5 $0- 5 ex no- ^3 ^ FIELDIANA WDlOOrifBWBy Geology NEW SERIES, NO. 43 The Intramandibular Joint in Squamates, and the Phylogenetic Relationships of the Fossil Snake Pachyrhachis problematicus Haas Olivier Rieppel Hussam Zaher March 31, 2000 Publication 1507 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, U.S.A. Text: Manuscripts must be typewritten double-spaced on standard- weight, 8Vi- by 1 1 -inch paper with wide margins on all four sides. If typed on an IBM-compatible computer using MS-DOS, also submit text on S^-inch diskette (WordPerfect 4.1, 4.2, or 5.0, MultiMate, Displaywrite 2, 3 & 4, Wang PC, Samna, Microsoft Word, Volks- writer, 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 floristics. Journal of Ecology, 51: 567-601. Langdon, E. J. M. 1979. Yage among the Siona: Cultural patterns in visions, pp. 63-80. I_n 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, DC. 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 "X0.8," 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 8% by 1 1 inches (22 X 28 cm) and may not exceed 1 \Vi by 16Vi 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 corre- sponding 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 paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). OEOUOGf UBRAIW FIELDIANA Geology NEW SERIES, NO. 43 The Intramandibular Joint in Squamates, and the Phylogenetic Relationships of the Fossil Snake Pachyrhachis problematicus Haas Olivier Rieppel Department of Geology Field Museum of Natural History 1400 South Lake Shore Drive Chicago, Illinois 60605-2496 U.S.A. Hussam Zaher Departamento de Zoologia Instituto de Biociencias Universidade de Sao Paulo Caixa Postal 11461 05422-970 Sao Paulo, SP Brasil Accepted July 28, 1999 Published March 31, 2000 Publication 1507 PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY © 2000 Field Museum of Natural History ISSN 0096-2651 PRINTED IN THE UNITED STATES OF AMERICA Table of Contents Abstract 1 Introduction 1 Materials and Methods 2 The Intramandibular Joint in Squamates 3 The Intramandibular Joint in Varanus and Lanthanotus 3 The Intramandibular Joint in Mosasaurs .... 7 The Compound Bone of the Ophidian Mandible 8 The Intramandibular Joint in Scolecophi- dians 9 The Intramandibular Joint in Anilioids 10 The Intramandibular Joint in Basal Ma- crostomatans 18 The Skull and Lower Jaw of Pachy rhachis 24 Character Evidence for the Monophyly of the Pythonomorpha 29 Cladistic Analysis 49 The Phylogenetic Relationships of Pachy RHACHIS, DlNILYSlA, AND DlBAMUS 61 Discussion: Snake Origins, and Homolo- gy Versus Convergence 62 Notes Added in Proof 65 Acknowledgments 66 Literature Cited 66 3. Lower jaw of Lanthanotus borneensis 6 4. Lower jaw of Platecarpus 7 5. Lower jaw of Anilius scytale 11 6. Transverse section through the lower jaw of Anilius scytale 12 7. Lower jaw of Cylindrophis ruffus 12 8. Lower jaw of Cylindrophis maculatus 14 9. Lower jaw of Melanophidium puncta- tus, Platyplecturus madurensis, and Pseudotyphlops philippinus 16 1 0. Lower jaw of Plecturus perroteti 17 11. Lower jaw of Python reticulatus 19 1 2. Lower jaw of Lichanura trivirgata ro- seofusca 21 13. Lower jaw of Calabaria reinhardti and Charina bottae 22 14. Lower jaw of Boa constrictor impera- tor 23 15. Skull of Pachyrhachis problematicus .... 26 1 6. Lower jaw of Pachyrhachis problema- ticus 28 17. Snake interrelationships 61 List of Tables Data matrix for the analysis of squamate interrelationships 50 List of Illustrations 1 . Lower jaw of Varanus 4 2. Coronoid and splenial of Varanus 5 The Intramandibular Joint in Squamates, and the Phylogenetic Relationships of the Fossil Snake Pachyrhachis problematicus Haas Olivier Rieppel Hussam Zaher Abstract A review of the morphology of the lower jaw in varanoid squamates, including mosasaurs, and basal snakes (scolecophidians, anilioids, basal macrostomatans) reveals a greater degree of variability in the differentiation of the intramandibular joint than had previously been recorded. In particular, the mandibular joint of mosasauroid squamates and snakes differs fundamentally. In mosasaurs, the dentary is primarily suspended from the prearticular and the posteriorly concave splenial receives the anteriorly convex angular. In snakes, the dentary is primarily suspended from the surangular portion of the compound bone, and the angular is the receiving part in the mobile contact with the splenial. Characters of the intramandibular joint, along with those resulting from a review of the cranial anatomy of the fossil snake Pachyrhachis from the basal Upper Cretaceous of Ein Jabrud, are used in a review of squamate interrelationships. The results corroborate macrostomatan affinities of Pachyrhachis and do not support the hy- pothesis that snakes originated from mosasauroids, a clade of marine varanoid squamates from the Cretaceous. Introduction Mosasauroids are a clade of fossil marine squa- mates related to extant monitor lizards. Their ear- liest fossil occurrence is in shallow marine de- posits of early Cenomanian age (lower Upper Cretaceous) of southern Europe. These stem- group taxa, variously referred to as Aigialosauri- dae and/or Dolichosauridae, remain relatively poorly known compared to later members of the clade, the Mosasauridae. The crown-group mo- sasaurs adopted fully pelagic habits and include species that were among the largest predators of the late Cretaceous seas. Mosasaurs became ex- tinct at the close of the Cretaceous. The intramandibular joint has played a promi- nent role in discussions of mosasauroid relation- ships with snakes ever since Cope (1869) com- mented on the ophidian affinities of his order Py- thonomorpha. In the Pythonomorpha, Cope (1869) included two families of mosasaurs, the Clidastidae and the Mosasauridae. The ophidian affinities of the Pythonomorpha were established by Cope ( 1 869) on the basis of similarities of den- tition, the suspension of the lower jaw, and intra- mandibular kinetics. More recently, cladistic support has been build- ing in support of a monophyletic clade Pythono- morpha that would include platynotan (varanoid) squamates and mosasauroids as well as snakes (Lee, 1997). Configuration of such a clade has been corroborated by the redescription of a fossil snake with hind limbs from the basal Upper Cre- taceous of the Middle East (Caldwell & Lee, 1997; Lee & Caldwell, 1998). Originally de- scribed by Haas (1979, 1980), the status of this fossil snake taxon, Pachyrhachis problematicus, remains problematic. Although already consid- ered by some to be the ideal fossil link between snakes and mosasauroids (Carroll, 1988), it was also noted that those characters that are snakelike in Pachyrhachis resemble relatively advanced (macrostomatan) snakes instead of more basal members of the group (Haas, 1979, 1980; Riep- FTELDIANA: GEOLOGY, N.S., NO. 43, MARCH 31, 2000, PP. 1-69 pel, 1994). This controversy is still alive, as a re- analysis of the cladistic relationships of Pachy- rhachis showed it to be the sister taxon of ma- crostomatan snakes rather than a primitive snake providing a link between this group and mosa- saurs (Zaher, 1998). Considering Pachyrhachis as the most primi- tive snake and "an excellent example of a tran- sitional taxon" (Scanlon et al., 1999) between mosasauroids and snakes (Lee, 1998) has impor- tant consequences, as this pattern of relationships suggests that snakes had a marine rather than ter- restrial (fossorial) origin. Shared derived charac- ters that have been used in support of a mono- phyletic Pythonomorpha recall Cope's (1869) analysis and were derived from braincase mor- phology and its relation to jaw suspension, lower jaw anatomy, and characters of the dentition (Lee, 1997; Lee & Caldwell, 1998). We have previously critically assessed the characters derived in these latter studies from squamate tooth implantation and replacement (Zaher & Rieppel, 1999) and from braincase morphology and its relation to jaw suspension (Rieppel & Zaher, in press). The intra- mandibular joint has traditionally been an impor- tant character in discussions of snake relationships (Camp, 1923). In their classic monograph, Mc- Dowell and Bogert (1954) compiled a large num- ber of characters in support of an anguimorph, or varanoid, relationship of snakes, among which the intramandibular joint figured prominently. Many of the characters enumerated by McDowell and Bogert (1954) came under severe criticism (Un- derwood, 1957), but anguimorph, or varanoid, re- lationships of snakes continued to be discussed (McDowell 1972; Schwenk, 1988; see also Riep- pel, 1988, for a review). Interestingly, the first large-scale cladistic analysis of squamate interre- lationships (Estes et al., 1988) did not provide strong support for anguimorph, or varanoid, re- lationships of snakes, which in this study were classified as Scleroglossa (all non-iguanian squa- mates) incertae sedis. However, parsimony anal- ysis of this data set put snakes close to fossorial or burrowing squamates such as dibamids and amphisbaenians (see also Rage, 1982). Whereas this latter hypothesis has recently gained further support from morphological evidence (Haller- mann, 1998), molecular data support anguimorph relationships for snakes (Forstner et al., 1995; Reeder, 1995). An as yet unpublished total evi- dence approach, combining molecular (DNA) and morphological data, unambiguously supported a ((Snake + dibamid) amphisbaenian) clade (Reed- er, 1995). As is true for every phylogenetic analysis, hy- potheses of relative relationships are only as good as the character evidence they are based on (Riep- pel & Zaher, in press). We propose to review, in this study, the lower jaw anatomy of varanoid squamates and snakes in detail, bearing in mind that superficial and potentially misleading resem- blances can result from two factors. One is that increased mobility in the lower jaw, as much as increased cranial kinesis in general, results from a reduction in bone overlap, which in turn is likely to result from paedomorphosis (assuming the aki- netic condition to be plesiomorphic; Irish, 1989). The other factor results from structural constraints in the development of an intramandibular joint. As Gauthier (1982, p. 46; see also Underwood, 1957, p. 25) pointed out, "some similarity is to be expected, especially since there is but one place in a squamate mandible where a mobile joint could form — between the dentary-splenial and the postdentary bones." This point is partic- ularly well borne out by comparison with the con- vergently differentiated intramandibular joint in Hesperornis, a fossil bird (Gregory, 1951; Gin- gerich, 1973). The splenial, for example, will al- ways show a reduced posterior extent in those taxa that develop an intramandibular joint (Estes et al., 1988). Reference to the intramandibular joint in the analysis of snake relationships will therefore have to transcend superficial similarities or mere reduction characters in order to reveal details of morphology. The characters of Lee (1997) will, in the following, be referenced as L97; the character evidence of Lee and Caldwell (1998) will be referenced as LC98. Materials and Methods The specimens examined for this study are list- ed below. Institutional abbreviations are bmnh, British Museum (Natural History); fmnh, Field Museum of Natural History; HUJ-Pal., Paleonto- logicai Collections, Hebrew University, Jerusa- lem. Drawings were made with a Wild binocular M-8 equipped with a camera lucida. Anilius scytale, fmnh 11175, 35688, uncata- logued Boa constrictor imperator, fmnh 22353, 22363 Calabaria reinhardti, fmnh 31372 FIELDIANA: GEOLOGY Charina bottae, fmnh 31300 Cylindrophis ruffus, fmnh 13100, 131780 Cylindrophis maculatus, bmnh 1930.5.8.48, un- catalogued Lanthanotus borneensis, fmnh 747 1 1 Leptotyphlops emini, fmnh 56374 Lichanura trivirgata roseofusca, fmnh 8043 Melanophidium punctatus, bmnh 1930.5.8.119 Pachyrhachis problematicus, HUJ-Pal. 3659 Platecarpus sp., fmnh UC 600 Platyplecturus madurensis, bmnh 1930.5.8.111 Plecturus perroteti, bmnh 1930.5.8.105 Pseudotyphlops philippinus, bmnh 1978.1092 Python reticulatus, fmnh 31281, 31329 Typhlops sp., fmnh 98952 Varanus komodoensis, fmnh 22199 Varanus sp., fmnh 195576 Xenopeltis unicolor, fmnh 11524 The Intramandibular Joint in Squamates The Intramandibular Joint in Varanus and Lanthanotus Among extant varanoids {Heloderma, Lantha- notus, and Varanus), the intramandibular joint shows various degrees of differentiation, least de- veloped in Heloderma, most developed in Lan- thanotus. In Varanus (Fig. 1), the posterior ends of the dentary and of the splenial lie entirely in front of the apex of the coronoid process (L97: char. 72). In lateral view, the posterior end of the dentary shows a more or less distinctly developed bicon- cave posterior margin. A smaller, dorsally located concavity or indentation receives the anterior tip of the coronoid. The broad and, in some species, shallow concavity below the coronoid-dentary contact broadly overlaps the anterior end of the surangular. The ventral margin of the dentary is drawn out into a short posterior process that over- laps with the anterior end of the angular. The prin- cipal element on which the dentary is supported is the surangular. In lateral view, the splenial and angula/ form a broadly overlapping, obliquely oriented contact in Varanus. More precisely, the tapering posterior end of the splenial superficially overlaps the broad anterior end of the angular and curves around its ventral margin (L97: char. 74; LC98: char. B12). Sutural relations between the dentary and splenial and the postdentary bones are more com- plex in medial view of the tooth-bearing shelf of the mandible of Varanus. The anterior tooth-bear- ing part of the dentary forms a gentle slope (a discrete subdental shelf is absent; L97: char. 67), which in front of the splenial projects ventrally, thus overhanging Meckel's canal in medial view. The anterior part of Meckel's groove opens ven- trally relative to the sagittal plane of the mandib- ular ramus, as it is defined by the lateral wall of the dentary and the medial tooth-bearing shelf (L97: char. 69; LC98: char. B13). The anterior end of Meckel's groove opens medioventrally in the live animal because the lower jaw is rotated around its long axis in such a way as to bring the tooth row into an upright position and to expose Meckel's cartilage medioventrally for the inser- tion of anterior intramandibular muscles. More posteriorly, the tooth-bearing shelf merges into the septum that separates Meckel's canal from the more dorsolateral ly positioned canal for the al- veolar ramus of the mandibular division of the trigeminal nerve. The ventral part of the posterior margin of this intramandibular septum is deeply concave. Its concavity defines the dorsal and an- terior margin of the anterior inferior alveolar fo- ramen, the posterior and ventral margin of which is defined by the splenial as it contacts the medial surface of the dentary. Above the posterior con- cavity of the septum, the dentary is broadly over- lapped by the splenial in medial view. Dorsal to the dentary-splenial overlap, the dentary forms a very short coronoid process defining a postero- ventral recess (notched in lateral view but not in medial view) into which fits the anterior dorsal tip of the coronoid. The splenial (Fig. 2C) itself is roughly of an arrowhead shape in medial view. A slender and pointed posteroventral projection overlaps the an- terior end of the angular (L97: char. 73; LC98: char. B12). A broad posterior dorsal projection provides the medial closure of Meckel's canal at the level of the anterior end of the surangular. Posterodorsally, the splenial contacts the anterior process of the coronoid (L97: char. 77; LC98: char. B14), anterodorsally it contacts the posterior end of the dentary, and posteriorly it defines the anterior margin of the subcoronoid fossa (L97: char. 79; LC98: char. Bl 1), below which it over- laps the anterior end of the prearticular. Anteri- orly, the splenial is drawn out into a tapering pro- cess that reaches to about the midpoint of the den- tary. Along the posterior two thirds of the length of the splenial, a horizontal shelf projects from RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES VII hy VII hy Fig. 1. The lower jaw of Varanus (based on Varanus komodoensis, fmnh 22199). A, lateral view; B, medial view; C, disarticulated medial view. Not to scale. Abbreviations for all figures: ale, alveolar nerve canal; amf, anterior mylohyoid foramen; an, angular; ar, articular; c, coronoid; cp, compound bone; d, dentary; mf, mental foramen; pi, palatine; pmf, posterior mylohyoid foramen; pra, prearticular; sa, surangular; saf, anterior surangular foramen; sp, splenial; mc (or Mc), Meckel's cartilage; mg (or Mg), Meckel's groove; VII hy, chorda tympani foramen. FIELDIANA: GEOLOGY B Fig. 2. The coronoid and splenial of Varanus sp. (fmnh 195576). A, coronoid, lateral view; B, coronoid, medial view; C, splenial, lateral view. Scale bar = 10 mm. the lateral surface of the splenial at a level just below the anterior mylohyoid foramen. This hor- izontal shelf underlies Meckel's cartilage and cre- ates a groove between itself and the laterally curved ventral margin of the splenial, which re- ceives the medially curved ventral margin of the dentary. The anterior mylohyoid foramen opens medially, but a small slitlike opening between the medial vertical wall of the splenial and its later- ally projecting shelf at the level of the anterior mylohyoid foramen allows a branch of the ante- rior mylohyoid nerve to pass into the groove that receives the ventral margin of the dentary. The splenial tapers off at the ventral margin of the dentary (L97: char. 70). The coronoid (Figs. 2A, 2B) shows a V-shaped outline in medial view, the apex pointing upward and forming the coronoid process. Its ventral mar- gin is concave and defines the subcoronoid fossa located between the coronoid and the prearticular. In transverse section, the coronoid forms an in- verted V, the apex pointing upward, and the base straddling the dorsal rim of the surangular. The main body of the coronoid carries a distinct an- teroventral process, the anterior tip of which fits into a recess at the posterior end of the dentary. Below and shortly behind the coronoid-dentary contact, the coronoid forms a medial sheet of bone that extends in an anteroventral direction deep (i.e., lateral) to the broad posterodorsal ex- tension of the splenial but medial to the suran- gular and to Meckel's cartilage. The coronoid en- ters the posterior margin of the anterior inferior alveolar foramen lateral to the splenial, where it becomes drawn out into two slender and delicate processes that follow the dorsal and ventral mar- gins, respectively, of this foramen. The dorsal process of the coronoid is more extensively de- veloped than the ventral one, as it follows the dor- sal margin of the anterior inferior alveolar fora- men medial to the dentary and ventral to the tooth-bearing shelf to the level of the midpoint of the dentary; the ventral projection of the coronoid reaches up to the midpoint of the lower margin of the anterior inferior alveolar foramen. At the posteroventral base of the coronoid pro- cess, the medial shank of the coronoid is drawn out into a posteroventral process, medially over- lapping the ascending process of the prearticular, which, together with the coronoid, forms the an- terior and medial margin of the adductor fossa (L97: char. 78; LC98: char. BIO). Because the me- dial margin of the adductor fossa is lower than the lateral margin in Varanus, the fossa is exposed in both dorsal and medial views of the lower jaw (L97: char. 80; LC98: char. B15). The prearticular forms the floor of the adductor fossa behind the angular and its medial wall deep to the coronoid. At the anterior margin of the ad- ductor fossa, the prearticular forms a dorsal pro- cess extending upward between the (lateral) sur- angular and the (medial) coronoid, thereby defin- ing the posterior margin of the subcoronoid fossa. The prearticular continues anteriorly below the subcoronoid fossa. Its anterior end slips beneath the posterodorsal extension of the splenial in me- dial view (i.e., passes lateral to the splenial). Be- low (i.e., lateral to) the splenial, the anterior end of the prearticular tapers to a thin and pointed process, located ventral to the anterior ventral pro- cess of the coronoid, which runs along the ventral margin of the anterior inferior alveolar foramen. In summary, the dentary and splenial, on the RTEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES B VII hy SP amf amf Fig. 3. The lower jaw of Lanthanotus borneensis (fmnh 1371 1). A, lateral view; B, medial view; C, angular and splenial, ventral view. A, B: scale bar = 5 mm; C: scale bar = 2 mm. Abbreviations as in Figure 1. one hand, and the postdentary bones on the other show a complex and extensive pattern of overlap with one another in Varanus, even though this overlap is not revealed by the superficial suture pattern. Lateral to Meckel's cartilage, the domi- nant overlap is between the dentary (superficial) and the surangular. Medial to Meckel's cartilage, the dominant overlap is between the splenial (su- perficial) and the coronoid plus prearticular. The lower jaw of Lanthanotus (Fig. 3) differs from that of Varanus by an extended contact of the coronoid with the dentary, in both lateral and medial views of the lower jaw. Unlike in Varanus, the anterior end of the coronoid is essentially bi- furcated, as it embraces the posterior end of the dentary both laterally and medially. The lateral anterior prong of the coronoid is shorter than the medial anterior prong, which extends anteroven- trally to establish a broad contact with the anterior end of the prearticular and the posterodorsal cor- ner of the splenial (L97: char. 77; LC98: char. B14). Behind that anterior bifurcation, the coro- noid straddles the longitudinal dorsal shoulder of the surangular, as it does in Varanus. A postero- ventral process of the coronoid descends on the medial side of the lower jaw, medially overlap- ping an ascending process of the prearticular, which itself forms the anterior and medial margin of the adductor fossa (rather than the coronoid itself; L97: char. 78; LC98: char. BIO) and, at the same time, the posterior margin of the subcoro- noid fossa. The surangular establishes a broad overlap with the dentary lateral to Meckel's cartilage; the pos- teroventral corner of the dentary also overlaps the anterior end of the angular laterally. The splenial gains no exposure in lateral view, and in medial view it shows a reduction of the posteroventral FIELDIANA: GEOLOGY Fig. 4. The lower jaw of Platecarpus sp. (fmnh UC 600) in lateral view. process, which in Varanus is distinct and overlaps with the angular. Reduction of this process results in a superficially vertical suture at the contact be- tween angular and splenial in medial view of the lower jaw. The posterodorsal extension of the splenial is again less developed than in Varanus, which reduces but does not obliterate the overlap with the anteroventral process of the coronoid. Whereas the vertical suture on the medial side of the lower jaw suggests a relatively simple and mobile contact between splenial and angular, the ventral aspect of the lower jaw (Fig. 3C) reveals that the angular-splenial contact is more complex (L97: char. 73; LC98: char. B12). The ventral mar- gin of the splenial is concave, receiving the convex ventral margin of the angular. More importantly, the angular forms a distinct anteroventral process, which extends lateral to the posterior end of the splenial, intercalated between the latter and the dentary (L97: char. 74; LC98: char. B12). As in mosasaurs (see below), the splenial is the receiving part, the angular the received part in the intraman- dibular articulation of Lanthanotus. Unlike in Var- anus, the anterior tip of the relatively short splenial lies dorsomedial to the ventral margin of the den- tary (L97: char. 70), and in front of it Meckel's canal opens ventrally relative to the sagittal plane of the lower jaw between a lateral flange of the dentary and the prominent tooth-bearing shelf (L97: char. 69; LC98: char. B13). The splenial of Lanthanotus was not disarticulated. It was there- fore not possible to assess the presence of a hori- zontal shelf projecting from the lateral surface of the splenial, which together with the ventral margin of the splenial would form a groove to receive the ventral margin of the dentary, as is seen in Var- anus. If present, however, such a medial crest must be confined to the posteriormost part of the splen- ial, as its anterior tapering end lies above the ven- tral margin of the dentary. The Intramandibular Joint in Mosasaurs The lower jaw of mosasaurs was described by Camp (1942) and, in more detail, by Russell (1967). Relevant information is also provided by Bell (1997). The main difference in the lower jaw of Varanus and mosasaurs is that the latter have mobilized the intramandibular joint to a greater degree, largely through a reduction of bone over- lap (probably due to paedomorphosis, which is common in secondary marine reptiles; Rieppel, 1993a). In lateral view, the posterior end of the dentary of mosasaurs {Platecarpus, fmnh UC 600) appears truncated relative to that of Varanus, with a more or less straight posterior margin that slopes posteroventrally (Fig. 4). The posteroven- tral corner of the dentary forms an extensive lat- eral overlap with the splenial. There is, however, no evidence for any significant anterior extension of the surangular deep to the dentary. Instead, the surangular appears to be truncated at its anterior end and consequently fails to overlap with the dentary. In Platecarpus (fmnh UC 600), the den- tary does not overlap with the surangular at all. The contact of the splenial with the angular again is not an overlapping one (Fig. 4). Instead, the two elements abut against each other in a ball- and-socket joint (Bell, 1997; Russell, 1967), which in lateral and medial views translates into a more or less vertically oriented contact between the two elements (L97: char. 74; LC98: char. B12). The posterior surface of the splenial is broadened and round or elliptical in outline. It is concave and forms the socket into which the RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES rounded and convex anterior surface of the an- gular fits. In the intramandibular joint of mosa- saurs, therefore, the splenial is the receiving part, the angular is the received part. In medial view, the lower jaw of mosasaurs again differs significantly from Varanus. As de- scribed previously (Zaher & Rieppel, 1999), the teeth are set in a groove that runs along the dorsal margin of the splenial. As a consequence, a den- tary shelf overhanging Meckel's canal is not dif- ferentiated in mosasaurs (L97: char. 67). Meckel's canal is open along the anterior third of the length of the mandibular ramus, but unlike in Varanus, it opens on the medial surface of the lower jaw because the dentary forms a ventromedial flange in its anterior half, which wraps around the ventral surface of Meckel's cartilage and rises up again on the medial side of the lower jaws (L97: char. 69; LC98: char. B13). More posteriorly, the splen- ial closes Meckel's canal in medial view. In front of its articulation with the angular, the splenial develops into a high flange of bone, almost com- pletely covering the medial exposure of the den- tary. The anterior inferior alveolar foramen is not conspicuous in mosasaurs. A broken lower jaw of an unidentified mosasaur (Mosasauridae indet. fmnh PR 674) shows that, unlike in Varanus, the splenial forms no longitudinal crest projecting from its lateral surface, which together with the ventral margin of the splenial would form a groove to receive the ventral edge of the dentary. The broken cross-section of the splenial reveals it to be a simple vertical lamina of bone. Its poste- rior part lies medial and ventral to the vertical lamina formed by the posterior end of the dentary. The ventral margin of the splenial becomes in- creasingly thickened posteriorly, as the bone forms the socket to receive the anterior head of the angular. Anteriorly, the splenial tapers to a blunt tip that terminates at a level above the ven- tromedial dentary flange that wraps around the ventral surface of Meckel's cartilage, i.e., not along the ventral margin of the dentary, as in Va- ranus (L97: char. 70), but on the medial surface of Meckel's cartilage. The coronoid is reduced in mosasaurs as com- pared to Varanus. A posteromedial process may still be present, but if so, it always remains small (Bell, 1997; L97: char. 78; LC98: char. BIO). In front of the apex of the coronoid process, the el- ement extends as a relatively broad anterior pro- cess that, as in Varanus, "is a saddle-shaped bone straddling the longitudinal 'shoulder' ... of the surangular which is enclosed in a deep sulcus" (Russell, 1967, p. 53). It approaches, but does not contact, the dentary, and it remains widely sepa- rated from the splenial (L97: char. 75; LC98: char. B14). The height of the prearticular is greatly in- creased in mosasaurs, and together with the cor- onoid the prearticular completely conceals the surangular in medial view of the lower jaw in front of the adductor fossa (L97: char. 79). As noted by Gauthier (1982), the prearticular is the principal element from which the dentary is sus- pended in mosasaurs, as it extends anteriorly as a high blade that enters between the equally high posterior part of the splenial (medially) and the dentary (laterally). As in Varanus, the adductor fossa of mosasaurs is characterized by a medial margin (formed by the prearticular) that is lower than the lateral mar- gin (formed by the surangular). As a consequence, the adductor fossa opens dorsally as well as me- dially (L97: char. 80; LC98: char. B15). By comparison to Varanus, mosasaurs have lost the dentary-coronoid contact, lost the broad overlap of surangular and dentary lateral to Meck- el's cartilage, and transformed the overlapping splenial-angular contact into a ball-and-socket joint. Medial to Meckel's cartilage, mosasaurs have lost the splenial-coronoid overlap due to a reduction of the coronoid, but the relative height of the splenial and prearticular is increased. In summary, mosasaurs have increased one area of support, the prearticular-splenial-dentary overlap, at the expense of two other areas of support that are well developed in Varanus, the splenial-cor- onoid overlap and the dentary-surangular overlap. The increased mobilization of the intramandi- bular joint in mosasaurs is correlated with a loos- ening of the mandibular symphysis. There is no sutural contact between the anterior tips of the den- taries. Instead, their anterior tips are smooth and rounded, and the dentaries must have been in syn- desmotic or ligamentous connection with each oth- er (Cope, 1869; L97: char. 68; LC98: char. B8). The Compound Bone of the Ophidian Mandible In snakes, the surangular, prearticular, and ar- ticular fuse to form a single "compound" or "mixed" bone during embryonic development (Bellairs & Kamal, 1981; DeBeer, 1937), incor- porating both dermal and chondral elements. Among other squamates, a similar compound FIELDIANA: GEOLOGY bone is only found in dibamids (Greer, 1985; Rieppel, 1984a) and in amphisbaenians (Montero et al., 1999; Zangerl, 1944). Many details of the embryonic development of the lower jaw of snakes remain to be determined, but it is clear that the elements contributing to the compound bone fuse during embryonic development (Bellairs & Kamal, 1981; Parker, 1879). Few authors had crit- ical embryonic stages available to them, which would show the dermal elements of the compound bone present but not yet fused (Brock, 1929; Hal- uska & Alberch, 1989; Kamal et al., 1970; Peyer, 1912); or consideration of the development of the dermal bones of the lower jaw was not included in the study (Genest-Villard, 1966). Several stud- ies of cranial development in snakes do not ad- dress the ossification sequence and pattern of der- mal bones (see reviews in Bellairs & Kamal, 1981; Rieppel, 1993b). However, Backstrom (1931) described in detail the development of the dermal bones in the lower jaw of Natrix natrix and noted that the first ele- ments to appear are the dentary and splenial, fol- lowed by the surangular. All five dermal elements are present in the 6.8-mm stage (Backstrom, 1931, Fig. 17) but are still separate from one another. The prearticular is confined to the medial aspect of the lower jaw (of Meckel's cartilage). It entire- ly conceals the surangular in medial view, and wraps around the ventral surface of Meckel's car- tilage below the mandibular articulation. The sur- angular develops on the lateral and dorsal aspect of Meckel's cartilage, and it is the surangular that forms the large anterior projection that enters be- tween the two posterior prongs of the deeply bi- furcated dentary. This interpretation is supported by the position of the homologue of the anterior surangular foramen (at least in alethinophidians), and it is also in accordance with the observation of Estes et al. (1970) that it is the surangular por- tion of the compound bone that provides the main support for the dentary in the fossil snake Dini- lysia. In the adult jaw, the compound bone wraps around Meckel's cartilage. It may be raised into a coronoid process toward its anterior end, and it carries the adductor fossa in its posterior part. The relation of Meckel's cartilage to the adductor fos- sa differs in important ways in snakes as com- pared to nonophidian squamates. In squamates other than snakes, Meckel's cartilage is exposed at the bottom of the adductor fossa for approxi- mately half of its length, and fibers of the poste- rior adductor insert into it. The compound bone of scolecophidians appears to be a simple tubelike structure, at least in its posterior part. In alethin- ophidians (at least in anilioids and basal macros- tomatans), Meckel's cartilage enters its own canal at the bottom of the compound bone, beginning at the level of the anterior margin of the adductor fossa, and hence is not exposed in the latter. The coronoid is of a much simpler structure in snakes than in nonophidian squamates. It never forms a saddle-shaped structure straddling the dorsal lon- gitudinal shoulder of the surangular. Instead, the coronoid of snakes is a simple sheet of bone that is always applied to the medial side of the dorsal (surangular) portion of the compound bone. It may or may not project beyond the dorsal margin of the compound bone at the apex of the coronoid process. The formation of a compound bone is an autapomorphy of snakes (or a potential synapo- morphy shared by snakes, dibamids, and amphis- baenians), and because of its formation, a subco- ronoid fossa exposing the surangular between the coronoid and the angular on the medial surface of the lower jaw is a character that cannot be applied or compared to snakes (L97: char. 79; LC98: char. Bll). The Intramandibular Joint in Scolecophidians Scolecophidians are divergently derived and highly autapomorphic in the structure of their lower jaw, owing to their microphagous habits. In Anomalepis (Haas, 1968), the surangular, articu- lar, and prearticular have fused into a compound bone, as in all other snakes. The coronoid is roughly trapezoidal and forms a prominent coro- noid process. It is applied against the medial sur- face of the anterior part of the compound bone and medially overlaps the contact between the posterior end of the dentary and the anterior end of the splenial (List, 1966; angular of Haas, 1968). The anterior end of the compound bone overlaps and thereby supports the dorsal margin of the posterior half of the splenial, which with its anterior half (i.e., in front of the compound bone) underlies and supports the posterior end of the dentary. The posterior part of the dentary meets the dorsal margin of the anterior half of the angular in an oblique plane (L97: char. 73; LC98: char. B12). Meckel's cartilage is wedged between the coronoid medially and the splenial and den- tary laterally. In front of the coronoid, Meckel's cartilage comes to lie in a shallow groove that follows the ventromedial margin of the dentary. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES The splenial is a simple, flat strip of bone that tapers off along the ventral margin of the dentary (relative to the sagittal plane of the lower jaw). Main support for the dentary is provided by the splenial and by the coronoid in front of the com- pound bone. Liotyphlops closely resembles Anomalepis in the essential features of the lower jaw, although the shapes of individual elements are different in detail (Haas, 1964). The simple and flat splenial (angular of Haas, 1964) is again underlying the anterior end of the compound bone and the pos- terior end of the dentary. At the anterior end of the splenial, the Meckelian groove shifts to a po- sition on the medial side of the dentary (Haas, 1964, Fig. 20). The splenial tapers off along the ventral margin of the dentary (relative to the sag- ittal plane of the lower jaw). The lower jaw of Typhlops typically comprises five elements (Haas, 1930). The largest one is the compound bone, the anterior end of which is again applied against the lateral surface of the large, roughly triangular coronoid. The tapering anterior end of the compound bone reaches the posterior tip of the dentary and may be accom- panied up to this level by the anteroventral pro- cess of the coronoid. As in anomalepids, a rela- tively large splenial underlies the contact of the compound bone with the dentary, but in typhlo- pids it extends anteriorly up to the symphyseal tip of the mandible. At the level of the anterior end of the compound bone, the posterior part of the splenial is located ventral to Meckel's cartilage; in front of the compound bone, the splenial forms a deep trough closed dorsally by the dentary, within which lies Meckel's cartilage. Meckel's ca- nal (groove) is therefore not exposed on the me- dial side of the lower jaw (L97: char. 69: LC98: char. B13). Other than in anomalepids, most ty- phlopids show a second element at the ventral margin of the lower jaw, sometimes vestigial, and located behind the splenial, which supports the dentary. The posterior element is identified as an- gular (List, 1966), but, as in anomalepids, the main support for the dentary is provided by the splenial in typhlopids. In mosasaurs, it is the prearticular that provides principal support for the dentary; in alethinophidians (and leptotyphlo- pids), it is the surangular as part of the compound bone (Estes et al., 1970, p. 46; see below). In nonophidian squamates, it is the splenial that un- derlies the dentary, and if so interpreted, anoma- lepids and typhlopids share a plesiomorphic char- acter in that respect. However, the posterior ex- tension of the splenial to a level well behind the level of the posterior tip of the dentary may be a synapomorphy shared by typhlopids and anoma- lepids, whereas the position of the splenial entire- ly lateral to Meckel's cartilage may be a synapo- morphy of anomalepids. Brock ( 1 932) described five bones in the lower jaw of Leptotyphlops, i.e., the dentary, splenial, coronoid, angular, and the compound bone. The splenial is located entirely on the medial side of the dentary, partially closing the posterior part of the Meckelian groove, which is located entirely on the medial aspect of the dentary. The dentary is a relatively large, tooth-bearing element, with a sloping posterior margin that establishes an ex- tended and mobile (Haas, 1930) contact with the compound bone. The latter is shorter and more massively built than in other scolecophidians. An angular underlies the anterior end of the com- pound bone, which may meet the splenial in a simple abutting contact if the latter projects be- yond the posteroventral corner of the dentary (List, 1966). The coronoid is applied to the medial surface of the compound bone and remains widely separated from the splenial. Unlike in anomale- pids and typhlopids, it is the compound bone that provides the major support for the dentary in lep- totyphlopids. This is also the case in alethinophi- dians, although in this group, the contact of the compound bone with the dentary is established in a different manner. The Intramandibular Joint in Anilioids Anilius shows a weak expression of the poste- rior bifurcation of the dentary, which is charac- teristic of alethinophidian snakes (Fig. 5). Posterodorsally, the dentary is drawn out into a short posterodorsal process, which together with the coronoid and the compound bone forms the prominent coronoid process. Medially, the poste- rior end of the dentary forms a broad concavity that accommodates the anterior end of the com- pound bone. The mandibular division of the tri- geminal nerve, along with Meckel's cartilage, is enclosed in a canal within the compound bone. Further anteriorly, the compound bone opens me- dially, releasing Meckel's cartilage along with the mandibular nerve into Meckel's groove on the medial side of the dentary. The alveolar ramus of the mandibular division of the trigeminal nerve enters a separate, dorsolaterally positioned canal that leads up to the single mental foramen, which 10 FIELDIANA: GEOLOGY Fig. 5. The lower jaw of Anilius scytale (fmnh 35688). A, lateral view; B, medial view. Scale bar = 2 mm. Abbreviations as in Figure 1. opens at the anterior end of the dentary on its lateral surface (L97: char. 76; LC98: char. C14). Meckel's canal opens ventromedially in its pos- terior part as the ventral rim of the dentary ex- pands medially below Meckel's cartilage at the level below the anterior end of the compound bone. More anteriorly, however, Meckel's canal opens ventrally (relative to the sagittal plane of the lower jaw), as it does in Varanus and other nonophidian squamates except mosasaurs (L97: char. 69; LC98: char. B13). The dorsomedial ledge of the dentary that overhangs the Meckelian groove is closely comparable to the tooth-bearing shelf of Varanus (L97: char. 67). The compound bone of Anilius includes, as it does in other snakes, the articular, surangular, and prearticular. It encloses an elongate, deep and wide adductor fossa with a well-defined medial margin. As a consequence, the adductor fossa opens dorsally (L97: char. 80; LC98: char. B15). Deep to the dentary, the part of the compound bone located lateral to Meckel's cartilage extends further anteriorly than the medial cover of Meck- el's cartilage, which results in a medial opening of Meckel's canal at the anterior end of the com- pound bone. The part of the compound bone lo- cated lateral to Meckel's cartilage corresponds to the surangular, and its greater anterior extension corroborates the observation of Estes et al. (1970) in Dinilysia and the embryological observations of Backstrdm (1931) that, in alethinophidians, it is the surangular that provides the principal sup- port for the dentary. At the anterior end of the compound bone, be- tween the ventral margin of the latter and the pos- terior maxillary ledge that wraps around the ven- tral surface of Meckel's cartilage, lies a small splint of bone. It projects anteriorly to a level slightly in front of the medial component of the compound bone, and it is located ventromedial to the compound bone and, in front of the latter, ven- tromedial to Meckel's cartilage. By comparison to other basal alethinophidians, this element is per- haps best interpreted as a vestigial angular, al- though it could also represent a vestigial splenial. Its presence was confirmed in both the lower jaw ramus of Anilius fmnh 35688 (Fig. 5) and in a serially sectioned skull (Fig. 6, uncatalogued specimen). The coronoid is a small element located at the tip of the coronoid process on the medial surface of the compound bone and behind the postero- dorsally ascending process of the dentary. The bone is truncated posteriorly but carries a short RTEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 11 ac?, sp? Fig. 6. Transverse section through the lower jaw of Anilius scytale (uncatalogued). Abbreviations as in Fig- ure 1. anteroventral projection, which results in a weakly concave ventral margin. Given the fusion of the surangular and prearticular (and articular) to form the compound bone, a subcoronoid fossa exposing the surangular in medial view is absent in snakes (L97: char. 79; LC98: char. Bll). The intramandibular joint takes on a more com- plex structure in Cylindrophis ruffus (Fig. 7, fmnh 13100, 131780). The function of the intramandi- bular joint in Cylindrophis ruffus has been ana- lyzed in detail by Cundall (1995). The dentary is deeply bifurcated posteriorly, supported by a large anterior projection of the compound bone. As de- scribed by Zaher and Rieppel (1999), the teeth are ankylosed to the interdental ridges on the pleura of the dentary. At the anterior end of the dentary, this pleura is again developed into a tooth-bearing shelf (again without marginal thickening) that overhangs the anterior end of Meckel's groove (L97: char. 76). The latter opens ventrally relative to the sagittal plane of the mandibular ramus (L97: char. 69; LC98: char. B13). More posteri- orly, the tooth-bearing shelf is developed into the intramandibular septum with a concave medial surface, separating the medial Meckelian groove from the laterally positioned canal for the alveolar ramus of the mandibular division of the trigeminal VII hy Fig. 7. The lower jaw of Cylindrophis ruffus (fmnh 131780). A, lateral view; B, medial view. Scale bar = 2 mm. Abbreviations as in Figure 1. 12 FIELDIANA: GEOLOGY nerve. Anteriorly, the nerve emerges from the sin- gle, laterally placed mental foramen at the tip of the dentary (L97: char. 76; LC98: char. CI 4). The intramandibular septum reaches backward into the anterior half of the gap between the dorsal and ventral prongs of the posterior end of the dentary. The large, laterally placed anterior prong of the compound bone (surangular portion, see above) thus fits snugly into a recess bounded dorsally by the dorsal prong of the caudally bifurcated den- tary, ventrally by the posteroventral process of the dentary, and medially by the intramandibular sep- tum. Below this contact between the compound bone and the dentary, the vertically oriented suture be- tween angular and splenial is narrowly exposed in lateral view (L97: char. 74). The medial view of the lower jaw exposes the splenial and angular in their full size, pierced by the anterior and poste- rior mylohyoid foramen, respectively. From the lateral surface of the splenial, at a level narrowly below the anterior mylohyoid foramen, projects a lateral crest. This crest underlies Meckel's carti- lage, and, with the ventral edge of the splenial, it forms a distinct groove that receives the ventral edge of the dentary, which itself curves inward. At the level of the anterior mylohyoid foramen, the lateral shelf of the splenial is pierced by a small foramen, which allows a ventral branch of the anterior mylohyoid nerve to slip out of Meck- el's canal, entering between the splenial and the dentary. Medially, the splenial forms a dorsal ver- tical blade that entirely closes the posterior part of Meckel's groove in medial view. More anteri- orly, the vertical blade of the splenial gradually tapers to a pointed tip that lies in line with the ventral edge of the dentary (L97: char. 70). The coronoid is a roughly triangular bone that is received in a shallow facet on the medial side of the compound bone (surangular portion). To- gether with the compound bone, it forms a prom- inent coronoid process. The ventral margin of the coronoid is slightly concave. Anteroventrally, the coronoid is extended into a prominent anterior process that remains restricted to the medial sur- face of the compound bone and hence does not participate in the formation of the large surangular prong that enters between the two posterior pro- cesses of the dentary. The anteroventral process of the coronoid establishes an extended contact with the anterior part of the dorsal margin of the angular, but it remains separated from the splenial in our specimens of Cylindrophis ruffus (L97: char. 75; LC98: char. B14). The coronoid contacts the splenial in specimens figured by McDowell (1975, Fig. 6) and Cundall (1995, Fig. 6). (It should be noted that macerating skulls with com- mercial bleach easily dissolves thin marginal ar- eas of bones.) In superficial medial view, the anterior end of the angular matches the posterior end of the splen- ial in height. The two elements meet in a straight, slightly posteroventrally trending suture (L97: char. 73; LC98: char. B12). Disarticulation of the splenial shows that the posterior surface of the posteroventral corner of the splenial is thickened, flat, and sloping posteroventrally, whereas the an- terior surface of the angular is similarly thickened, flat, and trending anterodorsally. The two ele- ments meet face to face in a simple abutting con- tact. Deep to this contact, the dorsomedial edge of the angular forms a small, anteriorly projecting prong that, on its medioventral surface, is lined by a congruent projection of the compound bone. This latter prong originates from the compound bone medioventral to Meckel's cartilage. As de- scribed above, the anterior end of the angular is somewhat thickened, and, in addition to the com- posite medial prong described above, its lateral margin slightly projects anteriorly. As a conse- quence thereof, the anterior surface of the angular forms a vertically oriented shallow trough that is bounded dorsomedially by the composite prong described above and laterally by the projecting lateral margin of the angular. The posterior end of the splenial fits snugly into that trough on the an- terior surface of the angular, while the horizontal crest that projects from the lateral surface of the splenial locks against the ventral surface of the composite dorsomedial prong. Other than in mo- sasaurs, therefore, the angular is the receiving part in the intramandibular articulation, the splenial is the received part. Posteriorly the angular tapers to a blunt tip, which is located at the ventral margin of the compound bone. Behind the coronoid process, the compound bone forms the adductor fossa, which is not as deep and wide as it is in Anilius. However, its medial margin is only slightly lower than the lat- eral margin, such that the adductor fossa opens dorsally relative to the sagittal plane of the man- dibular ramus (L97: char. 80; LC98: char. B15). Cylindrophis maculatus (Fig. 8, bmnh 1930.5.8.48; serially sectioned skull, uncata- logued) is closely comparable to Cylindrophis ruf- fus in most of the essential characteristics of the lower jaw. The coronoid process is somewhat lower, but the anteroventral process of the coro- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 13 VII hy SP amf Pmf B cp ' ^^^* Mc Fig. 8. The lower jaw of Cylindrophis maculatus (bmnh 1930.5.8.48; serially sectioned skull, uncatalogued). A, medial view, scale bar = 2 mm; B, serial sections from the coronoid process to the anterior tip of the splenial. Abbreviations as in Figure 1. 14 FIELDIANA: GEOLOGY noid is larger and approaches the splenial very closely or establishes a contact with it. Also, the superficial splenial-angular suture as exposed in medial view is distinctly concave anteriorly in its ventral part, a trait that is only weakly developed in Cylindrophis ruffus. The anterior prong pro- jecting from the medial anterior margin of the an- gular, and complemented medioventrally by an anterior prong of the compound bone, is also pre- sent in Cylindrophis maculatus, although perhaps a little less prominently developed than in Cylin- drophis ruffus. This prong is located at the dorsal margin of the concavity formed by the anterior end of the angular, such that the splenial-angular contact is closely comparable to that described for Cylindrophis ruffus. The lower jaw of uropeltids (Figs. 9, 10) resem- bles that of Cylindrophis quite closely except for characters that result from reduction, probably as a consequence of paedomorphosis related to small overall size. The dentary of Melanophidium punc- tatum (Figs. 9A, 9B: bmnh 1930.5.8.119) retains the deep posterior bifurcation characteristic of al- ethinophidians, but the posteroventral process of the dentary is much reduced in Platyplecturus madurensis (Figs. 9C, 9D: bmnh 1930.5.8.111), vestigial in Plecturus perroteti (Fig. 10: bmnh 1930.5.8.105), and fully reduced in all other spe- cies examined (Pseudotyphlops philippinus, bmnh 1978.1092 [Figs. 9E, 9F]; Rhinophis drummond- hayi, bmnh 1930.5.8.67-68; Teretrurus rhodogas- ter, bmnh 1930.5.8.98, and Uropeltis woodman- soni, bmnh 1930.5.8.73-74). The hypothesis that the posteroventral process of the dentary is re- duced rather than absent within uropeltids re- quires corroboration by reconstruction of cladistic relationships within the clade, which is not avail- able at this time. However, Melanophidium punc- tatum is also plesiomorphic with respect to some other characters of its cranial anatomy, such as the presence of teeth on the palatine, the location of the optic foramen, and the retention of sutures in the occipital condyle delineating the basioccipital from the exoccipitals (Rieppel, 1977). The coronoid, an element of variable size and shape, is always applied to the medial surface of the compound bone and, together with the latter, forms a weakly expressed coronoid process. The coronoid retains relatively distinct anteroventral and posteroventral processes and a concave ventral margin in Plecturus perroteti (bmnh 1930.5.8.105) and, to a lesser degree, in Platyplecturus maduren- sis (bmnh 1930.5.8.111). The coronoid usually contacts the posterodorsal process of the dentary and at least marginally projects beyond the dorsal margin of the compound bone, except in Pseudo- typhlops philippinus (bmnh 1978.1092), where the coronoid is vestigial, restricted to the medial sur- face of the compound bone, and has lost the con- tact with the dentary. The coronoid never contacts the splenial in uropeltids. In taxa with a reduced posteroventral process of the dentary, the posterior end of this bone wraps around the anterior end of the compound bone in a manner very similar to that observed in Anilius. Meckel's groove is open medially in front of the anterior end of the compound bone except in those taxa where the dorsal margin of the splenial establishes a contact, in its posterior part, with the ventral margin of the tooth-bearing shelf of the dentary. Meckel's groove opens on the me- dial surface of the mandible above the splenial, but it opens ventrally (relative to the sagittal plane of the lower jaw ramus) in front of the splenial, with the exception of Plecturus perroteti (bmnh 1930.5.8.105), where the ventral opening of Meckel's groove is restricted to its anterior end. All uropeltids have a well-developed splenial and angular, each pierced by the anterior and pos- terior mylohyoid foramen respectively (Figs. 9, 10). The elements meet in a slightly curved, an- teriorly concave, but essentially vertically orient- ed suture apparent on the medial surface of the lower jaw. This suggests a similar articulation of angular and splenial, as is also observed in Cylin- drophis maculatus. The disarticulated lower jaw of Pseudotyphlops philippinus (bmnh 1978.1092), as well as a serially sectioned head of Plecturus perroteti (uncatalogued), reveals that the posterior surface of the posteroventral end of the splenial is somewhat thickened, as is the anterior surface of the anteroventral end of the angular. The two elements meet in a simple abutting contact, the splenial with a weakly convex surface, the angular with a weakly concave surface. The composite medial prong formed by the angular and the com- pound bone and locking the splenial in place in Cylindrophis is absent in uropeltids, which there- fore are characterized by a somewhat simplified articulation between angular and splenial. The serially sectioned head of Plecturus per- roteti (Fig. 10C) shows the posterior mylohyoid nerve leaving Meckel's canal through a small slit- like aperture in the anterior ventral margin of the compound bone to reach the posterior mylohyoid foramen in the angular. In transverse sections, this creates the impression that the compound bone is drawn out into two short anteroventral processes. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 15 VII hy VII hy Fig. 9. The lower jaw of Melanophidium punctatus (bmnh 1930.5.8.119: A, lateral view; B, medial view; scale bar = 1 mm), Platyplecturus madurensis (bmnh 1930.5.8.111: C, lateral view; D, medial view; scale bar = 1 mm), and Pseudotyphlops philippinus (bmnh 1978.1092: E, lateral view; F, medial view; scale bar = 2 mm). Abbreviations as in Figure 1. 16 FIELDIANA: GEOLOGY 4 V* -pmf Mc sp sp sp SP Mc Fig. 10. The lower jaw of Plecturm perroteti (bmnh 1930.5.8.105; serially sectioned skull, uncatalogued). A, lateral view; B, medial view; C, serial sections from the coronoid process to the anterior tip of the splcnial. Scale bar = 1 mm. Abbreviations as in Figure 1. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 17 These do not, however, bridge the angular-splen- ial contact. As in Cylindrophis, the splenial de- velops a horizontal crest that projects from its lat- eral surface at a level closely below the anterior mylohyoid foramen and that underlies Meckel's cartilage. This crest, together with the ventral edge of the splenial, again forms a groove that receives the ventral edge of the dentary. A small foramen located next to the anterior mylohyoid foramen allows a ventral branch of the anterior mylohyoid nerve to pass through the horizontal crest of the splenial and to continue anteriorly be- tween the latter and the dentary. The anterior tip of the splenial again lies at the ventral margin of the dentary. The disarticulated mandible of Pseu- dotyphlops philippinus (bmnh 1978.1092) reveals the same splenial morphology. The adductor fossa on the compound bone is elongate and deep in Melanophidium punctatum (bmnh 1930.5.8.119); this character again seems to represent the plesiomorphic condition, as it is closely comparable to Cylindrophis. In other uro- peltids, the adductor fossa tends to be reduced to a shallow and relatively short groove located on the anterior half of the bone, behind the coronoid process, and showing a foramen at its anterior and posterior end for the passage of the mandibular division of the trigeminal nerve. The adductor fossa opens dorsally and medially in Melanophi- dium punctatum (bmnh 1930.5.8.119), where it is well developed, but it opens dorsomedially or me- dially in those taxa where it is reduced to a shal- low groove. The Intramandibular Joint in Basal Macrostomatans The lower jaw of Xenopeltis is highly special- ized to allow for extreme mobility (Frazzetta, 1999; McDowell, 1975; Rieppel, 1977). The den- tary is short relative to the much elongated com- pound bone. It is deeply bifurcated posteriorly and carries a much elongated posterior dentigerous process. A sliverlike coronoid is attached to the dorsal surface of the compound bone in front of the adductor fossa, but it does not participate in the formation of a coronoid process (Hoge, 1964). The anterior end of the compound bone enters be- tween the two posterior prongs of the dentary. A small angular and an elongate and pointed splen- ial are applied to the medial surface of the anterior end of the compound bone. There is no mobile contact between angular and splenial. Indeed, Xenopeltis differs from other snakes in that the intramandibular joint lies between the dentary and the compound bone, angular and splenial being parts of the functional unit represented by the compound bone (Frazzetta, 1999). No lower jaw of Loxocemus hicolor was avail- able for disarticulation, which renders it impos- sible to comment on the internal structure of the articulation of the splenial with the angular. In its superficial structure, however, the mandible of Loxocemus resembles that of other basal alethin- ophidians and/or macrostomatans (McDowell, 1975; Rieppel, 1977). The dentary is deeply bi- furcated posteriorly and carries the elongate pos- terior dentigerous process characteristic of ma- crostomatan snakes. The coronoid is a relatively small sliver of bone that is applied to the medial side of the compound bone, lining the anterior margin of the coronoid process and contacting the dentary anteriorly. The splenial and angular meet in an abutting contact; the superficial suture on the medial surface of the lower jaw is anteriorly concave. An anterior mylohyoid foramen is ab- sent in the splenial. The anterior mylohyoid nerve passes through a notch in its dorsal margin, as is also the case in Python (see below). The coronoid narrowly approaches, but does not contact, the splenial. Meckel's groove opens medially along the splenial but ventrally (relative to the sagittal plane of the mandibular ramus) in front of the splenial. The adductor fossa is elongate and well developed, with a medial margin that is lower than the lateral margin such that the fossa opens medially and dorsally. The intramandibular joint of Python reticulatus (Fig. 11, fmnh 31329) corresponds in its essential traits to that of Cylindrophis, although it is some- what more elaborate. In lateral view, the dentary appears deeply bifurcated, with an elongate pos- terior dentigerous process overlapping the anterior lateral prong of the compound bone. The presence of such an elongate posterior dentigerous process is a synapomorphy of macrostomatan snakes (Zaher, 1998). Disarticulation of the mandible (Fig. 11) reveals an intramandibular septum with a concave medial surface, separating the medially open Meckelian groove from the laterally posi- tioned canal for the alveolar nerve. This canal opens anteriorly through the single mental fora- men on the lateral surface of the anterior tip of the dentary (L97: char. 76; LC98: char. C14). The intramandibular septum reaches relatively further back into the posterior bifurcation of the dentary than in Cylindrophis and, together with the pos- terior dentigerous process and the posteroventral 18 FIELDIANA: GEOLOGY B eP D Fig. 1 1. The lower jaw of Python reticulatus (fmnh 31329). A, lateral view of postdcntary bones; B, medial view of postdentary bones; C, splenial, medial view; D, dentary, lateral view; E, splcnial, lateral view. A-D, scale bar = 2 mm; E, scale bar = 5 mm. Abbreviations as in Figure 1. process of the dentary, forms a deep recess that receives the prominent anterior and lateral prong of the compound bone (surangular portion). The medial view of the dentary exposes the Meckelian groove, which opens medially in its posterior part as the dentary turns inward below Meckel's cartilage but ventrally (relative to the sagittal plane of the mandibular ramus) in front of the splenial (L-97: char. 69; LC98: char. B13). The tall but relatively narrow posterior portion of RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 19 the splenial covers only the posteriormost part of the Meckelian groove in medial view. A short and tapering process of the splenial follows the dorsal margin of Meckel's groove, whereas a much lon- ger anteroventral process of the splenial follows the ventral margin of Meckel's groove, tapering off along the ventral margin of the dentary at a more anterior level than is characteristic of ani- lioids (L97: char. 70). The horizontal lateral shelf that projects from the lateral surface of the splen- ial is set very low, close to the ventral margin of the element (Fig. HE). It underlies Meckel's car- tilage and, together with the ventral margin of the splenial, forms a groove that receives the medially curved ventral margin of the dentary. Posteriorly, the horizontal shelf gradually tapers off in front of the thickened posteroventral head of the splen- ial. Again, the lateral horizontal crest of the splen- ial is pierced by a foramen for the passage of a ventral branch of the anterior mylohyoid nerve, which comes to lie between splenial and dentary. An anterior mylohyoid foramen was not observed in Python reticulatus (fmnh 31329); the nerve passed through a notch on the posterior dorsal margin of the splenial, as is also the case in Py- thon sebae and Epicrates cenchris (Frazzetta, 1959). The relatively large and L-shaped coronoid is applied to the medial surface of the compound bone, together with which it forms the prominent coronoid process. All traces of a posteroventral process of the coronoid have disappeared, and its ventral margin therefore is straight (L97: char. 79; LC98: char. BIO). The anteroventral process, however, is well developed and broadly contacts the dorsal margin of the angular. Anteriorly, this process of the coronoid is developed into an an- terodorsally pointing spur that overlaps the pos- terodorsal corner of the splenial (L97: char. 75; LC98: char. B14). Superficially, the angular meets the splenial in a vertical suture (L97: char. 74), which on the medial margin of the mandible appears slightly convex anteriorly (L97: char. 73; LC98: char. B12). Disarticulation of the dentary and splenial reveals two anterior prongs or processes of the compound bone that, together with the anteriorly projecting lateral margin of the angular, hold the splenial in place. The larger of these processes is located dorsomedially to Meckel's cartilage, pro- jecting from the prearticular portion ventral and deep to the anterodorsally projecting tip of the coronoid. The smaller ventral process originates from that portion of the compound bone that is located ventromedial to Meckel's cartilage, and it corresponds to the single anterior process of the compound bone seen in Cylindrophis. In Python, the ventromedial anterior process of the com- pound bone is not complemented by an antero- medial prong of the angular, as it is in Cylindro- phis. The two medial processes of the compound bone together with the laterally projecting edge of the angular define a deep, essentially vertically oriented trough into which slides the posterior margin of the vertical lamina of the splenial. The ventral portion of the anterior part of the angular is broadened and forms a distinct socket on its anterior surface. Into this socket fits the equally thickened posteroventral head of the splenial. The intramandibular joint as a whole has reached a greater level of complexity in Python as compared to Cylindrophis or uropeltids. The adductor fossa forms a deep and wide trough on the compound bone, with well-defined lateral and medial margins. Of these, the medial margin is lower than the lateral margin, such that the adductor fossa opens dorsally and medially (L97: char. 80; LC98: char. B15). The lower jaw of basal erycines rather closely matches the pattern established for Python. Li- chanura trivirgata roseofusca (Fig. 12, fmnh 8043) and Charina bottae (Figs. 13C, 13D: fmnh 31300), but not Calabaria reinhardti (Figs. 13 A, 13B: fmnh 31372), show the development of a distinct posterior process from the posterodorsal corner of the intramandibular septum that sepa- rates Meckel's groove from the canal for the su- perior alveolar nerve. This process is only vesti- gial in Python and Calabaria (Figs. 13 A, 13B), but in Lichanura (Fig. 12) and Charina (Figs. 13C, 13D), it is distinct and comes to lie in a well- delineated groove or facet on the medial aspect of the anterior prong of the compound bone (suran- gular portion) that supports the dentary, thus add- ing to the firmness of the dentary suspension. Only the lower jaw of Lichanura was disarticu- Fig. 12. The lower jaw of Lichanura trivirgata roseofusca (fmnh 8043). A, lateral view; B, medial view; C, postdentary bones, medial view; D, splenial, medial view; E, dentary, medial view; F, splenial, lateral view. Scale bar = 2 mm. Abbreviations as in Figure 1. 20 FIELDIANA: GEOLOGY B VII hy amf RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 21 B D Fig. 13. The lower jaw of Calabaria reinhardti (fmnh 31372: A, lateral view; B, medial view) and of Charina bottae (fmnh 31300: C, lateral view; D, medial view). Scale bar = 2 mm. Abbreviations as in Figure 1. lated (Fig. 12), and it revealed some differences in the angular-splenial contact as compared to Py- thon. As in the latter genus, an anterior medial prong originates from the compound bone in a position located dorsomedial to Meckel's cartilage and medial to the surangular portion, distinctly projecting beyond the anterior margin of the an- gular. And again as in Python, a second anterior medial prong originates from the compound bone ventromedial to Meckel's cartilage, but whereas this process remains very small in Python, it is elaborated into a tall vertical flange in Lichanura and establishes contact with the more dorsally lo- cated anterior medial projection of the compound bone (Fig. 12C). This arrangement results in a combined anterior projection of the compound bone medial to the angular, whereas the latter shows an anteriorly projecting lateral margin. Thus, the compound bone and the angular togeth- er define a deep vertical trough, into which slides the posterior margin of the posterior vertical lam- ina of the splenial. The tall ventral part of the 22 FIELDIANA: GEOLOGY medial anterior projection of the compound bone carries a distinct notch at the depth of the trough that is formed by itself and the anteriorly project- ing lateral margin of the angular. Into that notch fits the knoblike posterior head of the horizontal crest that projects from the lateral surface of the splenial (Fig. 12F). The intramandibular joint of Boa constrictor (Fig. 14, fmnh 22363) resembles that of Licha- nura more closely than that of Python, although important differences are also noted. As in other macrostomatans, the dentary is again deeply bi- furcated posteriorly. The two posterior prongs of the dentary, together with the intramandibular septum, form a recess that receives the strongly developed anterior prong (surangular portion) of the compound bone. As in Lichanura and Cha- rina, the intramandibular septum forms a distinct posteromedial projection. In Lichanura and Cha- rina, this projection comes to lie in a facet on the VII hy B pmf an Fig. 14. The lower jaw of Boa constrictor imperator (fmnh 22363). A. postdcntary bones, medial view; B, dentary, medial view; C, postdentary bones, medial view; D, splenial, lateral view. A, B, scale bar = 10 mm; C, D, scale bar = 5 mm. Abbreviations as in Figure 1 . RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 23 medial aspect of the compound bone and itself remains superficially exposed in medial view. In Boa, this process forms a distinct facet on its me- dial surface, which receives distinct anterior pro- cesses of the coronoid and splenial. Unlike in Python, the angular is applied to the ventromedial surface of the compound bone, with no exposure in lateral view of the lower jaw. Its anterior end its thickened, and its anterior lateral margin is projecting anteriorly. Superficially, the angular meets the splenial in a sigmoidal suture. Unlike in Python, the splenial carries a distinct and pointed posterodorsal process that fits snugly into a triangular recess formed by the anterior dor- sal margin of the angular and the anterior end of the compound bone (Figs. 14C, 14D). Medial to the angular, the compound bone forms a pointed, cup-shaped anterior projection that, together with the anteriorly projecting lateral margin of the angular, forms a deep trough that receives the posterior end of the splenial. This ar- rangement very closely resembles Lichanura ex- cept for the fact that the two components (dorsal and ventral) that form the anterior projection of the compound bone deep to the angular in the latter genus cannot be distinguished in Boa be- cause they are completely fused. At the bottom of the trough between the angular and the anterior medial projection of the compound bone, the lat- ter carries a very distinct notch. Unlike in Python, but as in Lichanura, the splenial is pierced by the anterior mylohyoid foramen in Boa. As in Li- chanura, a very distinct lateral horizontal crest develops from the lateral surface of the splenial at the level just below the anterior mylohyoid fo- ramen. Together with the ventral margin of the splenial, this horizontal crest forms a groove that receives the medially curved ventral margin of the dentary. A ventral division of the anterior mylo- hyoid nerve pierces the lateral horizontal shelf of the splenial immediately lateral to the anterior mylohyoid foramen. Unlike in Python, the pos- terior end of the horizontal shelf of the splenial is developed into a distinct articular knob, more dis- tinctly so than in Lichanura, which fits into the notch in the anterior medial projection of the com- pound bone deep to the angular. Of all the snakes described in this study, Boa shows the most com- plex differentiation of the intramandibular joint. The differences observed in the differentiation of the intramandibular joint indicate that a broader survey of its structure could reveal a number of characters that might be useful in the analysis of the interrelationships of basal snakes. The Skull and Lower Jaw of Pachyrhachis In order to assess the phylogenetic relationships of Pachyrhachis, it is necessary to revisit the mor- phological description and cranial reconstruction given by Caldwell and Lee (1997) and Lee and Caldwell (1998). After the initial description of the dorsal aspect of the skull (Haas, 1979), the specimen was embedded in epoxy and the ventral side of the skull was prepared and described (Haas, 1980). The epoxy resin covering the dorsal surface of the skull renders it difficult to assess some morphological details. Parts of the skull are poorly preserved, such as the nasal complex com- prising the nasals, vomers, and septomaxillae, as well as parts of the palate. In general, however, the skull is fairly well known as far as it is pre- served, and it will suffice in the present context to review selected parts of its anatomy. The interpretation of the circumorbital bones in basal snakes has long been controversial (Haas, 1964, 1968), and Pachyrhachis is no exception (L97: char. 23; LC98: chars. C7, D3). Haas (1979) described three bones surrounding the orbit of Pa- chyrhachis, a dorsally located postfrontal, a pos- teriorly located postorbital, and an ectopterygoid that appears to floor the orbit. Lee and Caldwell (1998) considered Haas's (1979) postorbital to represent a postorbital fused with a postfrontal, the latter represented by an elongate anterior pro- cess lining the dorsal margin of the orbit and not indicated by Haas (1979). The postfrontal as iden- tified by Haas (1979) was interpreted as a jugal by Lee and Caldwell (1998). Re-examination of the holotype of Pachyrhach- is (Fig. 15) showed the postorbital to be a distinct element applied to the lateral wing of the parietal at the posterodorsal corner of the orbit, with a ventral process forming an extensive postorbital bar. The dorsal head of the postorbital is roughly of a triradiate structure. There is a thickened an- terodorsal head, which is sutured to the lateral end of the posterior surface of the transverse ridge that is developed on the anterior end of the parietal. From the posterior margin of this dorsal articular head projects a small yet distinct lappet, which is applied against the laterodorsal surface of the lat- eral parietal wing. Although the dorsal head of the postorbital is smaller than it is in Python, its re- lation to the parietal closely resembles the post- orbital-parietal contact in the latter taxon. At the dorsal margin of the right orbit, an elon- gate element is exposed that was interpreted as 24 FIELDIANA: GEOLOGY the postfrontal by Lee and Caldwell (1998), who thought it was originally fused to the postorbital and would have lined the dorsal margin of the orbit. Closer inspection reveals, however, that this bony element represents a sheet of bone that ex- poses its lateral edge dorsally and dips medially as it disappears below the parietal. The bone in question passes below the lateral wing of the pa- rietal and below the proximal head of the post- orbital. As such, it cannot represent a postfrontal, which would originally have been fused to the postorbital, but rather corresponds to a broken part of the laterally descending flange of the pa- rietal. We therefore conclude that Pachyrhachis lacked a postfrontal. As preserved, the ventral tip of the postorbital is pushed against the posterior end of an elongated element that lies in the floor of the orbit on top of the posterior end of the maxilla, pointing an- teromedially (the postfrontal of Haas, 1979). This element was interpreted as a jugal by Lee and Caldwell (1998), which would appear reasonable, given its topological relation relative to the max- illa and postorbital and its location mostly in front of the latter element (L97: char. 29; LC98: char. Dl). However, the element in question shows a distinct broadening of its anterior end, whereas a jugal would be expected to have a tapering ante- rior process lining the ventral margin of the orbit. For this reason, the element in question could just as well be interpreted as the anterior end of an ectopterygoid that has been broken across the posterior end of the maxilla upon dorsoventral compression of the skull during fossilization. We favor this latter interpretation because it is also supported by parsimony analysis. The parietal of Pachyrhachis is pythonomorph in that there is a distinct transverse ridge on its anterior end from which originates a sagittal crest that extends posteriorly (L97: char. 19; LC98: char. C21). The sagittal crest ends in a knoblike projection (the supraoccipital of Haas, 1979), which indicates that the parietal overhung a ver- tically oriented supraoccipital, just as in Python. The dorsolateral surface of the parietal supports the elongate supratemporals (L97: chars. 25-27; LC98: chars. C5, El), which have free-ending posterior processes from which the quadrates are suspended (Haas, 1979; Lee & Caldwell, 1998). The quadrate of Pachyrhachis is autapomorphous as it develops a broad anterior lateral extension. The mandibular condyle of the right quadrate is preserved in articulation with the right mandible, and the cephalic condyle of the same quadrate is in articulation with the posterior tip of the supra- temporal. From the posterior margin of the right quadrate there projects a distinct medial flange, positioned at a right angle to the broad anterior lateral extension of the quadrate. This posterior medial flange is most prominently developed shortly above the mandibular condyle, but it re- cedes along the upper half of the posterior margin of the quadrate, tapering off toward the cephalic condyle. The medial surface of the broad anterior lateral extension of the quadrate shows a weakly developed ridge that trends from the anterodorsal corner in a posteroventral direction toward the medially projecting shelf. Below this ridge there is a shallow groove against which the quadrate ramus of the pterygoid is articulated as it extends toward the mandibular condyle of the quadrate. Above this ridge on the medial surface of the quadrate, between it and the medial flange pro- jecting from the posterior margin of the latter, is located a shallow yet distinct notch that must have received the cartilaginous distal end of the stapes. The medial flange projecting from the posterior margin of the quadrate may therefore, in part at least, correspond to the stylohyoideal process on the quadrate of snakes. The morphology of Pa- chyrhachis is again fairly closely comparable to that of Python, except that the stylohyoideal pro- cess is better defined as it projects from the pos- terior medial margin of the quadrate in Python. Also, the contact between stapes and quadrate seems to be in a more ventral position in Pachy- rhachis as compared to Python. However, the sta- pes-quadrate articulation of Pachyrhachis is dis- tinctly different from the morphology observed in mosasaurs or in basal alethinophidian snakes such as anilioids (L97: 45). The shaft of the right stapes of Pachyrhachis is well preserved as it projects laterally from below the pterygoid between the quadrate posteriorly and the coronoid anteriorly. This element was identified as a supratemporal by Haas (1980) but as a questionable squamosal by Lee and Caldwell (1998). This slender blade of bone cannot repre- sent the supratemporal, as the latter is seen in ar- ticulation with the cephalic condyle of the quad- rate. Its identification as a possible squamosal fol- lowed from the fact that Lee and Caldwell (1998) interpreted another rod-shaped element as the sta- pes. According to their interpretation, the stapes would project posteriorly, emerging from between the posterior tips of the quadrate ramus of the pterygoid and the supratemporal, respectively. For most of its exposed part, this latter element ap- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 25 I I 26 FIELDIANA: GEOLOGY pears rod-shaped indeed, but it quickly broadens anteriorly, as is also particularly well shown (on both sides of the skull) on radiographs. In fact, the stapes as identified by Lee and Caldwell (1998) represents a posterior opisthotic process that forms a rudimentary paroccipital process, as is developed in some basal macrostomatans such as Python. A comparable paroccipital process is absent in scolecophidians and anilioids. Also, the distal end of the stapes of Pachyrhachis forms an elongate, slender, and flattened blade of bone, as is again seen in some basal macrostomatans such as Python but which is different from the short and robust stapedial shaft characteristic of anilioid snakes. Little detail can be identified in the dermal pal- ate and basicranium of Pachyrhachis beyond the general contours of the pterygoids and the elon- gate dentigerous processes of the palatines. How- ever, both lower jaws are comparatively well pre- served (Fig. 16) and have been described as being closely similar to the mandibles of mosasauroids (Lee & Caldwell, 1998). The Meckelian canal would be located on the medial surface of the dentary (L97: char. 69; LC98: char. B13), the an- terior tip of the splenial would again be located on the medial aspect of the dentary (L97: char. 70), and the splenial would meet the angular in a vertical suture indicative of intramandibular mo- bility (L97: chars. 73, 74; LC98: char. B12). The coronoid carries a short posteroventral process (L97: char. 78; LC98: char. BIO) and a long an- teroventral process, which is reconstructed by Lee and Caldwell (1998) to contact the splenial (L97: char. 75; LC98: char. B14). In fact, the illustration given by Lee and Cald- well (1998, Fig. 4) shows an elongated angular in an overlapping contact with a broad splenial in the right mandible. In the left mandible, the splen- ial is shown in a position medial to the dentary, but its posterior margin forms a vertical suture with the compound bone rather than with the an- gular, from which it remains separated. In the ventral view of the specimen, the com- pound bone of the left mandible is preserved in medial view. The adductor fossa on its posterior end faces medially and shows a sharp medial mar- gin almost as high as the lateral margin. In front of the adductor fossa, the dorsomedial margin of the compound bone is lined by an elongate ante- rior process of the coronoid, which posteriorly is developed into the large coronoid process auta- pomorphic for Pachyrhachis. As in other snakes, the coronoid is applied to the medial side of the compound bone rather than straddling the dorsal margin of the surangular, as is characteristic of nonophidian squamates. The dentary of the left mandible is crushed but exposed in ventral view. Meckel's canal is well exposed toward the ante- rior end of the dentary, and it is bordered on both sides by sharp edges. This indicates that Meckel's canal originally opened ventrally relative to the sagittal plane of the lower jaw, as is characteristic of other snakes as well. However, the lateral and medial components of the dentary surrounding Meckel's canal have been crushed, as the ventro- lateral marginal zone of the dentary can be seen to be broken off from the rest of the bone and flipped medially. This morphology is not compa- rable to the mosasaur jaw, where Meckel's canal forms a sulcus with smooth, rounded margins on the medial surface of the dentary. Lee and Caldwell (1998) identified a remnant of the angular along the ventral margin of the left mandible, although it is represented by nothing more than a broken splint of bone and remains separated from the supposed splenial by what ap- pears to be the compound bone. By comparison to the right mandible, the angular thus identified appears to be too narrow, and the bone in question appears to be a crushed remnant of the postero- ventral process of the dentary instead. By con- trast, there is what appears to be a V-shaped su- ture line, the apex pointing backward, on the me- dial surface of the mandible somewhat in front of the posterior end of the coronoid, which may de- lineate the posterior end of the angular. If cor- rectly identified, the angular represents a relative- ly broad, elongate, platelike element comparable to, yet somewhat wider than, the corresponding element of the right mandible. If correctly iden- tified, the angular of Pachyrhachis would have to Fig. 15. A, radiograph of the skull of the fossil snake Pachyrhachis problematicus Haas from the lower Middle Cretaceous of Ein Jabrud; B, the right orbit of Pachyrhachis and its surrounding bones in dorsal view; C. the left suspensorium of Pachyrhachis in ventral view. A, scale bar = 20 mm; B, C, scale bar = 5 mm. Abbreviations: c, coronoid; ec, ectopterygoid; f, frontal; f.stp, facet for stapes; md, mandible: op. opisthotic; p. parietal; po, postorbital; pt, pterygoid; q, quadrate st, supratemporal; stp, stapes. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 27 an? Fig. 16. The mandibles of Pachyrhachis problematicus Haas as exposed on the ventral side of the specimen. A, left lower jaw; B, right lower jaw. Scale bar = 5 mm. Abbreviations as in Figure 1. be reconstructed in broad sutural contact with the coronoid. Following the angular anteriorly, the relation- ships of the elements of the lower jaw become more obscured by extensive breakage. There is an anteriorly tapering structure pressed between the lateral and medial margins of Meckel's canal that, in its totality, could represent the angular in artic- ulation with the splenial. However, the vertical posterior end of the splenial, shown in articulation with the compound bone by Lee and Caldwell (1998, Fig. 4), appears to represent a break. Fur- thermore, the anterior tip of the supposed splenial is separated from the dentary along the dorsal margin of Meckel's canal by what appears to be a very delicate fracture rather than a suture. But even if this were a suture, the splenial would be associated with the wrong (dorsal) margin of Meckel's canal by comparison with other squa- mates. The exact nature of the angular-splenial contact cannot be unequivocally established on the left mandible. Also, a posterior mylohyoid fo- ramen cannot be identified on the left angular in a comparable position to the one on the right el- ement, although it might be represented by a fo- ramen located more anteriorly in a depression. The morphology and relations of the angular and splenial are better preserved and more fully ex- posed on the right mandible. The preservation of the right mandible is more complicated, but the compound bone is preserved in more or less straight ventral view (the adductor fossa facing upward), and it is exposed to a level far anterior along the lower jaw. The dentary has been twisted during dorsoventral compression of the skull due to the overlying maxilla. The lateral surface of the dentary is therefore more or less exposed in the ventral view of the specimen. There is no indication of a shallow Meckelian groove on the anterior end of the dentary, as is illustrated by Lee and Caldwell (1998, Fig. 4b). Two elongated elements are located along the me- dial margin of the compound bone and have been interpreted as angular and splenial by Haas (1980) and Lee and Caldwell (1998). The posterior one of these two elements is a fairly broad and elon- gate, platelike angular with a slightly concave ventral margin and a broken posterior end. The posterior mylohyoid foramen is located somewhat in front of its midpoint. The anterior end forms a distinct and pointed anteroventral projection that overlaps the posterior margin of the adjacent el- 28 FIELDIANA: GEOLOGY ement, the splenial. Above this anteroventral pro- jection, the sloping anterior margin of the angular is distinctly notched and thus forms the posterior margin of the anterior mylohyoid foramen, which is completed anteriorly by the splenial. The splenial is an unusually broad element that has been preserved with a broken anterior tip. It seems to taper in its anteriormost part only and must have reached far anteriorly along the ventral margin of the lower jaw. It remains unclear whether a splint of bone lying alongside the den- tary represents the thin anterior extremity of the splenial. But even if this possibility is discounted, the splenial of Pachyrhachis reaches further an- teriorly than is typical of basal alethinophidians and hence is a character that the latter genus shares with macrostomatans. Pachyrhachis is au- tapomorphous, however, with respect to the broad, platelike appearance of the angular and splenial. The two elements also do not meet in a vertical suture but in an overlapping contact, which fur- thermore appears to enclose the anterior mylo- hyoid foramen. Another matter of debate is the number of men- tal foramina present in the dentary of Pachy- rhachis (L97: char. 76; LC98: char. C14). Non- ophidian squamates typically have a series of mental foramina lining the dentary below the tooth row. By contrast, snakes have a single men- tal foramen located toward the anterior tip of the dentary. Pachyrhachis was described as retaining two mental foramina, another supposedly primi- tive character of the genus (Lee & Caldwell, 1998). There is no indication of any foramen on the damaged lateral surface of the right dentary (ex- posed in the ventral view of the skull). By con- trast, two foramina appear to be located toward the anterior end of the left dentary, exposed in the dorsal aspect of the specimen, which is now em- bedded in epoxy. Of these, the smaller posterior one is an undisputed mental foramen; it is also clearly identifiable in the photograph published by Haas (1979, Fig. 6B). By contrast, the position of the larger anterior foramen corresponds to a notch in the eroded ventrolateral margin of the dentary in the photograph, which shows the specimen be- fore it was embedded in epoxy (Haas, 1979, Fig. 6B). Indeed, the erosion of the margin of the den- tary is obscured by the epoxy, on the "ventral" (lower) surface of which a thin film of matrix was left in place, lining the dentary and suggesting a complete lateral (ventral) margin of the supposed foramen. The bone surface is easily distinguished from the film of matrix, however, not only by its shiny appearance but also by a reddish hue that is absent in the matrix. The impression of a fo- ramen is further reinforced by a bubble, which appears to have formed in the notch of the eroded margin of the dentary as it was embedded in ep- oxy. In conclusion, Pachyrhachis is characterized by the presence of a single mental foramen, as is characteristic of all snakes. Character Evidence for the M onophyly of the Pythonomorpha The monophyly of the Pythonomorpha, with Pachyrhachis as intermediate between mosasau- roids and snakes, has recently been supported by a global analysis of squamate interrelationships (Lee, 1998). In this section, we review the char- acter evidence used by Lee (1998; see this paper for the primary sources of characters, if applica- ble, and a discussion of characters other than the comments below) for the analysis of squamate in- terrelationships. For readers who do not have Lee (1998) at their disposal, we give an abbreviated definition of each character as conceived by Lee (1998). Subtleties of character definitions may be lost in our rendering, and the reader is encouraged to refer to Lee (1998) to avoid possible misun- derstandings. Our review of this data set (Table 1 , pp. 50 ff.) is not intended to provide a better un- derstanding of global squamate relationships but rather to test the monophyly of the Pythonomor- pha on the same grounds on which it was pro- posed. If Pachyrhachis is indeed the link between mosasauroids and snakes, the consequence could be that snakes have had a marine origin and that the fossorial ecomorph evolved independently among nonophidian squamates and within snakes (Lee, 1998; Scanlon et al., 1999). Our review therefore focuses on those characters that are rel- evant to the placement of Pachyrhachis and to the potential phylogenetic relationships of snakes with varanoid lizards (including mosasauroids), on the one hand, or with amphisbaenians and di- bamids on the other. Character 1: Premaxillary palatal foramina pres- ent (0), absent (1). As defined by Lee (1998), the premaxillary foramina in nonophidian squa- mates are synonymized with the premaxillary channels of snakes (see Kluge, 1989, for ter- minology). RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 29 Character 2: Premaxillary lateral foramina present (0), absent (1). There are no such lateral premax- illary foramina in snakes, which therefore cannot be compared and are coded as unknown (?). Character 3: Premaxilla-maxilla contact sutural (0), nonsutural (1). Character 4: Alveolar ridge of maxilla straight (0), upturned at anterior end (1). Character 5: Dorsal process of maxilla at or in front of (0) or behind ( 1 ) midpoint of maxilla. The character is redundant, as it combines the information of a retracted naris (character 16) and the degree of posterior extension of the maxilla below the orbit (character 7). Pachy- rhachis is miscoded: if Lee and Caldwell's (1998) description is correct, the maxillary-pre- frontal contact is autapomorphic for this taxon. However, personal inspection of the holotype suggests (especially on the left side of skull) that the "dorsal process of the maxilla" is, in fact, the anterior portion of the prefrontal of similar proportions as seen in anilioids. We therefore code Pachyrhachis 0. Character 6: Dorsal process of maxilla extends dorsomedially (0), dorsolateral^ (1). Character 7: Posterior process of maxilla long (0), short (1). Character 8: Lacrimal present (0), absent (1). Character 9: Lacrimal separate (0), fused with prefrontal (1). Character 10: Lacrimal foramen single (0), dou- ble (1). Character 11: Lacrimal foramen at least partly bordered by facial elements (0), entirely within prefrontal (1). The primitive position of the lac- rimal foramen in squamates is between the lac- rimal and the prefrontal. Since snakes lack a lacrimal, the lacrimal foramen lies between the maxilla and the prefrontal for those terminals we included. We have been unable to verify Lee's (1998) assessment of the position of the lacrimal foramen in mosasaurs (Camp, 1942; Russell, 1967), but if he is correct, the position of the lacrimal within the prefrontal would be autapomorphic for mosasaurs. As such, the character is uninformative and hence is deleted. Character 12: Jugal present (0), absent (1). Per- sonal investigation of the holotype of Pachy- rhachis indicates that the jugal sensu Lee and Caldwell (1998) most probably represents the anterior end of the ectopterygoid, which broad- ly overlaps the posterior end of the maxilla. The ectopterygoid of Pachyrhachis appears to be broken across the posterior end of the maxilla on both sides of the skull, a consequence of the dorso ventral compression of the skull. Dinilysia has also been described with a jugal (Estes et al., 1970), an assessment that has been ques- tioned by Lee (1998). We have coded Dinilysia with or without jugal in separate analyses to test the effect of either assumption. Character 12 codes Dinilysia with a jugal (0), whereas char- acter 142 codes Dinilysia without a jugal (1). Using either one of these characters requires de- letion of the alternative. Character 13: Jugal does not (0), does (1) extend anteriorly beyond midpoint of orbit. This char- acter again could not be coded accurately for Dinilysia because of the poor preservation of the bone identified as a jugal (Estes et al., 1970). Character 14: Nasals paired (0), fused (1). Va- ranus should be coded polymorphic for this character until the derived nature of paired na- sals is confirmed by cladistic analysis of var- anid interrelationships. Scolecophidians are polymorphic in this character. The nasal(s) of Pachyrhachis is/are not known. Character 15: Nasal-prefrontal contact present (0), absent (1). Character 16: External naris not retracted (0), slightly retracted (1; frontal excluded), strongly retracted (2; frontal enters external naris). Lan- thanotus has to be coded 1 (Rieppel, 1983) or polymorphic. The nature of the external naris is not known in Pachyrhachis (?), although it is most conceivable that it was retracted. Lee (1998) coded snakes 1 for this (ordered) mul- tistate character, but exclusion of the frontal from the posteriorly retracted external naris is most probably a reversal due to the burrowing habits of scolecophidians and anilioids. Character 17: Prefrontal smooth (0), rugose (1) at orbital margin. 30 FIELDIANA: GEOLOGY Character 18: Frontal(s) single (0), paired ( 1 ). Character 19: Frontal with straight or weakly concave (0), strongly concave (1) lateral mar- gin. Lee (1998) coded Varanus 1 for a deeply concave orbital margin of the frontal, yet coded mosasaurs 0 for a slightly concave orbital mar- gin. In fact, the concavity of the orbital margin of the frontal is closely comparable in both groups (1) (Russell, 1967, Fig. 83). A straight frontal margin is derived within mosasaurs (Bell, 1997). Character 20: Frontal flange underlying nasal ab- sent (0), present (1). Character 21: Frontoparietal suture complex, in- terdigitating (0), simple, straight transverse line (1). We propose a redefinition of this character as follows: Superficial delineation of frontopa- rietal suture complex and distinctly interdigitat- ing (0), essentially a straight transverse line ( 1 ), frontal invading parietal (2), frontals postero- lateral^ embraced by parietal (3). Nonophidian squamates have a frontoparietal suture that su- perficially forms a more or less straight line (1). In mosasaurs, the frontal tends to develop pos- terior processes of variable shape overlapping the parietal (2) (Bell, 1997). In basal snakes (Anilioidea), the parietal tends to form antero- lateral processes embracing the frontals in a curved suture (3). Scolecophidians (Haas, 1964, 1968; List, 1966) are polymorphic in this char- acter (1 and 3). Anelytropsis shows character state 1 (Greer, 1985), but Dibamus (Rieppel, 1984b; not Anelytropsis: Greer, 1985) shows character state 3. Most amphisbaenians show state 0. Pachyrhachis has state 1. Character 22: Frontal enters (0), is excluded ( 1 ) from dorsal margin of orbit. Pachyrhachis has to be coded as unknown (?) for this character due to the uncertain nature of the splint of bone exposed at the dorsal margin of the right orbit (see discussion of character 24, below). Dini- lysia is coded 0. This character is not applicable to scolecophidians (?). Character 23: Postfrontal large (0), small ( 1 ), ab- sent (2). This character is partially redundant with some of the following. We retain it as cod- ed by Lee (1998) except in Pachyrhachis and snakes, which we interpret as lacking a discrete postfrontal (2). Pachyrhachis is coded as un- known (see discussion of character 24, below). Character 24: Postfrontal separate (0), fused to postorbital in adult ( 1 ). This character is prob- lematic, as it makes assumptions about ontog- eny in fossils. By comparison to Varanus, we agree that the postfrontal and postorbital fuse to form a postorbitofrontal in mosasaurs (Bell, 1997). However, Lee (1998) codes Pachyrhach- is for a fused postorbitofrontal. This coding is based on the assumption (Lee & Caldwell, 1998) that a splint of bone exposed at the dorsal margin of the right orbit of the holotype of Pa- chyrhachis is the anterior process of the post- frontal, which establishes a contact with the prefrontal at the dorsal margin of the orbit (not present in Varanus or mosasaurs). Alternative interpretations would be to identify this splint of bone as postfrontal, separate from the post- orbital, or to identify this splint of bone as a supraorbital (present in Python [Frazzetta, 1959], with which Pachyrhachis shares the crested anterior end of the parietal), lying in front of the postorbital. Re-examination of the holotype of Pachyrhachis suggests, however, that neither of these interpretations is correct, as the fragment of bone exposed at the dorsal margin of the orbit seems to be a splint of the laterally descending flange of the parietal (see above). Given the uncertainty of interpretation, we conservatively code Pachyrhachis as un- known (?) for this character (as well as for char- acters 22 and 23). Other snakes lack the post- frontal and hence are not comparable (?). Bau- meister (1908) found a sutural separation of the anterolateral process of the parietal that lines the dorsal margin of the orbit in a uropeltid (Rhinophis) and considered this process to rep- resent the postfrontal. Although accepted by Rieppel (1977), we here discard this interpre- tation until the observation of Baumeister (1908) is independently corroborated. Character 25: Postfrontal not forked (0), forked (1) medially. The postfrontal (postorbitofrontal) is forked dorsally, i.e., clasps the frontoparietal suture in all nonophidian squamates that retain the mesokinetic axis (Rieppel, 1984b), includ- ing mosasaurs. Snakes lack a postfrontal, but in Python, the dorsal end of the postorbital is forked, without, however, clasping the fronto- parietal suture. This morphology is approached by Pachyrhachis. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 31 Character 26: Palpebral ossification absent (0), present (1). Character 27: Postorbital present (0), absent (1). This character is difficult to interpret without adding the specification of whether the post- frontal is present or absent as a discrete ossifi- cation (redundant with character 24). However, as coded by Lee (1998), the character has no bearing on snake relationships and therefore is retained unchanged. Character 28: Ventral process of postorbital long (0), short (1). Character 29: Posterior margin of orbit complete (0), with small gap (1), with large gap (2). Be- cause we interpret the jugal of Pachyrhachis as an ectopterygoid, it becomes impossible to de- cide whether Pachyrhachis has a complete or an incomplete postorbital arch. Yet the taxon shows a long ventral process of the postorbital, for which reason we code the taxon as 1. Character 30: Parietal(s) paired (0), fused (1). Character 31: Parietal tabs (triangular flanges ex- tending anteriorly into fossae on ventral surface of frontals) present (0), absent (1). We have some difficulty understanding this character. In Varanus, the parietal bears lateral tabs project- ing anteriorly that overlap a facet on the frontal as part of the mesokinetic joint (Rieppel, 1979a). A similar relation of these bones, but not as strongly expressed as in Varanus, is seen in other nonophidian squamates. This character is absent, however, in Platecarpus (amnh 04909) and in other mosasaurs (Bell, 1977). Comparable anterolateral tabs of the parietal ar- ticulating in a facet on the frontal are also ab- sent in snakes (including Pachyrhachis), am- phisbaenians, or dibamids, which also have lost mesokinesis! However, given our difficulty in understanding Lee's (1998) character definition, we retain the coding he chose for this character. Character 32: Parietal half as long as skull or shorter (0), longer (1) than half of skull length. Character 33: Pineal foramen present (0), absent (1). Character 34: Pineal foramen within parietal (0), on frontoparietal suture (1), within frontal (2). Character 35: Origin of jaw adductor muscles re- stricted to ventral surface (0), invades dorsal surface (1) of parietal. This character overlaps with character 57. For reasons discussed below (character 57), we retain this character in our analysis but delete character 57. Character 36: Supraoccipital exposed in dorsal view (0), concealed in dorsal view (1) by pa- rietal. Character 37: Posterolateral process of parietal distinct (0), short or absent (1). Character 38: Upper temporal arch complete (0), incomplete (1). Character 39: Jugal does not (0), does (1) contact squamosal. Character 40: Squamosal present (0), absent (1). Pachyrhachis is coded for presence of a squa- mosal, but personal inspection of the holotype confirmed that the squamosal identified by Lee and Caldwell (1998) is, in fact, the stapes and that the stapes identified by Lee and Caldwell (1998) is, in fact, a posterior opisthotic (par- occipital) process, which is also present in a variety of basal macrostomatans (Zaher, 1998; see also Frazzetta, 1959). The squamosal is ab- sent in Pachyrhachis (1). Character 41: Dorsal process of squamosal pre- sent (0), absent (1). This character is not appli- cable to Pachyrhachis given the reassessment of the previous character. Character 42: Upper temporal fenestra not re- stricted (0), restricted (1) by postorbital. Character 43: Upper temporal fenestra not re- stricted (0), restricted (1) by postf rental. Characters 42^3: These characters were coded 0 for Pachyrhachis and snakes by Lee (1998), when in fact they are not applicable to these taxa (?). Character 44: Supratemporal absent (0), present (1). Character 45: Supratemporal on dorsolateral (0), ventrolateral (1) surface of parietal. 32 FIELDIANA: GEOLOGY Character 46: Supratemporal confined to skull roof (0), forms part of paroccipital process and/ or braincase ( 1 ). In nonophidian squamates, the supratemporal may or may not contact the distal end of the paroccipital process. The posterior tip of the supratemporal establishes a broad contact with the distal tip of the opisthotic in Varanus and a somewhat more extended but otherwise identical contact in mosasaurs (Riep- pel & Zaher, in press). In snakes, the paroccip- ital process is much reduced. If present, it cor- responds to a small posterior projection of the opisthotic as seen in some basal macrostoma- tans, and it is not in contact with the supratem- poral (although the slight mobility of the supra- temporal may change these relations in dried skulls). Such a paroccipital process is present in Pachyrhachis, but its relations to the supra- temporal are obscured through dorsoventral crushing of the skull. The posterior tip of the opisthotic (stapes of Lee & Caldwell, 1998) is separate from the supratemporal, however, which indicates lack of a contact. In no case is the supratemporal (a dermal element) part of the braincase (endocranium), although the su- pratemporal may be superimposed on braincase elements. We conclude that Pachyrhachis and snakes have to be coded 0 for this character. Character 47: Supratemporal less than half (0), at least half (1) of maximum skull width. This character is misleadingly coded for mosasaurs by Lee (1998). In a skull roof of Platecarpus (amnh 01820), the supratemporal is less than half the maximum width of the skull. This is also the case for other mosasaurs (Camp, 1942; Russell, 1967). In basal snakes, the supratem- poral is absent or small (scolecophidians and anilioids); it is also small or absent in dibamids and amphisbaenians. In macrostomatans, the supratemporal is elongated and carries a free- ending posterior process, as it also does in Pa- chyrhachis (Zaher, 1998). We therefore propose to modify this character in order to account for the presence of a free-ending posterior process of the supratemporal in Pachyrhachis and ma- crostomatans: supratemporal without (0) or with ( 1 ) free-ending posterior process. Character 48: Supratemporal does not (0), does (1) contact prootic. In Varanus, the posterior tip of the supratemporal lies against the lateral sur- face of the opisthotic at the distal tip of the paroccipital process and is in a loose syndes- motic connection with the prootic. In mosa- saurs, the posterior tip of the supratemporal is somewhat expanded and forms a deeply inter- digitating suture with the prootic (Rieppel & Zaher, in press). In snakes, the supratemporal shows very different relations. Due to the rel- ative size increase of the braincase (Rieppel, 1984b), the latter comes to lie in the same plane as the dermatocranium (rather than being sus- pended within it. as in mosasaurs and other nonophidian squamates). The supratemporal thus comes to lie on top of the otic capsule, lateral to the reduced posterolateral (supratem- poral) processes of the parietal, which in turn also lie on top of the otic capsule. The prootic- supratemporal contact in snakes (including Pa- chyrhachis) is therefore the result of a funda- mentally different remodeling of the snake skull (Rieppel, 1984b; Rieppel & Zaher, in press) and is not comparable to the morphology seen in mosasaurs and Varanus. We therefore delete this character from the analysis (the coding re- tained in the data matrix is that of Lee, 1998). Character 49: Quadrate monimostylic (0), strep- tostylic and suspended from supratemporal, squamosal, and opisthotic ( 1 ), suspended main- ly from supratemporal (2), suspended mainly from opisthotic (3), suspended mainly from squamosal (4). As worded by Lee (1998), the definition of this character is somewhat confus- ing. It remains unclear whether character state 1 requires equal participation of the supratem- poral, squamosal, and opisthotic in the quadrate suspension and whether character state 2 does or does not allow some participation of the squamosal and opisthotic in the quadrate sus- pension. As shown by Russell (1967, Fig. 20; see also Rieppel & Zaher, in press), the relation of the paroccipital process (opisthotic) to the supratemporal, squamosal, and quadrate is closely similar in mosasaurs and Varanus, where the quadrate articulates with the ventro- lateral surface of the distal tip of the paroccip- ital process by means of an intercalary carti- lage. We therefore code mosasaurs as having a squamosal-supratemporal-opisthotic articula- tion of the quadrate (see also Camp. 1942. p. 35). Character 50: Quadrate shaft slanting strongly anteroventrally. almost horizontal (0). slanting slightly anteroventrally or vertical ( 1 ). slanting slightly or greatly posteroventrally (2). We re- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 33 define this character as quadrate slanting strongly anteroventrally, almost horizontal (0), slanting slightly anteroventrally (1), positioned vertically (2), or slanting posteroventrally (3). These character states reflect an ontogenetic transformation (Bellairs & Kamal, 1981), and their assessment therefore requires that adult material be examined. A posteroventrally slant- ing quadrate has been designated a macrosto- matan character (Rieppel, 1988), and although the reconstructions of Pachyrhachis provided by Lee and Caldwell (1998) show a vertically positioned quadrate, it may have had a postero- ventrally slanting quadrate in life. As preserved, the quadrates extend posterolaterally from their articulation with the supratemporal, yet the tip of the right lower jaw still reaches the anterior tip of the right maxilla, which indicates that the quadrate and mandible have been shifted slight- ly forward. We accordingly code Pachyrhachis for a posteroventrally sloping quadrate (3). Character 51: Tympanic crest on quadrate well developed (0), weakly developed (1), absent (2). We code Varanus 1 for a reduced tympanic crest, not 0 for a prominently developed tym- panic crest, as coded by Lee (1998). Character 52: Quadrate with (0), without (1) an- teromedial lappet. Character 53: Orbitonasal fenestra wide (0), nar- row (1). Character 54: Ventromedial processes of frontal end free ventrally (0), abutting or sutured to each other below olfactory tracts (1), contact parabasisphenoid (2). As worded by Lee (1998), this character would refer to medially descending flanges of the frontal. These are an autapomorphy of alethinophidians (our charac- ter 231), however, which is why we believe that Lee (1998) is, in fact, referring to laterally de- scending frontal flanges. In amphisbaenians, the lateral ventral flanges of the frontals meet each other and contact the orbitosphenoid ventrally. The coding retained in the data matrix is that of Lee (1998). Character 55: Orbitosphenoid absent (0), present (1). This character is an autapomorphy of Am- phisbaenia and, as coded by Lee (1998), groups Sineoamphisbaena with amphisbaenians. Its presence in Sineoamphisbaena must be critical- ly assessed (see character 59). Character 56: Parietal downgrowths absent or weak (0), prominent (1). The well-preserved parietal of Platecarpus (amnh 01563) is closely comparable to that of Varanus. Mosasaurs are here coded 0. Character 57: Parietal downgrowths pointed ven- trally (0), sheetlike (1). Following the character definition of Lee (1998), mosasaurs as well as all varanoids should be coded 1. However, as we understand it, we believe this character is misleading. The parietal may have laterally de- scending flanges or not: If such flanges are ab- sent, the jaw adductors originate from the ven- tral surface of the parietal only. If such flanges are present, the jaw adductors "migrate" onto the "dorsal surface" of the parietal (Estes et al., 1988), i.e., they invade the lateral surface of these flanges. This is the same character as character 35. Another character is the presence or absence of a distinct ventral projection from the parietal as seen in some "lizards" such as skinks (Greer, 1970). Redefined along these lines, the character loses its importance for the analysis of snake-mosasaur relationships and is hence deleted from the analysis (the coding re- tained in the data matrix is that of Lee, 1998). Character 58: Parietal-prootic contact absent or restricted (0), extensive (1). We believe this character to be poorly defined. Nonophidian squamates in general, mosasaurs included, have an alar process of the prootic made of Zuwachs- knochen sensu Starck (1979; the term Zuwachs- knochen refers to a membrane bone extension added to an element, the rest of which is pre- formed in cartilage) that contacts the parietal. Dibamids (Rieppel, 1984a; Greer, 1985), am- phisbaenians (Montero et al., 1999, contra Rieppel, 1984a), and snakes lack the alar pro- cess of the prootic, which is also absent in Sphenodon. They also have an extensive pari- etal-prootic contact. We therefore redefine this character as alar process on prootic absent (0), or present (1). Character 59: Parietal downgrowths end free ven- trally (0), contact parabasisphenoid (1), contact orbitosphenoid (2). Pachyrhachis should tech- nically be coded as unknown (?), although we agree to code it 1. As such, character state 1 is 34 FIELDIANA: GEOLOGY a synapomorphy of snakes. The coding used here for Sineoamphisbaena is the same as in Lee (1998), although the presence of an orbi- tosphenoid and its relations to the parietal need to be critically reassessed in this taxon (M. Kearney, personal communication). Character 60: Optic foramen not enclosed by bone (0), enclosed partially or entirely by fron- tal (1), enclosed by orbitosphenoid (2). We be- lieve that this character has to be broken down in order to account for the different positions of the optic foramen in scolecophidians, henophi- dians, and caenophidians sensu Underwood (1967). The character thus becomes: optic fo- ramen not enclosed by bone (0), enclosed by frontal (1), enclosed by orbitosphenoid (2), en- closed by frontal and parietal (3), enclosed by frontal, parietal, and parasphenoid (4). Acro- chordids are autapomorphic for the position of the optic foramen in the parietal, a character that is not relevant to the present analysis and hence is here ignored. Pachyrhachis has to be coded as unknown (?). The coding used here for Sineoamphisbaena is the same as in Lee (1998), although the presence of an orbitosphe- noid and its relations to the optic foramen need to be critically reassessed in this taxon (M. Kearney, personal communication). Character 61: Anterior brain cavity not floored by bone (0), floored by orbitosphenoid (1), floored by wide cultriform process of para- sphenoid (2). We believe this character to be poorly defined. First, the term "anterior brain cavity" is a poor choice of words: since the olfactory tracts and bulbs are part of the brain, the brain cavity extends anteriorly to a level in front of the frontal. As coded by Lee (1998), this character is a synapomorphy shared by mo- sasaurs and snakes. However, the basicranium of mosasaurs generally resembles that of Va- ranus, with two exceptions: the sella turcica is less recessed below the dorsum sellae, and the dorsum sellae is less developed (Rieppel & Zaher, in press). But mosasaurs, like any other nonophidian squamates, have a sella turcica in front of which the basicranium is abruptly con- stricted to a short rostrum basisphenoidale, which in turn terminates in a point (or narrow cultriform process), indicating a tropibasic skull. In snakes, the sella turcica lies at the back end of the parabasisphenoid, and the latter ex- tends in front of the sella turcica as a broad element underlying the brain, with the Vidian canal exposed on the dorsal surface and with the lateral edges sutured to the parietal down- growths (Rieppel, 1979b). In front of the pari- etal, i.e., below the frontals, the parabasisphen- oid may form a relatively broad (e.g., Cylindro- phis) or narrow (e.g., Anilius) cultriform pro- cess. This type of parabasisphenoid (our definition for character state 2) is in fact a syn- apomorphy of snakes and reflects the funda- mental change of skull proportions in snakes versus lizards (much elongated postorbital re- gion, small orbits, short preorbital region). Fos- sorial lizards such as Anniella (Rieppel, 1978), amphisbaenians, and dibamids (Rieppel, 1984a) approach the morphology of the ophidian par- abasisphenoid because of similar changes in skull proportions. Character 62: Cultriform process of parasphenoid in lateral view curved upward (0), straight ( 1 ). Any aspect of this character that goes beyond the discussion of character 61 is a preservation- al artifact. The character is therefore deleted from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 63: Trigeminal foramen (foramina) open anteriorly (0), bordered anteriorly by pa- rietal ( 1 ), bordered anteriorly by orbitosphenoid and parabasisphenoid (2). Character 64: Alar process of prootic weak (0), extensive (1). This character is redundant with character 58 and therefore is deleted from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 65: Alar process of prootic directed dorsally (0), anterodorsally (1). This character is inapplicable to Pachyrhachis (unknown [?], contra Lee, 1998), snakes, amphisbaenians. and dibamids. Character 66: Crista prootica well developed (0), reduced (1). We believe that by comparison to snakes, dibamids, and amphisbaenians, the var- anoid genera Lanthanotus and Heloderma, as well as xenosaurs, should be coded 0. However, we agree to code these taxa for a weakly de- veloped crista prootica (1), but code snakes, amphisbaenians, and dibamids 2 for the absence of a crista prootica. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 35 Character 67: Lateral head vein not enclosed (0), enclosed (1) in bony canal formed by antero- ventral continuation of crista prootica. Character 68: Facialis foramen single (0), double (1). The single or double facialis foramen is variable within Varanus (even at the species level), as well as in mosasaurs (Rieppel & Za- her, in press). It is also variable in snakes, where some (scolecophidians, uropeltines) show an intracranial course of the palatine branch of the facial nerve. We therefore delete this character from the analysis (the coding re- tained in the data matrix is that of Lee, 1998). Character 69: Hypoglossal foramen well separat- ed (0), close or confluent (1) with vagus fora- men. Character 70: Occipital recess open (0), closed (1) laterally. As defined and coded by Lee (1998), this character requires that the presence or absence of the crista circumfenestralis of snakes is dealt with as a separate character. Oth- erwise, snakes would have to be coded 1 also (see character 73). Character 71: Otic capsule not expanded (0), ex- panded (1) laterally. Lee (1998) codes Sineoam- phisbaena, Amphisbaenia, and dibamids as having a laterally expanded otic region (1). This reflects the relative size increase of the laby- rinth organ in miniaturized squamate skulls (Rieppel, 1984b), which is also observed in oth- er fossorial squamates, including snakes. The latter therefore have to be coded 1 also. Character 72: Stapes light, footplate small (0), robust, footplate large (1). As dibamids and Amphisbaenia are coded 1 for a stapes with a robust shaft and a large footplate, basal snakes such as anilioids and some basal macrostoma- tans (Xenopeltis) also have to be coded 1 . Scin- cidae is polymorphic for this character (Riep- pel, 1981). Character 73: Stapedial footplate not surrounded (0), tightly surrounded by bony ridges project- ing from lateral surface of braincase (1). As de- fined and coded by Lee (1998), this character becomes a synapomorphy of mosasaurs and snakes. Note that, in contrast to Lee (1997, character 44) and Lee and Caldwell (1998, p. 1548), Lee (1998, p. 393) notes morphological differences of this character between mosasau- roids and snakes but continues to assume pri- mary homology of these structures in his cod- ing. However, the braincase of mosasaurs re- sembles that of Varanus rather closely and is not comparable to the otic region of snakes characterized by a crista circumfenestralis (Es- tes et al., 1970; Rieppel & Zaher, in press). We therefore propose to redefine this character as follows: a crista circumfenestralis, enclosing a juxtastapedial recess, is absent (0) or present (1). Pachyrhachis has to be coded as unknown (?). The character is a synapomorphy of snakes. Character 74: Basipterygoid processes long (0), short (1). As defined by Lee (1998), this char- acter is not applicable (not comparable) to snakes because basipterygoid processes of snakes (basal macrostomatans) are not homol- ogous to those of lizards (Kluge, 1991; Rieppel & Zaher, in press). Lee and Caldwell (1998, p. 1534) described short basipterygoid processes in Pachyrhachis, and Lee (1998) coded Pa- chyrhachis for short basipterygoid processes (1), despite his claim that the nature of these processes cannot be confirmed for this taxon (Lee, 1998, p. 442) or the claim that distinct basipterygoid processes are absent in this taxon (Scanlon et al., 1999). Pachyrhachis should be coded as unknown (?). Character 75: Articular facet on basipterygoid process subcircular (0), anteroposteriorly elon- gated (1). The same comments apply to this character as to the preceding one. Character 76: Basal tubera located posteriorly (0), anteriorly (1). In varanoids and mosasaurs, the basal tubera (sphenoccipital tubercles sensu Oelrich, 1956) are located anteriorly on the ba- sioccipital at the ventral end of the crista tub- eralis. In snakes, they are incorporated into the posteroventral part of the crista circumfenes- tralis, as the latter incorporates the crista tub- eralis (Estes et al., 1970; Rieppel & Zaher, in press). As defined by Lee (1998), this character is not applicable (not comparable) to snakes. Pachyrhachis should be coded as unknown (?). Character 77: Posterior opening of Vidian canal within basisphenoid (0), at basisphenoid-pro- otic suture (1), between prootic and epiphyseal ossification in the region of the basal tubera (2). 36 FIELDIANA: GEOLOGY Character 78: Posterior opening of Vidian canal situated well in front of (0), near ( 1 ) the pos- terior end of basisphenoid. This is a somewhat vague character, but as the braincase anatomy of mosasaurs is rather closely comparable to that of Varanus, they, too, should be coded 0. Snakes should be coded 1 for character 78, yet some snakes should be coded 1 for character 77 (Rieppel, 1979b). Character 79: Crista tuberalis weakly developed (0), flaring (1). As defined by Lee (1998), this character is difficult to separate from his char- acter 73. As coded, the character is a synapo- morphy of Lanthanotus and Varanus, and as such is irrelevant for the analysis of snake re- lationships. We retain it as coded by Lee (1998). Character 80: Supraoccipital separated from (0), in narrow contact with (1), in broad contact with (2) parietal. This character is poorly de- fined, and we propose to replace it by the fol- lowing: posttemporal fossae present (0), re- duced (1), absent (2). Character 81: Supraoccipital situated ventrad or posteroventrad (0), at same level (1) as parietal. Scincidae and Pygopodidae are polymorphic for this character. Character 82: Exoccipital separate (0), fused to opisthotic (1) in adult. Although we retain this character as defined and coded by Lee (1998), we believe its phylogenetic information content to be very limited. The skull of an adult Var- anus komodoensis (fmnh 22199; condylobasal length: 215 mm) retains a separate exoccipital. Character 83: Occipital condyle single (0), dou- ble ( 1 ). Although we retain this character as de- fined and coded by Lee (1998), we believe its phylogenetic information content to be limited. The double occipital condyle, formed by a pos- terior projection of the exoccipitals beyond the basioccipital, is strongly expressed within Gek- kota only (Rieppel, 1984c). Character 84: Posttemporal fenestrae present (0), absent (1). This character is redundant with character 80 and is therefore deleted from the analysis (the coding retained in the data matrix is that of Lee, 1998). Character 85: Septomaxilla extensively sutured (0), not sutured (1) to maxilla. Iguanids are polymorphic for this character (Oelrich, 1956; Etheridge, personal communication), yet in most lizards, the main body of the septomaxilla floors the anterior part of the external naris and is sutured to the maxilla, premaxilla, and vomer. In mosasaurs, the thin and bladelike septomax- illa is not in contact with the maxilla but lies dorsal to it and is sutured to the equally thin vomer (Camp, 1942). There does not appear to be any potential for independent mobility of the septomaxilla. Mosasaurs therefore differ from varanoids and other nonophidian squamates in this character. However, in shape and location, the septomaxilla in all nonophidian squamates (including mosasaurs) is radically different from (i.e., nonhomologous to) that of snakes, where the septomaxilla lies lateral to the vomer and, together with the latter, forms a chamber for Jacobson's organ, a synapomorphy of snakes. Pachyrhachis has to be coded as un- known (?) because the vomers are not known. Other snakes are also coded ? because the char- acter as defined by Lee (1998) is not compa- rable (not applicable). Character 86: Septomaxillae separated by carti- laginous gap (0), meeting on midline (1). We delete this character from the analysis (the cod- ing retained in the data matrix is that of Lee, 1998) because each septomaxilla is associated with its nasal capsule, and the two septomax- illae are always separated by the cartilage of the internasal septum and trabecula communis. Ob- servations to the contrary are artifacts of a dried skull (e.g., Bellairs & Kamal, 1981, Figs. 25, 28, 32, 65). Character 87: Medial flange on septomaxilla short (0), long ( 1 ). Among snakes, anomalepids have no medial flange of the septomaxilla that turns upward (Haas, 1964, 1968). Character 88: Septomaxillary roof for Jacobson's organ flat (0), domed (1). Character 89: Opening of Jacobson's organ con- fluent with choana (0), separated from choana by vomer and maxilla ( 1 ), separated by vomer and septomaxilla (2). Character 90: Vomer(s) paired (0), fused (1). RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 37 Character 91: Vomer without (0), with (1) exten- sive contact with maxilla behind Jacobson's or- gan. Character 92: Vomer less than half as long (0), at least half as long ( 1 ) as maxilla. Character 93: Vomer platelike (0), rodlike (1). As defined by Lee (1998), this character is prob- lematic because of the apomorphic vomer con- figuration in snakes. Although we retain the character, we code basal snakes for a broad vo- mer (0). Pachyrhachis has to be coded as un- known (?). Character 94: Vomer anterior or anteromedial to palatine (0), entirely medial to palatine (1). Character 95: Secondary palate absent (0), pre- sent (1). Character 96: Palatine-vomer contact short (0), extensive (1). Character 97: Palatine-vomer contact immobile (0), mobile (1). Character 98: Palatine as long as (0), half as long as (1) vomer. Lee (1998) codes Pachyrhachis as unknown (?) for characters 92 and 93 (vo- mer), appropriately indicating that the vomer of Pachyrhachis is not known, but for characters 91, 94, 96, 97, and 98, he codes Pachyrhachis 0 when in fact he should code it as unknown again (?). Character 99: Interpterygoid vacuity extending (0), not extending (1) between palatines. Character 100: Palatine without (0), with (1) dis- tinct medial process. Character 101: Choanal groove on palatine short or absent (0), long (1). Character 102: Ectopterygoid-palatine contact absent (0), present (1). Lee (1998) coded mo- sasaurs as lacking a contact of ectopterygoid and palatine (maxilla enters suborbital fenestra), but Russell (1967) described the ectopterygoid of Tylosaurus and Plotosaurus as meeting the posterior rim of the palatine (see also McDow- ell & Bogert, 1954, Fig. 10). The precise dis- position of these elements is not known in basal mosasauroids (aigialosaurs), for which reason we code mosasauroids as unknown (?). Character 103: Suborbital fenestra large (0), small (1), absent (2). Character 104: Pyriform recess open and wide (0), open and narrow (1), closed by broad para- sphenoid (2). Character 105: Pterygoid-vomer contact present (0), absent (1). Character 106: Pterygoid with (0), without (1) triangular depression on ventral surface, ex- tending from suborbital fenestra toward pala- tobasal articulation. Character 107: Anterior (palatine) process of pterygoid gradually merges with (0), is distinct- ly set off from (1) transverse (ectopterygoid) process. Character 108: Anterolateral process of pterygoid extending along lateral margin of palatine ab- sent (0), present (1). Character 109: Epipterygoid present (0), absent (1). Character 110: Mandibular symphysis rigid (0), mobile (1). As coded by Lee (1998), mosasaurs and snakes share a mobile mandibular symphy- sis, with the anterior ends of the dentaries being smoothly rounded. Although this character is shared with other tetrapods that show an intra- mandibular joint (e.g., Tyrannosaurus rex: C. Brochu, personal communication), we accept this character in the present analysis for the lack of better knowledge of mosasaur symphyseal structure. A dentary of Platecarpus (fmnh UC 600) shows the anterior end of Meckel's groove to taper out on the medial surface of the straight dentary. It remains unknown what role Meck- el's cartilage played in the formation of a mo- bile mandibular symphysis. In snakes, the an- terior tip of the dentary is usually curved in- ward, and Meckel's cartilage protrudes from Meckel's canal and extends beyond the anterior tip of the dentary as it relates to ligaments, mus- cle fibers, and accessory cartilages (cartilago symphyseum, derived in snakes) in the forma- tion of a mobile mandibular symphysis (Young, 1998). It is noteworthy that Meckel's cartilages 38 FIELDIANA: GEOLOGY of the two mandibular ramus fuse with each other to form a true mandibular symphysis in scolecophidians (Young, 1998). Character 111: Three or more (0), two or less (1) mental foramina on dentary. Personal inspec- tion of the holotype of Pachyrhachis reveals rather extensive breakage at the anterior end of the left mandible (lateral view). In spite of re- taining the character definition of Lee (1998), we point out that we can identify a single men- tal foramen only, which is a synapomorphy that Pachyrhachis shares with other snakes. Character 112: Dentary in lateral view with straight (0), concave (1) dorsal (alveolar) mar- gin. As defined and coded by Lee (1998), the straight dorsal margin of the dentary is a syn- apomorphy shared by Pachyrhachis and mosa- saurs, but personal inspection of the holotype of Pachyrhachis did not allow us to determine the correct character state for this taxon, which is therefore coded as unknown (?). In addition, the dentary is slightly concave dorsally in the lower jaw of Platecarpus (fmnh UC 600), and it is distinctly concave in Prognathodon (Lingham-Soliar & Nolf, 1989), such that mo- sasaurs have to be coded polymorphic for this character. Character 113: Dentary with large (0), small (1), without (2) posterodorsal process extending onto lateral surface of coronoid process. As Un- derwood (1957) and Gauthier (1982) have em- phasized (see discussion above), there is only one logical place to put a joint in the lower jaw, i.e., between dentary and postdentary bones (in the following, all characters correlated with the differentiation of an intramandibular joint will be called dp-characters [for dentary-postden- tary relation]). Character 114: Meckel's canal an open groove (0), closed with dentary lips in sutural contact (1), closed with dentary lips fully fused (2). Character 115: Anterior end of Meckel's canal at anteroventral margin (0), on medial surface (1) of dentary. The opening of Meckel's canal on the medial surface of the lower jaw, instead of along its ventromedial margin, is a mosasaur (mosasauroid?) autapomorphy. As discussed above, the anterior end of Meckel's canal opens ventral relative to the sagittal plane of the man- dibular ramus in all snakes (as in all nonophi- dian squamates). Intramandibular muscles in- sert into Meckel's cartilage, which, in order to be an effective site of muscle attachment, is ex- posed ventromedially along the ventral edge of the lower jaw. Inspection of the holotype of Pa- chyrhachis did not reveal a position of Meckel's canal on the medial side of the dentary com- parable to the mosasaur condition. Character 116: Intramandibular septum of den- tary does not approach (0), approaches or reaches (1) posteriormost tooth position. In Varanus, the posterior margin of the intraman- dibular septum lies right below the posterior- most tooth; the same is true for mosasaurs (Lingham-Soliar & Nolf, 1989). As defined by Lee (1998), the character is not applicable (not comparable) to scolecophidians (Haas, 1964, 1968). A serially sectioned head of Anilius scy- tale shows the intramandibular septum to ter- minate just in front of the anterior end of the compound bone, i.e., it terminates well in front of the posterior tooth. The same is true for Cy- lindrophis. In uropeltids (serially sectioned head of Plecturus perroteti), the intramandibu- lar septum extends to a level posterior to the posteriormost teeth, which, by comparison to other basal alethinophidians, reflects the short- ened tooth row of uropeltids. Macrostomatans become difficult to compare because of the elongated posterior dentigerous process (see character 133, below). Character 117: Subdental shelf large (0), weak (1), absent (2). As described elsewhere (Zaher & Rieppel, 1999), the subdental shelf is a char- acter that applies only to nonophidian squa- mates with labial pleurodonty. It is absent in Varanus, mosasaurs, Pachyrhachis, and snakes. Character 118: Posterior margin of lateral surface of dentary without notch (0), with shallow notch (1), with deep notch (2). This is a dp- character, which loses weight in pulling snakes to mosasaurs plus varanoids if treated unor- dered, as it should be. The deeply bifurcated posterior end of the dentary of Pachyrhachis and snakes (2) is not part of a morphocline, but results from a restructuring of the intramandi- bular joint, with the compound bone (surangu- lar portion) becoming the supporting element for the dentary. To account for polymorphism in snakes, the character should be further sub- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 39 divided to include the elongated posterior den- tigerous process of the dentary, which was de- fined as a separate character above. In spite of these difficulties, we retain the character as de- fined and coded by Lee (1998). Character 119: Overlap of dentary with postden- tary bones extensive (0), reduced (1). This is another dp-character that is difficult to under- stand because the dentary-postdentary articu- lation is fundamentally different in varanoids, mosasaurs, and snakes. We propose the follow- ing redefinition of character states: dentary principally supported by coronoid, surangular, and prearticular (0, nonophidian squamates, in- cluding varanoids), by prearticular (1, mosa- saurs), by surangular (2, snakes). Character 120: Splenial large (0), small (1), ab- sent (2). Lee (1998) coded snakes for a reduced splenial, but we code scolecophidians and ma- crostomatan snakes for a large splenial (0). Character 121: Splenial overlaps postdentary bones and does (0), does not ( 1 ) expand beyond apex of coronoid process, or splenial does not substantially overlap postdentary bones (2). This is another dp-character. We find the splen- ial to substantially overlap the postdentary bones in Varanus in medial view of the man- dible. This overlap is reduced in Lanthanotus. In ventral view of the mandible, however, the articular carries a long anterior process that broadly overlaps with the splenial. As described above, the posterior margin of the splenial of Lanthanotus is concave in ventral view, receiv- ing the convex angular. In mosasaurs, there is no overlap between the splenial and the post- dentary bones, but the anterior surface of the splenial is concave, receiving the convex pos- terior head of the articular. In snakes, the splen- ial-postdentary relations are as described above, with extensive overlap of the splenial with the postdentary (compound) bone in sco- lecophidians. In summary, we propose the fol- lowing redefinition of character states: splenial overlaps with angular (0, all nonophidian squa- mates except mosasaurs, scolecophidians); splenial meets angular in an abutting contact, the splenial being the receiving, the angular be- ing the received element (1, autapomorphic for mosasaurs); splenial meets angular in an abut- ting contact, the angular being the receiving, the splenial being the received element (2, alethin- ophidians). Pachyrhachis is coded unknown (?) for this character because the detailed nature of the angular-splenial articulation remains un- clear (see discussion above). Character 122: Anterior tip of splenial on ventral margin (0), on medial surface (1) of dentary. As shown above, the position of the anterior tip of the splenial on the medial surface of the den- tary is an autapomorphy (uninformative char- acter) of mosasauroids and is therefore deleted from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 123: Extensive (0), reduced (1) contact between splenial and dentary. As defined by Lee (1998), this character is difficult to under- stand. Mosasaurs have a very extensive splen- ial-dentary contact. In other taxa, this contact varies according to the degree to which Meck- el's canal is closed medially by the splenial. However, in all squamates examined for this pa- per except mosasaurs, the splenial carries a lat- eral shelf, which underlies Meckel's canal and which, together with the ventral margin of the splenial, locks against the ventromedial margin of the dentary in a solid contact. In view of the autapomorphic relation of the splenial to the dentary in mosasaurs, the character becomes uninformative and is deleted from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 124: Splenial-angular contact, in me- dial view, overlapping and irregular (0), straight (vertical) and abutting (1). As defined by Lee (1998), we find this character to overlap with character 121. The restriction of the view to the medial side of the mandible is artificial and does not account for the complexity and the dif- ferences of the splenial-angular relations in the different taxa. As the splenial is the receiving part of the intramandibular articulation in mo- sasaurs and is the received part in snakes, this character is not simply a synapomorphy linking snakes to mosasaurs as coded by Lee (1998). Given the redefinition of character 121 above, we delete character 124 from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 125: Anteromedial process of coronoid long (0), short (1). As defined by Lee (1998), we believe this character to be misleading. As 40 FIELDIANA: GEOLOGY described above, the coronoid is V-shaped in cross-section (apex pointing dorsally) and strad- dles the surangular in nonophidian squamates, including mosasaurs. Its anterior contact is re- duced in mosasaurs as compared to Varanus, as is correctly coded by Lee. In snakes, the coro- noid is a simple bony plate that lies against the inside of the compound bone and hence has no chance to overlap with the medial surface of the dentary. In scolecophidians, the dentary lies lat- eral to the coronoid (Haas, 1964, 1968), while in basal alethinophidians, the posteroventral process of the dentary (if present — vestigial or absent in uropeltids) may extend backward to a level behind the anterior tip of the coronoid (if present) and at a morphological level lateral to the coronoid. As defined, this character is not applicable (not comparable) to snakes. Character 126: Anterolateral process of coronoid absent (0), present (1). As defined by Lee (1998), this character is not applicable (not comparable) to snakes for the same reasons as those discussed in relation to the previous char- acter. Character 127: Coronoid contacts (0), does not contact (1) splenial. The coronoid contacts the splenial in scolecophidians and in several basal macrostomatans. The splenial-coronoid contact appears to be variable in Cylindrophis ruffus. The specimen discussed above shows the ab- sence of such a contact, which however is shown to be present in another specimen by McDowell (1975, Fig. 6). The drawing of the lower jaw of Cylindrophis maculatus discussed above shows the coronoid to approach the splenial very closely; serial sections show the gap between the two bones to be only 0.1 mm, which is only marginally wider than any other skull suture (syndesmosis). Character 128: Ventral margin of coronoid straight or convex (0), concave (1). Character 129: Subcoronoid fossa (exposing sur- angular on medial side of mandible) absent (0), present (1). Characters 128 and 129: As defined by Lee (1998), dibamids should be coded 1 or poly- morphic and amphisbaenians polymorphic for character 128 (straight or concave ventral mar- gin of the coronoid). If this is done, the codings for characters 128 and 129 become practically redundant. However, a subcoronoid fossa, ex- posing the surangular on the medial surface of the lower jaw, is a character that is, in fact, not applicable (not comparable) to those taxa that form a compound bone composed of surangu- lar, prearticular, and articular. We opt for the retention of character 128 with corrected coding as indicated above and replacement of character 129 with a new character. (Our) Character 129: Compound bone formed of surangular, prearticular, and articular absent (0) or present (1, dibamids, amphisbaenians, Pachyrhachis, snakes); some amphisbaenians may show incomplete fusion of the postdentary bones (Zangerl, 1944; Montero et al., 1999). Si- neoamphisbaena shares the presence of a com- pound bone (Wu et al., 1996). Character 130: Surangular extends medially to the surface of the dentary terminating in a point (0), terminating with a blunt end ( 1 ), surangular terminates with blunt end but does not extend far medial to the dentary (2), surangular extends far into lateral surface of the dentary and ter- minates in a point (3). This is another dp-char- acter, which was miscoded for mosasaurs by Lee (1998). In Varanus, the surangular does ex- tend medial to the dentary and terminates blunt- ly; probably the same is true for Lanthanotus (the lower jaw was not disarticulated). In mo- sasaurs, the surangular has no overlap at all with the dentary; instead, the overlap is with the prearticular. Dibamids and amphisbaenians should be coded as unknown. As was done by Lee (1998), Pachyrhachis is coded like all other snakes except for scolecophidians. Character 131: Surangular without (0), with (1) dorsal flange overlapping posterior part of cor- onoid process. Character 132: Angular present (0), absent (1). Character 133: Angular with wide (0), with nar- row or without exposure ( 1 ) on medial side of mandible. Character 134: Fingerlike angular process present (0), absent ( 1 ). Character 135: Prearticular does not (0), does (1) extend anteriorly past posterior dentary tooth RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 41 position(s). This is another dp-character which, as coded by Lee (1998), is a synapomorphy shared by mosasaurs, Pachyrhachis, and snakes. However, in mosasaurs, the prearticular extends anteriorly far beyond the posterior teeth (1). Pachyrhachis has to be coded as unknown (?). Scolecophidians code as 0, and alethino- phidians as not comparable (?). Character 136: Prearticular crest absent (0), mod- erately well developed (1), prominent (2). Character 137: Adductor fossa faces dorsomedi- ally (0), dorsally (1). As coded by Lee (1998), this character is difficult to assess. Following the descriptions above, we propose the follow- ing redefinition of character states: medial mar- gin of the adductor fossa on lower jaw is low and rounded (0), developed into a distinct ver- tical flange (1). The coding for mosasaurs is 1, for Pachyrhachis is 1, and for scolecophidians is 0; anilioids are polymorphic (0 and 1), and basal macrostomatans are 1. Character 138: Adductor fossa narrow trans- versely (0), inflated transversely (1). This is a synapomorphy of the Lacertoidea that reflects the entry of the posterior adductor into Meck- el's canal. Character 139: Articular fused with prearticular and surangular (0), fused with prearticular (1), separate (2). As defined by Lee (1998), this character is wrongly polarized (see our discus- sion of character 129). Character 140: Retroarticular process in line with rest of mandible (0), offset medially (1). Character 141: Retroarticular process extends straight posteriorly (0), extends posteromedially (1). As defined by Lee (1998), the character states of this character are often difficult to es- tablish and appear to be redundant with char- acter 140. The retroarticular process of mosa- saurs is clearly deflected medially. Varanus has a rather straight posterior retroarticular process by comparison, but the retroarticular process of Lanthanotus is more distinctly medially deflect- ed. Lee (1998) coded Serpentes as 1 but sco- lecophidians as 0. By comparison to other squa- mates, Pachyrhachis shares with alethinophi- dians a reduced, knobby retroarficular process, which should be added as a different character state (2). Character 142: Dorsal surface of retroarticular process with pit or sulcus (0), smoothly con- cave (1). Lee's (1998) coding for Serpentes is problematic. Scolecophidians have a retroarti- cular process that is circular in cross-section (Haas, 1964, 1968), whereas alethinophidians, including Pachyrhachis, are not comparable, owing to their short, knobby retroarticular pro- cess. Furthermore, the knobby process has a convex, not a concave, dorsal surface. To avoid redundancy with character 141, we propose to treat character 142 as not applicable (not com- parable) to Pachyrhachis and snakes. Character 143: Dorsomedial margin of retroarti- cular process smooth (0), with distinct tubercle or flange (1). Character 144: Retroarticular process tapering, narrow distally (0), not tapering, broad distally (1). Character 145: Retroarticular process not twisted posteriorly (0), twisted posteriorly (1). As de- fined by Lee (1998), this character is not appli- cable (not comparable) to Pachyrhachis and snakes. Character 146: Marginal tooth implantation ac- rodont (0), pleurodont (1), thecodont, shallow alveoli (2), thecodont, deep alveoli (3). The pre- maxillary, maxillary, and dentary teeth of mo- sasaurs are not thecodont but modified (fully) pleurodont, as are those of varanoids. Scoleco- phidians likewise are pleurodont. Alethinophi- dia are derived by the attachment of the tooth base to circular interdental ridges, a character they share with Pachyrhachis (Zaher, 1998). According to our analysis of squamate tooth implantation, we propose the following redefi- nition of character states: tooth implantation ac- rodont (0), labially pleurodont (1), modified (fully) pleurodont (2, varanoids, mosasaurs, and scolecophidians), or teeth ankylosed to the rim of a shallow socket that is homologous to the interdental ridge of nonophidian squamates (3, Pachyrhachis, alethinophidians). Character 147: Marginal teeth without (0), with (1) sharp carinae. Varanus has laterally com- pressed teeth with anterior and posterior cutting 42 FIELDIANA: GEOLOGY edges. The same is true for mosasaurs, although their teeth are less laterally compressed. We find carinae to be either very weakly developed or absent in extant snakes. If anything is pre- sent, it is a weakly developed anterior cutting edge. Pachyrhachis has carinae on the lateral surface of the tooth crown, an autapomorphy of this taxon. As defined by Lee (1998), we find this character misleading and delete it from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 148: Plicidentine absent (0), present (1). Lee codes Pachyrhachis 1 for the presence of plicidentine and refers to Lee and Caldwell (1998) for justification. However, Lee and Cald- well (1998, p. 1537) state that "marginal teeth are hollow cones," which was confirmed by personal observation of the holotype of Pa- chyrhachis (particularly clear in the broken an- teriormost tooth of the left palatine). The pres- ence of weak striations on the enamel surface does not, in itself, indicate the presence of pli- cidentine, which is the character addressed by Lee (1998). The latter taxon, as all other snakes, therefore has to be coded for the absence of plicidentine, the presence of which is a vara- noid synapomorphy. Character 149: Tooth crowns closely spaced (0), separated by wide gaps (1). As defined by Lee (1998), this character carries little phylogenetic information. The modified (fully) pleurodont teeth of varanoids have a flaring tooth base, and although the teeth meet each other at their base (i.e., narrow spacing of the tooth positions), the flaring of that base still results in a wider spac- ing of the tooth crowns than is characteristic for nonophidian squamates, which show labial pleurodonty (Zaher & Rieppel, 1999). In addi- tion, the rhythm of tooth replacement in Var- anus is timed such that the functional teeth tend to alternate with replacement teeth, which cre- ates gaps between the functional teeth, although tooth positions are closely spaced (Edmund, 1960). Mosasaurs again show a basal contact between the teeth, i.e., closely spaced tooth po- sitions, but because of a flaring tooth base, the tooth crowns appear more widely spaced. Flar- ing tooth bases and a basal contact between teeth are a derived character shared by mosa- saurs and varanoids but are absent in other squamates. In spite of these problems, we retain this character in our analysis, although it ap- pears at least partially redundant with the type of tooth implantation (character 146). Character 150: Replacement tooth positioned lin- gual (0) to functional tooth, posterolingual (1) to functional tooth, or absent (2). Character 151: Resorption pits at base of teeth (0), on bony tooth pedicel (1), absent (2). The development of resorption pits starts at the base of the teeth, i.e., in the bone of attachment, in all squamates (snakes included). However, var- anoids lack the development of large resorption pits within which the replacement teeth develop and which extend into the tooth crown. Mosa- saur teeth are autapomorphic in that they de- velop large resorption pits that hold the devel- oping replacement teeth but remain restricted to the tall tooth base formed from the bone of at- tachment. Scolecophidians develop large re- sorption pits, but other snakes have small re- sorption pits, restricted to the basal bone of at- tachment. In light of these variations, we offer the following redefinition of character states: re- sorption pits large, extending into tooth crown (0); resorption pits small, restricted to bone of attachment at the base of the tooth (1, vara- noids, alethinophidians); resorption pits large, restricted to tall tooth pedicel composed of the bone of attachment (2, autapomorphic for mo- sasaurs). Character 152: Replacement teeth erupt upright (0), erupt horizontally (1). The horizontal po- sition of replacement teeth is a synapomorphy of snakes (unknown in Pachyrhachis) that is absent in mosasaurs (Zaher & Rieppel, 1999). Character 153: Five or more (0), four or fewer (1) premaxillary teeth. Character 154: Median premaxillary tooth absent (0), present (1). Character 155: Median premaxillary tooth not enlarged (0), enlarged (1). Character 156: Premaxillary teeth similar in size to (0), distinctly smaller than ( 1 ) anterior max- illary teeth. Character 157: Thirteen or more (0), twelve to nine (1), eight or fewer (2) maxillary tooth po- sitions. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 43 Character 158: Thirteen or more (0), twelve to nine ( 1 ), eight or fewer (2) dentary tooth posi- tions. Character 159: Palatine teeth present (0), absent (1). Character 160: Palatine teeth small (0), similar in size to marginal teeth (1). Character 161: Pterygoid teeth present (0), absent (1). Character 162: Pterygoid teeth small (0), similar in size to marginal teeth (1). As defined by Lee (1998), this character is ambiguous because marginal teeth decrease in size from front to back. Pterygoid teeth, where present, are al- ways smaller than anterior maxillary or dentary teeth in all squamates. For this reason, we de- lete this character from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 163: Egg tooth single (0), paired (1). Character 164: Centra not constricted (0), slightly constricted (1), strongly constricted (2) in front of condyle. Character 165: Vertebral condyles facing poste- riorly (0), slightly dorsally (1), dorsally (2). Character 166: Shape of vertebral condyles in middorsal region oval (0), round (1). Lee (1998) coded snakes for a circular vertebral condyle (1), but scolecophidians, various ani- lioids, and macrostomatans all have a vertebral condyle with an oval outline and are coded ac- cordingly (0). Character 167: Centra notochordal (0), not no- tochordal (1). Character 168: Neural spines tall processes (0), low ridges (1). Character 169: Zygosphenes and zygantra pre- sent (0), absent (1). Character 170: Articular surface of zygosphene faces dorsally (0), lateroventrally (1). Characters 169 and 170: As defined by Lee (1998), these characters are redundant and have to be combined into one: zygosphene-zygan- trum absent (0); present with zygosphene artic- ular surface facing laterodorsally ( 1 , Lacertidae, Cordylidae, Gerrhosauridae: Hoffstetter & Gasc, 1969, Fig. 42); present with zygosphene articular surface facing ventrolaterally (2, mo- sasaurs, snakes, Teiidae, Gymnophthalmidae). We code character 169 accordingly but delete character 170 from the analysis (the coding re- tained in the data matrix is that of Lee, 1998). Character 171: Dorsal intercentra present (0), ab- sent (1). Character 172: Presacral vertebrae 22 or fewer (0), 23 to 25 (1), 26 (2), 27 to 50 (3), 50 to 1 19 (4), 120 or more (5). Character 173: Cervical vertebrae seven or fewer (0), eight (1), nine or more (2). Character 174: Hypapophyses present on fourth to sixth presacral (0), on seventh presacral and beyond (1). As defined by Lee (1998), this character is based on arbitrary morphological distinctions. The number of cervical hypapo- physes is increased in varanoids because of an elongation of the neck (character 173). Note that Varanus has nine cervicals with hypapo- physes, whereas mosasaurs have eight or seven cervicals (Russell, 1967). According to Russell, only the anterior six or seven cervicals carry hypapophyses in mosasaurs (in Mosasauridae indet., fmnh PR 2103, the last hypapophysis is on the sixth cervical). Snakes are not compa- rable because they have no easily defined cer- vical region of the vertebral column and the hy- papophyses extend backward far into the trunk, suggesting the presence of dorsal intercentra (absent in nonophidian squamates other than geckos). In the posterior trunk region, the hy- papophyses are reduced to a hemal keel, but they may be reduced more anteriorly also in burrowing species (Hoffstetter & Gasc, 1969). A redefinition of character states could account for the number of cervical vertebrae (not appli- cable, i.e., not comparable to snakes) or the presence versus absence of trunk intercentra. In view of the difficulties of establishing clear-cut character state relations beyond autapomor- phies, we delete this character in our analysis (the coding retained in the data matrix is that of Lee, 1998). 44 FIELDIANA: GEOLOGY Character 175: Dorsoposterior flange on atlas neural arch present (0), absent (1). Character 176: Cervical intercentral not sutured nor fused (0), sutured (1), fused (2) to preced- ing centrum. Character 177: Cervical intercentra neither su- tured nor fused (0), sutured (1), fused (2) to following centrum. Character 178: Caudal transverse processes sin- gle (0), double ( 1 ) in some caudals. Character 179: Caudal transverse processes: two prongs converge (0), diverge (1) distally. Character 180: Caudal transverse processes pro- ject laterally or posterolaterally (0), anterolat- eral^ (1). Character 181: Caudal autotomy septa present (0), absent (1). Character 182: Caudal autotomy septa anterior to or within (0), posterior to (1) transverse pro- cesses. The distinction of two character states appears arbitrary because, in some nonophidian squamates, the autotomy septum is anterior, in others it is within, and in still others it is pos- terior to the transverse processes (the coding retained in the data matrix is that of Lee, 1998). Character 183: Pedestals on caudal vertebrae for chevrons weakly developed (0), prominent (1). Character 184: Chevrons articulate with (0), fused to (1) caudal centra. Characters 183 and 184: For reasons discussed below (character 185), we consider chevrons to be absent in snakes rather than co-ossified with the centrum, as was assumed by Lee (1998). Character 185: Caudal chevron positioned at (0), in front of (1) posteroventral margin of cen- trum. As defined by Lee (1998), this character is not applicable (not comparable) to snakes, which have no chevrons but have hemapophys- es instead (Hoffstetter & Gasc, 1969), and these are located at the posterior end of centrum. As coded by Lee (1998), mosasaurs share with var- anoids the anterior shift of the chevrons. Character 186: First rib on third (0), fourth (1) cervical vertebra. This character is not appli- cable to snakes. Character 187: Proximal end of rib without (0), with (1) anteroventral pseudotuberculum. Character 188: Proximal end of rib without (0), with (1) posteroventral pseudotuberculum. Character 189: Lymphapophyses ("forked cloa- cal ribs" of Lee, 1998) absent (0), present (1). Lee and Caldwell (1998) described a distally expanded sacral rib for Pachyrhachis and hy- pothesized that the appearance of its distal bi- furcation might be due to breakage. Lee (1998) codes the distally forked sacral rib of Pachy- rhachis as comparable to the distally forked clo- acal ribs of other snakes, amphisbaenians, and dibamids. Indeed, Pachyrhachis shows at least one well-developed lymphapophysis (the sacral rib of Lee & Caldwell, 1998), but the presence of additional, more posteriorly located lympha- pophyses cannot be assessed owing to poor and/ or incomplete preservation. In those basal snakes that retain limb rudiments, these do not establish contact with the lymphapophyses, and the same might have been true of Pachyrhachis. Character 190: Scapulocoracoid large (0), re- duced (1), absent (2). Character 191: Emarginations on anterodorsal edge of scapula absent (0), present ( 1 ). Character 192: Anterior coracoid emargination absent (0), present (1). Character 193: Posterior coracoid emargination absent (0), present (1). Character 194: Clavicle present (0), absent (1). Character 195: Clavicle follows anterior margin (0), curves anteriorly away from (1) scapulo- coracoid. Character 196: Clavicles rodlike (0), expanded proximally with proximal notch or fenestra (1). Character 197: Interclavicle present (0), absent (1). Character 198: Interclavicle cross-shaped (0), simple rod ( 1 ). RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 45 Character 199: Anterior process of interclavicle short or absent (0), long (1). Characters 197 and 198: The interclavicle of nonophidian squamates shows greater complex- ity than is expressed by these two characters (Camp, 1923). The absence of an anterior pro- cess (character 199) may result in a T-shaped or arrow-shaped interclavicle (the coding re- tained in the data matrix is that of Lee, 1998). Character 200: Ossified (should read calcified) sternum present (0), absent (1). Character 201: Sternal fontanelle absent (0), pre- sent (1). Character 202: Five (0), four (1), three (2), two or fewer (3) pairs of ribs attach to the sternum. Character 203: Postxiphisternal inscriptional ribs not united (0), one or more pairs united along ventral midline (1). Character 204: Forelimbs large (0), small or ab- sent (1). Character 205: Ectepicondylar foramen on hu- merus present (0), absent (1). Character 206: Pelvis large (0), reduced (1), ab- sent (2). Character 207: Pelvic elements co-ossified into single bone (0), strongly sutured to one another (1), weakly united in nonsutural contacts (2). As defined by Lee (1998), this character is dif- ficult to assess throughout squamates. The skel- eton of mosasaurs, like that of other marine rep- tiles, is subject to skeletal paedomorphosis (Sheldon, 1997; see also Rieppel, 1993a), which accounts for the reduced ossification of the pelvic elements (joined together by cartilage in life). However, the pelvic elements are firmly sutured to one another in basal mosasauroids (aigialosaurs: Carroll & deBraga, 1992), which were coded accordingly (1). The pelvis of Pa- chyrhachis is much reduced by comparison to that of mosasaurs, which could be a conse- quence of its marine habits or of its being a snake, or both. The pelvic rudiments of other snakes are not easily comparable. Character 208: Sacral blade of ilium with (0), without (1) anterior process. Character 209: Pubis short, symphyseal process directed ventrally (0), intermediate in length, symphyseal process directed anteroventrally (1), elongated, symphyseal process directed an- teriorly (2). As defined by Lee (1998), this character is difficult to understand. The pubis points anteroventrally in all nonophidian squa- mates (except mosasaurs), with a medial incli- nation to form the pubic symphysis. In mosa- saurs, the pubis lies horizontally and points me- dially to form the symphysis (Russell, 1967). The pubis in Pachyrhachis is dislocated and its natural orientation unknown. It also remains un- known whether Pachyrhachis had a pubic sym- physis. To these problems of comparison, we add the observation that coding of the hind limb in Pachyrhachis can severely skew the analysis, depending into how many characters the hind limb is atomized. We delete character 209 from our analysis (the coding retained in the data ma- trix is that of Lee, 1998). Character 210: Pubic tubercle on posterodorsal end of pubis (0), on shaft of pubis (1). Character 211: Hind limbs well developed (0), rudimentary or absent (1). We adopt Lee's (1998) coding (1) for Pachyrhachis but note that the incompleteness of its limb (all elements distal to astragalus and calcaneum missing) ap- pears to be an artifact of preservation. Character 212: Femur gracile (0), stout (1). Lee (1998) coded a stout femur as a derived char- acter state shared by Pachyrhachis and mosa- saurs. Although we retain his coding, we note that the femur of mosasaurs is modified to form the proximal element in a paddle; that of Pa- chyrhachis is crushed (see also character 213, below). Character 213: Femur curved (0), not curved (1) in dorsoventral plane. Lee (1998) coded Pa- chyrhachis and mosasaurs for a straight femur. In fact, the two taxa are not comparable in fe- mur morphology. The femur of mosasaurs is much reduced in length but broadened as a con- sequence of the limb having been transformed into a paddle. By comparison, the femur of Pa- chyrhachis is elongate and relatively slender (some of the robustness of the femur of Pa- chyrhachis is due to crushing). The diaphysis of mosasaurs retains a distinct biconcave shape, which is not expressed in Pachyrhachis (anoth- 46 FIELDIANA: GEOLOGY er consequence of crushing?). Given the com- pression of the femur, the character state for character 213 also cannot be identified for Pa- chyrhachis. In view of these differences, and because this character is partially redundant with character 212, we delete character 213 from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 214: Distal end of tibia gently convex (0), with notch fitting into a ridge on astragalo- calcaneum (1). As defined by Lee (1998), this character is difficult to understand. In fact, this character reflects little more than the absence of separate epiphyseal ossification centers in mo- sasaurs, which is again due to skeletal paedo- morphosis, whereas Pachyrhachis may lack separate epiphyseal ossification centers because of its marine habits or because it is a snake, or both (all snakes lack separate epiphyseal ossi- fication centers: Haines, 1969). As such, this character is redundant with character 228 and hence is deleted from our analysis (the coding retained in the data matrix is that of Lee, 1998). Character 215: Astragalus and calcaneum fused (0), separate ( 1 ) in adult. This character is cod- ed by Lee (1998) as another putative synapo- morphy of mosasaurs and Pachyrhachis, but the lack of fusion of astragalus and calcaneum in mosasaurs is again due to skeletal paedomor- phosis (marine). The character is furthermore subject to ontogenetic variation among other nonophidian squamates yet is not applicable (not comparable) to other snakes. To these problems of comparison, we add the observa- tion that coding of the hind limb in Pachy- rhachis can severely skew the analysis, depend- ing on whether features of the hind limb are lumped into few or split into many characters. In spite of these problems, we retain this char- acter, as it adds to the strength of the test of the phylogenetic position of Pachyrhachis as sister taxon of Macrostomata (see further comments below). Character 216: Dorsal body osteoderms absent (0), present (1). Character 217: Ventral body osteoderms absent (0), present (1). Character 218: Separable cranial osteoderms ab- sent (0), present on periphery of skull table ( 1 ), present across entire skull table (2). Character 219: Separable cranial osteoderms few and large (0), many and small (1). Character 220: Separable cranial osteoderms tightly connected to skull roof (0), loosely con- nected to skull roof ( 1 ). Character 221: Rugosities on skull roof formed by overlying cephalic scales absent (0), with vermiculate sculpture (1), without vermiculate sculpture (2). Character 222: Scleral ossicles present (0), absent (1). Character 223: Fifteen or more (0), 14 (1), 13 or fewer (2) scleral ossicles. Character 224: Scleral ossicle shape complex and irregular (0), square ( 1 ). Character 225: Second epibranchials present (0), absent (1) in hyoid skeleton. Character 226: Second ceratobranchials present (0), absent (1) in hyoid skeleton. Character 227: Epiphyses on skull and axial skel- eton present (0), absent (1). Character 228: Epiphyses on appendicular skel- eton present (0), absent (1). Character 227 and 228: As defined and coded by Lee (1998), these two characters are redundant and should be coded as a single character. The only difference recorded by Lee (1998) is the polymorphic coding of mosasaurs for the pres- ence of separate epiphyseal ossification centers in the appendicular skeleton of mosasaurs. Bell (1997) describes the variable development of epiphyses on postcranial elements of mosa- saurs, but from his description it is not entirely clear what exactly his understanding is of the term epiphysis or, more exactly, of a separate epiphyseal ossification center. We were unable to confirm the presence of separate epiphyseal ossification centers on postcranial elements of mosasaurs in the Field Museum collections. Separate epiphyseal ossification centers are pre- sent, however, in stem-group mosasauroids RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 47 (Carroll & deBraga, 1992). In summary, char- acters 227 and 228 are therefore redefined as a single character (227): separate epiphyseal os- sification centers present (0, nonophidian lepi- dosaurs), absent (1, snakes). Because of the am- biguities of description and the presence of epiphyses in stem-group mosasauroids, mosa- saurs are coded polymorphic for this character in our analysis. Character 228 is deleted from our analysis (the coding retained in the data ma- trix is that of Lee, 1998). Character 229: Epiphyses fuse to diaphyses of long bones at the same time or after (0), before (1) fusion of braincase elements. Character 230: Postcloacal bones absent (0), pre- sent (1). To those characters of Lee, we add the follow- ing three characters, which emerged from the character discussion above: Character 231: Medial ventral flanges of frontal, separating olfactory tracts from one another, ab- sent (0), present (1), This character is a syna- pomorphy of Alethinophidia, absent in Dinily- sia (Estes et al., 1970), and unknown in Pa- chyrhachis. Character 232 (new): Cartilaginous processus as- cendens of supraoccipital present (0, mosasaurs, Varanus, indeed all nonophidian squamates ex- cept Gekkota and Dibamus), absent (1, Gek- kota, Dibamus, snakes). Amphisbaenia (and 57- neoamphisbaena), as well as Pachyrhachis and Dinilysia, have to be coded as unknown (?). Character 233: Elongated posterior dentigerous process of dentary absent (0), present (1). This character is a potential synapomorphy shared by Pachyrhachis and macrostomatans (Zaher, 1998). The list of characters used by Lee (1998) does not include some features that were subject to controversy in the analysis of the relationships of Pachyrhachis with- in squamates by Caldwell and Lee (1997), Lee and Cald- well (1998), and Zaher (1998). We propose to critically review these characters in comparison to those of Lee (1998). Below follows a list of the characters used by Cald- well and Lee (1997; abbreviated as CI to C8) in support of a sister-group relationship of scolecophidians and al- ethinophidians at the exclusion of Pachyrhachis. The number in parentheses preceded by a D refers to the corresponding character in Lee and Caldwell (1998). CI (Dl): Jugal present (0), absent (1). A jugal is here considered to be absent in Pachyrhachis. This feature is coded under character 12 (see above). C2 (D3): Posterior orbital margin complete (0), in- complete (discontinuous) ( 1 ). We interpret the purported jugal in Pachyrhachis to be part of the ectopterygoid (its anterior ramus). Pachyrhachis had an incomplete pos- terior orbital margin similar to boids (i.e., with a long postorbital, almost touching the dorsal surface of the ec- topterygoid). This character is coded under character 29 (see above). C3 (D4): Exoccipitals not in contact (0), in contact (1) dorsal to the foramen magnum. As pointed out by Zaher (1998) and acknowledged by Lee (1998, p. 442), this character cannot be scored in Pachyrhachis owing to the poorly preserved condition of the skull in this area. This character is thus not included in the present analysis. C4 (D5): Angular-coronoid contact absent (0), pre- sent (1). The contact is absent in Leptotyphlops (Mc- Dowell & Bogert, 1954, Fig. 13; personal observation), uropeltids (Figs. 9, 10), Xenopeltis (Hoge, 1964, Figs. 1, 2), Corallus (Kluge, 1991, Fig. 14), and Candoia (McDowell, 1979, Fig. 4), and it is variable in Typhlops (see Haas, 1930, Figs. 34, 41), Cylindrophis (Figs. 7, 8), Loxocemus (McDowell, 1975, Fig. 6; Kluge, 1991, p. 37), Eunectes (personal observation), and Boa (Kluge, 1991, p. 37; Fig. 14). Contrary to Lee's (1998, p. 442) claim that Kluge (1991) indicated "that the contact is present in anilioids," this author (Kluge, 1991, p. 37) pointed out that only some erycines (Eryx) and anilioids (Cylindrophis) "have a coronoid-angular contact." Per- sonal observations corroborated Kluge's observations on erycines and demonstrated that Cylindrophis is variable in respect to this character. All snake terminal taxa (ex- cept Dinilysia, as pointed out by Lee, 1998) used in this study, including basal anilioids and basal macrostoma- tans, are variable with respect to this character, preclud- ing its use in a phylogenetic analysis. C5 (D6): Mental foramina on dentary, two or more (0), one (1). As already pointed out above (see character 111), Pachyrhachis retains only one foramen, as in other snakes. This character is coded under character 29 (see above). Character 234 (C6 [D8]): When present, the pel- vis is external to the rib cage, sacral contacts usually present (0), lies within rib cage, sacral contact absent (1). This feature has been added to the data matrix as character 234. Dibamids show state 1. Character 235 (C7 [D10]): Femur well developed (0), small (1), vestigial or lost (2). This feature has been added to the data matrix as character 235. 48 FIELDIANA: GEOLOGY Character 236 (C8 [Dll]): Tibia, fibula, astraga- lus, and calcaneum present (0), absent (1). This character has been added to the data matrix as character 236. Dibamids are coded 0, although tibia and fibula only are present in the hind limb of Dibamus (Greer, 1985). Characters 235 and 236: We note that the coding of the limb in Pachyrhachis may be problem- atic because of the potential of oversplitting the character of the presence of a limb. It is obvious that the more limb characters that are included in the analysis, the more Pachyrhachis will be pulled toward the root of the ophidian clade, as its most basal member (Lee & Caldwell, 1998). However, in this analysis, we opt for the reten- tion of these limb characters, as they add to the severity of the test of the hypothesis that Pa- chyrhachis is the sister taxon of Macrostomata (Zaher, 1998). Lee and Caldwell (1998, p. 1550) proposed three additional characters: D2: Squamosal present (0), absent (1). We interpret the squamosal of Pachyrhachis as identified by Lee and Caldwell (1998) as the shaft of the stapes. Pachyrhachis lacks a squamosal. This feature is coded in the present study as character 40 (see above). D7: Neural spines well developed (0), reduced (1). Pachyrhachis and macrostomatans share character state 1. whereas the other snakes (including Dinilysia) show character state 0. This feature has been coded as char- acter 168 (see above). Lee and Caldwell (1998, p. 1550) also listed four characters (El to E4) found in Pachyrhachis and alethinophidian snakes. Among these, three (El, E2, E3) were discussed by Zaher (1998). Only character E4 is added here as follows: Below follows a list of the characters used by Zaher (1998) and retained in the present study: 18: Quadrate anteriorly directed (0), vertically or pos- teriorly directed ( 1 ). This character was discussed above as character 50. Character 238 (19): Tooth-bearing anterior pro- cess of the palatine absent (0), present ( 1 ). This character is added to the data matrix as char- acter 238. 20: Free-ending process of the supratemporal absent (0), present (1). This feature has been coded as character 47 (see above). 21: Dorsal surface of the prootic not concealed (0), concealed by the supratemporal ( 1 ). This character is here deleted from the analysis, as it is correlated with the elongation of the supratemporal (Lee, 1998). 22: Basipterygoid process well developed, with artic- ulating surface facing more laterally than vcntrally (0). reduced, with articulating surface facing vcntrally ( 1 ). This feature has been discussed under character 74 (sec above). Character 233 (23): Posterior dentigerous process of the dentary absent (0), short (1), enlarged (2). This feature has been modified to a binary character (dentigerous process absent [0] or present [1]) and has been added to the data ma- trix as character 233. Character 239 (24): Suprastapedial process of the quadrate present (0), absent (1). Contrary to Lee (1998, p. 443), a suprastapedial process is present and well developed in typhlopids, less developed in anomalepids, and ill-defined or absent in leptotyphlopids. Scolecophidians are thus considered to retain a suprastapedial pro- cess. This feature has been added to the data matrix as character 239. Character 240: Jugal present (0), absent ( 1 ). For further details see the discussion of character 12. Character 237 (E4): Palatine short and broad (0), narrow and long ( 1 ). As pointed out by Lee and Caldwell (1998), the palatine of Pachyrhachis and alethinophidian snakes is a long, narrow el- ement. It is short and broad in Dinilysia and short (vestigial?) in scolecophidians. This char- acter is added to the present data matrix as char- acter 237. Cladistic Analysis The cladistic analysis presented below is not intended to assess global squamate interrelation- ships, but should rather be viewed as a test of the conclusions reached by Lee (1998). As indicated in the character discussion above, we believe that RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 49 Table 1. The data matrix used to analyze the interrelationships of the fossil snake Pachyrhachis. Character definitions and discussion are given in the text. Pachyrhachis 1 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 iqn./de 1 Kuehneosauridae ? 9 0 0 0 0 0 0 0 9 ? 0 0 0 0 2 Marmoretta 0 0 0 0 0 0 0 0 0 9 ? 0 0 ? 0 3 Rhynchocephalia ? 1 0 0 0 0 0 0/1 0 0 0 0 0 0 0 4 Ancestor 0 0/1 0 0 0 0 0 0 0 0 0 0 0 0 0 5 Iguanidae 0 0 0 0/1 0 0 0 0 0 0 0 0 1 0 0 6 Agamidae 0 0/1 0 0 0 0 0 0/1 0 0 0 0 0/1 0 0 7 Chamaeleonidae ? f 0 0 0 0 0 0/1 0 0 0 0 1 0/1 0 8 Xantusiidae 1 0 0 0 0 0 ? ? 0 0 0 0 0 1 9 Gekkonidae 1 0 0 0 0 0 1 9 0 0 0 0/1 0 0/1 1 0 Pygopodidae 0/1 0 0 0 0 0 1 ? 0 0 0 0 0 0/1 1 1 Sineoamphisbaena ? 0 0 0 1 0 0 1 0 ? 0 1 0 0 1 2 Dibamidae 0 0 0 0 0 0 1 9 0 0 1 ? 0 1 1 3 Amphisbaenia 1 0 0 0 0/1 0/1 1 9 0 0 1 9 0 1 1 4 Lacertidae 1 0 1 0 0 0 0 0 0 0 0 0/1 0 1 1 5 Teiidae 1 0 1 0 0 0 0 0 0 0 0 1 0 0/1 1 6 Gymnophthalmidae 0 0 1 0 0 0 0/1 0/1 0 0 0 1 0 0/1 1 7 Cordylidae 0/1 0 0 0 0 0 0 0 0 0 0 0/1 0 0 1 8 Scincidae 0/1 1 0 0 0 0 0 0/1 0 0 0 0 0/1 0 0/1 1 9 Anguidae 0/1 1 0 0 0 0 0 0 0 0 0 0 1 0 0/1 2 0 Xenosauridae 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2 1 Heloderma 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 22 Lanthanotus 0 0 0 0 1 0 1 0 0 1 0 0 1 1 0 2 3 Varanus 0 0 0 0 1 0 1 0 0 1 0 0 1 0/1 • 9 2 4 Mosasauroidea 0 0 0 0 1 0 1 0 0 0 1 0 1 0 ? 25 Pachyrhachis 9 9 0 0* 0 0 9 9 0 1 * ?• ? 9 2 6 Scolecophidia 0 ? 0 0 0 0 9 9 0 1 9 0/1 0 27 Dinilysia 0 9 ? 0 0 0 9 9 0 0* 9 0 0 2 8 Anilioidea 0 ? 0 0 0 0 9 9 0 1 9 0 0 29 Macrostomata 0 9 0 0 0 0 9 9 0 1 9 0 0 Pachyrhachis 2 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 3 O ord. ord. ord 1 Kuehneosauridae 0 0/1 1 0 0 0 0 0 0 0 0 0 0 0 0 2 Marmoretta 0 0 0 0 9 0 0 0 0 0 9 0 0 0 1 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0/1 4 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0/1 5 Iguanidae 0 1 0 1 1 1 0 1 12 0 0 0 0 0 0 1 6 Agamidae 0 1 0 1 1 1 0 2 ? 0 0 0 0 0 1 7 Chamaeleonidae 1 1 0 0 1 0 0/1 2 9 0 0 0 0 0 1 8 Xantusiidae 0 0 0/1 1 0 1 0 0 9 1 0 1 ? 0 0/1 9 Gekkonidae 0 0 0 1 0 1 0 0 9 1 0 1 9 2 0/1 1 0 Pygopodidae 0 0 0 0/1 0 1 0/1 0 ? 1 0 1 9 2 0 1 1 Sineoamphisbaena 0 1 0 0 1 1 0 1 0 0 0 0 0 0 1 2 Dibamidae 0 0 0 0 1/3* 0 1/2 9 0&1 0 1 9 2 1 1 3 Amphisbaenia 0 0 0/1 0 0* 0 2 ? 0 0 1 9 2 1 1 4 Lacertidae 0 0 1 0 0 0 0 ? 1 1 1 9 0 1 1 5 Teiidae 0 0 1 0 0/1 0 0 0/1 1 0 0 1 0 1 1 6 Gymnophthalmidae 0 0 1 0 0 0 0 0 1 0 0 1 0 1 1 7 Cordylidae 0 0 0/1 1 0 0/1 0 0 0 1 0/1 0 1 0 1 1 8 Scincidae 0 0 1 0 1 0 0 0/1 1 0/1 0/1 1 0 1 1 9 Anguidae 0 0 0/1 1 0 1 0/1 0 0/1 1 1 0 1 0 1 20 Xenosauridae 0 0 0/1 1 0 1 0 0 1 1 1 0 1 0 1 2 1 Heloderma 0 0 0 0 1 1 0 ? 0 0 1 9 0 1 2 2 Lanthanotus 1 * 0 0 0 1 1 0 9 1 0 1 9 0 1 2 3 Varanus 2 0 1 0 1 0 0 1 1 1 0 1 1 1 24 Mosasauroidea 2 0 1 ' 0 2* 0 0 1 1 0/1 0 1 0 1 2 S Pachyrhachis 7* 0 0 9 1 ?' ?• ?* ?' 0 0 0 1 1 2 6 Scolecophidia 1 0 0 0 1 &3 9 2 9 9 0 1 9 ? 0/1 27 Dinilysia ? 0 0 0 0 0 1 0 9 0 0 1 2 1 2 8 Anilioidea 1 0 0 0 3 1 2 9 ? 0 0&1 1 2 1 2 9 Macrostomata 1 0 0 0 1 0 2 9 ? 0 0 0&1 1 &2 1 50 FIELDIANA: GEOLOGY Table 1. Continued. Pachyrhachla 3 3 1 32 33 34 35 36 37 38 3 9 40 4 1 42 4 3 44 4 5 ord. ord. 1 Kuehneosauridao 1 0 0 1 1 0 0 0 0/1 0 0 0 0 0 7 2 Marmoretta 0 0 1 ? 2 ? ? 0 0 ? ? 0 0 7 ? 3 Rhynchocephalia 0 0 0 0 1 0/1 0 0 1 0 0 0 0 0 0 4 Ancestor 0 0 0/1 0/1 1/2 0 0 0 0/1 0 0 0 0 0 0 5 IguankJae 0/1 0 0 1 0/1 0 0 0 1 0 0 0 0 1 6 Agamidae 0/1 0 0 1 0/1 0 0 0 1 0 0 0 0 1 7 Chamaeleonidae 0/1 0 0/1 2 1 /2 1 0 0 1 0 0 0 0 1 8 Xantusiidae 0 0 0/1 0 0 1 1 0 0 0 1 0 1 9 GekkonkJae 1 0 1 ? 0 0 0 / 1 1 ? 0 0 0 0/1 1 0 Pygopodidae 1 0 1 ? 0 0 0 1 ? 0 0 0 0 7 1 1 Sineoamphisbaena 1 0 1 7 0 0 0 0 1 0 1 0 0 7 1 2 Dibamidae 1 1 1 7 2 0 0/1 1 ? ? 0 0 7 7 1 3 Amphisbaenia 1 1 1 0 2 0/1 1 1 ? 1 0 0 0 7 1 4 Lacertidae 0 0 0 0 0 1 0 0 0 0 0 1 1 5 Teiidae 0 0 0 0 1/2 0 0 0 0/1 0 0 0 1 6 Gymnophthalmidae 0 0 1 7 0/1 0 0 0 0 0 0/1 0 1 7 Cordylidae 0/1 0 0/1 0 0 0/1 0 0 0 0 1 0 1 8 Seine idae 0/1 0 0 0 0 0 0 0 0 0 0 1 1 9 Anguidae 0 0 0 0/1 0 0 0 0 0/1 0/1 0 0/1 2 0 Xenosauridae 0 0 0 0/1 0/1 0 0 0/1 0 0/1 0 0 2 1 Heloderma 0 1 ? 0 0 0 1 ? 0 0 0 2 2 Lanthanotus 0 1 7 1 0 0 1 ? 0 0 0 23 Varanus 0 0 0 1 0 0 0 0 0 0 0 24 Mosasauroidea 0 0 0 1 7 0 0 0 0 0/1 0 0 25 Pachyrhachis 0 1 7 2 0 1 ? 1 • ?• ?• ?• 0 2 6 Scolecophidia 0 1 7 2 0 1 ? 7 ? ? 0 27 Dinilysia 0 1 7 2 0 1 ? ? ? ? 0 28 Anilioidea 1 1 7 2 0 1 ? ? ? ? 0 29 Macrostomata 0 1 ? 2 0 1 ? ? ? 7 0 Pachyrhachla 4 4 6 47 48 49 50 5 1 52 5 3 54 5 5 5 6 57 58 59 60 • delete ord. ord. delete 1 Kuehneosauridao ? ? 4 0 0 0 0 ? 0 7 7 0 7 2 Marmoretta ? ? 0 0 0 0 ? 0 7 7 0 7 3 Rhynchocephalia 0 0 0 0 0 0 0 0 7 0 0 0 4 Ancestor 0 0 0/4 0 0 0 0 0 0 7 0 0 0 5 kguanidae 0 0 0 0/1 0 0 0 0 7 0 0 6 Agamidae 0 0 0 1 0 0 0 0 7 0 0 7 Chamaeleonidae 0 0 2 1 0 0 0 0 7 0 0 8 Xantusiidae 0 0 0 1 0/1 0 1 0 0 0 9 Gekkonidae 0 0 0 1 1 0 0 7 0 0 1 0 Pygopodidae ? 7 0 1 1 0 0 7 0 0 1 1 Sineoamphisbaena 7* ? ? 2 1 1 1 1 0 0 0/2 1/2 1 2 Dibamidae o • 7 ? 2 0/1 1 0 1 1 0 0 7 1 3 Amphisbaenia 7* 7 7 1 • 2 1 1 1 1 1 0 2 2 1 4 Lacertidae 0 0 0 0 0 0 0 7 0 0 1 5 Teiidae 0* 0 0 0 0 0 0 1 0 0 0 1 6 Gymnophthalmidae 0 0 0 0 0/1 0 1 0 0 0 1 7 Cordylidae 0 0 0 0 0 1 0 0 0 1 8 Scincidae 0 0 0 0 0 0/1 0 0 0 1 9 Anguidae 0 0 0/1 0 0 0/1 1 0 0 2 0 Xenosauridae 0 0 0 0 0 0 7 0 0 2 1 Heloderma 0* 0 0 0 1 0 0 ? 0 0 22 Lanthanotus 0' 0 1 0 0 0 7 0 0 2 3 Varanus 0* 0 1 1 0 0 7 0 0 24 Mosasauroidea 0* 1 ' 2 0 0 0 0' 1 * 0 0 2 5 Pachyrhachis 0* 1 2 3* 2 7 ? 0 7 7* 2 6 Scolecophidia 0 0 3 0 2 2 0 0 1 2 7 Dinilysia 0 0 2 2 2 7 2 0 0 7 2 8 Anilioidea 0 0 2 2 2 2 0 0 1/3 29 Macrostomata 0 1 2 3 2 2 0 0 3/4 RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 51 Table 1. Continued. Pachyrhachis 5 6 1 6 2 63 6 4 6 5 6 6 67 6 8 6 9 70 7 1 7 2 7 3 7 4 7 5 delete | delete delete 1 Kuehneosauridae 0 ? 0 ? ? ? ? ? ? ? 0 ? ? 0 0 2 Marmoretta ? ? ? 7 ? ? 0 ? ? 7 7 ? 7 0 0 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 Iguanidae 0 0 0 0/1 0/1 0 0 0 0 0 0 0 0 0 0/1 6 Agamidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 Chamaeleonidae 0 7 0 0 0 0 0 0 0 1 0 0 0 0 0 8 Xantusiidae 0 ? 0 1 1 0 1 0 0 0 0 0 0 0 0 9 Gekkonidae 0 0 0 1 1 0 0/1 0 0 0 0 0 0 0 0 1 0 Pygopodidae 0 0 0 1 1 0 1 0 0 0 0 0 0 0 0 1 1 Sineoamphisbaena 1 ? 7 ? ? 0 0 0 7 1 1 1 0 0 0 1 2 Dibamidae 0 ? 0 0 ?• 2* 0 0 1 1 1 1 0 0 0 1 3 Amphisbaenia 1 1 2 1 ?• 2* 0 0 1 1 1 1 0 0/1 0 1 4 Lacertidae 0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 1 5 Teiidae 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 6 Gymnophthalmidae 0 ? 0 1 1 0 0/1 0 0 0 0 0 0 0 0 1 7 Cordylidae 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 8 Scincidae 0 0 0 1 1 0 0 0 0 0 0 0/1 * 0 0 0/1 1 9 Anguidae 0 0/1 0 1 1 0/1 0 0 1 0 0 0/1 0 0 0 2 0 Xenosauridaa 0 0 0 1 1 1 0 0 1 0 0 0 0 0 0 2 1 Heloderma 0 ? 0 1 1 1 • 0 0 1 0 0 0 0 0 1 2 2 Lanthanotus 0 7 0 1 1 1 0 1 1 0 0 0 0 0 1 2 3 Varanus 0 0 0 1 1 0 0 1 1 0 0 0 0 0 1 24 Mosasauroidea 0* 1 0 1 1 0 0 1 1 0 0 0 0* 1 1 25 Pachyrhachis 2 1 1 ?* ? ? ? ? ? ?• ?• ? ?• ? 2 6 Scolecophidia 2 1 ? ? 2 0 1 1 0 1 1 1 ? 7 27 Dinilysia 2 1 ? ? 2 0 ? ? 0 1 1 1 ? ? 28 Anilioidea 2 1 ? 7 2 0 1 1 0 1 1 1 ? ? 29 Macrostomata 2 1 ? ? 2 0 1 1 0 1 0&1 1 ? ? Pachyrhachis 6 7 6 77 7 8 79 80 8 1 8 2 83 8 4 8 5 8 6 87 8 8 89 9 0 ord. ord. delete delete 1 Kuehneosauridae 0 ? ? ? ? 0 0 0 0 ? ? ? ? ? ? 2 Marmoretta 0 0 0 ? ? ? 0 ? ? ? ? ? ? ? ? 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 Ancestor 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 Iguanidae 0 0 0 0 0* 0 0 0 0/1 • 0 0 0 0 0 6 Agamidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 Chamaeleonidae 0 0 0 0 0* 0 0 0 ? ? ? ? 0 1 8 Xantusiidae 0 0/1 0 0 0* 0 0 0 0 0/1 1 9 Gekkonidae 0 0 0 0/1 * 0 0 0 0 0 0/1 1 0 Pygopodidae 0 0 0 1/2* 0/1 • 0 0 0 0 1 1 1 Sineoamphisbaena ? 0 0 1 * 0 0 0 0 0 _J 1 0 1 2 Dibamidae 2 1 0 2* 1 0 1 0 0 2 0 1 3 Amphisbaenia 2 0 0 2* 0/1 0 1 0 0 1 0 1 4 Lacertidae 0 1 0 0 0/1 * 0 0 0 0 0 0 0 1 5 Teiidae 0 0 0 0 0* 0 0 0 0 0 0 0 1 6 Gymnophthalmidae 0 0 0 0 0* 0 0 0 0 0 0/1 0 1 7 Cordylidae 0 0/1 0 0 0* 0 0 0 0 0 0 0 1 8 Scincidae 0 0/1 0 0 0/1/2 ' 0/1 * 0 0 0 0 0/1 0/1 1 9 Anguidae 0 1 0/1 0 0/1 * 0 0 0 0 0 0 2 0 Xenosauridae 0 0/1 0 0 0* 0 0 0 0 0 0 2 1 Heloderma 0 1 1 0 0 0 0 0 0 0 0 22 Lanthanotus 1 1 0 1 0* 0 0 0 0 0 0 23 Varanus 1 0 0 1 0* 0 0 0 0 0 0 24 Mosasauroidea 1 0 0* 0 0* 0 0 0 1 1 0 0 25 Pachyrhachis ? ? ? ? 2 ? 7 ? 7 ? ? 2 ? 2 6 Scolecophidia ? 1 0 0 2 0 7 0&1 2 0 27 Dinilysia ? ? 0 0 2 0 ? ? ? 7 ? 0 2 8 Anilioidea ? 1 0 0 2 0 7 1 2 0 2 9 Macrostomata ? 0 0 0 2 0 ? 1 2 0 52 FIELDIANA: GEOLOGY Table 1. Continued. Pachyrhachls 7 9 1 92 93 94 95 9 6 97 98 99 1 00 101 102 1 03 1 04 105 ord. ord. 1 Kuehneosauridae 0 0 0 0 0 ? ? 0 ? 0 ? 0 0 1 0 2 Marmoretta 0 0 ? 0 0 0 ? 0 1 7 0 0 0 1 ? 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 0 0 0/1 0 4 Ancestor 0 0 0 0 0 0 0 0 0/1 0 0 0 0 1 0 5 Iguanidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0/1 6 Agamidae 0 0 0 0 0 0 0 0 1 0 0 0 0 0/1 7 Chamaeteonidas 0 0 1 0 0 0 0 0 1 0, 1 0/1 0 0 0 8 Xantusiidae 0 0 0 0 0 0 0 0 0 0/1 0 0 9 Gekkonidae 0 0 0 0 0 0 0 0 0 0/1 0 0 1 0 Pygopodidao 0 0 0 0 0 0 1 0 0 0/1 0 0 1 1 Sineoamphisbaena 1 0 0 0 1 0 0 0 0 1 2 0 1 2 DbamkJae 1 0 0 1 1 0 0 0 0 1 1 1 1 3 Amphisbaenia 1 0 0 0 1 0 0 0 0 1 2 2 0/1 1 4 Lacertidae 0 0 0 0 0 0 0 0 0 0 0 0/1 1 5 Teiidae 0 0 0 0 0 0 0 0 0 1 0 1 1 6 Gymnophthalmidae 0 0 0 0 0 0 0 0 0 0 0 1 1 7 Cordylidae 0 0 0 0 0 0 0 0 0 0 0 0 1 8 Seine idae 0 0 0 1 0 0 0 0 0 0 0 0 1 9 Anguidae 0 0 0 0 0 0 0 0 0 0/1 0 0 20 Xenosauridae 0 0/1 0 0 0 0 0 0 0 0 0 0 0 2 1 Heloderma 0 1 0 0 0 0 1 0 0 1 0 0 22 Lanthanotus 0 0 0 0 0 0 1 0 0 1 0 0 23 Varanus 0 1 0 0 0 0 1 0 0 0 1 0 0 24 Mosasauroidea 0 1 0 0 0 0 1 0 0 ?• 0 1 25 Pachyrhachis ?• 7 ?• 0 ?• ?• ?• 0 0 0 0 26 Scolecophidia 0 0 1 0 0 1 0&1 0 ? 0 0 27 Oinilysia 0 ? 1 0 1 1 0 0 0 0 0 28 Anilioidea 0 0 1 0 0 1 0 0 0 0 0 29 Macrostomata 0 1 1 0 0 1 0 0 0 0 0 Pachy rhachla 8 1 06 107 1 08 1 09 1 10 1 1 1 1 1 2 1 1 3 1 1 4 1 1 5 1 1 6 1 17 1 18 1 1 9 1 20 ord. ord. ord. ord. ord. 1 Kuehneosauridae ? 0 0 ? 1 0 0 0 0 ? ? ? 0/1 0 0 2 Marmoretta 0 7 7 0 0 0 0 2 0 0 ? 0 ? 0 ? 3 Rhynchocephalia 0 0 0 0 0 0 0/1 0 0 0 0 1 0/1 0 2 4 Ancestor 0 0 0 0 0 0 0 0/2 0 0 0 0/1 0/1 0 0/2 5 Iguanidae 0 0 0 0/1 0 0 1 2 0/1/2 0/1 0 1/2 1 0 0/1/2 6 Agamidae 1 0 0 0 0 0 0/1 2 0 0 0 ? 1 0 1/2 7 Chamaeleonidae 1 0 0 1 0 0 0 2 0 0 0 ? 1 0 2 8 Xantusiidae 0 0 0 0 0 0 0 1 2 ? 0 0 1 0 1 9 Gekkonidae 0/1 0 0 0 0 0 0/1 2 2 ? 0 0 1 0 1 1 0 Pygopodidae 0 0 0 0 0 0 0/1 2 2 ? 0 0 0/1 0 1 1 1 Sineoamphisbaena ? ? 1 7 0 7 0 2 ? ? ? 7 0 0 ? 1 2 DbamkJae ? 0 1 0/1 0 0 0 0 2 ? 0 0 0 0 2 1 3 Amphisbaenia 0/' ? 1 0/1 0 0 on 0 0/1/2 1 0 1/2 1 0 1/2 1 4 Lacertidae 0 0 0 0 0 0 1 2 0 0/1 0 0 1 0 0 1 5 Teiidae 0/1 0 0 0 0 0 1 2 0/1 0 0 0 1 0 0 1 6 Gymnophthalmidae 1 0 0 0 0 0 0/1 2 0/2 0 0 0/1 0/1 0 0/1 1 7 Cordylidae 0/1 0 0 0 0 0 1 1 0 0 0 0 1 0 0 1 8 Scincidae 0/1 0 0 0 0 0 1 1 0/1/2 0 0 0 1 0 0/1 1 9 Anguidae 0 0 0 0 0 0 1 1/2 0 0 1 1 0 0/1 2 0 Xenosauridae 0 0 0 0 0 0 1 1 0 0 1 1 0 0 2 1 Heloderma 0 0 0 0 0 0 1 2 0 0 2 1 0* 1 2 2 Lanthanotus 0 0 0 0 0 0 1 2 0 0 2 1 0* 1 23 Varanus 1 0 0 0 0 0 1 2 0 0 2 1 0* 0 24 Mosasauroidea 0 1 0 0 0 0/1 • 2 0 1 2* 0 1 0 25 Pachyrhachis ? ? 0 7 ?• 2 0 0 ? 2 2 2 0 26 Scolecophidia 1 ? ? 0 1 2 0 0 7 2 0 2 0 27 Dinilysia 1 0 0 0 1 2 0 0 ? 2 2 2 1 28 Anilioidea 0/1 1 0 0 1 2 0 0 0 2 2 2 1 29 Macrostomata 0/1 1 0 0 1 2 0 0 ? 2 2 2 0 RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 53 Table 1. Continued. Pachyrhachis 9 1 21 1 22 1 23 1 24 1 25 1 26 1 27 1 28 1 29 1 30 1 31 1 32 1 33 1 34 1 35 ord. delete delete dolete 1 Kuehneosauridae ? ? ? ? ? ? ? ? ? 0 ? 0 ? 0 ? 2 Marmoretta ? 9 ? ? 0 0 ? 9 ? ? 0 ? ? ? 9 3 Rhynchocephalia 9 9 ? ? 0 0 9 0 0 0 0 0 0 0 0 4 Ancestor ? ? ? ? 0 0 ? 0 0 0 0 0 0 0 0 S Iguanidae 0* 0/1 0 0 0 r 0/1 0 0* 0 0 0 0/1 0/1 6 Agamidae 0* 1 0 0 0 0 0 0 0 0 0 0 1 7 Chamaeleonidae 0* ? ? ? 0 0 9 0 0 0 0 0 0 0/1 8 Xantusiidae 0* 0 0 0 0 0 0 0* 0 0/1 1 ? 0 9 Gekkonidae 0 1 0 0 0 1 0 0* 0 0 0/1 1 0 1 0 Pygopodidao 0 1 0 0 0 1 0 0* 0 0 1 ? 0 1 1 Sineoamphisbaena 9 ? 9 ? ? 1 9 0 9 0 0 ? 0 ? 1 2 Dibamidae 0* 9 ? ? 1 0 9 0/1 1 • ?' 0 1 9 0 1 3 Amphisbaenia 0* 1 0 0 0/1 0 0 0/1 1 * ?* 0/1 0/1 0 0 9 1 4 Lacertidae 0 0/1 0 0 0 1 0 0* 0 0 0 0 1 5 Teiidae 0 1 0 0 0 1 0 0* 0 0 0 0 1 6 Gymnophthalmidae 0* 1 0 0 0 1 0 0* 0 0 0 0 1 7 Cordylidae 0 0 0 0 0 1 0 o • 0 1 0 0 1 8 Scincidae 0 0/1 0 0 0 1 0 0* 0 1 0 0 1 9 Anguidae 0 0 0 0 0 1 0 0* 0 0/1 0 0 20 Xenosauridae 0 0 0 0 0 1 0 0* 0 0 0 0 2 1 Heloderma 0* 0 0 0 1 0 0* 1 0 0 0 0 22 Lanthanotus o- 0 0 1 0 0 0* 2 0 0 0 0 0 2 3 Varanus 0* 0 0 0 0 0 o- 2 0 0 0 0 0 24 Mosasauroidea 1 * 1 1 0 1 0 0 2 0 0 0 0 1 25 Pachyrhachis ?* 1 ?• ?• ?• 0 1 • 3 0 0 0 0 ?• 2 6 Scolecophidia 0 0 ? 9 0 0 1 ? 1 0 0 0 0 27 Dinilysia 2 1 9 ? 0 0 1 3 0 0 0 0 ? 28 Anilioidea 2 1 9 ? 0/1 0 1 3 0 0/1 0 0 ? 29 Macrostomata 2 1 9 ? 0/1 0 1 3 0 0 0 0 9 Pachyrhachis 10 1 36 137 1 38 1 39 1 40 1 41 142 1 43 1 44 1 45 1 46 1 47 1 48 1 49 1 50 ord. ord. ord. delete 1 Kuehneosauridae 0 ? 0 0 0 0 0 0 1 0 0 0 0 0 2 Marmoretta ? 0 9 1 12 9 ? 9 9 9 9 0 0 0 0 3 Rhynchocephalia 0 0 0 2 0 0 0 0 0 0 0/1 0 0 0 0/2 4 Ancestor 0 0 0 1/2 0 0 0 0 0/1 0 0 0 0 0 5 Iguanidae 0 0 0 0 0 0 0/1 0 0 0 0 0 0 0 6 Agamidae 0 0 0 0 0 0 0 0 0 0/1 0 0 0 2 7 Chamaeleonidae 0 0 0 0 0 0 0/1 0 0 0/1 0 0 0 2 8 Xantusiidae 1 0 0 2 0 0 0 0 0 0 0 0 0 0 9 Gekkonidae 0 0 0 0 1 1 1 0 1 1 0 0 0 0 1 0 Pygopodidae 0 0 0 0 1 1 1 0 1 0 0 0 0 0 1 1 Sineoamphisbaena 9 9 0 0/1 0 9 9 9 9 9 0 0 0 2 1 2 Dibamidae 0 0 0 0* 0 0 1 0 1 1 0 0 0 1 1 3 Amphisbaenia 0 0 0 0/1 /2 0 1 1 0 1 1 0/1 0 0 0 1 1 4 Lacertidae 1 0 1 0 0 0 0 0 0 0 0 0 0 0 1 5 Teiidae 2 0 1 0 0 0 0 0 0 0 0 0 0 0/1 1 6 Gymnophthalmidae 2 0 1 0 0 0 0 0 0 0/1 0 0 0 0/1 1 7 Cordylidae 0 0 0 0 0 1 0/1 1 1 1 0 0 0 0 1 8 Scincidae 0 0 0/1 0 0 1 1 1 1 1 0 0 0 0/1 1 9 Anguidae 0 0 0 0 0 1 1 0 1 1 0 0/1 0 1 20 Xenosauridae 0 0 0 0 0 0 1 0 0 1 0 0 0 1 2 1 Heloderma 0 0 0 0 0 1 1 0 0 1 2 * 1 1 1 2 2 Lanthanotus 0 0 0 0 0 1 1 0 0 1 2* 1 1 1 2 3 Varanus 0 0 0 0 0 1 1 0 0 1 2* 1 1 1 2 4 Mosasauroidea 0 1 0 1 0 1 1 0 1 1 2* 0* 1 1 2 5 Pachyrhachis 0 1 0 ?• 9 2 ' ? 9 9 9 3 • 0 o- ? 2 6 Scolecophidia 0 0 0 9 0 0 9 0 0 9 2 0 0 1 27 Dinilysia 0 1 0 9 9 ? ? 9 9 9 3 0 0 9 2 8 Anilioidea 0 0/1 0 9 0 2 ? 0 0 9 3 0 0 1 29 Macrostomata 0 1 0 9 0 2 ? 0 0 ? 3 0 0 1 54 FIELDIANA: GEOLOGY Table 1. Continued. Pachyrhachis 1 1 151 1 52 1 S3 1 54 155 156 157 158 1 59 160 161 162 163 1 64 1 65 ord. ord. delete ord. ord. 1 Kuehneosauridae 0 7 0 0 ? 0 0 0 0 0 0 0 ? ? ? 2 Marmoretta 0 7 0 0 ? 0 0 0 0 0 0 ? ? ? ? 3 Rhynchocephalia 0 0 0 0 ? 0 0 0 0 0 0 0 ? ? ? 4 Ancestor 0 0 0 0 ? 0 0 0 0 0 0 0 ? ? ? 5 Iguanidae 0 0 0 0 0 0 0 0/1 0 0/1 0 0 0 6 Agamidae 7 ? 0 0 0/1 0 0 ? 1 ? 0 0/1 7 Chamaeteonidae 7 ? 1 0 1 0 0 7 1 ? 0 0 8 Xantusiidae 0 0 0 0 0 0 0 ? 1 ? 0 0 9 Gekkonidae 0 0 0 0 0 0 0 ? 1 ? 1 0 1 0 Pygopodidae 0 0 0 0 0 0 0 ? 1 ? 1 0 1 1 Sineoamphisbaena 7 ? 0 1 0 1 1 ? 1 ? ? 0 1 2 Dbamidae 0 0 0 1 0 1/2 1 ? 1 ? 0 0 1 3 Amphisbaenia 0 0 0 1 1 0/1 2 2 ? 1 ? 0 0 1 4 Lacertidae 0 0 0 0 0/1 0 0 ? 0/1 0 0 0 1 5 Teiidae 0 0 0 0 0/1 0/1 0 ? 0/1 0 0 0/1 1 6 Gymnophthalmidae 0 0 0 0 0 0 0 ? 1 0 0 0 1 7 Cordylidae 0 0 0 0 0 0 0 ? 0/1 0 0 0 1 8 Scincidae 0 0 0 0 0 0 0 ? 0/1 0 0 0 1 9 Anguidae 0 0 0 0 0 0 0 0 0/1 0 0 0 2 0 Xenosauridae 0 0 0 0 0 0 0 ? 0/1 0 0 0 21 Heloderma 1 • 0 0 0 1 1 1 0/1 0 0 0 0 0 2 2 Lanthanotus 1 • 0 0 0 1 1 1 0 0 0 ? 1 2 23 Varanus 1 • 0 0 0 1 1 1 7 1 ? 0 2 2 24 Mosasauroidea 2* 0* 1 0 ? 1 0 0 7 0 1 7 0 0/1 2 5 Pachyrhachis ?' 7 1 7 ? 7 0 0 0 1 0 1 ? 0 0 26 Scolecophidia 0 1 1 0 7 7 2 2 1 7 1 ? ? 0 0 27 Dinilysia 7 7 7 7 ? 7 0 1 0 ? 0 1 ? 1 ? 28 Anilioidea 1 1 1 0 7 1 1 0/1 0 1 0 1 0 0 0/1 29 Macrostomata 1 1 1 0 7 1 0 0 0 1 0 1 0 0/1 1 Pachyrhachis 12 1 66 1 67 1 68 1 69 170 171 172 173 174 175 1 76 177 178 179 180 delete ord. ord. delete ord. ord. 1 Kuehneosauridae ? ? 0 o • ? 7 1 1 0 0 0 0 0 ? 0 2 Marmoretta 7 ? 0 7 ? 7 ? ? ? ? 7 ? ? ? ? 3 Rhynchocephalia ? 0 0 1 • 0 0 1 1 0 0 0 0 0 ? 0 4 Ancestor ? 0 0 0/1 0 0 1 1 0 0 0 0 0 ? 0 5 Iguanidae 0 0 0/1 1 0/1 1 0 0 0/1 0 0/1 1 0 6 Agamidae 0 0 0* 7 0/1 1 0 0 0 1/2 0 ? 0 7 Chamaeteonidae 0 0 0* 7 0 0 0 0 0 0 0 ? 0 8 Xantusiidae 0 0 0* ? 0/1 2/3 1 0 0 0 0/1 0/1 1 0 9 Gekkonidae 1 0 0 0* ? 1/2/3 1 0 0 0 0 0 ? 0 1 0 Pygopodidae 1 0 0 0* 7 4 0 0 0 0 0 0 ? 0 1 1 Sineoamphisbaena 0 0 0* ? f 1 1 1 0 0 7 ? ? 1 2 Dbamidae 0 1 0* ? 4/5 0 1 1 2 0 0 ? 1 1 3 Amphisbaenia 0 1 0* 7 4/5 0 0/1 1 2 0 0 ? 1 1 4 Lacertidae 0 0 1 • 0 1/2/3 1 0 0 0/1 0/1 0/1 1 0 1 5 Teiidae 0 0 2* 1 1/2/3 1 0 0 0 1 1 1 0 1 6 Gymnophthalmidae 0 0 0 1 1/2/3 1 0 0 0 1/2 1 1 0 1 7 Cordylidae 0 0/1 1 • ? 2/3 0/1 0 0 0 0 0 ? 0 1 8 Scincidae 0 0 0" ? 3 1 0 0 1/2 0 0/1 0 0 1 9 Anguidae 0 0 0* ? 3/4 0/1 0/1 0 2 0 0/1 0 0 20 Xenosauridae 0 0 0" ? 2/3 1 0 0 2 0 0 0 2 1 Heloderma 0 0 0' 7 3 1 0 0 1/2 0 0 0 22 Lanthanotus 0 0 0* ? 3 2 1 0 1/2 0 0 0 23 Varanus 0 0 0* 7 3 2 1 0 2 0 0 0 24 Mosasauroidea 1 0 2* 1 3 7 1 1 0 0 0 0 25 Pachyrhachis ? 7 0 2' ? 5 ? 1 0 2 0 0 0 26 Scolecophidia 0 0 1 2 7 5 ? ? 1 2 0 0 0 27 Dinilysia ? 7 1 2 ? 5 ? ? 1 2 0 ? ? 28 Anilioidea 0/1 0 1 2 ? 5 ? ? 1 2 0 0 0 29 Macrostomata 0/1 0 1 2 ? 1 5 ? ? 1 2 0 0 0 RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 55 Table 1. Continued. Pachyrhachls 13 181 1 82 1 83 1 84 1 85 1 86 187 1 88 1 89 190 1 91 1 92 1 93 1 94 1 95 ord. 1 Kuehneosauridae 1 7 7 9 ? ? 0 0 0 0 0 0 0 7 7 2 Marmoretta 7 7 7 7 7 7 7 7 7 ? 7 7 7 7 7 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0/1 0 0 0 0 4 Ancestor 0/1 0 0 0 0 0 0 0 0 0 0/1 0 0 0 0 5 Iguanidae 0/1 0/1 0/1 0 0 1 0 0 0 0 0/1 0/1 0 0 6 Agamidae 1 ? 0 0 0 1 0 0 0 0 0 0/1 0 0 7 Chamaeleonidae 1 7 0 0 0 1 0 0 0 0 1 0 1 7 8 Xantusiidae 0 0 0 0 0 1 0 0 0 0 0 0 0 1 9 Gekkonidae 0 1 0 0 0 1 0 0 0 0 1 0/1 0 1 1 0 Pygopodidae 0 1 0 0 0 0 1 0 0 1 0 0 0 1 1 1 Sineoamphisbaena 7 7 7 7 7 1 0 0 7 0 7 7 7 7 1 2 Dibamidae 0 0 0 0 0 0 1 1 1 1 7 7 1 7 1 3 Amphisbaenia 0/1 0 7 1 0 0 1 1 1 0/1/2 0 0 0/1 1 1 4 Lacertidae 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1 S Teiidae 0 0 0 0 0 1 0 0 0 0 0/1 1 0 1 1 6 Gymnophthalmidae 0 0 0 0 0 1 0 0 0 0 0 1 0 1 1 7 Cordylidae 0 0 0 0 0 1 0 0 0 0 0 0 0 0/1 1 8 Scincidae 0/1 0 0 0 0 1 0 0 0 0 0/1 0 0 1 1 9 Anguidae 0/1 0 0 0 0 0/1 0 0 0 0/1 0/1 0 0 1 20 Xenosauridae 0/1 0 0 0 0 1 0 0 0 0 0 0 0 1 2 1 Heloderma 1 7 0 0 1 1 0 0 0 0 0 0 0 1 2 2 Lanthanotus 1 7 1 0 1 1 0 0 0 0 0 0 0 1 2 3 Varanus 1 7 1 0 1 1 0 0 0 0 0 1 0 0/1 2 4 Mosasauroidea 1 ? 1 0 1 0 0 0 0 0 0 0/1 0 0 7 2 5 Pachyrhachis 1 7 7 7 7 7 0 0 1 2 7 7 7 2 6 Scolecophidia 1 7 7 7 7 7 0 0 1 2 7 7 7 27 Dinilysia 7 7 7 7 7 7 0 0 7 7 7 7 7 7 2 8 Anilioidea 1 7 7 7 7 7 0 0 1 2 7 7 7 29 Macrostomata 1 7 7 7 7 7 0 0 1 2 7 7 7 Pachyrhachis 14 196 197 1 98 1 99 200 201 202 203 204 205 206 207 208 209 21 0 ord ord. ord. delete 1 Kuehneosauridae 7 7 7 7 7 7 7 7 0 0 0 1 0 0 0 2 Marmoretta ? ? 7 7 7 7 ? 7 0 7 7 7 7 7 ? 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 4 Ancestor 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 5 Iguanidae 0/1 0 0 0 0 0/1 1/2 0/1 0 0 0 1 0 0 0 6 Agamidae 0 0 0 0/1 0 1 0/1 0 0 0 0 1 0 0 0 7 Chamaeleonidae 7 1 7 7 0 0/1 3 1 0 1 0 1 1 0 0 8 Xantusiidae 1 0 0 1 0 0/1 1 0 0 0 0 0 1 2 9 Gekkonidae 1 0 0/1 1 0 0/1 1/2 0 0 0 0 1 1 1 1 0 Pygopodidae 0 1 7 7 0 0 3 0 1 7 1 0 1 7 1 1 Sineoamphisbaena 7 0 7 7 7 7 7 7 0 0 7 7 7 7 1 2 Dibamidae 7 1 7 7 0 0 3 1 1 7 1 0/1 0 7 1 3 Amphisbaenia 7 1 7 7 0/1 0 3 0 0/1 1 1 0 1 7 1 4 Lacertidae 0 0 1 0 1 1 0/1 0 0 0 1 0 2 1 5 Teiidae 0 0 1 0 1 1 0 0 1 0 0 0 1 1 6 Gymnophthalmidae 0 0 1 0 1 1 0 0 1 0 0 1 2 1 7 Cordylidae 0/1 0 0 1 0 0 1 12 0/1 0 0 0 0 0 2 1 8 Scincidae 0/1 0 1 0 0/1 1 0/1 0 0 0 0/1 1 2 1 9 Anguidae 0/1 0 1 0 0 1/2 0 0/1 0 0/1 0 0/1 2 20 Xenosauridae 0 0 0 0/1 0 0 1 0 0 0 0 0 0/1 1 2 1 Heloderma 0 0 1 7 0 0 1 0 0 0 0 1 1 1 2 2 Lanthanotus 0 0 0 1 0 0 3 0 0 0 0 1 1 1 23 Varanus 0 0 0 0/1 0 0/1 2 0 0 0 0 1 0 0 24 Mosasauroidea 0 0 1 7 0 0 0 0 0 0 1 * 1 0/1 25 Pachyrhachis 7 1 7 7 1 7 7 0 1 7 1 2 1 1 2 6 Scolecophidia 7 1 7 7 1 7 7 0 1 7 1 2 1 7 7 2 7 Dinilysia 7 7 7 7 7 7 7 0 7 7 7 7 7 ? 7 2 8 Anilioidea 7 1 7 7 1 7 7 0 1 7 1 2 1 7 7 2 9 Macrostomata 7 1 ? 7 1 7 7 0 1 7 1/2 2 1 7 7 56 FIELDIANA: GEOLOGY Table 1. Continued. Pachyrhachls 15 21 1 21 2 213 214 21 5 21 6 21 7 21 8 219 220 221 222 223 224 225 delete delete ord. ord. ord. 1 Kuehneosauridae 0 0 0 0 7 0 0 0 ? 7 0 ? ? 7 ' 2 Marmoretta ? ? ? 7 ? ? ? ? ? ? 0 ? ? ? ? 3 Rhynchocephalia 0 0 0 0 0 0 0 0 ? ? 0 0 0 0 0 4 Ancestor 0 0 0 0 0 0 0 0 ? ? 0 0 0 ? 0 5 Iguanidae 0 0 0 0 0 0 0 0 ? ? 0/1 0 1/2 0 0/1 6 Agamidae 0 0 0 0 0 0 0 0 ? ? 0/1 0 2 0 0/1 7 Chamaeleonidae 0 0 1 0 0 0 0 0 ? ? 0/1 0 2 0 1 8 Xantusndae 0 0 0 1 0 0 0 0 ? ? 0/2 0 1 0 0 9 Gekkonidae 0 0 0 1 0 0/1 0 0 ? ? 0 0 0 1 0 1 0 Pygopodidae 1 ? ? ? ? 0 0 0 ? ? 0 0 0 1 0 1 1 Sineoamphisbaena ? ? ? ? ? 0 0 0 ? ? 2 ? ? ? ? 1 2 Dibamidae 1 ? 7 ? ? 0 0 0 ? ? 0 1 ? ? 1 1 3 Amphisbaenia 1 7 7 7 ? 0 0 0 ? ? 0/2 0/1 2 ? 1 1 4 Lacertidae 0 0 0 1 0 0 0 1 0 0 2 0 1/2 0 0 1 5 Teiidae 0 0 0 1 0 0 0 0 ? ? 0/2 0 1 0 0 1 6 Gymnophthalmidae 0 0 0 1 0 0 0 0 ? ? 0/2 0 1/2 0 1 1 7 Cordylidae 0 0 0 1 0 1 2 0 0 2 0 2 0 0 1 8 Scincidae 0 0 0 1 0 1 2 0 0 0/2 0 1/2 0 0/1 1 9 Anguidae 0/1 0 0 1 0 1 2 0 0 0/1 0 1/2 0 0/1 20 Xenosauridae 0 0 0 1 0 0 2 1 0 2 0 1 0 1 2 1 Heloderma 0 0 0 1 0 0 2 1 0 1 0 2 0 0 2 2 Lanthanotus 0 0 0 1 0 0 2 1 1 0 0 2 0 1 23 Varanus 0 0 0 1 0 0/1 0 0/2 1 1 0/1 0 0 0 1 24 Mosasauroidea 0 1 1 0 1 0 0 0 ? ? 0 0 1 0 ? 25 Pachyrhachis 1 1 1 0 1 0 0 0 ? ? 0 ? ? ? 26 Scolecophidia 1 7 7 7 7 0 0 0 ? ? 0 ? ? 1 27 Dinilysia 7 7 ? ? ? 0 0 0 7 7 0 7 ? ? 28 Anilioidea 1 7 7 7 7 0 0 0 ? 7 0 ? ? 1 29 Macrostomata 1 7 7 7 7 0 0 0 ? ? 0 ? ? 1 Pachyrhachls 16 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 delete ord. ord.* 1 Kuehneosauridae 7 1 1 ? 7 ? 7 0 0 0 0 0 0 0 0 2 Marmoretta 7 7 7 7 7 7 ? 0 0 0 0 0 0 0 0 3 Rhynchocephalia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 Ancestor 0 0/1 0/1 0 0 0 0 0 0 0 0 0 0 0 0 5 Iguanidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 Agamidae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 Chamaeleonidae 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 Xantusiidae 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 9 Gekkonidae 0 0 0 1 1 0 1 0 0 0 0 0 0 0 0 1 0 Pygopodidae 0 0 0 ? 1 0 1 0 0 1 0 0 0 0 0 1 1 Sineoamphisbaena ? 0 0 0 7 0 ? 0 0 ? ? 0 0 0 0 1 2 Dibamidae 1 0 0 7 0 0 1 0 1 1 0 0 0 0 1 1 3 Amphisbaenia 0/1 0 0 1 0 0 ? 0 0 2 1 0 0 0 1 1 4 Lacertidae 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 5 Tendae 0/1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 6 Gymnophthalmidae 0/1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 7 Cordylidae 0/1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 8 Scincidae 0/1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 9 Anguidae 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2 0 Xenosauridae 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2 1 Heloderma 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2 2 Lanthanotus 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 23 Varanus 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 24 Mosasauroidea ? 0/1 • 0/1 ? 0 0 0 0 0 0 0 0 0 0 0 25 Pachyrhachis 7 1 1 ? 0 ? ? 1 0 1 0 1 1 1 26 Scolecophidia 1 1 7 7 0 0 ? 0 1 2 1 0 0 0 27 Dinilysia ? 1 ? ? ? 0 ? 0 ? ? ? 0 0 0 28 Anilioidea 1 1 ? 7 0 1 1 0 1 2 1 1 0/1 0 1 29 Macrostomata 1 1 7 7 0 1 1 1 1 2 1 1 1 1 RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 57 many of the characters are problematic, yet some were alternatively retained or deleted in the cla- distic analysis as coded by Lee (1998) in order to remain as close to the original character evidence as possible. A large number of character states are given as the relative size of continuous variables (e.g., 7, 19, 23, 28, 53, 66, etc.) without reference to some standard. Although this may cause prob- lems for other workers attempting to critically evaluate character-state assignments, we retained these character codings unaltered. The same is true for distinctions of character states that appear to be purely arbitrary, such as "three or more" versus "two or less" in character 111 (see also characters 153, 157, 158, 172, 173, 174, 223). Furthermore, the terminal taxa are coded at the family level, which most probably will result in a substantial degree of polymorphism as greater scrutiny is applied to character coding. As pointed out by Etheridge (personal communication), Ig- uanidae are coded with a single character state for characters 8, 15, 34, 36, 39, 45, 50, 96, 103, 132, 139, 144, 168, 179, and 180, all of which exhibit two or more states in that taxon. It remains to be seen to what degree polymorphism applies to oth- er terminal taxa. Anniella, for example, would greatly affect the coding for Anguidae if it were to be included in that family, and might affect the result of the analysis if it were included as ter- minal taxon. If a terminal taxon is polymorphic for a given character, then the plesiomorphic char- acter state, if known, should be used for that tax- on. Yet the plesiomorphic character state can only be determined by phylogenetic analysis in intra- familial relationships, and these are often not available. Pleurodont iguanians are a case in point (Frost & Etheridge, 1989), with potentially seri- ous consequences for coding of the Iguanidae. A global assessment of squamate relationships will necessitate breaking up terminal taxa into groups below the family level or rejecting previous ana- lyses of intrafamilial relationships in order to de- termine the plesiomorphic character state at the family level. Finally, we also note that work in progress is likely to change several character codings for am- phisbaenians as well as for Sineoamphisbaena (M. Kearney, personal communication) without threatening to change the major conclusions reached in this paper, however. At this point, our interest is not in the recovery of squamate history (which is beyond the scope of this paper). Instead, our goals are twofold. The first is to test the re- lationships of Pachyrhachis on the basis of the evidence proposed by Lee (1998), for which rea- son we propose to split up Serpentes as a terminal taxon. The second goal is to test Lee's (1998) con- clusion that the burrowing ecomorph evolved con- vergently in snakes and in dibamids plus amphis- baenians, which is why we critically reexamined the characters relevant to the groups involved, i.e., snakes, varanoids, and the amphisbaenian-dibam- id clade. The cladistic analysis of the modified data set was performed using the software package PAUP, version 3.1.1, developed by David L. Swofford (Swofford, 1990; Swofford & Begle, 1993). The search settings invariably employed the heuristic search strategy with random stepwise addition (10 replications), and branch swapping (on minimal trees only) was effected by tree bisection and re- connection. A number of alternative analyses were performed that varied with respect to three parameters: rooting the analysis on the three out- group taxa (Kuehneosauridae, Marmoretta, Rhyn- chocephalia) or rooting it on the ancestor recon- structed by Lee (1998); ordering the multistate characters as indicated by Lee (1998; i.e., the characters 16, 23, 29, 34, 35, 50, 51, 77, 80, 103, 104, 113, 114, 117, 118, 120, 121, 136, 139, 146, 157, 158, 164, 165, 172, 173, 176, 177, 190, 202, 206, 207, 218, 221, 223) as well as ordering one of the newly added multistate characters (235) or leaving all multistate characters unordered; and retaining all characters or deleting the ones so designated (i.e., characters 11, 48, 57, 62, 64, 68, 84, 86, 122-124, 147, 162, 170, 174, 209, 213, 214, 228). Characters rendered uninformative by the choice of different outgroups were always ig- nored if not deleted. Bootstrap ( 1 ,000 replications, using identical heuristic search settings) and de- cay analyses were run for those analyses that most closely approach the search procedure employed by Lee (1998), i.e., with multistate characters or- dered, and rooting on the ancestor (our runs 2 and 4 below). All analyses were run in two alterna- tives. Assuming presence of a jugal in Dinilysia required retention of character 1 2 but deletion of character 240. Conversely, assuming a jugal to be absent in Dinilysia required character 12 to be de- leted and character 240 to be retained. In those analyses that retained character 240, this character was treated as ordered and unordered, respective- ly. The values of tree statistics in the discussion below that are not placed in brackets are those obtained by retention of character 1 2, deletion of character 240 (Jugal present in Dinilysia); con- versely, the values placed in brackets are those 58 FIELDIANA: GEOLOGY obtained by retention of character 240, deletion of character 12 (jugal absent in Dinilysia). Tree to- pologies were identical under both assumptions, but the assumption that a jugal is absent in Dini- lysia proved slightly more parsimonious. A first series of tests retained ordered multistate characters as indicated above. Retaining all the characters designated for deletion, deleting the an- cestor, and rooting the tree on Kuehneosauridae, Marmoretta, and Rhynchocephalia (run 1 ) yielded two equally parsimonious trees with a tree length (TL) of 645 [644] steps, a consistency index (CI) of 0.462 [0.463], and a retention index (RI) of 0.690 [0.691]. Lack of resolution was restricted to the outside of that part of the cladogram that em- braces the anguimorphs, the amphisbaenian-di- bamid clade, and snakes. The relative relation- ships of the latter taxa were fully resolved and read as follows: (Anguidae (Xenosauridae (Helo- derma ((Lanthanotus, Varanus) (Mosasauroidea (Sineoamphisbaena ((Amphisbaenia, Dibamus) (Scolecophidia {Dinilysia (Anilioidea (Pachy- rhachis, Macrostomata))))))))))). Anguimorpha is paraphyletic in this search, mosasauroids being the sister group to a clade including amphisba- enians, Dibamidae, and snakes. Amphisbaenians plus Dibamidae form the sister group of snakes, whereas within snakes, Dinilysia is the sister tax- on of Alethinophidia and Pachyrhachis is the sis- ter taxon of Macrostomata. Retaining the ordered multistate characters, re- taining all characters designated for deletion but deleting the outgroup taxa Kuehneosauridae, Mar- moretta, and Rhynchocephalia and rooting the analysis on the ancestor reconstructed by Lee (1998) yielded (run 2) two equally parsimonious trees, again with TL = 622 [621], CI = 0.477 [0.478], and RI = 0.690 [0.692]. The tree topol- ogy is the same. Retaining the ordered multistate characters but deleting all characters so designated and deleting the ancestor but rooting the analysis on the three outgroup taxa Kuehneosauridae, Marmoretta, and Rhynchocephalia (run 3) yielded four equally par- simonious trees, with TL = 606 [605], CI = 0.460 [0.461], and RI = 0.683 [0.684]. Resolution of the tree is greatly reduced, but varanoids, on the one hand, and amphisbaenians-dibamids-snakes on the other form monophyletic clades, respec- tively. The relationships among varanoids are (Heloderma (Mosasauroidea (Lanthanotus, Var- anus))); those of snakes are (Sineoamphisbaena ((Amphisbaenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrosto- mata)))))). Mosasauroids turn out to be the sister group of the Lanthanotus-Varanus clade, whereas amphisbaenians plus dibamids again form the sis- ter group of snakes, Dinilysia again is the sister taxon of Alethinophidia, and Pachyrhachis again is the sister taxon of Macrostomata. Retaining the ordered multistate characters but deleting all characters so designated and deleting the three outgroup taxa Kuehneosauridae. Mar- moretta, and Rhynchocephalia but rooting the analysis on the ancestor (run 4) yielded again four equally parsimonious trees, with TL = 584 [583], CI = 0.476 [0.477], and RI = 0.683 [0.685). The topology of the strict consensus tree is identical to that of the previous run. A second, parallel set of tests was run but with all multistate characters unordered. Retaining all characters designated for deletion, deleting the an- cestor, and rooting the analysis on the three out- group taxa Kuehneosauridae, Marmoretta, and Rhynchocephalia (run 5) yielded a total of four equally parsimonious trees, with TL = 607 [606], CI = 0.486 [0.487], and RI = 0.690 [0.691]. The strict consensus tree differs somewhat for the re- lationships outside the group that comprises an- guimorphs, amphisbaenians, Dibamidae, and snakes, yet the relationships among the latter taxa resemble the first run and are (Anguidae (Xeno- sauridae (Heloderma ((Lanthanotus, Varanus) (Mosasauroidea (Sineoamphisbaena ((Amphisba- enia, Dibamidae) (Scolecophidia (Dinilysia (Ani- lioidea (Pachyrhachis, Macrostomata))))))))))). Unordering the multistate characters therefore did not alter relationships among these latter taxa but resulted in a decrease in the tree length and a slight increase in the consistency index and reten- tion index. The same search but rooted on the ancestor (the three outgroup taxa Kuehneosauridae, Marmoret- ta, and Rhynchocephalia deleted), all multistate characters unordered, and all characters retained (run 6) resulted in two equally parsimonious trees, with TL = 587 [586], CI = 0.501 [0.502], and RI = 0.690 [0.691]. The relationships of the taxa un- der consideration remain the same, however, as those recovered in the previous search. Finally, with deletion of all characters so des- ignated, with all multistate characters unordered, and rooting the analysis on the three outgroup taxa Kuehneosauridae, Marmoretta, and Rhyn- chocephalia (deleting the ancestor; run 7). the analysis yielded a total of four equally parsimo- nious trees, with TL = 567 [566], CI = 0.487 [0.488], and RI = 0.684 [0.685]. The relationships RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 59 for the taxa under consideration have slightly changed to a monophyletic Varanoidea that in- cludes the mosasauroids, i.e., (Anguidae (Xeno- sauridae {(Heloderma (Mosasauroidea (Lanthan- otus, Varanus))) (Sineoamphisbaena ((Amphis- baenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrostomata))))))))). Deleting all characters so designated, retaining all multistate characters as unordered, deleting the three outgroup taxa Kuehneosauridae, Marmoret- ta, and Rhynchocephalia, and rooting the analysis on the ancestor (run 8) yielded two equally par- simonious trees again, with TL = 548 [547], CI = 0.502 [0.503], and RI = 0.683 [0.684]. The relationships for the relevant taxa remained the same as in the previous search. In a final note, we would like to point out that the stability of the (Sineoamphisbaena ((Amphis- baenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrostomata)))))) clade is not simply due to the addition of the char- acters 231 through 239 (240)! Running analyses parallel to the ones above with all the characters from 231 through 239 (240) excluded (multistate characters ordered and unordered; characters 1 through 230 all retained, or those so designated deleted; and rooting on the three outgroup taxa Kuehneosauridae, Marmoretta, and Rhynchoce- phalia or on the ancestor) all reproduced that clade, with one difference: Dinilysia invariably falls into an unresolved trichotomy with anilioids and the Pac/ryr/zac/n's-Macrostomata clade. The strict consensus tree thus reads (Sineoamphisba- ena ((Amphisbaenia, Dibamidae) (Scolecophidia (Dinilysia, Anilioidea (Pachyrhachis, Macrosto- mata))))). And in all these trees, the taxa Helo- derma, Lanthanotus, Varanus, and mosasauroids are related to that clade, either as a monophyletic unit or as a paraphyletic assemblage with mosa- sauroids closest to that clade. In essence, there- fore, the addition of new characters (23 1 through 240) did not influence the basic results other than tree statistics and some increase in resolution among snakes. Discussion of diagnostic characters is primarily based on DELTRAN character optimization, as it minimizes secondary loss of characters diagnostic at a higher level of inclusiveness. The synapo- morphy listings for the (Sineoamphisbaena ((Am- phisbaenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrostomata)))))) are remarkably consistent throughout the eight an- alyses performed and discussed above (runs 1 through 8), with one exception. Runs 3 and 4 re- tained the ordered multistate characters but delet- ed the characters so designated. This resulted in a loss of resolution, the (Sineoamphisbaena ((Am- phisbaenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrostomata)))))) clade falling into a polytomy with other squa- mates. That result lengthened the list of characters diagnostic for the root of this clade by three char- acters. The other nodes remained closely similar again to all other analyses performed. In the fol- lowing, we list all the synapomorphies for the successive nodes in that clade, along with the character state (in parentheses). These lists cor- respond to run 1; differences in other runs are listed consecutively. Unequivocal synapomor- phies (as determined in the first of the eight ana- lyses) optimizing the same way in the ACCTRAN and DELTRAN mode (i.e., with ci = 1) are char- acterized by an asterisk (*). Bootstrap support and decay indices relate to the node characterized by the outermost brackets of the groupings. (Sineoamphisbaena ((Amphisbaenia, Dibami- dae) (Scolecophidia (Dinilysia (Anilioidea (Pa- chyrhachis, Macrostomata)))))): 19 (0), 25 (0), 33 (1), 51 (2), 54 (1), 56 (1), 58 (0), 60* (1), 71* (1), 72* (1), 75 (0), 80 (1), 98 (0), 149 (0), 155* (1). Run 2 is identical; runs 3 and 4 add 76 (1), 157 (1), 158 (1), 174 (1), 175 (1), and 218 (0) to that list but delete 75, 98, and 149; runs 5 and 6 add 161 and delete 80; runs 7 and 8 add 76 (1), 118 (0), 175 (1), 216 (0), and 218 (0) and delete 75, 80, 98, and 149. Run 2: bootstrap support, 69%; decay index = 5. Run 4: bootstrap support, 80%; decay index = 5. ((Amphisbaenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrosto- mata))))): 8 (1), 12 (1), 23 (2), 29 (2), 35 (2), 37 (1), 38 (1), 40* (1), 46* (0), 49 (3), 66 (2), 80 (2), 81* (1), 84* (1), 116 (0), 129* (1), 168 (1), 172 (4), 189* (1), 190 (1), 194 (1), 197 (1), 204 (1), 206 (1), 211 (1), 222* (1), 232 (1), 235 (1). Run 2 is identical; runs 3 and 4 add 69 (1), 77 (1), 89 (2), 133 (0), 150 (1), 176 (2), 221 (0), 225 (1), and 226 (1) to that list and delete 116; runs 5 through 8 add 157 (2) and delete 190. Run 2: bootstrap support, 77%; decay index = 3. Run 4: bootstrap support, 85%; decay index = 7. (Scolecophidia (Dinilysia (Anilioidea (Pachy- rhachis, Macrostomata)))): 3* (1), 16 (1), 45 (0), 54 (2), 59* (1), 61 (2), 63* (1), 68 (1), 73* (1), 89 (2), 94* (1), 97* (1), 100* (1), 106 (1), 110 (1), 111* (1), 119* (1), 152* (1), 153(1), 154(0), 167 (0), 169 (2), 172 (5), 190 (2), 200* (1), 207 (2), 227 (1), 234 (1), 235 (2); 236 (1). Run 2 is 60 FIELDIANA: GEOLOGY identical; runs 3 and 4 add 117 (2), 128 (0), 146 (2), and 181 (1) to that list but delete 89; runs 5 and 6 delete 172 and 235; run 7 adds 128 (0) and deletes 172 and 235; run 8 adds 128 (0) and 181 (1) and deletes 172 and 235. Run 2: bootstrap support, 100%; decay index = 17. Run 4: boot- strap support, 100%; decay index = 21. (Dinilysia (Anilioidea (Pachyrhachis, Macro- stomata))): 49 (2), 50 (3), 102 (0), 118 (2), 121 (2), 124 (1), 130* (3), 137 (1), 146 (3), 159 (0), 161 (0). Run 2 is identical; runs 3 and 4 delete character 102 from that list; runs 5 through 8 add character 157 (0). Run 2: bootstrap support, 94%; decay index = 10. Run 4: bootstrap support, 98%; decay index = 8. (Anilioidea (Pachyrhachis, Macrostomata)): 87 (1), 107 (1), 141 (2), 151 (1), 160* (1), 231* (1), 237* (1). Runs 2, 5 and 6 are identical; runs 3 and 4, and 7 and 8, add character 156 (1) to that list. Run 2: bootstrap support, 72%; decay index = 2. Run 4: bootstrap support, 76%; decay index = 2. (Pachyrhachis, Macrostomata): 47* (1), 50 (3), 157 (0), 158 (0), 233* (2), 238* (1), 239* (1). Run 2 is identical; runs 3 and 4 add character 120 (0) to that list; runs 5 through 8 delete character 157. Run 2: bootstrap support, 97%; decay index = 4. Run 4: bootstrap support, 98%; decay index = 4. The most parsimonious result obtained in this analysis is run 8, which is based on the assump- tion that Dinilysia lacks a jugal (character 12 re- tained, character 140 excluded) and had all mul- tistate characters unordered, all characters so des- ignated deleted, and was rooted on the ancestor. The result is ((Heloderma (Mosasauroidea (Lan- thanotus (Varanus))) (Sineoamphisbaena ((Am- phisbaenia, Dibamidae) (Scolecophidia (Dinilysia (Anilioidea (Pachyrhachis, Macrostomata))))))) (Fig. 17). Note, however, that the bootstrap sup- port for the node linking mosasauroids or vara- noids (including mosasauroids) to the amphisba- enian-dibamid-snake clade was consistently less than 50%; the decay index for that node is 1. The Phylogenetic Relationships of Pachyrhachis, Dinilysia, and Dibamus Our analysis of snake interrelationships recog- nizes five terminal taxa, viz. Scolecophidia, Din- ilysia, Anilioidea, Pachyrhachis, and Macrosto- mata. The monophyly of macrostomatans is well Fig. 17. Cladogram of the interrelationships of snakes ohtained hy rcanalysis of the data of Lee (1998). For further discussion, sec text. corroborated (Rieppel, 1988), but the same cannot be said for scolecophidians or anilioids (Cundall et al., 1993). The test of the monophyly of Sco- lecophidia is beyond the scope of this paper and it is tentatively accepted here, but the monophyly of the Anilioidea has been corroborated in an in- dependent study (Zaher & Rieppel, unpublished data), which is why we retain this taxon. One character that supports the monophyly of the An- ilioidea in this latter work is the configuration of the perilymphatic foramen (Rieppel, 1979b). Our analysis indicates that, among the terminal taxa used in this context. Pachyrhachis is the sis- ter taxon of Macrostomata (Zaher, 1998). This re- sult is very robust, as it was obtained in all ana- lyses performed, with or without inclusion of the newly added characters 231 through 239 (240). Indeed, four unequivocal synapomorphies diag- nose the clade including Pachyrhachis and Ma- crostomata, viz. supratemporal at least half of maximum skull width (47, 1 ), posterior dentiger- ous process of the dentary enlarged (233, 2), tooth-bearing anterior process of the palatine pre- sent (238, 1). and suprastapedial process of the quadrate absent (239, 1). This corroborates Za- her's (1998) earlier findings and removes Pachy- rhachis from the position of a link between mo- sasauroid squamates and snakes (contra Carroll, 1988; Caldwell & Lee, 1997; Lee & Caldwell, 1998). The significance of the presence of poste- rior limbs in Pachyrhachis remains elusive at the present time. Either the limb was redeveloped from a rudimentary stage, as is still retained in basal alethinophidians, or relatively complete hind limbs were retained in a variety of fossil alcthin- ophidian snakes that remain unknown (Zaher, 1998). Arguments for the position of Pachyrhach- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 61 is that rely on the presence of hind limbs are ad hoc and ignore all the other data. The relationship of Dinilysia as the sister taxon of alethinophidian snakes (Rieppel, 1988) is also well corroborated. Only by deletion of the newly added characters 231-239 (240) does Dinilysia fall into an unresolved trichotomy with the Ani- lioidea and the Pac/ryr/iac/i/s-Macrostomata clade. Although there is a substantial list of char- acters supporting this hypothesis of relationships of Dinilysia, only one character is an unequivocal synapomorphy: surangular extends far into lateral surface of dentary and terminates in a point ( 1 30, 3). In other words, Dinilysia shares the alethino- phidian intramandibular joint. Another sister-group relationship that we find well supported is the clade that comprises dibamids and amphisbaenians. The bootstrap support for the node linking dibamids to amphisbaenians is 95% (runs 2 and 4), and its decay index is 6 (runs 2 and 4). These two taxa are linked by an impressive list of characters (DELTRAN character optimization; run 1): 1 (1), 15 (1), 27 (1), 32 (1), 70 (1), 77 (2), 82 (0), 83 (1), 91 (1), 96 (1), 103 (1), 104 (1), 108 (1), 113 (0), 120 (1), 144 (1), 146 (1), 147 (0), 173 (0), 180* (1), 185 (0), 186 (0), 187 (1), 188* (1), 202 (3). Run 2 is identical; runs 3 and 4 add char- acters 102 (1) and 126 (0) to that list and delete characters 120, 146, and 185; runs 5 and 6 delete characters 103, 104, and 202; runs 7 and 8 add 126 (0) and delete 103, 104, 146, and 185. The two unequivocal synapomorphies linking these two taxa are caudal transverse processes project anterolater- ally ( 1 80, 1 ), and proximal end of rib with poster- oventral pseudotuberculum (188, 1). This clade was also obtained by Lee (1998) and other authors (Caldwell, 1999; Evans & Barbadillo, 1998; Haller- mann, 1998). We feel less confident, however, ac- cepting other aspects of the cladogram(s) obtained in the present analyses, for reasons discussed below. Discussion: Snake Origins, and Homology Versus Convergence Based on his analysis, Lee (1998) concluded that the elongate fossorial ecomorph evolved in- dependently in nonophidian squamates (e.g., di- bamids and amphisbaenians) and in snakes. Al- though he recognized a suite of derived characters shared by snakes, dibamids, and amphisbaenians, he attributed these to miniaturization of fossorial forms (Lee, 1998, p. 415). Characters that diag- nose the fossorial ecomorph were judged to be correlated, and a case was made that such char- acters need to be downweighted in order to avoid a "cascade of effects that lead to apparently strong support for a (probably spurious) phylo- genetic hypothesis" (Lee, 1998, p. 417). This pro- cedure was justified by reference to the claim that independent a priori evaluation of the potential phylogenetic information content of characters (on the basis of functional anatomy, for example) may be necessary to avoid mistaken conclusions (Lee & Doughty, 1997). That way, a phylogenetic hypothesis is reconstructed that is believed to be better in line with an evolutionary scenario sup- ported, for example, by functional anatomical ex- planations (of the burrowing ecomorph in this ex- ample). In our view, and contrary to Lee (1998; see also Lee & Doughty, 1997), this procedure is circular because empirically empty a priori assumptions about an evolutionary process are allowed to in- fluence the phylogenetic analysis, when informa- tion about evolutionary processes, including func- tional anatomical explanations, should flow from the reconstructed phylogeny (e.g., Lauder & Liem, 1989). However, we agree with Lee (1998) that there is a serious potential for convergence in the evolution of the fossorial ecomorph owing to structural constraints that correlate with miniatur- ization and that affect not only dibamids, amphis- baenians, and snakes but also members of other "lizard" families as well, such as Anniella (Riep- pel, 1984b). However, it remains unclear from Lee's (1998) arguments why a clade grouping di- bamids with amphisbaenians should be retained and even named, although it is supported by char- acters of the fossorial ecomorph, while the same characters are claimed to support "spurious" phy- logenetic relationships of snakes and hence have to be downweighted if snakes are included in the analysis. We acknowledge, however, that the di- bamid-amphisbaenian clade survives even more severe downweighting of these characters than is necessary to break the snakes away from that clade. But just as snakes might group with amphis- baenians and dibamids on the basis of convergent characters correlated with fossorial habits, char- acters that are correlated with the evolution of the intramandibular joint might lend unjustified sup- port to the snake-mosasaur link. As shown in our analysis of the intramandibular joint, there are enough structural differences in varanoids, mo- sasaurs, and snakes to justify at least the suspicion 62 FIELDIANA: GEOLOGY that this functional complex evolved convergently in snakes and nonophidian squamates (Gauthier, 1982). Several of the characters used in support of a monophyletic Pythonomorpha by Lee (1997, 1998) and Lee and Caldwell (1998) reflect the dif- ferentiation of an intramandibular joint. Estes et al. (1988, p. 253) recognized the problem of po- tential character correlation in association with the differentiation of an intramandibular joint such as the limited posterior extent of the splenial (L97: char. 72; LC98: char. B14). Lee (1997) lists a total of nine mandibular characters that group aigialo- saurs, mosasaurs, and snakes. Of these, at least four are correlated with the development of an intramandibular joint: (L97: char. 68, 1; LC98: char. B8) mobile mandibular symphysis; (L97: char. 72, 1 ) posterior end of the splenial anterior to coronoid process; (L97: char. 73, 1; LC98: char. B12) splenial-angular contact abutting, straight, mobile; (L97: char. 75, 1; LC98: char. B14) cor- onoid not sutured to splenial. Characters used by Lee (1998) to analyze squamate interrelationships include even more potential synapomorphies that are correlated with the differentiation of an intra- mandibular joint (our dp-characters in the discus- sion of the character evidence above). But in con- trast to the characters that diagnose the fossorial ecomorph, no attempt or recommendation was made by Lee (1998) to investigate the influence of potential character correlation related to the in- tramandibular joint on his phylogenetic analysis by downweighting or deleting those. In essence, however, we believe the strategy of downweight- ing characters to be misguided. In the absence of testability, some kind of correlation (ontogenetic, allometric, functional, etc.) can be invoked for any number of characters, which renders it im- possible to establish objective criteria for justifi- able degrees of downweighting. As Lee (1998) postulates convergence of the burrowing ecomorph in the dibamid-amphisba- enian clade and in snakes, the phylogenetic link of snakes to marine mosasaurs becomes essential because it alone documents that snakes could have had a marine origin (Caldwell & Lee, 1997; Lee & Caldwell, 1998; Scanlon et al., 1999) and, consequently, that fossorial habits evolved inde- pendently within snakes. The phylogenetic rela- tionships of Pachyrhachis thus becomes a key is- sue in this controversy, and Lee (1998) goes to great lengths to refute Zaher's (1998) conclusion that Pachyrhachis is not the most basal snake but the sister group of Macrostomata (i.e., of relative- ly advanced snakes) instead. However, by treating Serpentes as only one ter- minal taxon, Lee's (1998) analysis did not test Zaher's (1998) hypothesis, because Pachyrhachis had nowhere else to go other than being the sister taxon of Serpentes. Breaking up Serpentes as a terminal taxon is therefore important, not only to eliminate polymorphism in this terminal taxon but also to properly test the phylogenetic position of Pachyrhachis. Lee (1998) might have thought it unimportant to further test the relationships of Pa- chyrhachis in the context of a global analysis of squamate interrelationships because the position of this genus as the most basal snake had previ- ously been obtained by Lee and Caldwell (1998). Yet the previous analysis of snake relationships conducted by Lee (1997) again constrained the search for the sister group of snakes to varanoid squamates, and many of the characters found in support of a monophyletic Pythonomorpha were used in the subsequent placement of Pachyrhachis (Lee & Caldwell, 1998). To provide as broad a basis as possible for the test of the phylogenetic relationships of Pachyrhachis, we added to the global squamate analysis presented above those characters that were used by Caldwell and Lee (1997), Lee and Caldwell (1998), and Zaher (1998), but were not included in Lee (1998). Our discussion of Lee's (1998) evidence above indicates that many of his character definitions are flawed. For some of them, we propose new defi- nitions, for others different codings; some we pro- pose to delete from the analysis; others, which would not seem to directly affect the position of Pachyrhachis and the relationships of snakes to the amphisbaenian-dibamid clade, we simply re- tain. On the basis of only revising character def- initions and/or codings as indicated above but without deletion of any character or addition of new characters, Pachyrhachis is already found to be the sister taxon of Macrostomata instead of be- ing a link between mosasaurs and snakes, irre- spective of whether multistate characters were or- dered or unordered. At the same time, snakes group with the dibamid-amphisbaenian clade in- stead of with mosasaurs and varanoids. The rele- vant part of the cladogram reads as (varanoids (mo- sasaurs ((dibamids, amphisbaenians) snakes))). Upon deletion of the problematic characters listed above, mosasaurs cluster within a monophyletic Varanoidea, which in turn forms the sister group of a clade that includes ((dibamids, amphisbaeni- ans) snakes). This result, although only very RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 63 weakly supported, is interesting, as it corresponds partly to the cladogram obtained by Evans and Barbadillo (1998). However, we believe that dif- ferent dichotomies of this hierarchy have to be evaluated at different levels of confidence. The position of Pachyrhachis as sister group of Macrostomata is here considered to be very ro- bust. It had been obtained by critical reassessment (Zaher, 1998) of another data matrix previously used by Caldwell and Lee (1997); it is obtained in this study by a critical review of the character evidence used by Lee (1998); and it is corrobo- rated by an independently compiled data set fo- cusing on cranial characters and the interrelation- ships of basal snakes (Zaher & Rieppel, unpub- lished data). With Pachyrhachis nested within snakes as sis- ter taxon of Macrostomata, the hypothesis that snakes had an aquatic origin (Caldwell & Lee, 1997; Lee & Caldwell, 1998) loses its empirical basis. The taxa Ophidia and Serpentes as defined by Lee and Caldwell (1998) become redundant. The name to be retained for reasons of priority is Serpentes Linnaeus, 1758 (Linnaeus included cae- cilians within his Serpentes; as pointed out by Kuhn [1967], the name Serpentes can already be found in the writings of Albertus Magnus). Unless another intermediate fossil is found to relate snakes to mosasauroids, the hypothesis of a ter- restrial origin of snakes is more parsimonious. Given our currently poor understanding of the sis- ter-group relationships of snakes within squa- mates, there is no good basis on which to try to optimize a terrestrial versus an aquatic origin of snakes, however. In light of our discussion of characters related to squamate dentition (Zaher & Rieppel, 1999), braincase structure (Rieppel & Zaher, in press), and intramandibular joint structure, the putative relationship of snakes with mosasauroids appears to have a weak morphological basis. It is certainly not supported after critical assessment of the data set used by Lee (1998). Conversely, the corrected data set still shows mosasauroids (or varanoids inclusive of mosasauroids) to be pulled into a sis- ter-group relationship with the amphisbaenian/di- bamid-snake clade, which indicates similarities shared at some level of the analysis. Corrobora- tion of the monophyly of Pythonomorpha (Lee, 1997) will require new and additional morpholog- ical characters but still remains a possibility be- cause molecular data do support a sister-group re- lationship of snakes and varanoids (Reeder, 1995; Forstner et al., 1995). The same is true of some soft anatomy characters (McDowell & Bogert, 1954; McDowell, 1972; Schwenk, 1988). Critical reassessment of the data matrix com- piled by Lee (1998) results in snakes being the sister clade to dibamids and amphisbaenians. This clade is very robust in our analysis and conflicts with Lee's (1998) argument for convergence. It is interesting to note that the only total evidence ap- proach to the phylogenetic relationships of snakes reported so far found the same result, i.e., snakes grouping with dibamids and amphisbaenians, al- though the separate analysis of the same DNA data yielded a grouping of snakes with varanoids (Reeder, 1995). This finding in itself suggests that the morphological data may be subject to conver- gence, and structural constraints resulting from miniaturization may provide a reasonable expla- nation for such rampant homoplasy of morpho- logical characters (Rieppel, 1984b). However, the acceptance of convergence cannot be an a priori (and hence empirically empty) assumption but must follow from phylogenetic analysis. The dif- ficulty here is that convergence, if indeed in- volved in this case, may result in such a strong signal that the node linking snakes to the dibam- id-amphisbaenian clade cannot be broken on an- atomical grounds. The solution cannot be an ar- bitrary weighting or ordering of characters. In- stead, the analysis of snake relationships among squamates would seem to be a classical case call- ing for a combination of molecular and morpho- logical data (Reeder, 1995). In addition, it should be noted that almost all of the morphological data that have so far been brought to bear on this ques- tion are osteological characters, many of which are particularly subject to structural constraints correlated with miniaturization. In addition to mo- lecular data, it would seem that the inclusion of soft anatomy characters may help to resolve the question of homoplasy versus homology in the comparison of snakes, dibamids, and amphisba- enians (Senn & Northcutt, 1973). Hallermann (1998), for example, used the ethmoidal region (nasal capsule and associated structures) in the analysis of phylogenetic relationships of squa- mates, and found a sister-group relationship of snakes with the dibamid-amphisbaenian clade on the basis of a character complex that would seem to be less subject to structural constraints resulting from miniaturization than would be the braincase and surrounding structures. Indeed, the long list of characters shared by Si- neoamphisbaena, amphisbaenians, dibamids, and snakes (see above, primarily run 1) includes, for 64 FIELDIANA: GEOLOGY the most part, characters that appear to be related to miniaturization and/or paedomorphosis in fos- sorial or burrowing squamates. However, there are also some shared characters that would not seem to be related to fossorial or burrowing habits. Rec- ognizing that the interpretation of morphological characters in terms of putative adaptations is fraught with difficulties, we propose the following loose groupings of the characters shared by Si- neoamphisbaena, amphisbaenians, dibamids, and snakes. Characters that appear to be related to paedo- morphosis coupled with miniaturization in fosso- rial or burrowing squamates are loss of the lacri- mal (8); loss of the jugal (12); loss of the post- frontal (23); postfrontal (postorbitofrontal), where present, not forked medially (25); incomplete pos- terior orbital margin (29); loss of posterolateral processes of the parietal (37); incomplete upper temporal arch (38); loss of squamosal (40); re- duction of crista prootica (66). Characters that are coupled with structural re- modeling of the skull in miniaturized fossorial or burrowing squamates, resulting primarily from an increased relative size of the braincase, are jaw adductor muscles invading the dorsal surface of the parietal (35); quadrate suspension mainly from opisthotic (49); parietal downgrowths prominent (56); alar process on prootic absent (58); otic cap- sule expanded laterally (71); stapes robust, foot- plate large (72); closure of the posttemporal fos- sae (80, 84); neurocranium and dermatocranium positioned at same level (81). Characters potentially directly related to fos- sorial or burrowing habits are pineal foramen ab- sent (33); tympanic crest on quadrate absent (51); neural spines are low ridges (168); elongation of trunk (172); scapulocoracoid reduced (190); clav- icle absent (194); interclavicle absent (197); fore- limbs small or absent (204); pelvis reduced (206); hind limbs rudimentary or absent (211, 235, 236). Finally, characters that do not appear to be re- lated to miniaturization and/or paedomorphosis nor to fossorial or burrowing habits are frontal with straight or weakly concave orbital margin (19); palatine as long as vomer (98); intramandi- bular septum of dentary does not approach pos- teriormost tooth position (116); compound bone in lower jaw present (129); tooth crowns closely spaced (149); medial premaxillary tooth enlarged (155); lymphapophyses present (189); scleral os- sicles absent (222); cartilaginous processus ascen- dens of supraoccipital absent (232). At an early date, Rage (1982) proposed a cladistic relation- ship of snakes with amphisbaenians and dibamids; this hypothesis continues to be worth testing. A relation to fossoriality does not, after all. preclude any character a priori from being homologous in a clade comprising amphisbaenians, dibamids. and snakes. Although the potential for conver- gence certainly exists, there exists also the pos- sibility that this clade evolved from a fossorial ancestor, from which the descendants inherited the characters of the fossorial ecomorph. At this point, we refrain from comments on other parts of the hierarchy obtained by the revi- sion of Lee's (1998) data matrix. The primary rea- son is that we critically reassessed characters and character states for dibamids, amphisbaenians. varanoids, mosasaurs, Pachyrhachis, and other snakes only, in order to test the relative relation- ships of these key taxa. In order to comment on other aspects of this phylogeny, critical reassess- ment of the character evidence would have to cov- er all other nonophidian squamate families, as was pointed out above (problems of varanoid codings have also been highlighted by Gao & Norell. 1998). In this context, we note that a tree only one step longer than the most parsimonious re- construction results in a dramatic loss of resolu- tion already outside the Sineoamphisbaena-axtx- phisbaenian/dibamid-snake clade. and a tree four steps (run 2) or three steps (run 4) longer results in a loss of all resolution outside the latter clade. As discussed in more detail above, the coding of nonophidian squamates at the family level only by Lee (1998) calls for greater scrutiny. This opens an avenue to a long-term project. When a new data matrix is built to investigate squamate interrelationships, the basic topology or the rela- tive support for the different nodes discussed in this paper may change dramatically. At this point, it may suffice to point out that the conclusions reached by Lee (1998) — namely, that Pachy- rhachis is the most basal snake, linking this group to mosasauroids (Caldwell & Lee, 1997; Lee & Caldwell, 1998); that snakes (therefore) may be inferred to have had a marine origin (Scanlon et al., 1999); and that the fossorial ecomorph (there- fore) evolved convergently in dibamids plus am- phisbaenians as opposed to snakes — do not pass the test of critical examination of the character evidence he used in their support. Notes Added in Proof While this paper was in press, Lee, Bell, et al. (1999) presented a gradualistic model for the evo- RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 65 lution of the ophidian feeding mechanism from the ancestral mosasauroid condition, with Pachyr- hachis as an intermediate stage. Mediolateral flex- ion of the mandible in mosasaurs (previously de- scribed by Telles Antunes [1964, pp. 156 ff. and Fig. 22], in a monograph overlooked by Lee, Bell, et al. [1999]) was interpreted as the starting point for the evolution of ophidian jaw mechanics. However, both the description of the intramandi- bular joint in squamates given above and the ma- crostomatan affinities of Pachyrhachis invalidate this scenario. Lee, Caldwell, et al. (1999) also redescribed Pachyophis woodwardi Nopcsa (1923) from the Cretaceous of Bosnia-Herzegovina. In order to analyze its relationships, Lee, Caldwell, et al. (1999) used the data matrix in Lee (1998), which we reviewed above, to which they added two characters: pachyostosis of mid-dorsal vertebrae and ribs, and the laterally compressed body. Pachyophis was found to be the sister taxon of Pachyrhachis on the basis of these two characters, and the two taxa were classified in a family of their own, the Pachyophiidae, again placed as sis- ter taxon to all other snakes. We were unable at this time to inspect the holo- type of Pachyophis, but based on the description by Lee, Caldwell, et al. (1999), we doubt the ophidian status of this taxon (see also Rage, 1984). With an estimated 120 presacral vertebrae, the vertebral count is lower in Pachyophis. The posterior part of the preserved vertebral column appears to us to closely approach the sacral or cloacal region. Pachyophis shows a greater degree of pachyostosis than Pachyrhachis, and pachyos- tosis persists into the posterior dorsal region in- stead of remaining restricted to the midtrunk as in Pachyrhachis. The latter taxon also shows elon- gated, nonpachyostotic ribs shortly in front of the "pelvic" region, before the last three or four ribs become abruptly shortened. In Pachyophis, the ribs gradually decrease in length in the posterior dorsal region, which accordingly would not have been laterally compressed as it was it in Pachy- rhachis. More important, Lee, Caldwell, et al. (1999) interpret a fragmentary bone as part of the right dentary. Their rendering of this fragment in their Figure 3b shows the anteriorly convex angular be- ing received by the posteriorly concave splenial. We understand from Nopcsa's (1923) description that this element is difficult to identify, but if the interpretation given by Lee, Caldwell, et al. (1999, Fig. 3b) is correct, Pachyophis shares the mosa- sauroid intramandibular joint, which is different from that of snakes and also different from that of Pachyrhachis (Lee & Caldwell, 1998, Fig. 4). Given the incompleteness of the material, we be- lieve the best solution is to retain Pachyophis (Pachyophiidae) as incertae sedis among squa- mates. Acknowledgments We would like to thank N. E. Arnold and C. McCarthy (bmnh), Harold Voris and Alan Resetar (fmnh), Eitan Tchernov (huj), and Guiseppe Puor- to (Instituto Butantan) for permission to study the collections under their care. N. C. Fraser, M. Kearney, and R. Etheridge kindly read an earlier draft of this paper, offering much helpful advice and criticism. The research of the junior author was supported by grants from fapesp (Fundacio de Amparo Persquisa de Sao Paulo, Brasil). Literature Cited Backstrom, K. 1931. Rekonstruktionsbilder zur Onto- genie des Kopfskeletts von Tropidonotus natrix. Acta Zoologica, Stockholm, 12: 83-143. Baumeister, L. 1908. Beitrage zur Anatomie und Phy- siologic der Rhinophiden. Zoologische Jahrbiicher, Abteilung fur Anatomie und Ontogenie der Tiere, 26: 423-526. Bell, G. L., Jr. 1997. A phylogenetic revision of North American and Adriatic mosasaurs, pp. 293-332. In Callaway, J. M., and E. L. Nicholls, eds., Ancient Ma- rine Reptiles. Academic Press, San Diego. Bellairs, A. d'A., and A. M. Kamal. 1981. The chon- drocranium and the development of the skull in recent reptiles. In Gans, C., and T. S. Parsons, eds., Biology of the Reptilia, 11: 1-263. Academic Press, London. Brock, G. T. 1929. On the development of the skull of Leptodeira hotamboia. Quarterly Journal of Micro- scopical Sciences, 73: 289-334. . 1932. The skull of Leptotyphlops (Glauconia nigricans). Anatomischer Anzeiger, 73: 199-204. Caldwell, M. W. 1999. Squamate phylogeny and the relationships of snakes and mosasauroids. Zoological Journal of the Linnean Society, 125: 115-147. Caldwell, M. W., and M. S. Y. Lee. 1997. A snake with legs from the marine Cretaceous of the Middle East. Nature, 386: 705-709. Camp, C. L. 1923. Classification of the lizards. Bulletin of the American Museum of Natural History, 48: 289-481. 1942. California mosasaurs. University of Cal- ifornia, Memoirs, 13: 1-68. 66 FIELDIANA: GEOLOGY Carroll, R. L. 1988. Vertebrate Paleontology and Evo- lution. W. H. Freeman and Co., New York. Carroll, R. L., and M. deBraga. 1992. Aigialosaurs: mid-Cretaceous varanoid lizards. Journal of Vertebrate Paleontology, 12: 66-86. Cope, E. D. 1869. On the reptilian orders Pythonomor- pha and Streptosauria. Proceedings of the Boston So- ciety of Natural History, 12: 250-266. Cundall, D. 1995. Feeding behaviour in Cylindrophis and its bearing on the evolution of alcthinophidian snakes. Journal of Zoology, London, 237: 353-376. Cundall, D., V. Wallach, and D. A. Rossman. 1993. The systematic relationships of the snake genus An- omochilus. Zoological Journal of the Linncan Society, 109: 275-299. DeBeer, G. 1937. The Development of the Vertebrate Skull. Clarendon Press, Oxford. Edmund, A. G. 1960. Tooth replacement phenomena in the lower vertebrates. Royal Ontario Museum, Life Sciences Division, Contributions, 52: 1-190. Estes, R., T H. Frazzetta, and E. E. Williams. 1970. Studies on the fossil snake Dinilysia patagonica Woodward. Part I. Cranial morphology. Bulletin of the Museum of Comparative Zoology, 140: 25-74. Estes, R., K. deQuieroz, and J. Gauthier. 1988. Phy- logenetic relationships within Squamata, pp. 1 19-281. In Estes, R., and G. Pregill, eds., Phylogenetic Rela- tionships of the Lizard Families. Stanford University Press, Stanford. Evans, S. E., and L. J. Barbadillo. 1998. An unusual lizard (Reptilia: Squamata) from the Early Cretaceous of Las Hoyas, Spain. Zoological Journal of the Lin- nean Society, 124: 235-265. Forstner, M. R., S. K. Davis, and E. Arevalo. 1995. Support for the hypothesis of anguimorph ancestry for the suborder Serpentes from phylogenetic anaysis of mitochondrial DNA sequences. Molecular Phyloge- netics and Evolution, 4: 93-102. Frazzetta, T H. 1959. Studies on the morphology and function of the skull in the Boidae (Serpentes). Part 1 . Cranial differences between Python sebae and Ep- icrates cenchris. Bulletin of the Museum of Compar- ative Zoology, 119: 453-472. . 1999. Adaptations and significance of the cra- nial feeding apparatus of the sunbeam snake (Xeno- peltis unicolor): Part I. Anatomy of the skull. Journal of Morphology, 239: 27-43. Frost, D. R., and R. Etheridge. 1989. A phylogenetic analysis and taxonomy of iguanian lizards (Reptilia: Squamata). University of Kansas Museum of Natural History, Miscellaneous Publication. 81: 1-65. Gao, K., and M. A. Norrell. 1998. Taxonomic revision of Carusia (Reptilia: Squamata) from the Late Cre- taceous of the Gobi Desert and phylogenetic relation- ships of anguimorphan lizards. American Museum Novitates, 3230: 1-51. Gauthier, J. A. 1982. Fossil xenosaurid and anguid liz- ards from the early Eocene Wasatch formation, south- eastern Wyoming, and a revision of the Anguioidca. Contributions to Geology, University of Wyoming. 21: 7-54. Genest-Villard, H. 1966. Developpcmcnt du crane d'un boide: Sanzinia nuulagascariensis. Mdmoires du Museum national d'Histoire naturelle. A40: 207-262. Gingerich, P. D. 1973. Skull of Hesperornis, and the early evolution of birds. Nature, 243: 70-73. Greer, A. E. 1970. A subfamilial classification of scin- cid lizards. Bulletin of the Museum of Comparative Zoology. 139: 151-183. . 1985. The relationships of the lizard genera Anelytropsis and Dibamus. Journal of Herpetology, 19: 116-156. Gregory, J. T. 1951. Convergent evolution: the jaws of Hesperornis and the mosasaurs. Evolution, 5: 345- 354. Haas, G. 1930. Ubcr das Kopfskclett und die Kaumu- skulatur dcr Typhlopidcn und Glauconiidcn. Zoolo- gische Jahrbucher, Abtcilung fur Anatomic und On- togenic der Tiere, 52: 1-94. . 1964. Anatomical observations on the head of Liotyphlops albirostris (Typhlopidac. Ophidia). Acta Zoologica, Stockholm, 45: 1-62. . 1968. Anatomical observations on the head of Anotnalepis aspinosus (Typhlopidac. Ophidia). Acta Zoologica, Stockholm, 49: 63-139. -. 1979. On a new snakclikc reptile from the Low- er Cenomanian of Ein Jabrud. near Jerusalem. Bulletin du Musdum national d' Histoirc naturelle (Scr. 4). 1: 51-64. 1980. Pachyrhachis problematicus Haas, snake- like reptile from the Lower Cenomanian: ventral view of the skull. Bulletin du Musdum national d' Histoirc naturelle (Ser. 4), 2: 87-104. Haines, R. W. 1969. Epiphyses and sesamoids. In Gans. C, T S. Parsons, and A. d'A. Bellairs, eds.. Biology of the Reptilia, 1: 81-115. Academic Press, London. Hallermann, J. 1998. The ethmoidal region of Dibamus taylori (Squamata: Dibamidac), with a phylogenetic hypothesis on dibamid relationships within Squamata. Zoological Journal of the Linncan Society. 122: 385- 426. Haluska, F, and P. Alberch. 1989. The cranial devel- opment of Elaphe obsoleta (Ophidia, Colubridac). Journal of Morphology, 178: 37-55. Hofstetter, R.. and J. -P. Gasc. 1969. Vertebrae and ribs of Modern reptiles. In Gans, C, T. S. Parsons, and A. d'A. Bellairs, eds., Biology of the Reptilia. 1: 201-310. Academic Press, London. Hoge. A. R. 1964. Nota sobrc Xenopeltis unicolor Rcin- wardt 1827. Memoric Instituto Butantan. 30: 31-33. Irish, F. J. 1989. The role of heterochrony in the origin of a novel bauplan: evolution of the ophidian skull. Gcobios. Mcmoirc Special. 12: 227-233. Kamal, A. M.. H. G. Hammouda. and F. M. Mokhtar. 1970. The development of the ostcocranium of the Egyptian cobra. I. The embryonic ostcocranium. Acta Zoologica. Stockholm. 51: 1-17. Klugk. A. G. 1989. Progress in squamatc classification. Hcrpctologica. 45: 368-379. . 1991. Boinc snake phylogcny and research cy- cles. Miscellaneous Publications. Museum of Zoolo- gy. University of Michigan. 178: 1-58. Kuhn. O. 1967. Amphibicn und Rcptilicn. Katalog dcr RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 67 Subfamilicn und hohcren Taxa mit Nachwcis des er- sten Auftretens. Gustav Fischer, Stuttgart. Lauder, G. V, and K. F. Liem. 1989. The role of his- torical factors in the evolution of complex organismal functions, pp. 63-78. In Wake, D. B., and G. Roth, eds.. Complex Organismal Functions: Integration and Evolution in Vertebrates. John Wiley and Sons, Chich- ester. Lee, M. S. Y. 1997. The phylogeny of varanoid lizards and the affinities of snakes. Philosophical Transactions of the Royal Society of London, B352: 53-91. . 1998. Convergent evolution and character cor- relation in burrowing reptiles: towards a resolution of squamate relationships. Biological Journal of the Lin- nean Society, 65: 369-453. Lee, M. S. Y., G. L. Bell, Jr., and M. W Caldwell. 1999. The origin of snake feeding. Nature, 400: 655- 657. Lee, M. S. Y, and M. W Caldwell. 1998. Anatomy and relationships of Pachyrhachis problematicus, a primitive snake with hindlimbs. Philosophical Trans- actions of the Royal Society of London, B352: 1521- 1552. Lee, M. S. Y, M. W Caldwell, and J. D. Scanlon. 1999. A second primitive marine snake: Pachyophis woodwardi from the Cretaceous of Bosnia-Herzego- wina. Journal of Zoology, London, 248: 509-520. Lee, M. S. Y, and P. Doughty. 1997. The relationship between evolutionary theory and phylogenetic analy- sis. Biological Reviews, 72: 471-495. Lingham-Soliar, T, and D. Nolf. 1989. The mosasaur Prognathodon (Reptilia) from the Upper Cretaceous of Belgium. Bulletin de lTnstitut Royal des Sciences Naturelles de Belgique, 59: 137-182. Linnaeus, C. 1758. Systema Naturae, 10th edition. List, J. C. 1966. Comparative osteology of the snake families Typhlopidae and Leptotyphlopidae. Illinois Biological Monographs, 36: 1-112. University of Il- linois Press, Urbana. McDowell, S. B. 1972. The evolution of the tongue of snakes, and its bearing on snake origins. Evolutionary Biology, 6: 191-273. . 1975. A catalogue of the snakes of New Guinea and the Solomons, with special reference to those in the Bernice P. Bishop Museum. Part II. Anilioidea and Pythoninae. Journal of Herpetology, 9: 1-79. . 1979. A catalogue of the snakes of New Guinea and the Solomons, with special reference to those in the Bernice P. Bishop Museum. Part III. Boinea and Acrochordoidea (Reptilia, Serpentes). Journal of Her- petology, 13: 1-92. McDowell, S. B., and C. M. Bogert. 1954. The sys- tematic position of Lanthanotus and the affinities of anguinomorphan lizards. Bulletin of the American Museum of Natural History, 105: 1-142. Montero, R., C. Gans, and M. L. Lions. 1999. Embry- onic development of the skeleton of Amphisbaena darwini heterozonata (Squamata: Amphisbaenidae). Journal of Morphology, 239: 1-25. Nopscsa, F 1923. Eidolosaurus und Pachyophis: Zwei neue Neocom-Reptilien. Palaeontographica, 65: 97- 154. Oelrich, T. M. 1956. The anatomy of the head of Cten- osaura pectinata (Iguanidae). Miscellaneous Publica- tions, Museum of Zoology, University of Michigan, 94: 1-122. Parker, W. K. 1879. On the structure and development of the skull in the common snake (Tropidonotus na- trix). Philosophical Transactions of the Royal Society of London, 169: 385-417. Peyer, B. 1912. Die Entwicklung des Schadelskelettes von Vipera aspis. Gegcnbaur's Morphologisches Jahr- buch, 44: 563-621. Rage, J.-C. 1982. La phylogenie des Lepidosauriens (Reptilia): Une approche cladistique. Comptes Ren- dus, Academie des Sciences, Paris, 294: 563-566. . 1984. Serpentes, pp. 1-80. In Wellnhofer, P., ed., Handbuch der Palaoherpetologie, Teil 11. G. Fischer, Stuttgart. Reeder, T. W 1995. Phylogenetic placement of snakes within the Squamata: evidence from molecules and morphology. American Society of Ichthyologists and Herpetologists, Program and Abstracts, p. 166. Uni- versity of Alberta, Canada. Rieppel, O. 1977. Studies on the skull of the Henophi- dia. Journal of Zoology, London, 181: 145-173. . 1978. The braincase of Annie lla pulchra Gray (Lacertilia: Aniellidae). Revue suisse de Zoologie, 85: 617-624. 1979a. A functional interpretation of the var- anid dentition (Reptilia, Lacertilia, Varanidae). Gegen- baurs Morphologisches Jahrbuch, 125: 797-817. . 1979b. The evolution of the basicranium in the Henophidia (Reptilia, Serpentes). Zoological Journal of the Linnean Society, 66: 411-431. 1981. The skull and jaw adductor musculature in some burrowing scincomorph lizards of the genera Acontias, Typhlosaurus and Feylinia. Journal of Zo- ology, London, 195: 493-528. ' 1983. A comparison of the skull of Lanthanotus horneensis (Reptilia: Varanoidea) with the skull of primitive snakes. Zeitschrift fur Zoologische Syste- matik und Evolutionsforschung, 21: 142-153. . 1984a. The cranial morphology of the fossorial lizard genus Dihamus with a consideration of its phy- logenetic relationships. Journal of Zoology, London, 204: 289-327. . 1984b. Miniaturization of the lizard skull: its functional and evolutionary implications. In M. W J. Ferguson, ed., The Structure, Development and Evo- lution of Reptiles. Zoological Society of London Sym- posia, 52: 503-520. Academic Press, London and New York. . 1984c. The structure of the skull and jaw ad- ductor musculature in the Gekkota, with comments on the phylogenetic relationships of the Xantusiidae (Reptilia: Lacertilia). Zoological Journal of the Lin- nean Society, London, 82: 291-318. 1988. A review of the origin of snakes. Evo- lutionary Biology, 22: 37-130. 1993a. Middle Triassic reptiles from Monte San Giorgio: recent results and future potential of analysis. Paleontologia Lombarda, n.s., 2: 131-144. . 1993b. Patterns of diversity in the reptilian 68 FIELDIANA: GEOLOGY skull. In Hanken, J., and B. K. Hall, cds.. The Skull. 2: 344-390. University of Chicago Press, Chicago. 1994. The Lepidosauromorpha: an overview with special emphasis on the Squamata. pp. 23-37. In Fraser, N. C, and H.-D. Sues, eds.. In the Shadow of the Dinosaurs. Early Mesozoic Tetrapods. Camhridge University Press, Cambridge. Rieppel, O., and H. Zaher. In press. The braincascs of mosasaurs and Varanus, and the relationships of snakes. Zoological Journal of the Linnean Society, 00: 000-000. Russell, D. A. 1967. Systematica and morphology of American mosasaurs. Pcabody Museum of Natural History, Yale. Bulletin, 23: 1-240. Scanlon, J. D., M. S. Y. Lee, M. D. Caldwell, and R. Shine. 1999. The palcoecology of the primitive snake Pachyrhachis. Historical Biology. 13: 127-152. Schwenk, K. 1988. Comparative morphology of the lep- idosaur tongue and its relevance to squamate phylog- eny. pp. 569-598. In R. G. Prcgill, cd.. Phylogcnetic Relationships of the Lizard Families. Stanford Uni- versity Press. Stanford. CA. Senn, D. G., and R. G. Northcutt. 1973. The forcbrain and midbrain of some squamates and their bearing on the origin of snakes. Journal of Morphology, 140: 135-152. Sheldon, A. 1997. Ecological implications of mosasaur bone microstructure, pp. 333-354. In Callaway, J. M., and E. L. Nicholls, cds.. Ancient Marine Reptiles. Ac- ademic Press, San Diego. Starck, D. 1979. Vergleichcndc Anatomic dcr Wirbel- tiere auf cvolutionsbiologischcr Grundlage. Vol 2. Springer Vcrlag, Berlin. SwonoRD. D. L. 1990. PAUP— Phylogcnetic Analysis Using Parsimony, Version 3.0. Illinois Natural History Survey, Champaign, IL. SwOFFORO. D. L.. and D. P. Begle. 1993. PAUP— Phy- logenetic Analysis Using Parsimony. Version 3.1. Laboratory of Molecular Systematics, Smithsonian In- stitution. Washington. DC. In i is Autunes. M. 1964. O Neoeretacico e o Ceno- zoico do Litoral de Angola. Junta de Investigacocs do Ultramar. Lisboa. UNDERWOOD, G. 1957. Lanthanotus and the anguino- morph lizards: a critical review. Copeia. 1957: 20-30. . 1967. A Contribution to the Classilication of Snakes. British Museum (Natural History). London. Wu. X.-C, D. B. Brinkman. and A. P. Rtssi i i . 19%. Sineoamphisbaena hexatabularis, an amphisbaenian (Diapsida: Squamata) from the Upper Cretaceous reedbeds at Bayan Mandahu (Inner Mongolia. Peo- ple's Republic of China), and comments on the phy- logcnetic relationships of the Amphisbaenia. Canadian Journal of Earth Sciences, 33: 541-577. Young. B. A. 1998. The comparative morphology of the intermandibular connective tissue in snakes (Reptilia: Squamata). Zoologischer Anzeiger. 237: 59-84. Zaher. H. 1998. The phylogcnetic position of Pachyr- hachis within snakes (Squamata. Lepidosauria). Jour- nal of Vertebrate Paleontology. 18: 1-3. Zaher. H.. and O. Rieppel. 1999. Tooth implantation and replacement in squamates. with special reference to mosasaur lizards and snakes. American Museum Novitatcs, 3271: 1-19. Zangerl. R. 1944. Contribution to the osteology of the skull of the Amphisbaenidae. The American Midland Naturalist, 31: 417-454. RIEPPEL AND ZAHER: INTRAMANDIBULAR JOINT IN SQUAMATES 69 A Selected Listing of Other Fieldiana: Geology Titles Available Status of the Pachypleurosauroid Psilotrachelosaurus toeplitschi Nopcsa (Reptilia: Sauropterygia), from the Middle Triassic of Austria. By Olivier Rieppel. Fieldiana: Geology, n.s., no. 27, 1993. 17 pages, 9 illus. Publication 1448, $10.00 Osteology of Simosaurus gaillardoti and the Relationships of Stem-Group Sauropterygia. By Oli\ior Rieppel. Fieldiana: Geology, n.s., no. 28, 1994. 85 pages, 71 illus. Publication 1462, $18.00 The Genus Placodus: Systematics, Morphology, Paleobiogeography, and Paleobiology. By Olivier Rieppel. Fieldiana: Geology, n.s., no. 31, 1995. 44 pages, 47 illus. Publication 1472, $12.00 Pachypleurosaurs (Reptilia: Sauropterygia) from the Lower Muschelkalk, and a Review of the Pachypleurosauroidea. By Olivier Rieppel and Lin Kebang. Fieldiana: Geology, n.s., no. 32, 1995. 44 pages, 28 illus. Publication 1473, $12.00 A Revision of the Genus Nothosaurus (Reptilia: Sauropterygia) from the Germanic Triassic, with Comments on the Status of Conchiosaurus clavatus. By Olivier Rieppel and Rupert Wild. Fieldiana: Geology, n.s., no. 34, 1996. 82 pages, 66 illus. Publication 1479, $17.00 Revision of the Sauropterygian Reptile Genus Cymatosaurus v. Fritsch, 1 894, and the Relationships of Germanosaurus Nopcsa, 1928, from the Middle Triassic of Europe. By Olivier Rieppel. Fieldiana: Geology, n.s., no. 36, 1997. 38 pages, 16 illus. Publication 1484, $11.00 The Status of the Sauropterygian Reptile Genera Ceresiosaurus, Lariosaurus, and Silvestrosaurus from the Middle Triassic of Europe. By Olivier Rieppel. Fieldiana: Geology, n.s., no. 38, 1998. 46 pages, 21 illus. Publication 1490, $15.00 Sauropterygia from the Middle Triassic of Makhtesh Ramon, Negev, Israel. By Olivier Rieppel. Jean-Michel Mazin, and Eitan Tchernov. Fieldiana: Geology, n.s., no. 40, 1999. 85 pages, 58 illus. Publication 1499, $25.00 Order by publication number and/or ask for a free copy of our price list. All orders must be prepaid Illinois residents add current destination tax. All foreign orders are payable in U.S. dollar-checks drawn on any U.S. bank or the U.S. subsidiary of any foreign bank. Prices and terms subject to change without notice. Address all requests to: FIELD MUSEUM OF NATURAL HISTORY Library— Publications Division 1400 So. Lake Shore Drive Chicago, Illinois 60605-2498, U.S.A. UNIVERSITY OF ILLINOIS-URBANA 3 0112 053323322 Field Museum of Natural History 1400 So. Lake Shore Drive Chicago, Illinois 60605-2496 Telephone: (312) 665-7055