Cretaceous Anuran And Dinosaur Footprints From The Patuxent Formation Of Virginia

17 Pages·1997·8.8 MB·English

by  R E Weems

PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
110(1):1-17. 1997.
Cretaceous anuran and dinosaur footprints from the
Patuxent Formation of Virginia
Robert E. Weems and Jon M. Bachman
(REW) Mail Stop 926A, U.S. Geological Survey, Reston, Virginia 20192, U.S.A.;
(JMB) 115 Windsor Circle, Fredericksburg, Virginia 22405, U.S.A.
—
Abstract. Footprints of an anuran (gen. et sp. indet.), a theropod dinosaur
(Megalosauropus sp.), and an ornithopod dinosaur {Amblydactylus sp.) have
been recovered from the Lower Cretaceous Patuxent Formation in Stafford
County, Virginia. These footprints are the first record of terrestrial vertebrates
from Cretaceous strata in Virginia, and their discovery suggests that the scarcity
of bones and teeth in the Patuxent probably is an artifact of preservation. The
anuran trackway provides the oldest known direct evidence for hopping loco-
motion among these amphibians.
The Patuxent Formation has yielded an may have existed while the slightly older
abundant fossil macroflora from Maryland Patuxent sediments were accumulating. Un-
and Virginia (Fontaine 1889; Ward et al. til the discovery of the footprints described
1905; Berry 1908, 1910a, 1910b, 1910c, here, however, no direct evidence for such
&
1911a, 1911b, 1911c, 191 Id; Clark Mil- a fauna was available. —
ler 1912; Brenner 1967; Wolfe 1972; Doyle Location, age, and geologic setting.
&
Hickey 1975; Skog 1982, 1988, 1992; The footprints described here were found
Hickey 1986), as well as a diverse paly- by Jon Bachman in strata of the Lower Cre-
noflora (Brenner 1963, Hughes & Moody taceous Patuxent Formation in a roadcut on
&
1966, Doyle 1977, Doyle Robbins 1977). the east side of U.S. Route 1, located 0.25
In sharp contrast, the only vertebrate ma- mi south of Potomac Creek in Stafford
terial reported from this unit is a fish skel- County, Virginia (Fig. 1). In its outcrop
eton, found somewhere in the James River belt, the Patuxent Formation is approxi-
&
valley of Virginia (Berry 1911a). There- mately 200 to 300 ft thick (Clark Miller
fore, it is significant that footprints of an 1912). The unit dips gently and thickens to
anuran and two kinds of dinosaur have been the east, becoming about 575 ft thick twen-
found north of Fredericksburg, Virginia, in ty-five mi east-southeast of Fredericksburg,
Stafford County (Fig. 1). These prints pro- near Oak Grove (Reinhardt et al. 1980). In
vide our first glimpse of the Cretaceous ver- the Fredericksburg region, pre-Mesozoic
tebrate fauna that once existed in Virginia. crystalline rocks directly underlie the Pa-
The Lower Cretaceous Arundel Formation, tuxent Formation with profound unconfor-
which overlies the Patuxent Formation in mity. These rocks crop out along the floor
Maryland but not in Virginia (Fig. 2), has of Potomac Creek well to the west of U.S.
produced a diverse but fragmentary verte- Route 1. Because the footprint-bearing out-
brate fauna that includes turtles, crocodiles, crop is located toward the eastern margin
and dinosaurs (Lull 1911a, 1911b; Gilmore of the Patuxent outcrop belt, the footprints
1921; Ostrom 1970; Galton & Jensen 1979; probably come from the upper half of that
Kranz 1989; Weishampel 1990; Martin & formation. However, faults are known irom
Brett-Surman 1992). The diversity of this the general vicinity of the footprint-1 aring
fauna suggests that a diverse fauna also outcrop (Mixon & Newell 1977 so the
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
SPOTSYLVANIA
CO
Fig. 1. Map showing Bachman locality (star) where footprints have been found in the Lower Cretaceous
Patuxent Formation north ofFredericksburg, Virginia (oblique rule). Outcrop belt ofLower Cretaceous Patuxent
Formation is shown in dark gray. Younger Tertiary and Quaternary sediments crop out to the east of this belt,
while older, pre-Mesozoic igneous and metamorphic rocks crop out to the west (both areas shown in light gray).
The Lower Cretaceous strata dip gently toward the east and thicken in that direction.
'
VOLUME NUMBER
110, 1
COLORADO/ BRITISH
PERIOD STAGE MARYLAND VIRGINIA TEXAS WYOMING COLUMBIA
97.5 FREDERICKSBURG —
GROUP__
ALBIAN PATAPSCO FM PATAPSCO FM. GLEN ROSE FM DAKOTA FM GETHTNG FM
ARUNDEL FM.
113
APTIAN LOWER TRINITY GROUP
119 BARREMIAN PATUXENT FM. PATUXENTFM CADOMINFM.
LOWER CRETACEOUS
124
HAUTERIVIAN
131
VALLANGINIAN
MINNES FM
138
BERRIASIAN
144 MYST MTN.FM.
Fig. 2. Comparison of the Lower Cretaceous geologic columns preserved in Maryland, Virginia, Texas,
Colorado/Wyoming, and British Columbia. Time intervals represented by preserved strata are shown; missing
time intervals are indicated by gray areas.
stratigraphic position of this outcrop could east came within thirty miles of the foot-
be lower than it appears to be. print-bearing locality during Patuxent time
The Patuxent, based on palynomorph (Reinhardt et al., 1980). These regional
&
studies (Brenner 1963, Doyle Robbins stratigraphic relationships indicate that the
1977, Reinhardt et al. 1980), is Barremian depositional environment of the Patuxent
to Aptian (124 to 115 Ma) in age. The rel- shifted southeastward from fluvial braided-
atively high horizon ofthe footprint-bearing stream through delta plain to marginal-ma-
bed suggests that an Aptian age (119 to 115 rine delta front environments (Reinhardt et
Ma) is more likely than Barremian. The al. 1980). Although the footprint-bearing
Aptian part of the Patuxent represents a part of the Bachman locality is dominated
portion of the Lower Cretaceous column by fine-grained sediments (see Table 1),
that is sporadically but widely preserved in other outcrops in the immediate vicinity
North America (Fig. 2). Thus, a number of tend to be dominated by cross-bedded to
Aptian units (the Arundel Formation of massive medium- to coarse-grained sands
Maryland, the Glen Rose Formation ofTex- and sandstones that often are pebbly. This
as, the basal Dakota Formation of Colorado local association of lithologies indicates
and Wyoming, and the Cadomin and basal that the strata at the Bachman locality were
Gething Formations of British Columbia) formed either in a temporary slack-water
are close in age to the Patuxent, and their area within the lower reaches of a fluvial
footprints and skeletal remains are especial- braided-stream system, or else it was part
ly relevant for comparison to the Patuxent of a deltaic environment that was briefly es-
footprints described here. tablished across the lower reaches of a flu-
In its outcrop belt, the Patuxent Forma- vial braided-stream system while its gradi-
tion has yielded only terrestrial fossil re- ent was lowered due to a marine incursion
&
mains (Clark Miller 1912, Berry 1911a), from the east.
and its sedimentary fabric indicates that The frog footprints were impressed on a
most of the unit accumulated in a fluvial clayey silt layer that dried out enough not
braided-stream environment (Glaser 1969). to be disrupted by deposition of an inch-
However, about thirty miles east-southeast thick (2.5 cm-thick) layer of ripple-cross-
of the Bachman locality near Oak Grove, laminated sandy and clayey silt containing
glauconite has been reported in the Patux- scattered stem and leaf debris. The top of
ent in its subcrop belt. Because glauconite this one inch-thick layer in turn was in the
forms under marine conditions, the pres- process of drying when the two types of
ence of this mineral indicates that marine dinosaurs came by and left their fo prints.
incursions across the Coastal Plain from the After this layer dried enough t become
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
4
—
Table 1. Section of Patuxent Formation on U.S. Route 1, 0.25 mile south of Potomac Creek.
Bed Description Thickness
8 Covered, sandy residuum overgrown with pines 8.5 ft
7 Sandstone, fine- to medium-grained, pale-yellowish-orange 1.8
(10YR8/6) to very-pale-orange (10YR8/2), clayey and silty, con-
tains scattered clayballs, fines upward
6 Siltstone, clayey, light-brownish-gray (5YR5/1), thinly laminated, 0.2
contains leaf impressions (mostly conifers)
5 Sandstone, medium- to coarse-grained, pale-yellowish-orange 1.5 to 2.0
(10YR8/6) to very-pale-orange (10YR8/2), poorly sorted, cross-
bedded, base has 6 inch amplitude southeast-trending swales cut
into bed below; fines upward to very fine- to fine-grained and
silty
4 Sandstone, dominantly fine-grained but very fine- to medium, silty, 0.5 to 1.0
yellowish-gray (5YR8/1), contains mudballs, carbon chips scat-
tered on bedding surfaces, stem fragments and impressions scat-
tered throughout, faintly laminated but mostly massive, symmet-
rical ripples and interference ripples on laminar surfaces,
dinosaur footprints within this interval
3 Siltstone, very fine sandy, micaceous, very pale-orange (10YR8/2), 2.5
abundant leaf impressions (mostly ferns), and stem fragments,
well bedded with Vs to xk in.-thick impersistent laminae; some
laminae contain symmetrical ripple marks; upper surface planar,
anuran prints on top of this unit
2 Sandstone, fine- to medium-grained, poorly bedded, very pale-yel- 1.5
lowish-brown (10YR7/2), scattered stem fragments present, pla-
nar bedded with 0.5 to 2.0 in.-thick laminae
1 Sandstone, medium- to coarse-grained, pale-yellowish-orange 10.0
(10YR8/6) to very pale-orange (10YR8/2), scattered mudballs
(up to 2 in. diameter) and quartz pebbles (up to 4 in. diameter),
subrounded to rounded, prominently crossbedded
Total thickness 27.0 ft
firm, it was in turn covered by a bed of of New Zealand, which are dominated by
silty, poorly sorted fine- to medium-grained broad-leafed conifers and ferns (Brenner
sand that contained abundant mudballs. 1963), and this implies that a similar cli-
This crudely graded layer ranges from 1.25 mate existed in Virginia during deposition
to 2.5 in. (3 to 6 cm) in thickness. The dis- of the Patuxent.
tribution of these layers appears to be quite Much of the Patuxent is unlithified, but
local, as would be expected in a lower flu- some horizons in the Stafford County area
vial braided-stream or upper deltaic envi- are semilithified to lithified. Between 1790
ronment. There are no indications that the and 1840, the lithified strata were quarried
local environment was paludal to lacustrine, and used as a building stone known as
as has been surmised for much ofthe Arun- "Aquia Creek sandstone" or "Virginia
del Formation (Glaser 1969). Although the freestone." Parts of the White House, Cap-
anuran suggests proximity to permanent itol, Treasury Building, and National Por-
fresh water, the presence of well-formed trait Gallery were built of this material
footprints of any kind indicates that the lo- (Withington, 1975). The strata in the foot-
cal environment was damp but subaerial at print-bearing outcrop are only in a semili-
the times the prints were formed. The flora thified state, and thus unsuitable for use as
of the Patuxent is strikingly similar to that building stone. Even so, their modest co-
of the modern warm-temperate rain forests herence greatly simplified collection and
VOLUME NUMBER
110, 1
Fig. 3. Stereophotos of part of an anuran trackway preserved on a counterpart slab (USNM 475487) from
the Lower Cretaceous Patuxent Formation of Virginia. Scale is in centimeters.
preservation of the footprints described notably smaller and less widely spread than
here. — the rear feet. Its gait placed left and right
Systematics. Specimens described in feet side-by-side during locomotion, rather
this paper have been donated to the collec- than in the obliquely-alternating pattern that
tions of the United States National Museum is normal for most tetrapod vertebrates dur-
(USNM). Their taxonomic placement is ing locomotion. This pattern indicates a
summarized as follows: hopping mode of progression. Although
much of the trackway is indistinct, one pair
Class Amphibia of rear and front feet are fairly well pre-
Order Anura served (Figs. 5, 6). These prints reveal the
Gen. et sp. indet. presence of three robust toes on each rear
Two contiguous slabs (USNM 475487 foot, with the middle toe elongated only
and USNM 475488) from the Bachman lo- slightly beyond the other two. The front
feet are less distinct, but they appear to be
cality preserve a single long trail of a small
elongate and rotated outward about 30°
quadrupedal animal. The preserved impres-
from the line of travel. As preserved, this
sions represent the counterpart filling of the
original trackway. All but one of the tracks trackway corresponds to no known animal.
on one slab are shown in Fig. 3, and the The relatively faint preservation of the en-
entire trackway is represented in Fig. 4. tire trackway, however, strongly suggests
Generally the tracks are poorly preserved, that at least some of the trackmaker's toes
in part due to the small size of the animal may not have impressed into the substrate.
that made them. Even so, the trackway pat- When our Cretaceous trackway is com-
tern is distinctive. The animal moved with pared to tracks made by a modern iog
left and right rear feet spread widely apart, (Rana clamitans) and a modern tor- JBufo
and it possessed front feet that appear to be woodhousei), the Cretaceous trac1 vay can
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
475487
v^i) &£
Rana
clamitans
475488
Bufo
woodhousei
10 cm
p
Fig. 4. Comparison of Patuxent anuran trackway (right) with footprints made in plaster by a modern frog
(Rana clamitans, upper left) and a modern toad (Bufo woodhousei, lower left). Tracks are light gray, except for
Bufo woodhousei where light gray areas are portions of tracks that are deeply impressed and unshaded areas are
portions of tracks only very lightly impressed. Dark gray area in upper right is an area of mineralization around
a woody stem or branch (black). Trackway at right is shown as a set of positive prints and thus is mirror-
reversed to Fig. 3. Dashed rectangle encloses footprints shown in Figs. 5 and 6. All prints drawn to same scale.
VOLUME NUMBER
110, 1
Fig. 5. Closeup stereophctos of the third set of counterpart track fillings in the anuran trackway illustrated
in Fig. 3, which is also the track set marked in Fig. 4 by the dashed rectangle.
be seen to be plausible for an anuran hop- marks represented the impression of the en-
ping across a substrate that was barely soft tire diminutive front foot of some very
enough to record the animal's passage. Al- enigmatic animal.
though anurans have four toes on their front The only characteristic of this trackway
feet and five toes on their rear feet, tracks that is exceptional for an anuran is the dis-
made for us by modern frogs and toads tance between hops, which proportionally is
hopping across wet plaster (Fig. 4) indicate much shorter than in modern frogs and
that usually only the central three toes (II- toads (Fig. 4). Given the great antiquity of
IV) of the pes leave clear marks. This in- this trackway, however, there is no com-
dicates that modern anurans primarily use pelling reason to expect that its maker
pedal digits II-IV to power their hops and should have been nearly so efficient at hop-
also implies that they use pedal digits I and ping as are modern frogs or toads. Because
V
primarily for balance. Similarly, the frog the trackmaker maintained a steady (albeit
only occasionally produced distinct manus plodding) gait, in that regard it was distinct-
prints, and the toad produced manus prints ly more like a toad than a frog. Toads can
that were lightly impressed except for digit afford to be deliberate in their gait, because
I, which was much more deeply impressed they possess poisonous skin glands that
than any of the other digits. From these ob- make them nearly immune to predation.
servations, we conclude that our Cretaceous Frogs, in contrast, must rely on their pow-
trackway probably was made by an anuran erful hops and erratic movements to avoid
that left only an imperfect representation of attack. Therefore, the deliberate and plod-
its feet. The Cretaceous anuran typically ding gait demonstrated by this trackway
left only impressions of pedal digits II-IV, strongly suggests that its makei, although
and its manus either did not leave visible probably too ancient to be a true toad, prob-
marks or left impressions only of digit I. In ably possessed skin glands that produced
this interpretation, the impressions of man- secretions to protect it from attack. Mem-
ual digit I are rotated 150° inward, which bers of the living family Discoglos tdae,
would be normal for an anuran, rather than known as far back in the fossil rr ord as
30° outward as would be indicated if these the Jurassic, produce distasteful sl< n secre-
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
Fig. 6. Interpretation of the stereophotos of counterpart anuran footprint fillings shown in Fig. 5. Footprints
are shown as counterparts for easier comparison with Fig. 5.
tions (Zweifel 1992). Therefore, it is rea- taceous, and four living families are known
sonable to suggest that at least some anu- to have been present by the Early Creta-
rans developed this type of protection by ceous in South America (Ascaphidae), Eu-
the Early Cretaceous. rope (Discoglossidae, Paleobatrachidae),
Anuran trackways have not been reported and Asia (Pipidae) (Benton 1987, Carroll
in the ichnotaxonomic literature (Haubold 1988). From these occurrences, it could be
1971). Therefore, by definition, this track- inferred that anurans were present in eastern
way represents an undescribed genus and North America in the Early Cretaceous, but
species. Unfortunately, the rather poor de- none of this skeletal material provides di-
tail of this trackway allows diagnosis of it rect proof of their presence. Second, the Pa-
only to the ordinal level of taxonomy. For tuxent specimen is attributable to an anuran
this reason, it is not suitable for use as a specifically because of its hopping mode of
type for a new ichnotaxon. Yet even at the locomotion. Thus, even though a hopping
ordinal level of identification, this trackway capability has been attributed to Prosalirus
is still of considerable interest. First, it is on anatomical grounds, our specimen pro-
the oldest direct evidence for the existence vides the oldest direct evidence for sus-
of anurans in Virginia or anywhere else in tained hopping locomotion in frogs.
the Eastern United States. The recent skel-
etal description of a functionally modern Class Reptilia
Order Theropoda
frog {Prosalirus bitis) from the Lower Ju-
rassic of Utah (Shubin & Jenkins 1995) in- Megalosauropus sp.
(USNM
dicates that anurans were present in western Several tridactyl footprints
USNM USNM
North America well before the Early Cre- 475489, 475490, and
VOLUME NUMBER
110, 1
Fig. 7. Stereophotos of a counterpart filling of a footprint of Megalosauropus (USNM 475489) from the
Lower Cretaceous Patuxent Formation of Virginia. Scale is in centimeters.
Fig. 8. Stereophotos of a counterpart filling of a footprint of Megalosauropus (USNM 47549f" om the
Lower Cretaceous Patuxent Formation of Virginia. Scale is in centimeters.
PROCEEDINGS OF THE BIOLOGICAL SOCIETY OF WASHINGTON
10
Fig. 9. A. Interpretation of footprint of Megalosauropus derived from the counterpart left footprint filling
shown in Fig. 7. B. Interpretation of footprint of Megalosauropus derived from the counterpart left footprint
filling shown in Fig. 8. C. Interpretation of right footprint ofMegalosauropus (USNM 475491, not figured). D.
Composite of footprint interpretations shown in Fig. 8A-C. Because the counterparts are of left feet and the
positive is of a right foot, symmetries match without mirror-symmetry adjustment.