter's Referred Material:UW 24215 ( 1997) assertion that the Pliosauridae as traditionally Ottarreau-Redwater Shale member of the defined was polyphyletic motivated a cladistic analysis Sundance Fonnatioo, Late Jurassic (Oxfordian); by O'Keefe (2001; see also O'Keefe in press a), who Natronaand Carbon Counties, Wyoming. found thatthe Pliosawidae contained members of three Diagnosis-A small plesiosaur with an unknown clades: the primarily Jurassic rhomaleosaurs and true number of cervical vertebrae, but probably less than pliosaurs, and the Cretaceous polycotylids. O'Keefe 31. Cervical vertebrae are much shorter than wide, are found strong support for a clade containing the not waisted, and do not have elongate articulations for Polycotylidae and the traditionally defined cervical ribs. The foramina subcentraJia are widely Cryptoclididae (including the Jurassic taxa spaced, and the rims of articular faces are poorly Cryptoclidus, Tricleidus, and Kimmerostnll"U3, as well ossified. The humerus possesses radial and ulnar as other, more derived Cretaceous fonns). O'Keefe also articulations that are subequal in length, articulations found that Muraenosaurs, often considered an for two supernumerary ossifications in the epipodial elasmosaur, was also a member of this clade. O'Keefe row, and a long. slender shaft. Thescapula possesses a (200I) therefore redefined Williston's (1925) medial process extending toward the midline but not Cryptocleidoidea to include Muraenosaurus and the contacting its neighbor, and certainly lacking a long Polycotylidae as well as other 'cryptoclidid'taxa. Given midline suture. Suture between scapula and coracoid in these relationships, the Upper Jurassic cryptocleidoid center of glenoid. Teeth narrow and recurved withlong taxa assume greater importance, because they are near roots, and striated all around. Anterior interpterygoid the base of a radiation giving rise to both vacuity present; the pterygoids behind this vacuity are plesiosauromorph and pliosauromorph taxa (O'Keefe, developed into a deep block of bone giving a distinct 2002). The bizarre and poorly-understood members of shelf to the basicranium. Pterygoid processes extend the Cimoliasauridae constitute a third lineage diverging caudally in a U-shape to effect articulation with the during this radiation. Cryptocleidoid material from basioccipital tubers. Wyoming bas the potential to shed light on the morphology and relationships of the clade at an early DESCRIPTION period in its history. This material is also biogeographically important, because Upper Jurassic Tatenectes laramiensis is represented by the cryptocleidoids are known almost entirely from following material: the neotype UW 15943 and UW Englandat present 24801, an articulated skeleton consisting of pectoral girdle, distal h�eriis, and partial axial skeleton SYSTEMATIC PALEONTOLOGY comprising nbs and dorsa] and two posterior cervical vertebrae; and UW 24215, a fragmentary skull and Suborder Plesiosauriade Blainville, 1835 articulated vertebral column. UW 15943 and UW Genus Tatenectes new genus 2480 I were collected from a Redwater Shale outcrop near Roughlock Hill in Natrona County, Wyoming Type Species--Tatenectes laramiensis, by (UW locality V-95010). UW 24215 was found in a monotypy. Redwater Shale outcrop (UW locality V -92066) near Diagnosis-as for species. the town of Arminto in Natrona County. The neotype specimen (UW15943 & UW 2480 l) was designatedas such because it contains the sameelements figured and 50 PALUDlCOLA. VOL. 4, NO. 2, 2003
descnl>ed by Knight (1900) in the lost holotype a consequence of ontogeny; Andrews (1910) skeleton (see Discussion). The features of the neotype demonstrated that the medial process is the last part of skeleton will be described first, followed by a the scapula to ossify in Cryptoclidus. Again. however, discussion of the craniaJ and cervical anatomy of the the humerus is well-ossified and the neural arches are referred specimen (UW 242 15). fused, indicating that the animal was not a juvenile. We The neotype specimen of Tatenectes laramiensis therefore accept the configuration of the medial process (UW 15943 & UW 24801 ) is illustrated here in Figure of the scapula as an adult feature. l. The humerus was given its own number because it The axial skeleton of Tatenectes is comprised was found as float just beneath the quarry that yielded almost entirely of dorsal vertebrae. The vertebrae are the concretion containing the axial skeleton and most clearly plesiosaurian in having deep, oval-shaped rib of the pectoral girdle. We believe it is likely that the articulations on the transverse processes, neural arches humerus fragment weathered out of the concretion more narrow than the centra, and possessing high, because the concretion contains a thoracic axial colwliit blade-like neural spines (O'Keefe 200l and references and associated pectoral girdle, and pieces of the therein); however, they lack differentiating characters coracoid (later reassembled) were found with the from within the Plesiosauria. Fortunately there are two humerus fragment beneath the quarry. The style of posterior cervical vertebrae preserved with the preservation is identical. specimen (Figure lb). The cervicals are not prepared The pectOral girdle of UW 15943 consists of the and are exposed on the surface of a weathered block articulated left scapula and coracoid (Figure 1). The from the concretion, but several characters are scapula is mostly complete, missing only a portion of apparent. First is the presence of two widely-spaced the dorsal process just antero-dorsal to the glenoid. The foramina subcentralia on each centra, proving these coracoid is complete anteriorly, but fragments vertebrae are plesiosaur cervicals (O'Keefe 200I). posteriorly where the element was weathering out of These foramina are more widely spaced than they are the limestone concretion. The glenoid fossa is well in PanJosaurus. Additionally, the centra are developed but rather large, its anterior edge is broken compressed antero-posteriorly, and while measurement away, and the scapula-coracoid suture is near its center was not possible, the vertebrae are clearly more as in most plesiosaurs (but unlike Tricleidus or compressed than those of Pantosaurus. The cervical rib Cryptoclidus, Brown 1981). The coracoid also broadly heads are rounded, not oblong, and are not carried on a resembles those in other crytpocleidoids, although its pedestal asthey are in Pantosaurus. Lastly, the rim of edges are fragmented and detailed comparison the articular facet of each centrum is not well ossified, impossible. There does not seem to have been a and the rim is poorly defmed, similar to Tric/eidus, pectoral bar formed by anterior extensions of the Cryptoclidus, and K1mmerosaurus, but unlike coracoids as in other cryptocleidoid taxa (Brown 1981 Muraenosaurus or Pantosaurus. The present material p. 331 ). The presence of this feature varies also lacks the fine striations of the ventral surfaces of ontogenetically (Brown 1981), but the advanced state the bottoms of the centra observed in the later two taxa of ossification of the humerus would seem to indicate It is impossible to know bow long the neck was in this that this specimen was an adult. The neural arches of taxon without more material; however, the amount of the dorsal and cervical vertebrae are also fused to the antero-posterior compression probably indicates that centra, another indicator of adult status (Brown 1981). the number of cervicals was probably less than 31, as The dorsal process of the scapula is unusual; the these two measurements are often correlated (O'Keefe portion that is preserved seems to suggest that the 2002). process projected anteriorly rather than postero The distal humerus fragment from the neotype of dorsally as in most plesiosaurs. Given the lack of Tatenectes is illustrated in Figure I c. The identification preservation and the possibility of post-depositional of this bone as a humerus seems secure given the fact deformation, however, caution should be exercised in that it carries a[ticulations for two supernumerary taking this morphology at face value. In contrast, the ossifications in the epipodial row; no known plesiosaur medial process of the scapula is well preserved and femur carrieS. two supernumerary articulations, and very unusual. The process is well-developed but does when this feature occurs it is always on the humerus. not extend to the midline, and is therefore intennediate The humerus of Tatenectes differs significantly from between early taxa (e.g. Plesiosaurus) that Jack a Pantosaurus in that the radial articulation is relatively suture of the scapulae on the midline, and later taxa shorter; this articulation is about as long as that for the (e.g. Cryptocluius, Tric/eidus, all elasmosaurs) where ulna, and is more similar to other Oxford Clay the scapulae meet in a long midline suture (O'Keefe cryptocleidoids (Andrews 1910) than to PanJosaurus 2001 and references therein). The medial process of the (O'Keefe and Wahl 2003). The radius itself is Tatenectes scapula is unlike that in any plesiosaur yet correspondingly small, and while it is larger than the known. The condition displayed by Tatenectes may be ulna it again resembles Oxford Clay cryptocleidoids in O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 51
8
A p
c E (,) L{)
FIGURE I. Elements of the neotype ofTa�necte& laramiemi& newgenus. UW 15943 &. UW 24801 Apectoral girdle. B. cervical .00 pectoral vertebrae,anterior the to left. C: distalof end humerus Ahhreviations arc: c, coracoid;glenoid; g. h. humerus; ra, radiusarticulation; scapula, s, scs, scapulalooracoid sutw-e:supernumerary Sta. oss ificationone ll'liculaoon; s,a, supernumerary ossificallootwo articulation; ua, ulna artJculation. 52 PALUDICOLA, VOL. 4, NO.2, 2003
relative size. The radius carries a depression in its exact nature of this articulation is not discernible due to anterior face, at least in the Knight specimen breakage.The ventral process is long and thin, and (illustrated in Mehl 1912); this feature is present in carries a long, shallow depression or socket for Muraenosaurus and Cryptoclidus but absent in articulation with the quadrate. The ventral process Pantosaurus and Tricleidus. The humerus also would have covered the quadrate almost entirely in possesses two articulations for supernumerary lateral view, a diagnostic character possessed by all ossifications rather than one, resembling Tricleidus or cryptodeidoids including the polycotylids (O'Keefe Colymbosaurus rather than Muraenosaurus, 2001). Cryptoclidus, or Pantosaurus. The forelimb of The pterygoids and basicranium are the most Tatenectes is very similar to that of Tric/eidus in most diagnostic-and most unusual-elements of respects, as acknowledged my Mehl; it is perhaps most Tatenectes, and are illustrated in Figure 2. The similar to Colymbosaurus given that that humeral shaft preserved portions of the left and right pterygoids is long and slender relative to Tricleidus. It certain'ly display the posterior margin of the anterior varies in many important respects from the forelimb of interpterygoid vacuity medially, but are fragmented Pantosaurus. anteriorly and laterally, resulting in a lack if Cranial Anatomy-Discussion of the cranial information on articulation with maxilla or anatomy of Tatenectes /aramiensis is derived from the ectopterygoid. The pterygoids meet in a median suture referred specimen (UW 24215), a partially articulated behind the anterior pterygoid vacuity, and the area of skeleton comprising a partial skull, an articulated series union is expanded dorso-ventrally into a deep block of of cervical vertebrae, and possibly some dorsal bone on the midline. This block fills the area of the vertebrae, although the exact contents of the posterior intcrpterygoid vacuities present in most other unprepared portions of the jacket are unknown. Unlike plesiosaurs. We believe this area is composed entirely most articulated skeletons from the Redwater Shale, of pterygoid, although there is no sign of a midline this skeleton is not in a limestone concretion, instead suture. This block of bone continues caudally and then being preserved in the unlaminated gray-green narrows dorso-ventrally while expanding laterally into glauconitic shale common to this member of the two processes. Of the two processes the right is the Sundance Formation. The nature of the matrix makes better preserved, and thls carries a shallow cup on its preparation easy, but has adversely affected dorsal surface. This feature was probably an preservation; gypsum infiltration is a problem in many articulation for the basioccipital tuber. The two lateral of the bones, and many are fractured and incomplete. processes fonn a U-shaped excavation in the posterior The referral of this specimen to Tatenectes is made on aspect of the pterygoids. A low boss protrudes from the the basis of the cervical vertebrae. These share the base of this excavation on the midline; we believe this proportions and other identifying characters present in process articulated with a shallow pit on the anterior the neotype specimen. face of the basioccipital.This pit is present on the two The cranial material of U W 24215 comprises isolated Redwater Shale basioccipitals (see below), as several isolated teeth, an essentially complete left well as in the basioccipital of Aristonectes (Chatterjee squamosal, large portions of the left and right and Small 1989). pterygoids, the paraspbenoid, and a fragment of the left If this supposition is correct, at least part of the frontal.The squamosal is illustrated here in Figure 2. midline block of bone must be composed of Thisbone is a rather delicate, triradiate element bearing basisphenoid, albeit completely enveloped by the extensive similarities to those of other cryptoclcidoid pterygoids anteriorly and ventrally.A small shelf of taxa such as Tricleidus, Cryptoclidus, or bone is (poorly) preserved on the dorsal surface of the Kimmerosaurus (Brown 1981; Brown et al 1986). The pterygoid block, and this may represent the dorsum dorsal process of the squamosal arches over the baclc of sellae and sella turcica. The presence of these features the slruU to contact its neighbor on the dorsal midline would identify tbis region as the basisphenoid, and to form the 'squamosal arch', the apomorphic fonn of furthermore would place these structures in the correct the occiput characteristic of all plesiosaurs (and position relative. to the supposed location of the Pistosaurus, O'Keefe 2001 char. 27). This dorsal basioccipital (O'Keefe in press b). If the supposed process is quite gracile, however, and ends in a small locations of the endochondral braincase elements is bulb for articulation with the opposite squamosal, most correct (i.e. basioccipital and basisphenoid), the palate similar to the dorsal process of Tric/eidus (Brown of Tatenectes is extremely derived in that the 1981) or the known portions of Kimmero.<:aurus pterygoids produce a deep ventral process on the (Brown et al. 1986). The anterior process is again midline just beneath the forebrain. This process gives a similar to these two taxa, being a gracile process distinct topography to the palate, one that is very forming the ventral margin of the deep temporal unusual in cryptocleidoids, a group in which the palate fenestra, and articulating with the jugal anteriorly. The is generally planar (O'Keefe in press a). O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIOOID PLESIOSAUR 53
ja
bs? \
FIGURE 2. Cranial elementsofUW 2421 S. Specimen Ais theleft squamosal in antcro-mcdialview. Specimen B oompriscsleft andrig)lt pterygoid fragmentsin vcnlnll (left)and lateral (right)VIews. Theventral view also includes the parBSpbenoul Abbrcvtations � qa.quadrate articulation;Ja. jugalarticullllJQn; pterygoid;sqa, squamosalarricuhltion;pt. aipv, llllcnor intcrptcrygoid vacuity;bolA, basioccipital tuber attJculation; boa, basiOCCipitalllticulstion; ps, paraspbeooid; bs, bastspbeooi In the ventral view in Figure 2, another element dorsal vertebral centra coUectedas float over an area of here interpreted as the parasphenoid is included. This about 25 square meters. The vertebral centra are element is a small splint of bone that is slightly curved, probably from the same individual and are referable to possessing no clear articulation on its anterior end, but Pantosaurus (O'Keefe and Wahl 2003). The possessing a wide boss with two clear articulations on basioccipital seems too small to belong with the its posterior end. These articulations presumably cervical centra, but too few of these are preserved to contacted the antero-ventraledge of the basioccipital in document this quantitatively. Both basioccipitals lack a a condition strongly reminiscent of Tricleidus groove anterior to the occipital condyle and possess (Andrews 191 0). The location of the paraspbenoid in exoccipital articulations that intrude into the dorsal Figure 2 is probably too posterior; in life this element surface of the condyle. The dorsal surface of the body probably reached about two-thirds of the way to the of the basioccipital carries a Y -shaped groove of anterior interpterygoid vacuity. However, without fmished bone between the exoccipital articulations. material found in articulation it is impossible -w These conditions closely resemble that of detennine the exact location of the parasphenoid, and Kimmerosaurs and differ from that of Muraenosaurus. the bizarre morphology in this region is difficult to The basioccipital tubers are confluent with the interpret relative to other plesiosaurs. There is some basisphenoid articulation, however, a feature shared by precedent for the paraspbenoid occurring ventral to the Tricleidus, Aristonecte.f, and the polycotylids, but pterygoids in plesiosaurs, as this condition occurs in lacking in Kimmerosaurus. Given the marked Dolichorhynchops; however, the morphology of the similarity between these isolated basioccipitals and that present taxon differs radically from any polycotylid. of Kimmerosaurus, it seems probable that they belong Figure 3 illustrates two other cranial elements. to Tatenectes rather than Pantosaurus; however, until The first is a fragment of the right frontal.This element articulated material is found this referral is provisional. is broken on all edges except the midline suture. The basioccipital is scored as belonging to Tatenectes However, the preserved portion does carry a shallow in the cladistic analysis below; exclusion of this depression on the lateral side, rimmed by a thickened material does not effect the resultingtopology. ridge running down the midline, and by a low ridge Axial Skeleton-Preparation ofUW 24215 has so trending antero-laterally from the midline. These far yielded 14 cervical vertebrae, two of which are the particulars are very similar to the ventral surface of the articulated atlas/axis complex. Two of these cervicaJs frontal of Kimmerosaurus illustrated by Brown ( 1981 are illustrated in Figure 4, along with two isolated p. 307), and allow identification of the bone, although centra from the float collection. The isolated centra neither anterior or posterior sutures are preserved. A represent the two morphotypes of sma11 plesiosaur small area of fmished bone edge is preserved on the cervicals occurring in the Redwater shale. The first, lateral edge of the fragment, demonstrating that the longer centrum is assignable to Pantosaurus (O'Keefe frontal was quite narrow in this region, and that the and Wahl 2003) by virtue of its possession of the prefrontal and postfrootal did not meet over the orbit. following characters: length of centrum subequal to Again these features are very similar to width; body of centrum constricted; cervical rib Kimmerosaurus. The last element is a single tooth articulation carried on a pedestal; cervical rib preserving most of the crown and a long, although not articulation elongate; fine striations present on the complete, root.The crown carries fme striations all ventral surface of the centrum near the articular faces; around as does Kaiwhekea (Cruickshank and Fordyce and rim of articular faces well ossified. The second 2002), although in Tatenectes these are more centrum is assignable to Tatenectes given the following developed on the lingual surface. The tooth crown characters: length of centrum much shorterthan width; curves linguaJJy and is slender relative to its length, centrum not constricted; cervical rib articulation not resembling those of Kimmerosaurus (although this carried on pedestal; cervical rib articulation round; no taxon lacks tooth striations).The tooth is more robust striations on veQtral surface, and poorly ossified than those preserved with Aristonectes (Chatterjee and articular rims. The UW 24215 cervicals are clearly Small 1989). assignable to 'lhe second of these morphotypes as they The last cranial element of interest is the possess all of the diagnostic characterslisted . basioccipital, two examples of which are .known from The foramjna subcentralia are prominent but not the Redwater Shale, and neither of which were otherwjse remarkable, and the centra lack both a associated with UW 24215. The first is a weathered ventral and lateral keel. The cervical ribs are large, specimen in the float coUection and was not found single-headed, and lack anterior processes. Several of associated with other material. The second the cervicals preserve neural arches, and these are basioccipital is larger and the preservationis better, and fused to the centra, although the suture between arch was found with UW 15938. This specimen is an and centrum is clearly apparent The neural spines are associated, but not articulated, group of cervical and not compressed, are angled backward, and are rather O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 55 A Scm 8 2 em ,· FIGURE3. Cranialclements ofUW 24215. SpecimenA is a fragmentof theleft frontal ventral in view. Specimen Bisolated isan toolh partially freed from matrix. 56 PALUDICOLA, VOL. 4, NO.2, 2003 A 8 2cm ,· FIGURE 4. Cervical vertebrae of Tatenecte:s, with representativevertebral eenb'aofRedwatu Shale plesiosaursfurcomparison. Topleft is an isolated float specimen,UW 24239,n:femble to PantosaunLftrtriaJUS. Top rightis an isolatedfloat specimen, UW unnumbered,referable to Tatenectes laramienm. Bottomare twovertebrae cervicals from UW 2421 S Tatenectes laramlensJSspecimen, referred leftis in leftlatetal view, rightis in posteriorview. O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 57 short. In general terms the cervicaJs of Tatenecte.v are is presented in Figure 7. Bootstrap percentages based closely comparable to those of Kimmerosaurus (Brown on I 000 replicates, as well as decay indices, are et al. 1986), the only difference being that the neural presented next to the relevant node on the cladogram. spines of the later taxon arenot angled backwards. The atlas neural arch contacts the atlas DISCUSSION intercentru.m laterally, excluding the atlas centrum from the rim of the occipital articulation. This is the Taxonomy of Tatenectes--The taxonomic issues condition in most plesiosaurs (and in amniotes surrounding Tatenecte.s laramiensis are complex. The generally, Romer 1956) but not Muraenosaurus or taxon was originally erected by W.C. Knight (1900) as Cryptoclidus (Andrews 191 0). The atlas carries a well 'Cimoliosauru.s' laramiensis, based on a specimen developed atlas rib, accompanied by a very large rib on consisting of a partial axial skeleton and a nearly the axis. Ventrally, the axis intercentrum carries a complete front limb. MehJ (1912) reexamined Knight's prominent ridge on the midline; thls feature was termed ' • material, and some new material of his own, and the 'hypapophysial ridge' by Andrews ( 1910 p. 168), concluded that two plesiosaur taxa were present in the and occurs in Muraenosaurus, Cryptoclidus, Redwater Shale. Mehl therefore took 'Cimo/iosaurus' Tnc/eidus, and Po/ycoty/us, but is poorly developed in /aramiensis as a valid taxon, but felt that the genus Dolichorhynchops, Trinacromerum (pers. obs.), and Cimo/iosaurus was a nomen dubium. He then assigned Aristonectes (Chatterjee and Small 1989). This feature the name 'Tricleidus?' laramiensis to the taxon to is absent in elasmosaurs (Welles 1943, plate 22), reflect similarities between its humerus and that of the pliosaurs (Andrews 1913) and in more primitive Oxford Clay taxon Tncle1dus. However it is unclear plesiosaurs (Andrews 1909). The contact between the from Mehl's work that this second taxon is in fact atlas and axis intercentra on the ventral midline congeneric with Tric/eidus. This confusion stems at illustrated by Williston (1903) and included as a leastpartially from Mehl's almost exclusive reliance on cladistic character by O'Keefe (200 I) linking the humerus morphology, which is diagnostic in the case polycotylids and cimoliasaurid cryptocleidoids- does of Pantosaurus but probably not for the second taxon seem to be present in Tatenectes, although the state of (alsosee discussion in O'Keefe and Wahl 2003). preservation prevents absolute certainty. The humerus of Tatenectes is in fact quite similar to that of Tricleidus: the radial and ulnar articular CLADISTIC ANALYSIS facets are subequal in length; the distal end of the humerus possesses clear articulations for two In order to develop a hypothesis of relationship supernumerary ossifications in the propodial row, the for TaJenectes, a preliminary cladistic analysis was second of which makes a roughly 90 degree angle with performed on this taxon, here taken to consist of the the epipodial articulations; and the anterior edge of the neotype material (UW 15943 and UW 2480 I, UW humerus carries a continuation of the radial articulation 24215), and the isolated basioccipitals found as float. (similar to Tricleidus and Muraenosaurus but differing Inclusion or removal in the basioccipital characters did from Pantosaurus; Figure 5). The only particular in not affect the resulting tree topology. The data matrix which the Wyoming humerus differs is the shaft, which is an extensively revised and updated version of that is significantly longer and more gracile than in found in O'Keefe 200I, and is identical to the one in Tricleidus (see MehJ's iUustration of Knight's original O'Keefe in press a. The matrix contains 13 taxa scored specimen, reproduced here as Figure 5. This limb is for 95 morphological characters (for characters see certainly a forelimb, because the tuberosity is offset to Appendix I; data matrix is Appendix 2). All analyses the posterior aspect of the shaft, as is the case in many were performed using PAUP• 4.0 {Swofford 2001). cryptocleidoid humeri but no known femur [Andrews Sixty-two of the characters are parsimony-informative; 1910, O'Keefe and Wahl 2003], and possesses autapomorphies were retained in the matrix to aid in articulations for twq_ supernumerary ossifications, a the diagnosis of individual genera. The outgroup feature also found only in humeri). Therefore, if we (Pie.siosauru.s and Brancasaurus) was defined prior to accept MeW's chlh-acterization of his taxon 'Tricleidus'? parsimony analysis and constrained to be paraphyletic laramiensis on the basis of characters of the humerus to reflect the topology in O'Keefe 200I, although the alone, it can be demonstrated that it differs from same clade topology is obtained with this constraint not Pantosaurus. However, it cannot be conclusively in force. Parsimony analysis was performed using the demonstrated that it differs from the Oxford Clay branch-and-bound algorithm and yielded four most Tricleidus. To complicate matters further, the original parsimonious trees (MPTs) having a tree length of 160, holotype of 'Cimoliosaurus' /aramiensis Knight 1900 a consistency index (CI) excluding uninformative has been lost. Knight figured the humerus of his characters of .675, and a rescaled consistency index holotype, but not the axial column. (RCI) of .532. A strict consensus tree of the four MPTs 58 PALUDICOLA, VOL. 4, NO. 2, 2003 Lastly, it is clear from the pectoral girdle of the neotype that 'Cimo/iosaurus' laramiensis is not congeneric with Tricleidus. We therefore erected a new genus for the taxon, creating Tatenectes laramiensis Knight 1900 (new combination). Cladistic Analysis and Aristonutes--The cranial and cervical material of Tatenectes described here, although fragmentary, is sufficient to demonstrate that this taxon is very similar to Kimmerosaurus in many respects. The cervical vertebrae are almost identical, as are the squamosal, the frontal, and the general dimensions of the teeth. The cladistic analysis reflects this general impression, with Ta tenectes falling out as the sister taxon of Kimmerosaurus in all MPTs. Both are members of the family Cimoliasauridae as defmed by O'Keefe 200 1; this taxon of poorly known animals also includes Aristonectes from Antarctica and South America. The present analysis finds the recently described New Zealand taxon Kaiwhekea (Cruickshank and Fordyce, 2002) to be the sister group of this family, fo nning a monophyletic clade with good decay index and bootstrap support. The family Cimoliasauridae should probably be broadened to include Kaiwhe/cea; however, continued instability is seemingly guaranteed in this clade given the lack of knowledge concerning many of its members. Also, the genus Cimo/iasaurus is a taxonomic morass that must be revised, and the validity (or lack thereof) of the genus name may affect the family name. We therefore refrain from revising the fa mily until the taxonomy at the genus level stabilizes. FIGURE S. Mehl's 1912 illustration of the humerus of the holotype The issues surrounding the Cimoliasauridae have of'Cimoli030urui Knight 1900;labels have loramlerui.J beenadded. been further complicated recently by the publication by Abb�iatiOfU are: h, humerus; r, radius; s1, supernumerary Arislonectes ossificahooone; s1a. supernumerary ossification rwo articulation; u, Gasparini et al. on (2003). These authors ulnL have two cenual contentions, the fl. rst being that the genus 'Morturneria' (Chatterjee and Small 1989) is a Fortunately, it is possible to characterize Knight's junior synonym of ArisiOnectes Cabrera 1941. We have bolotype further from his original description, because viewed the 'Morturneria' material, but not that of he included measurements of two cervical vertebrae. It Aristonectes, and so cannot bold an infonned opinion is clear from these measurements that the vertebrae are about this issue; we have therefore accepted the compressed antero-posteriorly- that they are much suggested synonymy in this publication. However, the wider than they are long- and so represent the second second contention of Gasparini et al.-- that Aristonecres of the two cervical vertebra morphotypes identifiable in is an elasmosaur, and that there are no cryptocleidoid the Redwater Shale (see Figure 4). We therefore can plesiosaurs in the Late Cretaceous -- is more establish an association between the humerus figured problematical. The cladogram offered here (Figure 7) by Knight and the compressed cervical morphotype. clearly places Aristonectes within the Cryptocleidoidea This association is identical to that observed in one of and does not clus�er ii with the primitive elasmosaur the new specimens collected by the Tate Museum (UW Brancasaurus, and this is a well-supported fm ding 15943 & UW 2480 1). The humerus fragment of this (bootstrap support .95, decay index six). Furthennore, a specimen agrees in all particulars with that of Knight's cladistic analysis of the entire clade produces the same holotype specimen, and the cervical vertebrae are of resuJL This second analysis was based on a matrix of the compressed morphotype. Given this agreement, we 35 taxa and 170 characters, and is an updated version decided to designate (UW 15943 & UW 24801) as the of the matrix in O'Keefe (200 I). Constraining neotype of 'Cimoliosaurus' /aramien.ss,i because it Aristonectes to membership in a clade with the other overlaps well with the lost holotype, and shares all of elasmosaurs resulted in a tree nine steps longer than the its characters that are now possible to detennine. O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 59 A FIGURE6. PaJaaJ viewsofUSNM 419640 , thecrugmatic Cubanskull Abbreviations �: pdos.pterygoid fossa; ps.q(i)t. paraspheooid; quadt1lc llanðe of pterygoi d; bota, basiOCC ip ital tuberuticul atioo;occ oc, ipital condyle;bo, basioccipita 60 PALUDICOLA. VOL. 4, NO. 2, 2003 Cimoliasauridae Polycotylidae I I I (/) I (I) (I) g. E (/) 2 2 (I) i3 (I) 2 ::::s (I) ::::s (I) 2 2 en Cl) (I) 2 ::::s en ::::s (t) ::::s §_ Q) (I) Cl) Q) � ::::s ::::s en :-o Cl) - 0 en E en 0 ..:::: ::::s e ,-. u � >.. -E (I) 0 Q) ...... (I) (I) c: Q) Q) 0 e (t) � Q) c: 0 e 0 Q) .2 c:::: ..s::: 0 .Q � E .2 (.) m Cl) c: � � � cn .! ·- � .o <- .c: � ::::s \::: .E -� .g> 0 it co � 0 t--: � � <;( � & lJJ Q � .92/2 .64/2 Cryptocleidoidea FIGURE 7. Cladognunoftbe Cryptoclci doidca. Numbersbeneath are eachnode llootstnpvalues/decly indices. For discussionoftbe aoalysis see tat. O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 61 most parsimonious tree. The cladistic matrix in Tatenectes. The Cuban skull also possesses the distinct Gasparini et aL contains ten taxa scored for twenty processes for articulation with the basioccipital tubers characters, and those authors do not consider most of present in Tatenectes. However, the Cuban skull also the anatomical evidence presented both here and in lacks an anterior interpterygoid vacuity (present in O'Keefe 200 I. Because the Gasparini et al. data set is Ta tenectes), and in this region the pterygoids develop a so restricted. and because the analysis of a more deep fo ssa on the midline. This fo ssa is identical to one inclusive data set yields a strongly contrary result, we preserved on a large palatal fragment with the do not accept their conclusion that Aristonectes is an 'Morturneria' type material (not figured by Chatterjee elasmosaur. Clearly, the phylogeny of the and Small 1989; pers. obs.). While no teeth are Cimoliasauridae is a problem demanding immediate preserved with the Cuban skull, the alveoli indicate research attention. they were very slender, again as in Ari.vtonectes. The Tlltm�ct� and Future Study-The most Cuban skull, therefore, may be a Late Jurassic distinctive trait displayed by Tatenecres is the deep '· representative of the aberrant cryptocleidoid radiation structure developed by the pterygoids on the posterior thought to be restricted to the Late Cretaceous of the palate midline, and the concomitant lack of posterior southern hemisphere by Cruickshank and Fordyce interpterygoid vacuities. The possession of posterior (2002). interpterygoid vacuities is a hallmark of almost all The anatomy of the palate of Tatenectes is plesiosaurs(O'Keefe in press b, O'Keefe 200I); another bizarre, although not as bizarreas that displayed by the is the possession of a planar or nearly planar palate and Cretaceous cimoliasaurids, with whom it shares many basicranium. The suite of characters displayed by traits. Tatenectes retains an anterior interpterygoid Tatenectes is in fact shared by only one complete skull, vacuity as in Kimmerosaurus and other a presently undescribed specimen in the Smithsonian cryptocleidoids. The skull of Tatenectes is also similar Institution (USNM 419640) from the Late Jurassic of to that of Kimmerosaurus in other respects, such as the Cuba, here illustrated in Figure 6. This skull is squamosal, the frontal, and the dentition, and in apparently from the Jagua Fonnation (Oxfordian), Kimmerosaurus there is at least some development of although it was not listed with the other plesiosaur medial processes of the pterygoids (Brown 1981 P- slrulls in the catalogue of Cuban reptile material 308). Tatenectes can therefore be thought of as published by lturralde-Vinent and Norell ( 1996). intermediate between Kimmerosaurus on one hand and Additional information on the stratigraphy and the derived cimoliasaurids- Aristonectes, Kaiwhekea, provenance of Cuban marine reptiles can be found in and the Cuban taxon- on the other. As such it is of this reference; the exact provenance of USNM 419640 critical imponaoce, because it has the potential to is cWTently unknown. The relation of this sk:ull to that untangle the anatomy and relationships of this strange descnbed by Gasparini et at. 2002 has yet to be group of animals. There is no cladistic support for this determined. hypothesis as present apart from the sister relationship Unfortunately the Cuban skull is in very poor between Kimmerosaurus and Tatenectes; the taxa condition; the skull was apparently coUected in a Kaiwhekea and Aristonectes actually fall more basal limestone concretion and then acid-prepared, and than the former taxa in the cladogram in Figure 7. We damage to the bone surface is severe. At present the believe this result is due to the large amount of missing slrull is held together by a thick coat of varnish. with data for Kaiwhekea and Aristonectes; much of the fragments defoliating on all sides, including large detailed anatomythese of taxa issimply not available. portions of the left skull roof and left mandible. No It is becoming clear that a large radiation of original bone surface or suture is visible on the skull cryptocleidoid plesiosaurs occurred in the Late roof or on the mandible fragments. The overall shape Jurassic, giving rise to long-necked forms of the sk:ull indicates that it belongs to a cryptocleidoid (Muraenosaurus and Pantosaurus), short -necked plesiosaur similar to Tricleidus. Although this fo rms (the Polyc9tylidae), and the aberrant specimen is Wldoubtedly a new taxon we have chosen Cimoliasauridae. More research is badly needed on this not to name it at present, as the state of preservation interesting time in plesiosaur evolution, beginning with makes adequate description difficulL the fauna of the Redwater Shale and Tatenectes in The palate has suffered less from acid damage particular. This taxon may prove to be the sister taxon than has the skull roof, and the one suture visible on of the Cretaceous cimoliasaurids, but to demonstrate the skull- the midline suture on the palate- is this more and better cranial material of Tatenectes must preserved here. The palate is remarkable in that it be found. Lastly, research attention on the possesses the union of the pteJygoids on the posterior cimoliasaurids is long overdue. midline, the deep dorso-ventral development of the pterygoids in this region. and the loss of the posterior interpterygoid vacuities, all described above for 62 PALUDICOLA, VOL. 4, NO. 2, 2003 ACKNOWLEDGMENTS Iturralde-Vinent, M. and M. A. Norell. 1996. Synopsis of Late Jurassic marine reptiles fo r Cuba. J. Massare and an anonymous reviewer provided American Museum Novitates 3 164: 1-I 7. thorough and helpful comments on an earlier version of Knight. W. C. 1898. Some new Jurassic vertebrates this manuscript. M. Cassiliano was very helpful in from Wyoming. American Journal of Science, arranging access to the UW collections. Special thanks Fourth Series 5 (whole number 155):378-380. to D. Brown and the staff of the Tate Museum for Knight. W.C. 1900. Some new Jurassic vertebrates. access to collections and help of all kinds. The staffat American Journal of Science, Fourth Series 10 the American Journal of Science furnished useful and (whole number 160): 1 15-1 19. timely pdffiles of Knight references. Marsh, 0. C. I 891. Geological horizons as determined by vertebrate fo ssils. American Journal of LITERA 11JRE CITED Science 42:336-338. -·· Marsh, 0. C. 1893. Congress Geologique International, Andrews, C. W. 1909. On some newPlesiosauria from Compte Rendu de Ia 5me Session, Washington, the Oxford Clay of Peterborough. Annals and D. C. 1891:156-159. Magazine of Natural History, London 4(8):418- Marsh, 0. C. 1895. The Reptilia of the Baptanodon 429. beds. American Journal of Science 50: 405-406. Andrews, C. W. J 910. A descriptive catalog of the Mehl, M. G. 1912. Muraenosaurus? reedii, sp. nov. marine reptiles of the Oxford Clay, Part I. British and Tricleidus? /aramiensts Knight. American Museum (Natural History}, London, England. Jurassic plesiosaurs. Journal of Geology 20 (4): Andrews, C. W. 1913. A descriptive catalog of the 344-352. marine reptiles of the Oxford Clay, PartII. British O'Keefe, F. R. 200l. A cladistic analysis and Museum (Natural History), London, England. taxonomic revision of the Plesiosauria (Reptilia: Brown, D. S. 198 I. The English Upper Jurassic Sauropterygia). Acta Zoologies Fennica 213:1- Plesiosauroidea (Reptilia) and a review of the 63. phylogeny and classification of the Plesiosauria. O'Keefe, F. R. 2002. The evolution of plesiosaur and Bulletin of the British Museum of Natural pliosaur rnorphotypes in the Plesiosauria History (Geology)35(4):253-347. (Reptilia: Sauropterygia). Paleobiology 28(1): Brown, D. S., A. C. Milner, and M. A. Taylor. 1986. 101-112. New material of he plesiosaur /(jmmerosaurus O'Keefe, F. R. In press a. Onthe cranial anatomy ofthe langhami Brown from the Kimmeridge Clay of polycotylid plesiosaurs, with new cranial material Dorset. Bulletin of the British Museum ofNatural of Polycotylus latipinnus Cope. Journal of History(Geology) 40(5):225-234. Vertebrate Paleontology. Cabrera, A. 1941. Un plesiosaurio nuevo del CrctAceo O'Keefe, F. R. In press b. Neoteny and the del Chubut. Revista del Museo de La Plata 2: plesiomorphic condition of the plesiosaur ll3-130. basicranium. in M. T. Carrano, T. J. Gaudin, R. Chatterjee, S., and B. J. Small. 1989. New plesiosaurs W. Blob and J. R. Wible, eds. Amniote from the Upper Cretaceous of Antarctica. Pages Paleobiology: Phylogenetic and Functional 197-215 in Crarne, J. A., ed. Origins and Perspectives on the Evolution of Mammals, Birds Evolution of the Antarctic Biota. Geological and Reptiles. University of Chicago Press, Society Special Publication 47. Chicago, II. Cruickshank, A. R. 1., and R. E. Fordyce. 2002. A new O'Keefe, F. R. and W. Wahl. 2003. Current taxonomic marine reptile (Sauropterygia) from New status of the plesiosaurPantosaurus striatus from Zealand: further evidence for a Late Cretaceous the Upper Jurassic Sundance Formation, austral radiation of crytptoclidid plesiosaurs. Wyoming. Paludicola 4(2):37-47. Palaeontology 45(3):557-575. Romer, A. -S. 1956. Osteology of the Reptiles. Gasparini, Z., N. Bardet, J. E. Martin, and M. University of Chicago Press, Chicago, Tllinois. Fernandez. 2003. The elasmosaurid plesiosaur Swofford, D. 200I. Pylogenetic Analysis Using Aristonectes Cabrera from the latest Cretaceous Parsimony • 4.0. Sinauer Associates, Inc. of South American and Antarctica. Journal of Sunderland, Massachusetts. Vertebrate Paleontology23(1): 104-115. Welles, S. W. 1943. Elasmosaurid plesiosaurs with Gasparini, Z., N. Bardet. and M. lturralde-Vinent description of new material from California and (2002) A new cryptoclidid plesiosaur from the Colorado. Memoirs of the University of Oxfordian (Late Jurassic) of Cuba. Geobios California 13(3): 125-254. 35:201-211. O'KEEFE AND WAHL-ABERRANT CRYPTOCLEIDOID PLESIOSAUR 63 Williston, S. W. 1903. North American plesiosaurs, Part I .Field Columbian Museum Publication (Geology) 73(2): 1-77 _ .- 64 PALUDlCOLA, VOL. 4, NO. 2, 2003 Appendix I. Cladistic characters used in phylogenetic analysis. For more complete character descriptions of characters l-88 see O'Keefe 200I. Character States/Coding I Relative skull length primitive/ 'nothosaurian' (0), large (I), small (2) 2 Relative neck length primitive (0), long (I), short (2) 3 Relative length of ischium/pubis subequal (0), ischium longer (I), pubis longer (2) 4 Relative length ofbumerus/femur subequal (0), humerus longer (I), fe mur longer (2) 5 Preorbital and postorbital skull length subcqual (0), longer prcorbital (I), shorter (2) 6 Fin aspect mtio high (0), low (I) 7 Elongate rostrum absent (0), premaxilla only (I), very long with maxilla in"'eeoded (2), elongate and hoop-like/ uoconstricted (3) 8 Dorso-medial process of premaxma contacts frontal (0), contacts parietal at pineal fo ramen(2), contacts anterior extension of the parietal (I) 9 Premaxilla/ externaJ naris contact present (0), absent (1) 1 0 Frontals paired/fused in adult paired (0), fused ( 1) ll Frontal with or without distinctpostero- without processes (0) lateml process with processes (I ) 12 Frontal enters margin of tempomJ doesnot (0) fe nestra does narrowly ( l) 13 Frontal contacts external naris does contact (0), does not contact (l) 14 Pineal fo ramen bordered anteriorly by not bordered by frontal (0), bordered by frontal (I) frontals on dorsal skull surface 15 Frontal process projects into orbit absent (0), present ( 1) 16 Parietal skull table relatively broad (0), constricted (1), sagittal crest (2) 17 Squamosal produces long, thin process no medial process (0), medial process and socket-like covering quadrate lat.emJly squamosal (I) 18 SquamosaV postorbital contact contact (0), no contact (I) 19 Jugal extends anteriorly along ventral anterior margin (0), middle of orbit (I), restricted to orbital margin posterior margin (2) 20 Jugal fo rms narrow barbetween orbit does not (0), does ( 1) and tempomJ emargination 21 Maxilla/squamosal contact no contact (0), contact (I), expanded posterior flange (2) 22 Exoccipital participates in fo rmation of do not participate(0), doparticip ate (I) occipital condyle 23 Occipital condyle morphology hemispherical with groove (0), short with no groove (I) 24 Pamoccipital process morphology gracile (0), robust (I) 25 ParaoccipitaJ process articulation squamosal exclusively (0), quadrate exclusively (I), both quadrate and squamosal (2) 26 VentmJ extentofparaoccipital process does not extend ventmJ to occipital condyle (0), extends past condyle (I) 27 Nature of paraoccipital process/ quadrate no contact (0), contact at latemJ_articulation only (I), pterygoid flange contact long contact aJong bodies ofproc esses (2) 28 Quadrate produces distinct process for process absent (0), process'present (I) articulation with pterygoid flange 29 Dorsal wing of epipterygoid broad/ columnar (0), reduced ( 1) 30 Epipterygoid dorsal processcontacts contact (0), no contact (I) parietal 31 Quadrate embayedl dished-shaped massive quadrate (0), dished anteriorly (I) anteriorly 32 Supraoccipital depth/sigmoid suture Shallow (0), deep antero-posteriorly/ sigmoid suture with exoccipital and prootic (l) O'KEEFE AND WAHL-ABERRANTCRYPT OCLEIOOIDPLESIOSAUR 65 33 Squared lappet of pterygoid underlies no squared lappet (0), squared lappet ( 1) quadrate pterygoid flange 34 Anterior interpterygoid vacuity absent (0), slit-like (1), broad withround ends (2) 35 Pterygoids meet posterior to posterior pterygoids do not meet (0), pterygoids meet ( 1 ), meet interpterygoid vacuity but arecovered by posterior parasphcnoid process (2) 36 Pterygoids meet between anterior and do not meet between vacuities (0), do meet between posterior interpterygoid vacuities vacuities (I) 37 Columnar ectopterygoid contacts no contact (0), contact (I) postorbital bar 38 Dished pterygoids absent (0), present (I) 39 Posterior pterygoid/parasphenoid Absent (0), present (I) contact 40 Parasphenoid morphology Iong,mpering anteriorly (0), short and blunt( I) 41 Paraspbenoid exposure anteriorto anterior parasphenoid not exposed on palate surface(0), posterior interpterygoidvacuities exposed via extension of posteriorinterpterygoid vacuities ( 1 ), exposed with lateral pterygoid sutures (2) 42 Parasphenoid/ basioccipital contact on absent (0), present (I) midline. 43 Basioccipital tubers reduced not reduced (0), reduced/tuber facets confluent with basisphenoid articulation (I) 44 Palatine/ internalnaris palatine enters internal naris border (0), excluded by vomer/maxilla contact (I) 45 Premaxilla/ anterior border of internal premaxilla enters anteriorborder (0), is excluded by naris vomer/ maxilla contact ( 1) 46 Sub-orbital fenestration absent (O), present (l) 47 Vomers extend farposterior to internal do not (0), extend posteriorand meet pterygoids in wide nares on midline interdigitatingsuture ( 1) 48 Mandibularsymphysis short(0), somewhat enforced (I), scooplike (2), long (3) 49 Splenial participates in symphysis doesnot participate(0), does participate (1), angulars extend past symphysis (2) 50 Coronoid present (0), absent ( 1) 51 Coronoid exposed on lateraljaw surface no exposure (0), exposure (I) 52 Prearticular shelf/groove absent (O), present (l) 53 Jaw articulation in relation to tooth row above or at collinear withtooth row (0), lower than toothrow (I) 54 One or two caninifonn teeth on maxilla present ( 1 ), absent (0) 55 tooth form gracile, small root, narrow, no wear (0) robust. large root, wear (I), very smalVneedle-like (2) 56 number of premaxillaryteeth 5 (0), 6(1), 7(2), greater than 7(3) 57 Maxillary teeth less than twenty (0), more twenty to thirty (I), many more than thirty(2) 58 Articulationof axis rib broad articulationwith atlas centrumand/or other elements (0), head confinedto axis centrum (I) 59 Atlas/axis morphology no lateral exposure of atlas centrum on cup face (0), lateral exposure (I), no lateral expo_sure, but atlas and axis intercentra exclude atlas centrum ventrally(2) 60 Number of cervical vertebrae primitive (0), increased (1), reduced (2), greater than 50 61 Proportions of cervical centra length equal to height (0), length greater than height ( 1 ), length Jess than height (2) 62 Distinct change in zygopopbyseal angle no change in angle (0), change (I) along cervical column 63 Ventral keel on cervical vertebrae absent (O), present (1) 64 Binocular shaped anterioccervical cen1ra absent (O), pnesent(J) 65 Width of cervical zygopopbyses wider than centrum (0), subequal with centrum (I), more narrow than centrum(2) 66 PALUDICOL� VOL. 4, NO. 2, 2003 66 Posterior articulation for s-ucceeding absent (0), present (I) neural spine, cervical vertebrae 67 Cervical rib articulation greatly cireular or subcircular (0), elongate (I) elongate/ cervical ribs expanded and blade-like 68 Anterior processof cervical ribs present (0), absent (I) 69 Anterior neural flange on cervical neural absent (0), present (I) spines 70 Neural spines, cervical vertebrae angled backward (0), not angled ( 1) 71 Lateral compression of neural spines, not compressed (0), compressed and blade-like (I) dorsal and cervical vertebrae 72 Interclavicle posterior process present (0), absent ( 1) 73 Presence of clavicles and interclavicle present (0), interclavicle absent ( J ). both absent (2) 74 Clavicle median symphysis symphysis (0), separatedby interclavicle (1), meet only behind notch (2) 75 Scapulae meet in anterior median separated by clavicles/interclavicle (0}, meet medially symphysis but leave notch for dermal elements (1), meet in long symphysis with no notch (2) 76 Anterior intrascapular fe nestra abseot (O), present (l) 77 Longitudinal pectoral bar absent (0), formed by clavicle and coracoid (1), fo rmed by scapula and coracoid (2) 78 Median coracoidperf orations absent (0), present (I) 79 Posterior coracoid extension with deep absent (0), present (1) median embayment 80 Pubis ventral (medial) margin convex ( 1 ), concave (0) 81 Median pelvic bar absent (0), present (I) 82 Angled humerus absent ( 1 ), present (0) 83 Distal end of humerus has two distinct absent (0), present (l) planes in adult 84 Distinct facet on distal humerus for abseot (O), present (l) supemumery ossification 85 Distinctly lunate ulna absent (0), present (1) 86 EpipodiaJ morphology longer than broad (0), equal or broader than long (I) 87 Supemumery ossifications, forelimb none (0), epipodial row/pisiform (I), propodial (2), both (3) 88 Interlockingdistal phalanges anterior to absent (0), present (I) fifthphalangeal row 89 Marked groove around margin dorsal of absent (0), present (1) and cervical vertebral articular surfaces 90 Processof postorbital extends present (1), absent (0) posteriorly along lateral margin of temporal fenestra 91 Deep notch in posterior margin of clivus present (I), absent (0) 92 Length of retroarticular process short (0), long (l) _ 93 Width of central pterygoid plate central plate absent (0), narrow ( 1 ), -broad (2) 94 Height of sagittalcrest low (0), high (1) 95 Sigmoid humeral shaft absent (0), present ( 1) O'KEEFE AND WAHL- ABERRANT CR YPTOCLEIDOID PLESIOSAUR 67 Appendix 2. Cladisticcharacter data used in phylogenetic analysis. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Plesiosaurus 0 0 2 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 1 Muraenosaurus 2 1 0 2 0 1 0 0 0 0 0 0 0 1 0 2 1 0 2 Cryptoclidus 2 0 0 2 0 1 0 0 0 0 0 0 1 1 0 2 1 0 2 Tricleidus 2 0 ? 0 2 1 0 0 0 0 ? 0 ? 0 0 2 1 ? 2 Kimmerosaurus ? ? ? ? 2 ? 3 0 ? 0 0 0 ? 1 0 2 1 ? ? Aristonectes ? ? ? ? 2 1 3 ? ? ? ? ? ? ? ? ? ? 1 ? Edgarosaurus 0 2 ? ? 0 1 1 0 1 0 0 0 1 0 0 2 1 0 2 Polycotylus ? 2 1 0 ? 1 ? -·- ? ? ? ? ? ? ? ? ? ? ? ? Dolichorhynchops 1 2 1 0 1 1 2 1 0 X 0 0 ? 0 1 2 1 0 2 Trinacromerum 1 2 1 1 1 1 2 1 0 X 0 0 ? ? 1 2 1 0 2 Brancasaurus 2 1 2 0 0 0 0 0 0 1 0 1 ? 1 0 2 1 0 2 Tatenectes ? ? ? ? ? ? ? 0 ? 0 ? ? ? ? 0 ? 1 ? ? Kaiwhekea 2 0 ? ? 2 1 0 1 0 X 0 ? 1 0 0 2 1 ? ? 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Plesiosaurus 0 1 ? 0 0 ? ? ? 1 1 1 1 0 0 1 0 0 0 0 Muraenosaurus ? ? 0 1 0 ? 0 ? ? ? ? 1102 0010 Cryptoclidus 1 0 1 1 0 1 0 0 0 ? ? 110 20010 Tricleidus 1 0 0 0 0 1 1 0 0 ? ? 1 ? 0 2 0 0 1 0 Kimmerosaurus 1 ? 1 1 0 0 0 0 0 ? ? 1 1 0 2 0 0 ? 0 Aristonectes ? ? 0 0 1 0 ? 1 ? ? ? 1 1 ? ? ? 0 ? 0 Edgarosaurus 0 2 0 0 1 ? ? ? 0 ? ?? 102101 1 Polycotylus ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Dolichorhynchops 0 2 0 0 0 1 1 1 0 0 011022011 Trinacromerum 0 2 ? 0 0 1 1 ? ? 0 0 1 ? 1 2 2 0 ? 1 Brancasaurus 0 1 0 0 0 ? 0 ? ? ? ? 0 1 ? 0 ? ? ? 0 Tatenectes 1 ? 1 1 ? ? ? ? ? 0 ? ? ? ? 2 ? 1 ? 0 Kaiwhekea ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 Plesiosaurus 0 0 1 0 0 0 0 0 0 0 ? ? 0 0 0 0 0 0 ? Muraenosaurus 0 1 1 0 0 1 0 0 1 0 0 0 1 1 0 0 0 0 0 Cryptocldusi 0 1 2 0 0 0 ? 0 0 0 001 100010 TricJeidus 1 1 2 1 1 ? 0 ? 0 0 0 0 ? 1 1 0 0 0 ? Kimmerosaurus 1 ? 2 1 0&1 ? ? ? ? 0 0 1 . 1 1 1 0 2 3 2 Aristonectes ? 1 2 1 1 0 1 ? ? ? ? ? ? 0 1 0 2 2 2 Edgarosaurus 1 1 2 ? ? ? ? 0 ? 2 1 0 1 1 1 1 1 1 0 Polycotylus ? ? ? 1 ? ? ? ? ? ? ? 0 1 1 1 ? 1 ? ? Dolichorhynchops 1 1 2 1 1 0 1 1 0 3 2 0 1 1 1 0 0 0 0 Trinacromerum 1 1 2 1 1 ? ? 1 ? 3 2 ? 1 1 1 0 0 0 1 Brancasaurus 0 0 ? 0 0 0 1 ? 1 ? ? ? ? ? ? 0 0 0 ? Tatenectes 1 1 X 1 1 ? ? ? ? ? ? ? ? ? ? ? 2 ? ? Kaiwhekea ? ? ? ? ? ? ? ? ? 0 0 ? ? ? 0 ? 2 2 2 68 PALUDICOLA, VOL. 4, NO. 2, 2003 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 Plesiosaurus ? 0 1 0 ? 0 0 2 1 0 0 0 1 1 ? 1 0 0 0 Muraenosaurus 0 1 1 1 0 0 ? 2 1 1 1 1 1 1 ? 0 1 1 1 Cryptoclidus 0 1 0 0 1 0 0 2 0 1 1 1 1 1 ? 0 0 1 1 Tricleidus ? 1 0 0 1 0 0 2 1 0 1 1 1 0 0 0 2 1 0 Kimmerossurus ? ? ? 2 ? 0 0 2 ? 1 1 ? 1 0 ? ? ? ? ? Aristonectes 0 2 ? 2 ? ? 1 2 ? 0 ? ? ? ? ? ? ? ? ? Edgarosaurus 1 2 2 2 ? 1 0 ? ? 0 ? ? ? ? ? ? ? ? ? Po/ycoty/us 1 2 2 2 0 ? -� 0 1 1 0 ? 1 1 0 X 0 2 0 1 Dolichorhynchops 1 2 2 2 1 ? 0 2 1 0 ? 1 1 0 X 0 2 0 1 Trinacromerum 1 2 2 2 1 ? 0 2 1 0 ? 1 1 0 X 0 2 0 1 Brancasaurus 1 0 1 1 ? ? 0 2 1 0 ? 0 0 1 1 0 0 1 ? Tatenectes ? 2 ? 2 ? 0 1 2 0 0 1 1 0 0 ? ? ? 1 ? Kaiwhekea ? ? 1 2 ? 1 1 2 ? 0 1 ?11 ????? n 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 P/esiosaurus 1 0 0 1 1 0 0 0 1 0 100111000 Muraenosaurus 2 0 0 1 0 1 1 1 0 1 1 0 1 ? 0 0 1 0 0 Cryptoclidus 2 0 0 1 0 1 1 0 0 1 1 0 1 0 0 0 1 0 0 Tricleidus 2 0 0 1 0 1 1 1 0 1 2 ? 1 0 0 0 1 0 0 Kimmerosaurus ? ? ? ? ? ? ? ? ? ? ? ? 1 0 ? 0 1 0 ? Aristonectes ? ? ? ? ? 1 ? ? ? 1 ? ? 1 ? ? ? X ? ? Edgarosaurus ? ? ? ? ? 1 1 1 0 1 ? 1 ? 0 ? 0 2 0 ? Polycotylus 1 1 ? 1 0 1 1 1 0 1 3 ? 1 ? ? 0 1 0 1 Dolichorhynchops 1 1 0 1 0 1 1 0 0 1 3 1 1 0 0 1 2 1 1 Trinacromerum 1 1 0 1 0 1 1 0 0 1 3 1 1 0 0 1 2 0 1 Brancasaurus 2 0 1 0 1 1 ? ? 0 1 ? ? 0 0 ? 1 0 0 0 Ta tenectes 0 0 0 ? ? 1 1 1 ? 1 2&3 ? 1 ? ? ? X ? 0 Kaiwhekea ? ? ? ? ? ? ? ? ? 1 ? ? 1 ? ? 0 ? 1 ? - . ;