Content uploaded by Darrin Lowery
Author content
All content in this area was uploaded by Darrin Lowery on Jun 02, 2014
Content may be subject to copyright.
73
Chapter 5
New Evidence for a Possible Paleolithic
Occupation of the Eastern North American
Continental Shelf at the Last Glacial Maximum
Dennis Stanford, Darrin Lowery, Margaret Jodry, Bruce A. Bradley,
Marvin Kay, Thomas W. Stafford and Robert J. Speakman
D. Stanford () · M. Jodry
Department of Anthropology, Smithsonian Institution, 20560 Washington, DC, USA
e-mail: Stanford@si.edu
M. Jodry
e-mail: JODRYM@SI.edu
D. Lowery
Geology Department, University of Delaware, 101 Penny Hall,
19716 Newark, DE, USA
e-mail: darrinlowery@yahoo.com
B. A. Bradley
Department of Archaeology Laver Building, University of Exeter, EX4 4QE, Exeter, UK
e-mail: b.a.bradley@ex.ac.uk
M. Kay
Anthropology Department, Old Main 330, University of Arkansas, 72701 Fayetteville, AR, USA
e-mail: mkay@uark.edu
T. W. Stafford
Stafford Laboratories Inc., 5401 Western Avenue, Suite C,
80301 Boulder, CO, USA
e-mail: TWSTAFFORD@stafford-research.com
R. J. Speakman
Center for Applied Isotope Studies, University of Georgia, 30602 Athens, GA, USA
e-mail: archsci@uga.edu
Introduction
Researchers have postulated the presence of submerged archaeological deposits on
the Middle Atlantic continental shelf of North America for decades (Emery and
Edwards 1966; Edwards and Emery 1977; Kraft et al. 1983). However, archaeologi-
cal discoveries on the continental shelf made during commercial shellfish dredging
have gone unrecorded or have escaped detection. By contrast, numerous vertebrate
remains including the bones, teeth, and skulls of mammoth, mastodon, and walrus
A. M. Evans et al. (eds.), Prehistoric Archaeology on the Continental Shelf,
DOI 10.1007/978-1-4614-9635-9_5, © Springer Science+Business Media New York 2014
74 D. Stanford et al.
have been reportedly discovered by deep-sea fishermen and dredgers on the conti-
nental shelf (Edwards and Merrill 1977; Whitmore et al. 1967).
In 1974, Captain Thurston Shawn and the crew of Cinmar, a scallop trawler
working 100 km east of the Virginia Capes, were dredging at a depth of 70 m
(Fig. 5.1). Just after starting their run, the dredge became very heavy and when
reeled in, it contained a mastodon skull. While cleaning the bone from the dredge,
a large bifacially flaked rhyolite knife was discovered. Shawn carefully plotted the
water depth and the exact location of the find on his navigation charts and noted
that all of these items were dredged at the same time. To expedite getting back to
dredging, the Cinmar crew broke up the skull and removed the tusks and teeth for
souvenirs, throwing the rest of the bone overboard. Later the tusks were sawn into
pieces and distributed among the crew.
Fig. 5.1 Last Glacial Maximum Susquehanna River drainage showing locations of the Cinmar
site and Ryolite Quarry
75
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
Captain Shawn retained for himself a tusk section, a complete tooth and the
biface, and gave one of the molars to his sister, Mrs. Sylvia Cannon of Mathews,
Virginia. Shawn was not an artifact or fossil collector and, subsequently, sold his
specimens to Dean Parker of Hudgens, Virginia. Parker in turn loaned them to the
Gwynn’s Island Museum where they have been on exhibit since 1974 (Stanford and
Bradley 2012).
The significance of the Cinmar’s discovery was not recognized until Darrin
Lowery conducted an archaeological survey in Mathews County, Virginia, and saw
the biface, mastodon tooth, and tusk segment at the museum. Subsequent interviews
with Captain Shawn and his sister confirmed the fact that all of the specimens were
recovered at the same time and place, as described here. The importance of the
Cinmar evidence concerning the timing of the New World settlement and human
occupation of the now-submerged coastal settings initiated the study reported here.
The find location, designated the Cinmar site, is on the edge of the outer conti-
nental shelf, south of the last glacial maximum (LGM) Susquehanna Paleo-River
Valley, which is referred to as the Cape Charles channel (Fig. 5.1). During the LGM,
19,000–26,500 years ago (Clark et al. 2009), sea stand is estimated to have been
130 m below the present sea level (Milliman and Emery 1968; Belknap and Kraft
1977). The site was on the edge of the LGM James Peninsula, immediately west of
a LGM barrier island and channel. This terrestrial landscape, which existed between
at least 14,500 years ago and possibly more than 25,000 years ago, would have
been 10–14 meters below sea level (mbsl) by the time Paleoindians occupied North
America approximately 13,500 years ago (Waters and Stafford 2007).
The Cinmar stone tool is a large, thin knife with evidence of well-controlled per-
cussion thinning flake scars on both faces (Fig. 5.2). It represents the workmanship
of a highly skilled flint knapper because rhyolite is very difficult to flake correctly.
The obverse face has a full face, possibly large overshot flake across the basal half.
Because the overshot flake resulted in the removal of an excessive portion of the
artifact’s surface, subsequent flaking adjustments were made, resulting in a slight
longitudinal curve and variable thickness. For measurements and proportions of the
Cinmar stone tool, see Table 5.1.
Table 5.1 Measurements and proportions (in mm) of the Cinmar stone tool
Length 186 (est. 190) Length/maximum width 3.5
Maximum width 54 Maximum width/thickness 7.7
Thickness 6 (at maximum width)
Width at 1/4 length 44 Width at 1/4 length 7.3
Width at 1/2 length 53 Width at 1/2 length 6.6
Width at 3/4 length 39 Width at 3/4 length 4.3 (at “stack”)
Thickness at 1/4 length 6
Mean width ( n = 4)
6.5
Thickness at 1/2 length 8
Thickness at 3/4 length 9
76 D. Stanford et al.
Use-Wear Evaluation of the Cinmar Biface
High-power microscopic examination of the artifact using a reflected light, dif-
ferential-interface binocular microscope with polarized light and Nomarski optics
ranging between 100 and 400 diameters identified linear microstriations and polish-
es typical of knife use, including up-and-down and back-and-forth movements on
the distal surfaces of the blade. The proximal end (base) exhibits microscopic linear
striations that are typical of haft wear. Evidence that the knife was resharpened
before being lost or discarded consists of noninvasive percussion retouch along the
distal edges that is not overprinted by use-wear traces. The preserved condition of
the flake scars, together with the preservation of surface microstriations and polish,
indicate that the biface did not experience episodes of redeposition by water trans-
port, nor was it abraded by surf action.
Separating the irregular crystal polygons evident on the banded rhyolite refer-
ence specimens and the artifact from microscopic wear traces proved a simple task,
as the wear traces are largely striated residues, or additive “microplating” features,
Fig. 5.2 The Cinmar Biface
77
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
which develop progressively with tool use (Kay 1996, 1998). Microplating residues
are impervious to ultrasonic cleaning with concentrated strong alkali (KOH) and
acid (HCL), and occur on siliceous artifacts from varied depositional environments
and ages in excess of at least 100,000 years. Experimentation demonstrates that mi-
croplating residues develop and harden coincident with tool use, are a biochemical
byproduct of moisture and direct contact with a material worked by a stone tool or
adhering to it, and, in an elegant way, express tool motion kinematics. Characteristic
of microplating residues are flow features; among them are filled-in striations, des-
iccation cracks on drying, abrasive particle capturing, and crystallization filaments.
Abrasive particles and crystallization filaments occur on the trailing edge or surface
opposite the direction(s) of movement of a tool stroke. They are also instructive of
handholding the tool or complementary movement of the tool in its handle. Micro-
plating features are ubiquitous on the artifact and overprint other tool use-related
abrasion and abrasive wear traces. They do not occur on an examined banded rhyo-
lite comparative specimens and are easily distinguished from the irregular crystal
polygons. These wear traces fall into two complementary categories, due to either
use or from movement in a handle.
The Cinmar biface has diagnostic and common wear traces characteristic of hav-
ing been a hafted knife. The haft wear traces do not resemble those far less devel-
oped and readily observed from handholding an experimental stone knife. The haft
element wear traces (Fig. 5.3, i) are the mirror opposite of the blade element. These
wear traces are indicative of complementary movement within the handle as a result
of tool use. Haft wear includes more extensive abrasive rounding of arises but not
true abrasive planing. The cutting wear traces (Fig. 5.3, j) are invasive and originate
on either blade edge and at the broken tip. Multiple tool strokes are recorded oblique
to the two blade edges and from the tip. The final tool strokes appear to be directed
from the tip, and either further penetrated or were withdrawing from the worked
material. Blade edge angles vary from 55° to 75°. Haft edge angles are 50° or less.
Blade edge steepening with use and resharpening seem likely, especially since the
blade edges only occasionally have use-wear. Most often the wear traces are on the
older and higher flake arises, and it was easiest to track them to these spots. The
blade edges are damaged, mostly rounded and crushed with micro step fractures.
The invasive cutting wear, the tool edge damage, and experimental analogs all point
to this knife having cut through a material that enveloped its surfaces while break-
ing and dulling the tool edges too. The contact material would have had paradoxical
qualities—hard and unyielding and yet soft and allowing deep penetration. Consis-
tent with experimental tool use-wear analogs, the likely and predominant contact
material would have been bone and cartilage within a carcass. This tool appears to
have been a heavy-duty butchering knife that was sharpened at least once and that
ultimately failed in use (Fig. 5.4).
78 D. Stanford et al.
Identification of the Source of the Rhyolite Used to Make
the Cinmar Biface
Volcanic rocks, such as obsidian and rhyolite can be linked to their geologic source
with a high degree of reliability by using analytical techniques such as instrumental
neutron activation analysis (INAA), X-ray fluorescence (XRF), and inductively-
coupled plasma mass-spectrometry (ICP-MS). These volcanic rocks typically occur
Fig. 5.3 Oriented photomicrographs of microplating residues and kinematic diagrams of stria-
tions ( colored linear features), abrasive particles ( ovals) and crystallization filaments ( gray-white
“cloud” features) for areas i and j on reverse face of the Cinmar artifact. The inferred directions
of movement for each location are the mirror opposite of the other: area i pertains to the haft ele-
ment, area j the blade element. The final movement is, respectively, diagonal to the longitudinal
axis for the haft element ( i) and just slightly off parallel to the longitudinal axis and bidirectional
for the blade element ( j)
79
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
in spatially discrete and relatively localized contexts. Such sources are typically
chemically homogeneous, and individual sources have unique chemical character-
istics. With sufficient field and laboratory work, the spatial extent of a specific geo-
chemical type of volcanic rock, including primary and secondary deposits, can be
established such that a source area can be defined (Speakman et al. 2007; Glascock
et al. 1998).
As a starting point for the geochemical source study, more than 350 vouchered
rhyolite specimens from eastern US localities ranging from Maine to North Caro-
lina (e.g., rhyolite, metarhyolite, and felsite) housed in the Rock and Ore Collection
of the Smithsonian Institution’s National Museum of Natural History, Department
of Mineral Sciences, were visually examined. More than 30 samples exhibiting
banding, as well as a few random samples, were analyzed by XRF and compared to
data from the Cinmar biface. When compared to other geologic samples from the
Eastern US, the Cinmar biface is chemically and visually distinct because of its high
(> 800 ppm) Zirconium (Zr) content and its unique banding and color.
Of the eastern US rhyolite samples examined in the National Museum of Natural
History’s mineral collection, only one was identified as a likely match: a sample of
banded metarhyolite (NMNH 60892) from the Catoctin formation of South Moun-
tain, Pennsylvania. The specific provenance of the sample is listed as “Maria Fur-
nace Road, 1 mile from Tom’s Creek Railroad Trestle.” The sample presumably was
collected by Smithsonian archaeologist W.H. Holmes who visited and described
the quarry in 1893–94 (Holmes 1897). Maria Furnace is ca. 10 miles southwest of
Gettysburg on Toms Creek, which is a branch of the Monocacy River. Following
the identification of the probable source as South Mountain, the authors visited the
Maria Furnace locale and collected additional rhyolite samples for XRF analysis.
Fig. 5.4 Use history diagram for the Cinmar biface artefact. a Primary ( gray shaded areas) and
mostly secondary flaking that crosscuts the recently damaged areas exposing the original cortex
( shaded yellow) on both faces. b Functional zones identified by microscopic evaluation of use-
wear on reverse face (haft element is shaded green, blade element cutting wear is red). c Blade
element resharpening ( shaded blue) on both faces
80 D. Stanford et al.
XRF analyses were conducted using a Bruker AXS Tracer III-V XRF. The analy-
ses permitted quantification of the following elements: Mn, Fe, Ga, Rb, Sr, Y, Zr,
Nb. The artifact and geologic specimens were analyzed as unmodified samples. The
instrument is equipped with a rhodium tube and a SiPIN detector with a resolution
of ca. 170 eV FHWM for 5.9 keV X-rays (at 1,000 counts/s) in an area 7 mm
2
. All
analyses were conducted at 40 keV, 15 µA, using a 0.076-mm copper filter and
0.0306-mm aluminum filter in the X-ray path for a 200-s live-time count. Peak
intensities for the above listed elements were calculated as ratios to the Compton
peak of rhodium, and converted to parts-per-million (ppm) using linear regressions
derived from the analysis of 15 well-characterized rhyolitic glasses that previously
had been analyzed by neutron activation analysis (INAA) and/or XRF.
Metarhyolite from South Mountain is widely recognized as a major lithic source
used for production of prehistoric stone tools throughout the US Middle Atlantic
Region (Stewart 1984, 1987) and an unpublished INAA study (Bonder 2001) has
demonstrated that metarhyolites from South Mountain are chemically discrete from
other sources in Maryland, Virginia, and North Carolina. Both visual examination
and chemical analysis confirm that the material used to manufacture the Cinmar
biface originated from the South Mountain Catoctin formation. Examination of the
XRF spectra (Fig. 5.5) and the plots of the data (Fig. 5.6) demonstrate that rhyo-
lite from outcrops near Maria Furnace are most similar chemically to the Cinmar
biface. The authors caution, however, against stating that the stone used to produce
the Cinmar biface originated from the vicinity of Maria Furnace given that numer-
ous Catoctin formation metarhyolite outcrops occur throughout the South Mountain
area of the Pennsylvania Blue Ridge.
Fig. 5.5 Comparison of XRF spectra from the Cinmar biface ( blue) and NMNH 60892 ( red)
81
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
The Mastodon, Carbon Fourteen Dates, and Environment
The diameter of the mastodon tusk section is small, measuring 83 × 73 mm. The
tooth, an upper right third molar, is also small, measuring 90 mm in width across
the tritoloph and 155 mm in length. Wear on the tooth has entered stage 2, indicating
a mature animal of approximately 30 years of age (Saunders 1977). The size and
age characteristics of the molar and tusk indicate that the Cinmar mastodon was a
small female.
Two sections of the tusk were sampled to obtain bone collagen for accelera-
tor mass spectrometry, 14C dating. The resulting age was 22,760 ± 90 RCYBP
(UCIAMS-53545). This age determination is consistent with the LGM sea level
data and led the authors to conclude that the mastodon died on the outer margin of
the continental shelf during the initial phase of the last glacial maximum.
Limited data are available for environmental reconstruction of the mid-Atlantic
outer continental shelf during the last glacial maximum. Freshwater peat dated to
15,500 years ago was dredged from depths of 64–66 m (210–216 feet) near the
Washington Canyon, north of the Cinmar site (Emery et al. 1967). Pollen extracted
from the peat suggests that spruce, water lily, sedge, pine, oak, and fir were growing
Fig. 5.6 Plot of zirconium and strontium-based ten logged concentrations for samples analyzed by
XRF. Data for the Cinmar biface (two replicates are projected against the 90 % confidence ellipse
calculated from six Maria Furnace geologic samples and two replicate analyses of NMNH 60892,
also from Maria Furnace)
82 D. Stanford et al.
on the continental shelf shortly after the last glacial maximum. Another pollen
sample, recently extracted from a soil sample taken from the Miles Point site dated
to greater than 25,500 years ago on the Eastern shore of the Chesapeake, and re-
vealed krummholz yellow birch, red spruce, balsam fir, and C3 grasses (Lowery
et al. 2010). These data are evidence that the adjacent terrestrial vegetation likely
extended as an unbroken biome onto portions of the continental shelf that were dry
land during the LGM. The likelihood of abundant freshwater springs and ponds
along the margin of the continental shelf (Faure et al. 2002), and the shrubby en-
vironment of the adjacent inter barrier island lagoon, as well as a relatively large
number of mastodon remains reported from the continental shelf (Whitmore et al.
1967), indicate an ideal environment to support a reasonable mastodon population.
Rhyolite Artifact Weathering and Patination in Coastal
Plain Environments
The Atlantic coastal plain contains a mix of chemical conditions and environ-
ments that can differentially affect rhyolite (Lowery and Wagner 2012). Rhyolite
or methyolite artifacts that are buried quickly retain a fresh appearance (Fig. 5.7a).
Conditions for rapid burial usually include anthropogenic features created by hu-
man activities. Natural processes also rapidly bury rhyolite artifacts and can re-
sult in fresh unweathered appearances. If a rhyolite artifact erodes from an upland
setting via fetch-related coastal processes, prolonged exposure to the “swash and
berm” zone results in abraded, smoothed surfaces with rounded edges (Fig. 5.7d).
The rhyolite specimen shown in Fig. 5.7d is also patinated, however, the edges
and surface of the point are rounded and polished due to prolonged tumbling and
abrasion in the surf. Inevitably, tidal marshes accrete over former uplands as a by-
product of marine transgression.
As a result of the formation of an overlying tidal marsh, iron-rich rhyolite ar-
tifacts situated within the submerged upland stratum, like Fig. 5.7a, will undergo
sulfidization in the organic-rich anaerobic surroundings. Because the iron in the
rock is chemically altered to dark-colored iron sulfide, the outward appearance
of the rhyolite artifacts in these settings may become darker, resembling the fresh
forms of rhyolite. Slow rates of sea-level rise can erode the archaeological deposits
from the drowned upland stratum beneath the tidal marsh peat. In these environ-
ments, the sulfidized rhyolite artifacts from beneath the anaerobic tidal marsh will
be subjected to aerobic conditions in the nearshore area. During periods of rapid
marine transgression, however, tidal marshes become inundated and bioturbation
by marine organisms will reintroduce oxygen to the underlying archaeological de-
posits. In either of these aerobic settings, the reoxygenated iron-rich artifacts will
undergo the sulfuricization process. As a result, a uniform chemically related sulfu-
ric acid corrosion patina will develop (Fig. 5.7b) on rhyolite artifacts. In buried or
intact settings, rhyolite artifacts will retain sharp cutting edges and any original use
wear (Fig. 5.7c). In abrasive, exposed nearshore and offshore areas, artifacts can
83
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
become dislodged from the intact drowned archaeological deposits. Under these
conditions, the edges of artifacts become heavily rounded (Fig. 5.7d–e). Depending
on the proximity to the coastline, the dislodged artifacts are tumbled and rounded
by currents in offshore sub-tidal settings or transported to abrasive nearshore areas.
Artifacts exposed for protracted periods of time on the surface of the water bottom
will generally accumulate a mix of attached marine organisms (Fig. 5.7f–h). For a
full-detailed discussion of both the sulfidization and the sulfuricization process in
nearshore coastal settings see Lowery and Wagner (2012, pp. 690–697). In contrast
to the variables impacting rhyolite artifacts along shorelines, similar artifacts de-
posited on interior upland archaeological site surfaces will only develop a patina on
surfaces that have been exposed skyward.
The unrounded surfaces retaining use wear with relatively sharp cutting edges
and even patination of the Cinmar biface indicate that the artifact originated from an
intact archaeological deposit. As a result of elevated sea levels at the end of the Ice
Age, ca. 14,500 years ago, tidal marsh peat developed over the archaeological depos-
it containing the Cinmar biface. As a result, the iron in the rhyolite underwent the sul-
fidization process associated with a brief exposure to the organic carbon-rich anaero-
bic tidal marsh surroundings. The accelerated sea level rise during Meltwater Pulse
1A quickly inundated the tidal marsh, and bioturbation from the offshore marine
organisms reintroduced oxygen to the underlying archaeological deposit. The intro-
duction of oxygen caused the Cinmar biface to undergo the effects of sulfuricization.
Like the specimen shown in Fig. 5.7b–c, a uniform, chemically mediated, corro-
sion patina developed. Unlike the specimen shown in Fig. 5.7b–c, the sulfuric acid
Fig. 5.7 Rhyolite artifacts exposed to marine and near shore environments. a Unaltered artifact
from a buried onshore environment. b, c Artifacts patinated by chemical corrosion. d Edge of arti-
fact subjected to prolonged abrading in the surf. e Chemically corroded artifact subjected to surf
abrasion. f–h Marine organisms attached to rhyolite artifacts
84 D. Stanford et al.
corrosion patina on the Cinmar biface is noticeably less. The degree of patina can
be equated to the duration of exposure to the anaerobic conditions when buried be-
neath an overlying tidal marsh. Given the slow rates of late Holocene sea level rise
(Fig. 5.8) for the Middle Atlantic (Nikitina et al. 2000) and the documented one-
meter thickness of tidal marsh peat overlying the drowned archaeological deposit
that produced the artifact shown in Fig. 5.7b–c, the authors conclude that this artifact
was subjected to at least 1,000 years of exposure to sulfidizing anaerobic conditions.
Lowery and Wagner (2012) have concluded that sulfuricization can occur very
rapidly once aerobic conditions are restored. In contrast, the setting associated with
Fig. 5.8 The Cinmar site relative to Middle Atlantic Isostatic and Global Eustatic Sea Level data.
(Based on Mallinson et al. 2005; Oldale et al. 1993)
85
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
the Cinmar biface was exposed to sulfidizing anaerobic conditions of a tidal marsh
for only a short period of time. The shortened duration and exposure to sulfidizing
conditions resulted in limited patination to the surface of the Cinmar biface once
the site was completely drowned and aerobic conditions were restored. The rates of
sea-level rise (3.7–4 m per century) postulated at the onset of Meltwater Pulse 1A c
14,500 years ago (Weaver et al. 2003) would mean that the Cinmar site was situated
in a nearshore tidal marsh environment for only a short period of time. The resultant
situation limited the artifact’s exposure to sulfidizing conditions and the rapid rates
of marine transgression inundated the site before the archaeological deposit was
eroded or disturbed.
A detailed overview of the chemical conversion of iron oxides to iron sulfides
in coastal setting soils has been presented by Fanning et al. (2010) and the same
process seems to impact iron-rich silicate artifacts in coastal tidal marsh settings
(Lowery and Wagner 2012). The rapid conversion of iron oxides in stone artifacts to
iron sulfides takes place chemically by reaction with dissolved sulfide in sea water.
The chemical reaction represents the microbial reduction of sulfate during oxidation
of organic matter in tidal marshes. The reduction of iron oxides in stone artifacts
by hydrogen sulfide results in the formation of both iron monosulfides and iron di-
sulfides (pyrite). The black monosulfides that result tend both to form quickly and
to fade quickly upon exposure to oxygen. Exposure to oxygen can be the result of
bioturbation in an offshore setting or simply a by-product of being brought to the
surface. Pyrite (FeS
2
) takes more time to form in an artifact and it is more persistent
after formation. With respect to the patination observed on the Cinmar biface, some
portion of the iron oxide in the parent rock was also altered to pyrite by long-term
exposure to the anaerobic conditions of a tidal marsh. When a stone artifact experi-
ences an aerobic environment, acidity is generated from the oxidation of the sul-
fides, and the hydrolysis of the iron. Bioturbation within a drowned tidal marsh peat
deposit introduces oxygen into the deeper anaerobic strata. In an aerobic setting, the
surface of the artifact creates its own chemical weathering patina.
When the Cinmar biface was dredged from the bottom, it was already patinated.
The conditions outlined above would explain why the Cinmar biface is uniformly
weathered or patinated on all surfaces, which is unlike the asymmetrical patination
typical of rhyolite artifacts lying on the surface in a terrestrial environment.
Questions of Association
The question of whether or not the biface was associated with the mastodon remains
is critically important for an accurate interpretation. Did Paleolithic people use the
knife while butchering the mastodon or was the close spatial relationship fortu-
itous? There are three kinds of events that might have produced a spurious associa-
tion between the artifact and the mastodon remains: lateral transport, prehistoric
coincident, and fraud. These possibilities are dealt with in turn.
86 D. Stanford et al.
Lateral Transport
The biface was initially deposited elsewhere and was transported to a location near
the mastodon remains via fluvial or tidal processes, or perhaps even dredged from
another location some distance away from the mastodon bone. The authors reject
this hypothesis for the following reasons:
Redeposition of a large stone tool in a high energy water transport system is
known to produce taphonomic alterations that modify flake scars and tool edges,
and overprint microscopic use-wear polish and striations with signatures of trans-
port that include fractured or rounded tool edges (Shea 1999; Grosman et al. 2011),
flattening of dorsal ridges, and patterns of abrasion that are similarly expressed
on the distal and proximal ends of artifacts subjected to redepositional forces. The
combined effects of sediment and debris-laden ocean currents tumbling the knife,
had it washed out to sea, would have compromised the flake ridges and knife edges
and obliterated the microscopic polish and use-wear scars. Moreover, lithic artifacts
eroded from coastal prehistoric sites stay within the “swash and berm” zone (Low-
ery 2003) and move laterally along the shoreline, and over the long-term they are
redeposited inshore, not offshore (Lowery 2008).
It might be conjectured that the knife was dredged and dragged from another
location before the trawler hit the mastodon remains. This hypothesis is rejected for
the following reasons:
• The dredge consists of a welded iron frame with a flat iron bar that drags along
the bottom. As the flat iron bar at the bottom of the dredge scrapes the sea floor,
scallops and other objects on the surface of the sea floor enter the dredge and are
captured. Behind the dredge is a large enclosure with a series of welded iron bars
with interwoven iron rings, or a monofilament seine-like bag. The sizes of the
interwoven iron rings vary but the mesh is generally between 4 and 5 in. (~ 10
and 13 cm). The mesh size limits what is retrieved from the bottom. Smaller
objects usually slip through the rings and only the larger objects are brought to
the surface. Archaeological objects on the Middle Atlantic continental shelf may
have escaped detection by commercial shell fishermen because small artifacts
such as debitage, flake tools, and projectile points less than 10–15 cm lying on
the surface of the continental shelf would easily fall through the larger size of
mesh of the equipment used to scrape the bottom for scallops before being lifted
to the surface. The large size of the dredge scalloping mesh may explain why
large Ice Age animal remains are commonly reported from drowned localities
on the Middle Atlantic continental shelf, while lithic or bone artifacts are rarely
reported. In the case of the discovery made by Captain Thurston Shawn, the large
size of the artifact allowed it to be recovered.
• The scallop dredge is tethered to the boat by a line or a large cable. Scallop
fishermen prefer to dredge stretches of the ocean floor with common or uniform
bathymetric depths to ensure that the dredge remains on the bottom. The distance
that a dredge travels across the bottom at a common or uniform bathymetric
87
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
depth can vary greatly. Generally, the captain or crew gauge the dredge retrieval
time based on the stresses placed on the boat. The stresses on the boat are caused
by the weight of material trapped in the dredge. As such, a scallop dredge can
be pulled across the sea floor for either short or long distances. The distance
traveled across the floor depends on how quickly the dredge is filled with scal-
lops and other debris. In the case of the Cinmar discovery, the stress caused by
the weight of a mastodon skull and associated tusks caused the transect run to
be terminated and the dredge pulled and cleaned as soon as the remains were
encountered. Because the biface was only slightly damaged by the iron frame,
bars and rings of the dredge the artifact was not pulled for any distance across the
sea floor, and therefore was dredged at relatively the same time as the mastodon
remains.
The interpretation is that, the skull and knife were deposited together as part of a
single archaeological assemblage. Again, if they were moved for any significant
distance by the dredge, they would have been heavily damaged by tumbling in the
metal framework of the trawler’s net. Moreover, Shawn reported that they had just
begun a transect when they encountered the heavy weight of the bones causing the
net to be reeled in unexpectedly. Trawlers in this area run parallel to the coastline
in order to maintain a constant fishing depth, so if the knife was not associated with
the bone, it would have been situated at the same elevation, and because of the an-
aerobic modification of the rhyolite it would have had to have been dredged from
an ancient saltwater marsh, as were the bones. Thus, given the fresh untumbled sur-
faces and sharp edges of the Cinmar biface, and the matching amount of oxidation
color change on both the biface and mastodon remains, the authors conclude that
the knife originated from an archaeological deposit associated with the mastodon or
near where the mastodon was dredged from the sea floor.
Prehistoric Coincident
The knife was lost or deliberately thrown overboard by a prehistoric mariner
traveling the ocean subsequent to sea-level rise and it came to rest with or near the
mastodon.
The authors submit that hypothesis 2 is a priori extraordinarily weak because of
the near absence of laurel leaf bifaces in the later middle Atlantic archaeological
record, coupled with the odds against some prehistoric hunter losing a knife while
on an ocean voyage some 100 km out in the Atlantic and that same knife settling
down over 70 m onto the same area in the ocean floor where a mastodon died mil-
lennia earlier. Moreover, if such an event had transpired, the artifact would not have
been subjected to the chemical environment that caused the geochemical patination
and weathering of its surfaces. The 16 m or more rise in eustatic sea level in less
than 300 years quickly drowned the tidal marsh and the sediments were partially
oxidized.
88 D. Stanford et al.
The short-term exposure of the iron-rich stone artifact to the anaerobic condi-
tions of the tidal marsh limited the degree of patination; however, the localized
acidic condition created by the reoxygenation of the site resulted in a uniform color
change of both faces of the knife.
Fraud
The association of the knife and the mastodon skull was fabricated. This hypothesis
is rejected because the discoverers received no glory from their find, unlike the
typical archaeological fraud. Moreover, it would seem unlikely that from all the ar-
tifacts that would have been accessible for fraudulent activities, the perpetrators of
a fraud would likely not have had a rare laurel leaf biface, let alone that they might
understand the cultural and temporal significance of a laurel leaf. It is important to
remember that both the mastodon remains and the biface had also been on display
since 1976 with a label outlining the circumstances of their discovery.
Thus, the authors conclude that the Cinmar discovery has major implications
for understanding New World prehistory. If the artifact is associated with the
22,760 RCYBP radiocarbon date, it would imply that humans were living on the
LGM continental shelf of eastern North America at least 10,000 years before any
other reliable radiocarbon dated archaeological sites. If it is not associated with the
mastodon in the freshwater marsh, the biface would be no younger than the salt-
water marsh formed at the onset of Meltwater Pulse 1A, making it at its youngest
2,000 years before the advent of Clovis, and is the oldest dated formal tool yet found
in the Americas.
The distance from the Cinmar site to the rhyolite sources in Pennsylvania is near-
ly 320 km, suggesting that by the time the Cinmar biface was manufactured, early
cultures had explored the interior of the Chesapeake drainage basin and discovered
useable stone resources. Therefore, the Cinmar date is an estimate for the timing of
human occupation in eastern North America and it nearly doubles the length of hu-
man occupation in the New World.
Chesapeake Bay Bifaces
The Cinmar biface is typologically unusual for the Middle Atlantic region. It is
significant that only three laurel leaf-shaped bifaces were identified during an
inventory of the Smithsonian Institution’s extensive archaeological collection of
nearly 300,000 artifacts from the Middle-Atlantic region representing Paleo-Indian
through historic time periods. The authors also examined private collections from
the region, collections at artifact shows, and conducted an artifact identification
weekend at Gwynn’s Island that resulted in identifying eight additional bifaces.
89
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
All but one of these bifaces were found within the Chesapeake Bay drainage
system (Fig. 5.9). The single outlier came from sand dredged from offshore to re-
plenish the beach at Ocean City, Maryland. Another was found while leveling a
LGM sand dune (Fig. 5.9, 4) and a third was found eroding out of the LGM terrace
adjacent to the Susquehanna River in Dauphin Co., Pennsylvania (Fig. 5.9, 5). A
large quartzite biface in the Smithsonian’s collection was found at Hampton, Vir-
ginia and donated to the museum in 1868 (Fig. 5.9, 2). The rest of these specimens
were dredged from the Chesapeake Bay.
A specimen made of local quartzite was dredged from the shallow water between
Tar Bay and Punch Island Creek off Dorchester County, Maryland (Fig. 5.10c). This
Fig. 5.9 Locations of Chesapeake Bay laurel leaf bifaces: 1 Cinmar site; 2 Hampton, Virginia;
3 Ocean City, Maryland; 4 Gore site; 5 Dauphin County, Pennsylvania; 6 Tar Bay, Maryland; 7
Taylor’s Island, Maryland; 8 and 9 Mopjack, Bay Virginia
90 D. Stanford et al.
specimen was at one time in the near shore zone and is an example of damage seen
on artifacts that have been heavily tumbled by “swash and berm action.” Another
specimen (Fig. 5.10d) was found within a drowned upland landscape underneath a
thick covering of tidal marsh peat on the west side of Taylor’s Island, in Dorchester
County, Maryland. The knife is made of jasper; however, because it was sulfidized
in the tidal marsh, it is highly stained, preventing identification of the source ma-
terial. A large knife (Fig. 5.10a) made of quartzite was dredged from the bottom
of Mopjack Bay near Norfolk, Virginia. Use-wear studies suggest that it was not
hafted, but rather it was hand-held. A heavily resharpened biface (Fig. 5.10e), was
also dredged from Mopjack Bay. Like the Cinmar biface, this tool was made of
banded rhyolite and was used as a hafted knife.
It is important that these specimens were found in circumstances indicating that
they were used and lost on the now-submerged continental shelf or the adjacent
lowlands along the LGM Susquehanna River channel. It is also evident that they
were all heavy-duty tools; likely used for butchering larger animals such as mast-
odons rather than smaller fauna.
If people settled eastern North America sometime between 23,000 and
15,000 years ago, why has this earlier archaeological record been so elusive? Per-
haps one reason is that the initial population of Paleolithic people favored the rich
diversity of the terrestrial and aquatic habitats of the now-submerged Continental
margins and adjacent major drowned river systems. As these coastal ecosystems
shifted westward due to rapidly rising sea levels approximately 14,500 years ago
and as the human population increased, settlement accelerated into the upland inte-
rior, whose archaeological record is not buried as it is on the inundated coastal plain.
Fig. 5.10 Laurel leaf bifaces from underwater contexts. a Mopjake Bay. b Cinmar. c Heavily
tumbled biface from Tar Bay. d Taylor’s Island. e Heavily resharpened knife from Mopjack Bay
91
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
It is important to note that the manufacturing technology used to produce the
Chesapeake Bay bifaces and the tool types themselves reflect the same technol-
ogy as that used by the Solutrean people of southwestern Europe during the LGM
(Stanford and Bradley 2012). Although more evidence is needed, it is not beyond
the realm of possibility to hypothesize that this early settlement of the East Coast
of North America resulted from a European Paleolithic maritime tradition. There
is little question that the Cinmar discovery indicates that exciting new chapters
in the story of Paleolithic people will be uncovered as archaeologists continue to
investigate the continental shelves of oceans worldwide (Earlandson 2001). (Note:
Funding has been obtained to conduct remote sensing survey of the area of sea floor
noted by Capt. Shawn during the summer of 2013).
Acknowledgements We thank our friends from Gwynn Island for their support, especially Jean
Tanner, Dean Parker, Thurston Shawn and the Crew of the Cinmar. We also thank Marcia Bakry,
Sarah Moore and Chip Clark for their help in the preparation of the text and illustrations.
References
Belknap, D. & Kraft, J. C. (1977). Holocene sea level changes and coastal stratigraphic units on
the northwest flank of the Baltimore Canyon trough geosyncline. Journal of Sedimentary Pe-
trology, 47(2), 610–629.
Bonder, G. H. (2001). Metarhyolite use during the Transitional Archaic in Eastern North America.
Paper presented at the 66th Annual Meeting of the Society for American Archaeology in New
Orleans, Louisiana, April 2001.
Clark, P., Dyke, A., Shakun, J., Carlson, A., Clark, J., Wohifarth, B., Mitrovica, C., Hostetler, S. &
McCabe, A. (2009). The last glacial maximum. Science, 325, 710–714.
Earlandson, J. (2001). The archaeology of aquatic adaptations: paradigms for a new millennium.
Journal of Archaeological Research, 9, 287–350.
Edwards, R. L., & Emery, K. O. (1977). Man on the continental shelf. In W. S. Newman & B.
Salwen (Eds.), Amerinds and their Paleoenvironments in Northeastern North America. Annals
of the New York Academy of Science, 228, 35.
Edwards, R. L., & Merrill, L. (1977). Reconstruction of the continental shelf areas of eastern North
America for the times 9,500 B.P. and 12,500 B.P. Archaeology of Eastern North America, 5, 1.
Emery, K. O., & Edwards, R. L. (1966). Archaeological potential of the Atlantic continental shelf.
American Antiquity, 31, 733.
Emery, K. O., Wigley, R., Barlett, A., Meyer, R., & Barghoorn, E. (1967). Freshwater peat on the
continental shelf. Science, 158, 1301–1307.
Fanning, D. S., Rabenhorst, M. C., Balduff, D. M., Wagner, D. P., Orr, R. S., & Zurheide, P. K.
(2010). An acid sulfate perspective on landscape/seascape soil mineralogy in the US Middle
Atlantic region, Geoderma, 154, 457–464.
Faure, H., Walter, R. & Grant, D. (2002). The coastal oasis: Ice age springs on emerged continental
shelves. Global and Planetary Change, 33, 47–56.
Glascock, M. D., Braswell, G. E. & Cobean, R. H. (1998). A systematic approach to obsidian
source characterization. In M. S. Shackley (Ed.), Archaeological obsidian studies: Method and
theory (pp. 15–65). New York: Plenum.
Grosman, L., Sharon, G., Goldman-Neuman, Y., Smikt O., & Smiansky, U. (2011). Studying post
depositional damage on Acheulian bifaces using 3-D scanning. Journal of Human Evolution,
60, 398–406.
92 D. Stanford et al.
Holmes, W. H. (1897). Stone implements of the Potomac-Chesapeake tidewater province. Fif-
teenth annual report of the Bureau of American Ethnology. pp. 3–152.
Kay, M. (1996). Microwear analysis of some clovis and experimental chipped stonetools. In G.
H. Odell (Ed.), Stone tools: Theoretical insights into human prehistory (pp. 315–344). New
York: Plenum.
Kay, M. (1998). Scratchin’ the surface: Stone artifact microwear evaluation. In M. B. Collins
(Ed.), Wilson Leonard: An 11,000-year archeological record of hunter-gatherers in central
Texas. Volume III: Artifacts and special artifact studies (pp. 743–794). Austin: Texas Archeo-
logical Research Laboratory, University of Texas at Austin and Archeology Studies Program,
Report 10, Texas Department of Transportation Environmental Affairs Division Studies in
Archeology 31.
Kraft, J. C., Belknap, D. F., & Kayan, I. (1983). Potentials of discovery of human occupation sites
on the continental shelves and near shore coastal zone. In G. H. Odell (Ed.), Quaternary coast-
lines and marine archaeology (pp. 87–120). New York: Academic.
Lowery, D. (2003). A landscape sculpted by wind and water: Archaeological and geomorpho-
logical investigations at the upper ridge site (44NH440) on Mockhorn Island in Northamp-
ton County, Virginia. Manuscript on file at the Virginia Department of Historic Resources,
Richmond, VA.
Lowery, D. (2008). Archaeological survey of the Chesapeake Bay shorelines associated with
Mathews County Virginia: An erosion threat study. Survey and Planning Report Series. Vir-
ginia Department of Historic Resources, Richmond, VA.
Lowery, D., & Wagner, D. (2012). Geochemical impacts to prehistoric iron-rich siliceous artifacts
in the nearshore coastal zone, Journal of Archaeological Science, 39, 690–697.
Lowery, D., O’Neal, M., Wah, J., Wagner, D., & Stanford, D. (2010). Late Pleistocene upland
stratigraphy of the western Delmarva Peninsula. Quaternary Science Review, 29, 1472–1480.
Mallinson, D., Riggs, S., Robert Thieler, E., Culver, S., Farrell, K., Foster, D., Corbett, D., Horton,
B., & Wehmiller, J. (2005). Late neogene and quaternary evolution of the Northern Albermarle
Embayment, Marine Geology, 217, 97–117.
Milliman, J. & Emery, K. (1968). Sea levels during the past 35,000 years. Science, 162, 1121–1123.
Nikitina, D. L., Pizzuto, J. E., Schwimmer, R. A., & Ramsey, K. W. (2000). An updated Holocene
sea-level curve for the Delaware coast. Marine Geolology, 171, 7–20.
Oldale, R., Colman, S., & Jones, G. (1993). Radiocarbon ages from two submerged strandline
features in the western Gulf of Maine and a sea-level curve for the Northeastern Massachusetts
coastal region. Quaternary Research, 40, 38–45.
Saunders, J. (1977). Late Pleistocene vertebrates of the western Ozark Highland, Missouri. Illinois
State Museum Reports of Investigations, 33, 1–118.
Shea, J. (1999). Artifact abrasion, fluvial processes and “living floors” from the Early Paleolithic
Site of ’Ubeidiya (Jordan Valley, Israel). Geoarchaeology, 14, 191–207.
Speakman, R. J., Glascock, M. D., Tykot, R. H., Descantes, C. H., Thatcher, J. J., & Skinner,
C. E. (2007). Selected applications of laser ablation ICP-MS to archaeological research. In
M. D. Glascock, R. J. Speakman, & R. S. Popelka-Filcoff (Eds.), Archaeological chemistry:
Analytical methods and archaeological interpretation, ACS Symposium Series Series 968
(pp. 275–296). Washington, DC: American Chemical Society.
Stafford, T., Jr., Hare, P., Currie, L., Jull, A., & Donahue, D. (1991). Acceleratoar radiocarbon dat-
ing at the molecular level. Journal of Archaeological Sciences, 18, 35–72.
Stanford, D., & Bradley, B. (2012). Across Atlantic ice: The origin of America’s Clovis culture.
Berkeley: University of California Press.
Stewart, R. M. (1984). South Mountain (meta) rhyolite: A perspective on prehistoric trade and ex-
change in the Middle Atlantic Region. In J. F. Custer (Ed.), Prehistoric lithic exchange systems
in the Middle Atlantic Region, monograph 3 (pp. 1–13). Newark: Center for Archaeological
Research.
Stewart, R. M. (1987). Rhyolite quarry and quarry-related sites in Maryland and Pennsylvania.
Archaeology of Eastern North America, 15, 47.
Waters, M., & Stafford, T. Jr. (2007). Redefining the age of Clovis: Implications for the peopling
of the Americas. Science, 315, 1122–1126.
93
5 New Evidence for a Possible Paleolithic Occupation of the Eastern …
Weaver, A. J., Saenko, O., Clark, P. U., & Mitrovica, J. X. (2003). Meltwater pulse 1A from Ant-
arctica as a trigger of the Bolling-Allerod warm interval. Science, 299, 1709–1713.
Whitmore, F., Emery, K., Cooke, H., & Swift, D. (1967). Elephant teeth from the Atlantic conti-
nental shelf. Science, 156, 1477–1481.