New material of the basal Thyreophoran Scutdlosaurus lawleri from the Kayenta Formation (Lower Jurassic) of Arizona

PaleoBios 20(1): 13-23, April 15, 2000 < 2000 University of California Museum of Paleontology New material of the basal Thyreophoran Scutdlosaurus lawleri from the Kayenta Formation (Lower Jurassic) of Arizona JASON N. ROSENBAUM and KEVIN PADIAN 1 Museum of Paleontology, University of California, Berkeley, CA 94720-4780; kpadian@socrates.berkeley.edu As a basal member of the thyreophoran lineage of ornithisehian dinosaurs, Scutdlosaurus lawleri is important, yet still incompletely known. Material recovered from the Lower Jurassic Kayenta Formation of Northern Arizona in 1983 reveals new details about its anatomy. Portions of the skull, ankle, forearm, and pelvis are particularly important because they were incompletely preserved or missing in the holotype. The phylogeny of the Ornithischia has been modified cladistically since Colbert's description, warranting a reassessment of the comparative morphol ogy and phylogcnetic position of S. lawleri. Skull fragments indicate an emendation to the clade Thyrephora is necessary. New limb material suggests that S. lawleri was not an obligate quadraped. Histological observations indicate that S. lawleri'grew slowly throughout its life. INTRODUCTION Scutdlosaurus lawleri is a small ornithisehian dinosaur known from the Kayenta Formation of Northern Arizona (Colbert 1981). It is important not only because it is thought to be a basal member of the thyreophoran lineage of ornithisehian dinosaurs, but also because it is one of the earliest relatively well known members of the Ornithischia. Any information about S. lawleri can potentially elucidate the origins and history of both of these groups. The animal has been described so far only from the holotype and the paratype specimens, both collected by Douglas Lawler for the Museum of Northern Arizona (MNA) (Colbert 1981). Expeditions to the Kayenta Formation by Harvard University in the late 1970s and early 1980s and by a joint Harvard-University of Texas team in 1997 procured additional specimens that are yet undescribed (F.A. Jenkins Jr., personel communication). Colbert (1981) based his description of S. lawleri primarily on comparisons to Lcsothosaurus diagnosticus, then thought to be the most appropriate taxon against which to compare a new basal ornithisehian, and he classified it with I.esothosaiirusas a member of Fabrosauridae, the most basal group of ornithischians then recognized. However, ornithisehian phylogeny has been modified cladistically since Colbert's (1981) description. Sercno (1986) first recognized Scutdlosaurus as the most basal member of Thyreophora, the group that contains the two major clades Ankylosauria and Stegosauria and their relatives. In several possible trees (many of which were published after Colbert's monograph of,s". lawleri), Hcterodontosaurus tucki would be at least as good a comparative specimen as L. diagnosticus (Benton 1990), because the latter is among the earliest members of the Ccrapoda, the sister taxon to Thyreophora (Sercno 1986). In cladograms constructed by Norman (1984a, b), Cooper (1985), and Sercno (1986), Hcterodontosaurus is a member of the sister group to the 1 author for correspondence Thyreophora. In Weishampel and Witmer's (1990) tree, H. tucki is considered the earliest member of Ccrapoda (the sister taxon to Thyreophora) and a basal member of the sub-clade Ornithopoda. This paper introduces new anatomical comparisons to H. tucki, and reevaluates many of Colbert's comparisons in this light. James M. Clark collected the six specimens described in this paper in 1981 and in 1983 for the University of California Museum of Paleontology at Berkeley (UCMP), from localities along the Adeci Fxchii Cliffs in northern Arizona on the lands of the Navajo Nation. Only sparse fragments represent four of the six individuals. The remaining two specimens were disarticulated. The less complete of these two (UCMP 170829) is represented by fragments, many of which appear to be distorted, but enough pieces are present and recognizable that the specimen does have comparative value. Unfortunately, in the most complete specimen (UCMP 130580) nearly all of the bones are extensively fragmented. Most of the fragmentation probably occurred relatively recently, however, because the fragments often fit together very well, and do not show much wear on the break facets. SYSTEMATIC PALEONTOLOGY ARCIIOSAURIA: DINOSAURIA: ORNITHISCHIA: THYRKOPHORA. Genus: Scutclhsaurus SPECIES: Scutdlosaurus lawleri Colbert 1981 Holotype MNA PI. 175, a relatively complete skeleton. Paratype MNA PI. 1752, incomplete material of an individual somewhat larger than the holotype. Referred Specimens-UCMP 130580, UCMP 170829, UCMP 130581, UCMP 175166, UCMP 175167, UCMP 175168. Locality All six LJCMP specimens were recovered from the Kayenta Formation in northern Arizona (Locality number V85010). Locality information and field notes are on file at UCMP.

14 PALEOBIOS, VOL. 20, NUMBER 1, 2000 Description- Skull: The skull bones of UCMP 130580 are mostly missing. However, portions of both frontals, both quadrates, an off-center portion of the parietal, the right jugal, and both dentaries are present (Figs. 1, 5A-G, 6A-G). Additionally, the basioccipital is nearly complete (Figs. 1G, 5F, 6F). The left frontal is missing only its most rostral portion (Fig. IF). The right frontal, however, is missing all but the most rostral process. The frontal is roughly triangular; the longest side forms the dorsal margin of the orbit, and the shortest side abuts the parietal caudally. The parietal fragment articulates with the left frontal. The suture is not fused, however. The small fragment of the parietal shows the beginning of ridges along its sides that meet posteriorly (Fig. 1H). The shapes of the parietal and the frontals are reminiscent of the skull of Hypsilophodon (Sues and Norman 1990). The jugal is an inverted T-shape, with the horizontal bar thickening towards the junction with the vertical process. On it's medial side (which faces the interior of the skull), the rostral process has a small pinched peak. The vertical process has a groove on its rostral side. This groove likely held another bone in place, as it does in Emausauriis (Haubold 1990). The condyles of both the left and right quadrate are present, and they arc thicker medially than laterally (Figs. 1C-D, 5E, 5G, 6E, 6G). Numerous small fragments are probably cranial, but are too fragmentary for identification. UCMP 175168 also has a large portion of the left frontal, and UCMP 175166 has the condyle of the left quadrate. Axial Skeleton: UCMP 130580 preserves 33 complete centra (five cervical, eight thoracic, two sacral, 18 caudal) (Fig. 61.). Additionally, thirty centra fragments are less reliably identifiable. Only the caudal vertebrae have neural arches fused to the centra, and this fusion becomes more pronounced caudally. There are numerous neural arch fragments, nearly all of which are fragmentary. Only one prccaudal arch is still associated with a centrum. Two of the more anterior caudal vertebrae retain relatively simple, nonelaborated neural arches. The more posterior (and smaller) centra of the caudal series also retain a very simple arch. Of special note are the atlas intercentrum, the axis neural arch, and the odontoid (Figs. 5H-J, 6H-J). UCMP 130580 also has innumerable rib fragments, but there is a complete sacral rib (Fig. 4G) that is substantially wider than the thoracic ribs. It is robust at its proximal articulation with the sacrum and bears complex articulation facets. It narrows to a short neck, and then immediately broadens at its distal end, giving the entire bone an overall shape like a flattened anvil. Shoulder Girdle and Forelimb: The incomplete scapulae of UCMP 130850 are crushed and slightly flattened, but retain much of their original shape (Figs. 4A-B). The left scapula is more complete than the right. Both bones lack most of the margins, so it is impossible to determine the full dimensions of the bone. Neither coracoid is present except for tiny fragments. The humeri of UCMP 130850 are highly fragmented and possibly distorted, but mostly complete (Figs. 4C-D). It is not possible to measure the length of the humeri with absolute certainty; however, they are complete enough that estimates can be made with a good degree of reliability (Table 1). The humeral shafts of UCMP 130580 seem to have slightly more torsion than those of MNA PI.175, but this difference may be more taphonomic than morphologic. The right radius is distorted, but complete. The left radius, however, is preserved in its entirety, with no apparent distortion (Fig. 4E). This provides the length of the forearm relative to other limb bones, which was heretofore only estimable (Table 1). Neither ulna is complete, but the proximal and distal ends of both the left and right ulna arc present (Fig. 4F). The distal end of the radius of UCMP 170829 is preserved. Two very small bones, one cuboidal and the other roughly pyramidal, are the only preserved bones of the carpus (Fig. 5H). One metacarpal is present, probably number 3 or 4 (Fig. 51). Neither the carpals nor the metacarpals can easily be assigned to the left or right side because of the dearth of adequate comparative material. This same problem holds for the phalanges. There is one ungual and one proximal phalanx, and three phalangeal fragments, which are all quite small. The proximal articular surface of the phalanx is concave, and the phalanx tapers distally. At about the midline, the phalanx broadens into a trapezoidal shape, which is round on its articular surface. It is unclear whether the phalanx is the most proximal of its digit. Pelvic Girdle and Hindlimb: The pelvis of UCMP 130850 is mostly absent, but the few fragments that remain are quite well preserved. Two long, flat, slightly sigmoidal bones represent the shafts of the left and right ischia (Figs. 2B, 4L-M). These shafts are strikingly similar to the ischia of Lesotbosanrus diajjnostiais (Thulborn 1972). The proximal portion of the left ischium, including its articulations with the ilium and pubis, is also present (Figs. 3B, 41,). Again, it is very similar to the same bone in /.. dinijnostkus (Thulborn 1972), with a broad, thick process ventrally, and a smaller, peg-like articulation dorsal to it. There is a very small fragment from the central portion of the right ilium, as well as a much larger portion of the left ilium. The fragment of the left ilium retains most of the slender anterior process (Figs. 2A, 4H). The process tapers toward its cranial end, where it expands again, and ends in a squared-off process of bone. UCMP 170829 also has a small fragment of the ilium. The femora of UCMP 130850 are both crushed. The right femur is only represented by a few small fragments, including a small portion of the distal end and what is possibly the lesser trochanter. The entire proximal end of the left femur is present, albeit substantially crushed and distorted (Fig. 41). The distal end of the left femur is intact and undistortcd (Fig. 4N). UCMP 170829 includes the distorted distal end of a femur.

ROSENBAUM & PADIAN-NEW MATERIAL OF SCUTKLI.OSAURUS 15 Fig. 1. ScutcttosauruslawleriskaH fragments, UCMP 130580. 3x. A. Right jugal, external lateral view. B. The same, internal lateral view. C. I-cft quadrate, caudal view. D. The same, ventral view (condyle). E. Left frontal in dorsal (above) and ventral (below) views. Caudal end at left. The caudal end of the frontal forms the suture with the parietal (Fig. 1 J), and the medial side forms the suture with the right frontal (Fig. IF). F. Right frontal in dorsal (above) and ventral (below) views, caudal end at right. The medial edge forms the suture with the left frontal (Fig. IK). G. Basioccipital in internal (dorsal) and external (ventral) views. H. Parietal, external (left) and internal (right) views, rostral end at top. The rostral edge forms a suture with the left frontal (Fig. IE).

16 PALEOBIOS, VOL. 20, NUMBER 1, 2000 Fig. 3. Scutetlosauruslaivleri, UCMP 130580. 3x. Proximal end of left ischium, lateral view. 1 cm Fig. 2. Scutellosaurttslawkripclvic fragments, UCMP 130580. 2x. A. Anterior process of the right ilium, lateral view above, medial view below. B. Shaft of left ischium, in lateral (left) and medial (right) views; proximal end at top. The entire right tibia is present, save for a small portion of the shaft near the proximal end, and a narrow ridge of bone along the shaft toward the distal end (Fig. 4J). The shaft of the right tibia is slightly compressed and fractured. Of the left tibia, only small fragments of the proximal and distal ends remain (Fig. 40). The proximal ends of both fibulae (Fig. 4K) of UCMP 130850 are well preserved, along with a portion of one shaft. The preserved shaft is very straight, and it is circular in cross section, which hinders any attempt to determine to which side the shaft fragment belongs. The proximal end of one fibula from UCMP 170829 is preserved. The left astragalus of UCMP 130850 is well preserved, as is a small, kidney-shaped bone that may represent distal tarsal 2 (Figs. 5M-N, 6M). Metatarsals from both feet are well preserved (Figs. 50- P), and appear similar to those of Hcterortontosaurus (Santa I.uca 1980). The first metatarsal is represented by the left proximal end and most of the shaft, and by the right distal end and most of the shaft. Both seem to be slightly distorted, but the first metatarsal is substantially shorter than the other metatarsals. The proximal end is triangular, and bulges proximally in a convex articular surface. The longest side of the triangle rests against the second metatarsal. The shaft is flattened medio-laterallv, but becomes round in cross-section toward the distal end. The second left metatarsal is represented by the distal end and part of the shaft. The right third metatarsal is nearly complete, and the left is present only in the distal end and a fragment of the shaft. The third metatarsal is straighter than the fourth. It is wider anteroposteriorly than mediolaterally at its proximal end, but the shaft flattens distally, and is wider mediolaterally than anteroposteriorly at the distal end. The right fourth metatarsal is complete, and the left is missing the central portion of the shaft. The bone curves medially in anterior view, and sigmoidally in lateral view. Its proximal articular facet is triangular, with the upper lateral edge rounded. The pes is represented by three proximal phalanges, two ungual phalanges, and five fragments (Figs. 5Q-R). These are similar in general shape to those of the mantis, bin larger and longer. The phalanx of the first digit is longer than those of the other three digits. Colbert's (1981) diagrams of the pedal phalanges of S. Ituvlcri indicate a substantial extensor pit on the dorsal surface toward the distal end of each phalanx. There are small indentations in the phalanges of the UCMP and MNA specimens, but not as substantial as Colbert's illustrations suggest. Osteoderms: UCMP 130580 has an extensive amount of scute material (Fig. 6N). The osteoderms tend to be relatively uniform in thickness, with a midline ridge that is slightly asymmetrical. The osteoderms are pitted and grooved on the ventral surfaces, with fewer larger pits on the dorsal surface. Two osteoderms are particularly thick compared to the other osteoderms from the specimen. In one of these osteoderms, both dorso-ventral edges are thickened and rugose, but the other only shows such elaboration on one edge. Much of the morphological range in the osteoderms described by Colbert (1981) is covered by the material from UCMP 130580, but because the skeleton was both fragmentary and disarticulated, it is

ROSHNBAUM & PADIAN-NEW MATERIAL OF SCUTELLOSAURUS 17 Tabic 1. Measurements (in millimeters) and ratios of Scutellosatirus lawleri(after Colbert 1981), Hctcrodontosauvus tucki(after Santa Luca 1980), and Lesotbosaurus diaijiwsticns (after Colbert 1981). c - estimated value; * - calculation based on estimated value. Scutcllosa u rus In wleri H. tucki L. diapnosticus CHARACTER PI.175 I loloiype PI.1752 Paratypc UCMP UCMP 130580 170829 UCMP 175168 UCMP 175166 SAM K1332 thickness of the frontal bone 3.2 width of scapula at neck 11.0-7.9 - length of humerus 68.5-66.0e - 83.5 58 proximal width, humerus 16.6-16.0 21.0 19.0 20.6 distal width, humerus 14.6-13.7 18.1 17.0 17.1 length oi radius 57e - 39.8-58.5 3" proximal width, radius 8.4-7.1-9.7 distal width, radius - - 6.5 12.0 9.4 proximal width, ulna lie - 10.5-15.2 distal width, ulna - - 7.0-7.6 length of femur 93.2 114.0 - - 112.0 104 proximal width, lemur 23.7 32.5 22.7 - distal width, femur 18.8 24.5 18.1 26.9 20.0 length of tibia 95.7 125.5 95.0c - 145.0 129 proximal width, tibia 16.5 25.3 21.7-11.7 distal width, tibia 20.4 27.9 20.0-22.3 width astragalus - - 14.0 18.5 length of ungual phalanx, manus 4.0-5.5c - 18.2 length of manus - - 38e - length of proximal phalanx, pes _ 14.1 length of ungual phalanx, pes 11.4 12.0 20.9 lengt h of pes 106e Ratio, presacral series to caudal series 43 63 Ratio, humcrus/radius 120* 166 143 154 Ratio, femur/tibia 97 77 73 Ratio, forelimb/hindlimb 58 52* 35 53

PALEOBIOS, VOL. 20, NUMBER 1, 2000 Fig. 4. Scntcllosniiriis lawlerilimb girdles, UCMP 130580. A. Left scapula, internal (ventral) view. B. Fragments of the right scapula, internal (ventral) view. C. Left humerus, dorsal view. D. Right humerus, dorsal view. E. Right and left radii, dorsal view. F. Right and left ulnae, dorsal view. G. Sacral rib, dorsal view. H. Anterior process of the right ilium, medial view. I. Left femur, proximal end, posterior view. J. Right tibia, anterior view. The larger portion of the tibia is a cast. K. Proximal end of fibula. L. Left ischium, lateral view. M. Right ischium, lateral view. N. Left femur, distal end, anterior view. O. Fragment of distal end of left tibia, anterior view.

_... ROSF.NBAUM & PADIAN-NEW MATERIAL OF SCUTKLLOSAURUS L9 H K M N 1 cm Q ' :».. R ky^aaa. Fig.5. ScutcUosauruslawkriskull Fragments and appendages, UCMP 130580. A. Left dentary, lateral (external) view. B. Ixft and right Frontals and parietal, dorsal view. The Fragments are aligned as they would be in life, with their sutures in contact. C. Right jugal, external lateral view. D. Right dentary, lateral (external) view. E. Ixft quadrate, posterior view. F. Basioccipital, internal (dorsal) view. G. Right quadrate, posterior view. H. Two bones of the carpus. I. Metacarpal (third or fourth). J. Axis neural arch, lateral view. K. Axis intcrccntrum, dorsal view, L. Odontoid, caudal view. M. IxFt astragalus, articulation with metapodium. N. Ixft (perhaps second) tarsal. O. Left metapodium, dorsal view. Left to right are the proximal end of the first metatarsal, shaft fragment of the third metatarsal, and proximal and distal ends of the fourth metatarsal. P. Left metapodium, dorsal view. Ixft to right are the fourth metatarsal, third metatarsal, and distal end of the first metatarsal. Q. Phalanges, dorsal view. R. Ungual phalanx, lateral view.

20 PALEOBIOS, VOL. 20, NUMBER 1, 2000 B > -- G, 1 cm 'ihiwiiiiii ' H. ;' i - A \x a M N 3' { V -... Fig. 6. Sciitellosniirttslawleriskull fragments, vertebrae, and osteodcrms, UCMP 130580. A. Left dentary, medial (internal) view. R. Left and right frontals and parietal, ventral view. The fragments are positioned as they would be in life, with their sutures aligned. C. Right jugal, internal medial view. D. Right dentary, medial (internal) view. E. I.eft quadrate, anterior (internal) view. F. Basioeeipit.il, external (ventral) view. G. Right quadrate, anterior (internal) view. H. Axis neural arch. The arch is split down the longitudinal axis. I. Axis neurocentrum, ventral view. J. Odontoid, rostral view. K. Astragalus. L. Vertebrae, dorsal view. The vertebrae are arranged left to right, most rostral to most caudal. One cervical, two thoracic, one sacral, and two caudal. M. Scute, lateral view. This scute is from the parasagittal series. N. Osteodcrms, dorsal view. An assortment of osteodcrms, ranging in size and thickness. The most robust are on the left.

ROSENHAUM & PADIAN-NEW MATERIAL OH SCUTKLLOSAURUS 21 Stegosauria Ankylosauria Heterodontosaurus ORNITHOPODA XT ^Scelidosaurus ^v ^ ^ Thyreophoroidea^v Scutellosaurus EUORNITHOPODA Thyreophora \ ^^-"" Lesothosaurus GENASAURIA ORNITHISCHIA Fig. 7. Relationships of the taxa discussed in this paper, after Sercno (1986). Our study confirms his phylogeny, with the single emendation thai the character "medial portion of quadrate condyle much more robust than lateral portion" is found to be a synapomorphy of Thyreophora, not merely Thvrcophoroidca, because it is also found in Scutellosaurus. impossible to reconstruct the scute pattern of the living animal. Colbert postulated a pattern for the osteoderms based on a variety of morphological characters. He concluded that the tall, wedge-shaped parasagittal osteoderms ran in two rows along the sides of the vertebral column, and converged to become a single row along the top of the caudal vertebrae. Size: UCMP 130580 is very slightly smaller than MNA PI. 175. Comparable pieces of MNA PI. 1752, UCMP 170829, UCMP 175168, and UCMP 175166 are all substantially larger than both UCMP 130580 and MNA PI.175 (Table 1). DISCUSSION Phylogeny Although the Thyreophora are a well-established and accepted group, the position of Scutellosaurus within the group has been questioned (Weishampel et al. 1990). When Scutellosaurus is excluded, the group is diagnosed by eight synapomorphies (Sereno 1986). The problem lies in the fact that six of these eight characters are cranial, and skull material from S. lawleri'is relatively poorly known. What fragments exist in UCMP 130580 show a similarity to the skull of Emausaurus (Haubold 1990), particularly the jugal. Several specimens of S. lawkri collected from the Kayenta Formation of northern Arizona bv the Museum of Comparative Zoology (Harvard University) base some cranial material, but they have yet to be described. Thus, it is difficult to test some crucial character state hypotheses relating to the affinity of S. lawkri. When S. lawkri is included in the Thyreophora, the number of synapomorphies drops to three (Sereno 1986; Weishampel et al. 1990): a sub-orbital bar that is transversely broader than tall, a parasagittal sequence of keeled osteoderms, and a lateral sequence of keeled osteoderms. This restricts the other five synapomorphies to Thyreophoroidea (Sereno 1986), a group nested within the Thyreophora (Scelidosaurus, Ankylosauria, and Stegosauria). Two of these, a sinuous curve to the dentary tooth row, and the complete incorporation of a supraorbital into the orbit, are definitely absent in Scutellosaurus. However, one thvrcophoroidcan character appears to be present in Scutellosaurus; the quadrate condyle of Scutellosaurus is more robust medially than laterally, and can be seen in UCMP 130580 and UCMP 175166. The remaining two characters cannot yet be assessed in.v. lawkri, because those portions of the cranium are not preserved in the specimens described to date. The heavy reliance on cranial characters is problematic when assessing remains of basal thyreophorans, because the crania of both stegosaurs and ankylosaurs are so heavily modified (Coombs and Maryanska 1990; Galton 1990). Excluding the osteoderm material,.s'. lawkri generally retains the post-cranial ornithischian morphology characteristic of "fabrosaurids" such as Lesothosaurus, which are considered the sister taxa to the rest of the Ornithischia (Sereno 1986; Weishampel et al. 1990). Scelidosaurus harrisonii, a slightly larger ornithischian from the Early Jurassic of southwestern England, is well accepted as a basal member of Thyreophora (Coombs, et al, 1990). The osteoderms of S. lawkri resemble those of Scelidosaurus described by Owen (1861 1863) and Padian (1987: UCMP 130056). Owen described the osteoderms as "triedral," with the smallest side forming the base and the two larger sides creating a peak. This is consistent with what Colbert classified as "Type IV scutes" in Scutellosaurus. Owen described the surface of the osteoderms as pitted, with vermiform striations, and this also characterizes the osteoderms of.v. lawkri. The osteoderms of Scelidosaurus are somewhat hollowed out vcntrally (Padian 1987). Those of S. lawkri, especially

22 PALKOBIOS, VOL. 20, NUMBER 1, 2000 those that are peaked, also tend to be hollow ventrally, maintaining a "uniform thickness" (Colbert 1981). Colbert's (1981) proposed pattern for the scutes of Scutellosaurus is also similar to the pattern observed in Scelidosaurus (Owen 1863), especially in having a row of keeled scutes along each side of the spine that merges to a single row posterodorsally on the tail. Colbert's pattern was inferred from the morphology of the osteoderms, without apparent reference to Scelidosaurus. Locomotion The length of S. lawlerfs forearm could previously only be estimated. UCMP 130580 shows that previous estimates of forearm length were too generous (sec Table 1 for data). Because UCMP 130580 is only slightly smaller than MNA PL 175, the fact that the radius is substantially shorter than Colbert's estimate requires a second look at the preferred mode of locomotion in S. lawleri. When Colbert's original estimate of the length of the radius is corrected, the forelimb to hindlimb ratio decreases from 0.58 to 0.52. This is a smaller ratio than the original estimate, but not as dramatic as the ratio for L. diagnosticum (.33). Additionally, the forelimb as a whole is slight compared to the hindlimb, and does not have substantial or elaborated structures for muscle attachment, as docs H. tucki (Santa Luca 1980; Weishampel and Witmer 1990). This would seem to support the inference that S. lawleri was not an obligate quadruped and may not have used its forclimbs extensively in locomotion. The extreme length of the tail compared to the precaudal vertebral series also likely aided the animal in balancing on the hind limbs alone (Colbert 1981). However, Lesotbosaurus had more gracile, elongate hindlimbs and reduced forclimbs, suggesting greater speed and agility than in Scutellosaurus. Ontogenetic Stage Colbert questioned whether the holotype specimen was an adult or a juvenile. UCMP 130580 is almost identical in size to the holotype, MNA 175.PI, and both are significantly smaller than the other specimens (Table 1). Neurocentral sutures in the vertebrae are not fused in either specimen except in the caudal series. This is a juvenile feature in many tetrapods, but it is noteworthy that the larger specimens also lack neural sutures, and that L. diagnosticus lacks them in its (ostensibly) adult form (Thulborn 1972). UCMP 130580 also lacks fused cranial sutures, however. This would seem to indicate that the smaller forms are juveniles, rather than smaller adult individuals, or sexual dimorphs. The absence of cranial material in the larger specimens makes it difficult to say with certainty. CONCLUSION Morphologically, Scutellosaurus lawleri is in many respects a good example of a basal, generalized ornithischian dinosaur. It is smaller and lacks the more developed armor that characterizes more derived members of the thyreophoran lineage, including Scelidosaurus And the stegosaurs and ankylosaurs. However, the presence and pattern of dermal osteoderms, along with three features of the skull indicate that Scutellosaurus is united with the rest of the Thyreophora. Particularly interesting are details of the Scutellosaurus skull and locomotor}' apparatus, which are both extensively remodeled in more derived thyreophorans. Preliminary histological analysis of S. lawleri supports the conclusion that it grew relatively slowly compared to most dinosaurs such as the hadrosaurs Maiasaura and Hypacrosaurus, as well as theropods and most sauropods (Horner et al., in press; Ricqles ct al., in press; Padian et al., unpublished data). The presence of some specimens that are around 25 percent larger than others suggests that the animal grew slowly throughout life, perhaps because its adult size was small. Alternatively, there may have been a pronounced sexual dimorphism in Scutellosaurus, but either hypothesis must await the description of more specimens. Just as important will be the discovery of additional basal ornithischian taxa, especially in the Late Triassic, when the initial divergence of ornithischian and saurisehian dinosaurs occurred, and during which interval the record of ornithischians is nearly blank. ACKNOWLEDGMENTS We thank Ellen Lamm (Museum of the Rockies) and Russell Main and Mark Goodwin (UCMP) for histological preparation of the specimens; members of the UCMP field crew including J.M. Clark, D.E. Fastovsky, S.M. Gatesy, and K. CoBabe, for their labors in the Kayenta Formation; F.A. Jenkins Jr. for information about the Kayenta expeditions of Harvard University; Deborah Hill, Collections Manager, Museum of Northern Arizona, for access to specimens; and J.R. Horner and A. de Ricqles for histological advice. We are grateful to the people of the Navajo Nation for permission to collect specimens on their tribal lands, and to the National Geographic Society for funding the field work. The photographs were taken by Jim Hendel of the UC Berkeley Scientific Photograph}' Lab. Additional funding for this research was provided by the UCMP and the UC Berkeley Alumni Club of Stanislaus Count}'. This is UCMP Contribution No. 1720. LITERATURE CITED Benton, M.J. 1990. Origin and interrelationship of dinosaurs, pp. 11-30 in D.B. Weishampel, P. Dodson, and H. Osmolska (eds.). The Dinosauria. University of California Press, Berke ley and Los Angeles, California. Colbert, K.H. 1981. A primitive ornithischian dinosaur from the Kayenta Formation of Northern Arizona. Museum of Northern Arizona Press Bulletin Series 53. 61 pp. Coombs, W.P., Jr., D.B. Weishampel, and I..M. Witmer. 1990. Basal Thyreophora. pp. 427-34 in D.B. Weishampel, P. Dodson, and II. Osmolska (eds.). The Dinosauria. University of California Press, Berkeley and Los Angeles, California. Coombs, W.P., Jr., and T. Maryanska. 1990. Ankylosauria. pp. 456-83 in D.B. Weishampel, P. Dodson, and H. Osmolska

ROSENBA UM & PADIA A/-NEW MATERIAL OE SCUTEI.LOSAURUS 23 (cds.). The Dinosauria. University of California Press, Berkeley and Los Angeles, California. Cooper, M.R. 1985. A revision of the orinithischian dinosaur Knngnasaurus coetzeei Haughton, with a classification of the Ornithischia. Annals of the.south African Museum 95:281-317. Galton, P.M. 1990. Stegosauria. pp. 435-55 in D.B. Weishampel, P. Dodson, and II. Osmolska (eds.). The Dinosauria. University of California Press, Berkeley and Los Angeles, California. Haubold, H. 1990. Ein neuer Dinosaurier (Ornithischia, Thyreophora) aus dem unteren Jura des Nordlichen Mitteleuropa. Revue dc Paleobiology. 29 pp. Norman, D.B. 1984a. A systematic appraisal of the reptile order Ornithischia. pp. 157-162 in W.-E. Reif and F. Westphal (eds.). Third Symposium on Mesozoic Terrestrial Ecosystems: Short Papers. Atlempto Verlag, Tubingen. Norman, D.B. 1984b. On the cranial morphology and evolution of ornithopod dinosaurs, pp. 521-547 in M.W.J. Ferguson (ed.). The Structure, Development and Evolution ol Reptiles. Symposia of the Zoological Society of I.ondon 52. Academic Press, London. Owen, R. 1861. A monograph of the fossil Reptilia ot the I.iassic Formations. Part First. Scelidosaurus barrisonii. Palaeontographical Society, I.ondon, 14 pp. Owen, R. 1863. A monograph of the fossil Reptilia of the I.iassic Formations. Part Second. Scelidosaurus barrisonii. continued. Palaeontographical Society, London, 26 pp. Padian, K. 1987. Presence of the dinosaur Scelidosaurus indicates Jurassic age for the Kayenta Formation (Glen Canyon Group, northern Arizona). Geology 17:438-41. Padian, K., J.R. Homer, and A. de Ricqlcs. 1999. Dinosaurian growth rates and the evolution of life history strategies. Journal of Vertebrate Paleontology 19 (3, suppl.). Santa Luca, A.P. 1980. The post-cranial skeleton of Hctcrodontosaurus tucki (Reptilia, Ornithischia) from the Stromberg of South Africa. Annals of the South African Museum 79(7): 159-211. Sereno, P.C. 1986. Phylogeny of the bird-hipped dinosaurs (Order Ornithischia). National Geographic Research 2:234-56. Sues, H.D., and D.B. Norman. 1990. Hypsilophodontidae, Tcuontosaurus, Dryosauridae. pp. 498 509 in D.B. Weishampel, P. Dodson, and H. Osmolska (eds.). The Dinosauria. University of California Press, Berkeley and Los Angeles, California. Thulborn, R.A. 1972. The post-cranial skeleton of the Triassic ornithischian dinosaur Fabrosaurus australis. Paleontology. 15:29-60. Weishampel, D.B., and I..M. Witmer. 1990. Ileterodontosauridac. pp. 486-97 in D.B. Weishampel, P. Dodson, and II. Osmolska (eds.). The Dinosauria. University of California Press, Berkeley and Los Angeles, California.

24 PALEOBIOS, VOL. 20, NUMBER 1, 2000 Guidelines for Contributors http://www.ucmp.berkeley.edu/museum/pbios/pbguide.html Manuscript Submission PaleoBios publishes submissions written in English. Manuscripts must be typed, one side only, on 8-1/2x11" paper, and doublespaced throughout, including references, figure legends, and tables. All margins should be at least 25mm or 1 inch wide. Text should not be right justified nor should text be hyphenated at the right margin. Three unmarked, clean copies of" the manuscript, including tables and figures, should be submitted. The first mention in the text of the scientific name of a taxon should be accompanied by the taxonomic authority, including the year. Underline scientific names and other words to be printed in italics. Metric and celsius units are preferred. In most cases, major headings of" the manuscript should be as follows: title page, abstract, introduction including an explanation of the abbreviations used, materials and methods, results, discussion, acknowledgments, cited literature, figure legends, tables, and figures. Put page numbers on every page of the manuscript (excluding figure legends, figures, rabies and title page). The title page should be a separate sheet and should include the title, name(s) of author(s), and addresses of author(s). The abstract should be 250 words or less and should not include references. Literature cited Literature citation in the body oft he manuscript should include the am hor and date. Use the format as follows for one author: (Phillips, 1981). Use the format as follows for two authors: (Phillips and Smith, 1982). Use the following format for three or more authors (Phillips et al., 1983). When author and date are given only as authority for a taxonomic name, references need not be cited for names above the generic level. The "literature cited" section should include all references cited in the text, and only references cited in the text, listed in alphabetical, then chronological, order by author. Each citation must be complete, with all journal titles unabbreviated, and in the following forms. Hanging indents must be five spaces (not a tab). Periodicals: Hell, ('.. J. 1993. Fossil lizards from the Elsinore fault /.one. Riverside County, California. PaleoBios 15(2):17-26. Books: Simpson, G. G. 1951. Horses; the Story of the Horse Family in the Modern World and Through Sixty Million Years of History. Oxford University Press, New York. 247 pp. Edited Volumes: Rose, K. D. and T. M. Bown. 1993. Species concepts and species recognition in Eocene primates, pp. 299 330 ;;; W. H. Kimbel and L. B. Martin (eds.). Species, Species Concepts, and Primate Evolution. Plenum Press, New York. Tables, figures and plates Each table must be typed on a separate sheet. Tables should beheaded by a brief caption. Figures must be submitted ready for publication. Captions should be listed on a sheet following the literature cited. Text figures should be in black ink and completely lettered. Figures should be designed with page format and column size of the final page in mind (single column width: 3.4 inches/8.7 cm/20.5 picas; double column width: 7 inches/17.8 cm/42 picas; column height: 9.2 inches/23.3 cm/55 picas). Where appropriate, a scale bar must be used in the figure. Ideally, photographs should be submitted as digital images in TIFF image format on disk along with three hard copy printouts. Black and white illustrations should be scanned as black and white (not grayscale) images. I ligh quality Xerox copies of figures or photographs are only suitable for reviewers' copies of submitted manuscripts. It is the author's responsibility to ensure that lettering in the figure will be legible after any necessary reduction. Sans-serif fonts are preferred for figure lettering. All figures, tables and plates should be numbered. Use one consecutive set of Arabic numbers for all illustrations (do not separate "plates" from "text figures"). Processing of manuscripts PaleoBios is a peer reviewed journal. The editor may reject manuscripts without review in cases of inappropriate subject matter, inappropriate presentation, or failure to adhere to format guide lines. Manuscripts are evaluated by at least two reviewers. Based on reviewer evaluations the editor makes a conditional decision to accept or reject. The editor's decision and the reviewers' comments are sent to the author for consideration and further action. A final, revised manuscript should be returned to the editor with review copies. Unless requested, only one copy of the final manuscript needs to be returned to the editor. Publication Costs PaleoBios accepts manuscripts of 50 or fewer double-spaced pages in length. Linger manuscripts will be considered, but the author may be charged page costs for pages over the specified length. The cost of abstract volumes will be borne by the organization or host institution of the meeting, not by PaleoBios. Electronic Submission Manuscripts must be submitted for review as hard copy. Final revisions of the manuscript may be submitted on disk or uploaded via Anonymous FTP in either Microsoft Word for Windows or Word for Macintosh format. If uploading a manuscript electronically, FTP to ucmplhcrkeley.edu and log in as Anonymous. Give your email address as the password. Access is then granted. Go to the directory /pub/paleobios/submission and put the tile in that directory. Please then send an email to PaleoBios (pbioscrtucmpl.berkeley.edu) notifying the editor of the upload. Submitting manuscripts Send all manuscripts and contributions toward publication costs to: University of California Berkeley Editor PaleoBios Museum of Paleontology 1101 VLSB#4780 Berkeley, CA 94720 4780 USA Correspondence on editorial matters may be sent to the above address or via e-mail: pbiosfa ucmpl.berkeley.edu.