BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.
PRESERVATION OF TRACE FOSSILS AND MOLDS OF TERRESTRIAL
BIOTA BY INTENSE STORMS IN MID?LAST INTERGLACIAL (MIS
5c) DUNES ON BERMUDA, WITH A MODEL FOR DEVELOPMENT OF
HYDROLOGICAL CONDUITS
Author(s): PAUL J. HEARTY and STORRS L. OLSON
Source: Palaios, 26(7):394-405. 2011.
Published By: Society for Sedimentary Geology
URL: http://www.bioone.org/doi/full/10.2110/palo.2010.p10-132r
BioOne (www.bioone.org) is an electronic aggregator of bioscience research content, and the online home to over
160 journals and books published by not-for-profit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne?s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
PALAIOS, 2011, v. 26, p. 394?405
Research Article
DOI: 10.2110/palo.2010.p10-132r
PRESERVATION OF TRACE FOSSILS AND MOLDS OF TERRESTRIAL BIOTA BY INTENSE STORMS
IN MID?LAST INTERGLACIAL (MIS 5c) DUNES ON BERMUDA, WITH A MODEL FOR
DEVELOPMENT OF HYDROLOGICAL CONDUITS
PAUL J. HEARTY1* and STORRS L. OLSON 2
1Department of Environmental Studies, University of North Carolina at Wilmington, North Carolina 28403, USA; 2Department of Vertebrate Zoology, National
Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560, USA
e-mail: heartyp@uncw.edu
ABSTRACT
We document massive deposition of carbonate sands along the south shore
of Bermuda that were emplaced during one or two great storms during the
last interglacial. As determined by their stratigraphic position and
geochronological data, these deposits formed during marine isotope
substage (MIS) 5c ca. 100 ka ago. Within a leeward set of eolian beds,
evidence of a living landscape was preserved that includes delicate
footprints of a shorebird (Scolopacidae, Catoptrophorus) preserved in
frothy dune foreset beds. In the same stratigraphic unit, outlines of a
standing forest of palm trees (Sabal bermudana), some evidently with
fronds in place, were molded in the fine carbonate dune sand. All available
evidence points to an MIS 5c sea level positioned several meters below the
present datum, which would require great intensity of storms to transport
such voluminous deposits well above present sea level. Waves and storm
currents transported loose sediments from the shallow shelf onto the
shore, where hurricane winds piled up sand sufficiently deep to bury
established forests of 8- to 10-m-tall trees. Evidence of such powerful
storms preserved in the rock record is a measure of the intensity of past
hurricanes, and a possible bellwether of future storm events. Entombment
of the trees involved rapid burial and cementation creating external molds
in limestone, a process that is important in the development of vertical
hydrological conduits commonly observed in eolianites.
INTRODUCTION
In an open ocean setting 1040 km ESE of Cape Hatteras, North
Carolina (Fig. 1), the subaerial portion of the Bermuda platform is fully
exposed to a variety of storm systems, including westerly trade winds,
Western Atlantic storms (so-called nor?easters), and tropical hurri-
canes. The literature on intense storms during the Quaternary generally
centers on overwash of coarse sediments into back-barrier environ-
ments during the mid to late Holocene (e.g., Donnelly et al., 2001; Nott,
2004; Stone and Orford, 2004 and papers therein; Mann et al., 2009).
The burial of trees in rapidly accreting coastal carbonate dunes
(eolianites) may also create long-lasting vertical conduits in limestone.
Elsewhere these have been known to develop into highly indurated
columns composed of mixed materials; a process we describe in a
model.
Bermuda is a relatively small (56 km2) oceanic island that preserves
an extensive record of interglacial coastal sedimentation, modulated by
sea-level changes throughout the Quaternary. The island formations
mapped by Vacher et al. (1989) were later correlated with marine
isotope stages (MIS) through independent U/Th dating (Harmon et al.,
1983; Ludwig et al., 1996) and aminostratigraphic techniques (Hearty et
al., 1992, 2004; Hearty and Olson, 2010), with some subsequent minor
modifications of the stratigraphic column (Hearty, 2002; Table 1). A
litho-, chrono-, and biostratigraphic framework based on limestones
and fossils (Hearty et al., 1992; Hearty and Olson, 2010) provides a
reliable means to correlate and date most of the deposits from the
island.
Interglacial highstands are represented in Bermuda by subtidal and
intertidal marine facies marking the approximate position of ancient
shorelines, and wind-deposited eolianite facies farther inland. Interca-
lated and oxidized red or orange terra rossa soils developed during
glacially lowered stands of sea level (Vacher et al., 1995). During pauses
in dune accretion, perhaps coinciding with minor sea-level regressions
during interglacials, tan to gray protosols (Vacher and Hearty, 1989)
formed when plants and animals actively colonized dune surfaces.
These interdune intercalations may contain rhizoliths (sensu Kraus and
Hasiotis, 2006), land snails, invertebrate, and vertebrate fossils.
Bermuda has a rich paleontological record, with hard parts of
vertebrates, including birds, reptiles, bats, and numerous terrestrial
invertebrates (e.g., land snails and crabs) preserved in a wide variety of
sedimentary environments (e.g., eolianite, protosols, foreshore deposits,
cave and fissure fills, pond sediments), ranging in age from early to
middle Pleistocene to late Holocene (Olson et al., 2005). While both
continental and marine birds are known from abundant bones and even
fossilized eggs (Olson and Hearty, 2003; Olson et al., 2005), bird
trackways have not been previously documented from Bermuda,
although plant impressions are noted in the literature (e.g., Verrill,
1902; Livingston, 1944; Vacher and Rowe, 1997).* Corresponding author.
FIGURE 1?Located 1040 km ESE of Cape Hatteras, North Carolina (inset),
Bermuda has a current land area of 56 km2. Study sites along the South Shore of
Bermuda (individual sections in Fig. 3) are as follows: GB 5 Grape Bay; SC 5
Southcourt Ave., HB 5 Hungry Bay, AS 5 Ariel Sands; RB 5 Rocky Bay (type
section), SH 5 Saucos Hill, SP 5 Spittal Pond, and CH 5 Charles Island.
Copyright g 2011, SEPM (Society for Sedimentary Geology) 0883-1351/11/0026-0394/$3.00
Dune accretion and island growth occurred asymmetrically on the
Bermuda platform (Fig. 1). As the broad North Lagoon effectively de-
energizes wave intensity along the north shore, significant sediment flux
and shoreline deposition from the north occurred only during shifts to
significantly higher-than-present sea levels, before accommodation
could be filled, such as during late MIS 5e and 11 (Hearty et al.,
1999, 2007; Olson and Hearty, 2009). Furthermore, prevailing winds
and the source of the strongest storms are mainly from the southern
quadrants. Hence, cumulative land accretion, mostly through dune
deposition, has been greater on the stormy and higher energy southern
exposure, accounting for the present-day asymmetrical orientation of
the Bermuda islands (56 km2) on the 650 km2 platform (Fig. 1).
Importance of Storms in Preservation of Trace Fossils and
External Molds
On Bermuda, storm events have clearly played an important role in
the formation and preservation of trace fossils and the impressions of
plants through rapid and voluminous sedimentation, as opposed to
slowly accreting dunes that would be constantly bioturbated by plants
and animals that largely destroy primary eolian physical sedimentary
structures (bedding), along with any traces of organisms that may have
been present. Such heavily bioturbated sands are common as protosols
within eolianites. On the other hand, exceptional circumstances are
necessary to preserve continuous sets of foresets in 3-dimensional dune
architecture. Conditions must prevail under which deposition is too
rapid to allow reworking and destruction of dune bedding by physical
processes or bioturbation. We, therefore, view extensive, uninterrupted
sets of foreset bedding as an indication, not of decadal-type hurricanes,
but extreme events that might occur only a few times during an
interglacial period.
In Bermuda, several authors interpreted sediment- or soil-filled pipes
as fossilized palmetto stumps (Verrill, 1902; Livingston, 1944; Vacher
and Rowe, 1997), although Herwitz (1993) and Herwitz and Muhs
(1995) reinterpreted these as being attributable to stemflow driven, top-
down limestone dissolution that formed vertical conduits. Below, we
revive the original bottom-up interpretation of palmetto stumps as the
external molds of tree trunks (Hearty, 2002) that were later filled with a
mixture of sediment types. Curran and White (2001) and Curran et al.
(2008) investigated the extent of terrestrial trace and body fossils on
several islands in the Bahamas, while Kindler et al. (2008) described
bird trackways in Eleuthera.
Additional evidence in the Bahamas for great storms and rising
seas at the end of MIS 5e comes from vast chevron ridges over
flatlands of several islands in which extensive oolitic sands entombed
palmettos and standing trees (Neumann and Hearty, 1996). Also, late
in MIS 5e, waves catapulted giant boulders onto present 20-m-high
ridges on open-ocean coasts (Hearty, 1997). Although deposition by
tsunami cannot be excluded, Hearty and Neumann (2001), based
primarily on the internal structure of the chevron ridges, suggested
that storm waves were the most likely process concurrently depositing
both the boulders and chevron ridges. Some authors contend that the
boulders, now at the crest of a complex middle and late Pleistocene
eolianite ridge (Hearty, 1997), rolled down from an easterly ridge and
came to rest on younger units, or alternatively, that the boulders are
tower karst, peaks of limestone left as relicts from deep karst erosion
of a higher eolianite ridge (Mylroie, 2008, p. 71 and 73, and his fig.
13). There is no evidence, however, of a hypothetical elevated
landscape to cause such rolling, or from which the karst towers could
have been fashioned. Furthermore, old and high rocks as hypothe-
sized would not fit stratigraphically or diagenetically into the geologic
context of the site. Primary dune morphology is preserved even in the
oldest (MIS 11) of the eolianite deposits (Hearty, 1998; Hearty et al.,
1999) and there is no evidence of deep surficial karst erosion in
the area.
Eolianites in Bermuda
Carbonate eolianites of Bermuda are composed of skeletal fragments
of once-living organisms that include marine taxa inhabiting the
shallow shelf such as mollusks, corals, foraminifera, calcareous algae,
and echinoderms. These fragments are comminuted by various
processes and reduced to sand, transported to the shore by currents,
and into dunes farther landward by wind. As dunes become inactive
and colonized by organisms, induration takes place rapidly over tens to
hundreds of years. Massive seas and intense winds episodically deposit
sediments well above mean sea level because Bermuda lies in the path of
major western Atlantic storms. Consequently, a large proportion of the
present-day Bermuda is formed of wind-deposited calcareous dunes.
Exposed lee-side foreset bedding and lower-angle eolian bed forms are
TABLE 1?Correlation table of nomenclature associated with the stratigraphy of Bermuda.
Vacher et al. (1989); Hearty et al. (1992) Hearty (2002); Hearty and Olson (2010) [sedimentary facies] Hearty et al. (2004) (MIS) correlation
Recent Recent Unit uv to z Admiral?s Cave 1?4
Southampton Fm. Southampton Fm. [eolianite and multiple protosols] Unit u; Wilkinson Q High fissure 5a
Hungry Bay Fm. (former Pembroke dune) [eolianite, protosol,
plant body fossils, and bird tracks]
5c
Harrington soil [weak reddened paleosol] 5e/5c
Pembroke dune Devonshire marine (max. + 6?9 m), including Spencer?s Pt.
[foreshore, eolianite, multiple protosols] Rocky Bay Formation
Unit s/t Late 5e
Harrington soil Mid-5e regression [protosol] Unit r/s Middle 5e
Devonshire marine or dune Grape Bay Mb (former Belmont marine (+2.5 m) of Rocky Bay
Formation) [marine, eolianite, ??palmetto stumps?? in upper
levels]
Unit p/q/r Early 5e
Shore Hills Geosol Red geosol (?) [complex paleosol] Unit o/,p Crane fauna Late 6 to early 5e
Belmont marine and dune (+2.5 m) Harvey Rd Q [eolianite] 7 (now 5e)
Ord Rd. Geosol Ord Rd. Geosol [complex paleosol] Middle Pleistocene 8
U. Town Hill Upper Town Hill [eolianite with multiple protosols] 9
Paleosol Harbour Road Geosol [red to orange paleosol] 10
L. Town Hill Lower Town Hill [foreshore, eolianite, multiple protosols] 11
Big Red soil (BRS): Castle Harbour
geosol [deep red, thick, clayey
paleosol in solution pipes in
Walsingham eolianite]
12?26?
Walsingham Fm. Walsingham Fm. [eolianite with multiple protosols] Early Pleistocene 27/35?
PALAIOS BERMUDA TRACE FOSSILS 395
oriented in all compass directions (Mackenzie, 1964a, 1964b; Vacher et
al., 1989; 1995).
Aiding the energetic deposition of marine sediments and dunes on the
south shore is the narrow shelf and steeply inclined ramp. A water
depth of 2100 m occurs only 800 m offshore of Devonshire and Rocky
Bays. With a steep ramp on the south, massive storm events, even
during lower-than-present sea levels, seas could potentially well up and
transport sediment tens of meters above the modern datum.
During intense storms, evidence of land organisms and traces of their
activity may be buried by voluminous sediment fluxes, providing
glimpses into the position and even activity of terrestrial biota. Here we
provide details of such a massive depositional event (or events) during
the last interglacial period, and reconstruct the environmental
conditions that led to the preservation of trace fossils and tree molds,
and how the latter initiated the formation of vertical conduits in coastal
carbonate dunes.
RESULTS
Stratigraphy, Age, and Geographic Range of the Sedimentary Deposits
External molds of plants occur in the last interglacial sedimentary
complex (MIS 5, sensu lato, 130?75 ka) comprising the Rocky Bay,
Hungry Bay (see discussion in Hearty, 2002), and Southampton
Formations (Table 1; Fig. 2). Large plant impressions occur in several
marine and eolian units, but are most common in eolianites positioned
stratigraphically above the Harrington soil and Rocky Bay marine units
(early to mid MIS 5) and below the Southampton Fm (late MIS 5).
Several deposits of this stratigraphic age are located along the South
Shore between Grape Bay (Fig. 1) in the west, to Charles Island in the
east, and at the mouth of Castle Harbour (Fig. 3).
Sedimentology and Aminostratigraphy of the Enclosing Sediments
The fine bioclastic sand enclosing the trace fossils and molds shows
little or no diagenesis and minimal grain-contact meniscus cementation,
appropriate for its vadose setting in these stratigraphically young
eolianites. Primary colors?such as in the pink foraminiferan Homo-
treme rubrum and in various sedimentary components of other marine
invertebrates?and textures of grains are clear and vivid. This unaltered
condition suggests that the recrystallization and dissolution typical of
karstic formations that include solution conduits have not had time to
form and thus cannot readily account for the voids that we interpret as
external molds of tree trunks.
Whole-rock (WR) D-alloisoleucine/L-isoleucine (A/I) data represent
the duration of MIS 5 range from 0.41 (GBF) to 0.27 (HBF), whereas
A/I values for the faster epimerizing land snail Poecilozonites range
from 0.60 (GBF) to 0.40 late MIS 5a (SHF) (Hearty et al., 1992 for a
review of the methodology and results; Hearty, 2002; Hearty and
Olson, 2010) (Fig. 2; Table 2). Poecilozonites from MIS 5 Harrington
Member have a mean and standard deviation of 0.49 6 0.03 (N 5 36).
The bioclastic eolianite containing trace fossils and molds yields a
small range of WR A/I values centered on 0.27 6 0.03 (15) (Table 2).
Along with the conspicuous sedimentary facies and lithology, these A/I
values correlate with the eolianite at numerous study sites across the
southern shoreline (Fig. 3). This lithologic unit correlates with the
Hungry Bay Formation (HBF), formerly the Pembroke Member of the
Rocky Bay Formation (Table 1; Fig. 3) (see discussion in Hearty,
2002). The Charles Island deposits yield an equivalent WR mean of 0.27
6 0.01 (N 5 2). Poecilozonites shells (0.45 6 0.05; N 5 5) from Unit II
FIGURE 3?Stratigraphic correlation of study sites along the south shore of Bermuda. The Hungry Bay Formation (HBF; former Pembroke Member) is identified across the
South Shore sections with thick dashed lines. Site abbreviations are defined in Figure 1. Stratigraphic abbreviations: SG.s 5 St. Georges soil; S 5 Southampton Formation;
HBF 5 Hungry Bay Formation; Hp 5 Harrington protosol; Dm 5 Devonshire Member (marine); De 5 Devonshire Member (eolianite); GBm 5 Grape Bay Member (marine).
FIGURE 2?Composite stratigraphic column, nomenclature, and associated amino
acid racemization values from the last interglacial of Bermuda. A) composite South
Shore stratigraphy (key to symbols in Fig. 3); B) stratigraphic nomenclature from
Hearty (2002); C) whole-rock A/I from Hearty et al. (1992) and Hearty and Olson
(2010); D) A/I from Poecilozonites land snails (Hearty et al., 1992; Hearty and Olson,
2010); E) Marine Isotope Stages (MIS).
396 HEARTY AND OLSON PALAIOS
of the HBF deposits (Fig. 2; Table 2) correlate with Poecilozonites from
the Southampton Formation (0.40 6 0.04 (N 5 36) (Hearty et al.,
1992), which, along with marine facies, are correlated with MIS 5a
(Vacher and Hearty, 1989; Ludwig et al., 1996; Dorale et al., 2010). The
Harrington Member postdates MIS 5e (120 ka) and is correlated with a
sea level regression during late MIS 5e, 5d and perhaps into MIS 5c
(Hearty, 2002). Thus, the preferred correlation of the HBF, including
the former Pembroke Member and Charles Island deposits, is with MIS
5c around 100 ka ago, based on stratigraphy and AAR values.
The extension of foreset bedding below the present datum and
absence of any washover or stillstand marine facies above the
Harrington soil (until late Southampton time) indicates a sea level
somewhat lower than the present one at the time of deposition of both
Units I and III of the HBF. Sea level was well above the platform
margin, however, as platform flooding is required for substantial
bioclastic sediment formation and subsequent transport to the shore
(, 220 m). These observations are supported by previous studies on
Bermuda that identified a sea level during MIS 5c at , 210 to 215 m
(Vollbrecht, 1990; Hearty, 2002).
We maintain that all the trace fossil? and mold-bearing deposits we
describe in Bermuda are the result of powerful and sustained storms
(see also Olson and Hearty, 2009). This view is based on the volume and
lack of complexity of the eolian deposits, the sedimentary conditions of
burial, and exquisite preservation of delicate footprints of birds.
Such a massive sediment flux indicates powerful and prolonged
storms accompanied by a large and constant supply of available
sediment, probably late in the highstand cycle. Extreme storm
conditions blur the interface between air, sea, and sand and result in
deposition of wet, air-filled sediment, cascading over the crest and into
the lee of coastal dunes. In this case, we describe extreme storm deposits
from Bermuda characterized by extensive, uninterrupted sets of
slumping eolian foresets composed of poorly sorted and high-porosity
sediment that rapidly buried terrigenous landscapes.
Trace Fossils and External Molds Preserved by Intense Storms During
MIS 5
Description of Bird Tracks.?A trackway was preserved on the
bedding planes of a huge tilted slab (23u?27u) of indurated calcarenite
that was detached from the outcrop on Charles Island (CH on Fig. 1)
during the high seas generated by Hurricane Fabian on 5 September
2003. Four stratigraphic units are exposed (Fig. 4): two thick eolianites,
each capped by weak soils, numbered I?IV from base to top of the
section. The trackway occurs midway in a lenticular eolianite of Unit
III (thin arrow Fig. 4) that varies laterally in thickness between 2 and
7 m.
In its present position, the tabular bed imprinted with the trackway is
nearly vertical, but when in place would have been inclined at ,25u in
foreset dune beds near the top of the exposure. The surface is that upon
which the bird actually walked, not a counterslab. Olson visited the
locality on 30 April 2004 and took photos, notes, and measurements.
Olson and Hearty returned to the site on 21 February 2005 and made
further observations and photographs and collected additional samples
for dating. Unfortunately, due to the unstable and precarious position
of the fallen block, images of the trackway from a vertical perspective
could not be obtained.
The trackway consists of 7 or 8 visible 3-toed footprints, obviously
avian, extending ,0.5 m in a relatively straight line (Fig. 5). There was
no imprint of the hallux (hind toe) or the metatarsal pad and lengths of
the toes were extrapolated from the estimated point of convergence of
the three toes. The stride?distance between tracks made by the same
foot?is ,14 cm, making the step (distance between tracks made by
alternate feet) ,7 cm. The length of the middle toe (and thus the total
length of the track) was 33.5 mm and the lateral toes ,29 mm. The
widths of the tracks were ,40 mm. Some prints preserved faint
impressions of the edge of a web between the toes. This did not extend
to the tips of toes, however, indicating a semipalmate condition of the
foot. Other possible footprints of the same size are on the same surface
as the trackway, but these are so obscure that absent the distinct set of
tracks, their possible identity as footprints would probably have gone
unrecognized.
Interpretation of the Bird Tracks.?The size and spacing of the
footprints indicate a medium-large wading or shore bird. The
semipalmate webbing is a highly diagnostic trait. Such birds as gulls
would be fully webbed, whereas herons, ibises, and such large
shorebirds as godwits (Limosa) and curlews (Numenius) would show
little to no webbing (Ray, 1678). The willet (Catoptrophorus semi-
palmatus), among the larger species of North American shorebirds, as
its specific name implies, has the toes only partly webbed (Coues, 1903,
p. 137, fig. 49), very much as in the Bermuda trackway.
The size and spacing of the tracks are close to what would be
expected for a willet. Length measurements (mm) of the toes of four
fluid-preserved specimens were as follows: digit III 37?39 (10%?16%
larger than tracks); digit II 30?32 (0.3%?10% larger); digit IV 31?35
(10%?20% larger). The discrepancy with the tracks could arise from
imprecision in knowing from where to measure at the posterior end of
the toe in the museum specimens (i.e., which part of the foot actually
touches the sediment surface) and also from possible shrinkage of the
TABLE 2?Amino acid data from whole-rock (WR) and Poecilozonites land snails
from Charles Island. See Figure 2 for stratigraphic position and aggregate mean
values for associated deposits.
AAL# Field # Material 1st run 2nd run Mean St. dev.
Unit I (A) and III (B)
4359 A OCI-04 WR 0.285 0.275 0.280 0.007
4359 B OCI-04 WR 0.260 0.271 0.266 0.008
Mean 0.273 0.009
Unit II
4588 UCH1b Land snail 0.398 0.395 0.397 0.002
4589 UCH1b Land snail 0.466 0.464 0.465 0.001
4590 UCH1b Land snail 0.486 0.482 0.484 0.003
4591 UCH1b Land snail 0.390 0.389 0.390 0.001
4592 UCH1b Land snail 0.479 0.490 0.490 0.008
Mean 0.445 0.048
AAL# 5 Amino Acid Laboratory number at Northern Arizona University (D.
Kaufman, Director); Field # 5 field number; St. dev. 5 1 standard deviation (1s).
FIGURE 4?Landward (northward) view of south side of Charles Island showing
stratigraphy and major units. Approximate stratigraphic level of bird trackways
indicated by thin black arrow. Possible casts of standing trees identified by
larger arrows.
PALAIOS BERMUDA TRACE FOSSILS 397
wet sand of the tracks upon drying. The fossil tracks were quite similar
in overall preservation?imprints of three anterior toes but no hallux or
metatarsal pad?to those shown by Elbroch and Marks (1999, p. 126)
for a willet walking in mud.
In walking birds, the femur is held horizontally and is essentially
fixed, so that forward movement involves swinging at the knee joint
(Gatesy, 1995), with the length of the stride becoming a function of the
length of the tibiotarsus; thus, as S.L. Olson (unpublished data, 1990)
has determined from field studies, the length of a step in an avian
trackway is roughly equivalent to the length of the bird?s tibiotarsus.
Willets vary in size geographically (western birds are larger) and
sexually (females larger) as well as individually (Lowther et al., 2001).
The length (mm) of the tibiotarsus (from the proximal articular surface
to the distal condyles) in skeletal specimens of willets was as follows:
males 70.1?83.3 (mean 75.5, n 5 7; females 79.1?87.8 (mean 84.1, n 5
13). Thus, the length of the step in the Bermuda trackway (,70 mm)
would fit reasonably well with a willet, particularly one walking slowly.
Elbroch and Marks (1999, p. 126) give somewhat higher values,
although they mistakenly use the word stride when the word step, as
defined above, was intended.
FIGURE 5?Montage of trackway views in Hungry Bay Formation at Charles Island, Bermuda, St. Georges Parish, south side (mouth) of Castle Harbour located at 32u
20.3129 N; 64u 40.3069 W. A) exposure on north side of Charles Island showing eolianite slab with the bird trackways. Oval indicates standing person partially hidden by rock.
Thin black arrow in A indicates approximate stratigraphic level of the slab before collapse. Apparent tree molds are indicated by larger arrows. B?D) General and more detailed
views of the trackway (2003 photos).
398 HEARTY AND OLSON PALAIOS
Willets occur in Bermuda only as rather uncommon but regular
migrants, almost entirely in fall migration with peak abundance from
July through September (Amos, 1991), generally coinciding with the
early half of the North Atlantic hurricane season. In migration they
inhabit beaches and shorelines and could be expected to take refuge
from storms by walking to the lee side of beach dunes.
The trackway is preserved in tabular beds composed of medium to
fine bioclastic sand of Unit III on foreset bedding dipping to the north
at ,23u?27u, very near the angle of repose (Fig. 5B). The bird trackway
traverses the foreset bed, climbing obliquely at an angle equally
dividing strike and dip. The beds were clearly wet and of fine sediment
texture, as preservation of detailed foot impressions (Figs. 5C?D)
would not have been possible in dry or coarser sediment, particularly
when inclined near the angle of repose. If the sand were dry when
imprinted, no detail would be preserved. If the trackway were exposed
for days or weeks, the sand would dry and run on a 25u slope, and detail
would be lost or completely obliterated. To preserve these details,
imprint, burial, and preservation must have been nearly synchronous.
Jackson et al. (2010) have shown that moisture content of the
medium (5substrate) has a great effect on the preservation of tracks,
with the best preservation being in moist, as opposed to dry or
saturated, sand. From the detailed sole or flow marks on the foreset
beds (e.g., Fig. 5B), the wet sand was apparently shifting and sliding
during the height of the storm, ultimately attaining the angle of repose.
The trackway was then buried by an additional 1?2 m of dune sand
perhaps as the storm waned, indicating that foul weather continued for
some time after the bird traversed the lee side of the dune. Immediately
adjacent to the beds containing the trackway, several hollow tree molds
,4 m high are also observed in the deposits (wide arrows; Figs. 4, 5A).
We conclude that the late Pleistocene avian trackway on Charles
Island, Bermuda, represents the path of a lone bird, likely a willet
(Catoptrophorus semipalmatus, Scolopacidae), sheltering from the
effects of a major storm on the lee side of a beachridge. The period
of willets? main abundance in Bermuda coincides with the late summer
or early fall hurricane season. Hitherto, there has been no evidence of
this species in the fossil record of Bermuda (Olson et al., 2005).
Palmetto Fronds.?Impressions of what we and others have
interpreted as palm fronds occur along South Shore coastal exposures
between Grape Bay, Southcourt Avenue, and Castle Roads (southern
entrance to Castle Harbour) (Fig. 1). Most commonly observed in the
dune sand are linear corrugations of significant number (Figs. 6, 7A?
B). These molds are characterized by 2- to 3-cm-wide, zigzag patterns,
radiating from a curving central axis or petiole (Figs. 7A?B).
Impressions sometimes occur stacked, like cards, standing vertically
in arching planar zones up to 50?100 cm wide. The corrugated
interfaces are readily exploited by erosion (Fig. 6). The impressions of
the fronds crosscut the dune bedding planes, which are generally
dipping to the north. This implies that the plants were standing upright
when they were buried, as opposed to lying flat on the ground as in the
case of dead, detatched fronds. The striking similarity of these traces
compared with living palmetto fronds, including the arching petiole
(Figs. 7A?B), precludes their formation as solutional rillenkarren as
suggested by Herwitz (1993).
A Buried Forest of Sabal Palm??Spectacular mold features exposed
within a vertical cliff section (Figs. 7C, 8A?B) at Hungry Bay?
documented photographically in May 1989?have since been destroyed
during cliff erosion by several storms since the 1990s. Numerous near-
vertical, cylindrical, empty or partially filled molds, 30?60 cm in
diameter, in various orientations (vertical, sloping, and horizontal), and
more than 8 m tall were exposed in the cliff section ,30 m length along
the shore. Directly associated with the trunk molds were the corrugated
frond impressions described above. In some cases the fronds are
pointing downward from near the top of the cylindrical structures
(Figs. 7C, 8) as if they were hanging from the trunk when buried.
Because of their consistent diameters (,50 cm) and association with
the radiating corrugations, many of the cylindrical forms are
interpreted as trunks of the endemic Bermuda palmetto, Sabal
bermudana, which lives on the islands today. Although we have no
knowledge of the morphometrics of mature virgin forests of S.
bermudana in times past, their length and trunk diameters and the
association with corrugated fronds invites an association with the
endemic S. bermudana, living examples of which average 25 cm in trunk
diameter (Britton, 1918, p. 56). This association is highly likely as the
only other common endemic tree of Pleistocene Bermuda was Juniperus
bermudiana, which has a quite different morphology.
An entire forest of S. bermudana was rapidly buried beneath more
than 8?10 m of uninterrupted, thick and broad sets of cascading dune
sand in both windward and leeward sets. We suggest that a single
depositional event buried the plants while either alive or standing dead
with leaves still attached. Sabal bermudana palms are well adapted to
salt, sand, and partial burial and presumably could survive most
conditions short of complete burial. Many of the trees were standing
while the fronds were still attached to crown of the tree, based on the
orientation of the molds (Fig. 7C). Other tree impressions were oriented
horizontally at different levels (circular molds in Fig. 8B), indicating
they had fallen before burial as in an established mature palm forest,
although perhaps they were toppled or broken off during the storm.
The basal soils in which the trees grew are not exposed, although
sinuous plant traces (Fig. 8B) that could be from roots were observed in
the outcrop.
Summary of Evidence for Abnormally Intense Storms
Evidence for great storms, perhaps no more than two, along the
south shore of Bermuda in MIS 5c consists of: (1) continuous,
uninterrupted sets of foreset bedding in 3-dimensional architecture,
whereas any cessation of deposition would be made evident by
protosols, the absence of foreset beds or changes within them,
structureless sand, or erosional discontinuities; (2) complete and rapid
burial of standing trees 10 m in height with fronds still attached; (3)
entombment of palmettos indicating live burial of vertical fronds as
opposed to the dead and decaying leaves expected with partial or
shallow burial; (4) preservation of avian trackways on a 25u slope made
in wet sand and buried before they could dry and be blown away; (5)
with maximum sea level being well below present, land transport of
sediments must have been due to high energy sources in order to deposit
such great volumes on land above present sea level; and (6) the existence
of only one (or possibly two at Charles Island) homogenous facies units
composed only of parallel foresets in the entire HB formation during a
highstand of several thousand years. Only exceptional events could
account for the coincidence of such circumstances.
FIGURE 6?Corrugations typical of molds of palmetto fronds at Grape Bay,
Bermuda. The hammer handle in the lower left of the photo is 3.5 cm wide.
PALAIOS BERMUDA TRACE FOSSILS 399
FIGURE 7?Plant impressions and modern counterparts in rocks correlated with the Hungry Bay Formation of mid last interglacial age (MIS 5c). A) corrugated impression of
a frond with arching petiole at Southcourt Avenue (Feb 2005 photo; hand for scale); B and D) The modern endemic palm Sabal bermudana. The trunk in view D is
approximately 40 cm in diameter. C) Hollow tree mold with fronds (arrows), many pointing downward from apex of the trunk (at Hungry Bay, 10 May 1989 photo).
400 HEARTY AND OLSON PALAIOS
A Model for Tree Mold-Cast Conduit Formation and Evolution
Bermuda offers exquisite examples of tree molds that exhibit
youthful stages (,125 ka) of evolution and diagenesis of vertical
hydrological conduits within carbonate dunes. Combined with similar
but older (ca. 400 ka) features in more advanced stages of development
at the Pinnacles in Western Australia at Nambung National Park
(NNP) (Lowry, 1973; McNamara, 1995; Hearty and O?Leary, 2008;
Lipar, 2009), we can develop an empirical model to explain the initial
formation and evolution of conduits created by burial of standing trees
in dunes (Fig. 9).
Stage I: Burial of Trees and Initial Conduit Formation.?Dune sand
may accrete with sufficient speed during intense storms to bury living
trees growing landward of the coastal beach-dune ridge. Rapid
deposition and large volumes of sediment are considered essential to
the process, as initial cementation of dune sand and molding of the
trunk must occur while the standing trunk is intact, previous to or
during its subsequent decay.
Highly soluble aragonite is abundant in the vadose environment of
the dune that would precipitate around the trees as a result of the
release of abundant organic acids from decaying plant material. Fluid
flow and precipitation of CaCO3 as calcrete would occur along the
cylindrical plane of the tree trunks forming molds deep within the
dunes. Curran et al. (2008; Fig. 5C) observe that calcrete rinds may
begin to form while the trees are still living.
Vadose solutions tracking along the mold-cast interface would also
move laterally and irregularly, filling voids and penetrating some of the
enclosing dune sediments. The result would be cementation and
induration of dune bedding beyond the circumference of the original
mold (Fig. 10A).
Stage II: Tree Decay and Initial Infilling of Void.?Molds within
carbonate sand (Fig. 9A) are progressively filled by a mixture of
materials collapsing, seeping, precipitating, and slowly entering into
the voids (Fig. 9B) as the tree begins to decay. The variety of infilling
materials (Fig. 9C) reflects polygenetic processes within the dune, the
cylindrical void, and at the surface over a considerable amount of
time. Materials deposited in the voids created by the decaying trunks
early in the process might include: (a) carbonate sand from adjacent
unconsolidated dunes or younger overlying dunes (often preserving as
hourglass-like sand-flow cross-bedding within the casts); (b) quartz
sand from continental dunes migrating across the landscape in the
case of Western Australia. Materials deposited later might include: (c)
breccia and rhizoliths from plants growing on the dune surface whose
roots exploit the conduit for moisture and nutrients; (d) reddened
soils and silts including atmospheric dust, translocated downward
(Fig. 9C); and (e) micrite or microcrystalline calcite progressively
filling interstitial pores.
Stage III: Cementation and Calcretization.?Over tens to hundreds of
thousands of years, secondary plant growth and roots exploit the
conduit by wedging and breaking the rock, eventually to form a
network of rhizoliths and brecciated limestone. Progressive thickening
of calcrete tubes and lenses would occur over time, just as they thicken
in soils and calcrete capping the dune limestone (Hearty, 1998).
Seasonal wetting and drying, and precipitation of vadose solutions,
would progressively fill voids with micrite. The result is a slightly
distorted replica of the morphology of the original tree, generally
tapering upward from the base of the dune, reflecting the initial
formation of casts from the bottom up. Irregular cementation of high-
angle hourglass bedding within and adjacent dune bedding beyond the
margin of the original trunk would explain the numerous observations
of cross-bedding within and around the Western Australia pinnacles
(Grimes, 2009; Lipar, 2009).
Stage IV: Exhumation and Sculpting.?Over several hundred
thousand years (multiple subsequent glacial-interglacial cycles) the
vertical casts become highly indurated as a complex aggregate of
polygenetic sedimentary components (Figs. 9C?D). At the eolianite
surface, a pedogenic calcrete may progressively thicken to several
decimeters over this same time interval. The high level of induration of
the filled tree cast would extend into the surrounding aeolianite matrix
(Hearty and O?Leary, 2008; Lipar, 2009). In the case of the Pinnacles at
NNP in Western Australia, subsequent erosion by wind and water
removed the softer eolianite matrix, leaving the stronger pinnacles in
positive relief (Figs. 9C?D) often with a relict cap of calcrete indicating
the approximate position of the original dune surface.
Alternative Models for Conduit Formation and Development
There is ongoing debate regarding the formation of vertical conduits
in limestone, but we suggest that the confusion originates from two
distinct processes that slightly converge in their morphological end
products. The fundamental question distinguishing those processes is
the initial formation of the void within the limestone. Are conduits
formed over hundreds of thousands of years by dissolution and
diagenesis of limestone forming a solution pipe, or are they formed
instantaneously (geologically) by molding and subsequent casting of
vertical tree trunks? We emphasize that our conduit model above differs
significantly in both fundamental process and time scale from the
pedogenic evolution of solution pipes that develop in such aging
limestone (see discussion in Herwitz, 1993; Lundberg and Taggart,
1995) as the Big Red Soil (BRS) in the early Pleistocene Walsingham
Formation of Bermuda (Hearty, 2002) (Fig. 10B), which clearly is not a
fossilized relict of a buried forest.
FIGURE 8?Preservation of a forest of trees at Hungry Bay. A) photograph showing
several hollow and refilled casts of suspected buried Sabal bermudana forest. B)
Sketch interpreting the above photograph showing standing and fallen trunks and a
large number of frond molds and rhizomorphs.
PALAIOS BERMUDA TRACE FOSSILS 401
FIGURE 9?Successive stages of evolution of mold and cast tree forms. A) Stage I: unfilled mold of palmetto in 100 ka eolianite at Hungry Bay, Bermuda. B) Stage II: mostly
filled cast of a large tree mold and cast in Nassau, Bahamas, 125 ka dune (note large void and rhizomorphs at top of cast). C) Stage III/IV: horizontally filled cast at Nambung
National Park (NNP, Western Australia) showing tapered upward trunk morphology?pinnacle molds and casting are estimated to have been initiated ca. 400 ka (Hearty and
O?Leary, 2008). D) Stage III/IV: field of pinnacles at NNP with conical or cylindrical morphology that have weathered out of softer sediments?millennia of subsequent
exposure to wind abrasion and sand-blasting have sculpted the Pinnacles at NNP into a myriad of unique shapes.
402 HEARTY AND OLSON PALAIOS
Solution Pipes and Top-Down Formation of Conduits.?The forma-
tion of solution pipes is dominated by dissolution and pedogenesis
forming enlarging vertical shafts in limestone that are simultaneously
(syngenetically?) filled with soils, rhizoliths, and eventually terra rossa
soils (Lundberg and Taggart, 1995; Grimes, 2009). This process might
be described as top-down conduit formation because the development
of the pipes enlarge from the surface downward, perhaps initially
guided by hydrologic stemflow from plants. As observed in well-
developed soils and karst surfaces of numerous Quaternary sections
(Hearty et al., 2005; Hearty and O?Leary, 2008; Olson and Hearty,
2009), this top-down pedogenic process results in meter-scale deep
shafts, narrowing with depth and filled with red or deep red clayey soil
(such as the BRS), formed on a time scale of over 0.5 to over 2.0 Ma
(Fig. 10B). In this latter case, if the enclosing softer dune matrix were
removed by erosion, the top-down, calcrete-lined, conical solution pipes
would be top-heavy, break off at their bases, and ultimately be strewn
about horizontally as a lag on the erosion surface, not vertically
standing as strong pillars in the case of the pinnacles. The solution pipes
are generally filled with terra rossa clayey soil, unless excavated by
inundation or drained by encountering subsurface voids.
At the Pinnacles at NNP in Western Australia, one group favors
initial, early development of conduits from the land surface downward
of the limestone by dissolution (Lowry, 1973; McNamara, 1995;
Grimes, 2009).
Tree Molding and Casting: Bottom-Up Formation of Conduits.?In
contrast (Hearty and O?Leary, 2008 and herein), we argue that the
composition and morphology of the pinnacles at NNP indicate an
initial bottom-up origin and evolution of conduits from tree molds and
casts. Formation of the conduit can occur as quickly as cementation
takes place in as little as a few hundred to a few thousand years (Halley
and Harris, 1979; Curran et al., 2008). In our bottom-up model,
pedogenesis and dissolution play a secondary role as they are only
related to thickening of calcrete in the subsurface and translocation of
soil material from the surface to build the cast. With removal of the
enclosing dune matrix, the calcrete-hardened casts would be left
standing vertically as in the case of the pinnacles of Western Australia.
CONCLUSIONS
During the mid last interglacial period, as powerful storms, probably
late summer hurricanes, slammed the southerly coast of Bermuda,
extensive dunes were deposited along nearly 10 km of coastline at a time
when sea level was above the platform margin, yet well below the
present datum. The voluminous sediment deposition, perhaps amplified
by waves along a steep shelf margin, has been dated to MIS 5c around
100 ka ago. The slurry of sea, sand, and air at the height of an intense
oceanic storm discharged huge volumes of sediment on land. Within the
dunes, delicate footprints of a willet (Catoptrophorus semipalmatus)
were imprinted on soft, wet sand at the onset of hurricane conditions,
then immediately buried. Vivid impressions of fronds presumably of
Sabal bermudana are preserved extensively throughout the deposit.
Hollow or partially filled molds of tree trunks represent the rapid burial
of entire standing trees in a mature forest while still alive or recently
dead.
The extent and consistent character of the foreset bedding of the
HBF eolianites suggest an exceptionally powerful storm that deposited
over 8 m of sediment on land during perhaps a single event that
occurred when sea level was positioned well below present. To the east
at Charles Island, the thick eolianites are interrupted by a well-
developed protosol containing abundant snails, suggesting reestablish-
ment of vegetative cover and slope stability, implying more than one
storm event. Thus, we suggest that the entire HBF on land in Bermuda
may have been deposited in as few as two intense storms that occurred
during the MIS 5c highstand.
Perhaps due to convergent forms, there is debate and considerable
confusion surrounding the origin of vertical conduits in limestone. The
evolutionary stages we suggest here differ considerably from stemflow,
or other means of concentration of fluids that through karstic processes
and pedogenesis, dissolve vertical shafts in limestone that are generally
backfilled with clayey terra rossa soils. The rapid burial, molding and
subsequent casting of tree structures, the evolution of the resulting
vertical conduit, and the ultimate possible exhumation comprise the
various stages of a model that explains such features in Quaternary
dunes. Understanding the processes and speed of entombment of
coastal landscapes may provide a bellwether of future storminess
associated with the continuing natural rise of the MIS 1 sea level with
possible exacerbation by anthropogenic global change.
ACKNOWLEDGMENTS
The bird trackways were discovered originally by Taran Card. We
are especially indebted to Patrick Talbot for showing Olson to the track
site and providing photographs and information, after which David
Wingate took us on a second visit. Lisa Green and Wolfgang Sterrer,
Bermuda Aquarium, Museum, and Zoo (BAMZ) were, as usual,
instrumental local contacts. We thank A. Conrad Neumann for sharing
his original concept of the intermingling of air, seawater, and sand
during intense storms and for discussions with M. Lipar that helped to
clarify the processes related to the formation of vertical conduits.
FIGURE 10?Contrasting processes and time scales. A) advanced stage of bottom-
up evolution of molds and casts over a several hundred thousand years at the
Pinnacles of Western Australia. Note cemented cross-bedding around margins of
casts. B) example of top-down development of solution pipes in recrystallized
limestone filled with deep red soils (Big Red Soil of Hearty, 2002) in the upper several
meters of the Walsingham Formation (.1 Ma?) capped by the Town Hill Formation
(MIS 11) in Wilkinson Quarry, Bermuda.
PALAIOS BERMUDA TRACE FOSSILS 403
Thanks to the editors of PALAIOS, and reviewers T. Martin and H.A.
Curran, for providing constructive comments that significantly
improved this manuscript. This is contribution #179 Bermuda
Biodiversity Project (BBP), of the BAMZ.
REFERENCES
AMOS, E.J.R., 1991, A Guide to the Birds of Bermuda: Published by the author,
Warwick, Bermuda, 206 p.
BRITTON, N.L., 1918, Flora of Bermuda: Charles Scribner?s Sons, New York, 585 p.
COUES, E., 1903, Key to North American Birds: 5th edition, 2 volumes: The Page
Company, Boston, 1152 p.
CURRAN, H.A., and WHITE, B., 2001, Ichnology of Holocene carbonate eolianites of
the Bahamas: SEPM (Society for Sedimentary Geology) Special Publication, v. 71,
p. 47?55.
CURRAN, H.A., WILSON, M.A., and MYLROIE, J.E., 2008, Fossil palm frond and tree
trunk molds: Occurrence and implications for interpretation of Bahamian
Quaternary carbonate eolianites, in Parks, L.E., and Freile, D., eds., Proceedings
of the Thirteenth Symposium on the Geology of the Bahamas and Other
Carbonate Regions: Gerace Research Center, San Salvador, Bahamas, p. 183?195.
DONNELLY, J.P., ROLL, S., WENGREN, M., BUTLER, J., LEDER, R., and WEBB, T., III,
2001, Sedimentary evidence of intense hurricane strikes from New Jersey: Geology,
v. 29, p. 615?618.
DORALE, J.A., ONAC, B.P., FORNO?S, J.J., GINE?S, J., GINE?S, A., TUCCIMEI, P., and PEATE,
D.W., 2010, Sea-Level Highstand 81,000 Years Ago in Mallorca: Science, v. 327, p.
860, doi: 10.1126/science.1181725.
ELBROCH, M., and MARKS, E., 1999, Bird Tracks & Sign: Stackpole Books,
Mechanicsburg, Pennsylvania, 456 p.
GATESY, S.M., 1995, Functional evolution of the hindlimb and tail from basal
theropods to birds, in Thomason, J., ed., Functional Morphology in Vertebrate
Paleontology: Cambridge University Press, Cambridge, UK, p. 219?234.
GRIMES, K.G., 2009, Solution pipes and pinnacles in syngenetic karst, in Gines, A.,
Knez, M., Slabe, T., Dreybrodt, W., eds., Karst Rock Features Karren Sculpturing
Postojna, Karst Research Institute ZRC SAZU, Ljubljana, p. 513?526.
HALLEY, R.B., and HARRIS, P.M., 1979, Freshwater cementation of a 1000 year old
oolite: Journal of Sedimentary Petrology, v. 49, p. 969?988.
HARMON, R.S., MITTERER, R.M., KRIAUSAKUL, N., LAND, L.S., SCHWARCZ, H.P.,
GARRETT, P., LARSON, G.J., VACHER, H.L., and ROWE, M., 1983, U/Th and amino
acid racemization geochronology of Bermuda: Implications for eustatic sea-level
fluctuation over the past 250,000 years: Paleogeography, Paleoclimatology,
Paleoecology, v. 44, p. 41?70.
HEARTY, P.J., 1997, Boulder deposits from large waves during the last interglaciation
on North Eleuthera, Bahamas: Quaternary Research, v. 48, p. 326?338.
HEARTY, P.J., 1998, The geology of Eleuthera Island, Bahamas: A Rosetta Stone of
Quaternary stratigraphy and sea-level history: Quaternary Science Reviews, v. 17,
p. 333?355.
HEARTY, P.J., 2002, Revision of the late Pleistocene stratigraphy of Bermuda:
Sedimentary Geology, v. 153, p. 1?21.
HEARTY, P.J., HOLLIN, J.T., NEUMANN, A.C., O?LEARY, M.J., and MCCULLOCH, M.,
2007, Global sea-level fluctuations during the last integlaciation (MIS 5e):
Quaternary Science Reviews, v. 26, p. 2090?2112.
HEARTY, P.J., KARNER, D.B., RENNE, P.R., OLSON, S.L., and FLETCHER, S., 2005,
40Ar/39Ar age of a young rejuvenation basalt flow: Implications for the duration of
volcanism and the timing of carbonate platform development during the
Quaternary on Kaua?i, Hawaiian Islands: New Zealand Journal of Geology and
Geophysics, v. 48, p. 199?211.
HEARTY, P.J., KINDLER, P., CHENG, H., and EDWARDS, R.L., 1999, Evidence for a
+20 m middle Pleistocene sea-level highstand (Bermuda and Bahamas) and partial
collapse of Antarctic ice: Geology, v. 27, p. 375?368.
HEARTY, P.J., and NEUMANN, A.C., 2001, Rapid Sea-Level and Climate Change at the
Close of the Last Interglaciation (MIS 5e): Evidence from the Bahama Islands:
Quaternary Science Reviews, v. 20, p. 1881?1895.
HEARTY, P.J., and O?LEARY, M.J., 2008, Carbonate eolianites, quartz sands, sea-level
cycles, Western Australia: Quaternary Geochronology, v. 3, p. 26?55.
HEARTY, P.J., and OLSON, S.L., 2010, Geochronology, biostratigraphy, and changing
shell morphology in the land snail subgenus Poecilozonites during the Quaternary
of Bermuda: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 293, p. 9?29,
doi: 10.1016/j.palaeo. 2010.04.026.
HEARTY, P.J., OLSON, S.L., KAUFMAN, D.S., EDWARDS, R.L., and CHENG, H. 2004,
Stratigraphy and geochronology of pitfall accumulations in caves and fissures,
Bermuda: Quaternary Science Reviews, v. 23, p. 1151?1171.
HEARTY, P.J., VACHER, H.L., and MITTERER, R.M., 1992, Aminostratigraphy and ages
of Pleistocene limestones of Bermuda: Geological Society of America Bulletin, v.
104, p. 471?480.
HERWITZ, S.R., 1993, Stemflow influences in the formation of solution pipes in
Bermuda eolianite: Geomorphology, v. 6, p. 253?271.
HERWITZ, S.R., and MUHS, D.R., 1995, Bermuda solution pipe soils: A geochemical
evaluation of eolian parent materials: Geological Society of America, Special
Paper, v. 300, p. 311?323.
JACKSON, S.J., WHYTE, M.A., and ROMANO, M., 2010, Range of experimental dinosaur
(Hypsilophodon foxii) footprints due to variation in sand consistency: How wet was
the track?: Ichnos, v. 16, p. 197?214.
KINDLER, P., CURRAN, H.A., MARTY, D., and SAMANKASSOU, E., 2008, Multiple
sedimentary sequences, bird tracks and lagoon beaches in last interglacial oolites,
Boiling Hole, North Eleuthera Island, Bahamas, in Proceedings of the Thirteenth
Symposium on the Geology of the Bahamas and Other Carbonate Regions, Parks,
L.E., and Freile, D., eds., Gerace Research Center, San Salvador, Bahamas, p.
169?181.
KRAUS, M.J., and HASIOTIS, S.T., 2006, Significance of different modes of rhizolith
preservation to interpreting paleoenvironmental and paleohydrologic settings:
Examples from Paleogene paleosols, Bighorn basin, Wyoming: Journal of
Sedimentary Research, v. 76, p. 633?646.
LIPAR, M., 2009, Pinnacle syngenetic karst in Nambung National Park, Western
Australia: Acta Carsologica, v. 38(1), p. 41?50.
LIVINGSTON, W., 1944, Observations on the structure of Bermuda: Geographical
Journal, v. 104, p. 40?48.
LOWRY, D.C., 1973, Origin of The Pinnacles, Nambung, WA: Australian
Speleological Foundation Newsletter, v. 62, p. 7?8.
LOWTHER, P.E., DOUGLAS, H.D., III, and GRATTO-TREVOR, C.L., 2001, Willet
Catoptrophorus semipalmatus: The Birds of North America, v. 579, p. 1?32.
LOWTHER, P.E., DOUGLAS, H.D., III, and GRATTO-TREVOR, C.L., 2001, Willet (Tringa
semipalmata), in A. Poole, ed., The Birds of North America Online, Ithaca: Cornell
Lab of Ornithology; Retrieved from the Birds of North America Online: http://bna.
birds.cornell.edu/bna/species/579, doi:10.2173/bna.579.
LUDWIG, K.R., MUHS, D.R., SIMMONS, K.R., HALLEY, R.B., and SHINN, E.A., 1996,
Sea-level records at ,80 ka from tectonically stable platforms: Florida and
Bermuda: Geology, v. 24(3), p. 211?214.
LUNDBERG, J., and TAGGART, B.E., 1995, Dissolution pipes in northern Puerto Rico:
An exhumed paleokarst: Carbonates and Evaporites, v. 10(2), p. 171?183.
MACKENZIE, F.T., 1964a, Geometry of Bermuda calcareous dune cross-bedding:
Science, v. 144, p. 1449?1450.
MACKENZIE, F.T., 1964b, Bermuda Pleistocene eolianites and paleowinds: Sedimen-
tology, v. 3, p. 51?64.
MANN, M.E., WOODRUFF, J.D., DONNELLY, J.P., and ZHANG, Z., 2009, Atlantic
hurricanes and climate over the past 1,500 years: Nature, v. 460, p. 880?885, doi:
10.1038/nature08219.
MCNAMARA, K., 1995, Pinnacles: Western Australian Museum, Perth, Australia, p.
92?108.
MYLROIE, J.E., 2008, Late Quaternary sea-level position: Evidence from Bahamian
carbonate deposition and dissolution: Quaternary International, v. 183, p. 61?75,
doi:10.1016/j.quaint.2007.06.030.
NEUMANN, A.C., and HEARTY, P.J., 1996, Rapid sea-level changes at the close of the
Last Interglacial (Stage 5e) recorded in Bahamian Island geology: Geology, v. 24,
p. 775?778.
NOTT, J., 2004, The tsunami hypothesis: Comparisons of the field evidence against the
effects, on the Western Australian coast, of some of the most powerful storms on
Earth: Marine Geology, v. 208, p. 1?12.
OLSON, S.L., and HEARTY, P.J., 2003, Probable extirpation of a middle Pleistocene
breeding colony of Short-tailed Albatross (Phoebastria albatrus) on Bermuda by a
+20 m interglacial sea-level rise: Proceedings of the National Academy of Sciences,
v. 100(22), p. 12825?12829.
OLSON, S.L., and HEARTY, P.J., 2009, A sustained +21 m highstand during MIS 11
(400 ka): Direct fossil and sedimentary evidence from Bermuda: Quaternary
Science Reviews, v. 28, p. 271?285.
OLSON, S.L., WINGATE, D.B., HEARTY, P.J., and GRADY, F.V., 2005, Prodromus of
vertebrate paleontology and geochronology of Bermuda, in Alcover, J.A., and
Bover, P., eds., Insular Vertebrate Evolution: The Palaeontological Approach:
Monografies de la Societat d?Histo`ria Natural de les Balears, Sept. 2003, Mallorca,
Spain, p. 1?22.
RAY, J., 1678, The Ornithology of Francis Willughby: Royal Society, London, 448 p.
STONE, J.W., and ORFORD, J.D., 2004, Storms and their significance in coastal
morpho-sedimentary dynamics: Marine Geology, v. 210, p. 1?5.
VACHER, H.L., and HEARTY, P.J., 1989, History of stage-5 sea level in Bermuda: With
new evidence of a rise to present sea level during substage 5a: Quaternary Science
Reviews, v. 8, p. 159?168.
VACHER, H.L., HEARTY, P.J., and ROWE, M., 1995, Stratigraphy of Bermuda:
Nomenclature, concepts, and status of multiple systems of classification, in Curran,
H.A., and White, B., eds., Terrestrial and Shallow Marine Geology of the Bahamas
and Bermuda: Geological Society of America Special Paper, v. 300, p. 271?294.
404 HEARTY AND OLSON PALAIOS
VACHER, H.L., and ROWE, M.P., 1997, Geology and hydrology of Bermuda, in
Vacher, H.L., and Quinn, T.M., eds., Geology and Hydrology of Carbonate
Islands: Developments in Sedimentology 54, Elsevier Science B.V., Amsterdam,
p. 35?90.
VACHER, H.L., ROWE, M., and GARRETT, P., 1989, Geologic map of Bermuda: Oxford
Cartographers, UK, 1:25,000 scale.
VERRILL, A.E., 1902, The Bermuda Islands: Their scenery, climate, productions,
natural history and geology: Transactions of the Connecticut Academy of Arts and
Sciences, v. 11, p. 413?916.
VOLLBRECHT, R., 1990, Marine and meteoric diagenesis of submarine Pleistocene
carbonates from the Bermuda carbonate platform: Carbonates and Evaporites, v.
5, p. 13?95.
ACCEPTED APRIL 8, 2011
PALAIOS BERMUDA TRACE FOSSILS 405