This pdf file is licensed for distribution in the form of electronic reprints and by way of personal or institu-
tional websites authorized by the author(s). 
A copyright license has been purchased to make this possible.
VERTEBRATE TRACKWAYS AMONG A STAND OF SUPAIA WHITE
PLANTS ON AN EARLY PERMIAN FLOODPLAIN, NEW MEXICO
WILLIAM A. DIMICHELE,1 SPENCER G. LUCAS,2 AND KARL KRAINER3
1Department of Paleobiology, NMNH Smithsonian Institution, Washington, DC 20560, USA, ,dimichel@si.edu.; 2New Mexico Museum of Natural
History and Science, Albuquerque, NM 87104, USA, ,spencer.lucas@state.nm.us.; and 3Institute of Geology and Paleontology, University of
Innsbruck, Innsbruck A-6020, Austria, ,karl.krainer@uibk.ac.at.
VERTEBRATE TRACKWAYS AMONG A STAND OF SUPAIA WHITE
PLANTS ON AN EARLY PERMIAN FLOODPLAIN, NEW MEXICO
WILLIAM A. DIMICHELE,1 SPENCER G. LUCAS,2 AND KARL KRAINER3
1Department of Paleobiology, NMNH Smithsonian Institution, Washington, DC 20560, USA, ,dimichel@si.edu.; 2New Mexico Museum of Natural
History and Science, Albuquerque, NM 87104, USA, ,spencer.lucas@state.nm.us.; and 3Institute of Geology and Paleontology, University of
Innsbruck, Innsbruck A-6020, Austria, ,karl.krainer@uibk.ac.at.
ABSTRACT?Little is known about the habit and spatial distribution of Early Permian tropical vegetation, a sharp
contrast with the Pennsylvanian from which many in-situ ??T0?? assemblages are known. Even less is known about
the potential interaction of plants and vertebrates. Here we report the discovery of a small stand of 34 probable
Supaia White plants from the Abo Formation of New Mexico. The plants were growing on a mudflat, subject to
periodic flooding and exposure. The same mudflat hosts trackways of vertebrates that appear to have walked
around or between the Supaia plants. The stems are preserved as molds, and vary from 20 mm to 70 mm in diameter,
averaging 42.4 mm, indicating heights of approximately 2.5?4 m. The plants, which may be described as small trees
given their estimated height, are as close as 110 mm to their nearest neighbor and average nearly 300 mm apart. A
series of lines or filled fissures, which we interpret as roots, radiates from the base of each stem. Leaves of Supaia
thinfeldioides White are the only foliage found in association with these stems, on bedding planes above and at the
base of the lowest expression of the stem molds. Associated vertebrate trackways either congregate around some of
the stems or wend their way between the stems and include those of a single large temnospondyl amphibian
(Limnopus Marsh) and many of small predatory parareptiles (Dromopus Marsh). This study demonstrates that S.
thinfeldiodes were small-statured, weedy, opportunistic plants. It also shows that contemporaneous vertebrates
prowled such environments, presumably either for food, shelter, or both given detectable pace and path.
INTRODUCTION
REAL-TIME OR so-called T
0 data (Johnson, 2007; DiMichele
and Falcon-Lang, 2011) are among the most sought after
in the fossil record. Under ideal circumstances they may reveal
spatial patterns, growth conditions, behaviors of animals, and
interactions among organisms comparable to those attainable
from certain kinds of modern ecological studies. The most
celebrated data of this type are trackway assemblages of
vertebrates and invertebrates, which reveal aspects of behavior
difficult to ascertain from most body fossils (e.g., Haubold, 1984;
Buatois et al., 1998; Buatois and Ma?ngano, 2007). For plants, the
closest example to the trackway record is that of vegetation
buried in situ by ash falls (Wing et al., 1993; Pfefferkorn and
Wang, 2007; Oplus?til et al., 2009), floods (Gastaldo et al., 2004;
DiMichele et al., 2009; Falcon-Lang et al., 2011), or by other
catastrophic means (e.g., DiMichele and Falcon-Lang, 2011 for
summary of Paleozoic data), which allow determination of plant
density and spacing, inferences about size, and habitats of
growth. A combination of plant and vertebrate in situ records
has the power to reveal trophic interactions, even if only by
inference, but still one based on clear behavioral data; there are
very few such examples known from the fossil record (e.g.,
Parker and Balsley, 1989; Carpenter, 1992).
Reported here is a stand of Early Permian tropical plants
of slight stature, preserved on a mud flat in a seasonal
environment, subject to periodic flooding and drying. The
plants are likely Supaia thinnfeldioides White (1929) based on
the monotypic associated foliage litter. Alone this would be a
singular discovery because so little is known about the growth
habits and physical distribution of Early Permian tropical
plants; many kinds of foliage are well known, but the habit of
the plants to which such foliage was attached remains virtually
unknown for broadleaved forms. In addition to the plants,
however, this deposit contains trackways of at least two kinds
of vertebrates, a large temnospondyl amphibian and small,
predatory parareptiles.
Supaia White (1929) was erected to encompass a number of
kinds of small forked fronds bearing pinnately compound
foliage of xeromorphic aspect, from the earliest Permian of the
Grand Canyon of Arizona. Originally thought to be confined
to Early Permian age rocks of the western United States (Read
and Mamay, 1964), Supaia has subsequently been described
from the Permian of China (Wang, 1997; Deng et al., 2009),
Europe (Gand et al., 1997; Galtier and Broutin, 2008) and
possibly Sumatra (Booi et al., 2009), and thus appears to have
been distributed across the Pangaean low latitudes. Wang
(1997) suggested an association with Autunia-type fructifica-
tions, which would support a peltaspermalean affinity, a
conclusion supported by DiMichele et al. (2005). From its
earliest description, the plant has been associated with
environmental conditions indicative of seasonal dryness, with
some indication of tolerance of severe moisture stress (White,
1929; Gand et al., 1997; DiMichele et al., 2007). Such
suppositions, however strongly supported they may be by
sedimentological and foliar morphological evidence, are
uninformed by any knowledge of the stature of the Supaia
plant or its population structure?common problems for
many-to-most fossil plant species. Essentially, it is not known
if this plant was tree, shrub, vine or ground-cover. These
deposits suggest it was a small, upright, tree-like plant, several
meters in height at most.
The vertebrate footprint ichnotaxa reported here are
characteristic of red-bed ichnoassemblages across a wide part
of the Early Permian Pangean tropics (Hunt and Lucas, 2006).
They represent well known temnospondyl and parareptile
trackmakers. However, their direct association with in situ
Permian ?trees? is very unusual, perhaps unique in the Early
Permian track record.
GEOLOGICAL SETTING
Location.?The fossil assemblage documented here was
discovered in an outcrop of the lower part of the Abo
Formation exposed along Cuchillo Negro Creek on the US
Journal of Paleontology, 86(4), 2012, p. 584?594
Copyright ? 2012, The Paleontological Society
0022-3360/12/0086-0584$03.00
584
Geological Survey Cuchillo, New Mexico 7K? Quadrangle to
the northwest of the town of Truth or Consequences (Fig. 1).
The map coordinates are in the locality database of the New
Mexico Museum of Natural History (NMMNH) as locality
8376.
Physical geology and age.?The thickness of the Abo
Formation outcrop exposed along the Chuchillo Negro Creek
is 21 m (Fig. 2). The exposed Abo red beds are mostly
siltstone-mudstone (comprising 60% of the section) with two
intercalated sandstone units (comprising 40% of the section)
(Figs. 2, 3.1). The siltstone to mudstone is commonly finely
laminated (lithofacies Fl) or appears massive (lithofacies Fm)
(lithofacies codes after Miall, 1996). Pedogenic carbonate
nodules and lenses (lithofacies P) are present in the siltstone-
mudstone at the base of the lower sandstone unit. In the
siltstone-mudstone succession between the two sandstone
units, pedogenic carbonate nodules float in the fine-grained
sediments and two nodular, lenticular, calcareous pedogenic
horizons (lithofacies P) are present. The upper pedogenic
horizon contains poorly-developed rhizoliths.
The lower sandstone unit is 3.1 m thick, composed of fine-
grained sandstone displaying trough cross-bedding (lithofacies
St), low-angle cross-bedding (lithofacies Sl) and fine-grained
sandstone with horizontal lamination (lithofacies Sh). The
lower contact is erosional and grain-size and bed thickness
decrease upward. In the horizontally laminated upper part,
desiccation cracks occur in several layers. Intercalated in the
fine-grained sandstone are two thin siltstone-mudstone beds
that are 0.2 and 0.6 m thick. The footprints and fossil plants
documented here are in the upper part of this unit (Fig. 2, Bed
4; Fig. 3.2).
The upper sandstone (Fig. 2, units 11?17, Fig. 3.1) unit is
6.5 m thick and is composed of individual sandstone beds
displaying trough cross-bedding (lithofacies St), low-angle
cross-bedding (lithofacies Sl), horizontal lamination (lithofa-
cies Sh) and rare ripple cross lamination (lithofacies Sr). The
FIGURE 1?Location of study site (black dot) on Cuchillo Negro Creek, east of the Mud Springs Mountains and northwest of the town of Truth or
Consequences in southern New Mexico.
DIMICHELE ET AL.?VERTEBRATES AMIDST THE SUPAIA?PERMIAN OF NEW MEXICO 585
FIGURE 2?Lithologic log of the Abo Formation at the study site on Cuchillo Negro Creek. Plant stems and vertebrate trackways studied occur in
Bed 4.
586 JOURNAL OF PALEONTOLOGY, V. 86, NO. 4, 2012
lower contact is erosional. Several thin siltstone-mudstone
layers are intercalated. Mudcracks occur in the upper part of
the sandstone unit within horizontally laminated sandstone.
Sandstone units in the Abo Formation commonly occur
as sandstone sheets, which can be traced laterally over long
distances (e.g., Krainer and Lucas, 2010; Lucas et al.,
2005a). Sandstone sheets correspond to the architectural
element CH (channel) of Miall (1996). We suggest that at
Cuchillo Negro the sandstone units form such sandstone
sheets, although the limited local exposure does not allow
tracing them laterally.
Sandstone sheets were deposited in broad, shallow channels
of a probable low sinuosity river system. The sandstone units
are similar to the sandstone sheets described from the Abo
Formation at Jemez River in northern New Mexico and at the
type locality near Abo Pass in central New Mexico (Eberth
FIGURE 3?Outcrop on Cuchillo Negro Creek. 1, exposure of Early Permian Abo Formation overlain by Pleistocene Palomas Formation; plant fossils
and vertebrate trackways occur on the bedding surface exposed at the bottom of the outcrop to the left of the person standing at the outcrop base;
2, bedding surface on which stem casts and trackways are preserved; the Limnopus trackway can be seen to the right of the Jacob?s Staff.
DIMICHELE ET AL.?VERTEBRATES AMIDST THE SUPAIA?PERMIAN OF NEW MEXICO 587
and Miall, 1991; Lucas et al., 2005a; Krainer and Lucas,
2010), but finer-grained than farther north.
The siltstone-mudstone (lithofacies Fl, Fm) is assigned to
the element FF (floodplain fines) of Miall (1996), representing
mainly floodplain deposits that formed by the settling of fine-
grained sediment from sheetfloods on extensive floodplains.
The presence of fine lamination (lithofacies Fl) indicates that
the sediments were deposited from suspension.
Compared to the Abo Formation at Jemez River or near
Abo Pass, the Abo Formation at Cuchillo Negro Creek is
developed in a more distal facies with fine-grained sandstone
being the coarsest sediment. The main sources of the sediments
of the Abo Formation were the Uncompahgre and Zuni
uplifts to the north and the Pedernal uplift to the east of the
current location of the Mud Springs Mountains (e.g., Kues
and Giles, 2004, fig. 11). Towards the south, grain-size
decreased continuously, and the non-marine red beds of the
Abo Formation interfinger with the shallow marine carbon-
ates of the Hueco Group about 25?75 km to the south of the
Mud Springs Mountains in the southern Caballo, Don?a Ana
and Robledo Mountains.
Although the Abo Formation in this area has not been
definitively dated either biostratigraphically or chronostrati-
graphically, the Abo Formation regionally is known to be of
middle-late Wolfcampian (approximately Sakmarian-Artins-
kian) age. This is based primarily on the fact that in northern
and central New Mexico it disconformably overlies the mixed
marine-nonmarine lower Wolfcampian Bursum Formation
(e.g., Lucas and Krainer, 2004; Krainer and Lucas, 2010), and
in southern New Mexico the Abo Formation laterally
interfingers with middle-late Wolfcampian marine strata of
the Hueco Group (e.g., Lucas et al., 1995, 1998, 2011). The
Mud Springs track and plant locality is in the lower part of the
Abo Formation, so it is almost certainly of middle Wolfcam-
pian age.
Fossiliferous deposit.?The plant stems (Fig. 4.1, 4.2) occur
at the base of, and project through, a 7-cm-thick interval of
siltstone and claystone beds; approximately seven such beds
comprise the total thickness of the exposed interval (Fig. 2,
Bed 4; Fig. 3.1, at base of exposure; Fig. 3.2). Immediately
below the stem- and footprint-bearing layer is a dark claystone
surface covered with mudcracks. Neither the stems nor their
presumed roots were identifiable at that level. In the
sedimentary layers comprising the 7-cm zone of silt-clay
lamination above the base of the deposit, there is at least one
surface with raindrop imprints, indicating subaerial exposure.
Beds at the top of this interval contain isolated specimens of
Supaia thinnfeldioides foliage (Fig. 4.3). The S. thinnfeldioides
specimens, isolated leaves only, were identified by the
following features: Once compound leaf architecture, the
rachis wide, composed both of the vascular strand and thin
flanking lamina created by the confluence of adjacent
pinnules. Pinnules elongate, 4 to 5 times longer than wide,
decurrent and narrowly confluent, slightly constricted at the
base, widely spaced/clearly distinct. The only clearly visible
pinnule venation, the pinnule midvein, is decurrent, strong,
and extends nearly to the tip of the pinnule. As described by
White (1929) and in other studies of large populations (e.g.,
DiMichele et al., 2007), there is considerable morphological
and size variation in Supaia. This variation is demonstrated by
the many species described by White (1929) and by the
variation in some of these, especially S. thinnfeldioides, the
most common form, within the scope of which the Cuchillo
Negro specimens fall.
The plants and trackways are exposed on an irregularly
shaped surface with maximum dimensions of 3 by 5 m. Most
of the tree molds are found in an area of 3 by 2 m.
Approximately one half of the total surface area represented
by the maximum extent of the exposed, stem-mold bearing
beds, was missing from the outcrop due to erosion and local
mining of flagstone. These beds dip gently to the north
(Figs. 3, 4.4) and are buried by alluvium.
We interpret this deposit to have originated in periodic
floods of sediment-bearing water into a shallow depression.
This depression may represent a small, largely abandoned
shallow channel or channel margin. The depositional envi-
ronment was regularly and perhaps mostly exposed following
periodic flooding and was close enough to more permanently
exposed areas that tetrapods of various sizes could cross its
surface shortly after floodwaters subsided.
PLANT AND ANIMAL REMAINS
Fossil plants.?Thirty-four stem molds were identified on
the exposed bedding surfaces. These molds are nearly circular
in cross-sectional aspect and average 42.4 mm in diameter,
ranging from 20 mm to 70 mm (Fig. 4.1, 4.2). They are filled
with dark brown or black siltstone that contrasts with the
reddish-brown bedding plane surface on which they are
exposed. The stem molds were inferred to be at least 70 mm
in height, based on two specimens that were exposed on a
bedding surface 70 mm above the large bedding plane on
which the majority of the specimens were identified (Figs. 3,
4.4). Although these specimens were not exposed in longitu-
dinal section, they appear to project upward from below; given
the rarity of this kind of preservation, it seems reasonable to
assume that these specimens are not a second stand of stems
originating at a different level in the sediment.
Radiating concentrically from each mold are several thin,
crack-like lines filled with the darker matrix material. These
cracks probably mark the course of roots. Some could be
traced as far as 50 cm from the stem mold. They usually
followed a sinuous course through the siltstone matrix
(Fig. 4.1, 4.2). If these cracks represent the former courses of
roots, their presence around the two specimens buried in later
deposited sediment suggests that they may have formed
adventitiously from the stems and that the plants were
shallowly, but broadly rooted.
Foliage of Supaia thinnfeldioides, inferred to have been
produced by the stems, was identified on several different
bedding surfaces in this deposit. The best-preserved specimens
(Fig. 4.3) were found between the basal and the upper-most
bed in which the stem molds appear. We interpret this foliage
to have been shed from the stems. Based on studies of Supaia
foliage from other places (e.g., White, 1929; Wang, 1997),
these leaves were likely to have remained attached to the
parent plant for multiple seasons. Although relatively small,
most being 20 cm or less from base to apex, the compound,
forked Supaia leaf appears to have had a woody petiole and
thick lamina, requiring considerable metabolic investment.
The distance to nearest neighboring stem averaged 296.2 mm,
and ranged from 110 mm to 800 mm. Edge effects and the
irregular surface area introduce biases, particularly for stems
with few others in close proximity, which ought to skew the
results slightly towards greater inter-stem distances. If the
outlying measurements of 750 and 800 mm are eliminated, the
average distance between stems is reduced slightly to 266.2 mm.
The heights of these plants are estimated to fall between 2.5
and 4.5 m. This was determined from the model of Niklas and
Spatz (2004), which was derived from the empirical study of
588 JOURNAL OF PALEONTOLOGY, V. 86, NO. 4, 2012
extant woody, tree-sized plants, supplemented with data from
self-supporting herbaceous plants and juvenile woody trees.
The critical constants in this model were estimated empirically,
which is not possible for the extinct pteridosperms, particu-
larly the peltasperms, for which few whole plants are known
on which such determinations could be based. The Supaia
stems are large enough to fall within the linear, two-thirds
slope portion of the Niklas and Spatz (2004) log-log curve of
stem height versus stem diameter. Supaia may have been
physically constructed quite differently from woody trees and
self-supporting herbs, however, possibly having a pterido-
sperm-type vine-like, vascular architecture consisting of
multiple vascular bundles (e.g., Stidd, 1981). Thus, these
estimates should be treated with caution, and allowance must
be made for the plants to have been somewhat smaller than
these projections, given the growth architectures illustrated by
such studies as that of Pfefferkorn et al. (1984) or Falcon-Lang
(2009).
Vertebrate trackways.?Two very different kinds of tetrapod
footprints are preserved in concave epirelief at the Mud Springs
tracksite (NMMNH Locality 8376, Fig. 5.1?5.4, 6). Most
common are footprints that display diagnostic features of the
ichnogenus Dromopus Marsh (1894), which mostly consist of 2
to 4 nearly parallel digit marks that are long, thin, slightly curved
and have pointed tips. The best-preserved specimens (e.g.,
Fig. 5.4) show an evident increase in digit length from digits I
through IV. These lacertoid tracks are up to 30 mm long, lack
sole imprints and do not form organized trackways. Because of
this and their generally incomplete preservation, manus tracks
cannot be distinguished confidently from pes tracks.
Dromopus is one of the most common kinds of tetrapod
footprints found in the Abo Formation across New Mexico
FIGURE 4?Plant fossils and the fossiliferous layer. 1, fossil stem cast with radiating sediment-filled cracks, inferred to be the remains of adventitious
roots, scale in centimeters (black and white bars); 2, a pair of fossil stem casts with radiating sediment-filled cracks inferred to be roots, scale55 cm;
3, foliage of Supaia thinnfeldioides found on the lower bedding surface of the interval bearing the fossil stem casts, tip of mechanical pencil51 mm;
4, fossiliferous layer; stem casts originate on the lowermost layer and project through several layers of siltstone and claystone. Scale is 25 cm (yellow tape).
DIMICHELE ET AL.?VERTEBRATES AMIDST THE SUPAIA?PERMIAN OF NEW MEXICO 589
FIGURE 5?Selected tetrapod footprints at the Mud Springs locality. 1, overview of part of the temnospondyl trackway, Limnopus sp., geology
hammer528.5 cm in length; 2, 3, close-up views of selected Limnopus footprints, scale55 cm; 4, parareptile (Dromopus) footprints, finest division of scale
in millimeters. Trackway abbreviations: RP5right pes; RM5right manus; LM5left manus; LP5left pes; number refers to footprint sequence.
590 JOURNAL OF PALEONTOLOGY, V. 86, NO. 4, 2012
(e.g., Haubold et al., 1995; Lucas et al., 2005b, 2005c; Hunt
and Lucas, 2006; Minter and Braddy, 2009). It has long been
identified as the footprint of a relatively small predatory
parareptile, most likely an araeoscelid (e.g., Haubold, 1971,
1984). The ichnogenus Dromopus is widely distributed in Early
Permian red bed ichnoassemblages, especially in the western
United States, eastern Canada and western Europe (Hunt and
Lucas, 2006). Dromopus occurrences like that at the Mud
Springs tracksite, in which there are numerous poorly
preserved footprints not organized into trackways, are
characteristic occurrences in the Abo Formation. At the
Mud Springs tracksite, the Dromopus tracks show no evident
pattern, except they are more greatly concentrated in two
areas?in quadrants D-E/2-4 and B-D/7-8 (Fig. 6).
The other kind of tetrapod footprint at locality 8376 is
identified by us as Limnopus Marsh (1894). These form a
trackway in the northeastern portion of the trackway surface
that consists of at least six associated manus-pes sets and two
less clear manus(?) tracks. These tracks are relatively large for
Early Permian footprints?pes imprints are as much as 14 cm
long and manus imprints are up to 11 cm long?though size
appears to have been exaggerated by the spreading out of the
trackway imprint in the substrate in which these undertracks
are registered (Fig. 5.1?5.3). Manus imprints are directly in
front of pes imprints or are overstepped, and the few digits
discernable suggest some rotation of both manus and pes
toward the trackway midline. Because the digits are not very
well preserved, exact digit counts cannot be made for manus
or pes (at minimum there are four) and the relative lengths of
the digits cannot be determined. However, the preserved digits
are short, thick and have blunt tips. Most significant is that
they are short relative to the large, nearly round and
plantigrade sole imprints. Based on the length between manus
and pes imprints on the same side of the body, in which there
is no overstepping, the gleno-acetabular length is estimated at
,50 cm.
There are only two possible Early Permian ichnogenera to
which this large trackway could be assigned?Ichniotherium
Pohlig (1892) and Limnopus. Ichniotherium is unlikely on
morphological grounds, primarily because the Mud Springs
tracks lack any hint of the relatively long, dumbbell-shaped
digit imprints characteristic of Ichniotherium and because the
short digits of the Mud Springs tracks are only a small
proportion of the entire footprint length, which is mostly the
sole imprint. In Ichniotherium the digit imprints are longer
than the sole imprint (e.g., Voigt and Haubold, 2000; Voigt et
al., 2007). Also, in New Mexico and elsewhere Ichniotherium is
known only from inland fluvial facies (e.g., Hunt and Lucas,
2006), so its occurrence at the Mud Springs locality in distal
floodplain facies near the Hueco shoreline would be unusual.
Instead, the large trackway at the Mud Springs locality can
be assigned with confidence to Limnopus on morphological
grounds, especially the short blunt toes and large plantigrade
sole, which exemplify the ichnogenus (e.g., Haubold et al., 1995;
Hunt et al., 1995). Limnopus has long been identified as the
track of a relatively large temnospondyl amphibian (e.g.,
Haubold, 1971, 1984). It is a characteristic tetrapod footprint
found in Early Permian red-bed ichnoassemblages in the
western United States, eastern Canada and Western Europe
(Hunt and Lucas, 2006). Indeed, the Mud Springs Limnopus
FIGURE 6?Map of a portion of the fossiliferous bedding surface where vertebrate trackways were exposed most prominently. One trackway of a large
temnospondyl amphibian (Limnopus) indicates a single animal moving through the Supaia plants, the basal casts of which are shown as dark ovals.
Numerous trackways of smaller parareptiles (Dromopus) can be seen.
DIMICHELE ET AL.?VERTEBRATES AMIDST THE SUPAIA?PERMIAN OF NEW MEXICO 591
tracks are among the largest Limnopus tracks known from the
Abo Formation (cf. Haubold et al., 1995; Hunt et al., 1995),
although as noted above their size has arguably been
exaggerated somewhat by the taphonomy of their preservation.
Perhaps most interesting about the Limnopus trackway
at the Mud Springs site is its unusual gait and trackway
pattern, which suggests an amphibian that walked very slowly,
meandering through the Supaia stems (Figs. 6, 7). Compari-
son of the Mud Springs Limnopus trackway with characteristic
(?normal?) Limnopus trackways in the published literature
(Fig. 7) indicates that the trackway pattern and gait of the
Mud Springs Limnopus trackmaker (a temnospondyl amphib-
ian) was unusual. Compared to ?normal? trackways, the Mud
Springs Limnopus trackway is narrow, has a wide pace
angulation and an irregular stride with overstepping of the
manus by the pes in some manus-pes pairs.
If the larger imprints are identified as pes and smaller as
manus (as is the case in Limnopus), then the trackway can be
interpreted as two complete strides evidently affected by the
trees along its borders. Thus, the imprint identified as the
initial right pes track (RP1 in Fig. 7) strides forward to RP2
and to RP3, whereas the initial left manus/pes pair (LM1/LP1
in Fig. 7) strides forward to LM3/LP2 and LM3/LP3. This
interpretation is supported by the stride lengths on both sides
of the body, which have a fairly consistent value of ,50 cm. If
LM2 is interpreted as the second left manus imprint, then it
indicates that the left manus was placed outside of the carriage
as the temnospondyl pivoted on pes imprint LP1. Moving RP1
and RM1 to RM2/RP2, though, is difficult to interpret as it
suggests that the right manus would either have had to be
swung over the Supaia tree in quadrant A4 (Fig. 6), or moved
to the left of it. The overstepping evident in RM2/RP2 and
RM3/RP3 indicates deliberate flexure of the body and great
extension of the hindlimb to place it anterior to the manus,
which would have taken place in a slow gait.
Given the poor preservation of the Limnopus trackway at
the Mud Springs site, other interpretations are possible, but
clearly it differs substantially from a ?normal? Limnopus
trackway (Fig. 7). These differences are readily understood
as reflecting the temnospondyl trackmaker changing gait to
accommodate the obstructions posed by the Supaia trees.
DISCUSSION
The small stand of plants described here appears to be a
cohort that colonized a freshly exposed mudflat on a
floodplain bordering a watercourse or within a braidplain,
an environmentally heterogeneous setting, one that includes
significant areas of disturbance (Naiman and De?camps, 1997).
FIGURE 7?Comparison of characteristic Limnopus trackway (after Baird, 1965) and the Limnopus trackway at the Mud Springs locality. Trackway
abbreviations: RP5right pes; RM5right manus; LM5left manus; LP5left pes; number refers to footprint sequence.
592 JOURNAL OF PALEONTOLOGY, V. 86, NO. 4, 2012
The stand survived several subsequent floods and repeated influx
of thin layers of sediment. Based on associated foliage, we believe
these plants are most likely Supaia thinnfeldioides. Plant fossils
occur in numerous exposures of Abo Formation (or equivalent)
siltstone-sandstone sheet and channel-form deposits throughout
central and southern New Mexico (Hunt, 1983; MacDonald,
1994; DiMichele et al., 2007). In almost all instances, the flora
consists of either the conifer Walchia or the peltasperm Supaia in
monotypic parautochthonous to allochthonous assemblages,
subject to degrees of transport that are difficult to determine.
This is, to our knowledge, the only in situ, autochthonous stand
of plants reported from the Early Permian, not only from the
Abo Formation but from anywhere in North America.
The plants show no evidence of deep rooting, instead
appearing to have rooted adventitiously into the surrounding
sediment, including new layers of sediment that were
introduced in floods subsequent to initial plant-establishment.
This kind of environment was subject to regular flooding
disturbance of varying degrees of intensity, occasionally severe
enough to kill local vegetation, as indicated by the truncation
of the entire stand of Supaia stems by one of the flood layers.
Height estimates of approximately 4.0 m, based on allometric
calculations (Niklas and Spatz, 2004), suggest that these plants
were little more than shrubs to small trees. The morphology of
Supaia foliage, which is strongly xeromorphic (White, 1929),
suggests multi-year leaf retention, consistent with growth in a
habitat subject to extreme variations in moisture availability.
These various lines of evidence suggest an opportunistic plant
that colonized disturbed substrates, probably had a relatively
short life span, and occurred in low diversity to monotypic
stands. The channel-margin to braidplain setting provides
ample opportunity for dispersal and colonization of flood-
disturbed surfaces. What is perplexing is that more such stands
of stems or stem molds of this type have not been seen?this
may simply be a ?search image? matter, and more such
discoveries might be expected.
As has been remarked on elsewhere (Hunt, 1983; DiMichele
et al., 2007), the flora of the red-to-yellow Abo Formation
siltstone-sandstone beds is of remarkably low diversity. The
flora is best preserved in upward fining sheet siltstones, which
are the main hosts for trace fossils and macrofossils,
particularly where thin beds of siltstone are separated by clay
partings or ?drapes?. These clay-drape deposits are frequently
mudcracked and bear raindrop imprints, strong evidence of
alternating flooded and exposed conditions. The most
commonly encountered fossils are branches and branch-
fragments of conifers, most often Walchia, though Ernestio-
dendron also has been observed. These can occur as isolated
specimens or as accumulations of multiple specimens, up to
dense mats of material. Supaia foliage is more restricted in
occurrence, but it too tends to occur in isolation of any other
taxa, and vary from isolated leaves to dense accumulations.
The uppermost Abo beds have revealed rare occurrences of
other taxa such as callipterids and sphenopsids. These patterns
of widespread monodominance appear to reflect the compo-
sition of the source vegetation and thus to have been drawn
from low diversity assemblages. If most of the Supaia
thinnfeldiodes plants were colonists of disturbed streamsides
or mudflats this may explain their patchy distribution, but it
still fails to account for why they so rarely occur in mixed
assemblages with conifers. Doubly perplexing are the seem-
ingly identical depositional environments (as distinct from
environments of growth), as far as we have been able to
determine, of the deposits differentially dominated by conifers
or Supaia. Based on the sizes of the conifer branches that have
been encountered, some nearly a meter in length, it appears
that the conifers were considerably larger plants than the
small, weedy Supaia plants reported here, which may be key to
their generally distinct occurrences. Supaia may have preferred
environments in periodically disturbed habitats, too consis-
tently disturbed for conifers to survive to reproductive age.
The vertebrate trackmakers at the site are parareptiles and a
relatively large temnospondyl amphibian. The aggregation of
predatory parareptiles may reflect feeding on smaller animals
such as arthropods, occurring among the Supaia trees. The
amphibian clearly changed gait as it moved between trees.
Thus, even though the vertebrates present were not herbivores,
a case can be made for the presence of the Supaia trees
affecting their behavior, and thus providing a rare ?snapshot?
of a very localized setting of vertebrates and plants during the
Early Permian.
ACKNOWLEDGMENTS
WD thanks the National Museum of Natural History Small
Grant Fund and Smithsonian Institution Endowment funds
for support of field research. We thank D. S. Chaney, W. J.
Nelson and J. A. Spielmann for assistance with fieldwork. We
are grateful for helpful comments on an earlier draft of this
paper, provided by M. Krings, an anonymous reviewer,
Associate Editor, R. Lupia and Editor, B. Pratt.
REFERENCES
BAIRD, D. 1965. Footprints from the Cutler Formation. U. S. Geological
Survey Professional Paper, 503C:C47?C50.
BOOI, M., I. M. VAN WAVEREN, AND J. H. A. vAN KONIJNENBURG-vAN
CITTERT. 2009. Comia and Rhachiphyllum from the Early Permian of
Sumatra, Indonesia. Review of Palaeobotany and Palynology, 156:418?
435.
BUATOIS, L. A. AND M. G. MA?NGANO. 2007. Invertebrate ichnology of
continental freshwater environments, p. 285?323. In W. Miller III (ed.),
Trace Fossils: Concepts, Problems, Prospects. Elsevier, Amsterdam.
BUATOIS, L. A., M. G. MA?NGANO, J. F. GENISE, AND T. N. TAYLOR.
1998. The ichnological record of the continental invertebrate invasion:
evolutionary trends in environmental expansion, ecospace utilization,
and behavioral complexity. Palaios, 13:217?240.
CARPENTER, K. 1992. Behavior of hadrosaurs as interpreted from
footprints in the ??Mesaverde?? Group (Campanian) of Colorado, Utah,
and Wyoming. Rocky Mountain Geology, 29:81?96.
DENG, S.-H., C.-B. WAN, AND J.-G. YANG. 2009. Discovery of a Late
Permian Angara-Cathaysia mixed flora from Acheng of Heilongjiang,
China, with discussions on the closure of the Paleoasian Ocean. Science
in China Series D, Earth Sciences, 52:1746?1755.
DIMICHELE, W. A. AND H. J. FALCON-LANG. 2011. Pennsylvanian ?fossil
forests? in growth position (T0 assemblages): origin, taphonomic bias
and palaeoecological insights. Journal of the Geological Society,
168:585?605.
DIMICHELE, W. A., H. KERP, M. KRINGS, AND D. S. CHANEY. 2005. The
Permian peltasperm radiation: evidence from the southwestern United
States. New Mexico Museum of Natural History and Science Bulletin,
30:67?79.
DIMICHELE, W. A., D. S. CHANEY, W. J. NELSON, S. G. LUCAS, C. V.
LOOY, K. QUICK, AND J. WANG. 2007. A low diversity, seasonal
tropical landscape dominated by conifers and peltasperms: Early
Permian Abo Formation, New Mexico. Review of Palaeobotany and
Palynology, 145:249?273.
DIMICHELE, W. A., W. J. NELSON, S. ELRICK, AND P. R. AMES. 2009.
Catastrophically buried Middle Pennsylvanian Sigillaria and calamitean
sphenopsids from Indiana, U.S.A.: what kind of vegetation was this?
Palaios, 24:159?166.
EBERTH, D. A. AND A. D. MIALL. 1991. Stratigraphy, sedimentology and
evolution of a vertebrate-bearing, braided to anastomosed fluvial
system, Cutler Formation (Permian?Pennsylvanian), north-central New
Mexico. Sedimentary Geology, 72:225?252.
FALCON-LANG, H. J. 2009. A Macroneuropteris scheuchzeri tree preserved
in growth position in the Middle Pennsylvanian Sydney Mines
Formation, Nova Scotia, Canada. Atlantic Geology, 45:74?80.
FALCON-LANG, H. J., N. A. JUD, W. J. NELSON, W. A. DIMICHELE, D. S.
CHANEY, AND S. G. LUCAS. 2011. Pennsylvanian coniferopsid forests in
DIMICHELE ET AL.?VERTEBRATES AMIDST THE SUPAIA?PERMIAN OF NEW MEXICO 593
sabkha facies reveal the nature of seasonal tropical biome. Geology,
39:371?374.
GALTIER, J. AND J. BROUTIN. 2008. Floras from red beds of the Permian
Basin of Lode`ve (Southern France). Journal of Iberian Geology, 34:57?72.
GAND, G., H. KERP, C. PARSONS, AND E. MARTI?NEZ-GARCI?A. 1997.
Palaeoenvironmental and stratigraphic aspects of animal traces and
plant remains in Spanish Permian red beds (Pen?a Sagra, Cantabrian
Mountains, Spain). Geobios, 30:295?318.
GASTALDO, R. A, I. M. STEVANOVIC-WALLS, AND W. N. WARE. 2004.
Erect forests are evidence for coseismic base-level changes in
Pennsylvanian cyclothems of the Black Warrior Basin, U.S.A., p.
219?238. In J. C. Pashin and R. A. Gastaldo (eds.), Sequence
Stratigraphy, Paleoclimate, and Tectonics of Coal-Bearing Strata.
A.A.P.G. Studies in Geology, 51.
HAUBOLD, H. 1971. Ichnia Amphibiorum et Reptiliorum fossilium.
Encyclopedia of Paleoherpetology, 18:1?124.
HAUBOLD, H. 1984. Saurierfa?hrten. Neue Brehm-Bucherei, 231 p.
HAUBOLD, H., A. P. HUNT, S. G. LUCAS, AND M. G. LOCKLEY. 1995.
Wolfcampian (Early Permian) vertebrate tracks from Arizona and New
Mexico. New Mexico Museum of Natural History and Science Bulletin,
6:135?165.
HUNT, A. 1983. Plant fossils and lithostratigraphy of the Abo Formation
(lower Permian) in the Socorro area and plant biostratigraphy of Abo
red beds in New Mexico. New Mexico Geological Society Guidebook,
34th Field Conference, Socorro Region II, p. 157?163.
HUNT, A. P. AND S. G. LUCAS. 2006. Permian tetrapod ichnofacies.
Geological Society of London, Special Publication, 265:137?156.
HUNT, A. P., S. G. LUCAS, H. HAUBOLD, AND M. G. LOCKLEY. 1995.
Early Permian (late Wolfcampian) tetrapod tracks from the Robledo
Mountains, south-central New Mexico. New Mexico Museum of
Natural History and Science Bulletin, 6:167?180.
JOHNSON, K. R. 2007. Palaeobotany: forests frozen in time. Nature,
447:786?787.
KRAINER, K. AND S. G. LUCAS. 2009. Cyclic sedimentation of the upper
Carboniferous Bursum Formation, central New Mexico: tectonics
versus glacioeustasy. New Mexico Geological Society Guidebook,
60:167?182.
KRAINER, K. AND S. G. LUCAS. 2010. Sedimentology of the Pennsylva-
nian?Permian Cutler Group and Lower Permian Abo Formation,
Northern New Mexico, p. 25?36. In S. G. Lucas, J. W. Schneider, and J.
A. Spielmann (eds.), Carboniferous?Permian Transition in Can?on del
Cobre, northern New Mexico. New Mexico Museum of Natural
History and Science Bulletin, 49.
KUES, B. S. AND K. A. GILES. 2004. The late Paleozoic ancestral Rocky
Mountains system in New Mexico, p. 95?136. In G. H. Mack and K. A.
Giles (eds.), The Geology of New Mexico: A Geologic History. New
Mexico Geological Society Special Publication 11.
LUCAS, S. G. AND K. KRAINER. 2004. The Red Tanks Member of the
Bursum Formation in the Lucero uplift and regional stratigraphy of the
Bursum Formation in New Mexico. New Mexico Museum of Natural
History and Science Bulletin, 25:43?52.
LUCAS, S. G., O. J. ANDERSON, A. B. HECKERT, AND A. P. HUNT. 1995.
Geology of Early Permian tracksites, Robledo Mountains, south-
central New Mexico. New Mexico Museum of Natural History and
Science Bulletin, 6:13?32.
LUCAS, S. G., A. B. HECKERT, J. W. ESTEP, AND C. W. COOK. 1998.
Stratigraphy of the Lower Permian Hueco Group in the Robledo
Mountains, Don?a Ana County, New Mexico. New Mexico Museum of
Natural History and Science Bulletin, 12:43?54.
LUCAS, S. G., K. KRAINER, AND R. M. COLPITTS. 2005a. Abo-Yeso
(Lower Permian) stratigraphy in central New Mexico, p. 101?115. In
S. G. Lucas, K. E. Zeigler, and J. A. Spielmann (eds.), The Permian of
Central New Mexico. New Mexico Museum of Natural History and
Science Bulletin 31.
LUCAS, S. G., N. J. MINTER, J. A. SPIELMANN, A. P. HUNT, AND S. J.
BRADDY. 2005b. Early Permian ichnofossil assemblage from the Fra
Cristobal Mountains, southern New Mexico, p. 140?150. In S. G.
Lucas, K. E. Zeigler, and J. A. Spielmann (eds.), The Permian of
Central New Mexico. New Mexico, Museum of Natural History and
Science Bulletin 31.
LUCAS, S. G., N. J. MINTER, J. A. SPIELMANN, J. A. SMITH, AND S. J.
BRADDY. 2005c. Early Permian ichnofossils from the northern Caballo
Mountains, Sierra County, New Mexico, p. 151?162. In S. G. Lucas,
K. E. Zeigler, and J. A. Spielmann (eds.), The Permian of Central New
Mexico, New Mexico, Museum of Natural History and Science Bulletin 31.
LUCAS, S. G., K. KRAINER, L. CORBITT, J. DIBENEDETTO, AND
D. VACHARD. 2011. The Trans Mountain Road Member, a new
stratigraphic unit of the Lower Permian Hueco Group, northern
Franklin Mountains, Texas, p. 93?109. In R. M. Sullivan, S. G. Lucas,
and J. A. Spielmann (eds.), Fossil Record 3: New Mexico Museum of
Natural History and Science Bulletin.
MACDONALD, J. 1994. Earth?s First Steps: Tracking Life Before the
Dinosaurs. Johnson Books, Boulder, Colorado, 290 p.
MARSH, O. C. 1894. Footprints of vertebrates in the Coal Measures of
Kansas. American Journal of Science, 3rd Series, 283:81?84.
MIALL, A. D. 1996. The Geology of Fluvial Deposits. Sedimentary Facies,
Basin Analysis, and Petroleum Geology. Springer, Berlin, 582 p.
MINTER, N. J. AND S. J. BRADDY. 2009. Ichnology of an Early Permian
tidal flat: the Robledo Mountains Formation of the Robledo
Mountains, Southern New Mexico, U.S.A. Special Papers in Palaeon-
tology, 82:1?107.
NAIMAN, R. J. AND H. DE?CAMPS. 1997. The ecology of interfaces: riparian
zones. Annual Review of Ecology and Systematics, 28:621?658.
NIKLAS, K. J. AND H.-C. SPATZ. 2004. Growth and hydraulic (not
mechanical) constraints govern the scaling of tree height and mass.
Proceedings of the National Academy of Sciences of the United States
of America, 101:15661?15663.
OPLUS?TIL, S., J. PS?ENIC?KA, M. LIBERTI?N, A. R. BASHFORTH, Z. S?IMU?NEK,
AND J. DAS?KOVA?. 2009. A Middle Pennsylvanian (Bolsovian) peat-
forming forest preserved in situ in volcanic ash of the Whetstone
Horizon in the Radnice Basin, Czech Republic. Review of Palaeobotany
and Palynology, 155:234?274.
PARKER, L. R. AND J. K. BALSLEY. 1989. Coal mines as localities for
studying dinosaur trace fossils, p. 353?360. In D. D. Gillette and M. G.
Lockley (eds.), Dinosaur Tracks and Traces. Cambridge University
Press, Cambridge, U.K.
PFEFFERKORN, H. W. AND J. WANG. 2007. Early Permian coal-forming
floras preserved as compressions from the Wuda District (Inner
Mongolia, China). International Journal of Coal Geology, 69:90?102.
PFEFFERKORN, H. W., W. H. GILLESPIE, D. A. RESNICK, AND M. H.
SCHEIHING. 1984. Reconstruction and architecture of medullosan
pteridosperms (Pennsylvanian). The Mosasaur (Delaware Valley
Paleontological Society), 2:1?8.
POHLIG, H. 1892. Altpermische Saurierfa?hrten, Fische und Medusen der
Gegend von Friedrichroda i. Thu?ringen, p. 59?64. In Anonymous (ed.),
Festschrift zum 70. Geburtstag von Rudolf Leuckardt, Engelmann,
Leipzig, Germany.
READ, C. B. AND S. H. MAMAY. 1964. Upper Paleozoic floral zones and
floral provinces of the United States. U.S. Geological Survey
Professional Paper, 454K:1?35.
STIDD, B. M. 1981. The current status of medullosan seed ferns. Review of
Palaeobotany and Palynology, 32:63?101.
VOIGT, S. AND H. HAUBOLD. 2000. Analyse zur Variabilita?t der
Tetrapodenfa?hrte Ichniotherium cottae aus dem Tambacher Sandstein
(Rotliegend, U-Perm, Thu?ringen). Hallesches Jahrbuch Geowissenschat
B, 22:17?58.
VOIGT, S., D. S. BERMAN, AND A. C. HENRICI. 2007. First well-
established track-trackmaker association of Paleozoic tetrapods based
on Ichniotherium trackways and diadectid skeletons from the lower
Permian of Germany. Journal of Vertebrate Paleontology, 27:553?570.
WANG, Z. Q. 1997. Permian Supaia fronds and an associated Autunia
fructification from Shanxi, China. Palaeontology, 40:245?277.
WHITE, D. 1929. Flora of the Hermit Shale, Grand Canyon, Arizona.
Carnegie Institution of Washington Bulletin, 405:1?221.
WING, S. L., L. J. HICKEY, AND C. C. SWISHER. 1993. Implications of an
exceptional fossil flora for Late Cretaceous vegetation. Nature,
363:342?344.
ACCEPTED 2 MARCH 2012
594 JOURNAL OF PALEONTOLOGY, V. 86, NO. 4, 2012