A large collection of well-preserved theropod dinosaur swim tracks from the Lower Jurassic Moenave Formation, St. George, Utah

Full text

315
Harris et al., eds., 2006, The Triassic-Jurassic Terrestrial Transition. New Mexico Museum of Natural History and Science Bulletin 37.
A LARGE COLLECTION OF WELL-PRESERVED THEROPOD DINOSAUR SWIM TRACKS
FROM THE LOWER JURASSIC MOENAVE FORMATION, ST. GEORGE, UTAH
ANDREW R.C. MILNER1, MARTIN G. LOCKLEY2AND JAMES I. KIRKLAND3
1St. George Dinosaur Discovery Site at Johnson Farm, City of St. George, 2180 East Riverside Dr., St. George, UT 84790, E-mail: amilner@sgcity.org;
2Dinosaur Tracks Museum, University of Colorado at Denver, P.O. Box 173364, Denver, CO 80217, E-mail: mlockley@carbon.cudenver.edu;
3Utah Geological Survey, 1594 West North Temple, Suite 3110, P.O. Box 146100, Salt Lake City, UT 84114-6100, E-mail: jameskirkland@utah.gov
Abstract—A large and exceptionally well-preserved collection of dinosaur swim tracks, attributable to the
ichnogenus Characichnos, is preserved as natural casts and reported in detail for the first time from the St. George
Dinosaur Discovery Site at Johnson Farm (SGDS) in southwestern Utah. Approximately 3200 individual claw
marks, typically in sets of three and rarely singular or paired, were made predominantly by small theropod
dinosaurs in a size range consistent with the ichnogenus Grallator. About 95% of all dinosaur footprints on
multiple track-bearing horizons at the SGDS are Grallator ichnites. Larger and less abundant Eubrontes-type
swim tracks are associated with Grallator-type swim tracks at the SGDS. “Typical” Eubrontes and Grallator
footprints occasionally occur among the swim tracks, providing further support for referring to swim tracks as
“Eubrontes-type” and “Grallator-type.” Abundant invertebrate grazing traces and burrows indicate organic-rich,
well-oxygenated sediments within the upper 1-2 cm of mudstone directly below the infilling sandstone unit of the
SGDS “Main Track Layer” in which the swim tracks are preserved. Well-preserved sedimentary structures
associated with a marginal lacustrine shoreline paleoenvironment suggest multiple animals swimming and/or
floundering, mostly in a southerly direction, against a north-flowing current that paralleled the paleo-shoreline.
Simultaneous formation of swim tracks on a clay-rich substrate, together with rapid burial of the traces, has
resulted in exceptional preservation of skin impressions, scale scratch lines, and possible fine details on the cuticle
of claw tips.
INTRODUCTION
Tracks made by swimming dinosaurs and other vertebrates, such
as crocodylomorphs and turtles, have become popularly known among
paleontologists as “swim tracks.” They provide a conjecture-inspiring
but controversial topic for investigation. They are of interest because
they provide insight into the behavior of ancient vertebrates in aquatic
environments, but they are controversial because they display irregular
morphologies that are very often difficult to interpret. Unlike a track
made by an animal walking on firm ground that supports most or all of its
weight, a swimming animal may touch the subaqueous substrate while
fully or partially buoyant. As a result, swim tracks rarely show regular
step and stride patterns. Instead, they are often incomplete, occurring in
irregular and confusing configurations. This incompleteness often makes
it difficult to identify the track maker or to distinguish between manus
and pes prints if the track maker was quadrupedal.
In this paper, we present a preliminary description of one of the
largest and best-preserved true swim track assemblages ever recorded.
They are from the Lower Jurassic (earliest Hettangian) Moenave Forma-
tion at the St. George Dinosaur Discovery Site at Johnson Farm (SGDS)
in southwestern Utah (Fig. 1). We also demonstrate that these ichnites
were unequivocally made by theropod dinosaurs of the type that made
typical terrestrial locomotory tracks nearby (i.e., Grallator and
Eubrontes). In places, it is even possible to demonstrate that there are
gradations between walking tracks and slide or swim tracks. We briefly
compare these swim tracks with examples from other localities.
VERTEBRATE SWIM TRACKS: A HISTORICAL PERSPECTIVE
In western North America, swim tracks have been reported from
a number of Mesozoic deposits. These include Triassic tracks attributed
to amphibians and phytosaurs (Boyd and Loope, 1984; Lockley et al.,
2005, Lockley and Milner, this volume), and Cretaceous tracks, first
assigned to ornithischian dinosaurs (McAllister, 1989a, b) and ptero-
saurs (Gillette and Thomas, 1989), but later identified as crocodylomorph
tracks (Bennett, 1992; Lockley and Hunt, 1995). Likewise, Late Jurassic
tracks in France attributed to hopping dinosaurs (Bernier et al., 1982,
1984) were later reinterpreted as the swim tracks of turtles (Thulborn,
1989, 1990). These examples serve to indicate the potential for confu-
sion in correctly attributing swim tracks to the actual track makers.
Focusing only on purported dinosaurian swim tracks, there has
been lively debate about so-called “swim tracks” of sauropods, theropods,
and ornithopods. In all cases, such debate has had significant ramifica-
tions for our views into the behavior of the dinosaurs in question. Bird
(1944) claimed to have found the manus-dominated trackway of a swim-
ming sauropod in the Early Cretaceous of Bandera County, Texas. Fol-
lowing his original example, others (Coombs, 1975; Bird, 1985; Ishigaki,
1989; Thulborn, 1990; Czerkas and Czerkas, 1990; Norman, 1985; Lee
and Huh, 2002) then interpreted these and other manus-dominated sau-
ropod trackways as indicative of swimming ability in sauropod dino-
saurs. This generated a “sauropod swim tracks paradigm.” Thus, incom-
plete manus-only or manus-dominated trackways were used to support
the notion of aquatic/swimming sauropods. This paradigm contrasts
with the idea that the trackways were, in fact, undertracks of walking
sauropods (Lockley and Rice, 1990; Lockley 1991; Lockley et al., 1994;
Lockley and Meyer, 2000) that were predominantly terrestrial, rather
than aquatic, animals.
Purported ornithopod swim tracks from the upper Hettangian
Przysucha Formation in the Holy Cross Mountains of Poland (Gierliñski
and Potemska, 1987; Pieñkowski and Gierliñski, 1987) are, like those of
the sauropods discussed above, probably undertracks (Lockley, 1991).
In contrast, theropods have traditionally been viewed as hydro-
phobic, and many older texts indicate that sauropods escaped from
theropods simply by entering the water (e.g., Colbert, 1945, p. 73;
Andrews, 1953, p. 58-59; Hotton, 1963, p. 95; Colbert, 1965, p. 95,
133; Paul, 1988, p. 44-47). This speculative scenario was called into
question by Coombs (1980) while describing purported theropod swim
tracks from the Lower Jurassic East Berlin Formation at Dinosaur State
Park, Connecticut. According to Coombs (1980), the theropod track
maker was probably the same kind of animal responsible for making
abundant Eubrontes tracks at the same locality. This example is highly

316
relevant to the SGDS site at which both walking Eubrontes tracks and
somewhat similar swim tracks coincide in Lower Jurassic deposits of the
Moenave Formation (Kirkland et al., 2002; Milner et al., 2004, 2005a, b,
this volume; Milner and Lockley, 2006). This will be discussed in more
detail below.
Contrasting with the tendency of ichnologists to name and clas-
sify “typical” fossil tracks, few supposed swim tracks had been named
until recently. Exceptions include the purported hopping dinosaur tracks
from France (named Saltosauropus, meaning “hopping saurian track”
[Bernier, 1984]) that are now considered turtle swim tracks (Thulborn,
1989), and the recently named Characichnos (meaning “scratch mark”)
from the Middle Jurassic Saltwick Formation of England (Whyte and
Romano, 2001). Characichnos has since been identified in the Lower
Jurassic Zagaje Formation of Poland (Gierliñski et al., 2004). Additional,
albeit unnamed, theropod swim tracks are reported from the Lower
Cretaceous of Spain (Ezquerra et al., 2004) and from the Whitmore Point
Member of the Moenave Formation and the Kayenta Formation in Zion
National Park, Utah (DeBlieux et al., 2003, 2005, this volume).
Proposed dinosaur swim tracks were also reported by Kvale et al.
(2001a, b) from the Middle Jurassic Gypsum Springs Formation of
Wyoming. They only referred to these ichnites as dinosaur swim tracks
and made no determination as to whether they pertain to a theropod or
ornithopod. Uncertainty by the authors (Kvale et al., 2001a) was also
expressed as to whether these proposed swim tracks are dinosaurian or
were made by a “crocodylian.”
All of these examples of purported or definitive dinosaur swim
tracks are based on relatively small sample sizes. The SGDS has, in
contrast, yielded more than 3200 traces, most of which are very well
preserved and easily correlated with the foot morphology of abundant
Grallator and associated Eubrontes ichnites.
MATERIALS AND METHODS
Soon after the time of the original discovery (May 4, 2001 by
ARCM), approximately 80 talus blocks containing about 250 swim
tracks were salvaged from the spoil piles excavated from Bodega Bay
Development Corporation (BBDC; called the Darcy Stewart [DS] sites
in Milner et al., this volume) and Washington County School District
(WCSD) properties. Subsequently, in January, 2003, during excavations
on WCSD property for the construction of Fossil Ridge Intermediate
School, an approximate 14 x 12 m (= 168 m2) area was excavated and
partially mapped. Designated as “Dinosaur Swim Track Quarry #1”
(DSQ1), the area produced about 140 large sandstone blocks ranging in
size from 0.2 x 0.1 x 0.1 m to 3 x 1.1 x 0.6 m (weighing from 0.02-11
metric tons). As a result, an estimated 2500 swim track claw marks were
identified and collected. The following year, in February, 2004, during
further excavations of BBDC property, “Dinosaur Swim Track Quarry
#2” (DSQ2) was opened immediately adjacent to, and as a continuation
of, DSQ1. DSQ2 resulted in the collection of an additional 23 in situ
blocks (most of very large size) that were carefully mapped and ex-
tracted from an area measuring about 16 x 5 m (= 80 m2). This led to the
identification of approximately 450 additional swim tracks. During these
excavations, data collection, although constrained by tight time schedul-
ing, was facilitated by the cooperation of many partners, leading to the
success of the day-to-day exploration, salvage, and transportation of
excavated material during the active development of these properties
(see Acknowledgments below).
Preparation of the swim track blocks is complicated, and has
never before been attempted on such a large scale. In general, the swim
tracks represent natural casts of narrow striations with high relief, some
of which terminate in well-preserved claw and phalangeal pad impres-
sions. When initially excavated (overturned), the undersurfaces were
embedded in a finer grained material that was typically 1-2 cm thick.
This material weathered rapidly on exposure, so volunteers were orga-
nized to clean surfaces carefully and restore and harden the delicate casts
using cyanoacrylate and Acryloid B-72 in acetone solution. Subsequently,
representative material was traced, photographed, and replicated. Prepara-
tion of these blocks is ongoing.
The most comprehensive documentation is being accomplished
by photographing all the surfaces with an 8.0 megapixel Nikon Coolpix
8700 digital camera, and then piecing together a photo-mosaic of the
reconstructed surface. This project also is ongoing, and will be published
elsewhere. This method enabled the development of a coherent under-
standing of the orientation of the swim traces across a large area. How-
ever, not all swim track blocks could be restored to their original orienta-
tions because many had been moved by private excavation activities
prior to the initiation of a systematic scientific method.
GENERALIZED STRATIGRAPHY, SEDIMENTOLOGY AND
PALEOENVIRONMENTAL INTERPRETATION
At the SGDS, the Moenave Formation is about 74 m thick. It is
divided into the lower Dinosaur Canyon Member (about 56.5 m thick)
and the upper Whitmore Point Member (about 17.5 m thick) (Fig. 2;
Kirkland and Milner, 2005, this volume). The Moenave unconformably
FIGURE 1. Index maps showing location of the St. George Dinosaur Discovery
Site at Johnson Farm in southwestern Utah.

317
overlies the Upper Triassic Owl Rock Member of the Chinle Formation
(per local usage), and, in turn, is unconformably overlain by the Springdale
Sandstone Member of the Lower Jurassic Kayenta Formation (Lucas
and Tanner, this volume). Toward the west in Nevada, this unconformity
is angular at a regional scale with the Whitmore Point Member eroded
away such that the Springdale Sandstone Member directly overlies the
Dinosaur Canyon Member (Marzolf, 1993, 1994). The Triassic-Juras-
sic boundary is located somewhere within the Dinosaur Canyon Mem-
ber (Kirkland and Milner, this volume). Based on some of the track
types (Eubrontes,Batrachopus,and Anomoepus) and body fossils, the
age of the upper portion of the Moenave is interpreted as Early Jurassic
(earliest Hettangian).
This study focuses on ichnites from the lower 5 m of the Whitmore
Point Member, particularly from the base of the main track-bearing
sandstone called the “Main Track Layer” (MTL; Figs. 2-3). Tracks have
been identified at 25 stratigraphic levels in the immediate vicinity of the
SGDS (Fig. 2), and several of these layers have been mapped in situ
(Milner et al., this volume; Williams et al., this volume).
The first track-bearing horizon discovered was the MTL at the
SGDS, revealing tracks and associated mudcracks preserved as robust
sandstone natural casts at the base of a thick (30-70 cm), well-sorted,
fine-grained sandstone bed, about 53 m above the Owl Rock-Moenave
contact (Figs. 2-3). Track casts and associated sedimentary structures at
the SGDS museum locality (the original discovery site), and nearby
several localities to the northwest, represent features that were made in
subaerially exposed conditions (Fig. 4). The berms and swales of the
“Top Surface” tracksite (Fig. 2), situated at the top of the main track-
bearing sandstone bed at the original discovery site east of Riverside
FIGURE 2. Stratigraphic section showing horizons mentioned in this paper. Abbreviations: MTL, “Main Track Layer” at base of main track-bearing
sandstone; TS, “Top Surface” of main track-bearing sandstone; DS-W, “Stewart-Walker Tracksites.”

318
FIGURE 3. Map and paleoenvironmental interpretation of the SGDS MTL.
A, Transect from museum site (A) on SGDS property (in gray) to swim
track quarries (B) on WCSD and former BBDC properties. The estimated
position of the paleo-shoreline is indicated (hatched line) for the MTL
based on orientation and preservational types from tracks, invertebrate
traces and sedimentary structures. B, Cross-section of transect A-B in A
showing variation in bed thicknesses of the MTL and Top Surface. The
estimated shoreline is for the MTL only. C, Map of the swim track quarries
showing position of trough and transect A’-B’ in B.D, Cross-section showing
trough containing abundant swim tracks on the MTL surface, the variability
in thickness of the MTL sandstone, and the Top Surface topography.
Drive, are oriented perpendicular to the shoreline at the SGDS, and
represent reworking of the sand sheet by shoreline erosion and
redeposition (see Kirkland and Milner, this volume, for further discus-
sion). This is supported by cross-cutting of pre-existing bedding planes
and track horizons.
Important relationships between local paleogeography, trackway
orientations, and sedimentary structures at the SGDS are evident, lead-
ing to important paleoenvironmental interpretations of the Moenave
Formation in this region. It is possible to trace and partially map the
MTL from the SGDS museum site (“onshore” location; Figs. 3-4) to-
ward the northwest and over to DSQ1 and DSQ2, where abundant dino-
saur swim tracks (Characichnos), subaqueous invertebrate traces (cf.
Palaeophycus), and sedimentary structures represent an “offshore” lo-
cation along the same MTL bedding surface (Figs. 3, 5). The dinosaur
swim tracks are actually concentrated in a channel-like depression filled
by the main track-bearing sandstone that paralleled the paleo-shoreline
and situated approximately 80 m to the northwest (Fig. 3).
This same pattern can be seen in sedimentary structures pre-
served on the MTL across the entire area, evidence of being formed
onshore and offshore. Sedimentary structures, such as mudcracks (Fig.
4A), sulfate salt crystal casts (Fig. 4B), and rain drop impressions (Fig.
4C) can typically be formed only when the sediment is subaerially ex-
posed. Offshore, near-shore, and shoreline aqueous-subaqueous sedi-
mentary structures include a variety of current and symmetrical ripples
(Figs. 5A-B), tool marks (Fig. 5C), flute casts, and scratch circles (Fig.
5D; Metz, 1991, 1999; Droser et al., 2002, fig. 2A; Rygel et al., 2004, fig.
9). In the offshore, channel-like feature mentioned above, current flow
was from the north and paralleled the shoreline, based primarily on
scratch circles, tool marks, flute casts, and other sedimentary structures
(Fig. 5).
Dinosaur swim tracks at the SGDS are interpreted as being pre-
served along the western margin of freshwater Lake Dixie (Kirkland et
al., 2002; Milner and Lockley, 2006). Natural casts of dinosaur foot-
prints and swim tracks formed from infilling by fine-grained, well-sorted
sand. The main track-bearing sandstone covers a 15 cm-thick horizon of
purplish-gray, silty shale, and mudstone-claystone. Abundant inverte-
brate grazing traces and burrows indicate organic-rich, probably well-
oxygenated sediments within the upper 1-2 cm of mudstone directly
below the MTL.
The MTL in the area west of Riverside Drive (Figs. 1, 3) that
preserves the natural casts of the swim traces was laid down predomi-
nantly by unidirectional currents (toward the north, as mentioned above).
Thinner bedding planes near the base of the MTL at the SGDS are
separated by clay-rich mudstone drapes that may represent fluctuations
in sedimentation rate prior to a possible “main depositional event” influx
of sediment that initially buried the track-bearing surface. Lateral varia-
tion in thickness (in some cases completely pinching out) of the main
track-bearing sandstone reflects its subsequent erosion and that the sedi-
ment was deposited with local thickness variation due to underlying
topography. Fluvial depositional environments are typically associated
with asymmetrical ripples produced by unidirectional currents. How-
ever, sedimentary features of the main track-bearing sandstone – thin,
laterally extensive bed geometry and climbing-ripple cross-bedding – do
not support a fluvial channel origin, but instead indicates deposition in
an offshore lacustrine setting, perhaps by longshore currents (Kirkland
and Milner, this volume).
Areas located northwest of the paleo-shoreline have an MTL that
is bioturbated with abundant invertebrate feeding-grazing trails and bur-
rows (Lucas et al., 2005, this volume; Fig. 3). The main track-bearing
sandstone in this area ranges in thickness from 10-20 cm from the
paleoshoreline to where it thickens in the channel-like trough described
above (Fig. 3). The MTL below this thinner main track-bearing sand-
stone bed has abundant scours, flute casts (with current orientations
toward the north), and rare, scoured-out dinosaur tracks (Fig. 5F) and
scoured mudcracks. This surface likely indicates a shallow, near-shore
environment. Vertebrate tracks and traces crossing the mudflat and beach
indicate that the track makers possibly ventured out into shallow waters
of the lake, and that they could potentially leave footprints in a substrate
showing varying states of cohesion and submergence. Mudcracks and
some of the poorly preserved dinosaur tracks would have formed during
a lake level regression and may represent an extension of the same surface
(MTL) preserved to the east of Riverside Drive. Scouring of these
mudcracks and tracks, and the formation of flute casts, probably oc-
curred contemporaneously with the initial deposition of the basal por-
tion of the MTL sandstone in this area. This contemporaneous deposi-
tion may have been responsible for the simultaneous and rapid burial of
swim tracks, other ichnites, and sedimentary structures as the sand swept
across the exposed, underlying fine sediments. Changes in substrate and
environment would likely give rise to a wide range of preservational track
types (Whyte and Romano, 2001).
DESCRIPTION OF SGDS SWIM TRACKS
AND ASSOCIATED TRACES
Swim Track Morphotypes and Swimming Orientations
The SGDS reveals a variety of track types. These include typical

319
walking to running theropod tracks of Grallator and less common
Eubrontes (Fig. 4D), slightly elongate tracks resembling scratch marks
(animals that may have purposely scratched at the substrate in search of
food, etc., in the opinion of ARCM and JIK) or slide marks (in the
opinion of MGL) (Fig. 4E), and highly elongate tracks that we refer to
herein as swim tracks (Fig. 5E). It should be kept in mind that the vast
majority (~95%) of tracks at the site are those of Grallator, formed
subaerially while walking, with a few Eubrontes tracks in which the
typical foot morphology is clearly seen (Milner et al., 2005a). This leads
to the inference that the scratch/slide marks and swim tracks represent a
different mode of preservation of traces made by the same track makers.
The scratch/slide marks (Fig. 4E) are characterized by three paral-
lel traces that extend back from a more or less recognizable, tridactyl
theropod track. Such traces are relatively uncommon and typically occur
in isolation on surfaces where normal walking traces predominate. These
scratch/slide marks have only been found on the SGDS “Top Surface”
FIGURE 4. Sedimentary structures and ichnites that indicate an “onshore” paleoenvironment for the SGDS MTL. A, Mudcracks (SGDS.10). B, Salt crystal
casts (SGDS.40). C, Rain drop impressions with Pagiophyllum branch (SGDS.491). D, A left Eubrontes track showing the asymmetry of digit III with the
wider part of the digit situated laterally and the narrower part medially (SGDS.9). E,Grallator scrape or sliding track showing sediment mounded up caudally
(black arrow) (SGDS.74).

320
FIGURE 5. Ichnites and sedimentary structures that indicate paleoenvironments on the SGDS MTL and Top Surface. A, Current ripples formed prior to
Eubrontes and Grallator trackways (SGDS.18). B, Symmetrical wave-form ripples indicate onshore or beach environments (SGDS.630). C, Tool marks
(SGDS.621). D, Subaqueous scratch circles (specimen number pending). E, Original swim tracks (Grallator-type) discovered on May 4, 2001 by ARCM
(SGDS.47). F, Scoured cf. Grallator track (Field # DS.128).

321
tracksite (Figs. 2-3). In contrast, swim tracks generally consist of more
elongate, parallel to subparallel “scrape marks” (as the name Characichnos
implies) that occur in high densities, almost invariably without associated
walking traces.
Generally, a mudstone compaction ratio is about 3:1, but it may
be less in continental sediments depending on many factors (Nadon,
1997). The majority of claw marks and swim tracks were likely close to
perpendicular to bedding when they were made. Most of the claw marks
and thin swim tracks have been laterally distorted and vertically com-
pressed in that they are now preserved at an average angle of about 45º or
greater due to compaction.
At least three categories of swim tracks at the SGDS appear to be
present, all preserved as natural casts in high density assemblages (Fig.
6A). These include:
(1) Inferred down-current traces that:
- have variable claw impression depths (typically 5-7 cm)
- always have the deepest and longest trace made by the
mesaxonic digit (presumably digit III; Fig. 6B)
- are accompanied by Grallator tracks that resemble normal
walking footprints (Fig. 6C);
(2) Traces oriented up-current that:
- consist of ubiquitous, parallel scrape marks (Characichnos)
- vary considerably in overall length (ranging in size from 5-40
cm long and 5-8 cm wide)
- are usually in sets of three, although occasionally possess only
two or one claw marks (digit III is assumed to always be
present)
- digit III marks are longer and deeper, while digits II and IV are
shorter and shallower than digit III (Figs. 5E, 6A, D-E); and
(3) Cross-current swim tracks that:
- are usually oriented more in an up-current than down-current
direction
- are usually situated at an angle of approximately 45° to current
flow direction
- have similar lengths and widths as the down-current swim
tracks (Fig. 6F).
It is important to note that down-, up- and cross-current traces are all
found on the same surfaces in the channel-like trough at the base of the
main track-bearing sandstone described above (Fig. 3). Down- and cross-
current traces (Grallator) are uncommon relative to the up-current swim-
ming traces.
All observations were made independently of those of Whyte and
Romano (2001), who had made similar observations about Middle Juras-
sic, British swim tracks. Whyte and Romano (2001) concluded that the
two styles of preservation they observed actually represented two dif-
ferent track-making episodes that related to changing water levels through
time. This raises an important taxonomic consideration: are Characichnos
traces, therefore, merely extramorphological variants of Grallator and
other tridactyl theropod track types made during identical behaviors as
produce “typical” Grallator tracks, or are they distinct “behavioral vari-
ants?” This matter is revisited below.
Detailed Preservation of Swim Tracks
Preserved claw tips on the SGDS swim tracks project backward,
indicating that the track-making animals were swimming in the opposite
direction to claw tip orientation (i.e., caudally-projected claw tips: Figs.
6C, E, G-H). Some claw tips at the SGDS are so well preserved that
details of the in vivo claw tips can be recognized (Figs. 6G-H). Morpho-
logical details of the living animal can be identified, such as smooth areas
cut by the claw cuticle and the boundary between cuticle and scaly skin,
with the fleshy part of the toe prominently raised above the cuticle by as
much as 0.8 cm. Further delicate details of the skin include very common
scale scratch lines (Figs. 6G-H) and rare skin impressions. Scale scratch
lines on the sides of swim tracks range in width from 1-1.5 mm.
Since claw tips of the well-preserved SGDS swim tracks project
backward, it is also possible to determine whether a swim track was
made by a left or right foot, both for Grallator and Eubrontes. The SGDS
swim tracks are so well-preserved that, on many specimens, impres-
sions made by the distal phalangeal pads can be clearly distinguished
from the claw marks themselves, making this collection unique. The
majority of SGDS swim tracks preserve marks made only by the claws
and show no evidence of the distal phalangeal pads. Digit III is asym-
metrical in each ichnotaxon when formed subaerially, with the more
prominent and rounded portions of the phalangeal pads on the lateral
side of the toe and a straighter medial side of the digit; in the swim tracks,
the claw is also positioned more medially than laterally on each impres-
sion of digit III (Fig. 4D). Therefore, when looking at the SGDS theropod
swim tracks, the medial sides of both the distal phalangeal pad and the
claw are narrow, and the lateral sides are thicker.
Comparison of Swim Tracks with
Known SGDS Track Ichnotaxa
Grallator-type Swim Tracks
Due to the large quantity of smaller dinosaur swim tracks at the
SGDS localities, and because 95% of all dinosaur footprints at the SGDS
are those of Grallator, we interpret that most of the smaller swim tracks
are “Grallator-type.” The widths between digits II-IV of “typical”
Grallator tracks at the SGDS range from 8-14 cm, whereas in swim track
sets, widths range from 5-8 cm (Figs. 4E, 5E, 6C). In theory, appression
between digits II-IV while swimming is probably due to the animals
purposely forming their appendage into more of an oar or paddle-like
structure to better push against the water in order to create better for-
ward propulsion.
The widths of the bases of the claws of Grallator-type swim
tracks range from 0.5-1.0 cm, and the distal phalangeal pads of the toes
are wider, with a range of 3.0-3.5 cm, giving a ratio of 1:3. Subaerially-
formed Grallator tracks at the SGDS (Figs. 4E, 5F, 6C) have the same
1:3 ratio. All of these measurements were taken from Grallator tracks
formed subaerially at the SGDS and from Grallator-type swim tracks
from the MTL sites.
Scale scratch line (Figs. 6B, F-G) widths on Grallator-type swim
tracks have a width range of 1.0-1.5 mm. Unfortunately, no skin impres-
sions from the distal phalangeal pads are preserved on any of the SGDS
subaerially-formed tracks recovered thus far, although scale scratch lines
have been identified on SGDS Grallator footprints on this part of the
toe. Additionally, scales with a variety of diameters occur on different
parts of the foot both on Grallator and Eubrontes tracks, with Eubrontes
scales only slightly larger when comparing similar portions of the foot.
Likewise, scale scratch line widths on Eubrontes tracks and Eubrontes-
type swim tracks are also in the same size range as those seen in Grallator
tracks and Grallator-type swim tracks.
Eubrontes-type Swim Tracks
Subaerially-formed Eubrontes tracks (Fig. 4D) are much less com-
mon than Grallator (Milner et al., this volume). The pattern described
for Grallator-type swim tracks is also seen with purported Eubrontes-
type swim tracks from DSQ1 and DSQ2, including asymmetry of digit
III, scale scratch lines, and occasional preservation of distal phalangeal
pads. Aside from subaerially formed Eubrontes tracks, two other large
forms of theropod track types must be considered at the SGDS site:
Gigandipus and cf. Kayentapus (see Milner et al., this volume). The
widths between digits II-IV of “typical” Eubrontes tracks at the SGDS
range from 18-28 cm, but in the only measurable swim track set (Fig.
7A), the width is about 30 cm. A narrower width is expected. However,
the swim track in question may have been produced by a larger indi-
vidual, or the digits could have been splayed farther apart by the way
they impacted the substrate. As mentioned above, like the aforemen-
tioned Grallator footprints and Grallator-type swim tracks, digit III is
asymmetrical in Eubrontes tracks and swim tracks. This asymmetry is

322
FIGURE 6. Grallator-type swim tracks from the SGDS MTL. A, Block showing high density of parallel swim tracks (Field # SW.29). B, Down-current swim
track set (Field # SW.104). C, “Normal” Grallator track in a down-current orientation (Field # SW.90). D, Elongate, up-current swim tracks (SGDS.167-
4). E, Shorter, up-current swim tracks (Field # SW.103). F, Swim track set cross-cutting current in a more up-current orientation (Field # SW.77). G,
Grallator-type swim track details showing claw mark, claw tip and scale scratch lines (SGDS.361). H, Close-up of claw tip, from G showing possible cuticle
details (SGDS.361). Abbreviations: CFD, current flow direction; CT, claw tip; DT, direction of travel; PP, distal phalangeal pad; SD, swim direction; SSL,
scale scratch lines. Scales in cm.

323
FIGURE 7. Possible Eubrontes-type swim tracks. A, Large, possible Eubrontes-type swim tracks (white arrows) associated with abundant Grallator-type
swim tracks (black arrows) (SGDS.805). B-E,Eubrontes-type swim track toe and claw mark (SGDS.507). B, Distal view, showing claw and asymmetry of
digit. Scale bar = 2 cm. C, Distal view of toe and claw showing asymmetrical shape of digit and claw. Scale bar = 2 cm. D, Medial view of digit showing claw
mark and clear scale scratch lines. E, Lateral view of digit showing claw mark and scale scratch lines. Abbreviations: CM, claw mark; DPP, distal phalangeal
pad; LPP, lateral side of distal phalangeal pad; MPP, medial side of distal phalangeal pad.

324
clearly shown on specimen SGDS.507 (Figs. 7B-C), which represents an
isolated example of an undetermined and large theropod digit. Scale scratch
lines are prominent on the sides of these specimens (Figs. 7D-E).
The claw base widths of Eubrontes-type swim tracks range from
0.7-1.3 cm, and the distal phalangeal pad of the toes range from 5.3-5.5
cm (Figs. 7B-C). Because of the rarity of interpreted Eubrontes-type
swim tracks, it is more difficult to determine which of the digits (i.e. II,
III, or IV) made the large swim tracks. Digits II and IV tend to be nar-
rower than digit III, rendering the 1:3 claw/distal phalangeal pad ratio
estimate inaccurate.
Several Eubrontes tracks (and possible scoured-out tracks) are
preserved among the swim tracks at DSQ1 and DSQ2 (Fig. 8A). An
impressive Eubrontes track with a metatarsal impression, associated
scale scratch lines, and slide marks (Figs. 8B-C) indicates an interesting
movement of the animal. This deep left track has an estimated foot length
of about 30 cm (54 cm with metatarsal) and approximate width of 19 cm
(Fig. 8B). This track likely represents a theropod dinosaur at rest, stand-
ing in the water with a north-flowing current toward its cranial left side,
while the animal was facing northeast. The orientation of scratch circles
and other sedimentary structures support this interpretation. Judging
from the slide marks on the substrate (Fig. 8C), it appears as though the
dinosaur attempted to advance forward and was pushed off balance to
the left (i.e., in an oblique, down-current direction). The animal then
advanced slightly forward – but the trace then grades into features too
chaotic to allow clear interpretation. Abundant Grallator-type swim
tracks are associated on the same block with several of them formed after
the Eubrontes track maker moved away (Figs. 8B-C).
Other Types of Swim Tracks or Possible Swim Tracks at the
SGDS
An unusual, elongate tail-drag-like trace, oriented at an angle of
approximately 45° to the down-current flow direction (Fig. 9A) exists in
DSQ1. The thinnest end of the trace (down-current direction) is approxi-
mately 1.5 cm wide. The thickest end is in the up-current direction and
about 5 cm wide, with about 2 cm of relief. Along the thinner end is a
sharp, prominent ridge extending about 20 cm (Fig. 9A) that grades to a
smooth, rounded surface with decreasing width. The entire trace has long
scale scratch lines on both sides, and has a total length of 114 cm (span-
ning two blocks; field # SW.103 and SW.15; specimen SW.103 has the
majority of the trace and measures about 74 cm in length). The exact
affinities of this trace are unclear; however, it is certain that this was
made by a reptile, and likely a theropod dinosaur based on associated
swim tracks and Eubrontes and Grallator footprints. It might be a tail-
drag or swipe mark, or may have been caused by an animal placing an
appendage on the substrate and dragging it down-current.
Another trace of interest found in DSQ1 is an isolated,
Batrachopus-like footprint that measures 3 cm long and 2.5 cm wide
(Fig. 9B). This natural cast specimen has 0.8 cm of relief, shows four
digits, and represents a right pes track. What makes this footprint of
interest is how it came to be preserved with subaqueous traces and
sedimentary structures. It may have been made prior to swim track
formation and burial of the track surface during a regressive lake phase,
but given that this surface was buried quite rapidly, this suggestion
seems unlikely. Batrachopus tracks are usually attributed to
crocodylomorphs, and Early Jurassic crocodylomorphs, such as
sphenosuchians, have limbs that suggest an erect, digitigrade stance, and
are therefore more adapted for cursorial habits than modern crocodylians
(Parrish, 1987; Wu and Chatterjee, 1993, p. 78). Protosuchians were
probably more terrestrial than modern crocodylians, but not as special-
ized as sphenosuchians in the Triassic and Early Jurassic (Parrish, 1987;
Carroll, 1988, p. 281). However, it is possible that a member of either of
these crocodylomorph clades could have produced a track while par-
tially or completely submerged to create the Batrachopus-like track in
Figure 9B.
Several important cf. Batrachopus trackways are found at the
FIGURE 8. Eubrontes tracks. A, Right Eubrontes footprints associated with
dinosaur swim tracks (Field # SW.103). B-C,Eubrontes track and deep
metatarsal impression and scale scratch marks, with associated Grallator-
type swim tracks (Field # SW.69). B, Cranial is toward the top of the photo.
C, Close-up of scale scratch lines in B. Lines begin at original position of
foot (bottom left of photo) and extend toward the track maker’s left
(toward top of photo). CFD, current flow direction.

325
SGDS museum site in the in situ “Top Surface” track layer (Figs. 2-3).
All specimens consist of walking trackways transitioning into swim
tracks; the best-preserved example of this is SGDS.18.T5 (Figs. 9C-E),
which is associated with four other trackways that moved in the same
direction (SSW), paralleling the paleo-shoreline. All begin with clear
Batrachopus trackways that show manus/pes sets (Figs. 9C-D). As the
topography of the tracked surface drops in elevation, these walking
tracks transition into sets of parallel claw marks arranged in sets of two
to four (Figs. 9C, E). There are six sets of alternating left and right swim
tracks, with the sixth set turned westward. Associated current ripples
are oriented in a NW direction (Fig. 9C), so perhaps the crocodylomorphs
responsible for these trackways were being turned down-current as they
entered the water and began to swim. The longest trackway (SGDS.18.T5)
is approximately 95 cm long but may prove longer with future excava-
tion.
Another Batrachopus specimen, preserved in convex epirelief,
displays a partial trackway, possibly transitioning from a walk to a
swim. It shows three claw-scrape marks leading up to three clear and
well-preserved digit impressions (Fig. 9F). A fourth digit is present but
without a claw-scrape mark leading up to it. This entire track is 4.6 cm
long, and the claw-scrape marks are 1.8-2.2 cm long and 1 mm wide. The
claws on the clear toe impressions are also 1 mm wide, and very small
skin impressions are visible around the larger of the digits. The second
set of tracks in this short trackway is located 1.7 cm away and consists
of “swim” tracks showing three claw marks. The set is 1.5 cm wide, and
the claw marks are 1 mm wide and range from 1.7-2.8 cm in length.
Batrachopus swim tracks will be described elsewhere in the future.
Finally, an unusual swim-track specimen (SGDS.421; Fig. 9G)
was found in 2003 by Dr. Paul Bybee (Utah Valley State College, Provo,
Utah) in previously excavated rock piles northwest of the SGDS mu-
seum location, probably from the Stewart-Walker Tracksite level (Fig. 2;
see Milner et al., this volume for further discussion of this tracksite). It
is a set of about eight parallel claw marks, measuring 4.9 cm wide be-
tween the outer marks. These claw marks range in length from 1.6-6.2 cm
and are 0.15-0.4 cm wide. The differing depths and thicknesses of the
various claw marks make it look like coincidentally overlapping swim
tracks made either by different limbs of the same animal or, more likely,
different animals at different times. This trace could possibly be part of
a turtle swim track, but until better specimens are discovered, it will
remain of uncertain affinity.
DISCUSSION
As mentioned above, the ichnogenus Characichnos (Whyte and
Romano, 2001) could be considered merely an extramorphological vari-
ant of “walking” track types found associated with or near the swim
tracks. Because well-preserved swim tracks are found in close associa-
tion with Grallator and Eubrontes tracks at the SGDS sites, we can
attribute these swim tracks to the same track makers and refer to them as
FIGURE 9. Other types of swim tracks and associated, enigmatic tetrapod-produced structures. A, “Tail swipe mark” with associated swim tracks (Field #
SW.103). B, cf. Batrachopus track found associated with dinosaur swim tracks (Field # SW.29). C-E, “Top Surface” Batrachopus trackway transitioning
from walk to swim (SGDS.18.T5). Notice the orientation of current ripples and the change in direction of the swim tracks. F,Batrachopus specimen
transitioning from walk to swim (SGDS.176). G, Swim track of unknown affinity (SGDS.421).

326
Grallator-type and Eubrontes-type swim tracks. We do the same for
Batrachopus and Batrachopus-like swim tracks, since tracks of this
ichnotaxon also occur on the SGDS Top Surface tracksite transitioning
from walking to swimming traces.
We accept the ichnogenus name Characichnos as a valid descrip-
tion of a behavioral character of the track producers. Where swim tracks
are all that exist, they cannot be attributed to any particular subaerially
produced, “walking” track ichnotaxon. Cases such as the SGDS swim
track sites, at which swim tracks occur in close proximity to, and some-
times in direct association with, subaerially produced tracks, are rela-
tively rare. Given that morphology is the main ichnotaxonomic criterion,
Characichnos could be redefined as swim tracks of all bipedal dinosaurs,
whether ornithischian or saurischian, that show three parallel scrape
marks (see Lockley and Milner, this volume for further discussion of
non-dinosaurian swim tracks with similar morphologies). Now that
Characichnos has also been used as a label for a recently defined
ichnofacies (Hunt and Lucas, in press), it is likely that the name will
continue to be used by ichnologists. In short, the discovery of so many
well-preserved swim tracks at the SGDS enhances the need to subject
the Characichnos ichnogenus concept to further scrutiny.
We also infer that the “Eubrontes-swim tracks” described by
Coombs (1980) from Dinosaur State Park may not be swim tracks, but
are undertracks of Eubrontes. The interpretation of Coombs (1980) that
swimming theropods could make unique and recognizable tracks may be
valid, but the trace fossils may not be swim tracks (Galton and Farlow,
2003). Coombs also illustrated a “megalosaur” (Coombs, 1980, fig. 3;
Galton and Farlow, 2003, fig. 21) producing the proposed swim tracks
rather than a more likely producer of Eubrontes tracks, such as
Dilophosaurus (Paul, 1988,fig. 2-10) or a similar, large ceratosaur. The
overall morphological characteristics of the “swim tracks” described by
Coombs (1980) do not correspond well with most SGDS swim tracks,
which occur in sets of three with digit III displaying a longer and deeper
claw mark, extending caudally and cranially beyond the lateral digit traces
(II and IV) in most cases, unless moving in a down-current direction. The
Dinosaur State Park “swim tracks” may have been produced on a very
firm substrate as demonstrated by Farlow and Galton (2003), who ob-
tained nearly identical tracks produced by a rhea walking on plaster of
Paris that was nearly hardened (Galton and Farlow, 2003).
During the MTL depositional event, the majority of small
theropods were evidently floundering against a current in a depression
within the lake that paralleled the paleo-shoreline to the northwest of the
SGDS museum site. The sheer number of subparallel swim tracks made
by similarly-sized theropods implies that the track makers were moving
in an organized group. Gregariousness has been previously suggested for
Late Triassic and Early Jurassic theropods. The Whitaker Coelophysis
Quarry at Ghost Ranch, New Mexico, in the Upper Triassic Rock Point
Formation (Paul, 1988; Colbert, 1989, 1990), produced an assemblage
that includes a probable pack or packs of Coelophysis. Similar
coelophysoid assemblages involve multiple individuals of Megapnosaurus
rhodesiensis from southern Africa (Raath, 1977, 1990; Paul, 1988), and
a group of Megapnosaurus kayentakatae from the Early Jurassic
(Sinemurian) Kayenta Formation of northeastern Arizona (Rowe, 1989).
Megapnosaurus is presently the only theropod described by body fos-
sils in the Moenave Formation (Lucas and Heckert, 2001), so it is quite
possible that this animal, or one very similar, is responsible for the
Grallator tracks at the SGDS. The large concentration of SGDS swim
tracks support the hypothesis that these coelophysoid theropods exhib-
ited gregarious behavior, apparently a common phenomenon among Tri-
assic and Jurassic theropods (Ostrom, 1972; Raath, 1977, p. 19-21;
Colbert, 1989, p. 16-17, 148; Currie, 1997; Lockley and Matsukawa,
1999; Lockley et al., this volume).
CONCLUSIONS
The Megapnosaurus or Megapnosaurus-like dinosaurs that oc-
cupied the Lake Dixie shores during the Early Jurassic Epoch (approxi-
mately 198 million years ago) ventured out into the lake waters. The
excellent state of preservation of most of these traces suggests that sand
swept along by this current filled the traces simultaneous with their
formation. The high concentration of swim tracks indicates large groups
of small, probably ceratosaurian dinosaurs, fighting or floundering against
currents. The direction of natatory locomotion was parallel to the shore-
line in a more or less southerly direction, opposite the caudally-directed
claw marks preserved in the swim tracks. Occasionally, swim tracks are
oriented cross-current or in the direction of current flow twoard the
north.
Non-natatory Eubrontes trackways co-occur with the Grallator-
type swim tracks. This indicates that larger, probably Dilophosaurus-
like theropods were able to wade through areas where smaller theropods
presumably had to swim. A water depth of 1-1.25 m can therefore be
inferred from such Grallator swim track evidence. However, based on a
single large, Eubrontes-like swim track, water depth varied to as much as
1.75 m. A simpler explanation, considering that in outcrop Lake Dixie
was more than 100 km across, would be that the animal slipped on the
smooth mud in the strong current. Smaller swim track claw marks were
either made by smaller individuals (Grallator track maker juveniles and/
or subadults) or by the manual unguals striking the muddy bottom.
Batrachopus swim tracks also give insight into the behavior of small
crocodylomorphs as they transitioned from emergent areas to shallow
water along the shoreline and across areas of variable topography.
ACKNOWLEDGMENTS
We thank the Washington County School District, Quality Exca-
vation, City of St. George, Darcy Stewart and Bodega Development
Corporation, Sheldon and LaVerna Johnson, and Theresa Walker (former
City of St. George Tracksite Coordinator at the SGDS) for outstanding
contributions to the SGDS, and their efforts in preserving and protecting
this incredible site. We give special thanks to Aaron and Lester Jessop of
Steed-RW Construction for providing their excavation skills and digging
equipment for Dinosaur Swim Track Quarry #1. We thank Gonzalez
Construction for help in the tedious job of transporting many of the
rocks collected across the street to the future museum site. We are in-
debted to all of the dedicated volunteers at the SGDS and from the Utah
Friends of Paleontology. Special thanks to Dr. Sheldon Johnson for loan-
ing us his Bobcat for recovering isolated swim track blocks, and for his
front-end loader work for recovering float swim track specimens and
clearing the in situ excavation site in preparation for later excavation. We
thank Debra Mickelson and Paul Bybee (Utah Valley State College) for
helpful comments and suggestions. Partial funding was provided by the
City of St. George and the DinosaurAh!Torium Foundation. Thanks to
Jerald D. Harris (Dixie State College), Spencer G. Lucas (New Mexico
Museum of Natural History and Science), and Jennifer Cavin, Janae
Wallace, and Mike Lowe (Utah Geological Survey) for reviewing the
manuscript and making suggestions and comments that have greatly
improved it.

327
REFERENCES
Andrews, R.C., 1953, All about dinosaurs: New York, Random House, 146 p.
Bennett, S.C., 1992, Reinterpretation of problematical tracks at Clayton
Lake State Park, New Mexico: not a pterosaur but several crocodiles:
Ichnos, v. 2, p. 37-42.
Bernier, P., Barale, G., Bourseau, J.-P., Buffetaut, E., Demathieu, G., Gaillard,
C. and Gall, J.-C., 1982, Trace nouvelle de locomotion de chelonian et
figures d’émersion associées dans les calcaires lithographiques de Cerin
(Kimmeridgien superieur, Ain, France): Geobios, v. 15, p. 447-467.
Bernier, P., Barale, G., Bourseau, J.-P., Buffetaut, E., Demathieu, G., Gaillard,
C., Gall, J.-C. and Wenz, S., 1984, Découverte des pistes de dinosaures
sauteurs dans les calcaires lithographiques de Cerin (Kimmeridgien
superieur, Ain, France) – implications paléoecologiques: Geobios Mémoire
Spécial, v. 15, p. 177-185.
Bird, R.T., 1944, Did Brontosaurus ever walk on land?: Natural History, v.
53, p. 60-69.
Bird, R.T., 1985, Bones for Barnum Brown: adventures of a dinosaur hunter:
Fort Worth, Texas Christian University Press, 225 p.
Boyd, D.W. and Loope, D.B., 1984, Probable vertebrate origin for certain
sole marks in Triassic red beds of Wyoming: Journal of Paleontology, v.
58, p. 467-476.
Carroll, R.L., 1988, Vertebrate paleontology and evolution: New York,
W.H. Freeman and Company, 698 p.
Colbert, E.H., 1945, The dinosaur book: New York, The American Museum
of Natural History, Man and Nature Publication Handbook no. 14, 156
p.
Colbert, E.H., 1965, The Age of Reptiles: New York, W.W. Norton &
Company, 228 p.
Colbert, E.H., 1989, The Triassic dinosaur Coelophysis: Museum of North-
ern Arizona Bulletin, v. 57, p. 1-160.
Colbert, E.H., 1990, Variation in Coelophysis bauri,in Carpenter, K., and
Currie, P.J., eds., Dinosaur systematics: approaches and perspectives:
Cambridge, Cambridge University Press, p. 81-90.
Coombs, W.P., Jr., 1975, Sauropod habits and habitats: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 17, p. 1-33.
Coombs, W.P., Jr., 1980, Swimming ability of carnivorous dinosaurs: Sci-
ence, v. 207, p. 1198-1200.
Currie, P.J., 1997, Theropods, in Farlow, J.O., and Brett-Surnam, M.K.,
eds., The complete dinosaur: Bloomington, Indiana University Press, p.
216-233.
Czerkas, S.J. and Czerkas, S., 1990, Dinosaurs: a global view: Limpsfield,
Dragon’s World Press, 247 p.
DeBlieux, D.D., Kirkland, J.I., Smith, J.A., McGuire, J. and Santucci, V.L.,
2005, An overview of the vertebrate paleontology of Late Triassic and
Early Jurassic rocks in Zion National Park, Utah: The Triassic/Jurassic
terrestrial transition, abstracts volume, p. 2.
DeBlieux, D.D., Kirkland, J.I., Smith, J.A., McGuire, J. and Santucci, V.L.,
this volume, An overview of the vertebrate paleontology of Upper
Triassic and Lower Jurassic rocks in Zion National Park, Utah: New
Mexico Museum of Natural History and Science, Bulletin 37.
DeBlieux, D.D., Smith, J.A. McGuire, J.A., Kirkland, J.I., Santucci, V.L. and
Butler, M., 2003, A paleontological inventory of Zion National Park,
Utah and the use of GIS to create Paleontological Sensitivity Maps for
use in resource management: Journal of Vertebrate Paleontology, v. 23,
p. 45A.
Droser, M.L., Jensen, S. and Gehling, J.G., 2002, Trace fossils and substrates
of the terminal Proterozoic-Cambrian transition: implications for the
record of early bilaterians and sediment mixing: Proceedings of the
National Academy of Sciences, v. 99, p. 12572-12576.
Ezquerra, R., Costeur, L., Doublet, S., Galton, P.M. and Pérez-Lorente, F.,
2004, Lower Cretaceous swimming theropod trackway from “La Virgen
del Campo” (La Rioja, Spain): Palaeontological Association annual meet-
ing, Lille, December, 2004, abstracts volume, p. 150.
Farlow, J.O. and Galton, P.M., 2003, Dinosaur trackways of Dinosaur State
Park, Rocky Hill, Connecticut, in Letourneau, P.M. and Olsen, P.E.,
eds., The great rift valleys of Pangea in eastern North America, volume
2: New York, Columbia University Press, pp. 248-263.
Galton, P.M. and Farlow, J.O., 2003, Dinosaur State Park, Connecticut,
USA: history, footprints, trackways, exhibits: Zubia, v. 21, p. 129-173.
Gierliñski, G., NiedŸwiedzki, G. and Pieñkowski, G., 2004, Tetrapod track
assemblage in the Hettangian of Soltyków, Poland, and its
paleoenvironmental background: Ichnos, v. 11, p. 195-213.
Gierliñski, G. and Potemska, A., 1987, Lower Jurassic dinosaur footprints
from Gliniany Las, northern slope of the Holy Cross Mountains, Po-
land: Neues Jahrbuch für Geologie und Paläontologie Abhandlungen, v.
175, p. 107-120.
Gillette, D.D. and Thomas, D.A., 1989, Problematical tracks and traces of
Late Albian (Early Cretaceous) age, Clayton Lake State Park, New
Mexico, USA, in Gillette, D.D. and Lockley, M.G., eds., Dinosaur tracks
and traces: Cambridge, Cambridge University Press, p. 337-342.
Hotton, N., III, 1963, Dinosaurs: New York, Pyramid Publications, 192 p.
Hunt, A.P. and Lucas, S.G., in press, Tetrapod ichnofacies: a new paradigm:
Ichnos.
Ishigaki, S., 1989, Footprints of swimming sauropods from Morocco, in
Gillette, D.D. and Lockley, M.G., eds., Dinosaur tracks and traces: Cam-
bridge, Cambridge University Press, p. 83-86.
Kirkland, J.I., Lockley, M.G. and Milner, A.R., 2002, The St. George dino-
saur tracksite: Utah Geological Survey Notes, v. 34, p. 4-5, 12.
Kirkland, J.I. and Milner, A.R.C., 2005, The Moenave Formation at the St.
George Dinosaur Discovery Site at Johnson Farm (SGDS): The Triassic/
Jurassic terrestrial transition, abstracts volume, p. 8-9.
Kirkland, J.I. and Milner, A.R.C., this volume, The Moenave Formation at
the St. George Dinosaur Discovery Site at Johnson Farm: New Mexico
Museum of Natural History and Science, Bulletin 37.
Kvale, E.P., Hasiotis, S.T., Mickelson, D.L. and Johnson, G.D., 2001a,
Middle and Late Jurassic dinosaur fossil-bearing horizons: implications
for dinosaur paleoecology, northeastern Bighorn Basin, Wyoming, in
Hill, C.L., ed., Mesozoic and Cenozoic paleontology in the western
plains and Rocky Mountains: guidebook for field trips, Society of Verte-
brate Paleontology 61st Annual Meeting: Museum of the Rockies Occa-
sional Paper 3, p. 17-45.
Kvale, E.P., Johnson, G.D., Mickelson, D.L., Keller, K., Furer, L.C. and
Archer, A.W., 2001b, Middle Jurassic (Bajocian and Bathonian) dinosaur
megatracksites, Bighorn Basin, Wyoming, USA: Palaios, v. 16, p. 233-
254.
Lee, Y.-N. and Huh, M., 2002, Manus-only sauropod tracks in the Uhangri
Formation (Upper Cretaceous), Korea and their paleobiological impli-
cations: Journal of Paleontology, v. 76, p. 558-564.
Lockley, M.G., 1991, Tracking dinosaurs: a new look at an ancient world:
Cambridge, Cambridge University Press, 238 p.
Lockley, M.G. and Hunt, A.P., 1995, Dinosaur tracks and other fossil foot-
prints of the western United States: New York, Columbia University
Press, 338 p.
Lockley, M.G. and Matsukawa, M., 1999, Some observations on trackway
evidence for gregarious behavior among small bipedal dinosaurs:
Palaeogeography, Palaeoclimatology, Palaeoecology, v. 150, p. 25-31.
Lockley, M.G. and Meyer, C., 2000, Dinosaur tracks and other fossil foot-
prints of Europe: New York, Columbia University Press, 323 p.
Lockley, M.G. and Milner, A.R.C., this volume, Tetrapod tracksites from
the Shinarump Formation (Chinle Group, Upper Triassic) of southwest-
ern Utah: New Mexico Museum of Natural History and Science, Bulletin
37.
Lockley, M.G., Milner, A.R.C. and Lucas, S.G., 2005, Archosaur tracks from
the Chinle Group (Late Triassic), St. George area, southwestern Utah:
The Triassic/Jurassic terrestrial transition, abstracts volume, p.13-14.
Lockley, M. G. Milner, A.R.C., Slauf, D. and Hamblin, A.H., this volume,
Dinosaur tracksites from the Kayenta Formation (Lower Jurassic), Desert
Tortoise site, Washington County, Utah: New Mexico Museum of Natu-
ral History and Science, Bulletin 37.
Lockley, M.G., Pittman, J.G., Meyer, C.A. and Santos, V.F., 1994, On the
common occurrence of manus-dominated sauropod trackways in Meso-

328
zoic carbonates: Gaia, v. 10, p. 119-124.
Lockley, M.G. and Rice, A., 1990, Did “Brontosaurus” ever swim out to
sea?: evidence from brontosaur and other dinosaur footprints: Ichnos, v.
1, p. 81-90.
Lucas, S.G. and Heckert, A.B., 2001, Theropod dinosaurs and the Early
Jurassic age of the Moenave Formation, Arizona-Utah, USA: Neues
Jahrbuch für Geologie und Paläontologie Monatshefte, v. 2001, p. 435-
448.
Lucas, S.G., Lerner, A.J. and Milner, A.R.C., 2005, Early Jurassic inverte-
brate ichnofauna of the St. George Dinosaur Discovery Site, Moenave
Formation: The Triassic/Jurassic terrestrial transition, abstracts vol-
ume, p.14-15.
Lucas, S.G., Lerner, A.J., Milner, A.R.C. and Lockley, M.G., this volume,
Lower Jurassic invertebrate ichnofossils from a classic lake margin fa-
cies, Johnson Farm, southwestern Utah: New Mexico Museum of Natu-
ral History and Science, Bulletin 37.
Lucas, S.G. and Tanner, L.H., this volume, The Springdale Member of the
Kayenta Formation, Lower Jurassic of Utah-Arizona: New Mexico
Museum of Natural History and Science, Bulletin 37.
Marzolf, J.E., 1993, Palinspastic reconstruction of Early Mesozoic sedi-
mentary basin near latitude of Las Vegas; implications for Early Meso-
zoic cordilleran cratonic margin, in Dunne, G.C. and McDougall, K.A.,
eds., Mesozoic paleogeography of the western United States-II: Pacific
section, Society for Sedimentary Geology, v. 71, p. 433-462.
Marzolf, J.E., 1994, Reconstruction of the early Mesozoic Cordilleran
cratonal margin adjacent to the Colorado Plateau, in Caputo, M.V.,
Peterson, J.A. and Franczyk, K.J., eds., Mesozoic systems of the Rocky
Mountain region, USA: Denver, Society of Sedimentary Geology, p.181-
216.
McAllister, J.A., 1989a, Dakota Formation tracks from Kansas: implica-
tions for the recognition of tetrapod subaqueous traces, in Gillette, D.D.
and Lockley, M.G., eds., Dinosaur tracks and traces: New York, Cam-
bridge University Press, p. 343-348.
McAllister, J.A., 1989b, Subaqueous vertebrate footmarks from the upper
Dakota Formation (Cretaceous) of Kansas, U.S.A.: Occasional Papers
of the Museum of Natural History, University of Kansas, v. 127, p. 1-
22.
Metz, R., 1991, Scratch circles from the Towaco Formation (Lower Juras-
sic), Riker Hill, Roseland, New Jersey: Ichnos, v. 1, p. 233-235.
Metz, R., 1999, Scratch circles: a new specimen from a lake-margin deposit
of the Passaic Formation (Upper Triassic), Douglassville, Pennsylva-
nia: Northeastern Geology and Environmental Sciences Abstracts, v. 21,
no. 3, p. 2.
Milner, A.R.C., Kirkland, J.I., Lockley, M.G. and Harris, J.D., 2005a, Rela-
tive abundance of theropod dinosaur tracks in the Early Jurassic
(Hettangian) Moenave Formation at a St. George dinosaur tracksite in
southwestern Utah: bias produced by substrate consistency: Geological
Society of America Abstracts with Programs, v. 37, p. 5.
Milner, A.R.C. and Lockley, M.G., 2006, History, geology and paleontol-
ogy: St. George Dinosaur Discovery Site at Johnson Farm, Utah, in
Reynolds, R.E., ed., Making tracks across the southwest: the 2006 Desert
Symposium field guide and abstracts from proceedings. Zzyzx, Desert
Studies Consortium and LSA Associates, Inc., p. 35-48.
Milner, A.R.C., Lockley, M.G. and Johnson, S.B., this volume, The story of
the St. George Dinosaur Discovery Site at Johnson Farm: an important
new Lower Jurassic dinosaur tracksite from the Moenave Formation of
southwestern Utah: New Mexico Museum of Natural History and Sci-
ence, Bulletin 37.
Milner,A.R.C., Lockley, M.G., Kirkland, J.I., Bybee, P. and Mickelson,
D.L., 2004, St. George tracksite, southwestern Utah: remarkable Early
Jurassic (Hettangian) record of dinosaurs walking, swimming, and sitting
provides a detailed view of the paleoecosystem along the shores of Lake
Dixie: Journal of Vertebrate Paleontology, v. 24, p. 94A.
Milner, A.R.C., Lockley, M.G., Kirkland, J.I., Mickelson, D.L. and Vice,
G.S., 2005b, First reports of a large collection of well-preserved dinosaur
swim tracks from the Lower Jurassic Moenave Formation, St. George,
Utah: a preliminary evaluation: The Triassic/Jurassic terrestrial transi-
tion, abstracts volume, p.18.
Nadon, G.C. and Issler, D.R., 1997, The compaction of floodplain sedi-
ments: timing, magnitude and implications: Geoscience Canada, v. 24, p.
37-43.
Norman, D., 1985, The illustrated encyclopedia of dinosaurs: New York,
Crescent Books, 208 p.
Ostrom, J.H., 1972, Were some dinosaurs gregarious?: Palaeogeography,
Palaeoclimatology, Palaeogeography, v. 11, p. 287-301.
Parrish, J.M., 1987, The origin of crocodilian locomotion: Paleobiology, v.
13, p. 396-414.
Paul, G.S., 1988, Predatory dinosaurs of the world: New York, Touchstone,
Simon and Schuster Inc., 464 p.
Pieñkowski, G. and Gierliñski, G., 1987, New finds of dinosaur footprints in
Liassic of the Holy Cross Mountains and its paleoenvironmental back-
ground: Prezegl¹d Geologiczny, v. 4, p. 199-205.
Raath, M.A., 1977, The anatomy of the Triassic theropod Syntarsus
rhodesiensis (Saurischia: Podokesauridae) and a consideration of its biol-
ogy [Ph.D. dissertation]: Salisbury, Rhodes University, 233 p.
Raath, M.A., 1990, Morphological variation in small theropods and its
meaning in systematics: evidence from Syntarsus rhodesiensis,in Car-
penter, K. and Currie, P.J. (eds.), Dinosaur systematics approaches and
perspectives: Cambridge, Cambridge University Press, p. 91-105.
Rowe, T., 1989, A new species of the theropod dinosaur Syntarsus from the
Early Jurassic Kayenta Formation of Arizona: Journal of Vertebrate
Paleontology, v. 9, p. 125-136.
Rygel, M.C., Gibling, M.R. and Calder, J.H., 2004, Vegetation-induced sedi-
mentary structures from fossil forests in the Pennsylvanian Joggins
Formation, Nova Scotia: Sedimentology, v. 51, p. 531-552.
Thulborn, R.A., 1989, The gaits of dinosaurs, in Gillette, D.D. and Lockley,
M.G., eds., Dinosaur tracks and traces: Cambridge, Cambridge Univer-
sity Press, p. 39-50.
Thulborn, R.A., 1990, Dinosaur tracks: London, Chapman Hall, 410 p.
Williams, J.A.J., Milner, A.R.C. and, Lockley, M.G., this volume, The Early
Jurassic (Hettangian) LDS Tracksite from the Moenave Formation in
St. George, Utah: New Mexico Museum of Natural History and Science,
Bulletin 37.
Whyte, M.A. and Romano, M., 2001, A dinosaur ichnocoenosis from the
Middle Jurassic of Yorkshire, UK: Ichnos, v. 8, p. 233-234.
Wu, X.-c. and Chatterjee, S., 1993, Dibothrosuchus elpahros, a
crocodylomorph from the Lower Jurassic of China and the phylogeny
of the Sphenosuchia: Journal of Vertebrate Paleontology, v. 13, p. 58-
89.